Lactate and its evolution are associated with the prognosis of patients in shock, although there is little evidence in those assisted with an extracorporeal venoarterial oxygenation membrane (VA-ECMO). Our objective was to evaluate its prognostic value in cardiogenic shock assisted with VA-ECMO.

Study of patients with cardiogenic shock treated with VA-ECMO for medical indication between July 2013 and April 2021. Lactate clearance was calculated: [(initial lactate − 6 h lactate) / initial lactate × exact time between both determinations].

From 121 patients, 44 had acute myocardial infarction (36.4%), 42 implant during cardiopulmonary resuscitation (34.7%), 14 pulmonary embolism (11.6%), 14 arrhythmic storm (11.6%), and 6 fulminant myocarditis (5.0%). After 30 days, 60 patients (49.6%) died, mortality was higher for implant during cardiopulmonary resuscitation than for implant in spontaneous circulation (30 of 42 [71.4%] vs 30 of 79 [38.0%], P=.030). Preimplantation GPT and lactate (both baseline, at 6hours, and clearance) were independently associated with 30-day mortality. The regression models that included lactate clearance had a better predictive capacity for survival than the ENCOURAGE and ECMO-ACCEPTS scores, with the area under the ROC curve being greater in the model with lactate at 6 h.

Lactate (at baseline, 6h, and clearance) is an independent predictor of prognosis in patients in cardiogenic shock supported by VA-ECMO, allowing better risk stratification and predictive capacity.