Discussion
Cardiogenic shock is a serious and often lethal manifestation following cardiac insult with extremely high rates of in-hospital mortality. Historical rates have been reported as high as 80%15. However, more contemporary trials such as the SHould we emergently revascularize Occluded Coronaries for cardiogenic shock (SHOCK) trial and Global Registry of Acute Coronary Events (GRACE) registries have reported rates around 60%16,17. Few reports have achieved mortality rates below 50%18,19. Considerably less data exists in the cardiac surgical population in patients with cardiogenic shock. Existing literature is limited to smaller case series and case reports. Herein, we report our experience of 604 procedures performed on patients in cardiogenic shock.
Hemodynamic stabilization and bridging of the cardiogenic shock patient population is highly debated. In our series, we utilized intravenous inotropes most frequently (35.4%), followed by the use of intra-aortic balloon pumps (33.4%). The effectiveness of intra-aortic balloon pump in cardiogenic shock, especially in the setting of acute myocardial infarction, has been argued. The IABP SHOCK II Trial has demonstrated no mortality benefit of these pumps in addition to early revascularization2. Early revascularization, however, has been proven to offer early and long-term survival benefit in this patient population20. Other methods of bridging such as venoarteroal ECMO and/or Impella device were utilized infrequently in our series, comprising less than 4% of our surgical experience. ECMO bridge to cardiac surgery other than transplant or ventricular assist device has been described in prior case series and reports. Common indications were for mechanical complications of acute myocardial infarction including ventricular septal defect or papillary muscle rupture21–23. Long-term outcomes of hemodynamic support with veno-arterial ECMO for cardiogenic shock have not been well studied.
Timing of surgical intervention can be as important to patient outcomes as the operation itself. No specific guidelines exist for the patient in cardiogenic shock, and often, timing is individualized according to the patient’s clinical status. In our opinion, the timing of operation in these patients should be dictated by the disease process, baseline risk of the patient including comorbidity assessment, as well end-organ status. For example, a patient presenting in volume overload and marginal hemodynamics with an acute-on-chronic valvular process may benefit from intravenous diuresis, inotropic or temporary mechanical circulatory support, and improvement in end-organ status before commencing with surgery. Sometimes this is not possible such as in cases of acute valve leaflet perforation from endocarditis where the heart is not pre-conditioned to valvular insufficiency. These patients typically require emergent operations. A similar balance exists in coronary revascularization. Patients presenting with acute ST-elevation myocardial infarctions with ongoing ischemia require emergent revascularization, whether percutaneously or surgically. In a patient presenting late from an acute myocardial infarction who is in cardiogenic shock but has no ongoing signs of active ischemia, stabilizing hemodynamics and optimizing the patient for a short time period before revascularization may be appropriate in select cases.
The early mortality rates in our study were markedly lower than those described in large cardiogenic shock trials and likely a result of selection bias when deciding operative candidacy. Patients with higher comorbidity burden and perceived prohibitive surgical risk were likely turned down and not captured in this study, while previous trials investigating cardiogenic shock included all shock patients, both surgical and non-surgical candidates16,17. When evaluating cardiogenic shock patients for surgical intervention, we have identified several risk factors independently associated with increased hazards for mortality. This include advanced age, peripheral vascular disease, and renal failure requiring dialysis, all risk factors associated with increased morbidity and mortality in the non-shock population following cardiac surgery24–26. Careful patient selection for surgical intervention may help reduce short-term mortality and avoidance of futile procedures in non-salvageable patients. As observed in our study, if these high-risk patients are able to survive to hospital discharge, their longer-term outcomes are favorable. Moreover, we observed overall 1- and 5 -year survival of 71.7% and 62.1%. Additionally, heart failure readmission occurred in only 18.9% of our cohort with median readmissions per person of 0 (IQR 0-2 readmissions), and less than 1% of surviving patients required advanced heart failure surgical therapy. Such findings support efforts to perform cardiac surgical procedures in appropriately selected patients presenting with cardiogenic shock.
Each individual patient is different as is the approach that individual surgeon’s and other providers may have with these patients. There are certainly multiple factors which play into the decision-making process regarding surgical candidacy. The general approach our surgical group has is to attempt to assess the patient’s overall life expectancy should they survive surgery, as well as the potential reversibility of the anatomic or physiologic insults causing the shock state. For instance, a 90-year old presenting in cardiogenic shock on multiple vasoactive agents with marginal urine output will likely have poor outcomes. Similarly, a patient with significant comorbidity burden such as chronic lung disease with home oxygen dependence, dialysis, and peripheral arterial disease requiring extensive multi-component cardiac surgery may have poor outcomes and limited life expectancy even if they survive surgery.