Prasad, Amit

AbstractBackgroundVeno‐arterial extracorporeal life support (V‐A ECLS) has become a cornerstone in the management of critical cardiogenic shock, but it can also precipitate organ injury, e.g., acute kidney injury (AKI). Available studies highlight the effect of non‐cardiac organ injury on patient outcomes. Only very little is known about the impact of non‐cardiac organ recovery on patient survival. AKI occurs frequently during cardiogenic shock and carries a poor prognosis. We have developed descriptive models to hypothesize on the role of AKI severity versus that of recovery of renal function for patient survival.MethodsRetrospective, observational study including 175 patients who were successfully decannulated from V‐A ECLS. We assessed AKI severity using the “Kidney Disease: Improving Global Outcomes” (KDIGO) criteria. We defined recovered or preserved renal function (RPRF) prior to decannulation from V‐A ECLS as 0 (AKI with no improvement) or 1 (no AKI or AKI with improvement). We classified patient outcomes as alive or dead at hospital discharge.Results78% (n = 138) of all patients survived hospital discharge of which 38% (n = 67) never developed AKI. After adjusting for shock severity and non‐renal organ injury, RPRF emerged as an independent predictor of survival in both the overall cohort [OR (95% CI) – 4.11 (1.72–9.79)] and the AKI‐only sub‐cohort [OR (95% CI) – 5.18 (1.8–14.92)]. Neither maximum KDIGO stage nor KDIGO stage at the end of V‐A ECLS was independently associated with survival.ConclusionsOur model identifies RPRF, but not AKI severity, as an independent predictor of hospital survival in patients undergoing V‐A ECLS for cardiogenic shock. We hypothesize that recovered or preserved non‐cardiac organ function during V‐A ECLS is crucial for patient survival.

Abstract Peripheral veno‐arterial extracorporeal membrane oxygenation (pVA‐ECMO) has gained increasing value in the management of patients with critical cardiogenic shock (cCS), allowing time for myocardial recovery. Failure of myocardial recovery has life‐altering consequences: transition to durable mechanical circulatory support (dMCS), urgent heart transplantation, or withdrawal of support. Clinical factors controlling myocardial recovery under these circumstances remain largely unknown. Using a retrospective cohort, we developed a model for early prediction of transition to dMCS in patients undergoing pVA‐ECMO for cCS. To promote myocardial recovery, our clinical management centered around left ventricular pressure unloading, that is, targeting pulmonary capillary wedge pressures (PCWP) ≤18 mm Hg. We collected demographic data, laboratory findings, inotrope use, and two‐dimensional transthoracic echocardiography measurements, all limited to the first 72h of pVA‐ECMO (D1‐3). Out of 70 patients who were alive after pVA‐ECMO, 27 patients underwent implantation of dMCS. There was no significant difference in survival to hospital discharge between patients with or without transition to dMCS. Ejection fraction D1‐3 (per 10% increase, OR 0.37 [0.17‐0.79]) and amount of inotropic support D1‐3 (OR 4.77 [1.6‐14.18]) but neither myocardial wall tension nor PCWP emerged as significant predictors of transition to dMCS. Optimism‐corrected c‐index (0.90 [0.89‐0.90]) revealed an excellent discriminative ability of our model. In summary, our model for early prediction of transition to dMCS in patients with cCS undergoing pVA‐ECMO identifies indicators of inotropic state as relevant factors. Absence of markers for myocardial oxygen consumption or left ventricular pressure loading allows us to hypothesize sufficient cardiac unloading in our cohort with PCWP‐targeted management.

AbstractVeno‐arterial extracorporeal membrane oxygenation (VA‐ECMO) provides mechanical circulatory support for patients with advanced cardiogenic shock, facilitating myocardial recovery and limiting multi‐organ failure. In patients with severely limited left ventricular ejection, peripheral VA‐ECMO can further increase left ventricular and left atrial pressures (LAP). Failure to decompress the left heart under these circumstances can result in pulmonary edema and upper body hypoxemia, that is, myocardial and cerebral ischemia. Atrial septostomy can decrease LAP in these situations. However, the effects of atrial septostomy on upper body oxygenation remain unknown. After IRB approval, we identified 9 out of 242 adult VA‐ECMO patients between January 2011 and June 2016 who also underwent atrial septostomy for refractory pulmonary edema/upper body hypoxemia. We analyzed LAP/pulmonary capillary wedge pressure (PCWP), right atrial pressures (RAPs), PaO2/FiO2 ratios (blood samples from right radial artery), intrathoracic volume status, and resolution of pulmonary edema before and up to 48 h after septostomy. There were no procedure‐related complications. Thirty‐day survival was 44%. LAP/PCWP decreased by approximately 40% immediately following septostomy and remained so for at least 24 h. PaO2/FiO2 ratios significantly increased from 0.49 (0.38–2.12) before to 5.35 (3.01–7.69) immediately after septostomy and continued so for 24 h, 6.6 (4.49–10.93). Radiographic measurements also indicated a significant improvement in thoracic intravascular volume status after atrial septostomy. Atrial septostomy reduces LAP and improves upper body oxygenation and intrathoracic vascular volume status in patients developing severe refractory pulmonary edema while undergoing peripheral VA‐ECMO. Atrial septostomy therefore appears safe and suitable to reduce the risk of upper body ischemia under these circumstances.