Skip to Main content Skip to Navigation
Master Thesis

La metformine a t-elle un effet protecteur vis à vis de lésions majeures d’ischémie-reperfusion induites par un arrêt cardiaque ? Étude de survie et analyse avec score de propension

Abstract : Introduction: Cardiac arrest induces a massive ischemia-reperfusion (IR). Metformin is a worldwide- prescribed antidiabetic drug, with preliminary data suggesting a potential protective effect in various settings of sterile and non-sterile inflammation. Our hypothesis is that metformin, as a daily treatment before hospital admission, could improve the prognosis of patients admitted for a cardiac arrest. Method: We performed a retrospective single academic medical center study (University Hospital of Guadeloupe, French West Indies – Guadeloupe, France) about patients admitted for a resuscitated cardiac arrest in the ICU, no matter Out-of-Hospital or In-Hospital cardiac arrests. Comparison are scheduled to assess patients under metformin at the time of cardiac arrest versus others (MET versus NoMET groups), and survival at D28 (Alive versus Deceased groups). We derived a predicting model of mortality at the admission in the ICU by a logistic regression model (multivariate analysis). The Kaplan Meier estimator was used to compute a survival analysis (non-parametric logrank test). We performed a propensity score matched analysis, using a logistic multivariate analysis to perform an optimized model of risk factors associated to the D28 mortality with variables adjusted by the propensity score. Results: A total of 555 patients were included for analysis over the 6 years-study period, with n=62 patients in the MET group versus n=493 in the NoMET group. Survival at D28 was 32.3% vs 17.5% (MET vs NoMET), p<0.01. In the subgroup of diabetic patients (n=202), survival was, MET vs NoMET, 33% vs 20%, p=0.04. When comparing, at D28, alive patients (n=115) versus deceased (n=440), the frequency of metformin intake before the cardiac arrest was 17.3% vs 9.5%, p=0.01, as well as in the sub group of diabetic patients, 42.5% vs 27.0%, p=0.04. The multivariate analysis identified several independent criteria associated to D28 mortality, a treatment with metformin at the time of the first cardiac arrest remaining an independent criterion associated with survival (OR 0.477 [0.237-0.957]). The survival curves (Kaplan Meier) showed a better survival in the MET group through 28 days, log- rank test p=0.031), also focusing on the ischemia duration (Low-Flow cut-off = 20 minutes) combined to metformin treatment or not (stratified log-rank test p=0.027). The use of propensity score matching and multivariable regression modelling shows that three criteria previously reported in the multivariate analysis remains significantly associated to death (i.e. asystole on the first EKG, number of iterative cardiac arrest and SAPSII), while metformin remains significantly associated to survival. Discussion: This is a first clinical evidence of a protective effect of MET against the most severe IR, i.e. CA. These data will help to validate MET’s benefit against short-term IR consequences, while this should be addressed in non-diabetic patients and in sterile and non-sterile inflammation.
Document type :
Master Thesis
Complete list of metadata
Contributor : Pointe À Pitre Bhu Connect in order to contact the contributor
Submitted on : Tuesday, May 25, 2021 - 5:51:41 PM
Last modification on : Sunday, January 9, 2022 - 3:29:27 AM


Files produced by the author(s)


  • HAL Id : dumas-03235465, version 1



Barthélémy Vallée. La metformine a t-elle un effet protecteur vis à vis de lésions majeures d’ischémie-reperfusion induites par un arrêt cardiaque ? Étude de survie et analyse avec score de propension. Chirurgie. 2019. ⟨dumas-03235465⟩



Les métriques sont temporairement indisponibles