| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2009;9:296-300. doi:10.1510/icvts.2008.197681 © 2009 European Association of Cardio-Thoracic Surgery
Extracorporeal membrane oxygenation for treatment of cardiac failure in adult patientsDepartment of Cardiovascular Surgery, Fu Wai Hospital, Peking Union Medical College, No.167, Beilishi Road, Beijing, 100037, P.R. China Received 31 October 2008; received in revised form 21 March 2009; accepted 24 March 2009
*Corresponding author. Tel.: +86-010-68314466.
This report reviews our experience in venoarterial extracorporeal membrane oxygenation (ECMO) support treatment in adult patients with cardiac failure, as well as analysis of the risk factors of early mortality. From February 2005 to June 2008, 45 patients undergoing cardiogenic shock required temporary ECMO support. They were divided into three groups: post-cardiotomy (n=31) and post-transplantation (n=5) heart failure, decompensated heart failure (n=9). ECMO implantation was performed through the femoral vessels, or axillary artery, or through the right atrium and ascending aorta. Average support duration was 126.7±104.3 h. Twenty-seven patients could be successfully weaned from support (60%); additionally, five were bridged to heart transplantation. The in-hospital mortality was 42% (19/45). Twenty-six patients (58%) could be successfully discharged. Additional intra-aortic balloon pumps were used in 11 patients, and six of them were successfully discharged. The mortality rate was obviously high for patients with acute renal failure treated by continuous renal replacement therapy (CRRT) under ECMO support (7/9 patients). The dominant mode of death was multisystem organ failure (9/19). ECMO offers effective cardiopulmonary support in adults. The better outcome requires a multidisciplinary approach to prevent complications unique to itself and limit organ injury before and during this support.
Key Words: Heart failure; Extracorporeal membrane oxygenation
Extracorporeal membrane oxygenation (ECMO) circuits have been introduced for treatment of cardiogenic shock in adult patients since the 1990s and have been shown to provide excellent oxygenation and hemodynamic support [1, 2]. The distinct advantage of ECMO is that it can be rapidly deployed in an emergency situation for biventricular support. However, it is difficult to predict outcome in adult patients supported by ECMO. This report reviews our experience in 45 adult patients placed on venoarterial ECMO (VA-ECMO) to treat cardiac failure. The analysis was undertaken in order to identify possible risk factors of the early mortality for ECMO support.
From February 2005 to June 2008, 45 adult patients with heart failure were placed on VA-ECMO using heparin-coated circuit (Medtronics Inc, Minneapolis, MN, USA). The majority of the cases were post-cardiotomy (n=31) and post-transplantation (n=5) heart failure, whereas the remaining patients suffered from decompensated heart failure (n=9) due to cardiomyopathy or post-infarction ischemic heart disease (Table 1).
The criterion to use ECMO was severe heart failure not responding to conventional therapy. Inclusion criteria were mean arterial pressure (MAP) <60 mmHg, left atrial pressure (LAP) >20 mmHg, cardiac index (CI) <2 l·min–1· m–2 and systemic vascular resistance index (SVRI) >2400 dyn·s·cm–5·m–2, despite inotropy, afterload reduction or the use of IABP. For post-cardiotomy and post-transplantation patients, failure to wean from bypass on inotropic agents or IABP support prompted ECMO insertion. Arterial and venous access for ECMO support was obtained by using the femoral vessels (peripheral cannulation) in the majority of the patients (40 cases). In order to optimize distal limb perfusion, a small cannula (size 8F–10F) was inserted distal to the cannulation site and was connected to the sidearm of the arterial inflow. Because of the severe sclerosis of the femoral artery, the retrograde femoral arterial cannulation in one female patient had to be diverted to the axillary artery to achieve adequate ECMO support. In four patients (8.9%), the ECMO circuit was instituted by means of central cannulation (ascending aorta and right atrium), with sternal incision not closed during the support period. For modification, axillary artery and right atrium cannulation was used in one patient recently, so that the sternal incision could be closed routinely. No left atrial or ventricular drainage was used in these 45 patients. Initially, pump flow were chosen to supply at least adequate total systemic circulatory support (2.5 l·min–1·m–2). With the condition improved, it was gradually adjusted to 40 ml·kg–1·min–1 to achieve a mixed oxygen saturation of 70%. A continuous cardiac output was monitored with a Swan-Ganz catheter. Inotropic support was reduced (but not withdrawn) to decrease myocardial oxygen demand and facilitate myocardial recovery. Low tidal volume (5–6 ml/kg tidal breaths) mechanical ventilation with biphasic positive airway pressure was applied.Heparin was used to ensure an activated clotting time (ACT) of approximately 150–160 s. For postcardiotomy patients, in order to maintain sufficient hemostasis to prevent further bleeding complications, correction of the ACT to the desired level by means of heparin was not considered until the chest drainage was <100 ml/h for 5–6 h. Echocardiography was routinely performed multiple times during the support period to provide important information for assessment of myocardial recovery. When recovery was unlikely, patients were evaluated for transplantation. These patients were <65 years old and had failed to demonstrate adequate improvement in ventricular function for weaning from the ECMO circuit. If cardiac and lung function significantly improved, patients were weaned from ECMO as soon as possible. Weaning was achieved over a period of several hours. ACT was adjusted to 180 s when flow was reduced. Hemodynamic situation was continuously monitored during the process. All the patients on ECMO support were registered in the department's ECMO register. Additional data were obtained from the patient records. Eighteen possible risk factors were compared between the survivor and non-survivor groups.In an effort to provide a quantified index of the patient's condition, the dosages of inotropic infusions were used to derive an inotrope score (IS). IS was calculated as previously described [3]: IS (µg/kg/min)=dopamine+dobutamine +15 xmilrinone+100xepinephrine+100xnorepinephrine+100xisoprotenolol. All surviving patients were followed-up after hospital discharge by telephone to assess their clinic status and exercise capacity. Continuous variables were presented as means±S.D., whereas discrete variables were presented as frequencies. SPSS 11.5 (SPSS Inc, Chicago, Illinois, USA) software was used for analysis. The 2 or Fisher-exact test, whenever applicable, was used for discrete variables, with estimation of odds ratio and 95% confidence interval. The unpaired t-test was used to compare continuous variables. A P-value of <0.05 was regarded as statistically significant.
The average patient age was 49.0±14.1 years (range, 18–76 years). There were 34 male patients (76%). Fifteen patients (33%) had ECMO initiated in the operating room because of the inability to wean from cardiopulmonary bypass. For the other 30 patients, the ECMO was instituted in the intensive care unit, including 14 patients who suffered from cardiac arrest prior to ECMO. Clinic features of the studied patients are summarized in Table 1. The mean duration of ECMO support was 126.7±104.3 h (range, 5–648 h). Of the 45 patients, 27 (60%) could be successfully weaned from ECMO; additionally, five were bridged to heart transplantation, while 13 patients died on ECMO. Twenty-six patients (58%) were discharged alive. Nineteen patients died in-hospital (42%). In patients who could not be weaned from ECMO, the main causes of death were persistent heart failure without improvement (6/13) and severe multiorgan failure (MOF) (4/13). The main cause of mortality in patients after successful weaning from ECMO was sepsis with consecutive MOF (5/6). The outcomes of ECMO support for the different groups are also demonstrated in Table 1. For 31 postcardiotomy patients, the supporting results are listed on different operative procedure in Table 2. The best support outcomes were in coronary artery bypass graft (CABG) and correction of congenital heart disease. IABP was combining used with ECMO support in 11 patients, all in postcardiotomy group. The statistical analysis did not indicate that the combining using IABP could influence survival in this group (P=0.80).
Complications developed frequently in patients on ECMO. Table 3 lists the overall complications in this group. Rethoracotomy for bleeding or tamponade was the most common complication and had to be performed in 12 (27%) patients; furthermore, half of them suffered rethoracotomy more than one time. The rate of infection, defined as a positive sputum or blood culture, was 38% (17/45). Acute renal failure required continuous venovenous hemofiltration (CVVH) which occurred in 12 patients. The CVVH treatment was combining used with ECMO in nine patients and had to be used days after successful weaning from ECMO in another three patients. Although the distal limb perfusion was optimized in this group, limb ischemia was observed in three patients and thrombectomy had to be performed. Fasciotomy of the lower leg was also required in two patients because of compartment syndrome.
We failed to detect statistically significant differences between survivors and non-survivors regarding age, gender, weight, duration of the support and mean blood pressure prior to ECMO. There were no conclusive results regarding the maximum values during the support period of aspartate aminotransferase, blood urea nitrogen or bilirubin. However, combining using CRRT during the ECMO support was associated with a significantly higher mortality rate (P=0.024), odds ratio for mortality (95% confidence interval) was 4.8 (1.1–20.5). Suffering cardiac arrest prior to ECMO also influenced the survival in this group (P=0.011), odds ratio for mortality (95% confidence interval) was 3.4 (1.3–9.3). In survivor group, the ECMO support blood flow, the maximum value of serum creatinine and the IS after ECMO established for 3 h were statistically significant, lower than that of non-survivor group (Table 4).
All of the discharged patients were reassessed (mean follow-up 15.6 months). Three patients died during the follow-up period because of refractory heart failure; another one died due to neurologic complications. The other 22 survivors are in good condition with cardiac symptoms of New York Heart Association class I or II.
Since the first successful implantation by Hill and colleagues in 1972 [4], ECMO had expanded its utility greatly for life support in the critically ill patient. Although the initial results of ECMO with adult respiratory distress syndrome showed poor outcome [5], several subsequent studies have reported the acceptable results of ECMO for temporary circulatory support in patients with cardiopulmonary failure [1, 2, 6]. In cardiogenic shock patients, recent studies suggest that the rate of discharged alive with the use of ECMO vary from 24% to 63% [1, 2, 7–11]. Smith and colleagues [9] reported a 41% overall survival in patients with a mean age of 69 years. Importantly, all hospital survivors remained alive at a median of 21 months follow-up, and the majority reports a satisfactory quality of life. Bavaria and colleagues [12] demonstrated that ECMO support could increase the afterload of left ventricle. Some surgeons think that combining the use of IABP may reduce this negative influence. Additionally, IABP can afford pulsatile blood flow and better coronary flow. Several studies indicated that the use of IABP was a predictor for better survival [7, 8]. And they proposed that IABP should be routinely used in patients undergoing ECMO. But we failed to detect the significant difference between survivor and non-survivor regarding the combination use of IABP during ECMO support. This may be due to our limited cases. Only 24% of our patients combining used IABP. In order to deal with the possible increased afterload caused by ECMO, we kept the patients on inotropic support and adjusted the ECMO blood flow rate in time to maintain ventricular ejection during this period. In Wagner's 80 cases report [10], there were 12 patients who suffered cardiac arrest and were resuscitated prior to ECMO. Their analysis showed that cardiac arrest did not influence the survival. That was in contrast to our group. Thirty-one per cent (14/45) patients in our group suffered cardiac arrest prior to ECMO, and 10 patients died in hospital. There were statistical differences between survivor and non-survivor (P=0.011). Different from Wagner's report, in our group, there were six patients who suffered cardiac arrest who could not be resuscitated successfully until ECMO was instituted. These could be defined as ECPR (extracorporeal cardiopulmonary resuscitation). As other literature indicated [13], the primary goal of ECMO is to stabilize hemodynamics, restore oxygen delivery to promote healing before reperfusion injury becomes a factor. However, the primary goal of ECPR is to address the reperfusion injury issue while preventing the patient's immediate death. Both use a blood pump and oxygenator, but the strategy for support and perfusion in ECPR needs to be further examined. Mild renal function impairment and antidiuresis are common during ECMO support. In our survivor group, the mean serum creatinine level also rose to 164.5±77.3 mg/dl. But the acute renal failure (oliguric to anuric type) during the support period is associated with poor outcome. Our investigation suggested that combining use of CRRT and ECMO might be a risk factor for survival. This finding resembles another study [14]. The acute renal failure is a manifestation of multiple organ system failure due to acute decompensated heart failure, sepsis, aggravated by hemolysis, and activation of complement system during ECMO support. So it is important to accept more aggressive and early initiation of ECMO support for advanced cardiac failure before other important organ failure develops. The timely recognition and treatment of cardiogenic shock are crucial in reducing the incidence of death. Bleeding continue to be a major complication of ECMO support. Heparin-coated circuits and the delay of systemic heparinization have been recommended. But still, some patients experienced rethoracotomy for bleeding during the postoperative course. Although the distal limb perfusion was optimized in this group, limb ischemia was still observed. Distal limb perfusion by using Dacron T-graft (end-to-side anastomosis to the femoral artery) could be a better method to solve this problem [11]. In conclusion, this study proposes that VA-ECMO is a simple and effective technique for temporary, complete circulatory support in cardiogenic shock. It can be rapidly initiated and can reverse ischemia and anoxia in time. The better outcome of ECMO support requires a multidisciplinary approach to prevent complications unique to itself and limit organ injury before and during this support.
Related Article
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Introduction
