Carbon monoxide diffusing capacity: a reliable indicator of bleomycin-induced pulmonary toxicity.

Seventy-seven untreated patients with germ-cell tumors, of whom 20 had no pulmonary involvement, no clinical or radiological evidence of pulmonary disease, received 6 courses of cisplatin, vinblastine and bleomycin. Pulmonary function tests were performed before, every 3 weeks during treatment, and once a month from then on. The carbon monoxide diffusion capacity tests were normalized with respect to the actual hemoglobin concentration. During treatment a nonsignificant decrease (P greater than 0.05) was seen in the flow volume relationships, but as it was reversible it was attributed to the strenuous treatment. In contrast a significant decrease of 35% (P less than 0.01) in the corrected carbon monoxide diffusion capacity was seen in 15 of 18 evaluable patients with neither clinical nor radiological signs of pulmonary toxicity. The decrease was correlated to the increasing dose of bleomycin (r = 0.63).