Does every patient with enigmatic lung disease deserve a lung biopsy? The continuing dilemma

Define the negative predictive factors for survival in patients who undergo open-lung biopsy.

Retrospective analysis from July 1988 through January 1992 (prior to widespread use of video-assisted techniques). Factors included in analysis were: oxygen therapy, requirement for intubation, and HIV status. Also analyzed were the accuracy of the preoperative presumptive diagnosis and the outcome of therapy.

University hospital in an urban area with active medical and surgical critical care services.

All patients who underwent open-lung biopsy during the study design period. Patients excluded were those with lung transplants and one patient who had a thoracoscopic biopsy. Forty-eight lung biopsies were performed and 31 patients without AIDS and 15 with AIDS. Twenty-four (16 without AIDS, 8 with AIDS) were receiving oxygen therapy, met intubation criteria, or were intubated and ventilated at the time of biopsy. The indications were undiagnosed nodular or interstitial lung disease.

Presurgical evaluation included a preoperative diagnostic algorithm of transthoracic needle biopsy and nodular disease (9 patients) and at least one bronchoscopy with bronchoalveolar lavage and transbronchial biopsy in interstitial disease (22 patients). Seven patients with nodular disease and nine with infiltrate had no preoperative procedures.

Lung biopsy obtained at definitive diagnosis in 98 percent of patients. Biopsy significantly altered therapy in 79 percent (81 percent without AIDS, 73 percent with AIDS). Postoperatively, in-hospital mortality was 8 (17 percent): 5 (16 percent) without AIDS and 3 (20 percent) with AIDS p > 0.1. Mortality rates showed patients without respiratory compromise, 4.2 percent (7 percent without AIDS, 0 percent with AIDS); patients receiving oxygen therapy, 6 percent (10 percent without AIDS; 0 percent with AIDS); patients who met intubation criteria, 60 percent (33 percent without AIDS, 100 percent with AIDS); patients who were ventilated, 75 percent (66 percent without AIDS [p < 0.05], 100 percent with AIDS [p < 0.05].

We recommend that the decision for lung biopsy be individualized based on respiratory status.