Outcomes of Abdominal Exploration in ECMO-Supported Patients: A Multicenter ICU Cohort Study

Exploratory laparotomy, when required as a treatment for abdominal complications of post-extracorporeal membrane oxygenation (ECMO) initiation, carries a high mortality rate that approaches 80% to 100%. However, studies examining outcomes in this critically ill population remain scarce. We aim to share our multicenter experience to shed light on the outcomes of ECMO-supported patients undergoing abdominal exploration.

We conducted an institutional review board-approved multicenter retrospective cohort study across 3 hospital sites, focusing on ECMO-supported patients who underwent abdominal exploration for various abdominal complications. A comprehensive review of patient records was performed to assess baseline clinical, ECMO-related, and surgical parameters. Comparisons were made between survivors and nonsurvivors using univariate and multivariate analyses to identify mortality risk factors.

This study was conducted at 3 academic medical centers within a hospital health care system. The centers provided high levels of care, including advanced ECMO management, and served as referral hospitals for critically ill patients requiring complex surgical and intensive care support.

Among 1,386 ECMO-supported patients identified across the 3 centers, only 56 (4%) underwent abdominal exploration for various abdominal complications, such as mesenteric ischemia, intra-abdominal hemorrhage, abdominal compartment syndrome, and bowel perforation, identified on imaging or clinical deterioration requiring operative evaluation. Indications prompting surgical exploration included bowel pneumatosis, portal venous gas, free intraperitoneal air, mesenteric edema, or progressive abdominal distention accompanied by a rapid decrease in hemoglobin level, increasing vasopressor requirements, or rising ventilatory pressure.

The primary intervention was suspected abdominal complications arising during ECMO support. The surgical procedures included resection of necrotic bowel, repair of bowel perforations, and other interventions as clinically indicated, depending on the nature and severity of the abdominal pathology.

The cohort's in-hospital mortality rate was 57% (32 of 56). The median ages of nonsurvivors and survivors were similar at 57 years (interquartile range [IQR] 42-67) and 55 years (IQR 30-60), respectively. The majority of nonsurvivors were women (53%) and White (94%). Nonsurvivors had significantly shorter hospital (14 v 49 days) and intensive care unit (ICU) stays (14 v 38 days) than survivors. Univariate analysis showed nonsurvivors had higher Acute Physiology and Chronic Health Evaluation II (APACHE II) scores (17 v 9, p < 0.001), Sequential Organ Failure Assessment (SOFA) scores (8 v 4, p < 0.01) during ICU admission, and pre-ECMO lactate levels (11 v 3 mmol/L, p < 0.001) during ECMO initiation. Operative findings indicated that most ischemic injuries represented watershed hypoperfusion related to cardiogenic shock and ECMO physiology, while true thromboembolic occlusion was uncommon. Radiologic abnormalities correlated closely with pathology, with pneumatosis and free air corresponding to transmural necrosis and perforation.

Disease severity scores (APACHE II, SOFA) and the diagnosis of mesenteric ischemia are significant factors influencing mortality in ECMO-supported patients undergoing abdominal exploration. Multidisciplinary care, timely surgical interventions, and prompt management of abdominal complications are essential to improving outcomes in this high-risk population. These findings emphasize the importance of early recognition and aggressive management to potentially improve survival in critically ill patients undergoing ECMO therapy.