Morning-to-evening decline in pulmonary diffusing capacity and spirometry

Pulmonary function tests are performed throughout the day, yet time-of-day effects on spirometry and pulmonary diffusing capacities for nitric oxide (DLNO) and carbon monoxide (DLCO) are poorly characterized, and DLNO's diurnal behavior is unknown. We examined whether these measurements vary in healthy men and derived within-day limits of change. Thirty-eight men (median age, 29.8 years) underwent up to six testing sessions during one waking day. Each session included spirometry (FVC, FEV₁, SVC, and FEV₁/FVC), followed by single-breath DLNO/DLCO testing to obtain DLCO, DLNO, alveolar volume (V_A), transfer coefficients KCO and KNO, and the DLNO/DLCO ratio. Time-of-day effects were assessed using linear mixed-effects models with subject-specific random intercepts; within-subject variability was quantified using the within-subject standard deviation. After quality filtering and multiplicity adjustment, FVC, FEV₁, SVC, DLNO, DLCO, and V_A decreased across the day (q ≤ 0.014). KCO declined before but not after false-discovery-rate adjustment (p = 0.041, q = 0.058); FEV₁/FVC, KNO, and DLNO/DLCO did not change significantly. Within-day variability ranged from 1.5% for SVC to 3.5% for DLCO. One-sided 95% within-day change limits were approximately 11 mL/min/mmHg for DLNO (z-score, 0.59), 2.8 mL/min/mmHg for DLCO (z-score, 0.59), 0.28 L for V_A (z-score, 0.35), and 0.27 for DLNO/DLCO. Spirometric volumes and absolute diffusing capacities declined from morning to evening, consistent with decreasing V_A, whereas ratio and transfer-coefficient indices were largely stable. Serial measurements should be performed at consistent times. These limits reflect measurement error plus short-term physiological or procedural variation and are specific to this device, protocol, and population.