Twenty-two of 83 silicotic sandblasters in the New Orleans area had complicating mycobacterial infections: 10 with Mycobacterium tuberculosis, 9 with M. kansasii, and 3 with M. intracellulare. The place of residence was related to the variety of atypical infection. The mean age of the patients was 44 years and the average exposure to silica was less than 10 years. Sputum conversion occurred in all 18 patients with positive cultures, but 8 died in respiratory failure secondary to progressive silicosis. The roentgenographic features of the disease were often atypical: Lower lobe involvement occurred in 6 patients, hilar adenopathy without calcification in 8, and diffuse filling of air spaces in 5. Pulmonary function studies usually showed a mixed pattern of obstruction and restriction associated with impaired gas transfer. Pulmonary function often declined over relatively short intervals of time with corresponding adverse chest roentgenographic alterations.

To determine if something other than isoniazid causes the serum glutamic-oxalacetic transaminase (SGOT) elevations in isoniazid recipients, tuberculin-positive hospital employees receiving the drug were compared, by monthly SGOT measurements, with a comparable group of tuberculin-positive employees not receiving isoniazid. About 12% of the drug recipients developed SGOT levels higher than 100 mU/ml (upper limits of normal, 40 mU/ml). Once the drug had been withdrawn, their SGOT levels declined rapidly. In the control group, no one who had a base-line SGOT level below 100 mU/ml had subsequent values exceeding 100 mU/ml. Those subjects with elevations above 100 mU/ml (mean age, 50.2 years) were, on the average, older than the other isoniazid recipients (mean age, 36.9 years) whose SGOT levels remained below 100 mU/ml during treatment. The base-line mean SGOT level of the control group was slightly higher than in the drug group (excluding the 12% with significant SGOT elevations), but this relation reversed as soon as isoniazid treatment was started. This controlled prospective study shows, without question, that the SGOT elevations noted in the isoniazid recipients are related to the drug, not other factors.

SUMMARY _ Of 427 hospital employees receiving isoniazid chemoprophylaxis for 4 to 8 months, elevations of serum glutamic oxalacetic transaminase (SGOT) developed in 37. Isoniazid was discontinued in 5 because they also had symptomatic hepatitis and in 8 others who were asymptomatic because they also had abnormal serum values of total bilirubin or alkaline phosphatase or both. In 24 subjects, however, SGOT elevations were the only indications of hepatic injury. Testing their hepatic function biweekly or monthly enabled evaluation of criteria based on SGOT limits for interrupting their chemoprophylaxis. Treatment with isoniazid was stopped whenever the SGOT exceeded 100 units (upper limits of normal: 50 units) on two occasions, or whenever SGOT exceeded 250 units on one occasion. As a result, isoniazid was stopped in 5 subjects with elevations of SGOT. It was continued in 19 others, who remained asymptomatic; their tests after completion of chemoprophylaxis showed normal values for SGOT. In this population, these criteria for interrupting chemoprophylaxis provided a satisfactory margin of safety.

"Validity of Serum Glutamic Oxalacetic Transaminase Determinations in Isoniazid Recipients1–3." American Review of Respiratory Disease, 107(4), pp. 670–672