Continuous Capnography and Shorter Duration of Subclinical Hypoxemia During Sedated Gastrointestinal Endoscopy in...

Randomized trials have established that capnography reduces hypoxemia during sedated gastrointestinal endoscopy, but how capnography alters provider behavior in routine practice remains unclear. We aimed to quantify the association between capnography display visibility and hypoxemia episode duration in older adults undergoing sedated gastrointestinal endoscopy.

We conducted a multicenter clinician-blinded device evaluation across three tertiary hospitals in China (November 2022 - February 2024). Of 1229 patients aged ≥65 years screened; 1,189 were enrolled. Capnography hardware was attached to all patients, with waveform display and alarms accessible in the visible group (n=620) and obscured/muted in the concealed group (n=569). The primary outcome was cumulative hypoxemia duration (subclinical [90-95% SpO₂], moderate [85-90%], severe [≤85%]), analyzed using inverse probability of treatment weighting-adjusted zero-inflated negative binomial regression.

Hypoxemia incidence did not differ between groups at any threshold (subclinical: 28.7% versus 30.1%; P=0.569). Capnography visibility was associated with 30% shorter subclinical hypoxemia duration (incidence rate ratio 0.70, 95% CI 0.54-0.91; P=0.007) and a model-derived estimated 30-40 second earlier response interval. Moderate hypoxemia showed a directionally consistent trend (IRR 0.69, 95% CI 0.46-1.04; P=0.074); severe hypoxemia duration did not differ (P=0.514). Airway interventions were 86% more frequent in the visible group (RR 1.86, 95% CI 1.57-2.19; P<0.001), predominantly jaw-thrust maneuvers. Hypotension occurred in 1.8% of the visible group versus 0% of the concealed group (P=0.004). Sensitivity analyses demonstrated result stability (E-value ≥2.5; consistent across center exclusions).

Visible capnography was associated with shorter subclinical hypoxemia duration and more frequent early airway interventions, consistent with an information-mediated provider response mechanism. These findings characterize how capnography visibility is associated with early detection and provider behavioral change in routine practice, complementing incidence-reduction evidence from randomized trials. All findings are reported as associations; no causal claims are made.