ECMO and coronary artery bypass grafting
Post-cardiotomy cardiogenic shock (PCCS) has an incidence of 2-6% in patients undergoing cardiac surgery and of these, an estimated 0.5-1% are refractory to maximal-dose inotropic support and IABP support4-6. Refractory PCCS rapidly leads to multi-organ failure and has an almost 100% mortality rate which makes it fatal without the use of MCS using devices such as VA-ECMO; thus, there is a clinical indication for its use in this cohort.
A range of retrospective studies have looked at outcomes including survival to weaning from ECMO and survival to hospital discharge for patients who have undergone cardiac surgery, but only a few of these have looked at outcomes in patients who underwent coronary artery bypass graft (CABG) surgery specifically6, 9-11. The largest cohort of these studies analysed outcomes and survival in 517 patients treated with ECMO for PCCS, of which 37% had undergone isolated CABG and 16.8% CABG in combination with valve surgery6. The study results showed that undergoing isolated CABG was associated with better survival, with an in-hospital mortality rate of 44% which was better when compared to that of other cardiac procedures (95% Confidence Interval (CI) 0.29-0.68; P< 0.001). A similar association was shown in another study looking at 31 patients (28% of total cohort) who were treated with ECMO following isolated CABG which demonstrated that undergoing isolated CABG was associated with improved in-hospital mortality on EMCO when compared to other cardiac procedures (41.3% vs 18.8%)9.
By contrast, a 2010 retrospective study looking at 233 patients who had undergone cardiac surgery and required ECMO post-operatively over an 11-year period, reported an increased hospital mortality which was associated with having undergone isolated CABG (P= 0.0015), among other factors10. This was also demonstrated in another retrospective study involving analysis of outcomes for 101 patients who received central VA-ECMO at a single centre over an 8-year period which reported the survival rate for isolated CABG as 17.2%, a figure lower than the overall survival rate of 27.7% at 1-year follow-up and is lower than survival rates reported in similar studies11. Additionally, patients requiring VA-ECMO following isolated CABG had a significantly increased mortality risk (Odds Ratio (OR) 3.23; 95% CI 1.1-9.4; P=0.021) compared to patients who did not undergo coronary surgery. Outcomes in ECMO following CABG from the included studies are summarised in Table 3.
Whilst there is no consensus on outcomes for patients requiring ECMO following isolated CABG, the aforementioned studies all demonstrated that in patients who had undergone CABG in combination with another cardiac procedure, namely valve surgery, was associated with a decreased survival rate with some studies quoting survival rates as low as 12.5%6,9-11.
Given that mortality rates for refractory PCCS without ECMO are almost always 100% this demonstrated the survival benefit of ECMO in post-CABG patients even though reported survival rates are variable. None of the studies included looked at the effect of differing ECMO initiation times on mortality and morbidity rates making it difficult to ascertain the best timing for starting ECMO for patients undergoing CABG surgery. In the studies which reported survival rates greater 40%, ECMO was instituted either intra-operatively during the primary cardiac procedure or within 24 hours following surgery6, 9. Whilst in studies that reported lower survival rates, there was no standardisation for when ECMO was instituted with ECMO being used as late as after the sixth post-operative day or data for when ECMO was instituted was missing completely. This would seem to suggest that ECMO should be given within the first 24 hours to maximise the survival benefit, though more research analysing the effect of when to start ECMO is required going forward.