Accuracy of preoperative cardiac power output vs cardiac index in predicting in-hospital outcomes among cardiac surgery patients: a retrospective cohort study
D Dela Cruz, G Escasura, A D Sadian, R H Tiongco II, A M NobleAbstract
Background
Accurate preoperative risk stratification is crucial among patients undergoing cardiac surgeries. Conventional measures such as left ventricular ejection fraction (LVEF) and cardiac index (CI) have limitations in reflecting complex perioperative hemodynamics. Cardiac Power Output (CPO) is an integrative marker that combines flow (cardiac output) and mean arterial pressure (MAP), reflecting the true pumping power of the heart. Its prognostic value has been demonstrated in critical cardiac illnesses, but its role in the perioperative cardiac setting remains unclear.
Purpose
This study aimed to assess the accuracy of preoperative cardiac power output versus cardiac index based on 2D echocardiogram in predicting in-hospital outcomes among adult cardiac surgery patients.
Methods
This retrospective cohort study included adult patients who underwent cardiac surgery at two tertiary centers from April 2022 to October 2025. Preoperative CPO was calculated as (mean arterial pressure x cardiac output)/451. The primary outcome was the accuracy of CPO in predicting in-hospital all-cause mortality. Secondary outcomes included in-hospital cardiovascular mortality and need for mechanical circulatory support (MCS). Optimal cutoffs were identified using Youden’s index. Diagnostic odds ratios, receiver operating characteristic (ROC) curves, and accuracy parameters were computed. Subgroup analyses were performed according to preoperative LVEF.
Results
A total of 241 patients were included (median age 62 years; 77% male). Median LVEF was 57.7%, median CI was 2.2 L/min/m², and median CPO was 0.8 W. In-hospital all-cause mortality was 5.4%, while in-hospital cardiovascular mortality was 3.3%. MCS was required in 8% of the patients. Preoperative CPO did not demonstrate superior discriminative ability over CI for all-cause mortality (AUC 0.48 vs 0.49; p=0.829), with both parameters showing low sensitivity but high specificity and negative predictive value. CPO was significantly associated with cardiovascular mortality (diagnostic OR 13.17; p=0.014) and the need for MCS (diagnostic OR 6.63; p=0.003), whereas CI was not. CPO demonstrated acceptable discrimination for predicting MCS (AUC 0.70). In subgroup analyses, CPO retained a significant association with cardiovascular mortality and MCS among patients with LVEF <50%.
Conclusion
Preoperative cardiac power output did not outperform cardiac index in predicting in-hospital all-cause mortality. However, CPO was significantly associated with cardiovascular mortality and need for mechanical circulatory support, particularly among patients with reduced preoperative left ventricular ejection fraction. These findings support the role of CPO as a complementary hemodynamic marker that reflects cardiac reserve. It also has a potential to enhance perioperative risk stratification beyond conventional indices.ROC curves: CI vs. CPOOptimal cutoffs of CPO and CI