CT findings in pulmonary alveolar proteinosis: serial changes and prognostic implications.

Abstract

BACKGROUND:

Pulmonary alveolar proteinosis (PAP) is a rare disease of unknown origin, characterized by impaired surfactant metabolism. In some patients, residual disease is observed after treatment; notably, this may progress to end-stage pulmonary fibrosis. This study was performed to evaluate changes in lung abnormalities on serial CT scans performed on the lungs of patients diagnosed with PAP, as well as to identify factors that can be used to predict clinical improvement in PAP.

METHODS:

Twenty-five patients (16 men, nine women) were diagnosed with PAP at a single tertiary hospital. The extent and distribution of PAP were assessed on baseline and follow-up CT scans (median, 38 months; range, 2-96 months). Serial CT scans and clinical findings were analyzed to identify the predictive factors for clinical improvement in PAP.

RESULTS:

Baseline CT scans of patients diagnosed with PAP revealed that ground-glass opacity was the most common abnormality (100%); the second most common abnormality was interlobular/intralobular septal thickening (88%). Importantly, the final follow-up CT scans showed that the extent of lung abnormalities had decreased (n=13), including complete resolution (n=5), unchanged (n=9), and increased (n=3). Traction bronchiectasis and architectural distortion were detected in two patients (8%). On univariate and multivariate analyses, the change in the overall extent of lung abnormalities was a predictive factor for clinical improvement in PAP (odds ratio: 55.780; P=0.038).

CONCLUSIONS:

Most patients with PAP exhibited residual disease; however, progression to pulmonary fibrosis was rare. Analyses of overall changes, with respect to lung abnormalities on serial CT scans, may be predictive of the extent of clinical improvement in given patient.

• Authors: Bo Da Nam, Tae Jung Kim, Man Pyo Chung, Myung Jin Chung, Tae Sung Kim, Kyung Soo Lee
• Journal: Journal of thoracic disease (J Thorac Dis) Vol. 10 Issue 10 Pg. 5774-5783 (Oct 2018) ISSN: 2072-1439 [Print] China
• PMID: 30505485 (Publication Type: Journal Article)