Transbronchial lung biopsy for the diagnosis of lymphangioleiomyomatosis: the severity of cystic lung destruction assessed by the modified Goddard scoring system as a predictor for establishing the diagnosis

Abstract

A guide of patient selection for establishing the diagnosis of lymphangioleiomyomatosis (LAM) by transbronchial lung biopsy (TBLB) has not been established, although the pathological confirmation of LAM by lung biopsy is desirable, particularly when patients have no additional test results except typical findings of computed tomography (CT) of the chest. We retrospectively reviewed the medical records of LAM patients who visited at our hospital from January 2010 to September 2018. We found 19 patients who underwent TBLB and collected the following data to investigate which parameters could predict the TBLB diagnostic positivity for LAM: age, degree of exertional dyspnea, pulmonary function test, cystic lung destruction visually assessed by the modified Goddard scoring system (MGS), serum level of vascular endothelial growth factor-D, and TBLB-related data. The diagnosis of LAM was established by TBLB in 15 of 19 patients (78.9%) and no serious complications occurred. MGS was significantly higher in the TBLB-positive group than the TBLB-negative group. In LAM patients without pulmonary lymphatic congestion on CT (N = 16), multivariable logistic regression analysis revealed that MGS and FEV₁/FVC were independent contributing parameters for TBLB diagnostic positivity. However, the analysis of Bayesian inference demonstrated that MGS is a better predictor than FEV₁/FVC; the probability of establishing diagnosis exceeds 80% if MGS is > 2 (i.e., area of cystic destruction occupies > 25% of lung parenchyma on CT). MGS may be a helpful and convenient tool to select candidates for TBLB to establish the diagnosis of LAM pathologically.