Improved phenotyping of cardiogenic shock after out-of-hospital cardiac arrest

Abstract

Abstract Background Cardiogenic shock (CS) is an important cause of mortality after out-of-hospital cardiac arrest (OHCA). The SCAI classification is a simple grading system to reflect the spectrum of CS but fails to reflect phenotypes of CS based on haemodyamic/haemometabolic criteria [1,2]. Purpose We applied the SCAI grade and features of tissue hypoperfusion (i.e., elevated lactate) to develop a phenotypic classification of CS after resuscitated OHCA of cardiac aetiology and evaluate its association with outcome. Methods Patients with resuscitated OHCA of presumed cardiac aetiology were included in the King’s Out-of-hospital Cardiac Arrest Registry (KOCAR), and Uniwersyteckie Centrum Kliniczne w Gdańsku, Poland (UCK) between 1st May 2012 and 31st December 2020. Patients were classified into phenotypes of CS according to a combination of SCAI grade and lactate concentration on arrival at the hospital. Patients were then placed into one of 4 groups - those with SCAI grades A-B, i.e., no CS, were termed non-shocked if lactate <5, hypoperfusion group if lactate ≥5. Amongst those with CS (SCAI grades C-E), those with lactate <5 were considered to be in haemodynamic shock, and those with lactate ≥5 were in haemometabolic shock (Figure 1). The primary endpoint was mortality and cause of death at hospital discharge. Amongst surviving patients, the neurological outcome was defined by the Cerebral Performance Category (CPC) score. Results Of 627 patients (median age 62.9 ± 20.7, and 76.4% male), 218 patients (34.8%) were classified as non-shocked, 116 (18.5%) as hypoperfused, 106 (16.9%) as haemodynamic shock, and 187 (29.8%) as haemometabolic shock. Low flow time increased as the shock classification deteriorated, with the shortest time in the non-shocked group and the longest in the haemometabolic shock group (p <0.001). There were lower rates of initial shockable rhythm in haemometabolic shock followed by hypoperfused patients (p <0.001) and higher rates of adrenaline use in these groups (p< 0.001). Survival with good neurological outcome was most common in the non-shocked, then in haemodynamic shock groups, followed by the hypoperfusion group and with significantly less favourable survival in the haemometabolic shock group CS (Table 1; Figure 1; p<0.001). Modes of death differed significantly by shock classification group. Patients in the non-shock and hypoperfusion groups had low rates of cardiac multiple organ dysfunction syndrome (MODS) death, with the highest rates of neurological death seen amongst the hypoperfused group (35.7%). Patients in the haemodynamic shock group had substantially higher cardiac death rates, and those in the haemometabolic shock group had the highest cardiac and MODS death rates. Conclusion A classification based on SCAI grade and lactate levels identifies distinct phenotypes of CS after OHCA and may identify specific sub-groups for selection of mechanical circulatory support devices.Table 1Figure 1