Post-cardiotomy cardiogenic shock (PC-CS) affects 2–6% of cardiac surgery patients. While survival rates keep improving, 40% of intensive care (ICU) survivors experience a “post-intensive care syndrome”, encompassing physical, cognitive, and psychological symptoms that highly impact quality of life (QoL) and predict persistent decline for years.

In PC-CS, the use of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is a mainstream life-saving circulatory support but is associated with complications which may influence QoL [1]. Current knowledge suggest heterogeneously impaired QoL, that depends on the pathology, severity of illness, type of assistance and duration of mechanical support. A recent review highlighted the low level of evidence on post-ECMO QoL, limited by small sample sizes and the absence of comparable control population [2].

This prospective cohort aims to assess long term QoL in PC-CS patients with or without VA-ECMO to address their multifaceted needs. As recovery is a time-dependent variable that improves within 9–12 months and then stabilizes, QoL was assessed by a SF-36 questionnaire at least 12 months after discharge from the ICU.

All consecutive adults admitted to our ICU from June 2014 to June 2019 with moderate to severe PC-CS (see ESM1 for the definition) were contacted by phone between May and July 2020 and proposed to complete the SF-36. Exclusion criteria were: long-term circulatory assistance (Left Ventricular Assist Device or heart transplant), or veno-venous ECMO (ESM1).

Among 6000 cardiac surgery patients, 474 (7.9%) were diagnosed with moderate to severe PC-CS over the period. All 291 survivors (68/201 with a history of VA-ECMO and 223/273 without) were contacted, 234 answered the SF-36 and 231 were analyzed: 55 (23.8%) with VA-ECMO and 176 (76.2%) without (ESM2). Overall, 132 (57.1%) patients were male, with a median age at hospital admission for surgery of 67 [58–74] years. General characteristics and ICU events are reported in ESM3. More than one year after discharge, they achieved physical and mental SF-36 scores of 78 [57–88] and 80 [62–87] respectively, reflecting an overall satisfactory return to good QoL. In a multivariable linear regression analysis, patients with a history of VA-ECMO had a lower physical SF-36 score than the control group after adjusting for emergency surgery, SOFA score, length of stay in ICU, length of mechanical ventilation, pulmonary infection, renal replacement therapy, transfusion, length of catecholamine infusion, delirium, age and length of follow-up (ESM4). Thus, VA-ECMO was the only factor associated with significant impairment in the physical SF-36 (-11.1 ± 4.4 points, P = 0.012), mainly impacting role limitation-physical and bodily pain (Fig. 1) whereas VA-ECMO accounted for a non-significant 3.2 lower mental score [± 4.0 points, P = 0.421]. This result highlights persistent difficulties in resuming previous daily activities, and suggests that these limitations may be related to worse levels of fatigue, energy and physical pain. Moreover, physical health (66 [40–85] vs. 80 [61–88], P = 0.007), was more profoundly affected than mental health (76 [54–84] vs. 80 [61–88], P = 0.016) in our patients, aligning with previous findings [2] and confirming that while some aspects of QoL improve over time, significant challenges remain.