Efficacy and safety of extracorporeal membrane oxygenation for burn patients: a comprehensive systematic review and meta-analysis

Abstract

Background: Respiratory and circulatory dysfunction are common complications and the leading causes of death among burn patients, especially in severe burns and inhalation injury. Recently, extracorporeal membrane oxygenation (ECMO) has been increasingly applied in burn patients. However, current clinical evidence is weak and conflicting. This study aimed to comprehensively evaluate the efficacy and safety of ECMO in burn patients.

Methods: A comprehensive search of PubMed, Web of Science and Embase from inception to 18 March 2022 was performed to identify clinical studies on ECMO in burn patients. The main outcome was in-hospital mortality. Secondary outcomes included successful weaning from ECMO and complications associated with ECMO. Meta-analysis, meta-regression and subgroup analyses were conducted to pool the clinical efficacy and identify influencing factors.

Results: Fifteen retrospective studies with 318 patients were finally included, without any control groups. The commonest indication for ECMO was severe acute respiratory distress syndrome (42.1%). Veno-venous ECMO was the commonest mode (75.29%). Pooled in-hospital mortality was 49% [95% confidence interval (CI) 41-58%] in the total population, 55% in adults and 35% in pediatrics. Meta-regression and subgroup analysis found that mortality significantly increased with inhalation injury but decreased with ECMO duration. For studies with percentage inhalation injury ≥50%, pooled mortality (55%, 95% CI 40-70%) was higher than in studies with percentage inhalation injury <50% (32%, 95% CI 18-46%). For studies with ECMO duration ≥10 days, pooled mortality (31%, 95% CI 20-43%) was lower than in studies with ECMO duration <10 days (61%, 95% CI 46-76%). In minor and major burns, pooled mortality was lower than in severe burns. Pooled percentage of successful weaning from ECMO was 65% (95% CI 46-84%) and inversely correlated with burn area. The overall rate of ECMO-related complications was 67.46%, and infection (30.77%) and bleedings (23.08%) were the two most common complications. About 49.26% of patients required continuous renal replacement therapy.

Conclusions: ECMO seems to be an appropriate rescue therapy for burn patients despite the relatively high mortality and complication rate. Inhalation injury, burn area and ECMO duration are the main factors influencing clinical outcomes.

Keywords: Acute respiratory distress syndrome; Burns; Extracorporeal membrane oxygenation; Inhalation injury.

Figure 1

Flow diagram of study selection

Figure 2

Forest plot of mortality on ECMO for adult and pediatric burn patients using a random-effect model. ECMO extracorporeal membrane oxygenation, CI confidence interval

Figure 3

Forest plot of mortality on ECMO for burn patients with different burn severity using a random-effect model. Burn severity was defined as follows: minor burns: adult TBSA <30%, pediatric TBSA <15%; major burns: adult TBSA 30–50%, pediatric TBSA 15–30%; severe burns: adult TBSA >50%, pediatric TBSA >30%. ECMO extracorporeal membrane oxygenation, CI confidence interval, TBSA total body surface area

Figure 4

Forest plot of mortality on ECMO for burn patients with different percentage of inhalation injury using a random-effect model. Subgroup was stratified by the percentage of inhalation injury in different studies (≥50 vs <50%). ECMO extracorporeal membrane oxygenation, CI confidence interval

Figure 5

Forest plot of mortality on ECMO for burn patients with different ECMO duration. Subgroup was stratified by the ECMO duration in different studies (≥10 vs <10 days). ECMO extracorporeal membrane oxygenation, CI confidence interval

Figure 6

Bubble plot and meta-regression analysis of mortality on ECMO. (a) Bubble plot correlating age and mortality. (b) Bubble plot correlating burn area and mortality. (c) Bubble plot correlating percentage of inhalation injury and mortality. (d) Bubble plot correlating ECMO starting time and mortality. (e) Bubble plot correlating ECMO duration and mortality. (f) Summary result of meta-regression analysis of mortality. ECMO extracorporeal membrane oxygenation, CI confidence interval, SE standard error

Figure 7

Forest plot of successful weaning from ECMO for burn patients with different severity using a random-effect model. ECMO extracorporeal membrane oxygenation, CI confidence interval