7. Case 1-2010 — A 75-Year-Old Man with Hypertension, Hyperglycemia, and Edema

Graham T. McMahon, M.D., M.M.Sc., Michael A. Blake, M.D., and Chin-Lee Wu, M.D., Ph.D.

Presentation of Case

Dr. Jennifer L. Lyons (Medicine): A 75-year-old man was admitted to this hospital because of the recent onset of hypertension, hyperglycemia, and edema.

The patient had been in his usual state of health, with borderline hypertension and glucose intolerance, until 11 days earlier, when at a routine visit to his internist, the blood pressure was 171/75 mm Hg. The pulse was 73 beats per minute and the weight 101 kg. The physical examination was normal. The serum levels of total protein, albumin, globulin, bilirubin, alkaline phosphatase, cholesterol, lipids, creatine kinase, iron, iron-binding capacity, ferritin, vitamin B₁₂, free thyroxine (T₄), and total triiodothyronine (T₃) and tests of liver function were normal; other laboratory-test results are shown in Table 1. He was instructed to check his blood pressure and serum glucose levels at home, and a follow-up appointment was scheduled.

During the following week, the patient reported blood pressures from 160 to 186 mm Hg systolic and from 79 to 82 mm Hg diastolic, and a glucose level (according to finger-stick testing after an overnight fast) of 170 to 286 mg per deciliter (9.4 to 15.9 mmol per liter). He noted ankle swelling, weight gain of 4.5 kg, pain in both calves that made it difficult to walk, and intermittent double vision. Two episodes of left-sided epistaxis occurred, which resolved spontaneously. The night before admission, a frontal headache developed (rated 6 of 10 on a scale in which 10 is the most severe). He recorded a systolic blood pressure of 206 mm Hg. He came to the emergency department of this hospital at 1:30 a.m.

The patient reported noticing pedal edema intermittently for 1 month, which had increased during the previous week; one episode of hematuria had spontaneously resolved. He did not have shortness of breath, orthopnea, paroxysmal nocturnal dyspnea, fevers, chills, nausea or vomiting, lower abdominal pain, or bowel or urinary symptoms. A diagnosis of hemochromatosis (homozygous C282Y mutation) had been made 8 years earlier, after routine laboratory tests showed elevated levels of serum iron and reduced iron-binding capacity, and was treated with regular phlebotomy. A diagnosis of prostate cancer had been made 22 years earlier and was treated with radical prostatectomy; pathological examination of the tissue reportedly showed no evidence of lymph-node metastases, but the serum level of prostate-specific antigen (PSA) did not fall to 0 after the operation. Four years before admission, the serum level of PSA was 21.0 ng per milliliter, and administration of bicalutamide was begun. Six months later, the level of PSA was 2.3 ng per milliliter, and 8 months before admission it was 8.75 ng per milliliter. Bilateral adrenal nodules (2.1 cm by 1.8 cm on the left, and 2.3 cm by 1.1 cm on right) had been present and stable in size on multiple computed tomographic (CT) studies of the abdomen during the previous 6 years and were thought to represent adenomas.

Bilateral thyroid adenomas had been present for 15 years; 11 years before admission, a left thyroidectomy revealed a follicular adenoma. A nodule on the right thyroid had been stable; 2 years earlier, pathological examination of a specimen from a fine-needle aspiration biopsy showed benign follicular cells consistent with follicular adenoma. He had a history of colonic adenomas, and 6 months before admission, a tubular adenoma with high-grade dysplasia had been removed during colonoscopy, with no evidence of residual adenoma on examination of another biopsy specimen 6 weeks later. He had had hypothyroidism since his thyroidectomy; he also had gastroesophageal reflux disease, osteoporosis, vitamin D deficiency, hyperlipidemia, migraine headaches, allergic rhinitis, and depression related to his wife’s recent illness and death. An episode of nephritis had occurred when he was in his 20s, which had resolved without sequelae. Five years before admission, a cardiac stress test and echocardiogram had been normal. He had had knee and bilateral cataract surgery. His parents had died of congestive heart failure, and a sister of nephritis; a grandfather had had throat cancer, and a first cousin colon cancer; his children and grandchildren were healthy.

The patient was retired from an office position and had lived alone since his wife’s death 7 months earlier. He drank alcohol rarely, had smoked cigarettes for 10 years but stopped 40 years earlier, and did not use illicit drugs or herbal supplements. Medications included aspirin, lorazepam, bicalutamide, celecoxib, chondroitin sulfate, glucosamine, levothyroxine, a multivitamin without iron, cholecalciferol, and oxycodone–acetaminophen. He was allergic to doxazosin, pramipexole, and nicotinic acid.

On examination, the temperature was 36.6°C, the blood pressure 186/79 mm Hg, the pulse 74 beats per minute, the respiratory rate 20 breaths per minute, the weight 105 kg, and the oxygen saturation 98% while the patient was breathing ambient air. The jugular veins were distended to 8 cm when the head was elevated to 30 degrees. There were mild expiratory wheezes diffusely and minimal crackles at the both lung bases. The first and second heart sounds were normal; a soft systolic murmur (grade 1/6) was heard at the right upper sternal border. Pulses were 2+ in the carotid arteries and extremities; no bruits were heard. There was mild bilateral gynecomastia. The abdomen was soft, with no distention or rebound tenderness. The liver measured 9 cm in the midclavicular line. There was 2+ pitting edema to the knees bilaterally. Funduscopic examination revealed sharp disks; visual acuity was 20/25 in the right eye and 20/100 in the left eye, with no diplopia. There was a mild left facial droop, and plantar responses were equivocal. The remainder of the examination was normal.

Tests of coagulation and serum levels of phosphorus, magnesium, and lipase were normal; other laboratory results are shown in Table 1. Urinalysis revealed glucose 3+ and blood 1+ and was otherwise normal. An electrocardiogram showed sinus rhythm, with nonspecific ST-segment and T-wave abnormalities. A chest radiograph was normal. CT of the abdomen after the administration of intravenous contrast material revealed multiple lesions in the liver (up to 2.5 cm in diameter), which were hypodense relative to the hepatic parenchyma after the administration of intravenous contrast material, and a hemangioma in segment 6, which had not changed from previous studies; bilateral adrenal nodules had increased slightly in size from previous studies (2.8 cm by 2.6 cm on the left, and 2.9 cm by 1.8 cm on the right). The patient was admitted to the hospital.

Insulin on a sliding scale, hydralazine, furosemide, and potassium chloride were administered, and sodium was restricted. Laboratory-test results showed no evidence of myocardial infarction. On the second day, the blood pressures ranged from 138 to 175 mm Hg systolic and from 64 to 79 mm Hg diastolic, and the weight was 99 kg. CT of the chest performed without the administration of contrast material revealed two nodules in each lung, 2 to 4 mm in diameter, and paratracheal and subcarinal lymph nodes, 0.8 to 1.0 cm in diameter, that had not been present on a study 6 years earlier. Results of laboratory tests are shown in Table 1. Other test results were pending.

On the third day, the blood pressure was 180/91 mm Hg. A diagnostic procedure was performed.

Differential Diagnosis

Dr. Graham T. McMahon: This 75-year-old man had a recent onset of hypertension, edema, hyperglycemia, and hypokalemic metabolic alkalosis; a rising level of PSA; and new liver lesions suggestive of metastatic tumor.

The acute onset of hypertension in an older patient warrants investigation. Acute, severe, or refractory blood-pressure elevation suggests secondary hypertension. Pheochromocytoma and hypothyroidism usually cause symptoms that this patient did not have, such as palpitations and constipation, respectively. Physical examination can provide clues that may suggest renovascular disease, coarctation of the aorta, sleep apnea, and Cushing’s syndrome. A laboratory evaluation may be necessary to rule out primary hyperaldosteronism or primary hyperparathyroidism among others.¹ In this case, mineralocorticoid excess and volume expansion are suggested by hypertension that is accompanied by edema, jugular venous distention, hypokalemia, and an elevated level of basic natriuretic peptide. Glucocorticoid excess is suggested by the presence of hyperglycemia and neutrophilia. In high concentrations, cortisol or its metabolites may induce volume retention by overwhelming 11β-hydroxysteroid dehydrogenase type 2 and activating the mineralocorticoid receptor.

May we review the imaging studies?

Dr. Michael A. Blake: CT of the abdomen obtained with the administration of intravenous contrast material on admission (Figure 1A) shows multiple new, low-density lesions throughout the liver, which are highly suggestive of metastatic disease. The adrenal glands are enlarged, and the adrenal nodules (Figure 1B) are bigger than they were on a scan obtained 2 years earlier. Examination of the pelvis was interpreted as showing no changes from previous examinations. Chest CT at the level of the thyroid gland shows evidence of the previous left hemithyroidectomy. There is heterogeneity of the remaining right lobe, which was stable in appearance and had been previously sampled, with no evidence of malignant conditions. Near the level of the hila, four new pulmonary nodules were noted, 5 mm or less in diameter, that were suggestive of metastatic disease (Figure 1C).

Figure 1. Imaging Studies on Admission.

An axial CT scan of the abdomen obtained after the administration of contrast material (Panel A) shows multiple low-density lesions (arrows) with slightly irregular walls throughout the liver, features suggestive of metastases. The adrenal glands are enlarged (Panel B), and the adrenal nodules (arrows) are bigger than they were 2 years earlier. An image from a chest CT scan with a lung window (Panel C) shows two of the new bilateral, small (5 mm or less in diameter), noncalcified nodules (arrows) that are suggestive of metastases.

Dr. McMahon: In endocrine disorders, the pace of the presentation can be an important indicator of the relative benignity of the underlying disease. In addition, pattern recognition is an important component of endocrinologic diagnosis, since most hormonal disorders present with diffuse metabolic and clinical derangements (Table 2), as in this case.² The rapid onset of hypertension and metabolic changes and the presence of new liver lesions raise the specter of paraneoplastic Cushing’s syndrome, with severe hypercortisolemia.

Cushing’s Syndrome

Cushing’s syndrome results from sustained hypercortisolemia. The most common cause is administration of exogenous glucocorticoids. Secretion of corticotropin from the pituitary (Cushing’s disease) accounts for approximately 70% of endogenous cases; adrenal tumors and the ectopic production of corticotropin each account for approximately 15% of cases.³ The clinical and laboratory features of Cushing’s syndrome overlap with many other medical conditions, and very few patients fulfill the classic presentation of facial rounding, weight gain, striae, hirsutism, hypertension, and muscle weakness. The majority of patients have abnormal glucose tolerance, but edema and hypokalemic alkalosis, as seen in this patient, occur in only a minority. This patient had gynecomastia, which is not typical in Cushing’s syndrome but can be a manifestation of hypogonadism or a consequence of hyperestrogenemia from a hormonally active tumor. Gynecomastia occurs in approximately 10% of men treated with bicalutamide.

The diagnosis of Cushing’s syndrome requires the confirmation of hypercortisolism, generally with the measurement of 24-hour urinary cortisol excretion, measurement of midnight salivary cortisol levels, or both. Autonomous production of cortisol can be demonstrated with the use of a dexamethasone (1-mg) suppression test. Once hypercortisolism is established, a corticotropin level of more than 20 pg per milliliter (4.4 pmol per liter) suggests corticotropin dependency; a level below 5 pg per milliliter (1 pmol per liter) suggests an adrenal source. When corticotropin dependency is established, magnetic resonance imaging can identify pituitary tumors approximately 60% of the time.⁴

The patient’s left-sided facial droop and left-sided epistaxis are not readily attributable to a benign pituitary adenoma. Nevertheless, pituitary enlargement in response to a tumor that produces corticotropin-releasing hormone or a primary benign or malignant pituitary tumor could account for some of the patient’s visual symptoms.

The standard method for differentiating between Cushing’s disease and the ectopic production of corticotropin involves venous sampling from the inferior petrosal sinus. Cushing’s disease is the most likely diagnosis if the sinus-to-peripheral-vein ratio of plasma corticotropin is at least 2:1 before or at least 3:1 after injection of corticotropin-releasing hormone during sampling from the inferior petrosal sinus. Ectopic production of corticotropin is implicated if 8 mg of dexamethasone does not suppress the plasma cortisol level, although this test lacks sensitivity.

Adrenal Nodules

The accelerated development of Cushing’s syndrome in a patient with known adrenal tumors introduces the differential diagnosis of adrenocortical carcinoma. Adrenal tumors, adrenal hyperplasia, and primary pigmented nodular adrenocortical disease can all lead to Cushing’s syndrome, although the presentation of these conditions tends to be gradual.

Adrenocortical carcinoma is very rare, with an annual incidence of approximately two cases per 1 million population.⁵ Of the 60% of adrenocortical carcinomas that are functional (i.e., hormone-secreting), 45% secrete both androgens and glucocorticoids; the rest secrete glucocorticoids alone (45%), androgens alone (10%), or, rarely, aldosterone (<1%).² The clinical manifestations of hormone excess in adrenocortical carcinomas are rapidly progressive, as in this case.

Imaging can be helpful in differentiating benign from malignant adrenal masses. Low attenuation on a CT scan obtained without the administration of contrast material suggests a substantial lipid content that is most consistent with adrenal adenoma. Adrenal adenomas also tend to wash out at least 60% of the contrast material within 15 minutes after contrast administration.⁶ Adrenal cancers generally have an irregular appearance, high attenuation, and rapid growth. None of these are apparent in this case, and the relative stability of this patient’s tumors in recent years makes a diagnosis of adrenocortical cancer unlikely. It is more likely that the recent bilateral adrenal enlargement in this case is due to either adrenal metastases or adrenal hyperplasia in response to corticotropin than to the development of hormonally active adrenal nodules.

The Ectopic Corticotropin Syndrome

Cushing’s syndrome was first reported in 1928 in a patient with small-cell lung cancer⁷; in the 1960s, corticotropin was shown to be the link between tumors (benign and malignant) and Cushing’s syndrome.⁸ In patients with cancer, Cushing’s syndrome generally develops quickly and is associated with extremely high levels of corticotropin and severe hypercortisolemia. Metabolic abnormalities tend to predominate in the clinical presentation, as in this case. Hypokalemia is present in approximately 70% of patients with Cushing’s syndrome and is related to the degree of hypercortisolemia.⁹ The physical phenotype of Cushing’s disease (i.e., facial fullness, a dorsocervical fat pad, and striae) may be absent in patients with ectopic corticotropin production, since these signs take weeks or months to develop. Pigmentation appears to be common in patients with small-cell lung cancer, and neuropsychiatric abnormalities have a particular association with neuroendocrine tumors.

Small-cell lung cancers, bronchial carcinoids, thymic tumors, islet-cell tumors of the pancreas, medullary thyroid carcinomas, and pheochromocytomas have all been associated with ectopic corticotropin secretion. The source of ectopic corticotropin remains unidentified in approximately 10% of patients with the syndrome, but that percentage has declined with improvements in imaging.¹⁰

Colon and Thyroid Carcinomas

The patient had a colonic lesion with high-grade dysplasia. Although metastatic colonic adenocarcinoma has been associated with the ectopic corticotropin syndrome, these cases are rare.^11,12 The patient also had follicular nodules of the thyroid. In contrast to medullary thyroid carcinomas,¹³ follicular adenomas and follicular carcinomas have not been associated with the ectopic corticotropin syndrome.

Prostate Cancer

This patient had a history of prostate cancer and progressive elevation in his serum PSA level, despite treatment with the androgen-receptor blocker bicalutamide. Small-cell neuroendocrine carcinoma accounts for only 1 to 2% of prostatic carcinomas,¹⁴ and prostatic tumors account for less than 2% of cases of the ectopic corticotropin syndrome.⁹ Neuroendocrine cells are found throughout the prostate and can secrete various active compounds including corticotropin, serotonin, chromogranin A, neuron-specific enolase, bombesin, calcitonin, and parathyroid-hormone–related protein. Neuroendocrine differentiation in prostate cancer has been reported to occur in patients treated with androgen ablation¹⁵ and carries a poor prognosis.¹⁴ Case reports of patients with ectopic corticotropin syndrome due to prostatic neuroendocrine carcinoma suggest that edema and hypokalemia are common and widely metastatic disease is present at diagnosis; the level of PSA is variably elevated.^{16,17,18,19,20,21,22}

Summary

Though the differential diagnosis remains broad, this patient’s clinical picture, particularly the acute nature of the presentation, is most consistent with a diagnosis of metastatic neuroendocrine cancer complicated by ectopic corticotropin syndrome. While awaiting the results of hormonal testing and medically stabilizing the patient, I would obtain a biopsy specimen of one of the liver lesions to confirm the diagnosis.

Dr. Nancy Lee Harris (Pathology): Dr. Lyons, would you tell us what the clinical thinking was?

Dr. Lyons: Our suspicion was high for Cushing’s syndrome due to carcinoma metastatic to the liver. The initial workup included diagnostic testing for Cushing’s syndrome, including measurement of urinary free cortisol and salivary cortisol levels, and then we arranged for a liver biopsy.

Clinical Diagnosis

Metastatic neuroendocrine carcinoma (most likely of prostatic origin), with the ectopic corticotropin syndrome.

Dr. Graham T. McMahon’s Diagnosis

Metastatic neuroendocrine carcinoma (most likely of prostatic origin), with the ectopic corticotropin syndrome.

Pathological Discussion

Dr. Chin-Lee Wu: A fine-needle aspiration of the liver was performed (Figure 2). On the cores of liver tissue, small groups of malignant tumor cells were seen in lymphatic channels. The tumor cells are small and round with scant cytoplasm, ill-defined cell borders, hyperchromatic nuclei with finely granular chromatin, nuclear molding, and inconspicuous nucleoli. On immunohistochemical staining, the tumor cells expressed the epithelial marker cytokeratin, thyroid transcription factor 1 (TTF-1), and the neuroendocrine marker chromogranin A; many cells expressed corticotropin, evidence that they are the source of ectopic corticotropin hormone. The tumor cells were negative for PSA.

Figure 2. Liver-Biopsy Specimen.

A biopsy of the liver (Panel A, hematoxylin and eosin) reveals a cluster of tumor cells in a lymphatic channel of liver parenchyma. The tumor cells are small and round with scant cytoplasm, hyperchromatic nuclei, a high nucleus:cytoplasm ratio, and ill-defined cell borders. Immunostaining shows that the tumor cells are diffusely positive for the epithelial marker cytokeratin (Panel B) and negative for the prostate marker prostate-specific antigen (Panel C) and that many of the tumor cells express the neuroendocrine marker chromogranin A (Panel D), as well as corticotropin (Panel E) and thyroid transcription factor 1 (Panel F).

Small-cell carcinoma is an aggressive neuroendocrine cancer most commonly seen in the lung, but it can also arise in extrapulmonary organs, including the prostate.^20,23,24 In some patients who have had a conventional adenocarcinoma of the prostate, the disease may recur as a small-cell carcinoma after hormonal therapy and present with a paraneoplastic syndrome.^25,26 Approximately half the cases of small-cell carcinoma of the prostate express TTF-1, a protein that is frequently expressed in thyroid and pulmonary tumors, including non–small-cell carcinomas.^20,23 Prostate markers such as PSA are usually not expressed in the small-cell carcinoma of the prostate.^20,23,25

In summary, the pathological diagnosis is metastatic small-cell carcinoma. The origin of the tumor cannot be determined on the basis of the pathological findings alone. However, evaluation did not reveal pulmonary or other extrapulmonary cancers. Thus, the combined pathological and clinical features are most consistent with small-cell carcinoma of prostatic origin.

Dr. Blake: CT scans of the pelvis performed 1 month after admission show a mass in the region of the pelvis that is consistent with recurrent prostatic cancer (Figure 3A). Although the pelvic CT scan on admission was initially reported as showing no change from previous studies, in retrospect, this mass was present but was partially obscured by streak artifact from surgical clips. An ¹⁸F-fluorodeoxyglucose (FDG) positron-emission tomographic scan (Figure 3B) obtained 1 month later showed intense uptake of FDG in the liver metastases; there was also intense uptake in bony metastases that was not apparent on the initial CT scan. The adrenal masses showed only mild FDG uptake, suggesting that the increase in their size may have been due to the influence of corticotropin, as proposed by Dr. McMahon.

Figure 3. Follow-up Imaging Studies.

An axial image from a pelvic CT scan obtained after the administration of intravenous contrast material shows a mass (Panel A, arrow) in the region of the prostate that is highly suggestive of a prostatic tumor. The lesion was present on admission but was not recognized because of streak artifact from the surgical clips. An axial image from a positron-emission tomographic scan after the administration of ¹⁸F-fluorodeoxyglucose (FDG) shows intense uptake of FDG in the liver metastases (Panel B, arrows) and in a bony metastasis in the spine (arrowhead). Neither of the adrenal masses show significant uptake, suggesting that their increase in size may have been due to the influence of corticotropin. Coronal images from a technetium-99m–labeled methylene diphosphonate bone scan (posterior view) (Panel C) and a CT scan displayed on bone windows (Panel D), both obtained 3 months later while the patient was receiving chemotherapy, show multiple abnormal foci of radiotracer uptake (arrows) in the bones corresponding to sclerotic lesions (arrows) on the CT scan, features consistent with bony metastases.

Discussion of Management

Dr. McMahon: Medical management of hypercortisolemia is often necessary in preparation for surgery or for palliation. The principal treatments are metyrapone and ketoconazole.²⁷ Metyrapone would need to be used cautiously in this case, since inhibition of 11β-hydroxylase may increase androgen levels in the presence of an elevated corticotropin level. Ketoconazole may be especially appropriate for a patient with prostate cancer, since it should reduce androgen production in the adrenal gland.²⁸ Other drugs include aminoglutethimide,²⁹ mitotane,^29,30 mifepristone, and somatostatin analogues.³¹ Surgical adrenalectomy is occasionally necessary if excision of the corticotropin-producing tumor is either impossible or not curative.

Dr. Harris: I would like to ask Dr. Geoffrey Walford from Endocrinology and Dr. Atish Choudhury from Medical Oncology to discuss their management of this patient’s disease.

Dr. Geoffrey A. Walford (Endocrinology): The diagnosis of corticotropin-dependent Cushing’s syndrome was confirmed by markedly elevated urinary free cortisol, late-night salivary cortisol, and serum corticotropin levels. Insulin, spironolactone, and antihypertensive medications were used to manage hyperglycemia, hypokalemia, and hypertension, respectively. Treatment with ketoconazole and, later, metyrapone was begun to reduce the synthesis of cortisol. The patient had a good response to this regimen. Cortisol levels fell to values at the low end of the normal range, and prednisone was begun to prevent the expected hypoadrenalism. Edema of the lower extremities resolved, and he was able to discontinue insulin, spironolactone, and antihypertensive agents.

Dr. Atish Choudhury (Oncology): The standard first-line palliative chemotherapy regimen for small-cell carcinoma is carboplatin and etoposide, which was initiated while the patient was in the hospital. Soon after the initiation of chemotherapy, the corticotropin level decreased to normal levels. The patient’s blood pressure, edema, and levels of blood sugars and electrolytes all improved toward normal.

Approximately 3 months later, after four cycles of chemotherapy, restaging by means of CT scans showed that the liver metastases and pulmonary nodules had decreased in size. Ketoconazole and metyrapone were discontinued. Despite resolution of visceral disease, the bony disease progressed.

Dr. Blake: A bone scan with technetium-99m–labeled methylene diphosphonate (Figure 3C) and a CT scan displayed on bone windows (Figure 3D) were obtained after the patient had received the four cycles of chemotherapy. Multiple abnormal foci of radiotracer uptake in the bones correspond to newly seen sclerotic lesions on the CT scan, features consistent with bony metastases.

Dr. Choudhury: We suspected that the visceral disease and the bony disease may represent two different processes, with the bony disease representing progression of his original prostatic adenocarcinoma in the bone. Although his PSA level had been rising while he was taking bicalutamide at initial presentation, we thought that the progression of his prostatic adenocarcinoma while his corticotropin levels were high could have been related to excessive adrenal production of androgens, which would have overwhelmed the ability of bicalutamide to inhibit their action. After the corticotropin level normalized, we restarted bicalutamide and initiated treatment with a gonadotropin-releasing hormone agonist (leuprolide); the PSA level decreased from 17 to 2 ng per milliliter.

Unfortunately, despite an excellent initial response to chemotherapy, the small-cell cancer recurred within 2 months after completion of six cycles of therapy; also, levels of corticotropin and cortisol rose and symptoms recurred. At this time, the patient elected not to pursue further therapy, and he died in the hospital in the company of family members, approximately 7 months after the diagnosis.

Dr. Richard J. Lee (Medical Oncology): The patient expressed gratitude before his death that he was to be the subject of a case history, so that others could learn from his experience.

Anatomical Diagnosis

Secondary Cushing’s syndrome due to metastatic small-cell carcinoma of prostatic origin.

Presented at the Medicine Case Conference, February 27, 2009.

Dr. Wu reports being listed as a coinventor on patents related to the diagnosis of prostate cancer; as of publication, these patents have not been licensed. No other potential conflict of interest relevant to this article was reported.

We thank Drs. Robert E. Singer (Internal Medicine), Richard J. Lee (Medical Oncology), and Giuseppe Barbesino (Endocrinology) for their assistance in preparing the case presentation.
Source Information

From the Division of Endocrinology, Diabetes, and Hypertension, the Department of Medicine, Brigham and Women’s Hospital (G.T.M.); the Departments of Radiology (M.A.B.) and Pathology (C.-L.W.), Massachusetts General Hospital; and the Departments of Medicine (G.T.M.), Radiology (M.A.B.), and Pathology (C.-L.W.), Harvard Medical School.

References

1. Chiong JR, Aronow WS, Khan IA, et al. Secondary hypertension: current diagnosis and treatment. Int J Cardiol 2008;124:6-21. [CrossRef][Web of Science][Medline]
2. DeLellis RA, Xia L. Paraneoplastic endocrine syndromes: a review. Endocr Pathol 2003;14:303-317. [CrossRef][Web of Science][Medline]
3. Nieman LK, Ilias I. Evaluation and treatment of Cushing’s syndrome. Am J Med 2005;118:1340-1346. [CrossRef][Web of Science][Medline]
4. Buchfelder M, Nistor R, Fahlbusch R, Huk WJ. The accuracy of CT and MR evaluation of the sella turcica for detection of adrenocorticotropic hormone-secreting adenomas in Cushing disease. AJNR Am J Neuroradiol 1993;14:1183-1190. [Abstract]
5. Schteingart DE. Adjuvant mitotane therapy of adrenal cancer — use and controversy. N Engl J Med 2007;356:2415-2418. [Free Full Text]
6. Blake MA, Jhaveri KS, Sweeney AT, et al. State of the art in adrenal imaging. Curr Probl Diagn Radiol 2002;31:67-78. [CrossRef][Medline]
7. Brown W. A case of pluriglandular syndrome. Lancet 1928;2:1022-1023.
8. Liddle GW, Nicholson WE, Island DP, Orth DN, Abe K, Lowder SC. Clinical and laboratory studies of ectopic humoral syndromes. Recent Prog Horm Res 1969;25:283-314. [Medline]
9. Isidori AM, Lenzi A. Ectopic ACTH syndrome. Arq Bras Endocrinol Metabol 2007;51:1217-1225. [Web of Science][Medline]
10. Ray D, Tomar N, Gupta N, Goswami R. Protein tyrosine phosphatase non-receptor type 22 (PTPN22) gene R620W variant and sporadic idiopathic hypoparathyroidism in Asian Indians. Int J Immunogenet 2006;33:237-240. [CrossRef][Web of Science][Medline]
11. Anthoney DA, Dunlop DJ, Connell JM, Kaye SB. Colonic adenocarcinoma associated ectopic ACTH secretion: a case history. Eur J Cancer 1995;31:2109-2112. [CrossRef]
12. Stambaugh JE Jr, Redfield ES. Ectopic production of ACTH by a primary adenocarcinoma of the colon. J Med Soc N J 1978;75:925-926. [Medline]
13. Mure A, Gicquel C, Abdelmoumene N, et al. Cushing’s syndrome in medullary thyroid carcinoma. J Endocrinol Invest 1995;18:180-185. [Web of Science][Medline]
14. Sciarra A, Cardi A, Dattilo C, Mariotti G, Di Monaco F, Di Silverio F. New perspective in the management of neuroendocrine differentiation in prostate adenocarcinoma. Int J Clin Pract 2006;60:462-470. [CrossRef][Web of Science][Medline]
15. Hansson J, Abrahamsson PA. Neuroendocrine differentiation in prostatic carcinoma. Scand J Urol Nephrol Suppl 2003;212:28-36. [Medline]
16. Alwani RA, Neggers SJ, van der Klift M, et al. Cushing’s syndrome due to ectopic ACTH production by (neuroendocrine) prostate carcinoma. Pituitary 2009;12:280-283. [CrossRef][Web of Science][Medline]
17. Suzuki N, Kobayashi Y, Yasuhara K, et al. A case of prostatic carcinoma with ectopic ACTH syndrome. Nippon Hinyokika Gakkai Zasshi 1993;84:2027-2030. [Medline]
18. Kataoka K, Akasaka Y, Nakajima K, et al. Cushing syndrome associated with prostatic tumor adrenocorticotropic hormone (ACTH) expression after maximal androgen blockade therapy. Int J Urol 2007;14:436-439. [CrossRef][Web of Science][Medline]
19. Stathem BN, Pardoe TH, Mir MA. Response of ectopic prostatic ACTH production to metyrapone. Postgrad Med J 1981;57:467-468. [Free Full Text]
20. Wenk RE, Bhagavan BS, Levy R, Miller D, Weisburger W. Ectopic ACTH, prostatic oat cell carcinoma, and marked hypernatremia. Cancer 1977;40:773-778. [CrossRef][Web of Science][Medline]
21. Nimalasena S, Freeman A, Harland S. Paraneoplastic Cushing’s syndrome in prostate cancer: a difficult management problem. BJU Int 2008;101:424-427. [Web of Science][Medline]
22. Rickman T, Garmany R, Doherty T, Benson D, Okusa MD. Hypokalemia, metabolic alkalosis, and hypertension: Cushing’s syndrome in a patient with metastatic prostate adenocarcinoma. Am J Kidney Dis 2001;37:838-846. [Web of Science][Medline]
23. Wang W, Epstein JI. Small cell carcinoma of the prostate: a morphologic and immunohistochemical study of 95 cases. Am J Surg Pathol 2008;32:65-71. [Web of Science][Medline]
24. Yao JL, Madeb R, Bourne P, et al. Small cell carcinoma of the prostate: an immunohistochemical study. Am J Surg Pathol 2006;30:705-712. [CrossRef][Web of Science][Medline]
25. Oesterling JE, Hauzeur CG, Farrow GM. Small cell anaplastic carcinoma of the prostate: a clinical, pathological and immunohistological study of 27 patients. J Urol 1992;147:804-807. [Web of Science][Medline]
26. Têtu B, Ro JY, Ayala AG, Johnson DE, Logothetis CJ, Ordonez NG. Small cell carcinoma of the prostate. I. A clinicopathologic study of 20 cases. Cancer 1987;59:1803-1809. [CrossRef][Web of Science][Medline]
27. Shalet S, Mukherjee A. Pharmacological treatment of hypercortisolism. Curr Opin Endocrinol Diabetes Obes 2008;15:234-238. [Medline]
28. Sonino N. The use of ketoconazole as an inhibitor of steroid production. N Engl J Med 1987;317:812-818. [Web of Science][Medline]
29. Nieman LK. Medical therapy of Cushing’s disease. Pituitary 2002;5:77-82. [CrossRef][Medline]
30. Terzolo M, Angeli A, Fassnacht M, et al. Adjuvant mitotane treatment for adrenocortical carcinoma. N Engl J Med 2007;356:2372-2380. [Free Full Text]
31. de Bruin C, Feelders RA, Lamberts SW, Hofland LJ. Somatostatin and dopamine receptors as targets for medical treatment of Cushing’s syndrome. Rev Endocr Metab Disord 2009;10:91-102. [CrossRef][Web of Science][Medline]

6. Minimally Invasive Total Knee Arthroplasty for Osteoarthritis

Minimally Invasive Total Knee Arthroplasty for Osteoarthritis

Seth S. Leopold, M.D.

This Journal feature begins with a case vignette that includes a therapeutic recommendation. A discussion of the clinical problem and the mechanism of benefit of this form of therapy follows. Major clinical studies, the clinical use of this therapy, and potential adverse effects are reviewed. Relevant formal guidelines, if they exist, are presented. The article ends with the author’s clinical recommendations.

A 65-year-old woman with osteoarthritis of her right knee is referred by her primary care physician for orthopedic consultation. She is healthy except for some well-controlled hypertension. She is 5 ft 5 in. tall (165 cm) and weighs 160 lb (73 kg), so her body-mass index (the weight in kilograms divided by the square of the height in meters) is 26.6. Her arthritis is not limited to one part of her knee but is diffuse and severe. She has minimal joint deformity and good bone quality. For several years, she was able to control the pain in her right knee with indomethacin, but recently this has been insufficient. An intraarticular corticosteroid injection was performed several months ago, with limited effect. The surgeon recommends total knee arthroplasty. The patient has a friend who has told her that his surgeon used a “minimally invasive” approach for his total knee replacement, and it went well. The patient has investigated this approach on the Internet, and she isn’t sure what to do. She asks her primary care physician whether he recommends that she consider “minimally invasive” surgery.

The Clinical Problem

Osteoarthritis affects more people than any other joint disease¹ and is the most common cause of long-term disability in most populations over the age of 65 years.^2,3 Primary osteoarthritis is rare before the age of 40 years but becomes increasingly common each decade thereafter.¹ A report from the Third National Health and Nutrition Examination Survey estimated that 37.4% of adults in the United States who are 60 years of age or older have radiographic evidence of the condition.⁴ Although osteoarthritis is not a life-threatening disease, the morbidity associated with this condition is considerable; 80% of patients with osteoarthritis have limitation of movement, and 25% have difficulty performing major activities of daily living.⁵ The economic burden of osteoarthritis may exceed $60 billion per year in the United States.¹

Pathophysiology and the Effect of Therapy

The pathophysiology of osteoarthritis is complex and incompletely understood, although the hallmark of the disease is the loss of articular cartilage, with concomitant changes in the underlying bone.^6,7 Many factors appear to be associated with the development of this condition, including injury, genetics, changes in tissue structure, and chondrocyte aging.^7,8,9 Major trauma, such as an intraarticular fracture, clearly increases the risk of subsequent (post-traumatic) arthritis. Research is being done to try to characterize the degree to which minor traumatic events can precipitate cartilage breakdown and degenerative joint disease,¹⁰ particularly when the injury is compounded by obesity, joint malalignment, or other predisposing factors.¹¹

The anatomic features of osteoarthritis include loss of articular cartilage, eburnation (or “sclerosis”) of the subchondral bone, formation of osteophytes (or “bone spurs”), and the presence of degenerative subchondral cysts (Figure 1). In some patients, there is clinically significant inflammation, including effusions, warmth, and synovitis that is visible during surgery. When osteoarthritis of the knee becomes severe, joint deformities — most commonly, varus (“bowlegs”) or valgus (“knock-knees”) — can occur.

Figure 1. Radiograph of the Left Knee of a Patient with Osteoarthritis.

Panel A shows an anteroposterior view with visible eburnation of the subchondral bone (“sclerosis”) (arrow) and loss of articular surface cartilage (or joint-space narrowing) (arrowhead). Panel B shows a lateral view, in which patellar osteophytes are also readily seen.

Total knee arthroplasty is an operation that consists of removal of the damaged cartilage, correction of joint deformities, and replacement of the worn cartilaginous bearing surfaces (on the femur, tibia, and patella) with an artificial bearing (Figure 2). Arthroplasty is not a disease-modifying procedure but rather is a mechanical solution to a biologic problem.

Figure 2. Radiograph of the Right Knee of a Patient after Total Knee Arthroplasty.

Shown are an anteroposterior view (Panel A) and a lateral view (Panel B) of the tibial components (arrows) and the femoral components (arrowheads).

For some patients with severe joint damage, arthroplasty may be the only option that offers the possibility of restored mobility and freedom from pain. However, arthroplasty is a major operation, and recovery is difficult. Patients typically have substantial postoperative pain, which must be tolerated while they undergo the aggressive physical therapy that is required for a good outcome. They often use assistive devices for ambulation for 6 weeks or longer and require frequent physical therapy and narcotic analgesics for several months.¹² For these reasons, there is interest among surgeons and patients alike in methods that would make total knee arthroplasty less invasive.^13,14 A variety of different procedures and techniques, all intended to reduce the amount of tissue injury occurring during surgery, have been lumped together under the label “minimally invasive surgery” (Table 1). This general term is used to identify all such procedures in this review.

The purported advantages of minimally invasive approaches include less postoperative pain, a shorter hospital stay, an earlier return of control for quadriceps muscles (leading to a shorter period of dependence on external devices for ambulation), and a briefer convalescence.^{12,13,14,15,16,17,18,19} Although there is limited evidence suggesting that minimally invasive total knee arthroplasty results in longer-term benefits in function than does the traditional approach,^15,20 most surgeons think that any benefits that may accrue are confined to the recuperative period.

Clinical Evidence

The clinical studies evaluating the potential benefit of minimally invasive total knee arthroplasty have a number of limitations. Of the studies that have been conducted, very few are randomized, most are quite small, and some evaluate different methods or aspects of the minimally invasive approach, making a comparison among studies difficult. Most of these studies do not take into account the skill development or “learning curve” of the surgeons who are performing the procedures. In addition, all the studies have been performed in the context of the evolution of surgical practice. In large measure because of these issues, there remains real conflict in the orthopedic literature about the benefit of this surgical approach.

In one of the few randomized trials, Kolisek et al.²¹ compared a minimally invasive approach with the traditional approach in 80 patients. The investigators found no significant differences in clinical or radiographic results at 3 months of follow-up. However, the authors defined “minimally invasive” surgery entirely as a function of incision length, since no specialized instruments were used in the group undergoing minimally invasive procedures, and the surgeons everted the patella (i.e., rotated it 180° on its tendon) as part of the operation. The use of incision length alone, as in the study by Kolisek et al., is not considered an appropriate or sufficient determinant to classify the procedure as minimally invasive (Table 1).

One of the largest nonrandomized studies²² involved 200 consecutive total knee arthroplasty operations (100 using the traditional approach and 100 using the minimally invasive approach), which were performed by experienced joint surgeons. The investigators, McAllister and Stepanian, defined minimally invasive surgery as total knee arthroplasty performed without dissection into the quadriceps tendon, eversion of the patella, or dislocation of the tibiofemoral joint; these operations also involved the use of instrumentation designed specifically for minimally invasive surgery. Patients who underwent the minimally invasive procedure had less postoperative pain and an improved early range of motion. They also had a significantly decreased risk of requiring manipulation under anesthesia (14% vs. 2%, P<0.001), an additional procedure resulting from the failure of rehabilitation or pain control that exposes the patient to the risk of hemarthrosis, fracture, and tendon rupture.

In our own study,¹² we compared 100 minimally invasive procedures with 50 performed with the traditional approach. We found that the minimally invasive approach (with a definition very similar to that used by McAllister and Stepanian) reduced the length of hospital stay from 3.7 to 2.8 days (P<0.001). It also significantly reduced the percentage of patients receiving narcotics at 2 weeks and at 6 weeks and the need for assistive devices for walking at 2 weeks.

Clinical Use

Regardless of whether total knee arthroplasty is performed through a traditional or minimally invasive approach, it is a major surgical intervention, with all the attendant risks of surgery. In addition, not all patients who undergo total knee arthroplasty have relief of symptoms even after a full recovery: 8 to 23% of patients have residual pain or stiffness in the knee at long-term follow-up.²³ In light of these facts, it is prudent to exhaust reasonable nonsurgical alternatives before considering arthroplasty. Nonsurgical interventions for pain associated with osteoarthritis of the knee include nonsteroidal antiinflammatory drugs, certain non-narcotic analgesics (in particular, acetaminophen), and intraarticular injections (corticosteroids or viscosupplements).²⁴

It is reasonable to consider surgery when a patient who is medically fit and is willing to accept the risks associated with the operation has persistent, moderate-to-severe pain associated with activity despite nonsurgical interventions. The patient should have radiographic evidence of significant joint damage, and if there appears to be inconsistency between the radiographic image and symptoms, the physician should consider other explanations for the patient’s pain. Contraindications to total knee arthroplasty include active infection, thrombophilias, bleeding disorders, severe vascular disease or neurologic disease affecting sensory or motor function in the affected leg, and inadequate soft tissue to cover the joint.

An essential factor to take into account in considering surgery is the patient’s ability and willingness to participate in an aggressive regimen of postoperative physical therapy. Vigorous physical rehabilitation, including exercises specifically intended to require early and repetitive motion of the affected knee despite substantial pain, is necessary for a good result. Failures of rehabilitation, which often stem from problems in managing postoperative pain early on, can permanently prejudice the outcome.^25,26,27

Once the decision has been made to proceed with surgery, the discussion can turn to whether a traditional or minimally invasive approach is more appropriate. An essential factor to consider in this decision is the experience of the surgeon. In our study, we specifically evaluated the effect of experience and showed that a surgeon must typically perform 25 to 50 procedures using minimally invasive techniques before benefits of this approach can be expected.¹² A fairly high practice volume is also important in maintaining the skills necessary for minimally invasive surgery, a principle that obviously applies to conventional surgery as well.^28,29 Given the exclusion criteria of the major clinical studies, other relative contraindications to the minimally invasive approach may include previous open knee surgery,^{15,16,19,30,31} severe osteoporosis or rheumatoid arthritis,^14,30,31 obesity or increased limb girth,^12,14,15 and severe joint deformity.^12,14

Anesthesia can be performed with the use of any of a variety of approaches. Either general anesthesia or regional (spinal or epidural) anesthesia, with or without adjunctive peripheral-nerve block, is appropriate.

During the operation, the patient’s knee is typically positioned in some flexion, and tourniquet control is used to reduce bleeding. The traditional approach uses an anterior longitudinal incision of 6 to 9 in. in length, whereas the minimally invasive approach often uses a somewhat shorter anteromedial incision along the medial border of the patella, extending distally to the level of the tibial tubercle. The traditional approach may then involve a longitudinal incision through the quadriceps tendon; the minimally invasive approach opens the medial capsule and extends proximally and obliquely into either the midvastus plane or subvastus plane by a small amount (typically 1 to 3 cm) and avoids the quadriceps tendon (Figure 3). In the traditional approach, the patella is then everted, whereas in the minimally invasive approach, the patella is retracted laterally but not everted. In the traditional approach, the tibiofemoral joint is then dislocated and the knee hyperflexed. In the minimally invasive approach, that joint is left in situ without dislocation.

Figure 3. Elements of Minimally Invasive Total Knee Arthroplasty.

Panel A shows the anatomical relations of the deep-tissue surgical incision for total knee arthroplasty. The traditional incision typically extends into the quadriceps tendon. The minimally invasive incision spares the quadriceps tendon and extends into or beneath the vastus medialis muscle. Joint-space exposure is shown for traditional total knee arthroplasty (Panel B) and for the minimally invasive procedure (Panel C). In the traditional procedure, a larger incision is made, and retractors are placed in a fixed position for maximal exposure. The tibiofemoral joint is dislocated, and the patella is everted (rotated laterally 180° on its tendon). In the minimally invasive procedure, a relatively small incision is made, and retractors are shifted during surgery to create a “mobile window” for the minimum necessary exposure. The tibiofemoral joint is not dislocated, and the patella is retracted laterally without being everted.

In both techniques, cutting jigs and anatomic landmarks are used to determine the depth and orientation of tibial and femoral bone resections. In the minimally invasive procedure, the cutting guides are reduced in size, rounded, and designed to optimize accuracy through the smaller anteromedial window. Regardless of approach, careful attention to ligament balancing and protecting neurovascular structures must be maintained. Trial implants then are placed over the resected bone surfaces; joint stability, ligament balance, and range of motion then are assessed. If satisfactory, final components are inserted (Figure 4). Final hemostasis is then obtained, and the joint is irrigated and closed.

Figure 4. Components of Total Knee Arthroplasty.

Precise resections are made in the distal end of the femur, the proximal end of the tibia, and the posterior surface of the patella to fit the corresponding surfaces of the three arthroplasty components. The femoral component is typically made of metal (most commonly, a cobalt–chromium alloy). The patellar component is typically made of ultra-high-molecular-weight polyethylene (a plastic resin). The tibial implant is usually made of metal (either a titanium or a cobalt–chromium alloy). There is an exchangeable polyethylene bearing on the tibia, which makes it possible to replace the plastic articular surface without replacing the metal component if wear of the bearing surface occurs.

The period of convalescence varies. As noted, in our study,¹² we found that the mean hospital stay for patients undergoing minimally invasive total knee arthroplasty was 2.8 days, about 1 day shorter than the mean duration for the traditional approach. Physical therapy is initiated the day of surgery, with an emphasis on range of motion, gait training, safety, and transfers. A machine that provides continuous passive motion may be used to enhance the exercises performed with a physical therapist.³² Adequate analgesia during physical therapy is essential; pain is the most common threat to adequate progression of mobility. In general, patients who have undergone minimally invasive total knee arthroplasty require the use of a walker for about a week and the use of a cane for a week to 10 days, at which point unassisted ambulation is the norm. Physical therapy usually concludes by about 6 weeks with the minimally invasive approach; with the traditional approach, an additional month of therapy is usually required.

Appropriate thromboprophylaxis should be used after either traditional or minimally invasive surgery. A variety of agents, including aspirin, warfarin, unfractionated heparin, and low-molecular-weight heparin, are used for this purpose. Intermittent pneumatic-compression devices that prevent venous stasis and that may enhance fibrinolysis are also commonly used. There is currently substantial disagreement between the recommendations of the American Academy of Orthopedic Surgeons and those of the American College of Chest Physicians regarding the appropriate intensity of anticoagulation.^33,34

According to one study, the estimated mean cost of primary total knee arthroplasty was $29,290 on the basis of data collected from October 2005 through June 2006 at four high-volume centers.³⁵ The study did not distinguish between the cost of minimally invasive procedures and that of traditional operations. One expert has observed that although the cost of minimally invasive total knee arthroplasty could be less than that of the conventional procedure (as a result of a shorter hospital stay), current reimbursement policies may create incentives to keep the patient in the hospital longer.³⁶

Adverse Effects

Many of the complications of total knee arthroplasty are similar whether a minimally invasive or traditional procedure is performed. The most feared complication, infection of the joint, occurs in less than 1% of patients.³⁷ By contrast, thromboembolic disease is common even with appropriate thromboprophylaxis. Venographic studies indicate that 15% of patients may have deep venous thrombosis, but symptomatic thromboembolic events occur in only 2 to 3% of patients.³⁸ Nerve injuries, especially peroneal nerve palsy, occur in 1 to 2% of patients,³⁹ whereas arterial vascular injury is much rarer. As noted above, persistent pain or stiffness occurs in 8 to 23% of patients.²³ Prosthesis failure, typically requiring surgical revision, occurs in approximately 2% of patients at 5 years.⁴⁰

In some studies, specific problems associated with the minimally invasive approach include inferior mean alignment⁴¹ or an increased frequency of outliers in terms of alignment,^41,42,43 concern about wound healing,²¹ a longer surgical duration,⁴³ and an inability to validate claims regarding improvements in early recovery or the time to independent ambulation.^21,42 However, most of these reports come from studies that enrolled patients early in the surgeon’s learning curve or that describe an approach as minimally invasive solely on the basis of the incision length. For these reasons, the experience of the surgeon is an essential consideration in choosing minimally invasive total knee arthroplasty.

Areas of Uncertainty

As noted above, the designation “minimally invasive” total knee arthroplasty encompasses several different modifications in surgical technique. It has not been established which of these changes are essential to improving outcomes for the patient. Increasingly, all elements of the minimally invasive approach are being used together (Table 1). It therefore seems unlikely that any trial will be conducted to distinguish, for example, the benefit of sparing the quadriceps tendon from that of avoiding or minimizing tibiofemoral-joint dislocation.

Since the minimally invasive approach has been in common use for less than 5 years, it has not yet been established that long-term outcomes will be as good (even in the hands of experienced surgeons) as those of patients undergoing traditional total knee arthroplasty. Although the characteristics of the procedure itself do not suggest obvious grounds for concern in this regard, additional studies will be required to provide a definitive answer to this question.

A variety of further modifications of surgical technique have been described in the past few years. These include computer-assisted navigation to aid in the precision of bone excision and prosthesis alignment, as well as recurrent interest in unicompartmental (partial) knee arthroplasty, meniscus replacement, and other similar elaborations, some of which are sometimes described as “minimally invasive” techniques. The potential advantages of such approaches as compared with the operation outlined in this review are not known.

Guidelines

There are no formal guidelines on the subject of minimally invasive total knee arthroplasty. The American Academy of Orthopedic Surgeons offers only a brief comment on the guidelines page of its Web site and does not draw a distinction between minimally invasive total knee arthroplasty and minimally invasive total hip arthroplasty. The hip procedure has provoked substantial controversy because of a high frequency of major complications.^44,45 The group’s guidelines state that minimally invasive surgery for total joint replacement “is a promising, but evolving surgical technique that requires additional scientific evidence to validate its short- and long-term safety and effectiveness, in comparison to conventional joint replacement methods.”⁴⁶

In 2004, the American Association of Hip and Knee Surgeons released an advisory statement about minimally invasive hip and knee arthroplasty that cites both potential advantages and disadvantages of the two approaches.⁴⁷ Unfortunately, since most of the literature on minimally invasive total knee arthroplasty has been published since 2004, that statement is now out of date.

Recommendations

The patient who is described in the vignette may be a suitable candidate for minimally invasive total knee arthroplasty, but some further consideration of her specific circumstances is appropriate. I would begin by discussing the major therapeutic options with the patient; in particular, I would try to ascertain whether she feels strongly that her quality of life with medical therapy alone has become unacceptable. If her answer is yes, I would describe the advantages and disadvantages of both the traditional and the minimally invasive approaches for total knee arthroplasty. I would emphasize that the minimally invasive approach should be performed only by a surgeon with considerable expertise and experience. I would describe the operation itself briefly but would place more emphasis on what she should expect in the perioperative and immediate postoperative period. The patient should be informed that she would have substantial postoperative pain and that she would need to participate actively in an aggressive regimen of physical therapy in order to have a successful outcome: relief of symptoms and improvement in mobility.

Dr. Leopold reports receiving grant support from the Orthopedic Research and Education Foundation, which receives funding from several medical and surgical pharmaceutical and equipment manufacturers. No other potential conflict of interest relevant to this article was reported.
Source Information

From the Department of Orthopedics and Sports Medicine, University of Washington School of Medicine, Seattle.

Address reprint requests to Dr. Leopold at the Department of Orthopedics and Sports Medicine, University of Washington School of Medicine, Box 356500, Seattle, WA 98195, or at leopold@u.washington.edu

References

1. Buckwalter JA, Saltzman C, Brown T. The impact of osteoarthritis: implications for research. Clin Orthop Relat Res 2004;427:S6-S15. [CrossRef][Medline]
2. Felson D. The epidemiology of osteoarthritis: prevalence and risk factors. In: Kuettner KE, Goldberg VM, eds. Osteoarthritic disorders. Rosemont, IL: The American Academy of Orthopaedic Surgeons, 1995:13-24.
3. Praemer A, Furner S, Rice DP. Musculoskeletal conditions in the United States. Rosemont, IL: American Academy of Orthopaedic Surgeons, 1999.
4. Dillon CF, Rasch EK, Gu Q, Hirsch R. Prevalence of knee osteoarthritis in the United States: arthritis data from the Third National Health and Nutrition Examination Surgery 1991-94. J Rheumatol 2006;33:2271-2279. [Free Full Text]
5. WHO, Bone and Joint Decade. The global economic and healthcare burden of musculoskeletal disease. Geneva: World Health Organization, 2001. (Available at http://www.boneandjointdecade.org/.)
6. Felson DT, Lawrence RC, Dieppe PA, et al. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med 2000;133:635-646. [Free Full Text]
7. Goldring MB, Goldring SR. Osteoarthritis. J Cell Physiol 2007;213:626-634. [CrossRef][ISI][Medline]
8. Goldring SR, Goldring MB. The role of cytokines in cartilage matrix degeneration in osteoarthritis. Clin Orthop Relat Res 2004;427:Suppl:S27-S36. [CrossRef][Medline]
9. Fukui N, Purple CR, Sandell LJ. Cell biology of osteoarthritis: the chondrocyte’s response to injury. Curr Rheumatol Rep 2001;3:496-505. [Medline]
10. Lohmander LS, Atley LM, Pietka TA, Eyre DR. The release of crosslinked peptides from type II collagen into human synovial fluid is increased soon after joint injury and in osteoarthritis. Arthritis Rheum 2003;48:3130-3139. [CrossRef][ISI][Medline]
11. Felson DT. Risk factors for osteoarthritis: understanding joint vulnerability. Clin Orthop Relat Res 2004;427:Suppl:S16-S21. [CrossRef][Medline]
12. King J, Stamper DL, Schaad DC, Leopold SS. Minimally invasive total knee arthroplasty compared with traditional total knee arthroplasty: assessment of the learning curve and the postoperative recuperative period. J Bone Joint Surg Am 2007;89:1497-1503. [Free Full Text]
13. Bonutti PM, Mont MA, McMahon M, Ragland PS, Kester M. Minimally invasive total knee arthroplasty. J Bone Joint Surg Am 2004;86:Suppl 2:S26-S32.
14. Tria AJ Jr, Coon TM. Minimal incision total knee arthroplasty: early experience. Clin Orthop Relat Res 2003;416:185-190. [CrossRef][Medline]
15. Haas SB, Cook S, Beksac B. Minimally invasive total knee replacement through a mini midvastus approach: a comparative study. Clin Orthop Relat Res 2004;428:68-73. [CrossRef][Medline]
16. Laskin RS, Beksac B, Phongjunakorn A, et al. Minimally invasive total knee replacement through a mini-midvastus incision: an outcome study. Clin Orthop Relat Res 2004;428:74-81. [CrossRef][Medline]
17. Seon JK, Song EK. Navigation-assisted less invasive total knee arthroplasty compared with conventional total knee arthroplasty: a randomized prospective trial. J Arthroplasty 2006;21:777-782. [CrossRef][ISI][Medline]
18. Seon JK, Song EK, Yoon TR, Park SJ, Bae BH, Cho SG. Comparison of functional results with navigation-assisted minimally invasive and conventional techniques in bilateral total knee arthroplasty. Comput Aided Surg 2007;12:189-193. [CrossRef][ISI][Medline]
19. Song EK, Seon JK, Yoon TR, Park SJ, Bae BH, Cho SG. Functional results of navigated minimally invasive and conventional total knee arthroplasty: a comparison in bilateral cases. Orthopedics 2006;29:Suppl 10:S145-S147. [Erratum, Orthopedics 2007;30:327.] [ISI][Medline]
20. Flören M, Davis J, Peterson MG, Laskin RS. A mini-midvastus capsular approach with patellar displacement decreases the prevalence of patella baja. J Arthroplasty 2007;22:Suppl 2:S51-S57. [CrossRef]
21. Kolisek FR, Bonutti PM, Hozack WJ, et al. Clinical experience using a minimally invasive surgical approach for total knee arthroplasty: early results of a prospective randomized study compared to a standard approach. J Arthroplasty 2007;22:8-13. [ISI][Medline]
22. McAllister CM, Stepanian JD. The impact of minimally invasive surgical techniques on early range of motion after primary total knee arthroplasty. J Arthroplasty 2008;23:10-18. [CrossRef][ISI][Medline]
23. Lingard EA, Sledge CB, Learmonth ID. Patient expectations regarding total knee arthroplasty: differences among the United States, United Kingdom, and Australia. J Bone Joint Surg Am 2006;88:1201-1207. [Free Full Text]
24. The Cochrane Collaboration home page. (Accessed March 30, 2009, at http://www.cochrane.org.)
25. Anderson JG, Wixson RL, Tsai D, Stulberg SD, Chang RW. Functional outcome and patient satisfaction in total knee patients over the age of 75. J Arthroplasty 1996;11:831-840. [CrossRef][ISI][Medline]
26. Maloney WJ. The stiff total knee arthroplasty: evaluation and management. J Arthroplasty 2002;17:Suppl 1:S71-S73. [CrossRef]
27. Walton NP, Jahromi I, Dobson PJ, Angel KR, Lewis PL, Campbell DG. Arthrofibrosis following total knee replacement: does therapeutic warfarin make a difference? Knee 2005;12:103-106. [CrossRef][ISI][Medline]
28. Hervey SL, Purves HR, Guller U, Toth AP, Vail TP, Pietrobon R. Provider volume of total knee arthroplasties and patient outcomes in the HCUP-Nationwide Inpatient Sample. J Bone Joint Surg Am 2003;85:1775-1783. [Free Full Text]
29. Katz JN, Barrett J, Mahomed NN, Baron JA, Wright RJ, Losina E. Association between hospital and surgeon procedure volume and the outcomes of total knee replacement. J Bone Joint Surg Am 2004;86:1909-1916. [Free Full Text]
30. Tashiro Y, Miura H, Matsuda S, Okazaki K, Iwamoto Y. Minimally invasive versus standard approach in total knee arthroplasty. Clin Orthop Relat Res 2007;463:144-150. [Medline]
31. Lombardi AV Jr, Viacava AJ, Berend KR. Rapid recovery protocols and minimally invasive surgery help achieve high knee flexion. Clin Orthop Relat Res 2006;452:117-122. [CrossRef][Medline]
32. Milne S, Brosseau L, Robinson V, et al. Continuous passive motion following total knee arthroplasty. Cochrane Database Syst Rev 2003;2:CD004260-CD004260. [Medline]
33. American Academy of Orthopaedic Surgeons. Clinical guideline on prevention of pulmonary embolism in patients undergoing total hip or knee arthroplasty. (Accessed March 30, 2009, at http://www.aaos.org/research/guidelines/PE_guideline.pdf.)
34. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008;133:6 Suppl:381S-453S. [CrossRef][ISI][Medline]
35. Bozic KJ, Rubash HE, Sculco TP, Berry DJ. An analysis of medicare payment policy for total joint arthroplasty. J Arthroplasty 2008;23:Suppl 1:S133-S138. [CrossRef]
36. Bozic KJ, Hansen E. The economics of minimally invasive total knee arthroplasty. J Knee Surg 2006;19:149-152. [Medline]
37. Jämsen E, Huhtala H, Puolakka T, Moilanen T. Risk factors for infection after knee arthroplasty: a register-based analysis of 43,149 cases. J Bone Joint Surg Am 2009;91:38-47. [Free Full Text]
38. Douketis JD, Eikelboom JW, Quinlan DJ, Willan AR, Crowther MA. Short-duration prophylaxis against venous thromboembolism after total hip or knee replacement: a meta-analysis of prospective studies investigating symptomatic outcomes. Arch Intern Med 2002;162:1465-1471. [Free Full Text]
39. Schinsky MF, Macaulay W, Parks ML, Kiernan H, Nercessian OA. Nerve injury after primary total knee arthroplasty. J Arthroplasty 2001;16:1048-1054. [CrossRef][ISI][Medline]
40. Vessely MB, Whaley AL, Harmsen WS, Schleck CD, Berry DJ. The Chitranjan Ranawat Award: long-term survivorship and failure modes of 1000 cemented condylar total knee arthroplasties. Clin Orthop Relat Res 2006;452:28-34. [CrossRef][Medline]
41. Chin PL, Foo LS, Yang KY, Yeo SJ, Lo NN. Randomized controlled trial comparing the radiologic outcomes of conventional and minimally invasive techniques for total knee arthroplasty. J Arthroplasty 2007;22:800-806. [CrossRef][ISI][Medline]
42. Dalury DF, Dennis DA. Mini-incision total knee arthroplasty can increase risk of component malalignment. Clin Orthop Relat Res 2005;440:77-81. [CrossRef][Medline]
43. Huang HT, Su JY, Chang JK, Chen CH, Wang GJ. The early clinical outcome of minimally invasive quadriceps-sparing total knee arthroplasty: report of a 2-year follow-up. J Arthroplasty 2007;22:1007-1012. [CrossRef][ISI][Medline]
44. Bal BS, Haltom D, Aleto T, Barrett M. Early complications of primary total hip replacement performed with a two-incision minimally invasive technique. J Bone Joint Surg Am 2005;87:2432-2438. [Free Full Text]
45. Pagnano MW, Leone J, Lewallen DG, Hanssen AD. Two-incision THA had modest outcomes and some substantial complications. Clin Orthop Relat Res 2005;441:86-90. [CrossRef][Medline]
46. American Academy of Orthopaedic Surgeons. Clinical practice guidelines: advisory statement. (Accessed March 30, 2009, at http://www.aaos.org/Research/guidelines/guide.asp.)
47. American Association of Hip and Knee Surgeons. Physician advisory statement: minimally invasive and small incision joint replacement surgery: what surgeons should consider. (Accessed March 30, 2009, at http://www.aahks.org/member/resources/MIS_Surgeons.pdf.)

5. A 36-Year-Old Woman with Headache, Hypertension, and Seizure 2 Weeks Post Partum

Case 8-2009 — A 36-Year-Old Woman with Headache, Hypertension, and Seizure 2 Weeks Post Partum

Aneesh B. Singhal, M.D., W. Taylor Kimberly, M.D., Ph.D., Pamela W. Schaefer, M.D., and E. Tessa Hedley-Whyte, M.D. Dr. W. Taylor Kimberly: A 36-year-old woman was admitted to the hospital because of headaches, hypertension, and seizures.

Nineteen days before the current admission, she delivered healthy twins at 35.6 weeks of gestation by cesarean section (for breech presentation) at another hospital. She was discharged on the fifth day. Nine days before the current admission, she began to have intermittent, throbbing, bifrontal headaches, and 2 days later she saw her gynecologist. She rated the pain as 8 (on a scale of 0 to 10, with 10 being the most severe pain). The blood pressure was 150/72 mm Hg. She was referred to an internist the same day, but she did not see the internist because the headache resolved while she was in the waiting room, and she returned home. Headache recurred that evening, and she went to the emergency department of a second hospital, where the blood pressure was 190/80 mm Hg. Computed tomography (CT) and magnetic resonance imaging (MRI) of the brain and the results of laboratory tests were reportedly normal. Oxycodone–acetaminophen was given for pain; the blood pressure decreased to 168/70 mm Hg, and she was sent home.

Four days before admission, the patient saw the internist. She reported that the headaches were sudden in onset and were usually worse in the early morning, when they awakened her from sleep, and in the late afternoon. She described her current headache as dull and rated the severity of the pain as 2 out of 10. The blood pressure was 142/78 mm Hg; trace peripheral edema was present. The remainder of the examination was normal. Furosemide and potassium chloride were prescribed.

Two days before admission, a severe headache (10 out of 10 in severity) occurred, with nausea and photophobia. The patient returned to the emergency room of the second hospital; the blood pressure was 204/96 mm Hg. The hematocrit was 34.7%; the results of the remainder of the complete blood count were normal, as were the results of other laboratory tests, including measurements of serum electrolytes, magnesium, calcium, and phosphorus and tests of renal and liver function. A urinalysis revealed that the specific gravity was 1.020, the pH 6.0, and the protein 30 mg per deciliter, with 5 to 10 red cells per high-power field. Hydromorphone was administered intravenously. Shortly thereafter, a generalized tonic–clonic seizure occurred, with urinary incontinence and loss of consciousness for 2 minutes. A CT scan of the head was normal. Examination of the cerebrospinal fluid (CSF) revealed no cells, a protein level of 64 mg per deciliter (reference range, 15 to 45), and a glucose level of 54 mg per deciliter (3.0 mmol per liter) (reference range, 47 to 70 mg per deciliter [2.6 to 3.9 mmol per liter]); Gram’s staining and culture were negative. Metoclopramide, lorazepam, and fentanyl were given intravenously, and oxycodone and methyldopa orally; the systolic blood pressure ranged between 140 and 160 mm Hg, and a frontal headache (4 out of 10 in severity) persisted. Magnesium sulfate and hydralazine were administered by continuous intravenous infusion, and the patient was admitted to the intensive care unit (ICU).

The next day, MRI with magnetic resonance angiography (MRA) and venography revealed posterior white-matter changes on T₂-weighted sequences and multifocal narrowing and dilatation of all the intracranial arteries. On the morning of the third hospital day, the patient appeared confused. A 5-minute episode of difficulties in word-finding and pronunciation and right hemiparesis occurred. Aspirin was administered. A CT scan of the head showed no evidence of acute infarction or hemorrhage. The patient was transferred by ambulance to this hospital, arriving 4 hours after the episode of aphasia.

She had given birth to a healthy child 2 years earlier, after a full-term pregnancy, without complications. She had not had hypertension or headaches before or during her pregnancies, and routine prenatal screening was normal. She lived with her husband and children and did not smoke, drink alcohol, or use illicit drugs or oral contraceptives. Her mother had hypertension and her father had cardiovascular disease; there was no family history of eclampsia. Medications before admission included furosemide, iron, multivitamins, and oxycodone–acetaminophen. There were no known drug allergies.

On examination, the patient was alert and oriented, with a normal affect. The temperature was 36.8°C, the blood pressure 136/91 mm Hg, the pulse 95 beats per minute, the respiratory rate 24 breaths per minute, and the oxygen saturation 98% while she was breathing ambient air. There was 1+ peripheral edema. The remainder of the general physical examination was normal. Speech was fluent and clear, but responses were slow. The first cranial nerve was not tested; other cranial nerves were normal. Strength was graded as 4 out of 5 on the right and normal on the left. There was ataxia on tests of rapid alternating movements and finger-to-nose and heel–knee–shin testing; she was not able to touch her nose with her right forefinger. Gait was not tested. The remainder of the neurologic examination was normal. The D-dimer level was 1975 ng per milliliter (normal level, <500). The results of other laboratory tests, including a complete blood count; measurements of serum electrolytes, calcium, phosphorus, and magnesium; and tests of renal and liver function, were normal. The patient was admitted to the neurologic ICU.

Electroencephalography performed with portable equipment showed no abnormalities. Fosphenytoin, labetalol, and magnesium sulfate were given intravenously by continuous infusion. The systolic blood pressure ranged between 110 and 120 mm Hg. During the first hospital day, the patient’s level of alertness fluctuated and there was intermittent right hemiparesis. MRI and MRA of the brain showed extension of the parenchymal lesions and worsening of the multifocal arterial stenosis.

On the second hospital day, continuous electroencephalographic monitoring revealed intermittent slowing of delta and theta waves bilaterally, intermittent low-amplitude beta activity bilaterally, and no epileptiform discharges. The plasma triglyceride level was 168 mg per deciliter (1.9 mmol per liter) (reference range, 40 to 150 mg per deciliter [0.4 to 1.7 mmol per liter]), and the anticardiolipin IgM antibody level was 17.9 IgM phospholipid units (reference range, 0 to 15); the erythrocyte sedimentation rate was 51 mm per hour (reference value, <25). Measurements of other lipids, creatine kinase, troponin T, thyrotropin, and complement (C3 and C4) were normal. Tests for antibodies to proteinase 3, antibodies to myeloperoxidase, antineutrophil cytoplasmic antibodies, anticardiolipin IgG antibodies, antinuclear antibodies, and rheumatoid factor were negative. During the day, weakness developed in the right arm and both legs, and the patient’s level of consciousness decreased. She did not seem to track gaze but moved her fingers on command, withdrew arms and legs on stimulation, and occasionally answered questions. At times she was unresponsive, with her eyes open.

A diagnostic procedure was performed.

Differential Diagnosis

Dr. Aneesh B. Singhal: May we review the brain imaging studies?

Dr. Pamela W. Schaefer: MRI of the brain and MRA of the circle of Willis on postpartum day 11 were normal. Brain MRI performed the day before admission to this hospital, on postpartum day 18 (Figure 1A), revealed hyperintense regions involving the posterior parietal–occipital lobes, the left frontal cortex, and the subcortical white matter. These regions had elevated diffusion, which was consistent with vasogenic edema. MRA of the circle of Willis (Figure 1B) revealed multifocal stenoses in the proximal anterior, middle, and posterior cerebral arteries. Brain MRI performed 1 day later, on admission to this hospital, revealed extension of the previously identified hyperintense regions (Figure 1C), with new restricted diffusion (Figure 1D), a finding that was consistent with cytotoxic edema secondary to ischemia. MRA of the circle of Willis (Figure 1E) revealed worsening of the previously identified multifocal anterior, middle, and posterior cerebral-artery stenoses.

Figure 1. MRI Studies of the Brain.

A fluid-attenuated inversion recovery (FLAIR) image obtained on postpartum day 18 (Panel A) shows hyperintense regions in both parietal–occipital lobes (arrows) that had elevated diffusion (not shown), findings that are consistent with vasogenic edema. MRA of the circle of Willis (Panel B) shows multifocal stenoses in the proximal anterior, middle, and posterior cerebral arteries. A FLAIR image from MRI performed on admission to this hospital (postpartum day 19) shows a hyperintense lesion (Panel C) in the left frontal lobe (arrow). The lesion is hyperintense on diffusion-weighted images (Panel D, arrow), a finding consistent with ischemia. MRA (Panel E) shows worsening of the multifocal cerebral arterial stenoses. A FLAIR image from MRI performed on hospital day 6 (Panel F) shows bilateral cerebral infarction with edema and hemorrhage.

Dr. Singhal: I am aware of the diagnosis in this case. Recurrent headaches with hypertension developed in this 36-year-old woman 10 days after an unremarkable cesarean delivery of twins, which followed an uncomplicated pregnancy. Head CT, brain MRI and MRA, and the results of routine laboratory tests were initially normal. I shall begin by considering the differential diagnosis at this stage of her illness.

Primary Headache Disorders

The incidence of postpartum headaches (Table 1) is nearly 40% in the first week,^1,2 with migraine and tension-type headaches accounting for more than 75% of cases. This patient’s headaches were severe and reached peak intensity within minutes, which is atypical for migraine and is consistent with thunderclap headaches, as defined by the International Headache Society.³ Thunderclap headaches can be primary or secondary⁴; virtually every condition listed in Table 1 has been associated with thunderclap headache. Primary thunderclap headaches are associated with coughing, strenuous exertion, sexual activity, and even bathing. They can be diagnosed only after secondary causes are ruled out by diagnostic testing that includes brain imaging and CSF examination.

Subarachnoid Hemorrhage

The initial diagnostic evaluation of thunderclap headache should focus on subarachnoid hemorrhage from a ruptured brain aneurysm, which is found in approximately 25% of patients with thunderclap headache; 75% of patients with aneurysmal subarachnoid hemorrhage present with a thunderclap headache.⁵ Aneurysmal rupture seems unlikely as a cause of this patient’s headaches, because the imaging studies were initially normal, the CSF examination showed no red cells, and the thunderclap headaches recurred, which is distinctly uncommon for aneurysmal cerebral hemorrhage. The normal brain imaging, the results of CSF examination, and the absence of fever and nuchal rigidity also help rule out many other causes of secondary postpartum headache (Table 1).

Cerebral-Artery Dissections and Venous Sinus Thrombosis

The estimated incidence of cerebral venous sinus thrombosis is 12 cases in 100,000 deliveries, with cesarean delivery and pregnancy-related hypertension being important risk factors.⁶ Thunderclap headache develops in approximately 15% of patients with cerebral venous sinus thrombosis,⁷ and 20% of all patients with cerebral-artery dissections present with thunderclap headache.⁸ This patient’s magnetic resonance venogram showed no evidence of cerebral venous sinus thrombosis. Although the initial head MRA study was unremarkable, the neck arteries were not imaged, so the possibility of dissection remains.

Postdural Puncture Headache

Approximately 5% of postpartum headaches are postdural puncture headaches. They result from intracranial hypotension due to a persistent CSF leak after the administration of a spinal anesthetic or inadvertent dural puncture during the administration of an epidural anesthetic. These headaches are usually persistent and have a postural component, although they can be manifested as thunderclap headache.⁹ As is often the case, we do not have knowledge of the CSF opening pressure in this patient, which would be diagnostic. It is conceivable that changes in CSF pressure in some way contributed to the patient’s headaches and subsequent angiographic abnormalities.

Delayed Postpartum Eclampsia

In the second week after the onset of symptoms, the patient continued to have thunderclap headaches, which were associated with hypertension, pedal edema, and a generalized seizure. Repeated brain MRI showed bilateral lesions at the junction of the cortex and subcortex, which were suggestive of vasogenic edema. This imaging pattern is observed in several conditions, including eclampsia, and in the appropriate clinical setting is recognized as the reversible posterior leukoencephalopathy syndrome.^10,11 The time course of this patient’s symptoms and the clinical and imaging features are consistent with delayed postpartum eclampsia.^12,13

Eclamptic convulsions can occur before, during, or after delivery. Postpartum eclampsia occurs in 10 to 45% of women with eclampsia.^12,13 About half of the cases of postpartum eclampsia occur within 48 hours after delivery, and the remainder occur between 2 days and 4 weeks after delivery (delayed postpartum eclampsia), as in this case. The symptoms are identical to those of antepartum eclampsia and include occipital or frontal headaches such as thunderclap headaches, blurred vision, scotomas, photophobia, altered mental status, shortness of breath, and upper abdominal pain.

Magnesium sulfate is indicated to prevent further seizures in women with eclampsia and was administered in this case. Magnesium has cerebral vasodilatory effects, alters the expression of endothelin-1 receptors, and reduces the permeability of the blood–brain barrier. These actions are relevant, since regional vasoconstriction, altered cerebral autoregulation with cerebral hyperperfusion, endothelial dysfunction, and breakdown of the blood–brain barrier are central to the pathophysiology of vasogenic edema in patients with eclampsia.¹⁰

Pregnancy-Related Stroke

Confusion, transient aphasia, right hemiparesis, ataxia, and visual deficits developed in this patient, and repeated MRI showed ischemic strokes. Pregnancy-related stroke has an incidence of 34.2 cases per 100,000 deliveries.¹⁴ Risk factors, which are relevant to this patient, include an age of more than 35 years, multiple gestation, increased parity, and preeclampsia.¹⁴ The greatest period of risk is during the 6 weeks after delivery. Although pregnancy-related stroke can result from thrombophilia, embolism, cerebral venous sinus thrombosis, and other causes, in this case the repeated MRA showed progressive multifocal vasoconstriction. This raises specific diagnostic considerations.

Primary Angiitis of the Central Nervous System

Primary angiitis of the central nervous system is an inflammatory condition characterized by insidious headaches with multifocal neurologic deficits and elevated white-cell count and protein levels in the CSF.^15,16 The initial brain MRI is virtually always abnormal,¹⁶ with scattered small-vessel infarcts, often with diffuse white-matter changes. Cerebral angiography can be normal, since this disease affects small arteries beyond the resolution of conventional angiography. However, many patients have ectasia and narrowing of the medium-size arteries, as seen in this patient. Primary angiitis of the central nervous system is a progressive disorder requiring prompt immunosuppressive therapy. Serologic evaluation for the disorder in this patient was negative; however, serologic and imaging tests have limited use for this diagnosis.¹⁷ Nonetheless, in this patient, the dramatic clinical presentation with recurrent thunderclap headaches, the absence of CSF pleocytosis, and the initially normal MRI and MRA make primary angiitis of the central nervous system unlikely. The rapid changes on MRA also argue against intracranial atherosclerosis, infectious arteritis, fibromuscular dysplasia, and other pathological entities associated with angiographic “beading.”

Reversible Cerebral Vasoconstriction Syndrome (Postpartum Angiopathy)

This patient’s recurrent thunderclap headaches, benign CSF results, rapidly progressive brain edema, strokes, and dynamic arterial changes are highly suggestive of postpartum cerebral angiopathy, a reversible cerebral vasoconstriction syndrome.¹⁸

Reversible cerebral arterial narrowing has been reported since the 1960s, with variable nomenclature (eclamptic vasospasm, migraine angiitis, and central nervous system pseudovasculitis) that reflected the clinical setting or the presumed cause. In the past 20 years, a syndrome of cerebral vasoconstriction associated with diverse conditions (Table 2) but with similar clinical, imaging, and prognostic features,¹⁹ some cases of which had previously been thought to be primary angiitis of the central nervous system,^15,20 has been recognized. This syndrome, now known as reversible cerebral vasoconstriction syndrome, nonetheless remained underrecognized until the publication of review articles and reports on large case series that have helped to define and characterize it.^{18,21,22,23,24,25}

The reversible cerebral vasoconstriction syndrome typically affects relatively young persons (20 to 60 years of age), occurs twice as often in women as in men, and is characterized by recurrent thunderclap headaches and reversible segmental arterial vasoconstriction on serial brain imaging — all features of this case — with no histologic evidence of inflammation. Brain imaging is normal in approximately 70% of patients with the reversible cerebral vasoconstriction syndrome, and the rest have border-zone ischemic strokes, parenchymal hemorrhage, vasogenic edema, and nonaneurysmal subarachnoid hemorrhage overlying the cortical surface.^18,24,26 The reversible cerebral vasoconstriction syndrome has many features in common with both isolated thunderclap headache and the leukoencephalopathy syndrome,^26,27 which suggests that these entities may belong to the same spectrum of disorders. The angiographic abnormalities in reversible cerebral vasoconstriction syndrome, as in this patient, are dynamic and often subtle and typically resolve within 3 months.^18,24,25 Most patients recover completely, although neurologic impairment (and even death) from progressive vasoconstriction, stroke, and brain edema has been reported.^28,29 To prevent this progression, treatment with calcium-channel antagonists, corticosteroids, and blood-pressure–modulating agents is initiated, as it was in this patient.

Clinical Diagnosis

Delayed postpartum eclampsia with postpartum angiopathy (the reversible cerebral vasoconstriction syndrome) complicated by brain edema and ischemic and hemorrhagic strokes.

Pathological Discussion

Dr. Kimberly: The first diagnostic procedure was CT angiography of the brain.

Dr. Schaefer: CT angiographic examination (Figure 2A and 2B) on the second hospital day (postpartum day 20) confirmed the presence of multifocal severe proximal stenoses of the anterior, middle, and posterior cerebral arteries.

Figure 2. CT Angiography and Digital-Subtraction Angiography of the Brain.

A CT angiogram obtained on the second hospital day (postpartum day 20) (Panel A) shows multifocal stenoses in the proximal anterior cerebral arteries (arrows). There is also stenosis of the middle cerebral arteries (Panel B, arrowheads) and posterior cerebral arteries (Panel B, arrows). A magnified lateral view from a digital-subtraction angiogram obtained on hospital day 3 (Panel C) shows multifocal areas of vasoconstriction (arrows) in the branches of the right middle cerebral artery. A CT angiogram obtained on hospital day 4 after the injection of nicardipine (Panel D) shows nearly complete resolution of previously seen vasoconstriction in the territories of the anterior, middle, and posterior cerebral arteries.

Dr. Kimberly: These results supported the diagnosis of the reversible cerebral vasoconstriction syndrome. On the third hospital day, further neurologic deterioration occurred, with unresponsiveness, continuous lip smacking and chewing movements, and spasticity and hyperreflexia in all limbs, with extensor plantar responses. An external ventricular drain was placed to monitor intracranial pressure; the trachea was intubated, and cerebral angiography was performed while the patient was under general anesthesia.

Dr. Schaefer: A digital-subtraction angiogram obtained on the third hospital day showed severe segmental narrowing and dilatation of multiple intracranial arteries in the anterior, middle, and posterior vascular territories (Figure 2C).

Dr. Kimberly: Nicardipine was injected into the left vertebral and right and left internal carotid arteries, resulting in nearly complete resolution of the arterial narrowing. However, on the fourth day, head CT scans and CT angiographic images showed recurrent severe narrowing of the distal branches of the left middle cerebral artery and both posterior cerebral arteries, with increased mean transit times of blood through both posterior and distal territories of the middle cerebral artery. A transcranial Doppler study showed increased blood-flow velocities consistent with the presence of severe right and mild left vasospasm. Nimodipine was administered. Repeated cerebral angiography with the administration of intraarterial and intravenous nicardipine again resulted in improvement of the arterial narrowing. Treatment with intrathecal nicardipine, insulin on a sliding scale, cefazolin, nafcillin, propofol (titrated to intracranial pressure) and milrinone was begun.

During the next 24 hours, intravenous phenylephrine, fludrocortisone, and intravenous norepinephrine were administered in an effort to raise systemic blood pressure and improve cerebral perfusion. Transcranial Doppler studies showed continuing vasospasm. A head CT scan revealed progressive watershed-territory infarctions with extensive cerebral edema. Hypertonic therapy, mannitol, pentobarbital, acetylcysteine, and hypothermia were administered, but the patient’s condition did not improve. On the seventh hospital day, the pupils were dilated and did not react to light. Brain imaging showed extensive ischemia and edema in both cerebral hemispheres. After discussion with the patient’s family, supportive care was withdrawn, and the patient died on the eighth hospital day. An autopsy was performed.

Dr. Schaefer: After treatment with intraarterial nicardipine, both digital-subtraction and CT angiograms (Figure 2D) revealed nearly complete resolution of the previously identified vasoconstriction. Multiple head CT and brain MRI scans obtained during the patient’s hospital course showed the continued development of new ischemic regions and brain swelling. The final brain MRI scan on hospital day 6 (postpartum day 25) (Figure 1F) showed infarctions in the brain stem, bilateral thalami, and bilateral frontal, parietal, and occipital lobes, with hemorrhagic transformation in some regions of ischemia, diffuse brain swelling with effacement of the basilar cisterns, and transtentorial herniation.

Dr. E. Tessa Hedley-Whyte: Postmortem examination was restricted to the brain, which was swollen and soft and weighed 1504 g (normal range, 1250 to 1400). The vessels of the circle of Willis were normal in diameter and wall thickness. Coronal sectioning revealed a dusky cut surface with a blurred junction of gray and white matter, particularly in the border-zone areas (Figure 3A and 3B). A hemorrhage in the right frontal and parietal lobes, related to the ventricular drain track, extended across the midline into the left frontal lobe. The lateral ventricles were mildly dilated and contained blood clot. On microscopical examination, there was widespread recent infarction, approximately 48 to 72 hours old, with foci of hemorrhage consistent with reperfusion (Figure 3C). There was diffuse neuronal necrosis secondary to hypoxic–ischemic injury (Figure 3D). The small blood vessels in the white matter had scattered hemosiderin-laden macrophages in their adventitia, a feature suggestive of early hypertensive vascular changes. The arteries of the circle of Willis were normal, apart from one patch of subendothelial thickening in the posterior cerebral artery (Figure 3E and 3F). No specific morphologic abnormalities have been described that correlate with the phenomenon of diffuse vascular spasm.

Figure 3. Postmortem Examination of the Brain.

Coronal sections of the left (Panel A) and right (Panel B) cerebral hemispheres show hemorrhage, originating around the drain tract, that extends across the corpus callosum into the left hemisphere. The gray matter–white matter junctions are indistinct, and overall, the brain has a dusky appearance consistent with diffuse ischemic damage. A section of parietal cortex (hematoxylin and eosin) shows an acute infarction with neutrophils and axonal retraction balls (Panel C, arrow) and neurons (Panel D) with eosinophilic cytoplasm characterized by loss of Nissl substance, which indicates neuronal death (arrows). A section of the right posterior cerebral artery (Panel E, hematoxylin and eosin) shows normal elastica and mild endothelial thickening. An Epon-embedded section of the distal middle cerebral artery (Panel F, toluidine blue) shows normal arterial-wall structures. The internal elastic lamina has a normal appearance. Electron-microscopical examination of the arteries showed no abnormalities.

Dr. Kimberly: Because this patient had features shared by postpartum angiopathy and antepartum eclampsia, we investigated this relationship further. Blood levels of placental growth factor (PlGF) and a soluble PlGF receptor (sFlt-1)^30,31 (members of the vascular endothelial growth factor [VEGF] pathway) and a soluble form of the transforming growth factor β1 receptor (soluble endoglin)³² correlate with the presence of antepartum eclampsia and also predict its development. Plasma from the patient was analyzed in the laboratory of Dr. Ravi Thadhani at this hospital. We compared the results in this patient with the mean values reported in the literature for antepartum and postpartum patients^{30,31,32,33,34,35} (Table 3). This patient had a slightly elevated sFlt-1 level (121 pg per milliliter) but virtually no PlGF (<5 pg per milliliter) in her serum. Since sFlt-1 is an antiangiogenic factor that acts by binding the proangiogenic protein PlGF, the relative levels of these proteins may predict their functional status better than absolute levels. The sFlt-1:PlGF ratio is a better predictor of antepartum preeclampsia than the absolute levels, especially when the ratio is above 5.³⁶ This patient’s sFlt-1:PlGF ratio of 24.2 approaches the mean ratio of approximately 32 in patients with antepartum preeclampsia. These data suggest that a functionally low PlGF state, similar to that seen in antepartum preeclampsia, may have played a role in this patient’s disease process.

Dr. Nancy Lee Harris (Pathology): In a woman who is more than 2 weeks post partum, do you expect to find any PlGF in the serum?

Dr. Kimberly: Yes. Although it was first discovered in human placenta, PlGF is also produced by several other cell types, most notably endothelial cells, which release PlGF,³⁷ resulting in a mean PlGF level in nonpregnant women of 11.5 pg per milliliter.³⁸

Dr. Harris: Dr. Greene, would you like to comment?

Dr. Michael F. Greene (Obstetrics and Gynecology): I am not sure whether abnormalities in sFlt-1 and PlGF levels are a cause or a result of preeclampsia and eclampsia.

Dr. Kimberly: There is evidence that these proteins are causative. Administration of sFlt-1 in a pregnant-rat model produces both clinical and pathological changes of preeclampsia,³⁹ and coadministration of sFlt-1 with soluble endoglin causes a HELLP-like syndrome (hemolysis, elevated liver-enzyme levels, and a low platelet count).⁴⁰ In humans, inhibition of the VEGF pathway with bevacizumab (which is functionally analogous to excess sFlt-1) can cause symptoms similar to those of preeclampsia.^41,42

Dr. Singhal: Progression to death is uncommon in the reversible cerebral vasoconstriction syndrome. Unfortunately, this case illustrates that pharmacologic blood-pressure modulation, calcium-channel antagonists, and direct interventions such as balloon angioplasty or injection of vasodilators may not be effective in preventing disease progression. Opening the artery may expose the brain to the risks of reperfusion injury. Nevertheless, the prompt but transient relief of vasoconstriction with the use of a vasodilator supports vasospasm as the underlying mechanism. Further research should focus on uncovering precise mechanisms⁴³ for the various conditions included in the reversible cerebral vasoconstriction syndrome in order to develop specific therapies.

Anatomical Diagnoses

Multiple bilateral acute cerebral infarcts with large right intracerebral hemorrhage arising in a subacute infarct (secondary to arterial spasm).

Diffuse cortical neuronal necrosis (due to ischemia).

Hypertensive small-vessel changes, mild.

Cerebral edema.

(Reversible cerebral vasoconstriction syndrome.)

Dr. Hedley-Whyte reports having equity ownership in Becton Dickinson. No other potential conflict of interest relevant to this article was reported.
Source Information

From the Department of Neurology (A.B.S.), Partners Neurology Residency Program (W.T.K.), and the Departments of Radiology (P.W.S.) and Pathology (E.T.H.-W.), Massachusetts General Hospital; and the Departments of Neurology (A.B.S., W.T.K.), Radiology (P.W.S.), and Pathology (E.T.H.-W.), Harvard Medical School.

References

1. Goldszmidt E, Kern R, Chaput A, Macarthur A. The incidence and etiology of postpartum headaches: a prospective cohort study. Can J Anaesth 2005;52:971-977. [Free Full Text]
2. Stella CL, Jodicke CD, How HY, Harkness UF, Sibai BM. Postpartum headache: is your work-up complete? Am J Obstet Gynecol 2007;196(4):318.e1-318.e7.
3. Headache Classification Subcommittee of the International Headache Society. The International Classification of Headache Disorders: 2nd edition. Cephalalgia 2004;24:Suppl 1:9-160. [CrossRef][Medline]
4. Schwedt TJ, Matharu MS, Dodick DW. Thunderclap headache. Lancet Neurol 2006;5:621-631. [CrossRef][ISI][Medline]
5. van Gijn J, Kerr RS, Rinkel GJ. Subarachnoid haemorrhage. Lancet 2007;369:306-318. [CrossRef][ISI][Medline]
6. Lanska DJ, Kryscio RJ. Risk factors for peripartum and postpartum stroke and intracranial venous thrombosis. Stroke 2000;31:1274-1282. [Free Full Text]
7. de Bruijn SF, Stam J, Kappelle LJ. Thunderclap headache as first symptom of cerebral venous sinus thrombosis. Lancet 1996;348:1623-1625. [CrossRef][ISI][Medline]
8. Mitsias P, Ramadan NM. Headache in ischemic cerebrovascular disease. I. Clinical features. Cephalalgia 1992;12:269-274. [CrossRef][ISI][Medline]
9. Schievink WI, Wijdicks EF, Meyer FB, Sonntag VK. Spontaneous intracranial hypotension mimicking aneurysmal subarachnoid hemorrhage. Neurosurgery 2001;48:513-517. [CrossRef][ISI][Medline]
10. Hinchey J, Chaves C, Appignani B, et al. A reversible posterior leukoencephalopathy syndrome. N Engl J Med 1996;334:494-500. [Free Full Text]
11. Lee VH, Wijdicks EF, Manno EM, Rabinstein AA. Clinical spectrum of reversible posterior leukoencephalopathy syndrome. Arch Neurol 2008;65:205-210. [Free Full Text]
12. Matthys LA, Coppage KH, Lambers DS, Barton JR, Sibai BM. Delayed postpartum preeclampsia: an experience of 151 cases. Am J Obstet Gynecol 2004;190:1464-1466. [CrossRef][ISI][Medline]
13. Sibai BM. Diagnosis, prevention, and management of eclampsia. Obstet Gynecol 2005;105:402-410. [ISI][Medline]
14. James AH, Bushnell CD, Jamison MG, Myers ER. Incidence and risk factors for stroke in pregnancy and the puerperium. Obstet Gynecol 2005;106:509-516. [ISI][Medline]
15. Calabrese LH, Duna GF, Lie JT. Vasculitis in the central nervous system. Arthritis Rheum 1997;40:1189-1201. [ISI][Medline]
16. Salvarani C, Brown RD Jr, Calamia KT, et al. Primary central nervous system vasculitis: analysis of 101 patients. Ann Neurol 2007;62:442-451. [CrossRef][ISI][Medline]
17. Duna GF, Calabrese LH. Limitations of invasive modalities in the diagnosis of primary angiitis of the central nervous system. J Rheumatol 1995;22:662-667. [ISI][Medline]
18. Calabrese LH, Dodick DW, Schwedt TJ, Singhal AB. Narrative review: reversible cerebral vasoconstriction syndromes. Ann Intern Med 2007;146:34-44. [Free Full Text]
19. Call GK, Fleming MC, Sealfon S, Levine H, Kistler JP, Fisher CM. Reversible cerebral segmental vasoconstriction. Stroke 1988;19:1159-1170. [Free Full Text]
20. Hajj-Ali RA, Furlan A, Abou-Chebel A, Calabrese LH. Benign angiopathy of the central nervous system: cohort of 16 patients with clinical course and long-term followup. Arthritis Rheum 2002;47:662-669. [CrossRef][ISI][Medline]
21. Singhal AB, Caviness VS, Begleiter AF, Mark EJ, Rordorf G, Koroshetz WJ. Cerebral vasoconstriction and stroke after use of serotonergic drugs. Neurology 2002;58:130-133. [Free Full Text]
22. Singhal AB. Cerebral vasoconstriction syndromes. Top Stroke Rehabil 2004;11:1-6. [Medline]
23. Singhal AB, Bernstein RA. Postpartum angiopathy and other cerebral vasoconstriction syndromes. Neurocrit Care 2005;3:91-97. [CrossRef][ISI][Medline]
24. Ducros A, Boukobza M, Porcher R, Sarov M, Valade D, Bousser MG. The clinical and radiological spectrum of reversible cerebral vasoconstriction syndrome: a prospective series of 67 patients. Brain 2007;130:3091-3101. [Free Full Text]
25. Chen SP, Fuh JL, Chang FC, Lirng JF, Shia BC, Wang SJ. Transcranial color Doppler study for reversible cerebral vasoconstriction syndromes. Ann Neurol 2008;63:751-757. [CrossRef][ISI][Medline]
26. Singhal AB. Postpartum angiopathy with reversible posterior leukoencephalopathy. Arch Neurol 2004;61:411-416. [Free Full Text]
27. Bartynski WS, Boardman JF. Catheter angiography, MR angiography, and MR perfusion in posterior reversible encephalopathy syndrome. AJNR Am J Neuroradiol 2008;29:447-455. [Free Full Text]
28. Buckle RM, Duboulay G, Smith B. Death due to cerebral vasospasm. J Neurol Neurosurg Psychiatry 1964;27:440-444. [Free Full Text]
29. Williams TL, Lukovits TG, Harris BT, Harker Rhodes C. A fatal case of postpartum cerebral angiopathy with literature review. Arch Gynecol Obstet 2007;275:67-77. [Medline]
30. Levine RJ, Maynard SE, Qian C, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350:672-683. [Free Full Text]
31. Rana S, Karumanchi SA, Levine RJ, et al. Sequential changes in antiangiogenic factors in early pregnancy and risk of developing preeclampsia. Hypertension 2007;50:137-142. [Free Full Text]
32. Levine RJ, Lam C, Qian C, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med 2006;355:992-1005. [Erratum, N Engl Med 2006;355:1840.] [Free Full Text]
33. Hirshfeld-Cytron J, Lam C, Karumanchi SA, Lindheimer M. Late postpartum eclampsia: examples and review. Obstet Gynecol Surv 2006;61:471-480. [CrossRef][ISI][Medline]
34. Koga K, Osuga Y, Yoshino O, et al. Elevated serum soluble vascular endothelial growth factor receptor 1 (sVEGFR-1) levels in women with preeclampsia. J Clin Endocrinol Metab 2003;88:2348-2351. [Abstract]
35. Wikström AK, Larsson A, Eriksson UJ, Nash P, Nordén-Lindeberg S, Olovsson M. Placental growth factor and soluble FMS-like tyrosine kinase-1 in early-onset and late-onset preeclampsia. Obstet Gynecol 2007;109:1368-1374. [CrossRef][ISI][Medline]
36. Levine RJ, Thadhani R, Qian C, et al. Urinary placental growth factor and risk of preeclampsia. JAMA 2005;293:77-85. [Free Full Text]
37. Okamoto T, Niu R, Mizutani S, Yamada S. Levels of placenta growth factor in gestational trophoblastic diseases. Am J Obstet Gynecol 2003;188:135-140. [CrossRef][ISI][Medline]
38. Autiero M, Luttun A, Tjwa M, Carmeliet P. Placental growth factor and its receptor, vascular endothelial growth factor receptor-1: novel targets for stimulation of ischemic tissue revascularization and inhibition of angiogenic and inflammatory disorders. J Thromb Haemost 2003;1:1356-1370. [CrossRef][ISI][Medline]
39. Maynard SE, Min JY, Merchan J, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003;111:649-658. [CrossRef][ISI][Medline]
40. Venkatesha S, Toporsian M, Lam C, et al. Soluble endoglin contributes to the pathogenesis of preeclampsia. Nat Med 2006;12:642-649. [Erratum, Nat Med 2006;12:862.] [CrossRef][ISI][Medline]
41. Reversible posterior leukoencephalopathy syndrome and bevacizumab. N Engl J Med 2006;354:980-982. [Free Full Text]
42. Yang JC, Haworth L, Sherry RM, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N Engl J Med 2003;349:427-434. [Free Full Text]
43. Calabrese LH, Molloy ES, Singhal AB. Primary central nervous system vasculitis: progress and questions. Ann Neurol 2007;62:430-432. [CrossRef][ISI][Medline]

4. An 81-year-old man was admitted to this hospital because of massive rectal bleeding

Case 9-2009 — An 81-Year-Old Man with Massive Rectal Bleeding

David G. Forcione, M.D., Hasan B. Alam, M.D., Sanjeeva P. Kalva, M.D., and Joseph Misdraji, M.D. Dr. John C. Lamattina (Surgery): An 81-year-old man was admitted to this hospital because of massive rectal bleeding.

He had been in his usual state of health until 4 a.m. on the day of admission, when he awoke with a feeling of abdominal fullness, nausea, and rectal urgency. He passed a mixture (estimated volume, 0.7 liter) of bright-red blood and soft stool through the rectum. Emergency medical services were called. On their arrival, the blood pressure was 102/86 mm Hg, the pulse 46 beats per minute (which rose to 86), the respiratory rate 20 breaths per minute, and the oxygen saturation 89% while he was breathing ambient air. The skin was pink, warm, and dry. He was taken to the emergency department of another hospital. On arrival, he appeared alert. He reported no abdominal or rectal pain, heartburn, or dyspepsia, and he had no vomiting or dysphagia. The temperature was 36.6°C, the blood pressure 113/88 mm Hg, the pulse 101 beats per minute, and the respiratory rate 24 breaths per minute. The physical examination was normal except for the presence of blood in the rectum. Laboratory-test results are shown in Table 1. A percutaneous intravenous catheter was inserted. Radiographs of the chest revealed an ectatic aorta and prominence of the interstitium of the lungs peripherally and at the bases, with no focal pulmonary consolidation or free air under the diaphragm. A radiograph of the abdomen revealed a nonobstructive pattern of gas and scoliosis. Oxygen (3 liters per minute) was administered by nasal cannula, and furosemide was given intravenously.

Approximately 3 hours after arrival, three episodes of diarrhea with red blood occurred within a 19-minute period; each episode was preceded by the sensation of abdominal fullness and nausea. Results of laboratory tests are shown in Table 1. A second catheter was inserted, and normal saline (500 ml) was administered intravenously. The patient was transferred to this hospital, arriving in the emergency department 2 hours after the last episode.

Every year for the past few years, an episode of lower gastrointestinal bleeding had occurred. Two years earlier, endoscopic evaluation of the lower gastrointestinal tract at the time of a bleeding episode revealed diverticulosis throughout the colon, blood throughout the colon but no active bleeding, and no blood or mucosal abnormalities in the terminal ileum; esophagogastroduodenoscopy revealed superficial erosions and altered blood in the stomach, thought to be consistent with trauma from a nasogastric tube, and superficial erosions in the duodenum, without active bleeding. The patient had a history of hypertension, hyperlipidemia, congestive heart failure, coronary artery disease with a myocardial infarction (for which he underwent coronary-stent placement), chronic obstructive pulmonary disease, pulmonary fibrosis, gout, arthritis, spinal osteomyelitis, and a hip fracture. Laboratory-test results from 1 year earlier are shown in Table 1. He was a widower who lived alone. He had consumed alcohol daily and smoked in the past; he had stopped drinking several months earlier and had stopped smoking several years earlier. There was no family history of gastrointestinal cancer. He was allergic to penicillin, erythromycin, clindamycin, and vancomycin. Medications on admission included buprenorphine–naloxone, digoxin, modafinil, verapamil, aspirin, furosemide, escitalopram, simvastatin, folate, potassium chloride, finasteride, quinine, pantoprazole, galantamine, and a multivitamin. During the previous 5 months, he had used oxygen supplementation (4 liters per minute) at home.

On examination, the patient was in moderate respiratory distress and mildly diaphoretic. The blood pressure was 97/51 mm Hg, the pulse 95 beats per minute, the respiratory rate 28 breaths per minute, and the oxygen saturation 80% while he was breathing ambient air, which improved to 92% with oxygen (2 liters per minute by nasal cannula). The jugular venous pulse was visible at 10 cm. There were crackles in both lungs. The abdomen was soft, with hyperactive bowel sounds; no tenderness, distention, or hepatosplenomegaly was present. Red and maroon blood was present in the rectum. No hemorrhoids were seen. The feet were warm, the pulses were full, and the remainder of the examination was normal. Levels of serum electrolytes were normal; tests for creatine kinase cardiac isoenzymes and troponin I were negative. Other laboratory-test results are shown in Table 1.

Insulin, sodium polystyrene sulfonate, ondansetron, furosemide, esomeprazole, albuterol, normal saline (1 liter), fresh-frozen plasma (2 units), and packed red cells (4 units) were administered intravenously; the blood pressure rose as high as 110 to 117/45 to 60 mm Hg, with a heart rate of 90 to 95 beats per minute. A nasogastric tube was placed, and gastric lavage revealed no blood in the stomach. Hematochezia continued. Results of laboratory tests are shown in Table 1. Approximately 7 hours after arrival, a central intravenous catheter was inserted, and additional red cells (3 units) were infused. Dyspnea worsened, with respirations up to 32 per minute, and the trachea was intubated.

A diagnostic procedure was performed.

Differential Diagnosis

Dr. David G. Forcione: I am aware of the diagnosis in this case. This 81-year-old man with ischemic cardiomyopathy, chronic obstructive pulmonary disease, and chronic renal insufficiency presented with massive rectal bleeding for several hours. Evaluation for rectal bleeding 2 years earlier had disclosed colonic diverticula and no abnormalities of the upper gastrointestinal tract. In the emergency department, his hematocrit continued to fall, despite blood and fluid resuscitation to the point of fluid overload. The urgent questions are: Where is the bleeding coming from, and what is its cause?

Sources of Rectal Bleeding

Several clinical factors are helpful in determining the source and cause of acute (<3 days’ duration)¹ rectal bleeding in a case such as this. The nature of the bleeding (melena or hematochezia) and the rate (intermittent or acute and massive) should be considered. Melena would suggest a source of bleeding proximal to the ligament of Treitz (i.e., upper gastrointestinal bleeding), although slow bleeds from lower gastrointestinal sources (e.g., small bowel or right colon) may present with melena. Hematochezia (bright-red blood from the rectum), as seen in this patient, is most typical of colonic and small-bowel sources, but acute, massive upper gastrointestinal bleeding may also present with hematochezia, almost always with severe hemodynamic instability.

Aspiration of gastroduodenal contents with a nasogastric tube can help distinguish an upper from a lower gastrointestinal source of bleeding. Aspiration of nonbloody bilious contents is highly suggestive of a source distal to the ligament of Treitz (i.e., lower gastrointestinal bleeding). Bright-red blood or “coffee grounds” material signifies a high likelihood of an upper gastrointestinal source and mandates early upper endoscopy for diagnosis and treatment. In this case, not only did the nasogastric aspirate reveal no blood, but also no bile was aspirated. This result suggested that the source was distal to the ligament of Treitz but did not rule out a possible duodenal source. Acute lower gastrointestinal bleeding accounts for 20% of all major episodes of gastrointestinal bleeding; the majority of episodes occur in patients, like ours, who are over the age of 65 years.^2,3 Lower gastrointestinal bleeding may result in instability of vital signs, anemia, and the need for blood transfusion, as in this case. Although most cases are self-limited, the mortality rate in published series has been as high as 4%.⁴

The differential diagnosis of acute lower gastrointestinal bleeding is broad (Table 2). Taking a careful medical history is important, including noting a history of gastrointestinal disease; this patient had had gastric and duodenal erosions and colonic diverticula on previous examinations. Other medical conditions or medications (e.g., renal dysfunction, liver disease, alcohol use, and ulcerogenic, antiplatelet, or anticoagulant medications) may contribute to bleeding or complicate management. This patient had consumed alcohol until recently; he had chronic renal insufficiency and was taking both potassium chloride and aspirin.

The large bowel (colon and rectum) is the most common source of acute lower gastrointestinal bleeding. However, because of recent advances in imaging, the small bowel is increasingly identified as the source of acute lower gastrointestinal bleeding, accounting for approximately 5% of cases. Upper gastrointestinal sources account for up to 10% of cases of acute rectal bleeding.⁵ In this case, an upper gastrointestinal source is unlikely, because the patient was relatively stable hemodynamically, had a nasogastric aspirate that was negative for blood, and had a previous negative upper endoscopy. The small bowel has an intermediate probability of being the source. On the basis of the patient’s known colonic diverticulosis, the acute and massive nature of the gastrointestinal bleeding, and epidemiology, a colorectal source is most likely.

Causes of Rectal Bleeding

Patients who have neoplasms or vascular abnormalities rarely present with acute, massive hematochezia, and neither of these lesions was evident on recent colonoscopy. Despite the patient’s history of alcohol use, mesenteric or rectal varices are unlikely, because the physical examination and results of laboratory studies do not suggest portal hypertension. Diverticular hemorrhage is the most common cause of colorectal hemorrhage, and our patient was known to have colonic diverticulosis. Thus, the leading diagnostic contender is colonic diverticular hemorrhage. Nonetheless, bleeding from jejunoileal diverticula or a small-bowel neoplasm (e.g., gastrointestinal stromal tumor or hemangioma) is also possible.

Surgical Approach

Dr. Hasan B. Alam: The leading cause of complications and death in patients with massive gastrointestinal bleeding, such as this patient, is a delay or failure in controlling hemorrhage. Thus, the primary surgical goal is to control the hemorrhage rapidly, which is difficult when the source of bleeding is unclear. We need to strike the right balance between localizing the source of bleeding by analyzing the results of time-consuming diagnostic studies and proceeding with a lifesaving operation without preoperative localization. As a general rule, the precise cause of the hemorrhage is less important for planning the surgical approach than is the anatomic location. The following four questions dictate the choice of diagnostic studies and the urgency of operative intervention: First, where is the source of bleeding (in the stomach or duodenum, the colon, or the small bowel)? As compared with sources proximal to the ligament of Treitz, which are easy to identify on endoscopy, and colonic sources, which can be controlled with a total colectomy, primary small-bowel sources of hemorrhage are notoriously difficult to diagnose and treat. Second, is the patient actively bleeding? Third, how rapid is the bleeding, and is the patient hemodynamically unstable? Finally, is endoscopic or angiographic control possible at the source of bleeding, or will surgical intervention be required?

Resuscitation

A patient’s response to initial resuscitation typically guides the subsequent workup. Patients who do not have a response to initial resuscitation need to be rapidly transported to the operating room or the endoscopy suite (depending on the bleeding source) for control of the hemorrhage. If initial resuscitation results in normalization of the hemodynamic status, further diagnostic studies can be carried out when the bleeding has stopped or slowed.

Aggressive resuscitation with fluids (crystalloid solutions) alone may potentiate the cellular injury caused by shock and increase the risk of acute respiratory distress syndrome, multiorgan failure, hemodilution, and coagulopathy,⁶ and there is no evidence that the administration of large volumes of intravenous fluids in cases of uncontrolled hemorrhage improves outcomes.⁷ Thus, aggressive fluid resuscitation was not a substitute for rapid control of hemorrhage in this patient.⁸ In studies in animals, hypotensive resuscitation (in which the blood pressure was sufficient to permit perfusion but was not normalized) reduced the risk of death in models of hemorrhage.⁹ Thus, low-volume, careful resuscitation is reasonable when trying to provide a critically ill patient such as this one with definitive care. Patients with trauma who have requirements for massive packed red cells seem to have better outcomes when treated early with plasma and platelets (i.e., component therapy)¹⁰; therefore, we used fresh-frozen plasma and platelets, in addition to packed red cells during the initial resuscitation, to prevent the development of coagulopathy,^11,12 although this approach has not been validated by a prospective, randomized study.

Diagnostic Studies

This patient had a transient response but required ongoing resuscitation. Therefore, studies to determine the location of the hemorrhage had to be selected carefully to avoid unnecessary delays in definitive care.

The vast majority of massive lower gastrointestinal bleeding is caused by colonic disease. Although this patient had known colonic diverticulosis, active bleeding from diverticula had not been seen during previous episodes of rectal bleeding. An anorectal source was ruled out by physical examination; if indicated, one can also perform rigid sigmoidoscopy at the bedside. Return of nonbloody fluid from the nasogastric tube had ruled out a gastric but not a duodenal source. Thus, although the colon was the most probable source, we were not comfortable proceeding directly to colectomy. Our options for differentiating between primary small-bowel and colonic sources included endoscopy (colonoscopy or enteroscopy), angiography, and a nuclear scan of radioisotope-tagged red cells. When the source of bleeding remains unidentified preoperatively, it can be localized by laparotomy and intraoperative enteroscopy in 80% of cases¹³; however, surgical intervention often fails, so we elected to transport the patient to the angiography suite to locate the site of bleeding, with a plan for immediate transfer to the operating room.

Critical Care during Transport

To perform visceral angiography on this critically ill patient required transporting him from a critical care area (the emergency department) to the angiography suite, where critical care is not typically available. Morbidity resulting from the transport of critically ill patients ranges from 13%¹⁴ to 33%¹⁵ and requires major changes in care in up to 25% of cases.¹⁶ These risks must be weighed against the potential benefits of the diagnostic study. When transport is absolutely needed, a well-trained team must travel with the patient to deliver critical care en route — in essence, the intensive care unit (ICU) travels with the patient. The Society of Critical Care Medicine developed guidelines for the transport of critically ill patients¹⁷ that suggest that the hospital should have a formalized plan that addresses pretransport coordination and communication, transport personnel, transport equipment, monitoring during transport, and documentation. A full team of emergency room nurses, surgical residents, and an anesthesiologist accompanied this patient to the angiography suite, and an attending surgeon was ready to receive the patient in a prepared operating room after the procedure. In addition, because of its experience with critically injured trauma patients,¹⁸ this hospital has an acute care surgery team that can deliver emergency surgical care around the clock^19,20; this team was responsible for delivering preoperative, operative, and postoperative care to this patient.

Dr. Sanjeeva P. Kalva: Selective superior mesenteric arteriography (Figure 1A) revealed active extravasation of contrast material from a proximal jejunal branch. This branch was catheterized superselectively with a microcatheter. Arteriography through the microcatheter (Figure 1B) revealed focal extravasation of contrast material and pooling in the jejunum. There was a suggestion of increased vascularity at the site of contrast-material extravasation. We suspected a small-bowel tumor or diverticulum. We decided to embolize the bleeding vessel with the understanding that this would temporarily stop the bleeding and enable better hemodynamic stability, so that the patient could safely be taken to the operating room. The bleeding vessel was embolized with two coils and a Gelfoam pledget (Figure 1C). His hemodynamic status improved after embolization. The catheter was left in the jejunal branch to help the surgeon identify the bleeding segment quickly. The patient was sent to the operating room.

Figure 1. Mesenteric Angiography.

Superior mesenteric arteriography (Panel A) reveals active extravasation (arrow) of the contrast material in the proximal jejunum. Superselective angiography through a microcatheter placed in the jejunal artery (Panel B) shows active extravasation and pooling of contrast material (arrow) in the jejunum. After embolization with two coils (Panel C, arrow), there is no more active bleeding.

Jejunal Diverticulosis

Dr. Forcione: After preparing my differential diagnosis (without knowledge of the actual diagnosis), I reviewed the angiographic findings, which I believe are consistent with jejunal diverticular hemorrhage. Acquired diverticula may be seen throughout the small intestine: 79% in the duodenum, 18% in the jejunum and ileum, and 3% in all three segments.²¹ On the basis of data from upper gastrointestinal series with barium as the oral contrast agent, the overall frequency of duodenal diverticula is 6 to 7% of the population, and the overall frequency of jejunoileal diverticula is approximately 0.7%.²² Autopsy series, however, show that the prevalence of jejunoileal diverticula may be as high as 4.6%.²³ Jejunoileal diverticulosis is most commonly seen in men approximately the age of this patient, in the seventh or eighth decade of life. Among cases of jejunoileal diverticulosis, 35% are associated with diverticula of the colon, which was found in this patient, 26% with diverticula of the duodenum, and 2% with diverticula of the esophagus.²⁴

Of patients with jejunoileal diverticulosis, 46% have complications, as compared with 13% of patients with duodenal diverticulosis.²² Symptoms, including pain and malabsorption due to bacterial overgrowth, are seen in 40% of patients with small-intestine diverticulosis. Presentations with acute symptoms are seen in 18% of such patients and include perforation, diverticulitis, and hemorrhage.²⁵ Gastrointestinal hemorrhage is characteristically acute and massive, as it was in this case.²⁶ The diagnosis is rarely made preoperatively in patients with an acute onset of symptoms such as this one.

Clinical Diagnosis

Jejunal diverticulosis or tumor with hemorrhage.

Dr. David G. Forcione’s Diagnosis

Jejunal diverticulosis with hemorrhage.

Intraoperative Decision Making

Dr. Alam: In the operating room, inspection of the small bowel at laparotomy did not show evidence of blood in the lumen. The angiography catheter was easily located by palpation of the mesentery, and it was found to be pointing directly to a large jejunal diverticulum (Figure 2A). It had a small neck-to-sac ratio, which is typical of small-bowel diverticula, and renders them difficult to diagnose in studies involving an oral contrast agent, because of poor filling of the sac. Also, because of the small neck size, such diverticula are typically collapsed and difficult to identify in a fatty mesentery; nonetheless, in this case we were able to identify several diverticula clustered in a small segment of the proximal jejunum. We performed a segmental small-bowel resection (Figure 2B), which is the treatment of choice. No blood was seen in the lumen of the resected specimen; therefore, to confirm that we had removed the source of bleeding, we opened the diverticulum, which contained an exposed blood vessel with an overlying clot (Figure 2C).

Figure 2. Intraoperative Inspection of the Jejunum.

Multiple diverticula are present in the proximal jejunum (Panel A), located on the mesenteric side. The arrows mark the diverticula, and the arrowhead marks the diverticulum that was the source of hemorrhage and that was identified by palpating the angiographic catheter and coil. The specimen from the segmental jejunal resection (Panel B) shows multiple diverticula. The opened specimen (Panel C) shows a blood clot overlying the artery, which was the site of bleeding.

Pathological Discussion Dr. Joseph Misdraji: The segment of small intestine contained several diverticula along the mesenteric border; the largest were a few centimeters in diameter. Sections of the diverticula revealed herniation of the mucosa and muscularis mucosae through the muscularis propria (Figure 3A), findings consistent with acquired diverticular disease and inconsistent with congenital diverticula such as Meckel’s diverticula, which are surrounded by all layers of the bowel wall. Within the largest diverticulum were several areas of mucosal ulceration (Figure 3B). In one section, a large artery showing evidence of degeneration of its wall, fibrin, and reactive endothelial cells coursed underneath an ulcer. The location of this artery was consistent with the site of hemorrhage (Figure 3C).

Figure 3. Pathological Findings (Hematoxylin and Eosin).

A low-power microscopical view of the largest diverticulum (Panel A) shows mucosa (arrowhead) and muscularis mucosae (short arrow) herniating through the muscularis propria (long arrow), which is characteristic of acquired diverticula. A section from the base of the largest diverticulum (Panel B) at low power shows focal ulceration of the diverticular mucosa (arrows). Coursing beneath the ulcer is a large muscular artery (arrowheads). A high-power view of the artery (Panel C) shows inflammation extending into the wall, with mural degeneration, fibrin deposition (arrows), and endothelial reaction (arrowheads), features that are consistent with the site of hemorrhage.

The vast majority of cases of jejunoileal diverticulosis involve the jejunum, with a predilection for the proximal jejunum.^24,27,28,29 A minority of cases affect the ileum, either alone or together with the jejunum.^24,28 Jejunal diverticula tend to be multiple, as in this case, whereas ileal diverticula are just as likely to be solitary as multiple.^21,30 Small-bowel diverticula, like colonic diverticula, are presumed to be due to increased luminal pressure, resulting in herniation of mucosa through points of blood-vessel penetration through the bowel wall.^21,28 Pathologically, acquired diverticula occur on the mesenteric border of the bowel, where the vessels penetrate the wall; in contrast, congenital diverticula occur on the antimesenteric aspect of the bowel.²¹ As with other acquired diverticula, the diverticula in this case consist of mucosa, muscularis mucosae, and submucosa, but not muscularis propria. They can range in size from tiny to a pouch that is 4 to 5 cm in diameter.²⁴

Studies with barium as the oral contrast agent reportedly have shown uncoordinated peristalsis in the involved segment.²⁷ Although an underlying muscular or neural disorder could explain the altered intestinal motility, no pathologic abnormality other than thickened muscularis is found in most cases. In one study, examination of the resected intestinal segments of 10 patients showed changes of systemic sclerosis in 4 patients, visceral myopathy in 2, and visceral neuropathy with intranuclear inclusions in 1.³¹ These findings have not been duplicated in other series. We found no evidence of these disorders in the specimen in this case.

Dr. Lamattina: The patient made an uneventful recovery. He was transferred to the ICU postoperatively, the trachea was extubated the next morning, and he was transferred to the regular surgical ward on the second postoperative day. He was discharged home on the seventh postoperative day. About 2 months after discharge, I visited him in his home. He had returned to his baseline functional status and was doing well.

Dr. Nancy Lee Harris (Pathology): Are there any questions or comments?

Dr. Paul S. Russell (Transplant Surgery): What would have been the outcome if you had just placed the coil and had not performed the resection?

Dr. Kalva: It is likely that he would have bled again. The incidence of recurrent hemorrhage is high, and mortality rates for conservative and operative care are 80% and 14%, respectively.³² The causes of bleeding from jejunal diverticula include trauma to the mesenteric vessels, mucosal ulceration, and diverticulitis without ulceration,^29,32,33 so rebleeding may occur from the same or other diverticula.

Anatomical Diagnosis

Diverticular disease of the jejunum with hemorrhage.

Dr. Kalva reports receiving consulting and lecture fees from Cordis Endovascular and Johnson & Johnson and research support from AngioDynamics. No other potential conflict of interest relevant to this article was reported.

We thank Dr. Andrea Reid, the attending gastroenterologist for this patient in the Emergency Department, for her assistance in the preparation of the case history.
Source Information

From the Gastrointestinal Unit (D.G.F.); the Division of Trauma, Emergency Surgery, and Surgical Critical Care (H.B.A.); and the Departments of Radiology (S.P.K.) and Pathology (J.M.), Massachusetts General Hospital; and the Departments of Medicine (D.G.F.), Surgery (H.B.A.), Radiology (S.P.K.), and Pathology (J.M.), Harvard Medical School.

References

1. Davila RE, Rajan E, Adler DG, et al. ASGE guideline: the role of endoscopy in the patient with lower-GI bleeding. Gastrointest Endosc 2005;62:656-660. [CrossRef][ISI][Medline]
2. Gostout CJ, Wang KK, Ahlquist DA, et al. Acute gastrointestinal bleeding: experience of a specialized management team. J Clin Gastroenterol 1992;14:260-267. [ISI][Medline]
3. Longstreth GF. Epidemiology and outcome of patients hospitalized with acute lower gastrointestinal hemorrhage: a population-based study. Am J Gastroenterol 1997;92:419-424. [ISI][Medline]
4. Farrell JJ, Friedman LS. Gastrointestinal bleeding in the elderly. Gastroenterol Clin North Am 2001;30:377-407. [CrossRef][ISI][Medline]
5. Jensen DM, Machicado GA. Diagnosis and treatment of severe hematochezia: the role of urgent colonoscopy after purge. Gastroenterology 1988;95:1569-1574. [ISI][Medline]
6. Committee on Fluid Resuscitation for Combat Casualties, Institute of Medicine. Fluid resuscitation: state of the science for treating combat casualties and civilian trauma. Washington, DC: National Academy Press, 1999.
7. Kwan I, Bunn F, Roberts I. Timing and volume of fluid administration for patients with bleeding. Cochrane Database Syst Rev 2003;3:CD002245-CD002245. [Medline]
8. Alam HB, Rhee P. New developments in fluid resuscitation. Surg Clin North Am 2007;87:55-72. [CrossRef][ISI][Medline]
9. Mapstone J, Roberts I, Evans P. Fluid resuscitation strategies: a systematic review of animal trials. J Trauma 2003;55:571-589. [ISI][Medline]
10. Borgman MA, Spinella PC, Perkins JG, et al. The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma 2007;63:805-813. [ISI][Medline]
11. Holcomb JB. Damage control resuscitation. J Trauma 2007;62:Suppl:S36-S37. [CrossRef][ISI][Medline]
12. Velmahos GC, Alam HB. Advances in surgical critical care. Curr Probl Surg 2008;45:453-516. [CrossRef][Medline]
13. Lewis MP, Khoo DE, Spencer J. Value of laparotomy in the diagnosis of obscure gastrointestinal haemorrhage. Gut 1995;37:187-190. [Free Full Text]
14. Insel J, Weissman C, Kemper M, Askanazi J, Hyman AI. Cardiovascular changes during transport of critically ill and postoperative patients. Crit Care Med 1986;14:539-542. [ISI][Medline]
15. Smith I, Fleming S, Cernaianu A. Mishaps during transport from the intensive care unit. Crit Care Med 1990;18:278-281. [ISI][Medline]
16. Indeck M, Peterson S, Smith J, Brotman S. Risk, cost, and benefit of transporting ICU patients for special studies. J Trauma 1988;28:1020-1025. [ISI][Medline]
17. Warren J, Fromm RE Jr, Orr RA, Rotello LC, Horst HM. Guidelines for the inter- and intrahospital transport of critically ill patients. Crit Care Med 2004;32:256-262. [CrossRef][ISI][Medline]
18. MacKenzie EJ, Rivara FP, Jurkovich GJ, et al. A national evaluation of the effect of trauma-center care on mortality. N Engl J Med 2006;354:366-378. [Free Full Text]
19. Institute of Medicine. Hospital-based emergency care: at the breaking point. Washington, DC: National Academies Press, 2007.
20. Committee to Develop the Reorganized Specialty of Trauma, Surgical Critical Care, and Emergency Surgery. Acute care surgery: trauma, critical care, and emergency surgery. J Trauma 2005;58:614-616. [ISI][Medline]
21. Williams RA, Davidson DD, Serota AI, Wilson SE. Surgical problems of diverticula of the small intestine. Surg Gynecol Obstet 1981;152:621-626. [ISI][Medline]
22. Akhrass R, Yaffe MB, Fischer C, Ponsky J, Shuck JM. Small-bowel diverticulosis: perceptions and reality. J Am Coll Surg 1997;184:383-388. [ISI][Medline]
23. Wilcox RD, Shatney CH. Surgical implications of jejunal diverticula. South Med J 1988;81:1386-1391. [ISI][Medline]
24. Baskin RH Jr, Mayo CW. Jejunal diverticulosis: a clinical study of 87 cases. Surg Clin North Am 1952;:1185-1196.
25. Tsiotos GG, Farnell MB, Ilstrup DM. Nonmeckelian jejunal or ileal diverticulosis: an analysis of 112 cases. Surgery 1994;116:726-731. [ISI][Medline]
26. Patel VA, Jefferis H, Spiegelberg B, Iqbal Q, Prabhudesai A, Harris S. Jejunal diverticulosis is not always a silent spectator: a report of 4 cases and review of the literature. World J Gastroenterol 2008;14:5916-5919. [CrossRef][ISI][Medline]
27. Altemeier WA, Bryant LR, Wulsin JH. The surgical significance of jejunal diverticulosis. Arch Surg 1963;86:732-745. [Free Full Text]
28. Longo WE, Vernava AM III. Clinical implications of jejunoileal diverticular disease. Dis Colon Rectum 1992;35:381-388. [CrossRef][ISI][Medline]
29. Shackelford RT, Marcus WY. Jejunal diverticula — a cause of gastro-intestinal hemorrhage: a report of three cases and review of the literature. Ann Surg 1960;151:930-938. [CrossRef][ISI][Medline]
30. Wilcox RD, Shatney CH. Surgical significance of acquired ileal diverticulosis. Am Surg 1990;56:222-225. [ISI][Medline]
31. Krishnamurthy S, Kelly MM, Rohrmann CA, Schuffler MD. Jejunal diverticulosis: a heterogenous disorder caused by a variety of abnormalities of smooth muscle or myenteric plexus. Gastroenterology 1983;85:538-547. [ISI][Medline]
32. Taylor MT. Massive hemorrhage from jejunal diverticulosis. Am J Surg 1969;118:117-120. [CrossRef][ISI][Medline]
33. Wilcox RD, Shatney CH. Massive rectal bleeding from jejunal diverticula. Surg Gynecol Obstet 1987;165:425-428. [ISI][Medline]

3. A 23-Year-Old Woman with an Abnormal Papanicolaou Smear

Case 10-2009 — A 23-Year-Old Woman with an Abnormal Papanicolaou Smear

Mark A. Goldstein, M.D., Annekathryn Goodman, M.D., Marcela G. del Carmen, M.D., M.P.H., and David C. Wilbur, M.D. Dr. Naana Afua Jumah (Student, Harvard Medical School): A 23-year-old woman was seen in the gynecology clinic of this hospital because of a high-grade squamous intraepithelial lesion seen on pathological examination of a Papanicolaou (Pap) smear.

Four weeks earlier, the patient had come to the adolescent and young adult medicine clinic of this hospital to establish care and to receive counseling regarding oral contraception, screening for sexually transmitted infections, and vaccination for the human papillomavirus (HPV).

Menarche had occurred at the age of 13 years, and menses had been monthly and regular. She had been sexually active since the age of 21 and had had 18 male partners, most of whom had had other sexual partners, and she had no history of sexually transmitted infections or pelvic infections. She had used oral contraceptives for 3 months, when she first became sexually active, and reported consistent condom use for both contraception and protection against sexually transmitted infections, except on three occasions. Physical and gynecologic examination 20 months earlier had been normal; the Pap smear was normal, with endocervical cells present. Testing of cervical secretions for gonorrhea and chlamydia were negative. Five months later, a viral culture of a genital specimen and serologic tests for herpes simplex virus (HSV) type 1 and type 2 IgG and IgM antibodies were reportedly negative. Three months before this evaluation, she had been seen in the emergency department of this hospital for a possible accidental overdose of acetaminophen, which she had been taking for dental pain. The blood acetaminophen level was 25.9 mg per liter (reference range, 10 to 25; toxic level, >120); the level 3 hours later was normal. She was instructed on proper dosing of the drug and discharged. She had had multiple atypical nevi, and one of two previous excisions had reportedly shown dysplastic changes. She had had an adenoidectomy at the age of 10 years. The patient worked in an office and lived with roommates. She had been without health insurance since graduating from college 1 year earlier. She drank 5 to 10 alcoholic beverages 2 to 3 days per week, including during sexual activity, and reported several episodes of blackouts while drinking; she had smoked marijuana on a few occasions in the past and did not smoke tobacco. Her mother had a history of vitiligo and alcohol abuse; her father and two sisters had irregular nevi. There was no family history of coagulopathy or gynecologic cancer. She was allergic to codeine. Her only medication was lorazepam, as needed for anxiety during airplane flights.

On examination, she appeared well. The vital signs were normal; the weight was 60.2 kg, the height 162 cm, and the body-mass index (the weight in kilograms divided by the square of the height in meters) 23.0. The tonsils were enlarged. There were multiple nevi on the trunk, arms, and legs (one over the right scapula was 3 mm in diameter with irregular borders, and one in the left inguinal region was 4 to 5 mm with color variation). Sexual development was Tanner stage 5 (fully mature), with normal external genitalia. The cervix was friable and tender to palpation; the uterus was midline, retroverted, and antiflexed; and the ovaries were normal. The remainder of the examination was normal. Laboratory testing included a complete blood count; measurement of electrolyte and cholesterol levels; tests of renal, liver, and thyroid function; testing for syphilis and the human immunodeficiency virus; cervical screening for Neisseria gonorrhoeae and Chlamydia trachomatis; and a liquid-based Pap smear. The patient was counseled about her level of alcohol consumption. The first dose of HPV vaccine was administered, and a prescription for oral contraceptives was written. Follow-up for the second vaccination in the series and for management of the oral contraceptives was scheduled. She was referred to the dermatology clinic.

The next week, review of the Pap smear revealed a high-grade squamous intraepithelial lesion. All other laboratory-test results were normal, and tests for infectious diseases were negative.

Three weeks later, she was seen in the colposcopy clinic. The results of a pelvic examination were unchanged from the previous examination. Colposcopic examination, performed after application of 3% acetic acid, revealed the entire squamocolumnar junction with an area of aceto-white change with punctation between the 9 o’clock and 12 o’clock positions. A biopsy specimen of the lesion was obtained, and endocervical curettage was performed. Pathological examination revealed moderate-to-severe squamous dysplasia with koilocytosis. Review of the endocervical-curettage specimen revealed highly atypical glandular epithelium. Testing for gonorrhea and chlamydia was negative.

Four weeks later, the patient was seen in the gynecology oncology clinic of this hospital. A diagnostic procedure was performed.

Differential Diagnosis

Dr. Mark A. Goldstein: On her first visit to our clinic, this young woman, who had had many sexual partners since becoming sexually active 2 years earlier, was found to have a high-grade intraepithelial lesion, a precursor of cervical cancer. This process is invariably associated with infection with certain types of HPV, estimated to be the most common sexually transmitted infection in 15-to-24-year-old persons in the United States. Infection is most frequent in sexually active women under the age of 25 years, like this patient.¹

Risk Factors for HPV Infection

Adolescents and young adults such as this patient are at higher risk for sexually transmitted infections, including HPV, than older persons, because of both behavioral and biologic factors. Young persons are more likely to engage in high-risk behaviors, including unprotected intercourse, multiple sexual partners, and substance abuse in association with sexual acts, all of which this patient had done. In women, having sex during menses, having early coitarche, and being subject to coercive sex are also important risk factors; using alcohol, tobacco, marijuana, or oral contraceptives and having a new partner, especially if he is known for less than 8 months before initiation of sexual activity, are predictive of an incident HPV infection.^2,3,4,5 This patient used oral contraceptives and acknowledged using alcohol and marijuana, including during sex. Other risk factors may include inadequate knowledge of the risks of sexually transmitted infections and lack of access to appropriate medical care; this patient had been without health insurance since leaving college.

Biologic risk factors in young people include the more frequent presence of cervical ectopy and a larger transformation zone than in older women, inadequate production of cervical mucus, and increased susceptibility to trauma during intercourse. The presence of another sexually transmitted infection, such as vulvar warts or HSV infection, is an additional biologic risk factor for HPV infection in this age group.³ This patient had cervical ectopy but did not have other sexually transmitted infections.

Several studies have addressed the incidence of HPV infection in sexually active women in this patient’s age group. The studies show cumulative incidences of HPV infection during periods of 12 to 24 months of about 40% among those who had vaginal intercourse^2,4,5; 50 to 80% of sexually active teenage girls and young women test positive for HPV.^6,7 Consistent use of condoms reduces but does not eliminate the risk of cervical and vulvovaginal HPV infection.⁵ The risk is increased if the male partner has had multiple other partners or if his sexual history is not known.⁸ This patient had multiple new sexual partners during a short time and reported that on at least three occasions she failed to use condoms. We can thus assume that she was at high risk for an incident HPV infection.

Most HPV infections in female adolescents and young adults are transient and have little long-term clinical significance; 70% of women clear the infection within 1 year and more than 90% in 2 years.^2,3,7 Risk factors for persistent infection include infection with multiple strains, as well as immunocompromise.⁹ Persistent infection is associated with abnormal cervical cytology, and more than 90% of patients with low-grade squamous intraepithelial lesions,⁵ high-grade squamous intraepithelial lesions, or adenocarcinoma in situ are positive for high-risk HPV.¹⁰

Current guidelines recommend initiation of cervical-cancer screening within 3 years after first intercourse or by the age of 21 years, whichever occurs first.¹¹ Cervical screening may be performed earlier than stated in the guidelines if follow-up of the patient is unpredictable, the patient is immunocompromised, or there is concern that the patient may not have reported sexual abuse or sexual intercourse. This patient reported that her first vaginal sexual intercourse occurred when she was 21 years of age, and 7 months later she had her first Pap smear at her college health service. The specimen was adequate, endocervical cells were present, and it was negative for intraepithelial lesions and cancer. Her medical insurance lapsed, and she had no routine medical care after college graduation, resulting in a delay of cervical cytologic screening and of the initiation of the HPV vaccine. At the first visit to our clinic, 20 months after the initial negative cervical cytologic test, she reported multiple sexual partners and other high-risk sexual behaviors. The HPV vaccine was initiated, she was screened for sexually transmitted infections, and liquid-based cervical-cancer screening was performed in accordance with the published guidelines for adolescents and young adults (Table 1).^11,12,13

Dr. David C. Wilbur: A specimen from a liquid-based Pap test was obtained, and the slide was initially screened by a computerized scanning device, as is routine in the cytopathology laboratory at this hospital. The scanning device flagged the specimen as requiring manual review, which was performed by a cytotechnologist. Abnormal cells were identified, and the slide was then referred to the cytopathologist. The specimen contained a mixture of types of abnormal cells, ranging from those of a low-grade to those of a high-grade squamous intraepithelial lesion. Koilocytes (Figure 1A), which are cells with the classic features of a productive viral infection, were indicative of the presence of low-grade squamous intraepithelial lesions. These cells are representative of a transient infection with HPV. However, coexisting with the koilocytes were cells representative of a high-grade squamous intraepithelial lesion, which is a true neoplastic precursor lesion capable of progression to invasive carcinoma (Figure 1B). These cells are primitive in that they lack differentiation and have a high nucleus-to-cytoplasm ratio and a hyperchromatic nucleus with an irregular distribution of chromatin. The presence of a high-grade squamous intraepithelial lesion typically correlates with the presence of severe dysplasia or carcinoma in situ, known as grade 3 cervical intraepithelial neoplasia (CIN 3) on biopsy specimens.

Figure 1. Cytologic Specimen (Papanicolaou Stain).

A low-grade squamous intraepithelial lesion (Panel A) is characterized by cells with perinuclear halos (arrow) characteristic of a productive human papillomavirus cytopathic effect. The nuclei are enlarged and hyperchromatic and show irregular nuclear outlines. A high-grade squamous intraepithelial lesion (Panel B) is exemplified by an abnormal isolated cell (arrow) with a high nucleus-to-cytoplasm ratio, irregular chromatin distribution, and an irregular nuclear envelope.

Discussion of Management

Dr. Goldstein: The patient’s second cervical cytologic test, 2 1/4 years after her first intercourse, indicated the presence of a high-grade squamous intraepithelial lesion. Although the majority of low-grade squamous intraepithelial lesions regress,¹⁴ high-grade squamous intraepithelial lesions are likely to persist or progress, so that follow-up was indicated.¹⁵ Tests for human immunodeficiency virus were negative, and we referred the patient for colposcopy in accordance with recent consensus guidelines.¹⁶

Dr. Annekathryn Goodman: To determine the best treatment for this patient, an understanding of the natural history of cervical neoplasia is crucial. Persistent infection with HPV is strongly linked to the development of high-grade squamous intraepithelial lesions, which this patient already has, as well as invasive cancer. Although the time of progression from persistent HPV infection to high-grade squamous intraepithelial lesions has been thought to be 5 to 15 years, on the basis of cross-sectional studies,¹⁷ recent longitudinal studies of newly infected patients suggest that the interval may be much shorter, with cumulative 36-month progression rates to grade 2 or 3 CIN (CIN 2/3) from 7 to 20%, with the highest rates reported for HPV type 16 (HPV-16).^7,18,19,20 Thus, this patient’s diagnosis of high-grade squamous intraepithelial lesions within 2 years after becoming sexually active may not be as unusual as we might have thought.

What is the likelihood that this patient’s high-grade squamous intraepithelial lesions will progress to invasive cancer? Invasive cervical cancer is rare in adolescents and young women aged 13 to 20 years^7,21; however, few data exist on the natural history of high-grade cervical lesions in adolescents. Two retrospective studies estimated a likelihood of progression from CIN 3 to invasive cancer ranging from 2 to 12%.^22,23 The median age of women with invasive cancer is typically 10 years higher than that of women with CIN 3.⁷ Thus, although high-grade squamous intraepithelial lesions in adolescents and young women is unlikely to regress, it does not appear to progress rapidly to invasive cancer.

The 2006 consensus guidelines recommend colposcopic evaluation of patients with high-grade Pap smears, along with biopsy and endocervical curettage.¹⁶ Adolescents with high-grade squamous intraepithelial lesions are also immediately evaluated by colposcopy. Because of the low risk of invasive cancer in this population, diagnostic excisional procedures are avoided in adolescents, unless high-grade squamous intraepithelial lesions persist for a total of 24 months.

In this patient, I performed a colposcopic examination and applied acetic acid to distinguish normal from abnormal cervical epithelium and assess the extent of the high-grade dysplasia.²⁴ The normal, glycogenated epithelium, which has cells that have tiny nuclei and abundant cytoplasm, is not affected by this fixative, whereas dysplastic epithelial cells, which have larger nuclei and scant cytoplasm, become opaque and appear white (aceto-white) on examination. In this patient, a white area was evident, from which I took a biopsy specimen; I also performed an endocervical curettage.

Dr. Wilbur: Colposcopic biopsy showed evidence of a high-grade squamous dysplasia, predominantly CIN 2 but with focal areas indicative of CIN 3 (Figure 2A). As predicted by examination of the cytologic specimen, evidence of a productive viral infection in the form of koilocytotic change was also noted (Figure 2B). In addition, the endocervical curettage revealed an area of atypical endocervical epithelium (Figure 2C). Although a number of endocervical reactive processes may have similar features, such changes raise the possibility of endocervical neoplasia and must therefore be further investigated. Adenocarcinoma in situ is also associated with HPV. It is rarer than squamous dysplasia but has increased in incidence during the past 20 years, most likely because of better detection but also perhaps in association with increasing use of hormonal contraception.

Figure 2. Cervical-Biopsy Specimen (Hematoxylin and Eosin).

This specimen shows high-grade dysplasia with immaturity of the basal squamous cells extending into the midportion of the epithelium (Panel A). In addition to the maturation abnormality, another area of the epithelium (Panel B) shows atypical cells with perinuclear halos, nuclear enlargement, and multinucleation, known as koilocytosis (arrow). These changes are indicative of the viral cytopathic effect associated with human papillomavirus infection. Atypical endocervical epithelium was also present in the cervical-biopsy specimen (Panel C). The epithelium is pseudostratified and shows enlarged nuclei with hyperchromasia. This appearance is in contrast to normal simple endocervical epithelium (inset).

Dr. Goodman: The consensus guidelines recommend excision or ablation of carcinoma in situ and adenocarcinoma in situ at all ages. In adolescents and young adults who have solitary CIN 3 lesions that are confined to the exocervix, observation every 6 months for up to 2 years is appropriate. A lesion that extends into the cervical canal, is multifocal, or contains adenocarcinoma in situ, as this one does, should be excised (there are no accepted nonsurgical therapies for adenocarcinoma in situ). Therefore, I performed a cervical loop electrosurgical excision procedure (LEEP) to excise the high-grade lesions involving both the exocervix and endocervix. Dr. Wilbur: The LEEP produced the following three specimens: portions of exocervix and endocervix and an endocervical curettage. All the disease identified was in the exocervical specimen. One CIN 3 lesion (equivalent to severe dysplasia or carcinoma in situ) had an abnormality of maturation involving full or near full thickness (Figure 3A) and numerous mitoses in the upper levels of the epithelium. This process extended to the exocervical resection margin. In addition, multiple foci of endocervical adenocarcinoma in situ were present. In the lesion, pseudostratified epithelium abutted normal simple endocervical cells (Figure 3B). The initial endocervical curettage specimen did not show the pseudostratified epithelium with mitotic activity (Figure 3C). These changes are indicative of endocervical adenocarcinoma in situ, since no evidence of stromal invasion is noted in the specimen.

Figure 3. Specimen from a Loop Electrosurgical Excision Procedure.

In a lesion identified as grade 3 cervical intraepithelial neoplasia (Panel A, hematoxylin and eosin), an abnormality of maturation of squamous cells is noted to ascend to involve near full thickness. No koilocytosis is present in this lesion, since these cells are indicative of a neoplastic, not infectious, process, and viral particles are not produced. Endocervical adenocarcinoma in situ (Panel B, hematoxylin and eosin) is present adjacent to normal endocervical epithelium. The arrow points to the junction between adenocarcinoma in situ and normal-appearing cells. At higher magnification (Panel C, hematoxylin and eosin), the typical features of endocervical adenocarcinoma in situ are shown, including nuclear pseudostratification, enlargement, and hyperchromasia. Mitotic figures are prominent (arrows). A hyperchromatic, crowded group of cells from the patient’s original Papanicolaou test (Panel D, Papanicolaou stain) has features that may be seen in both high-grade squamous lesions and endocervical lesions. A vague columnar appearance is present at one margin (arrow), suggesting an origin from endocervical cells, although the rest of the group has a more syncytial appearance, a feature suggestive of high-grade squamous intraepithelial lesions.

In retrospect, the original cervical cytologic specimen showed very subtle changes suggestive of the presence of an endocervical lesion (Figure 3D). The Pap test was originally designed to detect the far more common squamous lesions, and its sensitivity for glandular lesions is generally thought to be less than 50%. Glandular lesions, which often reside high in the endocervical canal and deep in the endocervical crypts, may not produce cells for examination, and when they do, the features may be difficult to differentiate from squamous dysplasias and from reactive conditions.

Dr. Goodman: The rate of treatment failure after LEEP ranges from 1 to 25% depending on the size of the lesion, whether it is multifocal, and whether the cervical margins are negative. The majority of recurrences occur within 2 years. In general, for a patient with an excision with negative margins, Pap smear and colposcopy are performed at 6-month intervals; after two normal Pap smears, the patient can return to routine screening. The consensus guidelines state that yearly screening after two consecutive repeat cytologic examinations are negative must be continued for 20 years.¹⁶ In this patient, with positive exocervical resection margins and negative endocervical margins, I performed a repeat LEEP with endocervical curettage 3 months later; the specimen showed no evidence of dysplasia.

Dr. Marcela G. del Carmen: This patient had received one dose of quadrivalent HPV vaccine before the diagnosis of high-grade squamous intraepithelial lesions. Should she continue with the vaccination program? Two vaccines are currently available^25,26,27,28: a bivalent vaccine containing HPV-16 and HPV-18, which are associated with most cases of cervical cancer (approved in Europe and Australia and submitted to the Food and Drug Administration [FDA] for approval in the United States), and a quadrivalent vaccine containing HPV-6 and HPV-11 (associated with genital warts) and HPV-16 and HPV-18 (approved by the FDA for use in girls and women aged 9 to 26 years).²⁹ Both vaccines provided more than 90% protection against incident infection by the relevant HPV strains in clinical trials^30,31,32 and had more than 90% efficacy for prevention of CIN 2 or higher lesions in women who were HPV-negative at the end of the vaccination program.^33,34,35,36 The efficacy of the quadrivalent vaccine was assessed in subjects who were enrolled regardless of baseline HPV status. In the 21% of subjects who had evidence of infection with HPV-16 or HPV-18, the vaccine did not appear to alter the risk of the development of CIN.^34,35,36 Nonetheless, for this patient, the benefit of vaccination would include protection against the development of new HPV infection with other high-risk strains, as well as against genital warts.

Although HPV vaccination will be important in the prevention of cervical cancer, it will not replace the continued need to participate in other preventive strategies, the most important of which may be continued cytologic screening for cervical cancer. In this patient, completion of vaccination with the quadrivalent vaccine would be appropriate, despite the likelihood that she is already infected with at least one of the four HPV types in the vaccine.

Dr. Nancy Lee Harris (Pathology): Dr. Goldstein, can you tell us what has happened to this patient?

Dr. Goldstein: The patient completed the series of three HPV vaccinations. On repeat colposcopy 6 months after the second LEEP, there was no gross abnormality of the cervix. Pap smear and endocervical curettage disclosed atypical squamous cells; for this reason, HPV testing was performed, and it was negative for all the high-risk strains. She will follow up in 2 months in the gynecology clinic.

Dr. Goodman: It is reassuring that, even though this patient had a clinically significant cervical lesion that was due to HPV infection, she eventually cleared the HPV infection. As long as she is sexually active, she can develop a new HPV infection that is not covered by the vaccine, so annual Pap smears are still essential. We do not know why a high-grade lesion developed in this patient, which occurs in less than 1% of all adolescent girls and young women.

Anatomical Diagnoses

CIN 3 (severe dysplasia or carcinoma in situ) of the uterine cervix.

Endocervical adenocarcinoma in situ.

Dr. del Carmen reports receiving lecture fees from Ortho Biotech, and Dr. Wilbur receiving lecture fees from Becton Dickinson and grant support from TriPath Imaging and Cytyc. No other potential conflict of interest relevant to this article was reported.
Source Information

From the Divisions of Adolescent and Young Adult Medicine (M.A.G.) and Gynecologic Oncology (A.G., M.G.C.) and the Departments of Obstetrics, Gynecology, and Reproductive Biology (A.G., M.G.C.) and Pathology (D.C.W.), Massachusetts General Hospital; and the Departments of Pediatrics (M.A.G.), Obstetrics, Gynecology, and Reproductive Biology (A.G., M.G.C.), and Pathology (D.C.W.), Harvard Medical School.

References

1. Weinstock H, Berman S, Cates W Jr. Sexually transmitted diseases among American youth: incidence and prevalence estimates, 2000. Perspect Sex Reprod Health 2004;36:6-10. [CrossRef][ISI][Medline]
2. Ho GYF, Bierman R, Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998;338:423-428. [Free Full Text]
3. Moscicki AB, Hills N, Shiboski S, et al. Risks for incident human papillomavirus infection and low-grade squamous intraepithelial lesion development in young females. JAMA 2001;285:2995-3002. [Free Full Text]
4. Winer RL, Lee SK, Hughes JP, Adam DE, Kiviat NB, Koutsky LA. Genital human papillomavirus infection: incidence and risk factors in a cohort of female university students. Am J Epidemiol 2003;157:218-226. [Erratum, Am J Epidemiol 2003;157:858.] [Free Full Text]
5. Winer RL, Hughes JP, Feng Q, et al. Condom use and the risk of genital human papillomavirus infection in young women. N Engl J Med 2006;354:2645-2654. [Free Full Text]
6. Moscicki AB. HPV infections in adolescents. Dis Markers 2007;23:229-234. [ISI][Medline]
7. Moscicki AB, Schiffman M, Kjaer S, Villa LL. Updating the natural history of HPV and anogenital cancer. Vaccine 2006;24:Suppl 3:S3/42-S3/51.
8. Winer RL, Feng Q, Hughes JP, O’Reilly S, Kiviat NB, Koutsky LA. Risk of female human papillomavirus acquisition associated with first male sex partner. J Infect Dis 2008;197:279-282. [CrossRef][ISI][Medline]
9. Moscicki AB, Ellenberg JH, Farhat S, Xu J. Persistence of human papillomavirus infection in HIV-infected and -uninfected adolescent girls: risk factors and differences, by phylogenetic type. J Infect Dis 2004;190:37-45. [CrossRef][ISI][Medline]
10. Datta SD, Koutsky LA, Ratelle S, et al. Human papillomavirus infection and cervical cytology in women screened for cervical cancer in the United States, 2003-2005. Ann Intern Med 2008;148:493-500. [Free Full Text]
11. ACOG Committee on Practice Bulletins. ACOG practice bulletin: clinical management guidelines for obstetrician-gynecologists: number 45, August 2003: cervical cytology screening (replaces committee opinion 152, March 1995). Obstet Gynecol 2003;102:417-427. [CrossRef][Medline]
12. Saslow D, Runowicz CD, Solomon D, et al. American Cancer Society guideline for the early detection of cervical neoplasia and cancer. CA Cancer J Clin 2002;52:342-362. [Free Full Text]
13. Preventive Services Task Force. Screening for cervical cancer: recommendations and rationale. Rockville, MD: Agency for Healthcare Research and Quality, 2008.
14. Moscicki AB, Shiboski S, Hills NK, et al. Regression of low-grade squamous intra-epithelial lesions in young women. Lancet 2004;364:1678-1683. [CrossRef][ISI][Medline]
15. Wright JD, Davila RM, Pinto KR, et al. Cervical dysplasia in adolescents. Obstet Gynecol 2005;106:115-120. [ISI][Medline]
16. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D. 2006 Consensus guidelines for the management of women with abnormal cervical cancer screening tests. Am J Obstet Gynecol 2007;197:346-355. [CrossRef][ISI][Medline]
17. Huang YK, You SL, Yuan CC, et al. Long-term outcomes of high-risk human papillomavirus infection support a long interval of cervical cancer screening. Br J Cancer 2008;98:863-869. [CrossRef][ISI][Medline]
18. Winer RL, Kiviat NB, Hughes JP, et al. Development and duration of human papillomavirus lesions, after initial infection. J Infect Dis 2005;191:731-738. [CrossRef][ISI][Medline]
19. Insinga RP, Dasbach EJ, Elbasha EH, Liaw KL, Barr E. Progression and regression of incident cervical HPV 6, 11, 16 and 18 infections in young women. Infect Agent Cancer 2007;2:15-15. [CrossRef][Medline]
20. Rodriguez AC, Burk R, Herrero R, et al. The natural history of human papillomavirus infection and cervical intraepithelial neoplasia among young women in the Guanacaste cohort shortly after initiation of sexual life. Sex Transm Dis 2007;34:494-502. [ISI][Medline]
21. Moscicki AB, Ma Y, Wibbelsman C, et al. Risks for cervical intraepithelial neoplasia 3 among adolescents and young women with abnormal cytology. Obstet Gynecol 2008;112:1335-1342. [CrossRef][ISI][Medline]
22. Melnikow J, Nuovo J, Willan AR, Chan BK, Howell LP. Natural history of cervical squamous intraepithelial lesions: a meta-analysis. Obstet Gynecol 1998;92:727-735. [CrossRef][ISI][Medline]
23. Ostör AG. Natural history of cervical intraepithelial neoplasia: a critical review. Int J Gynecol Pathol 1993;12:186-192. [ISI][Medline]
24. MacLean AB. Acetowhite epithelium. Gynecol Oncol 2004;95:691-694. [CrossRef][ISI][Medline]
25. Lowy DR, Schiller JT. Prophylactic human papillomavirus vaccines. J Clin Invest 2006;116:1167-1173. [CrossRef][ISI][Medline]
26. Stanley M. Prevention strategies against the human papillomavirus: the effectiveness of vaccination. Gynecol Oncol 2007;107:Suppl 1:S19-S23. [CrossRef][ISI][Medline]
27. Stanley M, Lowy DR, Frazer I. Prophylactic HPV vaccines: underlying mechanisms. Vaccine 2006;24:Suppl 3:S3/106-S3/113.
28. Kirnbauer R, Booy F, Cheng N, Lowy DR, Schiller JT. Papillomavirus L1 major capsid protein self-assembles into virus-like particles that are highly immunogenic. Proc Natl Acad Sci U S A 1992;89:12180-12184. [Free Full Text]
29. Gardasil prescribing information. Whitehouse Station, NJ: Merck, 2007 (package insert).
30. Gall SATJ, Naud P. Substantial impact on precancerous lesions and HPV infections through 5.5 years in women vaccinated with the HPV-16/18 L1 VLP AS04 candidate vaccine. Presented at the Annual American Association for Cancer Research Meeting, Los Angeles, April 14–18, 2007.
31. Harper DM, Franco EL, Wheeler C, et al. Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial. Lancet 2004;364:1757-1765. [CrossRef][ISI][Medline]
32. Harper DM, Franco EL, Wheeler CM, et al. Sustained efficacy up to 4.5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from a randomised control trial. Lancet 2006;367:1247-1255. [CrossRef][ISI][Medline]
33. Paavonen J, Jenkins D, Bosch FX, et al. Efficacy of a prophylactic adjuvanted bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a phase III double-blind, randomised controlled trial. Lancet 2007;369:2161-2170. [Erratum, Lancet 2007;370:1414.] [CrossRef][ISI][Medline]
34. The FUTURE II Study Group. Quadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions. N Engl J Med 2007;356:1915-1927. [Free Full Text]
35. Ault KA, FUTURE II Study Group. Effect of prophylactic human papillomavirus L1 virus-like-particle vaccine on risk of cervical intraepithelial neoplasia grade 2, grade 3, and adenocarcinoma in situ: a combined analysis of four randomised clinical trials. Lancet 2007;369:1861-1868. [CrossRef][ISI][Medline]
36. Garland SM, Hernandez-Avila M, Wheeler CM, et al. Quadrivalent vaccine against human papillomavirus to prevent anogenital diseases. N Engl J Med 2007;356:1928-1943. [Free Full Text]

2. A Question Well Put

A Question Well Put

Brendan M. Reilly, M.D., Peter D. Hart, M.D., Susana Mascarell, M.D., and Hemant Chatrath, M.D.

In this Journal feature, information about a real patient is presented in stages (boldface type) to an expert clinician, who responds to the information, sharing his or her reasoning with the reader (regular type). The authors’ commentary follows.

A 51-year-old woman with a history of hypertension and depression reported progressively worsening pain in the left thigh over a period of several months, which had made her unable to walk for the past week. She also described generalized weakness and pains in her lower back, arms, and chest. She reported no weight loss, anorexia, trauma, or fever. She did not drink alcohol or smoke cigarettes. She reported having undergone hip surgery 6 months earlier to repair a “stress fracture” of her painful left leg.

Physical examination was unremarkable except for severe pain with any movement of the patient’s left thigh and tenderness to palpation of both upper arms, several ribs bilaterally, and her lower spine. Muscle strength and range of motion of the joints were otherwise normal.

Progressively increasing pain at a site of previous orthopedic surgery raises concern about infection or a mechanical complication of the procedure, including osteonecrosis of the femoral head or failure of the hardware. Radiographs of the site will address these concerns. More information is needed about the circumstances of the fracture itself. Stress fractures, which result from repetitive tensile or compressive stresses, may occur in the absence of any underlying bone pathology. However, I am concerned that this patient may have had a pathologic fracture, given her description of pain in other bony areas.

The usual causes of pathologic fractures are neoplastic and metabolic bone diseases. The absence of systemic symptoms and abnormal physical findings argues against a widely metastatic solid tumor, but hematologic malignant conditions, especially multiple myeloma, must be considered. Common metabolic bone diseases include osteoporosis, osteomalacia, Paget’s disease, and osteitis fibrosa cystica due to hyperparathyroidism. Although osteoporosis is the most prevalent of these disorders, diffuse bone pain would be inconsistent with this diagnosis, and osteoporosis severe enough to result in a femoral fracture would be highly unlikely in a woman of this age. It will be important to review her previous laboratory-test results and other medical records.

The patient had had normal menses until she underwent a hysterectomy without oophorectomy for uterine fibroids 4 years before presentation at our hospital. The patient’s hip surgery 6 months before presentation involved intramedullary nailing of her left femur for a nontraumatic fracture. An intraoperative biopsy specimen of the involved bone showed benign histologic findings. At that time, a bone survey revealed generalized osteopenia, confirmed by dual-energy x-ray absorptiometry (T scores, –2.1 in the neck of the left femur and –1.5 in the lumbar spine). Complete blood count, serum and urine protein electrophoresis, serum calcium and parathyroid hormone (PTH) levels, the serum creatinine level, and thyroid-function tests were normal. The serum 25-hydroxyvitamin D level was 16 ng per milliliter (normal range, 20 to 100).

Postoperatively, the patient was treated with 50,000 IU of oral ergocalciferol (vitamin D₂) weekly for 3 months and monthly thereafter, as well as physical therapy. She recovered from her surgery and was able to resume walking, but her pain continued, gradually worsening over the next several months.

The results of dual-energy x-ray absorptiometry confirm low bone density but cannot specify its cause. Laboratory-test results show no evidence of myeloma or hyperparathyroidism. Osteomalacia due to vitamin D deficiency is possible; the patient’s level of 25-hydroxyvitamin D is clearly below the desirable range (as distinct from the normal range) of more than 30 ng per milliliter. However, when bone disease becomes clinically evident in patients with osteomalacia due to vitamin D deficiency, the 25-hydroxyvitamin D level is usually lower (<10 ng per milliliter) than this patient’s level, and the PTH level is usually elevated. In addition, I would expect the patient to have improved with treatment if vitamin D deficiency were the sole cause of her disorder.

It will be important to inquire about any signs or symptoms of liver disease or malabsorption and to obtain radiographs of all painful areas. I would also review the histologic findings of the previous bone-biopsy specimen.

The patient reported no diarrhea, abdominal discomfort, or jaundice. Radiographs showed generalized osteopenia, multiple rib fractures, and transverse lytic lesions in the left femur and pubic ramus (Figure 1). A radionuclide bone scan showed increased uptake in all these areas. The patient was admitted to the hospital.

Figure 1. Radiograph of the Left Hip Joint.

Linear insufficiency (Looser) fractures are visible at the proximal portion of the left femur and the left inferior and superior pubic rami (arrows).

A complete blood count, urinalysis, serum and urine protein electrophoresis, and serum levels of electrolytes, calcium, PTH, and PTH-related protein were normal. The level of 25-hydroxyvitamin D was 54 ng per milliliter. The serum alkaline phosphatase level was 193 mg per deciliter (upper limit of normal, 110). Liver-function tests, including measurement of -glutamyltransferase levels, were normal.

Computed tomographic (CT) scans of the chest, abdomen, and pelvis and bilateral mammograms showed no evidence of neoplasm. Review of the histologic findings of the previous bone-biopsy specimen showed no pathologic process, but histomorphometry with tetracycline labeling was not performed.

There is no evidence of neoplasia. The previous bone biopsy, performed without fluorochrome labeling, is nondiagnostic, but the transverse lytic lesions in the femur and pubic ramus (Figure 1) strongly suggest Looser pseudofractures, features characteristic of osteomalacia. The most common causes of osteomalacia are phosphate depletion and disorders of vitamin D metabolism; thus, I would measure serum levels of phosphorus and 1,25-dihydroxyvitamin D. The elevated alkaline phosphatase level is nonspecific and is a finding consistent with bone fractures of many causes. However, this finding rules out hypophosphatasia (a rare heritable cause of osteomalacia that is associated with low serum alkaline phosphatase levels), which may present in adulthood with recurrent metatarsal stress fractures or symptomatic chondrocalcinosis.

The serum phosphorus level was 1.1 mg per deciliter (normal range, 2.5 to 4.5). The serum level of 1,25-dihydroxyvitamin D was 15 pg per milliliter (normal range, 15 to 60).

Hypophosphatemia, most often caused by renal phosphate wasting, is a potent stimulus of the production of 1,25-dihydroxyvitamin D. Thus, the low-normal serum level of 1,25-dihydroxyvitamin D is surprising. Combined with severe hypophosphatemia, it suggests rare disorders that cause both renal phosphate wasting and down-regulation of 25-hydroxyvitamin D-1-hydroxylase, resulting in osteomalacia. Most of these disorders, however, are hereditary (e.g., autosomal dominant hypophosphatemic rickets) and would be expected to present much earlier in life. An acquired disorder is much more likely.

I would review the patient’s family history. I would also review the patient’s previous medical records to determine whether her hypophosphatemia is long-standing or recent. In addition, I would measure the urinary excretion of phosphate.

The patient reported that she had 13 siblings and that neither they nor her parents have had known bone disease or unexplained fractures. Review of the patient’s laboratory studies revealed a serum phosphorus level of 1.0 mg per deciliter 6 months previously, but there were no measurements before that date. A 24-hour urine collection, obtained when the serum phosphorus level was 1.3 mg per deciliter, contained 972 mg of phosphate (54 mg per deciliter), 39 mg of calcium (2.6 mg per deciliter), and 1260 mg of creatinine (70 mg per deciliter). The serum creatinine level was 0.4 mg per deciliter.

In the presence of hypophosphatemia, these findings document renal phosphate wasting, defined as urinary excretion of more than 100 mg of phosphate per 24 hours or a fractional excretion of phosphate ([urine phosphatexplasma creatinine x100]÷[plasma phosphatexurine creatinine], all measured in milligrams per deciliter) greater than 5%. In the absence of persistent vitamin D deficiency or hyperparathyroidism, this patient’s fractional excretion of phosphate of 23% indicates a primary renal tubular disorder causing “phosphate diabetes.” The most common such disorder is Fanconi’s syndrome — renal phosphate wasting associated with glycosuria, aminoaciduria, hyperuricosuria, and type 2 renal tubular acidosis — but this is usually inherited. Acquired Fanconi’s syndrome in an adult would raise concern about the presence of myeloma, Sjögren’s syndrome, or the toxic effects of heavy metals or other toxins.

Serum levels of uric acid and bicarbonate were normal. Urinalysis did not show glycosuria.

There is no evidence of Fanconi’s syndrome. Thus, the patient’s hypophosphatemia is the result of an isolated defect in the reabsorption of renal phosphate. This finding, combined with an inappropriately low serum level of 1,25-dihydroxyvitamin D in an adult patient with generalized bone pain, osteopenia with pseudofractures, and a negative family history of bone disease or unexplained fractures, makes tumor-induced osteomalacia the only tenable diagnosis.

I would measure the serum level of fibroblast growth factor 23 (FGF-23), which is the phosphaturic hormone, or phosphatonin, most commonly associated with tumor-induced osteomalacia. In addition, I would review the previous CT scans. Occult mesenchymal tumors, often benign, are the usual cause of tumor-induced osteomalacia, but these tumors may be small and difficult to find.

A serum specimen was obtained for measurement of the FGF-23 level. Previous CT scans revealed multiple bone fractures of the ribs, pelvis, and spine, some with callus formation, but no apparent tumor.

I continue to suspect a mesenchymal tumor as the cause of the osteomalacia. Such tumors may be located in unusual sites such as the oral cavity or legs. Because many of these tumors have somatostatin receptors, an octreotide-labeled scintigram often can localize tumors that are not apparent on CT or magnetic resonance imaging scans. For the same reason, octreotide has been used as treatment (together with phosphorus supplementation and 1,25-dihydroxyvitamin D treatment) in cases in which the tumor cannot be localized or is malignant and metastatic to other sites.

If a tumor cannot be localized with the use of scintigraphy with labeled octreotide or other imaging, it is reasonable to consider performing genetic testing, despite the negative family history, to rule out a very rare, spontaneous mutation of the FGF-23 gene (defective in autosomal dominant hypophosphatemic rickets) or the PHEX gene (phosphate-regulating gene with homologies to endopeptidase on the X chromosome; defective in X-linked hypophosphatemic rickets).

The serum FGF-23 level was elevated, at 418 U (normal range [at the Mayo Clinic], 0 to 180). An octreotide scan (scintigraphy using octreotide labeled with indium-111) revealed intense focal uptake in the right posterior chest wall (Figure 2A). Another review of the chest CT revealed a round, well-circumscribed hyperdensity, 12 mm in diameter, in the posterior aspect of the right seventh rib (Figure 2B), which was previously attributed to callus formation at the site of a rib fracture.

Figure 2. An Octreotide Scan.

Scintigraphy with the use of octreotide labeled with indium-111 shows intense focal uptake in the right posterior chest wall at the 4-o’clock position (Panel A, arrow). A chest CT scan shows a round, well-circumscribed hyperdensity, 12 mm in diameter (Panel B, arrow) and an associated fracture of the posterior aspect of the right seventh rib.

Circulating levels of FGF-23, detectable in normal human serum, are not elevated in all patients with tumor-induced osteomalacia. Other phosphatonins have been implicated in some cases. In this patient, the elevated FGF-23 level and imaging studies strongly support the diagnosis of tumor-induced osteomalacia, but tissue confirmation is needed. The most common phosphatonin-secreting tumors are benign mixed connective-tissue types, such as ossifying fibromas and hemangiopericytomas, but in some cases the tumors are malignant. Surgical resection is the treatment of choice. The rib mass was resected surgically. Histologic studies revealed a hemangiopericytoma-like neoplasm, probably benign (Figure 3). Staining for FGF-23 was not performed, but postoperatively, the serum FGF-23 level fell to normal (80 U). Without further treatment, the serum phosphorus level and urinary excretion of phosphate also became normal (3.1 mg per deciliter and <5%, respectively).

Figure 3. Photomicrograph of Phosphaturic Mesenchymal Tumor (Hematoxylin and Eosin).

Histologic studies show a neoplasm, probably benign, with a hemangiopericytoma-like staghorn vascular pattern. It consists of epithelioid, hyperchromatic, and multinucleated cells with abundant cytoplasm (arrows).

The patient recovered uneventfully. She is walking normally and feeling well 1 year after surgery.

Commentary

In 1947, McCance described a patient whose vitamin D–resistant osteomalacia was cured by resection of a benign osteoid tumor of the femur.¹ Since then, remarkable progress has been made in understanding tumor-induced osteomalacia and the rare hereditary syndromes it resembles, including autosomal dominant hypophosphatemic rickets, X-linked hypophosphatemic rickets, and autosomal recessive hypophosphatemia.^2,3 In all these disorders, phosphatonins (phosphaturic hormones) inhibit both the renal tubular reabsorption of phosphate and the 1-hydroxylase enzyme that converts 25-hydroxyvitamin D to 1,25-dihydroxyvitamin D.^4,5 The net result is severe hypophosphatemia (caused by renal phosphate wasting without the compensatory increase in the intestinal absorption of phosphorus and calcium normally stimulated by 1,25-dihydroxyvitamin D) leading to osteomalacia and debilitating skeletal fractures. Patients with autosomal dominant hypophosphatemic rickets produce a mutant form of FGF-23 that resists normal (furin-mediated) degradation; the precise mechanisms underlying increased phosphatonin activity in X-linked hypophosphatemic rickets and autosomal recessive hypophosphatemia remain controversial.^3,6,7

In tumor-induced osteomalacia, unusual types of usually benign neoplasms secrete phosphatonins (FGF-23 and others), inducing a debilitating paraneoplastic syndrome with clinical effects that are typically worse than the neoplasm itself.^8,9,10 As in the present case, finding the causative tumor is often difficult but well worth the effort, because its removal cures the patient. Elucidation of these rare syndromes has stimulated research in phosphate homeostasis, which may have broad clinical implications.^11,12,13

Low bone density, as measured by dual-energy x-ray absorptiometry, is often viewed as diagnostic of osteoporosis, but it also occurs in cases of osteomalacia (because of reduced mineralization of the bone matrix). Thus, it is important to maintain a high index of suspicion for osteomalacia, the treatment of which is different from that of osteoporosis. Clinical presentations (e.g., severe, persistent pain or muscle weakness, tooth loss, or weight loss) or laboratory findings (e.g., abnormal serum levels of calcium, phosphorus, or alkaline phosphatase) that are inconsistent with a diagnosis of osteoporosis warrant further investigation. Bone radiographs often cannot distinguish osteomalacia from osteoporosis, but visualization of Looser pseudofractures, as in our patient, strongly suggests the diagnosis of osteomalacia. Bone-biopsy specimens with tetracycline labeling may be needed in some cases, but the clinical presentation and simple laboratory-test results usually are sufficient to make the diagnosis of osteomalacia and suggest its underlying cause.

Rare causes of osteomalacia include abnormal osteoid (e.g., osteogenesis imperfecta), impaired bioactivity of alkaline phosphatase (hypophosphatasia), inhibitors of calcification (e.g., aluminum), or an abnormal pH at the site of calcification (e.g., renal tubular acidosis). But in most cases, the cause of osteomalacia is either abnormal vitamin D metabolism or phosphate deficiency. Sorting this out, and identifying the specific disease responsible, requires an understanding of calcium and phosphate homeostasis in health and disease. For example, vitamin D deficiency causes osteomalacia primarily because of phosphate depletion that results from secondary hyperparathyroidism (and renal phosphate wasting) that was stimulated by impaired intestinal absorption of calcium. Thus, as the discussant noted, our patient’s normal serum PTH level was inconsistent with this hypothesis, despite documentation of a low serum level of 25-hydroxyvitamin D. In contrast, when hypophosphatemia unrelated to hyperparathyroidism causes osteomalacia, it typically stimulates a marked increase in production of 1,25-dihydroxyvitamin D to augment intestinal absorption of phosphate (and calcium). Our patient’s “inappropriately” normal serum level of 1,25-dihydroxyvitamin D in this context thus became the key to the puzzle. As the discussant knew, only phosphatonin-mediated disorders can explain this phenomenon.

But what if the discussant had not known about phosphatonins? More generally, how do clinicians compensate for imperfect knowledge when solving unusual clinical problems? Before consulting a colleague or the medical literature, experts take pains to “set” the problem or, according to Schon, “name the things to which we will attend and frame the context in which we will attend to them.”¹⁴ In this case, the discussant set the problem precisely: What causes adult-onset, vitamin D–resistant osteomalacia with renal phosphate wasting and normal serum levels of PTH and 1,25-dihydroxyvitamin D? Even if he had not known the answer, he knew how to put the question.¹⁵

No potential conflict of interest relevant to this article was reported.
Source Information

From the Department of Medicine, Weill Cornell Medical Center, New York (B.M.R.); and the Department of Medicine, Cook County (Stroger) Hospital (P.D.H., S.M., H.C.), and Rush Medical College (P.D.H., S.M.) — both in Chicago.

Address reprint requests to Dr. Reilly at the Department of Medicine, Weill Cornell Medical Center, 525 E. 68th St., Rm. M-522, Box 130, New York, NY 10065, or at brr2013@med.cornell.edu

References

1. McCance RA. Osteomalacia with Looser’s nodes (Milkman’s syndrome) due to a raised resistance to vitamin D acquired about the age 15 years. Q J Med 1947;16:33-46. [ISI]
2. Jan de Beur SM, Levine MA. Molecular pathogenesis of hypophosphatemic rickets. J Clin Endocrinol Metab 2002;87:2467-2473. [Free Full Text]
3. Lorenz-Depierieux B, Bastepe M, Benet-Pagès A, et al. DMP1 mutations in autosomal recessive hypophosphatemia implicate a bone matrix protein in the regulation of phosphate homeostasis. Nat Genet 2006;38:1248-1250. [CrossRef][ISI][Medline]
4. Kumar R. Tumor-induced osteomalacia and the regulation of phosphate homeostasis. Bone 2000;27:333-338. [CrossRef][Medline]
5. Schiavi SC, Moe OW. Phosphatonins: a new class of phosphate-regulating proteins. Curr Opin Nephrol Hypertens 2002;11:423-430. [CrossRef][ISI][Medline]
6. Jonsson KB, Zahradnik R, Larsson T, et al. Fibroblast growth factor 23 in oncogenic osteomalacia and X-linked hypophosphataemia. N Engl J Med 2003;348:1656-1663. [Free Full Text]
7. Shimada T, Muto T, Urakawa I, et al. Mutant FGF-23 responsible for autosomal dominant hypophosphataemic rickets is resistant to proteolytic cleavage and causes hypophosphataemia in vivo. Endocrinology 2002;143:3179-3182. [Abstract]
8. Cai Q, Hodgson SF, Kao PC, et al. Inhibition of renal phosphate transport by a tumor product in a patient with oncogenic osteomalacia. N Engl J Med 1994;330:1645-1649. [Free Full Text]
9. Shimada T, Mizutani S, Muto T, et al. Cloning and characterization of FGF23 as a causative factor of tumor-induced osteomalacia. Proc Natl Acad Sci U S A 2001;98:6500-6505. [Free Full Text]
10. Jan de Beur SM. Tumor induced osteomalacia. JAMA 2005;294:1260-1267. [Free Full Text]
11. Sommer S, Berndt T, Craig T, Kumar R. The phosphatonins and the regulation of phosphate transport and vitamin D metabolism. J Steroid Biochem Mol Biol 2007;103:497-503. [CrossRef][ISI][Medline]
12. Hsu C. FGF-23 and outcomes research — when physiology meets epidemiology. N Engl J Med 2008;359:640-642. [Free Full Text]
13. Levi M, Bruesegem S. Renal phosphate-transporter regulatory proteins and nephrolithiasis. N Engl J Med 2008;359:1171-1173. [Free Full Text]
14. Schon DA. The reflective practitioner: how professionals think in action. New York: Basic Books, 1983.
15. Dewey J. How we think. Boston: D.C. Heath, 1910.

1. A 47-Year-Old Man with Fever, Headache, Rash, and Vomiting

Case 11-2009 — A 47-Year-Old Man with Fever, Headache, Rash, and Vomiting

Sigall K. Bell, M.D., and Eric S. Rosenberg, M.D. Dr. Mary Berlik Rice (Medicine): A 47-year-old man was admitted to the hospital because of fever, headache, rash, and vomiting. The patient had been well until 8 days earlier, when severe pleuritic chest pain developed, worse on the right side than on the left, and a maculopapular rash appeared on his torso, which by the next day involved the scalp and the arms and legs, sparing the palms and soles. He also had temperatures of up to 39.1°C, chills, diaphoresis, a throbbing frontal headache that radiated to the vertex, a sore throat, swollen cervical lymph nodes, a cough productive of thick yellow sputum, and diffuse myalgias and arthralgias without joint swelling or erythema.

Five days before admission, the patient went to the emergency department of another hospital for these symptoms. Measurements of serum electrolytes and glucose and tests of renal function were normal; results of other laboratory tests are shown in Table 1. Chest radiographs and computed tomography (CT) of the chest and abdomen revealed multiple small nodules (the largest was 17 mm in diameter) in both adrenal glands; imaging revealed characteristics of an adenoma but was otherwise normal. Acetaminophen was prescribed, and he was sent home.

During the next 3 days, the symptoms did not improve, and the patient’s appetite decreased. Bleeding from the left nostril occurred twice, and he vomited once; his temperature rose to 38.9°C. Three days before admission, he was seen in the medical walk-in clinic of this hospital. On examination, he appeared uncomfortable. The temperature was 37.4°C, the blood pressure 112/75 mm Hg, and the pulse 93 beats per minute. Photophobia was present, with discomfort on upward gaze. The neck was supple. There were innumerable brown-gray macules on the trunk and face. The conjunctivae were slightly injected; the tonsils were diffusely red and enlarged, with a sparse white exudate. There was a tender, mobile, soft lymph node, approximately 4 cm in diameter, in the left submandibular area and smaller palpable anterior and posterior cervical and inguinal nodes bilaterally and in the right axilla. The chest was tender to palpation along the lateral ribs below the nipple line, more on the right side than on the left. The remainder of the examination was normal. The urine was positive for blood (4+) but was otherwise negative; results of other laboratory tests are shown in Table 1. Specimens of blood, urine, and sputum were cultured. Ketorolac was administered intramuscularly, and the patient was sent home with instructions to take doses of ibuprofen alternating with acetaminophen for fever and to follow up in 3 days, or sooner if the symptoms worsened.

Three days later, he came to the emergency department of this hospital because of persistent symptoms and increased nausea and vomiting. He reported photophobia, neck stiffness and pain, and some pressure on initiating urination. The pain in the chest and abdomen had resolved, and there was no diarrhea, shortness of breath, hematuria, pyuria, or dysuria; there also were no palpitations or changes to vision or hearing. The lesions on the chest and back had faded. A culture of the sputum from 3 days earlier grew abundant Streptococcus pneumoniae and Neisseria meningitidis, and cultures of the blood and urine were sterile. Nucleic acid testing of a urine specimen was negative for N. gonorrhoeae and Chlamydia trachomatis.

The patient had had syphilis at the age of 18 years, and he had lumbar disk disease with L5 disk herniation, for which he was receiving disability payments. Serologic tests for syphilis and the human immunodeficiency virus (HIV) had reportedly been negative 6 months earlier. He lived with a single male partner with whom he had been monogamous for the past 3 years. Four years earlier, his partner had received a diagnosis of HIV infection and did not take antiretroviral medications, reportedly because of a low viral load and normal CD4 count. The patient and his partner did not use condoms and reportedly did not practice anal sex. He owned cats, had been exposed to paint chips 2 weeks earlier when he scraped a room in his home, and had not traveled recently. His partner was not ill. The patient’s illness occurred during the winter, and he had not been exposed to mosquitoes or ticks. He smoked cigarettes and marijuana and had done so for many years; he drank alcohol rarely, after years of heavy use in the past, and he did not use intravenous drugs. Family members had had coronary artery disease, hypertension, diabetes mellitus, and polysubstance abuse; a sibling had died from cirrhosis related to alcohol and infection with hepatitis C virus. Medications included oxycodone as needed for pain, diazepam, acetaminophen, and ibuprofen. He had no allergies to medications.

On examination, the patient coughed frequently, the respirations were 28 breaths per minute, and the oxygen saturation was 98% while he was breathing ambient air. The temperature was 36.3°C; it rose to 38.3°C within 2 hours and was 39.2°C later in the day. Other vital signs were normal. There was bilateral posterior cervical lymphadenopathy; the oropharynx was erythematous, with no exudate or tonsillar enlargement. Neck flexion caused discomfort; range of motion was full. There was abdominal guarding. An erythematous, blanching rash covered the chest and upper back, with no papules or nodules. The remainder of the examination was normal.

Screening tests for influenza A and B antigens were negative. Measurements of serum electrolytes, total protein, and globulin and tests of liver and renal function were normal; other test results are shown in Table 1. An electrocardiogram was normal. An ultrasonogram of the abdomen revealed a region of hypoattenuation within the pancreatic head that was thought to represent a peripancreatic lymph node. A radiograph of the chest showed a perihilar linear opacity suggestive of mild subsegmental atelectasis. CT of the brain revealed mucosal thickening of the maxillary sinus. A lumbar puncture was performed; results of cerebrospinal fluid (CSF) analysis are shown in Table 2.

The patient was admitted to the hospital. Vancomycin, ceftriaxone, acyclovir, narcotic analgesia, sumatriptan, and acetaminophen were administered. Additional diagnostic tests were performed.

Differential Diagnosis

Dr. Sigall K. Bell: I am aware of the diagnosis in this case. Within a 24-hour period, 10 signs and symptoms developed in this previously healthy 47-year-old man: the classic mononucleosis-like triad (fever, sore throat, and lymphadenopathy), rash, chest pain, cough, myalgia, arthralgia, diaphoresis, and headache. The symptoms persisted for 8 days, and he ultimately presented with vomiting, neck stiffness, and evidence of meningitis on lumbar puncture.

The differential diagnosis includes a bacterial infection that spread to involve the meninges; infection with an organism such as an atypical bacterium, fungus, or parasite that would not be detected by routine cultures; a systemic viral illness progressing to aseptic meningitis; and a noninfectious process such as a rheumatologic disease mimicking an infectious process. The clinician’s priorities are to distinguish between bacterial and aseptic meningitis; to ascertain whether meningitis, encephalitis, or both were present; and to rapidly rule out life-threatening illnesses.

Bacterial Infection

The recovery of S. pneumoniae and N. meningitidis from the sputum, together with the finding of mucosal thickening on CT of the head, raises the possibility of a bacterial sinus infection progressing to meningitis. Meningococcemia can rapidly be fatal, necessitating prompt recognition and initiation of therapy. S. pneumoniae is a relatively common sinorespiratory pathogen, and meningitis could have developed by hematogenous spread or by direct extension from a sinus infection. The administration of antibiotics to address these organisms was appropriate.

There are several features of this case that weigh against a typical bacterial infection. Despite the cough, the physical examination and radiographic findings did not disclose changes consistent with pneumonia. Although thickened sinuses were seen on the CT scan of the head, there is insufficient evidence for a diagnosis of clinically active sinusitis. The CSF profile revealed the classic findings of aseptic meningitis: a mild, lymphocytic-predominant pleocytosis with elevated protein and normal glucose levels. A Gram’s stain of the CSF did not contain bacteria, and blood and CSF cultures, which were obtained before the initiation of antibiotic therapy, were negative. In routine clinical practice, physicians are often left to ponder whether a presentation consistent with aseptic meningitis is due to partially treated bacterial meningitis that results from antibiotics prescribed for antecedent symptoms. In this case, we can safely rule out classic bacterial meningitis. The sputum isolates may indicate a tracheobronchitis, but they are more likely to be colonizers in this longtime smoker.

Atypical Bacterial, Fungal, or Parasitic Infection

Are there organisms that would not grow in routine cultures (so-called culture-negative organisms) that could be associated with this patient’s systemic manifestations and aseptic meningitis? Like meningococcemia, rickettsial disease is considered a “fever and rash emergency,” and it is therefore important to rule it out rapidly. The abrupt onset of fever, headache, malaise, conjunctival injection, rash, myalgia, and arthralgia is a feature consistent with a diagnosis of rickettsial disease. However, the patient became ill in the winter and did not report any recent travel, so tickborne diseases, including Rocky Mountain spotted fever, are unlikely. The patient owned cats, and cervical lymphadenopathy was present, leading us to consider bartonella and toxoplasma infections. Bartonella can be manifested as encephalitis in the central nervous system, but this is uncommon in an immunocompetent host. The pace of the patient’s disease makes bartonella infection unlikely, and serologic testing for toxoplasmosis was negative.

Spirochetal infections, including secondary syphilis, Lyme disease, and leptospirosis, are important considerations. This patient had syphilis at 18 years of age, and his current illness could have been a result of a new exposure or relapse.¹ Meningeal involvement is common early in secondary syphilis, and fever, pharyngitis, headache, anorexia, lymphadenopathy, and macular or maculopapular rash over the trunk and extremities may occur (although the rash, like that of Rocky Mountain spotted fever, characteristically involves the palms and soles). After reaching an all-time low in 2000, the incidence of primary and secondary syphilis has increased dramatically in homosexual men in the United States.^2,3 Syphilis must remain at the forefront of the clinician’s considerations, but in this case the rapid plasma reagin test was nonreactive.

Viral Infection

This patient’s systemic symptoms and findings such as conjunctival injection, myalgias, arthralgias, rash, and gastrointestinal symptoms without peripheral leukocytosis suggest a viral infection. Many viruses are associated with myopericarditis, which could explain his chest pain. Enterovirus is the most common cause of aseptic meningitis, especially during the summer, although up to 10% of cases of aseptic meningitis may be attributable to enterovirus even during other seasons.⁴ Culture or polymerase-chain-reaction (PCR) testing of the CSF is often positive for enterovirus in these cases.⁵ West Nile virus can be associated with rash; this and other arboviruses would need to be considered if the patient had presented in the summer. Influenza, a common virus in the winter, was considered, but tests were negative. Infection with lymphocytic choriomeningitis virus, which is a rodentborne arenavirus, is also more common in the colder months, but no exposure to rodents was described.

Causes of fever and rash also include the classic childhood viral exanthems of parvovirus B19, measles, mumps, and rubella, although these are rare in adults. Parvovirus, the causative agent of fifth disease of childhood, causes fever, rash, and arthritis in adults; however, infection with parvovirus was ruled out by serologic evaluation. After remaining stable for many years, the incidence of mumps has increased recently in the United States, with a series of outbreaks in 2006 in several states, including Massachusetts.⁶ Before vaccination became routine, mumps caused 10 to 20% of cases of aseptic meningitis, which is one of the most common extrasalivary complications of mumps. Similarly, there was a sudden increase in the incidence of measles in the United States in 2008.⁷ There was no reported history of childhood measles or mumps in this patient, and we are unaware of his vaccination history, although evidence of previous vaccination or infection could be confirmed by serologic testing. A diagnosis of measles or mumps is unlikely, given the patient’s age, the character of his rash, and the absence of parotitis.

Infection with herpes simplex virus (HSV) or varicella–zoster virus can cause aseptic meningitis or encephalitis, fever, and rash, but it rarely causes disseminated disease in an immunocompetent host. However, it is critical to recognize these viruses, since effective treatment is available. The initial description of chest pain and truncal rash, especially if unilateral, raises concern for varicella–zoster virus, but the subsequent spread and character of the rash, which was nonvesicular, is not typical of the virus. Nucleic acid testing of the CSF for HSV DNA was negative, and we were not told of any change in mental status or brain function. Given these test results, acyclovir could safely be discontinued.

Mononucleosis-like Illness

The presence of posterior cervical lymphadenopathy, fever, pharyngitis, and rash is characteristic of a mononucleosis-like illness. Causes of mononucleosis-like illness include acute infection with Epstein–Barr virus (EBV), cytomegalovirus (CMV), and HIV. The use of ampicillin has a strong association with the development of a maculopapular rash in patients with acute EBV infection, but rash occurs in 10% of cases even without antibiotic use. EBV mononucleosis is typically associated with lymphocytosis or atypical lymphocytosis; neither was present in this case. However, these hematologic manifestations may be less pronounced in adults than in adolescents. The heterophile antibody test was negative, but the timing of the appearance of heterophile antibodies is variable, and a single negative test does not rule out a diagnosis of acute EBV infection. Since the patient had been symptomatic for 8 days, the test should have been repeated, or EBV-specific serologic testing could have been performed. However, acute EBV infection is unlikely in a 47-year-old man, since more than 90% of adults in the United States have serologic evidence of past EBV infection.⁸

Primary CMV infection is also unlikely, since 50 to 60% of adults in the United States have serologic evidence of previous CMV infection⁹; this percentage is increased among men who have sex with men. Primary CMV infection generally causes no or mild pharyngitis and mild lymphadenopathy and is almost universally associated with hepatitis, which was not seen in this case. It is unlikely that the patient had recently encountered CMV for the first time, but serologic testing for CMV should be performed to rule out a diagnosis of primary CMV infection. Aseptic meningitis is a rare complication of EBV or CMV mononucleosis.

Acute HIV Infection

The patient had a male partner who was HIV-positive. The patient was sexually active, did not use condoms, and was HIV-negative 6 months before presentation. Does his failure to acquire HIV infection despite repeated exposure over a period of several years suggest that he was somehow protected against infection? Homozygosity for the CCR5*32 mutation confers protection from infection, and this polymorphism is present in 1% of white persons.^10,11,12 Cohorts of highly exposed, persistently seronegative persons have been described in studies of commercial sex workers^13,14,15 and serodiscordant sexual partners.¹⁶ Although the mechanism of protection is not known, it is hypothesized that repeated exposure to the virus may augment host immune responses, similarly to a protective vaccine. If this patient was not resistant, his cumulative risk of HIV infection has become substantial by this time.

This patient’s presentation includes most of the symptoms and signs that have been described in the acute retroviral syndrome, including fever, headache, pharyngitis, nausea and vomiting, anorexia, myalgia and arthralgia, diaphoresis, and lymphadenopathy^17,18 (Table 3). Aseptic meningitis has been reported in 24% of cases of acute HIV infection,¹⁸ and myopericarditis has also been described.¹⁹ The presence of a maculopapular rash and oral or genital ulcers can help to distinguish this diagnosis from other mononucleosis-like viral illnesses, although these findings are not specific. Acute HIV infection accounts for about 1% of patients with a mononucleosis-like illness and a negative heterophile antibody test²⁰ and about 1% of patients with symptoms of a viral illness who present to an urban urgent care clinic.²¹ The diagnosis of acute HIV infection requires a high index of suspicion, and it is seldom considered on initial presentation, even in high-risk populations.^18,22 As in this case, it often takes several encounters with the medical system before the diagnosis is considered.

Noninfectious Diseases

Rheumatologic diseases such as systemic lupus erythematosus, Still’s disease, Kawasaki’s disease, Wegener’s granulomatosis, and sarcoidosis may mimic an infectious process. Epistaxis and hematuria in the absence of thrombocytopenia invite consideration of Wegener’s granulomatosis or sarcoidosis; however, the inactive urine sediment in this case makes vasculitic glomerulonephritis unlikely, and the chest imaging does not suggest sarcoidosis. Systemic lupus erythematosus is associated with central nervous system disease that typically occurs as psychosis or seizures and is more common in young women than in men.

Still’s disease affects both sexes equally and has a bimodal age distribution, of 15 to 25 years and 36 to 46 years of age, near the age of this patient at presentation. It is characterized by an evanescent salmon-colored rash involving the trunk and extremities that typically is accompanied by fever, myalgia, arthralgia, and pharyngitis, all of which were present in this patient. Pleurisy or pericarditis is not uncommon. However, as with lupus, the onset is typically less abrupt than in this case, and leukocytosis, hepatitis, and splenomegaly are common. Although there are reports of meningitis associated with Still’s disease,²³ it is exceedingly rare.

Summary

Taken together, the patient’s history of exposure to HIV, abrupt onset of illness, and constellation of symptoms makes acute HIV infection the most likely diagnosis. Acute HIV infection is characterized by high-level viral replication, a transient decline in the CD4+ T-cell count, and evolution of virus-specific antibodies. Occasionally, during the transient decline in the CD4+ T-cell count (and associated immunosuppression), patients present with a concurrent opportunistic infection. Therefore, when a patient such as this one presents with acute HIV infection, it is important to determine whether the illness is related to HIV alone or whether another pathogen may also be causing disease. In this patient, the presentation and findings are consistent with acute HIV infection alone. To make this diagnosis, a test of the HIV type 1 (HIV-1) viral load should be performed and interpreted in conjunction with HIV-1–antibody tests.

Dr. Eric S. Rosenberg: Dr. Peterson, would you tell us what your impressions were when you saw this patient?

Dr. Scott L. Peterson (Infectious Diseases): Those of us who saw this patient considered secondary syphilis and acute infection with toxoplasma, CMV, or HIV to be the most likely diagnoses. Since he had frequent sexual contact with an HIV-infected partner, we were most concerned that he had acute HIV infection presenting as a mononucleosis-like syndrome with meningitis.

Clinical Diagnosis

Acute HIV infection.

Dr. Sigall K. Bell’s Diagnosis

Acute HIV infection.

Pathological Discussion

Dr. Rosenberg: This patient had an extensive evaluation for common pathogens that cause mononucleosis-like syndromes. Serum tested for EBV-specific antibodies was negative for IgM and positive for IgG antibodies to the viral capsid antigen. Nucleic acid testing of plasma for the presence of EBV DNA was negative. These results indicate previous, but not acute, EBV infection. Similarly, testing for CMV IgM antibody was negative and for CMV IgG antibody was positive, indicating previous CMV exposure. Direct detection of CMV by means of antigenemia testing was negative, ruling out reactivation of latent virus.

Acute HIV-1 infection is characterized by a negative or weakly positive enzyme-linked immunosorbent assay (ELISA) for antibodies to HIV, a negative or indeterminate Western blot analysis for HIV-1, and high-level viremia detected by means of nucleic acid testing. This patient’s ELISA for HIV-1 and HIV-2 antibodies was weakly positive, and Western blot testing was negative for both HIV-1 and HIV-2. Quantitative testing for HIV-1 nucleic acids was positive at 45.7 million copies of HIV RNA per milliliter of plasma. Taken together, these results are diagnostic of acute HIV-1 infection. At the time of diagnosis, the patient’s absolute CD4+ T-cell count was 432 cells per cubic millimeter, a finding consistent with the transient decline in the CD4+ T-cell count that is characteristic of acute HIV-1 infection.^17,24,25 Although CSF was not tested for HIV, the meningitis was almost certainly secondary to acute infection with HIV-1.

Discussion of Management

Dr. Bell: The central management question is whether a patient such as this one should receive antiretroviral therapy during acute HIV-1 infection. This question has not been adequately addressed in the literature, and the best practice remains unknown. There are several reasons to consider initiating therapy during acute HIV-1 infection.²⁵ First, antiretroviral treatment has been shown to suppress viremia in more than 95% of patients²⁴ and thus may offer symptomatic relief to an ill patient with high-level viremia.²⁶ Treatment of this patient also could be considered on the basis of his neurologic involvement.²⁶ The most compelling reason to consider treatment is the potential for the preservation of HIV-specific T helper cells, an immune response that may be a critical component in the body’s long-term defense against replicating HIV.^26,27,28 It has been suggested that treatment that is initiated during acute infection and subsequently discontinued has the potential to improve control of viral replication and establish a lower viral load.²⁹ However, this effect may be transient.^30,31

There are currently no published data from randomized, controlled trials that have adequately compared treatment with no treatment during acute HIV-1 infection, and the long-term clinical benefits of early therapy are unknown. In addition, the introduction of antiretroviral medications during acute infection may result in a course of therapy many years longer than if treatment is delayed until the patient meets standard criteria for the initiation of treatment in chronically infected persons.³² Receipt of therapy for a longer time may increase the risks of medication-related adverse effects and viral resistance. In balancing these considerations, we offer early treatment to patients with acute HIV-1 infection who are ready, willing, and able to take medications consistently for many years.^25,32 Regardless of whether treatment is begun, making the diagnosis of acute infection may have important public health benefits by limiting the unknowing transmission of the virus from a person with the high level of viremia that is characteristic of acute HIV infection.^16,33,34,35

Dr. Rosenberg: Dr. Peterson, how did you treat this patient, and what is his current condition?

Dr. Peterson: When results of the CSF culture and HSV PCR assay were negative, antimicrobial therapy was discontinued. The patient’s condition slowly improved, and he was discharged 5 days after admission. HIV RNA was reported positive on hospital day 3, but the diagnosis of acute HIV infection was not definitively established until 2 days after discharge, when the HIV-1 Western blot test was negative. Five days after discharge, he continued to have disabling headaches. We decided to initiate antiretroviral therapy to help relieve his symptoms. A fixed-dose formulation of efavirenz, tenofovir, and emtricitabine was begun, with resolution of symptoms within a week. He had no side effects, and approximately 6 months later his viral load was less than 50 copies of HIV-1 RNA per milliliter of plasma, and his most recent CD4+ T-cell count was 819 cells per cubic millimeter.

Dr. Eugene P. Rhee (Medicine): Does the level of viremia affect the decision to treat?

Dr. Bell: Unless the level is unusually low, which suggests that the patient is spontaneously controlling HIV replication, the magnitude of viremia at peak should not in and of itself influence the decision to treat.

Dr. Viviany Taqueti (Medicine): Why do you think this patient did not become infected with HIV until after several years of exposure? Do you think he had a mechanism of resisting infection, or had he recently changed his behavior?

Dr. Bell: Since he eventually did become infected, he most likely did not have two copies of the mutant CCR5 (chemokine receptor 5) gene that is protective against infection with a CCR5-tropic virus, the most commonly transmitted form of HIV. We do not know whether he recently changed to higher-risk sexual behavior. One possible explanation is that his partner may have become more infectious, with a viral load that was transiently increased. Since viral transmission among men who have sex with men can occur during approximately 1 of every 100 sexual acts, depending on sexual practice,^36,37 the patient’s odds of eventually becoming infected were substantial.

Anatomical Diagnosis

Acute HIV-1 infection.

Dr. Rosenberg reports serving on the paid advisory board of Viral Genetics and serving as scientific advisor and having equity ownership in TBS Technologies. No other potential conflict of interest relevant to this article was reported.
Source Information

From the Division of Infectious Diseases, Beth Israel Deaconess Medical Center (S.K.B.); Infectious Disease Unit, Department of Pathology, Massachusetts General Hospital (E.S.R.); and the Departments of Medicine (S.K.B., E.S.R.) and Pathology (E.S.R.), Harvard Medical School.

References

1. Rolfs RT, Joesoef MR, Hendershot EF, et al. A randomized trial of enhanced therapy for early syphilis in patients with and without human immunodeficiency virus infection. N Engl J Med 1997;337:307-314. [Free Full Text]
2. Primary and secondary syphilis — United States, 2003-2004. MMWR Morb Mortal Wkly Rep 2006;55:269-273. [Medline]
3. Heffelfinger JD, Swint EB, Berman SM, Weinstock HS. Trends in primary and secondary syphilis among men who have sex with men in the United States. Am J Public Health 2007;97:1076-1083. [Free Full Text]
4. Elmore JG, Horwitz RI, Quagliarello VJ. Acute meningitis with a negative Gram’s stain: clinical and management outcomes in 171 episodes. Am J Med 1996;100:78-84. [CrossRef][ISI][Medline]
5. Lee BE, Davies HD. Aseptic meningitis. Curr Opin Infect Dis 2007;20:272-277. [ISI][Medline]
6. Dayan GH, Quinlisk MP, Parker AA, et al. Recent resurgence of mumps in the United States. N Engl J Med 2008;358:1580-1589. [Free Full Text]
7. Update: measles — United States, January-July 2008. MMWR Morb Mortal Wkly Rep 2008;57:893-896. [Medline]
8. Ascherio A, Munger KL, Lennette ET, et al. Epstein-Barr virus antibodies and risk of multiple sclerosis: a prospective study. JAMA 2001;286:3083-3088. [Free Full Text]
9. Staras SA, Dollard SC, Radford KW, Flanders WD, Pass RF, Cannon MJ. Seroprevalence of cytomegalovirus infection in the United States, 1988-1994. Clin Infect Dis 2006;43:1143-1151. [CrossRef][ISI][Medline]
10. Huang Y, Paxton WA, Wolinsky SM, et al. The role of a mutant CCR5 allele in HIV-1 transmission and disease progression. Nat Med 1996;2:1240-1243. [CrossRef][ISI][Medline]
11. Paxton WA, Martin SR, Tse D, et al. Relative resistance to HIV-1 infection of CD4 lymphocytes from persons who remain uninfected despite multiple high-risk sexual exposure. Nat Med 1996;2:412-417. [CrossRef][ISI][Medline]
12. Samson M, Libert F, Doranz BJ, et al. Resistance to HIV-1 infection in caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature 1996;382:722-725. [CrossRef][Medline]
13. Rowland-Jones SL, Nixon DF, Aldhous MC, et al. HIV-specific cytotoxic T-cell activity in an HIV-exposed but uninfected infant. Lancet 1993;341:860-861. [CrossRef][ISI][Medline]
14. Rowland-Jones SL, Dong T, Fowke KR, et al. Cytotoxic T cell responses to multiple conserved HIV epitopes in HIV-resistant prostitutes in Nairobi. J Clin Invest 1998;102:1758-1765. [ISI][Medline]
15. Rowland-Jones S, Pinheiro S, Kaul R. New insights into host factors in HIV-1 pathogenesis. Cell 2001;104:473-476. [CrossRef][ISI][Medline]
16. Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis 2005;191:1403-1409. [CrossRef][ISI][Medline]
17. Kahn JO, Walker BD. Acute human immunodeficiency virus type 1 infection. N Engl J Med 1998;339:33-39. [Free Full Text]
18. Schacker T, Collier AC, Hughes J, Shea T, Corey L. Clinical and epidemiologic features of primary HIV infection. Ann Intern Med 1996;125:257-264. [Erratum, Ann Intern Med 1997;126:174.] [Free Full Text]
19. Guillaume MP, Van Beers D, Delforge ML, Devriendt J, Cogan E. Primary human immunodeficiency virus infection presenting as myopericarditis and rhabdomyolysis. Clin Infect Dis 1995;21:451-452. [ISI][Medline]
20. Rosenberg ES, Caliendo AM, Walker BD. Acute HIV infection among patients tested for mononucleosis. N Engl J Med 1999;340:969-969. [Free Full Text]
21. Pincus JM, Crosby SS, Losina E, King ER, LaBelle C, Freedberg KA. Acute human immunodeficiency virus infection in patients presenting to an urban urgent care center. Clin Infect Dis 2003;37:1699-1704. [CrossRef][ISI][Medline]
22. Hightow LB, Miller WC, Leone PA, Wohl DA, Smurzynski M, Kaplan AH. Predictors of repeat testing and HIV seroconversion in a sexually transmitted disease clinic population. Sex Transm Dis 2004;31:455-459. [CrossRef][ISI][Medline]
23. Garrote FJ, Marco J, Obeso G, Rodriguez E, del Ser T. Aseptic meningitis and focal central nervous system involvement in a case of adult onset Still’s disease. J Rheumatol 1993;20:765-767. [ISI][Medline]
24. Kassutto S, Maghsoudi K, Johnston MN, et al. Longitudinal analysis of clinical markers following antiretroviral therapy initiated during acute or early HIV type 1 infection. Clin Infect Dis 2006;42:1024-1031. [CrossRef][ISI][Medline]
25. Kassutto S, Rosenberg ES. Primary HIV type 1 infection. Clin Infect Dis 2004;38:1447-1453. [CrossRef][Medline]
26. Gazzard BG. British HIV Association guidelines for the treatment of HIV-1-infected adults with antiretroviral therapy 2008. HIV Med 2008;9:563-608. [CrossRef][ISI][Medline]
27. Oxenius A, Price DA, Easterbrook PJ, et al. Early highly active antiretroviral therapy for acute HIV-1 infection preserves immune function of CD8+ and CD4+ T lymphocytes. Proc Natl Acad Sci U S A 2000;97:3382-3387. [Free Full Text]
28. Rosenberg ES, Billingsley JM, Caliendo AM, et al. Vigorous HIV-1-specific CD4+ T cell responses associated with control of viremia. Science 1997;278:1447-1450. [Free Full Text]
29. Rosenberg ES, Altfeld M, Poon SH, et al. Immune control of HIV-1 after early treatment of acute infection. Nature 2000;407:523-526. [CrossRef][Medline]
30. Kaufmann DE, Lichterfeld M, Altfeld M, et al. Limited durability of viral control following treated acute HIV infection. PLoS Med 2004;1:e36-e36. [CrossRef][Medline]
31. Streeck H, Jessen H, Alter G, et al. Immunological and virological impact of highly active antiretroviral therapy initiated during acute HIV-1 infection. J Infect Dis 2006;194:734-739. [CrossRef][ISI][Medline]
32. Hammer SM, Eron JJ Jr, Reiss P, et al. Antiretroviral treatment of adult HIV infection: 2008 recommendations of the International AIDS Society-USA panel. JAMA 2008;300:555-570. [Free Full Text]
33. Brenner BG, Roger M, Routy JP, et al. High rates of forward transmission events after acute/early HIV-1 infection. J Infect Dis 2007;195:951-959. [CrossRef][ISI][Medline]
34. Pilcher CD, Joaki G, Hoffman IF, et al. Amplified transmission of HIV-1: comparison of HIV-1 concentrations in semen and blood during acute and chronic infection. AIDS 2007;21:1723-1730. [CrossRef][ISI][Medline]
35. Pilcher CD, Tien HC, Eron JJ Jr, et al. Brief but efficient: acute HIV infection and the sexual transmission of HIV. J Infect Dis 2004;189:1785-1792. [CrossRef][ISI][Medline]
36. Wilson DP, Law MG, Grulich AE, Cooper DA, Kaldor JM. Relation between HIV viral load and infectiousness: a model-based analysis. Lancet 2008;372:314-320. [Erratum, Lancet 2008;372:1808.] [CrossRef][ISI][Medline]
37. DeGruttola V, Seage GR III, Mayer KH, Horsburgh CR Jr. Infectiousness of HIV between male homosexual partners. J Clin Epidemiol 1989;42:849-856. [CrossRef][ISI][Medline]