was due to pulmonary drug allergy. The symptoms were exacerbated by rechallenge in hospital, there was marked
eosinophilia in both blood and BAL fluid, and the lung biopsy specimen showed a focal process consistent with an allergic reaction. Finally, the chest roentgenogram returned to normal with drug withdrawal and corticosteroid therapy, and has not relapsed. The transfer factor was reduced at seven weeks, suggesting some residual interstitial reaction or damage to the capillary bed. Pulmonary eosinophilia has been a well-described side effect of many drugs, classically presenting 10 to 14 days after drug commencement, relapsing with rechallenge and resolving following withdrawal. u Sulfonamides have a strong association, 1·3 including reports in combination with pyrimethamine (Fansidar) ..., . Respiratory reactions to dapsone (a sulfone) are less common, 7 despite their widespread use at high dose in leprosy, with no reports of pulmonary eosinophilia. Until recently, pyrimethamine was believed to be an innocent bystander in Fansidar hypersensitivity. However, a recent report of noncardiogenic pulmonary edema with peripheral eosinophilia due to higher doses of pyrimethamine suggests otherwise.• In addition, pyrimethamine has been implicated in four patients with antimalarial pulmonary eosinophilia, three of whom took weekly maloprim/cbloroquine, and one who took pyrimethamine/cbloroquine.• As there are no reports (to our knowledge) of chloroquineinduced pulmonary eosinophilia, the maloprim was almost certainly to blame. The absence of specific reports of either component alone causing pulmonary eosinophilia suggests a synergistic effect. With the spread of acquired immunodeficiency syndrome (AIDS), dapsone and pyrimethamine are finding new roles in the treatment ofopportunistic infection-dapsone proving effective against Pneumocysffs carinii10 and pyrimethamine against toxoplasmosis. 11 Prescribers of pyrimethamine or dapsone in AIDS units (in addition to those using this combination for malaria prophylaxis) should be aware of this important side effect, which may mimic pulmonary infection as it did in our patient. REFERENCES
1 Feinmann L. Lung parenchymal changes due to ingested substances. Proc R Soc Med 1975; 68:440-41 2 Murphy RL, Phair JP. Systemic reaction to pyrimethamine and sulfadoxine. J Fam Pract 1986; 22:375-76 3 Leader. Sulphasalazine induced lung disease [editorial]. Lancet 1974; 2:504-05 4 Koch-\\\lser J, Hodel C, Leimer R, Style S. Adverse reactions to pyrimethamine/sulfadoxine. Lancet 1982; 2:1459 5 Svanbon M, Rombo L, Gustafl'son L. Unusual pulmonary reaction during short term prophylaxis with pyrimethamine sulfadoxine (Fansidar). BMJ 1984; 9.88:1876 6 McCormack D, Morgan WKC. Fansidar hypersensitivity pneumonitis. Br J Dis Chest 1987; 81:194-96 7 Grayson ML, Yung AP, Doherty RR. Severe dapsone syndrome due to weekly maloprim. Lancet 1988; 1:531 8 Pang JA. Non-cardiogenic pulmonary oedema associated with pyrimethamine. Respir Med 1989; 83:247-48 9 Davidson AC, Bateman C, Shovlin C, Marnnan M, Burton GH, Cameron IR. Pulmonary toxicity of malaria prophylaxis. BMJ 1988; 297:1240-41 10 Leoung GS, Mills J, Hopewell PC, Hughes W, Wofsy C.
308
Dapsone-trimethoprim for Pneumocy"'8 carlnti pneumonia in the acquired immunodeficiency syndrome. Ann Intern Med 1986; 105:45-8 11 Leport C, Raffi F, Matheson S, Katlama C, Regnier B, Sainot AG, et al. Treatment of central nervous system toxoplasmosis with pyrimethamine/sulfadiazine combination in 35 patients with the acquired immunodeficiency syndrome: efficiency of long-term continuous therapy. Am 1Med1988; 84:94-100
Disseminated Pneumocystosis Presenting as a Pleural Effusion* Robert L. Jayes, M.D.; Harry N. Kamerou\ M.D.; Susan M. HasselquiBt, M.D.; Morgan D. Delaney, M.D.; and
Daoid M. lbnmti, M.D.
Extrapulmonary pneumocystosis recently has been reported in a number of tissues. Most cases occurred in patients receiving aerosoli7.ed pentamidine prophylaxis. We report a case of disseminated pneumocystosis presenting as a large pleural effusion withoot apparent lung involvement where Pneurnocystil carinii was the only pathogen identi&ed. The absence of parenchymal lesions on chest x-ray film, the lack of hypoxemia and the minimal uptake of gallium all argue against significant lung involvement. The patient was successfully treated with chest tube drainage, intravenous and inhaled pentamidine and orally administered dapsone and trimethoprim. The addition of inhaled pentamidine to intravenously administered pentamidine may have increased pleural fluid levels substantially and its use coincided with the patient's improvement. (Chat 1993; 103:306-08)
P
neumocysffs carinii pneumonia is a common opportunistic infection in AIDS, but pleural effusions occur rarely and typically are small.1.1 Extrapulmonary pneumocystosis has been reported in a number of tissues including kidney, bone marrow, spleen, lymph node, and liver. 3 •4 Most cases occur in patients receiving aerosolized pentamidine prophylaxis3·4 and probably result from the poor systemic absorption. We report a case of disseminated pneumocystosis in such a patient who presented with a large pleural effusion without apparent lung involvement. We discuss the roles of intravenously administered and inhaled pentamidine in his successful treatment. CASE REPORT
A 27-year-old man was successfully treated 18 months earlier for
P carlnii pneumonia with dapsone and trimethoprim, since he was intolerant of trimethoprim-sulfamethoxizole. He began receiving zidovudine and inhaled pentamidine (300 mg/month). Eight weeks prior to admission, he developed cryptococcal meningitis which *From the Divisions of General Internal Medicine (Dr. Jayes), and InfecPulmonary Diseases (Ors. Hasselquist and Delaney~ tious Diseases (Dr. Parenti), Department of Medicine, and the Department of Pathology (Dr. Kamerow), the George Washington University Medical Center, Washington, DC. Ors. Jayes, Hasselquist, Delaney and Parenti currently are with Medical Faculty Associates, Washington, DC. Dr. Kamerow currently is at Centre Community Hospital, State College, Pa. Olssaminatad Pneumocyslollls '~.,al)
FIGURE 1. Chest x-ray film showing large free-Bowing right pleural elfusion with no apparent pulmonary parenchymal disease. responded rapidly to the experimental agent Schering 39304. A chest x-ray film showed a small free-Bowing right pleural elfusion and slight enlargement of the left hilum. He was brieBy hospitalized two weeks later with fever and pharyngitis. The CSF cryptococcal antigen titer had fallen, but the chest x-ray film showed a small free-Bowing right hydropneumothorax. Thoracentesis revealed a hazy, yellow Buid with a pH value of 7.41; protein value, 2.7 w'dl; LDH level, 406 U/L; glucose level, 124 mw'lOO ml; 512 RBCs/cu mm; and no WBCs. No organisms were seen on a Gram stain; no other stains were done. Cultures for bacteria, fungi, and mycobacteria were negative. His symptoms resolved without treatment. The hydropneumothorax was treated conservatively. One week prior to admission he complained of discomfort on the right chest wall. He was afebrile. A chest x-ray film showed a large free-Bowing right pleural elfusion with no apparent pulmonary parenchymal disease (Fig 1). Thoracentesis yielded 200 ml of bloody Buid with a pH value of 7.42; LDH 725 U/L; glucose level, 99 mw'dl; protein value, 1.9 w'dl; 370,000 RBCs/cu mm; and 136 WBC/ cu mm (S 12, L 4, M 76). Diff-Quick (Dade Diagnostics) and Papanicolaou stains showed erythrocytes, lymphocytes, foamy histiocytes and rare mesothelial cells. In addition, Diff-Quick-stained
~
....
...... FIGURE 2. Pleural Buid (100 x) stained with Diff-Quick method (Dade Diagnostics) showing erythrocytes, rare mesothelial cells, lymphocytes and histiocytes. Many P carinii trophozoites are clustered in the center of the figure and within histiocytes.
smears revealed innumerable P carinil trophozoites (Fig 2). A Gomori methenamine silver stain revealed cyst forms of P carinU. The trophozoites did not stain. Cultures for bacteria, fungi and mycobacteria were negative. Therapy with dapsone (100 mw'day) and trimethoprim (20 mg/kw'day) was started . . Four days later, the elfusion was substantially larger as evidenced on a chest x-ray film and he was hospitalized. Room air oxygen saturation (pulse oximetry) was 99 percent. On hospital day 1, a chest tube was inserted, 3,000 ml of bloody Buid was removed from the right side of the chest, and intravenously administered pentamidine (4 mw'kw'day) was started. A gallium scan showed minimal uptake of tracer in the lungs. Chest and abdominal CT scans showed hilar lympbadenopathy and hypodense splenic lesions. Pleurodesis with tetracycline was unsuccessfully attempted on day 2. The chest tube continued to drain 400 mVday with a persistent air leak. Pneumocystis carinil organisms were seen in the pleural Buid as late as ten days after starting intravenously administered pentamidine. We considered intrapleural instillation of pentamidine, but there has been no experience with this approach and we feared that this would induce serious arrhythmias.• After 14 days of intravenous therapy, pentamidine levels later measured by high pressure liquid chromatography (HPLC) (E. M. Bernard, Memorial Sloan-Kettering Cancer Center) were 64 ow'ml in the serum (trough) and 24 ow'ml (38 percent of serum level) in a simultaneous sample of pleural Buid. Dapsone (100 mw'day) was added to the regimen on day 14 and inhaled pentamidine (600 mg/day) on day 15. The patient's serum albumin level fell to 1.2 mw'lOO ml and peripheral hyperalimentation with albumin supplementation was begun. After the addition of inhaled pentamidine, random pleural Buid pentamidine levels were as high as 83 nw'ml. No further serum level.s were obtained. On day 20, no P carinii organisms were seen in the pleural Buid. Therapy with intravenous and inhaled pentamidine was discontinued after 23 days of intravenous and 10 days of inhaled therapy once the chest tube output ceased. He was discharged on a regimen of dapsone and trimethoprim. An abdominal CT scan three months later showed persistence of the splenic lesions. A chest x-ray film six months later showed only a persistently blunted right costophrenic angle. The patient survived for another year on a regimen of dapsone therapy without evidence of recurrent P carinU pneumonia. DISCUSSION
This patient developed a massive pleural effi.ision in which P carlnii was the only pathogen identified, with no apparent lung involvement. His hilar lymphadenopathy and splenic
lesions were consistent with disseminated pneumocystosis. •.• Pleural effi.isions are uncommon in pulmonary pneumocystosis and are generally small. •·u Nodular pleural lesions have been reported at lung biopsy and autopsy and may be associated with pulmonary vasculitis or disseminated pneumocystosis,3•4 but extensive involvement of the pleura without ·apparent lung infection has not been described. Pulmonary pneumocystosis in this patient carinot be completely excluded, since neither bronchoalveolar lavage nor lung biopsy was done. Indeed, subclinical Pneumocystis pneumonia has been reported at lung biopsy• and autopsy, u and gallium scanning may be less sensitive in patients receiving inhaled pentamidine. 10 However, the absence of parenchymal lesions on a chest x-ray film and the er and gallium scans, and the lack ofhypoxemia all argue against significant lung involvement. This patient's improvement followed the resumption of dapsone therapy and the addition of inhaled pentamidine. Although dapsone therapy failed initially, its combination with intravenous and inhaled pentamidine therapy may have CHEST I 103 I 1 I JANUARY, 1993
307
been helpful. However, it is more likely that the addition of inhaled pentamidine produced effective pentamidine levels in the pleural ftuid. Comparison of the pleural ftuid levels before and after the addition of inhaled pentamidine support this view. After ten days of intravenously administered pentamidine, the trough serum level-which should have reached a steady state-was typical of levels reported after an intravenous 4 mg/kg dose, 11 but the simultaneous pleural ftuid level was only 40 percent of this serum level. Higher pleural ftuid levels were measured after the addition of inhaled pentamidine. Inhaled pentamidine (300 mg) only slightly increases serum levels (about 14 nw'ml), 19 but it produces levels as high as 700 nw'ml in bronchoalveolar lavage ftuid. 13 Inhaled pentamidine may have increased the pleural ftuid level by transfer across the inftamed pleural membrane or through the air leak of the hydropneumothorax. Daily inhaled pentamidine may be useful for other patients with resistant pleural disease. REFERENCES 1 Suster B, Akerman M, Orenstein M, Wax MR. Pulmonary mamfestations of AIDS: review of 106 episodes. Radiology 1986; 161:87-93 2 Cbafrey MH, Klein JS, Gamsu G, Blanc P, Golden JA. Radiographic distribution of PneumocysffB carinU pneumonia in patients with AIDS treated with prophylactic inhaled pentamidine. Radiology 1990; 175:715-19 3 Telzak EE, Cote RJ, Gold Jw. Campbell SW, Armstrong D. Extrapulmonary PneumocyBffB carinU infections. Bev Infect Dis 1990; 12:380-86 4 Northfelt OW, Clement MJ, Safrin S. Extrapulmonary pneumocystosis: clinical features in human immunodeficiency virus infection. Medicine 1990; 69:392-98 5 Wharton JM, Demopulos PA, Goldschlager N. Tursades de pointes during administration ofpentamidine isothionate. Am J Med 1987; 83:571-76 6 Balacbandran I, Jones DB, Humphrey OM. A case of pneumocystis carinii in pleural ftuid with cytologic, histologic, and ultrastructural documentation. Acta Cytologica 1990; 34:48690. 7 Dyner TS, Lang W, Busch OF, Gordon PR. Intravascular and pleural involvement by PneumocyBffB carinU in a patient with acquired immunodeficiency syndrome. Ann Intern Med 1989; 111:94-95 8 Davey RT, Margolis D, Kleiner D, Deyton L, Travis W, et al. Digital necrosis and disseminated Pneumocystis carinU infection after aerosolized pentamidine prophylaxis. Ann Intern Med 1989; 111:681-82 9 Rao NA, Zimmerman PL, Boyer D, Biswas J, Causey D, Beniz J, et al. A clinical, histopathologic, and electron microscopic study of PneumocysffB carinU choroiditis. Am J Ophthalmol 1989; 107:218-28 10 Jules-Elysee KM, Stover DE, Zaman MB, Bernard EM, White DA. Aerosolized pentamidine: effect on diagnosis and presentation of Pneumocystis carinU pneumonia. Ann Intern Med 1990; 112:750-57 11 Conte JE, Chernoff D, Feigal OW, Joseph P, McDonald C, Golden JA. Intravenous or inhaled pentamidine for treating .Rmumocystis carinU pneumonia in AIDS: a randomized trial. Ann Intern Med 1990; 113:203-09 12 Conte JE, Golden JA. Concentrations of aerosolized pentamidine in bronchoalveolar lavage, systemic absorption, and excretion. Antimic Agents Chemother 1988; 32:1490-93 13 Montgomery AB, Debs RJ, Luce JM, Corkery KJ, Turner J, Brunette EN, et al. Selective delivery of pentamidine to the lung by aerosol. Am Bev Respir Dis 1988; 137:477-78
308
H-Type Tracheoesophageal Fistula and Congenital Esophageal Stenosls* Douglas N. Homnick, M.P.H., M.D., F.C.C.P.t
An 18-month-old boy was seen in the pediatric pulmonary clinic with a history of wheezing, stridor and intolerance to solid foods. Barium esophagram revealed distal esophageal stenosis and subsequently an H-type TEF at surgery. Following the surgery to repair both lesions the child (Chat 1993; 103:308-09) continues to do well. GT= gastrostomy tube; TEF = tracheoesophageal fistula; UGI= upper gastrointestinal
H
-type TEF without esophageal atresia is rare, comprising 4.2 percent of 1,058 cases ofTEF in a survey by the American Academy of Pediatrics.' H-type TEF occurring with esophageal stenosis is much rarer, with very few cases reported in the literature.1.3 Diagnosis may be difficult. The case described herein is of a young patient with this combination of anomalies. CASE REPORT
An 18-month-old boy was referred to the Pediatric Pulmonary Clinic because of cholcing spells associated with stridor occurring from 1 to 2 times a week to 10 to 12 times per day. The symptoms had occurred since near birth, with two episodes of mild bronchitis and wheezing occurring at 3 and Sl/2 months old, respectively. The patient had never been able to tolerate solid foods, regurgitating most, since their introduction at about 4 months of age. He had always been able to take liquid feedings well without regurgitation, cholcing or other respiratory symptoms. At 10 months of age, the child had undergone a UGI series with a small bowel study; reported results were normal. Examination in the Pulmonary Clinic showed no stridor or other abnormal respiratory signs. Height and weight were found to be less than the fifth percentile for age. Shortly after the Pulmonary Clinic visit, a barium esophagram revealed marked narrowing of the esophageal lumen at the junction of the proximal two thirds and distal third. A chest x-ray film done at that time was reported as normal. The child was referred to pediatric surgery with a presumptive diagnosis of distal esophageal stenosis. A retrospective review of the UGI done at 10 months of age revealed an incorrect diagnosis with the stenosis evident at that time. The child underwent esophagoscopy and attempted dilatation which revealed noninftamed mucosa at the site of the stenosis. A repeat esophagram showed the esophageal stenosis and no gastroesophageal reftux (Fig 1). A GT was placed for retrograde dilatation. Prior to the dilatation, but after GT placement, the child experienced an acute wheezing episode for which he was hospitalized. The history, prior to admission, revealed an upper respiratory infection and also feeding of crackers. A chest x-ray film, lateral decubitus and inspiratory/expiratory chest x-ray films revealed no infiltrate or differential hyperinftation to suggest foreign body aspiration. The episode resolved in a few days with therapy with aerosolized and intravenous bronchodilators. An attempt at retrograde dilatation was undertaken in two *From Michigan State University/Kalamazoo Center for Medical Studies, KaLimazoo, Mich. t Assistant Professor of Pediatrics and Director of Pediatric Pulmonary Specialty Clinics. H-Type Tracl-.opllllg88 Fistula (Doug/as N. Homnlck}