ABSTRACTS 283

species, whereas resistance to GM may have developed through additional mechanisms as well. Although this approach to penumonia prevention is clearly efficacious in this animal model, clinical studies are needed to define the frequency and significance of microbial resistance in human subjects. (Reprinted with permission.)

Unexpectedly High Incidence of Pneumocystis carinii Infec- tion After Lung-Heart Transplantation: Implications for Lung Defense and Allograft Survival. Gryzan S, Paradis IL, Zeevi A, Duquensnoy RJ, Dummer JS. Grifith BP, Har- desty RL, Trento A, Nalesnik MA, Dauber JH. Am Rev Respir Dis 137:1268, 1988.

Pneumonia due to Pneumocystis carinii (PCP) is regularly encountered in organ allograft recipients who are immuno- suppressed to prevent rejection. Recipients of lung/heart allografts may be particularly prone to pulmonary infection due to systemic immunosuppression and the fact that defense mechanisms in the transplanted lung may be further impaired through tissue incompatibility and the effects of surgery. In this study, we monitored 16 lung transplant recipients for infection with Pneumocystis carinii using serial bronchoalveolar lavage (BAL) and found the prevalence of Pneumocystis infection of the lung to be 88%. Six episodes were associated with the usual symtoms of pneumonia, whereas IO episodes were associated with minimal or no symptoms. In 3 of the 6 symtomatic episodes, a concurrent bacterial infection was also found. The total number of cells recovered from the lung by BAL, the proportion of T- lymphocytes, and the number of cytotoxic/suppressor and helper/inducer cells were elevated during infection with Pneumocystis compared to before and after. Spontaneous and interleukin-2-induced proliferation of BAL cells in vitro was also higher during infection, suggesting that there was an increased number of activated T-lymphocytes in the air- spaces of the infected allograft. BAL cells cultured with irradiated spleen cells from the donor proliferated at higher levels when obtained after Pneumocystis infection than when obtained before or during infection even for subclinical infections. These results indicate that in the absence of prophylaxis, the prevalence of Pneumocystis infestation is very high after lung/heart transplantation. Impaired defense of the transplanted lung does not seem to stem from the inability of activated T-lymphocytes to accumulate in the allograft. Because of the possibility of concomitant infection with bacteria and P. carinii, this organism must be excluded as a pathogen in cases of bacterial pneumonia in lung allograft recipients who survive longer than 2 months. The significance of an increased number of cells recognizing donor antigens in the lung allograft after resolution of Pneumocystis infection remains to be determined, but it is possible that this is associated with rejection. (Reprinted with permission.)

Utility of Airway Endoscopy in the Diagnosis of Respiratory Complications of Cardiac Transplantation. Schulman LL, Smith CR, Drusin R. Rose EA. Enson Y, Reemtsma K. Chest 93:960, 1988.

We evaluated 39 episodes (in 32 patients) of pulmonary parenchymal infiltrates following cardiac transplantation

with fiberoptic bronchoscopy (FOB) in a prospective study of 94 consecutive recipients. Initial FOB established the diag- nosis in 24/39 (62 percent) instances. Subsequent examina- tions included repeat FOB (five), open lung biopsy (five), needle aspiration (two), and autopsy (nine), establishing 49 diagnoses. Specific pathogens were identified in 45 instances, neoplasm in two, and idiopathic interstitial pneumonitis in two. Bronchoalveolar lavage alone yielded diagnosis in 63 percent and transbronchial biopsy and bronchial washings/ brushings in 46 and 43 percent, respectively. Transbronchial biopsy suggested idiopathic interstitial pneumonitis in 17 instances, but four had spontaneous clearing, and open lung biopsy or autopsy showed alternative diagnoses (particularly CMV and Aspergillus) in I 1. The main complication of FOB was moderate (25 to 100 ml) hemorrhage after transbron- chial biopsy (10 percent); no severe episodes occurred despite elevated pulmonary vascular pressures. In this population of immunocompromised hosts: (1) bronchoalveolar lavage is the most sensitive bronchoscopic techniques for detecting infec- tion; (2) transbronchial biopsy is not useful in detecting CMV or Aspergillus infection; (3) pulmonary hypertension is associated with some risk of moderate but not severe hemor- rhage after transbronchial biopsy. (Reprinted with permis- sion.)

Diagnosis of Pulmonary Hemorrhage in the Immunocom- promised Host. Kahn FW, Jones JM, England DM. Am Rev Respir Dis 136: 155, 1987.

The efficacy of bronchoalveolar lavage (BAL) in diagnos- ing pulmonary hemorrhage was studied in 51 immunosup- pressed patients with new pulmonary infiltrates. Similar studies were performed in a control group of 8 nonimmuno- compromised patients. Hemosiderin content in Prussian- blue-stained alveolar macrophages obtained by BAL was graded using a numerical scale. This “hemosiderin score” correlated closely with the degree of hemorrhage seen in corresponding histologic sections in the 26 patients from whom a lung biopsy or autopsy specimen was available. Severe pulmonary hemorrhage was ultimately diagnosed in 14 cases, and a mild degree of hemorrhage was found in an additional 19 cases. Thrombocytopenia and invasive fungal infections were statistically associated with severe hemor- rhage, as was an increased percentage of alveolar macro- phages in the BAL sample. This study demonstrates the efficacy of BAL in diagnosing occult pulmonary hemorrhage in the immunosuppressed host and highlights risk factors associated with hemorrhage in this setting. (Reprinted with permission.)

Protected Transbronchial Needle Aspiration and Protected Specimen Brush in the Diagnosis of Pneumonia. Larch DC, John JF, Tomlinson JR, Miller KS, Sahn SA. Am Rev Respir Dis 136:565, 1987.

Protected transbronchial needle aspiration (PTBNA) of pneumonic lung theoretically could bypass dislodged upper respiratory tract flora, a potential source of contamination of protected specimen brush (PSB) cultures. To evaluate the usefulness of PSB and PTBNA in establishing the etiology of pneumonia, we prospectively studied 20 patients with acute bacterial pneumonia not receiving antibiotics. After informed