Acute interstitial pneumonitis (AIP) is an acute, idiopathic interstitial lung disease characterized by rapidly progressive diffuse pulmonary infiltrates and hypoxemia requiring hospitalization. The case-fatality ratio is high. Previous reports suggested that survivors of the acute event have a favorable outcome. We undertook this study to examine the natural history of survivors. We had observed several patients who experienced recurrent episodes of AIP and chronic progressive interstitial lung disease. We sought to determine longitudinal survival in these patients and to compare our experience with that in the medical literature. Overall, we identified 13 biopsy-proven cases of AIP. The mean patient age was 54 years in our review, which is identical to previous reports. Twelve patients were hospitalized and all 12 required mechanical ventilation. Overall hospital survival was 67%. All patients demonstrated abnormalities in gas exchange at presentation. Radiographs typically demonstrated bilateral patchy densities that progressed to a diffuse alveolar filling pattern in nearly all cases. All biopsy specimens showed organizing diffuse alveolar damage. Longitudinal data were available for 7 patients. Two died of AIP recurrences. A third died of complications of heart failure shortly after hospital discharge. One patient progressed to end-stage lung disease and required lung transplantation. Two patients experienced persistent pulmonary symptoms, accompanied in 1 by progressive lung fibrosis. One patient had nearly complete recovery of lung function 2 years following AIP. (Follow-up information was unavailable for 2 survivors.) In our literature review, 5 of 7 patients reported experienced some recovery of lung function. One case of progressive interstitial lung disease requiring lung transplantation was reported. The reported mortality was much higher than in our experience (74% versus 33%). The mean time from symptom onset to death was 26 days, compared with 34 days in our experience. The use of corticosteroids did not appear to influence survival, although this has not been tested in a rigorous manner. The better survival in our series may be related in part to a survivor selection bias. In contrast to previous reports, we found that survivors of AIP may experience recurrences and chronic, progressive interstitial lung disease. We did not identify any clinical or pathologic features that predict mortality in these patients. Likewise, there were no features that predicted the longitudinal course in survivors. Further study to identify causal factors is required in the hope of preventing morbidity and mortality related to this disease.