Short-term mechanical circulatory support in adults with congenital heart disease: indications and outcomes

Abstract Background An increasing number of adults with congenital heart disease (ACHD) present with heart failure (HF), which may require short- or long-term mechanical circulatory support (MCS), often around cardiac surgery or other interventions. Even though MCS has entered ACHD practice over the last few years, little is known on its indications and outcomes in this population. Purpose To assess the use and outcomes of short-term MCS in ACHD patients in a specialist centre, and assess the predictive power of the Survival After Veno-Arterial Extracorporeal Membrane Oxygenation (SAVE) score in this population. Methods Data on all ACHD patients who received MCS since 2008 in a single tertiary centre were collected. Immediate, short and longer-term outcomes were also collected, including death, complications, and weaning success. Logistic regression and ROC analysis were performed to identify predictors of in-hospital mortality. Result There were 65 episodes of MCS in 63 patients during the study period: age 36.9+-13.6 years, 61.5% male. Most patients had CHD of moderate or severe complexity (49.2% moderate and 10.8% severe according to ESC guidelines). The most common anatomical diagnoses were left outflow tract obstruction in 35.4%, Marfan syndrome in 20.0% and tetralogy of Fallot or pulmonary atresia in 10.8%. The most common indications for ECMO were perioperative support (not endocarditis) in 55.6%, HF in 14.3%, pneumonia in 12.7% and endocarditis-related surgery in 7.9%. A critical preoperative status was present in 26.8%; median PEACH score (a validated ACHD peri-operative risk score) was 1.0[1.0-3.0] and ACHS score was 0.6[0.6-1.0]. Median SAVE score was -4.0[-9.0–2.0], with 21(43.8%) in SAVE class 4 or 5. VA-ECMO was used in 69.2%, VV-ECMO in 21.5%, BiVAD-ECMO in 4.6% and RVAD support in 4.6% of patients. The median duration of ECMO support was 198.6[133.0-354.2] hours. Complications included ischemic stroke (15.4%), intracerebral bleeding (9.2%), and other thromboembolism (23.8%). Successful weaning was achieved in 56.2%, whereas 4.6% were switched to more definitive MCS. In-hospital mortality was 47.7%, and this was higher in patients receiving VA support (OR 4.92, 95%CI:1.32-23.93, p=0.03). For patients on VA-ECMO or BiVAD support, the presence of Marfan syndrome (OR 4.92, 95%CI:1.32-23.93, p=0.03) and a lower (worse) SAVE score (OR 1.3 per point reduction, 95%CI:1.11-1.59, p=0.004) were associated with in-hospital mortality. An adjusted SAVE score that accounted for the presence of heritable thoracic aortic diseases (HTAD, -3 points) performed well at predicting in-hospital mortality (AUC increasing from 0.73 to 0.8, Figure). Conclusions Short-term MCS is a useful temporary rescue intervention in patients with ACHD, as a bridge to recovery or durable support, with 1 in 2 patients surviving to hospital discharge. The SAVE score adjusted for HTAD can reliably predict outcomes in patients with VA-ECMO and requires external validation.Figure