Clinical characterization of 39 patients with autoimmune pulmonary alveolar proteinosis.

ObjectivesWe aimed to explore the clinical manifestations, examination findings, diagnostic methods, and treatment efficacy of patients with pulmonary alveolar proteinosis and improve clinicians' understanding of this disease.MethodsWe retrospectively analyzed the clinical characteristics of 39 patients diagnosed with autoimmune pulmonary alveolar proteinosis admitted to Fuzhou Pulmonary Hospital, Fujian Province, from August 2013 to December 2024. We also analyzed the correlations between lactate dehydrogenase, arterial partial oxygen pressure, and pulmonary alveolar arterial oxygen differential pressure, lung function, tumor markers, blood lipids, and disease severity. Furthermore, the relevant literature was reviewed.ResultsOf the 39 patients, 23 were male, 16 were female, 18 were smokers, and 11 had pneumoconiosis. The main clinical symptoms were cough and shortness of breath. The typical manifestations of chest computed tomography were "map-like" changes or "crazy-paving pattern". The main manifestations of pulmonary function were diffusion dysfunction and restricted ventilation dysfunction. pulmonary alveolar proteinosis disease severity score was positively correlated with lactate dehydrogenase, pulmonary alveolar arterial oxygen differential pressure and the tumor markers carcinoembryonic antigen, cytokeratin 19 fragment antigen 21-1, neuron specific enolase, and carbohydrate antigen 153 (r = 0.689, P < 0.01; r = 0.801, P < 0.01; r = 0.414, P < 0.05; r = 0.570, P < 0.01; r = 0.488, P < 0.01; and r = 0.415, P < 0.05, respectively). The disease severity score was negatively correlated with the percentage of arterial partial oxygen pressure and carbon monoxide dispersion to the predicted value (r = -0.952, r = -0.554, P < 0.01). Lactate dehydrogenase was negatively correlated with carbon monoxide dispersion to the predicted value (r = -0.522, P < 0.01). Arterial partial oxygen pressure was negatively correlated with carcinoembryonic antigen, cytokeratin 19 fragment antigen 21-1, neuron specific enolase, and carbohydrate antigen 153 (r = -0.479, r = -0.607, r = -0.598, and r = -0.425, respectively, P < 0.01). In contrast, arterial partial oxygen pressure was positively correlated with carbon monoxide dispersion to the predicted value (r = 0.616, P < 0.01). pulmonary alveolar arterial oxygen differential pressure was positively correlated with carcinoembryonic antigen, cytokeratin 19 fragment antigen 21-1, neuron specific enolase, and carbohydrate antigen 153 (r = 0.529, P < 0.01; r = 0.546, P < 0.01; r = 0.554, P < 0.01; and r = 0.348, P < 0.05, respectively). Pulmonary alveolar arterial oxygen differential pressure was negatively correlated with carbon monoxide dispersion to the predicted value (r = -0.641, P < 0.01).ConclusionPulmonary alveolar proteinosis is a rare diffuse lung disease without specific clinical manifestations. Whole lung high-volume alveolar lavage is the most effective treatment for pulmonary alveolar proteinosis. Arterial partial oxygen pressure, pulmonary alveolar arterial oxygen differential pressure, lactate dehydrogenase, tumor markers, and carbon monoxide dispersion to the predicted value were correlated with the severity of pulmonary alveolar proteinosis.