Dynamic EtCO₂/PETCO₂ changes for predicting fluid responsiveness during preload challenge in mechanically ventilated adults: a systematic review and meta-analysis

Fluid responsiveness assessment guides hemodynamic management in mechanically ventilated critically ill and perioperative patients, but direct cardiac output monitoring is not always available. Under stable ventilation and carbon dioxide production, dynamic EtCO₂ changes may reflect preload-induced changes in pulmonary blood flow and cardiac output. We evaluated the diagnostic accuracy of EtCO₂/PETCO₂ changes for predicting fluid responsiveness during preload challenge in mechanically ventilated adults. PubMed, Embase, Web of Science, the Cochrane Library, and ClinicalTrials.gov were searched from inception to July 12, 2026. Eligible studies evaluated dynamic EtCO₂/PETCO₂ changes during passive leg raising, mini-fluid challenge, or volume expansion in mechanically ventilated adults against an independent cardiac output-, cardiac index-, stroke volume-, stroke volume index-, or LVOT-VTI-based reference standard. Pooled sensitivity and specificity were estimated using a bivariate random-effects model, and reported AUCs were synthesized after logit transformation using a random-effects model. Prespecified subgroup analyses and univariable meta-regression explored clinical and methodological sources of heterogeneity. Risk of bias and certainty of evidence were assessed using QUADAS-2 and GRADE. Twenty studies involving mechanically ventilated adults were included, contributing 1,515 diagnostic observations or test conditions. Eighteen studies contributed 2 × 2 data, with pooled sensitivity of 0.748 (95% CI, 0.680–0.806), specificity of 0.853 (95% CI, 0.797–0.895), and SROC AUC of 0.872. Across all 20 studies, the random-effects pooled reported AUC was 0.839 (95% CI, 0.788–0.879). Exploratory subgroup analyses suggested numerically higher estimates for PLR-based assessment, relative thresholds, septic shock cohorts, and patient-level analyses, without establishing subgroup superiority. QUADAS-2 raised frequent concerns about index-test thresholds and reference-standard reporting; GRADE rated the certainty of evidence as low. Dynamic EtCO₂/PETCO₂ changes may provide a useful non-invasive adjunct for assessing fluid responsiveness during standardized preload challenge in mechanically ventilated adults under stable ventilatory conditions. They should not be used as a standalone rule-in or rule-out test, and the available evidence does not support a universal threshold.