Abstract / Summary
Background Exertional desaturation in interstitial lung disease (ILD) is conventionally reduced to one value, usually nadir peripheral oxygen saturation (SpO 2 ) or walked distance. Continuous pulse oximetry now permits second-by-second characterisation of oxygenation and pulse-rate (PR) kinetics. We compared kinetic and conventional metrics during the six-minute walk test (6MWT) and one-minute sit-to-stand test (1STST) in ILD. Methods In this cross-sectional, single-centre analysis within the European ILD Registry (eurILDreg), 163 adults completed both tests on one day with continuous 1-Hz SpO 2 and PR recording. Panel time-series models estimated desaturation, recovery and PR slopes. Analyses were stratified by catheter-confirmed pulmonary hypertension (PH) and related to the GAP index. Results The 1STST desaturated 1.7-fold faster than the 6MWT (−0.022 versus −0.013%SpO 2 ·s −1 ) and loaded the circulation 9.7-fold faster (0.305 versus 0.031 bpm·s −1 ). Nadir SpO 2 was nevertheless lower during the 6MWT (86.9% versus 91.1%); after the 1STST it fell a further 24 s. Cross-test agreement was moderate for nadir SpO 2 (r=0.42), strong for peak PR (r=0.71). DLCO outperformed FVC for every significant endpoint, slope-based metrics most strongly (r up to 0.62), and was the dominant independent predictor after adjustment. PH (n=22) approximately doubled the 6MWT desaturation rate (−0.0220 versus −0.0111%·s −1 , p<0.001); the 1STST separated these patients only weakly. Desaturation slope steepened across GAP stages (p<0.001), yet GAP explained only 11% of its variance. Conclusions The 6MWT and 1STST impose distinct, complementary loads and are not interchangeable. Oximetry-derived slopes capture diffusion-driven gas-exchange failure more sensitively than amplitude-based endpoints and add information beyond established severity indices.