ContextContrary to tonometer gastric intramucosal pH, there is currently no validated threshold prognostic value for Pco2gap (tonometer gastric mucosal Pco2minus arterial Pco2) in the critically ill patient.ObjectiveTo demonstrate a relationship between Pco2gap and mortality in mechanically ventilated patients.Design and SettingInception cohort study from a 9-month prospective survey of 95 consecutively ventilated critically ill patients in a teaching hospital.PatientsAll the ventilated patients of the intensive care unit were included at their admission.Measurements and Main ResultsGastric Pco2using regional capnometry with air-automated tonometry, arterial gas, lactate, and organ system failure score were measured at admission and after 6, 12, 24, 48, 72, 96, and 120 hrs. For the entire population, the 28-day mortality was 44%. In multivariate analysis, independent predictors of death were organ system failure score (odds ratio, 2.12; 95% confidence interval, 1.02–3.14), 24-hr Pco2gap (odds ratio, 1.57; 95% confidence interval, 1.10–2.24), and 24-hr lactate (odds ratio, 1.48; 95% confidence interval, 1.06–2.05). We found a threshold value of 20 mm Hg for Pco2gap and 2.5 mmol/L for lactate, which was associated with a sensitivity of 0.70 and 0.72, respectively, and a specificity of 0.72 and 0.73, respectively.ConclusionThe Pco2gap is a marker of mortality in ventilated patients in the intensive care unit.