Obesity, Mortality, and ECMO

The use of extracorporeal membrane oxygenation (ECMO) in the obese population has been a hotly debated topic since the origin of ECMO. Early studies demonstrated a mortality reduction and decreased hospital length of stay for patients with a body mass index (BMI) of  35 or higher who received venovenous (VV) ECMO for pulmonary disease, despite increased cardiovascular and device-related complications.¹ However, a recent meta-analysis failed to reproduce this benefit, showing no change in mortality rate, length of stay, or duration of ventilator use.² Given these equivocal results, no upper limit has been placed on BMI when  patients are screened for VV ECMO cannulation.

The use of venoarterial (VA) ECMO in obese patients with pulmonary and cardiac failure has been screened historically by using the Survival After VA ECMO score; the investigators who developed that score showed that as BMI values increased, so did patient mortality rates.³ However, their study is more than a decade old and the results have been debated, especially as ECMO patient care algorithms and ECMO technology have significantly improved since its publication.^4,5 Adding to the complexity of the issue is a recent systematic review and meta-analysis that again failed to reproduce the association between BMI values and mortality rates for patients receiving either VA or VV ECMO.⁶ To address this issue, McCloskey et al analyzed the Extracorporeal Life Support Organization (ELSO) registry data to determine whether obesity is associated with VA ECMO mortality rates, with the hypothesis that as BMI increases, so does mortality.⁷

The ELSO registry is a database composed of patient outcomes and processes of ECMO care. McCloskey et al⁷ evaluated all registry patients treated with VA ECMO between January 1, 2015, and December 31, 2021, for the primary outcome of mortality at hospital discharge and for secondary outcomes of ECMO-defined complications, including both circuit-related and patient-centered complications. A unique aspect of this study is that the association between obesity and ECMO outcomes was assumed to not be linear, so fractional polynomials were used to transform BMI with flexible parameterization of continuous variables. BMI was thus transformed into a regression model, using S values to improve BMI analyses to fit a model that deviates from a linear relationship. Variables were assessed for multicollinearity with exclusions for large variance inflation factors. BMI was evaluated three ways: body surface area, World Health Organization BMI classes, and BMI as a continuous variable. To eliminate type I errors associated with the numerous complications analyzed, P < 0.001 was used.

The study is a retrospective analysis of 22,825 patients from the ELSO registry, of whom 12,275 (54%) died before hospital discharge. Deaths were more likely to occur in older, White, male patients with high BMIs. Important to the study outcome was that mortality rates were similar for patient populations before and after COVID-19 (2015-2019 and 2020-2021). The mean BMI for all patients was 29 ± 6.8, with survivors having a lower BMI (28.4 ± 6.5) compared with the total samples (P < 0.001). As obesity class increased, mortality rates also increased, with 47% of underweight patients dying versus 50% of normal-weight patients, 53% of pre-obese patients, and 56%, 58%, and 65% of obesity class 1, class 2, and class 3 patients, respectively (P < 0.001).

Using logistic regression to adjust for demographic characteristics, pulmonary and cardiac variables, ECMO-specific therapies, timing (before vs. after COVID-19), and patient outcomes, the investigators determined that a BMI with an S value of 3 (BMI³) was significantly associated with mortality (odds ratio [OR], 1.000005; P < 0.0001), with mortality rates increasing as BMI³ increased. Translated to clinical use, this finding means that when a BMI of 25 is used for comparison, the OR for mortality is 1.15 for a BMI of 35, 1.46 for a BMI of 45, and 2.12 for a BMI of 55. In addition, the ORs for mechanical complications and renal complications (both with and without the need for renal replacement therapy [RRT]) increased as BMI increased. The ORs for pulmonary complications decreased as BMI increased, similar to a protective pattern of obesity often occurring in patients with acute respiratory distress syndrome (ARDS) who receive ECMO.⁸
 
A novel discovery of this study is that when BMI is transformed into BMI³, several nonlinear relationships surface regarding complication rates and mortality in patients who receive VA ECMO. This association can be explained by several mechanisms. First, obesity often coincides with multiple comorbidities that lead to higher rates of mortality, such as diabetes mellitus, hypertension, hyperlipidemia, and chronic kidney disease. Second, a higher BMI often means a more difficult percutaneous cannulation process, which can lead to flow issues and bleeding complications. Third, obesity is independently associated with the development of heart failure, atrial and ventricular arrythmias, ARDS, increased need for RRT, higher rates of thrombosis, and sudden cardiac death.

This research is the first to evaluate the effect of BMI on such a large patient population. It is also the first to prove a nonlinear relationship among mortality, mechanical complications, renal-related complications, and increasing obesity. However, the study is not without some inherent limitations. A retrospective data analysis can only assume a correlation, not validate causation. Further, the ELSO registry is a voluntary database, with centers independently entering data; information can be omitted or inaccurate, creating potential selection bias in data interpretation. However, past audits have shown near-perfect accuracy of the ELSO database to patient information, so this bias is unlikely to alter the conclusions.⁹ Regardless of limitations, this study lends credibility to the addition of BMI cutoff values for VA ECMO cannulation criteria.


References

1. Peetermans M, Guler I, Meersseman P, et al. Impact of BMI on outcomes in respiratory ECMO: an ELSO registry study. Intensive Care Med. 2023 Jan;49(1):37-49.
2. Ng WW, Leung KC, Hui RW, Yeung Ng P, Ngai CW, Sin SW. Impact of obesity on outcomes in patients receiving extracorporeal membrane oxygenation: a systematic review and meta-analysis. Int J Artif Organs. 2025 Mar;48(3):211-215.
3. Schmidt M, Burrell A, Roberts L, et al. Predicting survival after ECMO for refractory cardiogenic shock: the survival after veno-arterial-ECMO (SAVE)-score. Eur Heart J. 2015 Sep 1;36(33):2246-2256.
4. Salna M, Fried J, Kaku Y, et al. Obesity is not a contraindication to veno-arterial extracorporeal life support. Eur J Cardiothorac Surg. 2021 Oct 22;60(4):831-838.
5. Heuts S, Mariani S, van Bussel BCT, et al. The relation between obesity and mortality in postcardiotomy venoarterial membrane oxygenation. Ann Thorac Surg. 2023 July;116(1):147-154.
6. Zaidi SAA, Saleem K. Obesity as a risk factor for failure to wean from ECMO: a systematic review and meta-analysis. Can Respir J. 2021 May 22;2021:9967357.
7. McCloskey CG, Hatton KW, Furfaro D, Engoren M. Obesity is associated with increased mortality in patients undergoing venoarterial extracorporeal membrane oxygenation. Crit Care Med. 2025 Mar 1;53(3):e567-e574.
8. Rudym D, Pham T, Rackley CR, et al. Mortality in patients with obesity and acute respiratory distress syndrome receiving extracorporeal membrane oxygenation: the multicenter ECMObesity study. Am J Respir Crit Care Med. 2023;208(6):685-694.
9. Dalton HJ, Butt WW. Extracorporeal life support: an update of Rogers' Textbook of Pediatric Intensive Care. Pediatr Crit Care Med. 2012 July;13(4):461-471.