INTRODUCTION

The incidence of cardiogenic shock following open-heart surgery has been reported between 3% and 5%^[1]. Despite most of the patients could be weaned from cardiopulmonary bypass (CPB) with pharmacotherapy and intra-aortic balloon pump (IABP), at least 1% evolves with progressive organ dysfunction in spite of optimized management and needed advanced mechanical circulatory support^[2]. Extracorporeal membrane oxygenation (ECMO) is the first line mechanical circulatory support, which allows time for “bridge to decision” or “bridge to recovery” for rescue patients in refractory cardiocirculatory failure^[3,4]. ECMO is a clinically demanding procedure, in which a multidisciplinary approach is required, with significant financial burden to the institution and public health system. The latter has hampered their use in Latin America.

The objective of this study was to assess the impact of ECMO in patients with cardiogenic shock after cardiac surgery in a single center with ECMO unit in Chile.

METHODS

RESULTS

Patient’s characteristics are shown in Table 1; 60% of them were men (n=10), 90% had hypertension (n=14), and 69% had left ventricular disfunction (n=11, X=25 ± 17). The mean EuroSCORE II was 12 ± 11, and Survival after Veno-Arterial ECMO score was -3; most of surgeries were performed in an urgent or emergency status. As several series, ECMO was associated with CABG in n=10 patients (62%), and six patients (38%) with complex cardiac surgery. Aortic cross-clamping time and CPB time are shown in Table 2.

Table 1 - Baseline characteristics of the entire cohort.

Table 2 - Intraoperative characteristics of the whole cohort.

ECMO is most often used for failure to wean from CPB, hence, we preferred to use the central or the mix cannulation in the OR (80%, n=13). Three (20%) patients were connected by peripheral cannulations using a cutdown technique in the ICU. The duration of support was 4.7 ± 2.6 days (1.5 - 12 days), which reflected that the recovery was earlier after cardiac surgery, and the weaning process was achieved in 88% (n=14). Although VA ECMO is the first line support in this pathology, IABP was used as the first approach in 56% of patients (n=9) undergoing CABG. We know that the benefit of the concomitant use of IABP with ECMO is unclear, however, once the IABP was implanted and the ECMO support was required, we preferred a simultaneous support aiming for left ventricular decompression and enhanced left side performance.

Complications in ECMO are common and increase over the time (Table 3); as it is shown in (Figure 1), patients who required longer ECMO support had higher mortality. The most frequent complication was surgical bleeding, requiring reoperation by cardiac tamponade in 56% of patients (n=9). Limb ischemia occurred in seven patients, two with peripheral cannulation and five with central cannulation. As it was expected, a higher mortality was highlighted in those with higher lactic acid level after 24 hours of ECMO assistance (Figure 2). In-hospital mortality was 44% (n=7) after discharge, three patients died, all of them by cardiac causes at follow-up. The freedom from all-cause mortality at one year of follow-up of the entire cohort was 38% (n=6) (Figure 3).

Table 3 - Complications during extracorporeal membrane oxygenation.

Fig. 1 - Blog-spot analysis of the time (days) in extracorporeal membrane oxygenation (ECMO) assistance and its relationship with mortality.

Download

Fig. 1 - Blog-spot analysis of the time (days) in extracorporeal membrane oxygenation (ECMO) assistance and its relationship with mortality.

Fig. 2 - Freedom from mortality at 30 days for the entire cohort of patients and its relationship with lactic acid level 24 hours after extracorporeal membrane oxygenation assistance.

Download

Fig. 2 - Freedom from mortality at 30 days for the entire cohort of patients and its relationship with lactic acid level 24 hours after extracorporeal membrane oxygenation assistance.

Fig. 3 - Kaplan-Meier curve showing a freedom from all-cause mortality of the entire cohort at one year of follow-up.

Download

Fig. 3 - Kaplan-Meier curve showing a freedom from all-cause mortality of the entire cohort at one year of follow-up.

DISCUSSION

Our ECMO program was started in 2016 to treat three prevalent pathologies in our country: i) hantavirus pulmonary syndrome, ii) adult respiratory distress syndrome, and iii) cardiogenic shock. This study aims to show our initial experience with VA ECMO utilization in patients with refractory cardiogenic shock after cardiac surgery in one single institution in Chile.

The incidence of VA ECMO utilization after open-heart surgery has been reported between 0.4 and 3.7%^[7]. Accordingly, we reported an incidence of 0.82% of ECMO utilization after open-heart surgery in patients with refractory cardiogenic shock. Although our reported incidence was low, we had a higher survival rate (38%) after one year of intervention, compared to the studies by Wang et al^[12]. (34%) and Khorsandi et al.^[10] (31%)^[8]. We could explain these differences due to the baseline characteristics of the patients included in other reports, such as, older patients, with prior myocardial infarction, usually had left main coronary disease, left ventricular dysfunction, prior open-heart surgery, and were frequently associated with combined surgeries^[9].

Interestingly, our data showed similar characteristics to almost 70% of the patients with severe left ventricular dysfunction, such as: i) > 60% of CABG in non-elective patients, and ii) high-risk scores (mean EuroSCORE II 12). Conversely, by the fact that the results showed in this article were the first data with ECMO utilization after cardiac surgery, which included our learning curve as team, our age population was younger than previous reports (58 years old).

In this cohort, the implementation of ECMO in all patients was unplanned, and the decision for support was guided by an increased requirement for vasoactive drugs followed by a close evaluation by transesophageal echocardiography in order to establish a true refractory cardiogenic shock. After that period, considering the high prevalence of patients with coronary disease, an IABP was utilized according to our institutional protocol, which explains that 56% of patients received one previously to ECMO implantation.

Despite the presence of an open-heart surgery, peripheral cannulation was frequently used as it reduces the possibility of mediastinal infection, avoids resternotomy, and allows for an uninterrupted chest compression during ECMO cannulation^[6,7]. Our institutional cannulation strategy was central cannulation if the patient’s sternum was still opened in the OR, that’s why only three patients had peripheral cannulation installed in the ICU unit.

In the other hand, central cannulation was associated with higher rates of bleeding but no difference in overall survival when compared with peripheral cannulation^[7,11]. We reported bleeding as the most frequent complication (81%) in our cohort because of recent surgery, usually with fragile tissue due to a reoperation and the early need of anticoagulation.

The mean duration of ECMO support was 4.7 days, as other publications who advocate that 48 to 72 hours support times are enough time to start the weaning. If insufficient recovery was observed and a more advanced mechanical support or a heart transplantation was needed, we transferred them to a center that perform heart transplantation^[12].

As it was reported by others^[7], we also founded a higher mortality but it was not statistically significant in patients with > 10 mg/dL lactic acid level after 24 hours of ECMO assistance. These patients were in salvage status or had a BMI > 34.

In spite of some enthusiastic ranging of weaning from VA ECMO (31-76%)^[7], we found that this was not correlated with the dramatic in-hospital mortality. We have had achieved 14 weaning in these five years, but our in-hospital mortality was 44%, and the principal causes were sepsis and arrhythmias. Doll et al.^[13] reported 76% in-hospital mortality over 219 patients after post cardiotomy refractory shock. Wang et al.^[12], in a meta-analysis of 20 observational studies, founded 34% of in-hospital mortality and one-year survival rate of 24%. Other metanalysis of Khorsandi et al.^[10] with 24 retrospective studies shows survival to in-hospital discharge of 30.8%.

CONCLUSION

Patients with refractory cardiogenic shock after cardiac surgery are still a very vulnerable population with a life-threatening condition. VA ECMO is now a well-known life-saving therapeutic option, but mortality and morbidity remain high. Our results are not by chance, managing these patients requires an ECMO program with a multidisciplinary teamwork with continuous training aiming to find the proper balance between patients benefits, ethical considerations, and public health financial input in South America.