Tuberculous Pleural Effusion* Twenty-Year Experience Allan F. Seibert, M.D.; johnson Haynes, Jr. , M.D., F.C.C.P.; Robert Middleton, M.D. ; and john B. Bass, Jr. , M.D. , F.C.C.P.

We reviewed the records of 1,738 cases of tuberculosis seen during the period from 1968 to 1988 in Mobile, Alabama. Seventy cases of tuberculous pleural effusion were identi· Sed and constituted 4.9 percent of all disease due to Mycobacterium tuberculoN during this period. Tuberculous pleural effusion was diagnosed if the patient had M tuberculoN cultured from sputum, pleura, or pleural Ruid and had a roentgenographic pleural effusion without an alternative explanation for the presence of the effusion. The diagnosis of tuberculous pleural effusion was made in the absence of a positive culture if the patient had an undiagnosed lymphocytic exudative pleural effusion and all

clinical and roentgenographic abnormalities resolved on antimycobacterial chemotherapy. The mean age of all patients was 47 ± 18.4 years. The 70 cases were evenly divided between 35 that were accompanied by roentgenographic pulmonary parenchymal inJiltrates and 35 that occurred in the absence of parenchymal infiltrates. We conclude that cultures of all potentially diagnostic specimens (sputum, pleural Ruid, and pleura) and an intermediatestrength skin test, are sensitive tests for the diagnosis of tuberculous pleural effusion. In addition, the age of patients with tuberculous pleural effusion appears to be increasing.

pleural disease due to Mycobacterium tuberculosis is generally categorized as extrapulmonary despite an intimate anatomic relationship between the pleura and the pulmonary parenchyma.• Tuberculous pleural effusion is thought to result from a delayed hypersen· sitivity reaction which occurs in response to the presence of mycobacterial antigens in the pleural space. 2 These mycobacterial antigens may gain access to the pleural space from the rupture of a small, subpleural caseous focus. 3 Tuberculous pleural effusion has been described as an acute granulomatous pleuritis occurring as a sequel to recent tuberculous infection in young adults and children who usually do not have roentgenographically apparent parenchymal tuberculosis. 4 .x However, it is now known that tuber· culous pleural effusions may occur in older adults and in patients with classic reactivation tuberculosis.9.1° Recent evidence suggests that the epidemiology of tuberculous pleural effusion may be changing. Establishing a diagnosis of tuberculous pleural ef· fusion can sometimes be difficult because all the classic findings of a lymphocytic exudative pleural effusion, pleural granulomata, and cutaneous sensitiv· ity to purified protein derivative (PPD) may not be present or available to the clinician. The diagnosis of tuberculous pleural effusion is probably most specific when some combination of the classic findings occur in a patient from whom M tuberculosis has been isolated by culture. Sputum, pleural Ruid, and pleural biopsy material have all been cultured in patients suspected of having tuberculous pleural effusion with

varying success. Pleural 8uid or pleural biopsy cui· tures which grow M tuberculosis have the highest specificity, but their diagnostic utility is limited by their sensitivity. Sensitivity can be increased by com· bining pleural histologic condition, pleural Ruid analysis, and multiple cultures of all potentially diagnostic specimens. However, acomprehensive evaluation does not guarantee a definite diagnosis of tuberculous pleural effusion. Despite a thorough evaluation, some exudative pleural effusions will defy a definite diagnosis. 11 The importance of considering a lymphocytic exudative effusion as tuberculous in etiology when no definite diagnosis can be made lies in Roper and Waring's 12 finding that 65 percent of patients with a positive tuberculin skin test and a serofibrinous effu· sion will develop active pulmonary tuberculosis within five years if they are not treated with anb'TB drugs. Because of the difficulty often encountered in establishing a diagnosis of tuberculous pleural effusion and the implications of missing the diagnosis, we reviewed our experience from 1968 to 1988 in patients with pleural tuberculosis. From this retrospective chart review, we hoped to determine the following: (1) the age distribution of patients with tuberculous pleural effusion; (2) the efficacy of mycobacterial cultures in the diagnosis of tuberculous pleural effu· sion; (3) the frequency with which tuberculous pleural effusion occurred in the setting of roentgenographic parenchymal disease; and (4) the prevalence of cuta· neous reactivity to PPD among patients with tuberculous pleural effusion.

*From the Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, University of South Alabama Colle~e of Medicine, Mohile. Manuscript re50 percent) exudative effusion (protein, >3.0 gldl, or lactic dehydrogenase (LDH). >200 IU) which resolved with appropriate antimycobacterial chemotherapy. All cases of tuberculous pleural effusion were treated with multidrug regimens. All tuberculin skin tests were performed by intradermal injection of5 tuberculin units ofPPD. Skin tests were performed at the time of initial evaluation. All skin tests were administered and read by a tuberculosis control nurse. Induration was measured in 48 to 72 hours and recorded in millimeters. Ten millimeters of palpable induration was used as the criterion for a positive test. All mycobacterial cultures were performed utilizing IAJwensteinJensen medium, and speciation was done with a standard niacin test. RESULTS

Seventy cases of tuberculous pleural effusion were identified, which represented 4.9 percent of all fonns of disease due to M tuberculosis during this 20-year period. The 70 patients consisted of 47 men (mean age, 51± 17.1 years), and 23 women (mean age, 39± 17.7 years). The mean age of all patients was 47 ± 18.4 years (mean± SD). Roentgenographic, skin test, and culture data are tabulated in Th.ble 1. All patients had a chest roentgenogram and at least three mycobacterial sputum cultures. Forty-three patients had an intermediatestrength tuberculin skin test applied and read, 50 had pleural fluid cultures, and 18 had pleural biopsy material cultured. Twenty-six patients had determinations made of pleural fluid total white blood cell counts with differential, protein, and LDH determinations. Fifty-six of the 70 cases occurred prior to 1982. Fourteen of the 70 cases occurred during the

period from 1982 to 1988. Although not tested for human immunodeficiency virus (HIV), none of these 14 was an intravenous drug abuser, a prostitute, or a homosexual, and none has developed an AIDS-associated opportunistic infection. Of the 70 patients, 35 had roentgenographic parenchymal infiltrates associated with their pleural effusion, and 35 had pleural effusion in the absence of parenchymal infiltrates. Thirteen of the 35 patients with infiltrates had cavitary infiltrates, and 22 had noncavitary infiltrates. The cavitary disease was located exclusively in the upper lobes or superior segments of the lower lobes. The noncavitary infiltrates were found in the upper lung fields in 14 cases and in the lower lung fields in eight. Distinct mediastinal or hilar lymphadenopathy was a roentgenographic finding in only two patients, both of whom had infiltrates in the lower lobes. Histologic findings from 13 of the 18 pleural biopsy specimens were available, and 11 of these 13 demonstrated granulomatous inflammation. All 11 cases in which pleural granulomata were found were also culture-positive. In 7 of the 11 cases only the pleural specimen yielded a positive culture. Six of these seven cases were without roentgenographic infiltrates. The overall sensitivity of the skin test was 93 percent (40/43). The three patients with negative (

Tuberculous pleural effusion. Twenty-year experience.

We reviewed the records of 1,738 cases of tuberculosis seen during the period from 1968 to 1988 in Mobile, Alabama. Seventy cases of tuberculous pleur...
4MB Sizes 0 Downloads 0 Views