Ten patients developed pulmonary fibrosis after bischloroethylnitrosourea (BCNU) therapy for malignancy. This was lethal in seven patients, four of whom had no evidence of tumor at autopsy. Presenting symptoms were either the insidious onset of cough and dyspnea or the sudden onset of respiratory failure. Physical findings were unremarkable. Chest roentgenogram usually showed interstitial infiltrates. Pulmonary function studies showed resting hypoxia with diffusion and restrictive defects. This complication of therapy does not appear to be dose related and may be made more likely by the concomitant administration of cyclophosphamide. Prednisone therapy did not benefit most patients. The literature and the implications of the use of BCNU alone or in combination are reviewed.

A randomized, double-blind, multicenter study in 181 afebrile cancer patients with ANC levels < 500/microL receiving myelosuppressive chemotherapy was undertaken to compare sargramostim (yeast-derived recombinant human granulocyte-macrophage colony-stimulating factor, RhuGM-CSF) and filgrastim (bacteria-derived recombinant human granulocyte colony-stimulating factor, RhuG-CSF) in the treatment of chemotherapy-induced myelosuppression. Patients received daily subcutaneous (SC) injections of either agent until ANC levels reached at least 1500/microL. There was no statistical difference between treatment groups in the mean number of days to reach an ANC of 500/microL, but the mean number of days to reach ANC levels of 1000/microL and 1500/microL was approximately one day less in patients receiving filgrastim. Fewer patients in the sargramostim arm were hospitalized, and they had a shorter mean length of hospitalization, mean duration of fever, and mean duration of i.v. antibiotic therapy compared with patients who received filgrastim. Both growth factors were well tolerated. No patient was readmitted to the hospital after growth factor was discontinued. Sargramostim and filgrastim have comparable efficacy and tolerability in the treatment of standard-dose chemotherapy-induced myelosuppression in community practice.

In 1991, an outbreak of Pseudomonas cepacia bacteremia (PCB) occurred among patients at an oncology clinic in Alabama. A case-patient was defined as any patient at Alabama Oncology Hematology Associates (AOHA) who had at least one blood culture positive for P. cepacia from 7 August through 31 October. Fourteen case-patients were identified; all required hospitalization (median duration, 17 days), but none died of PCB. A cohort study assessing risk factors for PCB focused on all patients who had been treated on the 8 days when case-patients had last visited AOHA during the period 7-21 August. Only patients with central venous catheters developed PCB (P < .001). Among patients with central venous catheters, PCB occurred only after visits to AOHA at which the catheters were flushed with heparin solution in the AOHA laboratory rather than in the treatment area (P < .001). P. cepacia was cultured from the only intravenous fluid bag used to prepare heparin flush solution in the laboratory during the interval 7-21 August. All outbreak-associated isolates of P. cepacia had an identical DNA ribotype pattern. These findings emphasize the importance of avoiding multiple use of single-use solutions, especially for high-risk patients with long-term indwelling central venous catheters.