Proof-of-concept study evaluates z-score interpretation in pulmonary conditions, highlighting diagnostic challenges.
Abstract Background The probabilistic interpretation of low lung diffusing capacity for carbon monoxide (DLCO) and nitric oxide (DLNO) has been recently standardized but the impact of different lower limits of normal (z-scores) on clinical assessment is not well established. Objective To assess the uncertainty zone of DLCO and DLNO z-score interpretation in computed tomography (CT)-determined interstitial pulmonary fibrosis (IPF) and pulmonary emphysema (PE). Design and methods In a combined retrospective (IPF) and prospective (PE), proof-of-concept study, standard lung function, including DLCO, and single-breath DLNO were measured. Z-scores derived from Global Lung Function Initiative (GLI) for DLCO and non-specific (ERS), sex- and device-specific (Munkholm et al.) or -corrected (Zavorsky & Cao) DLNO reference equations, analyzed. Results 120 adults subjects with available CT of the chest participated in the study, 66 of them with IPF and the other 54 with a CT pattern consistent with PE. 56 asymptomatic subjects served as a control group. DLCO from GLI and DLNO from any reference equations showed high sensitivity and specificity for both IPF and PE, with DLCO from GLI showing the highest diagnostic accuracy for PE and DLNO from Munkholm et al. for IPF. However, the best thresholds separating IPF and PE from asymptomatic control subjects were widely different and ranging from ~14th percentile (-1.095 z-score) for DLCO from GLI to ~3rd percentile (-1.900 z-score) for DLNO from ERS. The DLCO z-scores from GLI showed the strongest negative correlation with PE (r=-0.710, P<0.0001) and DLNO from ERS with IPF (r=-0.750, P<0.0001) but in females z-scores from ERS and Zavorsky & Cao were not significantly correlated with extent of IPF. Conclusion DLCO and DLNO thresholds separating subjects with IPF or PE from healthy controls may differ substantially from standard lower limits of normal of -1.645 and -1.960 z-scores. For DLCO it may be due to inhomogeneous impairment of blood-to-air barrier whereas for DLNO different devices and reference equations seem to play a major role. Clinical trial registered with https://register.clinicaltrials.gov/prs/beta/records (ClinicalTrials.gov Identifier: NCT07091838)
No takes yet. Share an insight, caveat, or question.
Barisione et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: