Oil red O staining for lipid-laden macrophage index of bronchoalveolar lavage: interobserver agreement and challenges to interpretation.
Academic Article
Overview
abstract
INTRODUCTION: Oil Red O (ORO) staining on cytologic specimens with calculation of the lipid-laden macrophage index (LLMI) is used as a part of the workup in a number of clinical settings, particularly when aspiration is of concern. As a part of ongoing internal quality improvement measures, the objective of the present study was to evaluate the interobserver agreement of the LLMI calculation and to identify factors that affect the variability of the calculation. MATERIALS AND METHODS: There were 9 study participants, which included 3 trainees, 3 cytotechnologists, and 3 cytopathologists. Each participant reviewed 100 ORO-stained bronchoalveolar lavage slides and assigned an LLMI score to each case. The scores were categorized into 3 groups according to the associated aspiration risk: low, LLMI <40; intermediate, LLMI 40 to 90; and high, LLMI >90. The participants were also requested to note any challenges to the calculation for each case. RESULTS: The interobserver agreement among all participants was fair (κ = 0.23). Stratified by participant group, the interobserver agreement among the trainees was fair (κ = 0.24), among cytotechnologists was fair (κ = 0.32), and among cytopathologists was moderate (κ = 0.60). In 70 cases, at least one participant scored the case at least one category higher than the other participants; in 47 cases there was a two category difference. A primary diagnostic challenge reported by participants was macrophage pigmentation (hemosiderin, anthracosis). CONCLUSIONS: We found only fair interobserver agreement among all 9 participants in the study. Hemosiderin and anthracotic pigmentation was a major factor impeding LLMI calculation resulting in overestimation of the LLMI.
