A55-31 Severe Early Hypercapnia Drives Higher ICU and 28-day Mortality in Non-ARDS Patients

Abstract Introduction Permissive hypercapnia is an accepted ventilatory strategy in Acute Respiratory Distress Syndrome (ARDS) but its impact on outcomes in non-ARDS critically ill patients remains unclear. We sought to determine whether early hypercapnia independently predicts Intensive Care Unit (ICU) and 28-day mortality, and affects ICU length of stay (LOS), in a real-world cohort. Methods We performed a retrospective cohort study of adult (≥18 y) ICU stays in the Medical Information Mart for Intensive Care (MIMIC-IV) database with at least one arterial PaCO2 measurement within 24 h of ICU admission. We excluded ARDS by ICD-9/10 codes and physiologic criteria (PaO2/FiO2 ≤ 300 mmHg + ventilation start ≤ 24 h) and restricted to complete cases with observed ICU/28-day mortality and LOS. Peak PaCO2 was categorized as normocapnia (35-45 mmHg; n = 9,280), mild hypercapnia (46-60; n = 6,590) and severe hypercapnia ( 60; n = 1,477). We adjusted for age, gender, Charlson comorbidity index, first-day SOFA, APS III score, and early mechanical ventilation. Logistic regression estimated adjusted odds ratios (OR) for ICU and 28-day mortality; LOS was modeled by ordinary least squares (OLS) on log-transformed ICU days, with results exponentiated to percent changes. Results Among 17,347 ICU stays, severe hypercapnia was independently associated with higher ICU mortality (OR 1.84; 95% CI 1.57-2.15; p 0.001) and 28-day mortality (OR 1.59; 95% CI 1.37-1.84; p 0.001) versus normocapnia. Mild hypercapnia showed a non-significant trend toward lower ICU mortality (OR 0.90; 95% CI 0.80-1.01; p = 0.064) and was protective for 28-day death (OR 0.76; 95% CI 0.69-0.85; p 0.001). After adjustment, mild hypercapnia shortened ICU LOS by 8.8% (p 0.001), while severe hypercapnia had no significant effect (-0.2%; p = 0.91). Age, comorbidity burden, and organ-failure scores each increased mortality and LOS; early ventilation was associated with lower mortality. Conclusions Severe early hypercapnia ( 60 mmHg) independently predicts worse survival in non-ARDS ICU patients, while mild hypercapnia may be protective or reflect unmeasured confounding. These findings warrant prospective study and may inform ventilator management beyond ARDS. This abstract is funded by: Self