Pneumonitis Complicating Low-dose Methotrexate Therapy for Rheumatoid Arthritis

patients who did not respond to nitroprusside treatment for reducing pulmonary hypertension prior to cardiac transplantation. These patients were given a bolus of amrinone (0.75 mglkg) followed by a prolonged continuous infusion (25 J.Lg/k!Y'min) for 48 h; 89 percent of these patients responded to the therapy evidenced by a decrease in PCWP, right atrial pressure, systemic vascular resistance (SVR), and PVR." In our two documented patients who had severe pulmonary hypertension and biventricular failure, the administration of amrinone brought about the dramatic change in the pulmonary indices. The PAP decreased by 20 percent and 23 percent while the PVR decreased by about 40 percent and 67 percent, respectively. We are aware that in case 2, because of the initial large V wave, the mean height of the A wave or the diastolic pressure of the PCWP should be used to calculate the PVR. Unfortunately, PCWP was not recorded; therefore, pulmonary arterial diastolic pressure was used in the PVR calculation. This probably underestimates the absolute severity of PVR, but the relative change in PVR remains valid. The resultant lower pulmonary resistance and pressure improved the right ventricular performance and reduced the wall stress of the right side of the heart, thus improving the left ventricular function and systemic BP. In conclusion, we demonstrated that the addition of amrinone can be life-saving in patients with severe pulmonary hypertension and biventricular failure following complicated valvular heart surgery, not responding to conventional (3-adrenergic (isoproterenol or dopamine) and vasodilator (nitroglycerin) drug therapy. A loading dose of amrinone, 1to 1.5 mglkg, followed by 25J.Lg/k!Y'min infusion increased the mean BP by 49 to 51 percent, CI by 36 to 37 percent, and decreased mean PAP by 20 to 23 percent, PVR by 40 to 67 percent, and HR by 13 to 22 percent in the moribund patients. REFERENCES 1 Levy JH , Bailey JM . Amrinone: pharmacokinetics and pharmacodynamics. J Cardiothorac Anesth 1989; 3(suppl2):10-4 2 Coenen M, Pedemonte 0 , Baele P, Col J. Amrinone in the management of low cardiac output after open heart surgery. Am J Cardiol 1985; 56:33-38B 3 Oslen KH , Kluger J, Fieldman A. Combination high dose amrinone and dopamine in the management of moribund cardiogenic shock after open heart surgery. Chest 1988; 94:503-06 4 Uretsky BF, Lawless CE , Verbalis JG, Valdes AM, Kolesar JA, Reddy PS . Improved hemodynamics and increased activation of the renin-angiotensin system with combined intravenous dobutamine and amrinone in patients with severe heart failure. Chest 1987; 92:657-62 5 Guinmond JG, Matuschak GM , Meyers F, Keating D. Augmentation of cardiac function in end stage heart failure and by combined use of dobutamine and amrinone. Chest 1986; 90:302-04 6 Hess W, Arnold B, Veit S. The haemodynamic effects of amrinone in patients with mitral stenosis and pulmonary hypertension. Eur Heart J 1986; 7:800-07 7 Ramsay JG, DeJesus JM, Wynands JE, Railey FE , O'Connor JP, Robbins GR, et al . Amrinone before termination of cardiopulmonary bypass: haemadynamic variables and oxygen utilization in the postbypass period. Can J Anaesth 1992; 39:342-48 8 Bolling SF, Deeb GM , Crowley DC, Badellino MM , Bove EL. Prolonged amrinone therapy prior to orthotropic cardiac transplantation in patients with pulmonary hypertension. Transplant Proc 1988; 20:753-56

1620

Pneumonitis Complicating Low-dose Methotrexate Therapy for Rheumatoid Arthritis* Discrepancies Between Lung Biopsy and Bronchoalveolar Lavage Findings Dimitri Leduc, M.D.; lbul De \ilyst, Ph .D .; Philippe Ueureux , M .D. ; Pierre-Aloin Gevenois, M.D .; DanieUe ]acobovitz, M .D. ; and lean-Claude Yemault, Ph.D ., F.C.C .P.

Two very similar cases of drug-induced pneumonitis com-

plicating treatment of rheumatoid arthritis with low-dose methotrexate are presented. Diagnosis was suggested by clinical history and findings, but the broochoalveolar lavage showed a high percentage of oeutrophils, an unusual feature in methotrexate-induced pneumonitis. Traosbroochialluog biopsies (TBB) confirmed the diagnosis by showing interstitial lymphocytic infiltrate with micrograoulomas. Although histologic findings are not strictly pathognomonic, when a differential diagnosis has to be made with infectious and rheumatoid lung disease, TBB appears to be of great promise. (Chnt 1993; 104:1620-23) BAL = bronchoalveolar lavage; RA =rheumatoid arthritis; TBB transbroocbiallung biopsy

=

low-dose methotrexate sodium administration W eekly (5 to 15 mg) is used in the treatment of rheumatoid

arthritis (RA) refractory to conventional antirheumatic drugs. 1 The dosages used are much lower than those used in the treatment of malignancies so that severe toxic reactions are infrequently reported.' Methotrexate-related interstitial pneumonitis, however, has been described. w We report two cases of methotrexate-induced pneumonitis after low-dose therapy for RA in which transbronchial lung biopsy (I'BB) was highly contributive to the diagnosis. CASE REPORTS CASE

1

This 68-yeaN>Id woman was a nonsmoker and had no history of respiratory disease. In 1967, she developed severe RA which required treatment with salazopyrine, salicylates, and corticosteroids. In July 1991, treatment with salicylates and corticosteroids was stopped because of slight impairment of renal function and osteoporosis. Starting on July 15, 1991, methotrexate sodium (10 mglwk) was administered orally with improvement of articular complaints. She received the last dose the day before her admission. From the end of September 1991, she recurrently complained of cough and fever, and these complaints progressively increased until admission on October 31 , 1991. She had no articular symptoms. Examination revealed an ill-appearing woman with severe respiratory distress and fever. Dry inspiratory crackles were heard in both lungs. A chest radiograph (Fig 1, top) and computed tomography scan (Fig 1, bottom) showed massive interstitial infiltration of both lungs. Lung function tests showed a severe restrictive ventilatory impairment (Table 1). While she was breathing room air, the Pa01 value was 37.2 mm Hg; PaC01 , 20 mm Hg, and arterial pH, 7.50. The white blood cell count was 4,000/mm•, with 81 percent neutrophils, *From Erasme University Hospital, Brussels, Belgium . Reprint requests: Dr. Leduc, Chest Department, Erasme Vnlvemty Hospital, 1070 Brussels. Belgium Pneumonitis Complicating Low-dose Me1hot!axate Therapy (Leduc et el)

Table 1-Lung Function Teat Re8ult.

Case

Predicted Values

Vital capacity, L Total lung capacity, L FEV,L CO diffusing capacity, ml!min/mm Hg

2.26 4.19 1.60 17.20

Vital capacity, L Total lung capacity, L FEV,,L CO diffusing capacity, ml!min/mm Hg

2.90 5.40 2.29 22.70

Before Methotrexate Treahnent 1.95 3.40 1.75

At admission

Two Months After Admission

0.59 3.02 0.33 4.00

1.76 4.21 1.40 8.20

1.60 3.80 1.05 6.20

2.12 4.25 1.39 24.00

2

11 percent lymphocytes, 6 percent monocytes, and 2 percent basophils. The serum creatinine value was 1.8 mg/100 ml (normal, 0.5 to 1.2 mg/100 ml}. Fiberoptic bronchoscopy was performed on the admission day and was normal. Bronchoalveolar lavage was performed in the right middle lobe {with 3150-ml aliquots} and the cell count was 260 x 10"/ml with 31 percent neutrophils, 5 percent

lymphocytes, and 64 percent macrophages. Stains and cultures for aerobic and anaerobic bacteria including Legionella, fungi, acid-fast organisms, virus, and Pneumocystis carinii were negative. No rise in viral or Mycoplasma titers was noted. Methotrexate was withdrawn and symptomatic treatment given. Because the clinical situation did not improve in the following 48 h, TBB was performed in the right upper and lower lobes. The biopsy specimens showed an extensive lymphocytic interstitial infiltrate with granuloma formation {Fig 2}. Corticotherapy was started {methylprednisolone, 2 mglkgtd) because of persistent hypoxemia. Clinical improvement occurred after 48 h, and the patient was discharged from the hospital15 days after admission. At that time, PaO. with the patient breathing room air was 75 mm Hg, and pulmonary infiltrates had completely disappeared on the chest x-ray film. Pulmonary function tests also improved (Table 1}. CASE2

A 73-yeaN>Id woman with RA was admitted to the hospital in December 1991 with a presumptive diagnosis of pneumonia. In 1983, she had developed erosive symmetrical polyarthritis. Rest, physical therapy, gold salts, salicylates, penicillamirie, and low-ilose prednisone failed to control her progressive arthritis. In July 1991, oral treahnent with methotrexate sodium (7.5 mgt wk} was initiated with progressive improvement of joint complaints. The patient was a nonsmoker and had no history of lung 'disease. Four weeks prior to admission, she developed progressively increasing breathlessness and a dry cough. She had taken the last dose of methotrexate on the admission day. She was admitted with tachypnea (30 breaths per minute). She had no fever. Inspiratory crackles were heard in both lungs. While she received oxygen (2 Umin}, the PaO. was 60 mm Hg; Paco., 31.8 mm Hg; pH, 7.53. The white blood cell count was 17.500/mm3 , with 89 percent neutrophils, 3

FIGURE 1A. 'lbp: Chest radiograph showing diffuse airspace consolidation, poorly marginated irregular opacities. Bottom: High resolution computed tomography showing heterogeneous ground glass opacities and septal lines. CHEST I 104 I 5 I NOVEMBER, 1993

1821

FIGURE 3. Histologic pattern showing lymphocytic interstitial infiltrate with a smallnonnecrotic granuloma. percent lymphocytes, 6 percent monocytes, and 2 percent eosinophils. The sedimentation rate was 100 mmlh. A chest radiograph and computed tomography scan showed diffuse interstitial infiltrates in boili lungs with patchy con8uent lesions in the two upper lobes. Pulmonary function tests indicated a moderate restrictive ventilatory impairment (Table 1). Fiberoptic bronchoscopy was normal. Bronchoalveolar lavage 8uid from the right middle lobe contained 420 X 10' nucleated cells/mi. with 70 percent neutrophils, 6 percent lymphocytes, and 24 percent macrophages. Stains and culture showed the presence of a few colonies (2 x 10'/ml) of Haemophilus inftuenzoe, and treatment consisted of administration of trimethoprim and sulfamethoxazole (trimethoprim, 8 m~d; sulfamethoxazole, 40 m~d intravenously) after stopping methotrexate therapy. Forty-eight hours later, the patient remained severely hypoxemic. Then TBB was performed in the right upper lobe. Pathologic examination showed a massive interstitial lymphocytic infiltrate with granuloma formation (Fig 3). Treatment with intravenously administered methylprednisolone, 2 mglkJ¥d, was begun. The patient rapidly improved. Concomitantly, the infiltrates evidenced on chest roentgenogram, resolved. Ten days later, at discharge, arterial blood gases with the patient breathing room air were Pa01 , 74 mm Hg; PaCO,, 29 mm Hg; and pH, 7.46. Lung function tests confirmed clinical improvement (fable 1). DISCUSSION

Evidence is accumulating that low-dose methotrexate is an effective therapy to control RA refractory to more conventional treatment. 1 Common adverse reactions from low-dose methotrexate treatment include an increase in serum transaminase activity, nausea, and diarrhea. 1 The incidence of interstitial pneumonitis is lower than 5 percent in the patients receiving this low-dose regimen for RA. 6 •7 Although methotrexate pneumonitis is generally considered as a hypersensitivity reaction, contribution of a direct cytotoxic effect is not excluded, since the exact mechanism of lung injury remains controversial.• That it can develop with low doses and that there is a wide range of delay between drug intake and onset of disease (15 days to 5 years) supports the hypersensitivity hypothesis. The clinical history and radiologic findings in our two patients were consistent with the diagnosis of methotrexateinduced lung disease.8 •8 In both cases, TBB assessed the diagnosis: massive interstitial infiltrate by lymphocytes and granuloma formation, the classic histologic feature of methotrexate-induced pneumonitis. 8 •8 1122

A very striking observation was the coexistence, in both cases, of a neutrophilic alveolitis with a well-demonstrated lymphocytic and granulomatous lung infiltration. The bronchoalveolar lavage (BAL) cell abnormality associated with hypersensitivity pneumonitis at the granulomatous stage classically is an elevated lymphocyte count; 10 lymphocytic alveolitis usually is found in BAL Ouid samples from patients with methotrexate-induced lung disease."·'" Our lavage findings contrasted with previous data, 11 •11 particularly those of White et al, 11 who systematically found in their six cases of methotrexate-induced lung disease a lymphocytic alveolitis with BAL lymphocyte count over 20 percent and a neutrophil count under 10 percent. We have no definite explanation for the discrepancy between lavage findings and lung biopsies. Since the computed tomographic scan confirmed diffuse involvement of both lungs, it is difficult to ascribe local differences in the alveolar cell populations. Also, it is difficult to attribute a role to the short delay (less than 48 h) between the two diagnotic procedures in these subacute diseases. 10•13 Neutrophilic alveolitis has been associated with rheumatoid lung" and this might interfere with BAL results in our cases; most reported cases of methotrexate-induced lung diseases with lymphocytic alveolitis are complications of malignancies without obvious underlying lung diseases. 11 •1i It must, however, be underlined that lesions of rheumatoid lung were not detected in the biopsy specimens. Since methotrexate withdrawal together with high-dose steroid treatment has been proposed to accelerate recovery of this disease, •.u• it appears important to recognize rapidly methotrexate-induced pneumonia (especially in critically ill patients). The dramatic improvement observed in our two patients tends to confirm the beneficial role of early steroid administration in this disease. In patients treated with methotrexate for RA who present with interstitial pneumonia, differential diagnosis of methotrexate-induced toxicity has to be made with rheumatoid lung disease and opportunistic lung infections (P carinii, etc). 18 In this particular clinical setting, histopathologic findings are specific enough to allow the diagnosis of methotrexate-induced pneumonitis. Since our data suggest that BAL findings are not always contributive to the diagnosis of methotrexate-induced disease and because it is a less invasive method to obtain lung tissue than open-lung biopsy, 17• 18 TBB appears to be a very interesting procedure for diagnosis in such cases. REFERENCES

1 Weinblatt ME, Coblyn JS, Fox DA, et al. Efficacy of low-dose methotrexate in rheumatoid arthritis. N Eng! ] Med 1985; 312:818-22 2 Engelbrecht JA, Calhoon SL, Scherrer JJ. Methotrexate pneumonitis after low-dose therapy for rheumatoid arthritis. Arthritis Rheum 1983; 26, n• 10:1275-1278. 3 Cannon GW, Ward JR, Clegg DO, Samuelson CO, Abott TM . Acute lung disease associated with low-dose pulse methotrexate therapy in patients with rheumatoid arthritis. Arthritis Rheum 1983; 26:1269-75 4 Saint-Clair WE, Rice JK, Snydermann R. Pneumonitis complicating low-dose methotrexate therapy in rheumatoid arthritis. Arch Intern Med 1985; 145:2035-38 5 Louie S, Lillington GA. Low-dose methotrexate pneumonitis in rheumatoid arthritis. Thorax 1986; 41:703-04 Pneumonitis Compllcatiug l.o¥Mio8e Metholnlxate Therapy (Leduc et a/)

6 Kremer JM, Lee JK. The safety and efficacy of the use of methotrexate in long term therapy for rheumatoid arthritis. Arthritis Rheum 1986; 29:822-31 7 Carson CW, Cannon GW, Elfer MJ, Ward JR, Clegg DO. Pulmonary disease during treatment of rheumatoid arthritis with low-dose pulse methotrexate. Semin Arthritis Rheum 1987; 16:186-95 8 Cooper JAD, White DA, Matthay RA. Drug-induced pulmonary disease: part 1. Am Rev Respir Dis 1986; 133:321-40 9 Sostmann HD, Matthay RA, Putman CE, Smith GJW. Methotrexate induced pneumonitis. Medicine 1976; 55:371-88 10 Costabel U. Alveoli tis in hypersensitivity pneumonitis. Rev Mal Resp 1989; 6:121-26 11 White DA, Rankin JA, Stover DE, Gellena RA. Methotrexate pneumonitis: bronchoalveolar lavage findings suggest an immunologic disorder. Am Rev Respir Dis 1989; 139:18-21 12 Almun G, Mayaud C , Touboul JL, Denis M, Milleron B, Perrot J. Use ofbronchoalveolar lavage in the evaluation of methotrexate lung disease. Thorax 1987; 42:652-55 13 Fournier E, Tonnel AB, Gosset PH , Wallaert B, Ameisen JC, Voisin C. Early neutrophil alveolitis after antigen inhalation in hypersensitivity pneumonitis. Chest 1985; 88:563-66 14 Garcia JGN, Parkami N, Killam D , Garcia PL, Keagh BA. Bronchoalveolar lavage Ouid evaluation in rheumatoid arthritis. Am Rev Respir Dis 1986; 133:450-54 15 Hasan FM, Mark EJ. Case records of the Massachusetts General Hospital: a28 yeaM>Id man with increasing dyspnea, dry cough and fever after chemotherapy for lymphoma. N Eng! J Med 1990; 323:737-47 16 Wollner A, Makle-Batouni J, Lambert RE , Perruquet JL, Raffin TA, McGuire JL. Pneumocystis carinii pneumonia complicating low-dose methotrexate treatment for rheumatoid arthritis. Thorax 1991; 46:205-07 17 Anders GT, Johnson JE , Bush BA, Matthews JI. Transbronchial biopsy without Ouoroscopy: a seven year perspective. Chest 1988; 94:557 18 Herf SM, Suratt PM . Complications of transbronchial lung biopsies. Chest 1978; 73:759

EMHT =extramedullary hematopoietic tumor

E

xtramedullary hematopoietic tumors (EMHTs) that resemble myelolipomas may occur when the normal function of bone marrow is disturbed.' In the posterior mediastinum, EMHTs are usually associated with chronic congenital hemolytic anemia. 1-6 These tumors present as asymptomatic, unilateral, or bilateral paraspinal masses. A single case of mediastinal EMHTs associated with macrocytosis related to vitamin deficiency has been reported.7 This report presents the first case (to our knowledge) associated with primary refractory anemia that was confirmed by histologic examination of the surgically removed mass. CASE REPORT

A 65-year-old man was referred to the Marie Lannelongue Hospital for a right femoral bypass. He had a history of partial gastrectomy at the age of 25 years and had smoked one pack of cigarettes a day for 45 years. A right paravertebral thoracic mass was incidentally discovered on preoperative chest radiographs. The lesion was already present on systematic radiographs performed 2 years before. The patient was not investigated further at that time. He had no thoracic symptom, no fever, and no weight loss. Laboratory findings included the following: a WBC count of 8 X 10"/L with neutrophils 65 percent, lymphocytes 29 percent, and monocytes 6 percent; an erythrocyte count of 3.35 x 1012/L with hemoglobin concentration of 127 giL, hematocrit of37 percent, and mean corpuscular volume of llO 0; and a platelet count of 250 X 10"/ L. Other pertinent laboratory serum values were as follows: normal creatinine and liver chemistry studies, total bilirubin of 22 j.l.moVL (N<20), free bilirubin of 20 j.t.moVL, vitamin 8, 2 of 353 ng!L 1 (N> 130), and folate of2.8 j.t.g/L (N> 1.8). Fiberoptic bronchoscopy revealed no lesion and aortography showed no argument for pulmonary sequestration. Computed tomography showed two paravertebral masses at the T9level (Fig 1). The right mass and the left measured 8 and 3.5 em, respectively. These tumors had a smooth margination and were composed of areas of low density suggestive of fat deposition and of regions of water density. There was no calcification. The bones were normal and not eroded.

Extramedullary Hematopoietic lllmors of the Posterior Mediastinum Related to Asymptomatic Refractory Anemia* Vincent Thomas De Montpriville, M .D.; Elisabeth M . Dulmet, M .D .; Alain R. Chapelier; M .D.; Philippe G. Dartevelle, M .D .; and jeanne M . \erley, M.D.

Two asymptomatic paravertebral thoracic masses occurred in a 65-year-old patient with isolated macrocytosis. The largest one measured 8 em and was surgically resected with a presumptive diagnosis of schwannoma. This thoracic mass was hemorrhagic, encapsulated, and composed of fat and hematopoietic tissue. While extramedullary hematopoietic tumors usually occur in patients with severe chronic hemolytic anemia, our report suggests that such lesions must be considered in the differential diagnosis of posterior mediastinal mass in patients without clinically evident anemia. (Chut 1993; 104:1623-24) *From the Departments of Surgical Pathology (Drs. Thomas De Montpreville, Ouimet, and Verley) and Thoracic and Vascular Surgery and Heart-Lung Transplantation (Drs. Chapelier and Dartevelle), Marie Lannelongue Surgical Center (Universite ParisSud), Le Plessis-Robinson, France.

FIGURE 1. Computed tomographic scan of the chest showing the two round paravertebral masses containing fat. CHEST I 104 I 5 I NOVEMBER, 1993 I

\

1623