Z-scores for Spirometry Interpretation: Implications for Classifying Impairments in Lung Function

Area of Science:

• Pulmonary Medicine
• Diagnostic Spirometry
• Clinical Interpretation

Background:

• Recent spirometry interpretation guidelines advocate for z-scores over percent-predicted values for classifying lung function impairment severity.
• The clinical implications of adopting z-scores for spirometry interpretation remain largely unexamined.

Purpose of the Study:

• To determine the proportion of individuals whose lung function severity classification changes when transitioning from percent-predicted values to z-scores.
• To assess the association between altered severity classifications and respiratory symptoms, as well as clinical outcomes.

Main Methods:

• Two cohorts were analyzed: NHANES III participants and patients with COPD from Johns Hopkins Health System.
• Individuals were classified using both percent-predicted and z-score thresholds based on FEV1, categorizing them as normal, mild, moderate, or severe.
• Changes in classification (less severe, more severe, unchanged) were evaluated for associations with symptoms (dyspnea, cough, etc.) and outcomes (mortality, exacerbations, hospitalizations) using regression analyses.

Main Results:

• The analysis included 14,863 (NHANES) and 14,238 (COPD) individuals.
• Z-scores led to reclassification to a lesser severity in 10% of the NHANES cohort and 49% of the COPD cohort; no individuals were reclassified to a higher severity.
• Reclassification to a lesser severity was associated with a reduced risk of dyspnea, mortality, COPD exacerbations, and other outcomes, excluding cough.

Conclusions:

• Transitioning to z-scores for spirometry interpretation significantly reclassifies lung function severity, especially in COPD patients.
• Individuals reclassified to a lesser severity using z-scores exhibit a lower risk of clinically significant outcomes.
• These findings support the adoption of z-score thresholds for more accurate spirometry interpretation and risk stratification.