Prognostic value of early ScvO2 and pCO2 gap kinetics and normalization in patients with cardiogenic shock: a new management target?
M Sobral Domingues, D Correia, A Bello, C Strong, J Presume, J Ferreira, C BrizidoAbstract
Introduction
Cardiogenic shock (CS) remains associated with high short-term mortality despite advances in management. Traditional haemodynamic parameters are often insufficient to assess tissue perfusion and guide therapy. Central venous oxygen saturation (ScvO2) and central venous–arterial pCO2 gap (pCO2 gap) are potential bedside surrogates of perfusion, yet their prognostic value and dynamic behaviour in CS are not fully established.
Purpose
To evaluate the prognostic impact of early ScvO2 and pCO2 gap normalization in patients admitted with CS.
Methods
We conducted a single-centre retrospective study on consecutive patients admitted to CICU with CS between 2015 and 2025. Patients with both baseline and 24 to 48h follow-up measurements of ScvO2 and/or pCO2 gap were analysed. Normalization thresholds up to 48h were defined as ScvO2 > 60% and pCO2 gap < 6 mmHg. The primary endpoint was 30-day all-cause mortality or advanced heart-failure therapy (heart transplantation or LVAD). Survival analysis was conducted using Kaplan–Meier estimates, and prognostic factors evaluated through univariate and multivariate Cox regression models. A discriminative analysis was also performed using receiver operating characteristic (ROC) curves.
Results
Among 446 patients with CS, 92 had paired ScvO2 measurements, and 79 had paired pCO2 gap measurements. Mean age was 67±17 years; 32% women. Regarding the CS aetiology, 159 (55%) presented with AMI-CS, 63 (23%) with HF-CS, and 58 (21%) with secondary-CS. Median ScvO2 and pCO2 gap were 58% [46-68] and 9 mmHg [7-12] at admission, and 60% [53-68] and 7 mmHg [5-9] at re-evaluation. Across patients with paired ScvO2 and pCO2 gap measurements available, 46 (50%) had ScvO2 normalized and 30 (38%) had pCO2 gap normalized at 48h. Normalization of ScvO2, but not pCO2 gap, within 48 hours was significantly associated with a lower incidence of the primary endpoint (log-rank p < 0.001 and 0.090, respectively). In multivariate analysis, normalization of ScvO2 (HR 0.36, 95% CI 0.18-0.70, p=0.003), but not pCO2 gap (HR 0.76, 95% CI 0.39-1.49, p=0.43), remained an independent predictor of the primary outcome after adjustment for age, sex, SCAI stage at admission, occurrence of cardiac arrest, and baseline lactate level. The area under the curve (AUC) for ScvO2 value on re-evaluation over the first 48h was moderate but statistically significant (AUC = 0.66 [0.56–0.77], p=0.002) whereas the AUC for pCO2 gap did not reach statistical significance.
Conclusions
In patients with CS, early ScvO2 normalization within the first 48 hours of intensive care was independently associated with improved short-term outcomes. In contrast, correction of the pCO2 gap did not demonstrate a significant prognostic impact in this cohort. These findings suggest that ScvO2 may serve as valuable, non-invasive tool for risk stratification and treatment guidance in acute cardiac critical care.For image description, please refer to the figure legend and surrounding text.