Azathioprine-Associated Pulmonary Dysfunction

Azathioprine-Associated Pulmonary Dysfunction

I V2 VJ 1/2 1/2 V4 Vs 1/2 It is possible but not certain that, with advancing severity of this disease, the intensity of the systolic murmur m...

864KB Sizes 22 Downloads 61 Views

I

V2

VJ

1/2

1/2

V4

Vs

1/2

It is possible but not certain that, with advancing severity of this disease, the intensity of the systolic murmur might vary from inspiratory decrease to inspiratory increase. However, long-term follow-up observations will be required to confirm the validity of such a hypothesis.

REFERENCES Lockhart A, Charpentier S, Bourdarias JP, Ismail MB, Ourbak P, Scebat L. Right ventricular involvement in obstructive cardiomyopathies; haemodynamic studies in 13 cases. Br Heart J 1966; 28:122

2 Morrow AG, Fisher RD. Fogarty TJ. Isolated hypertrophic obstruction to right ventricular outflow: clinical, hemodynamic, and angiographic findings befOre and after operative treatment. Am Heart J 1969; 77:814 3 Cardiel EA, Alonso M, DeJean JL, Menarguez L: Echocardiographic sign of right-sided hypertrophic obstructive cardiomyopathy. Br Heart J 1978; 40:1321 4 Kishimoto C, Kaburagi T, Takayam S, Kanyu I, Yokoyama S, lilkatsu Y: A case of hypertrophic obstructive cardiomyopathy with isolated obstruction to the right ventricular outflow: the echocardiographic findings. J Cardiography 1980; 10:881

Azathioprine-Associated Pulmonary Dysfunction* Michael]. Krowka, M.D.t; Richard I. Breuer, M.D.:!:; and

Thomas]. Kehoe, M.D.§ We present a case of azathioprine-associated alveolitis diagnosed by gallium-67 scanning and transbronchial biopsy. The patient denied respiratory symptoms, exhibited spiking fevers, and had normal chest roentgenograms. Allopurinol inhibition of azathioprine metabolism may have been a contributing factor.

P

ulmonary toxicity of immunosuppressive and cytotoxic agents has been recently reviewed.'·2 Respiratory dysfunction is usually characterized by dyspnea, cough, abnormal pulmonary function tests, and chest roentgenographic *From the Department of Medicine, Evanston Hospital, Northwestern University Medical School, Evanston, Illinois. tResident in Internal Medicine. :!:Associate Professor, Division of Gastroenterology. §Associate Attending, Division of Pulmonary Medicine.

-

Vs

FIGURE 4. Electrocardiogram showing right axis deviation and right ventricular hypertrophy.

changes in a febrile patient. Early detection of such toxicity, prior to chest roentgenographic abnormalities, has been described and may have a significant impact on morbidity and mortality. 3 We describe a patient with azathioprinerelated pulmonary dysfunction characterized by normal findings on chest roentgenogram and abnormal gallium-67(67Ga) lung scan.

CASE REPORT A 35-year-old man was admitted for evaluation of intermittent fever (39.4°C). Crohn's disease had been diagnosed when he was aged 17. Four bowel operations performed because of obstruction or fistulas resulted in an ileostomy in 1970. Subsequent medical therapy included a two-month trial of azulfidine because of radiologic demonstration of ileal recurrence two years prior to admission. Kidney stones were passed in 1971, 1972, and 1979. Stone analysis indicated 90 percent urate in each instance. Allopurinol therapy was begun two years prior to admission and continued at doses ranging from 100 to 200 m!¥day. Eight months prior to admission, ileal obstruction resolved with tube decompression and initiation of prednisone therapy, 60 mglday. The prednisone was decreased to 20 mglday, but four months later bleeding from the ileum recurred. Azathioprine, 50 m!¥day, along with prednisone, 40 m!¥day, successfully controlled the bleeding. By one month prior to admission, prednisone had been reduced to 40 mg every other day and azathioprine had been increased to 125 m!¥day. At no time did the patient complain of respiratory difficulties. Physical examination indicated blood pressure of 120/80 mm Hg without orthostatic changes. Temperature was 39.1°C orally, respirations 18 per minute, with a regular pulse rate of90 per minute. The chest was clear to auscultation and percussion. Cardiac examination was within normal limits. Abdominal examination was significant for mild, diffuse tenderness without rebound or hepatosplenomegaly. There was no adenopathy, and the neurologic examination was normal. Fever workup included cultures of induced sputum, urine, blood, spinal Ouid, and bone marrow. All were negative. Chest roentgenograms were normal. The WBC was elevated to 15,500/cu mm, with 65 percent polymorphonuclear leukocytes and no eosinophils. Atypical lymphocytes were not seen on the peripheral smear. Antinuclear antibody and rheumatoid factor titers were less than 1:20. Lymphocyte stimulation, immune complex, and complement studies were not done. Hepatorenal function was normal, and the urinalysis was unremarkable. Forty-eight hours after admission, azathioprine and allopurinol therapy was stopped, and prednisone administration continued at 40 mglday. A 07Ga scan obtained to locate a possible abdominal abscess was unremarkable. However, both lungs demonstrated spectacular uptake at 6, 24, and 48 hours after Azalhloprlne-8SSOCialed Pulmonary Dysfunction (Ktowlra et al)

volumes were unchanged but the Dco increased to 70 percent predicted. The patient remained afebrile without respiratory symptoms. Six weeks after discharge, repeated examinations showed improvement in Rows and volumes (averaging 10 to 15 percent), with the Dco 104 percent predicted. A 67Ga scan at this time showed no uptake in either lung. Prednisone therapy, 40 mg every other day, was continued. DISCUSSION

FICURE 1. 87Ga scanning shows diffuse uptake in both lungs 48 hours after injection. Mild uptake in the liver is normal. the radionucleotide injection (Fig 1). Arterial puncture for blood gas determination was refused. Pulmonary function tests performed five days after the medications were stopped showed a moderate restrictive defect with an abnormal single-breath diffusing capacity (FVC = 3.09 L, FEV, = 2.78 L, FEV,/FVC = 115 percent predicted, FEF 25-75%=4.17 US, VC=2.97 L [60 percent predicted], RV = 1.37 L (72 percent predicted], TLC = 4.34 L [66 percent predicted], and Dco=45 percent predicted). Bronchoscopic study with transbronchial biopsies indicated an alveolitis (Fig 2). Transbronchial specimens were negative for viral inclusions, acid-fast organisms, and fungal forms; stain for Pneumocystis carinii was negative. Sera for Mycoplasma and cytomegalovirus was not obtained. Three weeks after stopping the azathioprine and allopurinol, pulmonary function tests were repeated. Pulmonary flows and

FICURE 2. Transbronchial lung biopsy shows thickened alveolar septae with intra-alveolar infiltration by macrophages. Lymphocytes and macrophages predominate in the interstium. Fibrosis is not present. Vasculitis is not suggested. Atypical cellular morphology is not seen (hematoxylin and eosin, x 190).

Azathioprine-associated pulmonary toxicity appears to be a rare occurrence. In a review of the literature we found two case reports presumptive of azathioprine-induced lung injury characterized by fever, dyspnea, and diffuse pulmonary infiltrates. Rubin et al4 described a patient with ulcerative colitis who received cyclophosphamide two months prior to beginning a six-week course of azathioprine (100 mglday). Acute restrictive disease with decreased diffusion capacity was reported in a setting of normal arterial blood gases at rest. No tissue specimen was obtained, and the patient's symptoms and chest roentgenographic changes resolved over a three-month period, after cessation of administration of azathioprine and prednisone. Weisenberger' described a patient with nephrotic syndrome who had been taking azathioprine (150 mglday) for two years and suddenly developed pulmonary symptoms subsequent to the onset of rapidly progressive glomerulonephritis . Pulmonary function tests were not done, but arterial blood gas analyses indicated a mild hypoxemia with hypocarbia. Open lung biopsy was remarkable for thickened alveolar septae without evidence of alveolar bleeding or fibrosis. Symptoms and chest roentgenogram changes resolved over a two-week period, after drug withdrawal and administration of high-dose steroids and cephalothin. Neither patient had 67 Ga lung scanning as a diagnostic procedure. Our patient had a reversible alveolitis characterized by fever, a normal chest roentgenogram, and a diffusely abnormal 67 Ga scan of both lungs. The clinical significance of an abnormal 67 Ga scan in patients with a normal chest roentgenogram has been reported following drug treatment for hematologic malignancies, talc granulomatosis, P carinii infection, and after lymphangiography. MacMahon and Bekerman6 described six patients with hematologic malignancies who had received multiple cycles of combination chemotherapy (cyclophosphamide and vincristine common to all patients) who had diffuse, bilateral uptake of"7Ga in the lungs. Tissue biopsy, when obtained, was unremarkable for infection or fibrosis. These authors also reported two cases of talc granulomatosis in intravenous drug abusers and one case of mixed-cell, poorly differentiated lymphoma that presented with abnormal scans bilaterally. Turbiner et al7 reported a patient with well-differentiated diffuse lymphocytic lymphoma who, while receiving chemotherapy that included cyclophosphamide and vincristine, demonstrated abnormal 07Ga uptake diffusely and was noted to have P carinii on open lung biopsy. There are no reports of pulmonary 67 Ga uptake in patients receiving azathioprine as part of any chemotherapeutic regimen. Pulmonary dysfunction in our patient, who had no respiratory symptoms, was documented by abnormally low singlebreath diffusing capacities and restrictive lung volumes. These abnormalities reversed when azathioprine therapy was stopped. Diffusing capacity was the most significant CHEST I 83 I 4 I April, 1983

897

abnormal value initially and appeared to improve more rapidly than other pulmonary function parameters. Indeed, viral pneumonitis was considered in the differential diagnosis and could not be excluded as the etiologic agent with complete certainty. The mechanism of pulmonary dysfunction related to antimetabolites such as azathioprine may be a hypersensitivity reaction that evokes types 1 and 3 of the Gell and Coombs classification. • However, toxic drug levels must be considered as a possible means of alveolar injury. With reference to azathioprine metabolism, this agent is converted in the liver to 6-mercaptopurine, which is then metabolized to 6-thiouric acid by the enzymatic activity of xanthine oxidase. These products are then excreted in the urine. Our patient received allopurinol in low doses (100 to 200 mgfday) with relatively low-dose azathioprine (50 to 125 mgfday). The allopurinol inhibition of xanthine oxidase could have resulted in accumulated metabolites resulting in a hypersensitivity or toxic reaction to either azathioprine or 6-mercaptopurine. In addition, either process could have been attenuated by the concurrent administration of prednisone. The allopurinolazathioprine interaction has been previously described by Kaplan and Calabrese, 9 but without suggestion of pulmonary toxicity. Finally, it should be noted that the suspected drug interaction in our patient, by either mechanism described above, occurred in the setting of normal hematologic findings and normal bone marrow aspirate. The clinical implications of this case suggest that early detection of pulmonary dysfunction related to antimetabolite use can be accomplished with 67Ga lung scanning. To date there are no reports of false-positive, biopsy-proved cases of diffuse 67Ga lung uptake. 10 The subclinical documentation of probable azathioprine-related lung dysfunction, prior to chest roentgenogram changes, suggests that a greater spectrum of lung injury can occur and that injury, if detected quickly, could be reversed with the minimization of morbidity. Finally, the suggestion of drug toxicity, as opposed to hypersensitivity, emphasizes the need for quantitative measurements reflecting blood levels of commonly used cytotoxic agents. This may be especially important when toxicity cannot be detected by serial blood counts as in the REFERENCES present case. 1 Ginsberg S, Comis R. Pulmonary toxicity of antineoplastic agents. Semin Oncol1982; 9:34-51 2 Sustan H, Matthay R, Putman C. Cytotoxic drug-induced lung disease. Am J Med 1977; 62:608-15 3 Crystal R. Alveolitis [Editorial]. Thorax 1982; 37:1-10 4 Rubin G, Baume P, Vanderberg R. Azathioprine and acute restrictive lung disease. Aust NZ J Med 1972; 3:272-74 5 Weisenberger D. Interstitial pneumonitis associated with azathioprine therapy. Am J Clin Pathol1978; 69:181-85 6 MacMahon H, Bekerman C. Diagnostic significance of gallium lung uptake in patients with normal chest radiographs. Radiology 1978; 127:189-93 7 Turbiner R, et al. Abnormal gallium scanning in Pneumocystis carinii infection with a normal chest radiograph. Radiology 1978; 127:437-38 8 Weiss R. Hypersensitivity to cancer chemotherapy. Semin Oncol 1982; 9:5-13 9 Kaplan H, Calabrese P. Immunosuppressive agents. N Eng! J Med 1976; 298:1234-36 10 Ebright R, et al. The gallium scan. Arch Intern Med 1982; 142:244-53

698

Cholesteryl Ester Crystals in a Porcine Aortic Valvular Bioprosthesls Implanted for Eight Years* VICtor}. Ferrans, M.D., Ph.D.; Bruce McManus, M.D., Ph.D.; and William C. Roberts, M.D., F.C.C.P.

Masses of crystals, which were largely composed of cholesteryl esters, were found in a porcine aortic valvular bioprosthesis removed eight years after implantation in the mitral position in a patient with rheumatic mitral valvular stenosis. Histologic sections of grossly raised and nonraised yeUow lesions in the three cusps of this bioprosthesis revealed large clefts, which on frozen section contained lipid-positive, birefringent crystals. These crystals gave a positive reaction with the Schultz test for cholesterol. Biochemical analyses of isolated nodules revealed a cholesterol content of 40 nmole/mg of wet tissue. Of this cholesterol, 88 percent was esterified, and the remaining 1.2 percent was free cholesterol. These cholesterol deposits are most likely derived from blood lipids; however, they were not related to hyperlipidemia, since the patient had normal blood levels of cholesterol and triglycerides. lipid droplets are found relatively frequently in Smallconnective tissue of the leaflets of implanted cardiac valvular bioprostheses, as documented by previous electron microscopic observations from this laboratory; 1 however, crystalline deposits of cholesterol have not been reported previously in implanted bioprostheses. Such a finding forms the basis of the present report. CASE REPORT

A 76-year-old woman with a history of rheumatic mitral valvular stenosis, previous closed mitral commissurotomy (1966), and replacement of the mitral valve with a Hancock porcine aortic valvular bioprosthesis (1973) died in December 1981 in intractable chronic congestive heart failure. She had marked pulmonary arterial hypertension (100/10 mm Hg before mitral replacement and 75/15 mm Hg six months after operation) and elevation of left ventricular enddiastolic pressure (20 mm Hg in 1974); her systemic arterial blood pressure ranged up to 170190 mm Hg. The bioprosthetic heart valve was considered to have functioned normally until the patient's death. Blood lipids had been within normal limits; cholesterol was 189 ± 36 mgidl (mean± SD of nine values), and triglycerides were 107 mgldl (mean of three values). Anatomic Findings

At autopsy, the heart weighed 820 g. The coronary arteries were free of atherosclerotic plaques. Atherosclerotic changes in aorta, cerebral vessels, and peripheral systematic arteries were mild. The left atrium was markedly dilated and contained a small mural thrombus. Both ventricles were hypertrophied, and the left was dilated. Focal fibrous intimal thickening was present in the main left and right pulmonary arteries. The mitral prosthesis was wellsecured, without perivalvular leaks. The atrial and ventricular aspects of the prosthetic valve ring were covered by fibrous tissue; *From the Pathology Branch, National Heart, Lung, and Blood Institute National Institutes of Health, Bethesda, Maryland. Reprint ,;quests: Dr. Ferrans, National Institutes of Health, BldglO, Rm 7N208, Bethesda 20205

Cholesteryl Ester Crystals in Porcine Bioprosthesis (Ferrans, McManus, Roberts)