If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Pulmonary edema is very common at intensive care unit admission in patients undergoing ECMO support after cardiovascular surgery.
Pulmonary edema three and five days after ECMO implantation are associated with poor survival.
ECMO rotation was associated with the evolution of pulmonary congestion.
Venoarterial-extracorporeal membrane oxygenation (VA-ECMO) is a life-saving method for patients with low-output failure after cardiac surgery. However, VA-ECMO therapy may increase left ventricular afterload due to retrograde blood flow in the aorta, which may lead to progression of pulmonary congestion. We examined the predictive value of pulmonary congestion in patients that need VA-ECMO support after cardiovascular surgery.
We enrolled a total of 266 adult patients undergoing VA-ECMO support following cardiovascular surgery at a university-affiliated tertiary care centre into our single-center registry. Pulmonary edema was assessed on bedside chest X rays at day 0, 3, 5 after VA-ECMO implantation.
Median age was 65 (57–72) years, 69% of patients were male and 30-day survival was 63%. At ICU-admission 20% of patients had mild, 54% had moderate and 26% showed severe pulmonary congestion. Pulmonary congestion at day 0 was not associated with outcome (adjusted HR 1.31; 95%-CI 0.89–1.93;P = 0.18), whereas pulmonary congestion at day 3 (adj. HR 2.81; 95%-CI 1.76–4.46;P<0.001) and day 5 (adj. HR 3.01;95%-CI 1.84–4.93;P<0.001) was significantly associated with survival. Linear regression revealed that out of left ventricular function, cardiac output, central venous saturation, maximum dobutamine and norepinephrine dose as well as fluid balance solely ECMO rotation was associated with the evolution of pulmonary congestion (P = 0.007).
Pulmonary edema three and five days after ECMO implantation are associated with poor survival. Interestingly, a high VA-ECMO output was the most important determinant of worsening pulmonary congestion within the first five days.
Venoarterial-extracorporeal membrane oxygenation (VA-ECMO) is a life-saving method for patients with low-output failure after cardiac surgery. Although the use of VA-ECMO after cardiac surgery is rapidly evolving, in-hospital mortality remains high [
]. VA-ECMO therapy efficiently improves organ perfusion and oxygenation. However, the advantages of VA-ECMO for peripheral perfusion are associated with drawbacks on cardiac hemodynamics and coronary perfusion. VA-ECMO support increases left ventricular afterload due to retrograde blood flow in the aorta towards the left ventricle, resulting in further increased left ventricular filling pressures and distension of the left ventricle on top of the already high pressures and volume of the failing left ventricle. This results in increased left atrial pressures, which further leads to pulmonary edema [
]. Recruitment was conducted at the Vienna General Hospital, a university-affiliated tertiary centre. We included 266 adult patients undergoing venoarterial ECMO support following cardiovascular surgery. Primary endpoint was 30-day survival. Mortality data were obtained by screening the national registry of death including screening for the cause of death. The study protocol was reviewed and approved by the Ethics Committee of the Medical University of Vienna and conforms to the Declaration of Helsinki.
Indications for the initiation of ECMO support therapy included clinical signs of cardiogenic shock such as systolic arterial hypotension and signs of end organ failure, anaerobic metabolism and metabolic acidosis despite optimized supportive measures. Patients that underwent ECMO support at time of left ventricular assist device implantation were not included into the study. VA cannulation was implemented in the operating room following CV surgery. The ECMO circuit combined a centrifugal pump console (Bio-Console 560; Medtronic, Minneapolis, MN, USA or CARDIOHELP system, Maquet, Germany) with a membrane oxygenator (Affinity-NTTM; Medtronic, Minneapolis, MN, USA or HLS Set Advanced, Maquet, Germany). All components of the extracorporeal oxygenation system were coated with heparin.
2.2 Radiological assessment
Two experienced radiologists evaluated bedside chest X rays of all ICU patients. At day 0, 3, 5 after ECMO implantation, pulmonary edema was assessed. The radiologists were blinded to the patient history and clinical data, which could influence the evaluation of the imaging findings. PC was graded as mild, moderate and severe as well as the presence of total bilateral consolidation based on several characteristics previously described that included the evaluation of hilar vessels (blurring, enlargement, density, vascular pedicle width), cardiothoracic ratio, evidence of peribronchial cuffs, Kerley B lines or diffuse hyperlucency, respectively [
]. In our cohort, interclass correlation coefficient for intra-observer variability was 0.96 95%CI 0.93–0.98 and inter-observer variability was 0.98 95%CI 0.97–0.99).
2.3 Statistical methods
Baseline characteristics were compared using Kruskal–Wallis test and χ2 test, as appropriate. Cox proportional hazard regression analysis was applied to assess the effect of PC on survival. Results were expressed as hazard ratio (HR) for an increase of one category with the respective 95% confidence intervals (95% CI). To account for potential confounders we adjusted for SAPS-3 score, ECMO rotation, fluid balance, ECMO cannulation site, use of intra-aortic balloon pump (IABP) and type of cardiovascular surgery. Kaplan-Meier analysis was applied to evaluate the effect of PC change on survival and compared using log-rank test. Linear regression analysis was performed to identify factors related to the evolution of PC. Two-sided P-values of <0.05 indicated statistical significance. SPSS 18.0 (IBM Corporation, Armonk, NY, USA) was used for all analyses.
Median age was 65 (57–72) years, 69% of patients were male and 30-day survival was 63%. The median SAPS-3 score and the median EuroSCORE of the study population were 43 (37–52) and 10 (8–13), respectively. VA-ECMO support was initiated after valve surgery (31%), coronary artery bypass graft (CABG) surgery (13%), combined CABG-valve surgery (27%), heart transplant (19%) and other cardiovascular surgeries (10%). Detailed baseline characteristics for the study population are given in Table 1.
Table 1Baseline characteristics of ECMO study population (n = 266) stratified according course of pulmonary congestion.
At ICU-admission 20% of patients had mild, 54% had moderate and 26% showed severe PC. PC at day 0 was not associated with outcome (crude HR 1.17, 95%CI 0.77–1.77, P = 0.48; adjusted HR 1.38;95%-CI 0.71–2.69;P = 0.34), whereas pulmonary congestion at day 3 (crude HR 2.45, 95%CI 1.59–3.81, P<0.001; adjusted HR 3.71;95%-CI 1.84–7.45;P<0.001) and day 5 (crude HR 2.67, 95%CI 1.69–4.23, P<0.001; adjusted HR 4.62; 95%-CI 2.17–9.80;P<0.001) was significantly associated with survival, respectively. In particular, an increase of PC between admission and day 3 (Fig. 1A) was associated with poor prognosis (crude HR 1.84, 95%CI 1.12–3.03, P = 0.016; adjusted HR 4.48;95%-CI 2.08–9.64, P<0.001).
Linear regression revealed that out of left ventricular function, cardiac output, central venous saturation, maximum dobutamine and norepinephrine dose as well as fluid balance, solely ECMO rotation was associated with the evolution of pulmonary congestion (P = 0.007). Patients with a decrease of PC between day 0 and day 3 had a median ECMO rotation at baseline of 2800 (2400–3200) rpm, patients with stable pulmonary congestion had a median ECMO rotation of 3000 (2500–3550) rpm and the patients with worsening pulmonary edema had a median ECMO rotation of 3350 (3110–3800; P = 0.02;Fig. 1B). Of note, SAPS-3 score (P = 0.50), EuroSCORE (P = 0.93), baseline left ventricular function (P = 0.06), maximum catecholamine doses as well as fluid balance (P = 0.65) were not different in patients with regression, stable congestion and progression of pulmonary edema, respectively
This study shows a strong and independent association between PC and mortality after ECMO support following cardiovascular surgery. Interestingly, not baseline PC but pulmonary edema at later time points and in particular the course of PC over time was associated with survival. This strong association persists even after adjustment for SAPS-3 score, type of cardiovascular surgery, fluid balance, ECMO cannulation site and rotation and use of intra-aortic balloon pump (IABP).
Pulmonary edema was very common at ICU admission in this patient group. This could have been caused, at least in part, by positive fluid balance during cardiovascular surgery. However, PC at admission was not associated with outcome. In contrast, an increase in pulmonary edema during VA-ECMO therapy was strongly associated with mortality. PC may be associated with decreased systemic and coronary oxygen saturation leading to prolongation of VA-ECMO support and deterioration of myocardial recovery resulting in increased duration of VA-ECMO therapy that is associated with occurrence of major complications and a decreased survival [
]. From a mechanistic point of view, our data demonstrates for the first time the interdependence of VA-ECMO output with increase of pulmonary congestion and worsening of outcome. Only VA-ECMO rotation was associated with worsening pulmonary edema. Interestingly, left ventricular function, fluid balance and vasopressor support were not associated with the evolution of PC suggesting that the most important factor is the increased afterload due to VA-ECMO support.
Decongestion strategies like percutaneous LV drainage, trans-septal venting, Impella device or switching to central ECMO cannulation are complex and costly [
]. As Impella devices were not used at our institution during the study, conclusions about the effect of impella on the evolution of pulmonary congestion and clinical outcome cannot be drawn. However, recently, routine use of IABP or Impella after implantation of VA-ECMO was found associated with decreased pulmonary congestion and improved ECMO weaning [
]. A prospective randomized clinical trial is currently investigating whether the addition of early direct ventricular unloading using Impella decreases pulmonary congestion and leads to improved survival in cardiogenic shock patients undergoing ECMO support (NCT03431467).
In conclusion, pulmonary edema three and five days after ECMO implantation are associated with survival. Interestingly, a high VA-ECMO output was the most important determinant of worsening pulmonary congestion within the first five days. Our observation should nurture the intensivist to constantly aim for the lowest ECMO flow needed to preserve adequate organ perfusion without an excess increase of left ventricular afterload. Whether the adverse pulmonary effects of VA-ECMO might be essentially counteracted with the simultaneous use of intravascular microaxial left ventricular assist devices has to be analyzed by future studies.
This research received no specific grant from any funding agency.
Declaration of Competing Interest
All authors declare that they have no conflicting interests.
Duration of extracorporeal membrane oxygenation support and survival in cardiovascular surgery patients.