Postcardiotomy shock requiring mechanical circulatory support: experience from our centre

Postcardiotomy shock (PCS) is an infrequent but severe complication. Initial management consists of pharmacological support; however, in refractory cases, circulatory support becomes necessary.

The aim of this study is to analyze the characteristics of patients with PCS requiring veno-arterial extracorporeal membrane oxygenation (VA-ECMO) implantation, and to compare complications and survival outcomes with the use of this support in other clinical scenarios.

Methods: A descriptive, retrospective, single-centre study including all patients supported with VA-ECMO because of PCS between 2021 and 2025.

Patients were categorized into three groups: PCS, cardiac arrest with VA-ECMO (ECPR) and patients whose etiology did not correspond to either of the previous categories (Non-PCS/ECPR).

Results: A total of 147 VA-ECMO devices were implanted: 24.5% (n=36) for PCS, 29.9% (n=44) for ECPR, and 54% (n=67) for Non-PCS/ECPR. Variables from the EuroSCORE II were collected for the PCS group. Notably, 38.9% of patients presented in a critical preoperative condition. The EuroSCORE II showed a wide range, with a median of 9.2 [IQR: 3.2–21.8].

Regarding surgical type, 47.2% underwent coronary artery bypass grafting and 72.3% underwent valvular surgery. Most patients (n=22, 61.1%) underwent more than one procedure. The mean cardiopulmonary bypass duration was 249 ± 139 minutes, and the mean aortic cross-clamp time was 163 ± 85 minutes.

Peripheral cannulation was used in 75.0% (n=27) of PCS cases and 47.2% (n=17) required a left ventricular unloading device, performed with an intra-aortic balloon pump in all cases. There were no significant differences in the unloading strategy compared with the ECPR group (52.8%, p=0.308) or the Non-PCS/ECPR group (43.3%, p=0.701).

The incidence of complications during support was 72.2% (n=26) in the PCS group, not significantly different from ECPR (65.0%, p=0.499), although numerically higher than in Non-PCS/ECPR (52.0%, p=0.059). The most frequent complication in PCS was massive transfusion (77.1%, n=27).

Hospital mortality in PCS was 69.4% (n=25). The most common causes of death were bleeding (28.0%, n=7), multiorgan failure (24.0%), and lack of recovery (20.0%). The ECPR group showed similar mortality (63.6%, p=0.585), but neurological injury was more frequent (42.9%). In the Non-PCS/ECPR group, mortality was significantly lower (44.8%, p=0.017), with multiorgan failure being the most frequent cause (43.3%). These survival differences persisted at one year follow-up.

Conclusions: In our center, PCS patients requiring VA-ECMO are typically those undergoing urgent surgeries involving multiple procedures. The preoperative risk estimated by EuroSCORE II indicates a predicted mortality of 9.2%. In-hospital and one-year survival rates are comparable to those of ECPR but lower than those of Non-PCS/ECPR patients, although the pattern of complications differs, with a higher incidence of massive transfusion.

Baseline characteristics