Age and Preoperative Total Bilirubin Level Can Stratify Prognosis After Extracorporeal Pulsatile Left Ventricular Assist Device Implantation

Transcription

1 Circulation Journal Official Journal of the Japanese Circulation Society ORIGINAL ARTICLE Cardiovascular Surgery Age and Preoperative Total Bilirubin Level Can Stratify Prognosis After Extracorporeal Pulsatile Left Ventricular Assist Device Implantation Taro Shiga, MD, PhD; Koichiro Kinugawa, MD, PhD; Masaru Hatano, MD; Atsushi Yao, MD, PhD; Takashi Nishimura, MD, PhD; Miyoko Endo; Naoko Kato, PhD; Yasunobu Hirata, MD, PhD; Shunei Kyo, MD, PhD; Minoru Ono, MD, PhD; Ryozo Nagai, MD, PhD Background: In Japan, the TOYOBO left ventricular assist device (LVAD) has been commercially available for heart failure patients as of 2010, but clinical risk stratification before implantation has not been widely performed. Methods and Results: In the present study data from 47 patients (age 38.6±14.6 [SD] years, male 74.5%, nonischemic 74.5%) implanted with a TOYOBO LVAD between November 2002 and February 2010 were analyzed. Kaplan-Meier survival analysis showed significantly higher mortality in the patients who had cardiogenic shock preoperatively (P=0.031). Multivariate analysis revealed that the preoperative total bilirubin level (odds ratio [OR] 1.312, P<0.001) and age (OR 1.076, P=0.013) were independent risk factors for death. Perioperative necessity of a right ventricular assist device was also an independent risk factor for poor prognosis. Conclusions: LVAD implantation is preferable before the patient experiences hemodynamic collapse. The preoperative total bilirubin level can be used to predict prognosis after device implantation in end-stage heart failure patients. (Circ J 2011; 75: ) Key Words: Heart failure; Heart-assist device; Hemodynamics; Risk factor; Transplantation Needless to say, heart transplantation is the only complete solution for stage D heart failure (HF), which is refractory and requires specialized interventions, 1 but even in the USA where there are more than 2,000 transplantation cases per year, 2 not all stage D patients are able to undergo transplantation immediately and most of them have to wait. Successful bridge-to-transplantation (BTT) results in a good prognosis after heart transplantation, but the longer the waiting period, the harder it is to maintain stable hemodynamics. Temporary mechanical support, such as an intraaortic balloon pump (IABP) or percutaneous cardiopulmonary support (PCPS), is usually available for less than 2 weeks. To secure the minimally required cardiac output over months or sometimes years, the best technique is implantation of a (left) ventricular assist device ([L]VAD). Although the worldwide trend is definitely toward implantable continuous flow VADs, 3,4 the situation in Japan has hardly changed since the early 1990s. The extracorporeal pulsatile TOYOBO VAD 5 was approved by the national government in 1990, but it is still the only device available as of August Most patients implanted with the TOYOBO VAD have to be in hospital for more than 2 years because of prohibited usage of this device outside of hospitals and the extreme shortage of donor hearts in Japan. Editorial p 53 Currently, we run a heart transplantation program as one of the government-certified institutes in Japan, and since 2002 not a few stage D patients have been referred, many of whom have had a TOYOBO VAD implanted. 6 To determine the best indication for a VAD, the analysis of risk factors affecting prognosis should be indispensable for VAD candidates, 7 11 but because of the unusual VAD circumstances in Japan, the published data from the US and Europe for BTT are not applicable to the TOYOBO VAD because the waiting period before transplantation is normally 2 3 months in those reports. 3,10,12 17 Since 2002, VAD implantation has also been used in the US as terminal care for end-stage HF patients who are not eligible for transplantation; that is, destination therapy Received August 4, 2010; revised manuscript received August 25, 2010; accepted August 30, 2010; released online November 23, 2010 Time for primary review: 12 days Departments of Cardiovascular Medicine (T.S., K.K., M.H., A.Y., N.K., Y.H., R.N.), Cardiothoracic Surgery (M.O.), Organ Transplantation (M.E.) and Therapeutic Strategy for Heart Failure (T.N., S.K.), Graduate School of Medicine, University of Tokyo, Tokyo, Japan This paper was presented in part at the 73 rd Annual Scientific Meeting of the Japanese Circulation Society (Osaka, Japan, 2008) and at the 13 th Annual Scientific Meeting of the Heart Failure Society of America (Boston, MA, USA, 2009). Mailing address: Koichiro Kinugawa, MD, PhD, Department of Cardiovascular Medicine, The University of Tokyo Hospital, Hongo, Bunkyo-ku, Tokyo , Japan. kinugawa-tky@umin.ac.jp ISSN doi: /circj.CJ All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp

2 122 SHIGA T et al. Table 1. Baseline Characteristics of the Total Patient Group n 47 Age, years 38.6±14.6 Male, n (%) 35 (74.5) Body surface area, m ±0.24 Systolic blood pressure, mmhg 83.3±12.1 Heart rate, beats/min 103.1±25.1 Etiology of heart failure, n (%) Dilated cardiomyopathy 26 (55.3) Ischemic cardiomyopathy 12 (25.5) Other 9 (19.1) Mean survival after LVAD implantation, days 407.1±314.3 LV ejection fraction,% 21.9±14.3 Serum sodium, meq/l 132.8±6.6 Serum creatinine, mg/dl 1.16±0.71 Serum total protein, g/dl 5.81±0.86 Serum albumin, g/dl 3.23±0.61 Serum total bilirubin, mg/dl 3.90±4.76 Hemoglobin, g/dl 10.5±1.3 Platelet count, 10 4 /ml 13.7±9.1 Plasma B-type natriuretic peptide, pg/ml 1,299.9±1,008.3 C-reactive protein, mg/dl 8.78±8.85 Activated partial thromboplastin time, s 61.5±40.4 Right ventricular assist device, n (%) 8 (17.0) NYHA class IV, n (%) 47 (100) Values are means ± SD when appropriate. Dilated cardiomyopathy included patients in the dilated phase of hypertrophic cardiomyopathy. The other causes of heart failure were restrictive cardiomyopathy (1), fulminant myocarditis (2), postpartum cardiomyopathy (2), valvular heart disease (2), adriamycininduced cardiomyopathy (1) and congenital heart disease (1). LVAD, left ventricular assist device; LV, left ventricular; NYHA, New York Heart Association. (DT). 18 DT data are based on a longer observation period, similar to that with the TOYOBO VAD, but DT patients are significantly older and have more impaired end-organ function than patients who are eligible for transplantation. 4,18,19 Thus, data associated with the use of the TOYOBO VAD would be unique, because we can observe a remote prognosis in BTT patients but there is little solid evidence for the TOYOBO VAD, which leads to debate on the indications for implantation. Therefore, in this study, we examined the survival rate of patients implanted with a TOYOBO VAD from 2002 to 2010 and which preoperative factors were responsible for the worse prognosis. Methods Patient Population We researched 51 consecutive patients with end-stage HF who were treated with a TOYOBO VAD between November 2002 and February 2010 and followed at the University of Tokyo Hospital. We excluded 4 patients whose data were insufficient for analysis, so finally we analyzed the data for 47 patients (age 38.6±14.6 years, male 74.5%, non-ischemic HF 74.5%). All the patients were treated with general optimal medical therapy (digoxin, diuretics, β-blockers, angiotensin-converting enzyme inhibitors unless intolerant) before TOYOBO VAD implantation. Patients were treated with cardiac resynchronization therapy, if indicated. We usually decided to implant a TOYOBO VAD if the patient developed cardiogenic shock or there was progressive decline of end-organ function because of low cardiac output despite treatment with intravenous inotropic agents. Support with PCPS, IABP, and/or continuous hemodiafiltration was introduced if necessary. Every patient was considered as eligible for transplantation, at least at the time of TOYOBO VAD implantation. Written informed consent was given before implantation. The study protocol was approved by the Ethics Committee of Graduate School of Medicine, the University of Tokyo [application number 779 (1)]. Some patients required a right VAD (RVAD) in addition to a LVAD perioperatively (ie, a biventricular assist device [BiVAD]). RVAD therapy included an extracorporeal membranous oxygenation system and/or the TOYOBO VAD inserted in the right ventricle or atrium. A retrospective analysis of the preoperative clinical, echocardiographic, laboratory, and hemodynamic data was performed to determine the patients clinical characteristics and to determine the risk factors for death after TOYOBO VAD implantation. Three endpoints were defined: death, explantation of TOYOBO VAD, alive with TOYOBO VAD. Definition of Patients Profiles We classified each patient according to their hemodynamic status at the time of TOYOBO VAD implantation (ie, profile 1 [critical cardiogenic shock] and profile 2 [progressive decline with a use of a continuous intravenous inotrope]), based on the INTERMACS (Interagency Registry for Mechanically Assisted Circulatory support: classification. Patients supported by PCPS were all assigned to profile 1, as were patients who were dependent on IABP. We sometimes inserted a IABP in patients without cardiogenic shock in order to maintain stable hemodynamics until the scheduled VAD implantation and these patients were assigned to profile 2. Stratification by the Level of Total Bilirubin (Tbil) We stratified patients by the level of serum Tbil to evaluate the relationship between death and the extent of liver damage. We set 3 levels of Tbil ( 2.0 mg/dl, mg/dl, and 5.0 mg/dl) because Tbil >2.0 mg/dl has been reported as an independent risk factor for postoperative development of right ventricular HF. 11 Statistical Analysis We performed a statistical analysis with PASW Statistics 18 (SPSS Inc, Chicago, IL, USA). Continuous variables are expressed as means ± SD and compared using unpaired t-tests. Categorical variables are expressed as percentages and compared using the χ2 test. Kaplan-Meier analysis was used to estimate survival and survival between groups was compared by log-rank test. Univariate and multivariate analyses with a Cox hazard model were performed to calculate the adjusted odds ratio (OR) and 95% confidence interval (CI) to assess the association between death after TOYOBO VAD implantation and each clinical variable. Variables that were found to be significant (P<0.05) in the univariate analysis were entered into the multivariate analysis. If continuous and categorical variables were both significant in the univariate analysis, we chose the continuous one. Probability was 2-tailed, with P<0.05 regarded as statistically significant. Results Patients Baseline Characteristics We enrolled 47 patients who were treated with a TOYOBO

3 Risk Stratification for Extracorporeal LVAD 123 Table 2. Comparison of Patients Characteristics in Profile 1 and Profile 2 Profile 1 Profile 2 (n=33) (n=14) P value Age, years 38.3± ± Male, n (%) 26 (78.8) 9 (64.3) Body surface area, m ± ± Ischemic cause of heart failure, n (%) 11 (33.3) 1 (7.1) Mean survival after LVAD implantation, days 362.7± ± LV ejection fraction,% 24.2± ± * Serum sodium, meq/l 134.1± ± Serum creatinine, mg/dl 1.14± ± Serum total protein, g/dl 5.63± ± * Serum albumin, g/dl 3.09± ± * Serum total bilirubin, mg/dl 4.63± ± * Hemoglobin, g/dl 10.3± ± Platelet count, 10 4 /ml 11.3± ± * Activated partial thromboplastin time, s 71.6± ± * C-reactive protein, mg/dl 9.92± ± B-type natriuretic peptide, pg/ml 1,325.1±1, ,251.5± Percutaneous cardiopulmonary support, n (%) 15 (45.5) 0 (0) 0.002* Intraaortic balloon pumping, n (%) 25 (75.8) 3 (21.4) 0.001* Continuous hemodiafiltration, n (%) 11 (33.3) 0 (0) 0.020* *Significant difference (P<0.05) between profiles 1 and 2. Abbreviations see in Table 1. Figure 1. Kaplan-Meier survival analysis of 47 patients supported by a TOYOBO left ventricular assist device (LVAD). VAD between November 2002 and February 2010 and who could be followed at the University of Tokyo Hospital until the time of their endpoints. The baseline characteristics of all 47 patients are shown in Table 1. All patients had New York Heart Association class IV symptoms and most had non-ischemic cardiomyopathy (74.5%). Mean age was 38.6±14.6 years, and there were 35 males (74.5%). Mean support period with a TOYOBO VAD was 407.1±314.3 days. Mean levels of serum C-reactive protein, plasma B-type natriuretic peptide, serum creatinine and Tbil, and the activated partial thromboplastin time (APTT) were above the normal ranges. Average value of left ventricular ejection fraction by echocardiography, and concentrations of serum sodium, hemoglobin, total protein, and albumin were lower than normal. In the present study, there were 8 patients who needed RVAD therapy (17.0%). The number of patients who were finally transplanted was 12 (25.5%). Other endpoints were: death: 14 (29.8%), explantation of TOYOBO VAD with recovery of cardiac function: 6 (12.8%), and alive with TOYOBO VAD: 15 (31.9%). The baseline characteristics of the 2 profiles are shown in Table 2. Profile 1 was assigned to 33 (70.2%) patients. Mean left ventricle ejection fraction, serum total protein, serum albumin and Tbil levels, platelet count and APTT were sig-

4 124 SHIGA T et al. Figure 2. Comparison of survival between patients with profile 1 (critical cardiogenic shock) or profile 2 (progressive decline with use of continuous intravenous inotropes) by Kaplan-Meier analysis. Survival between profiles 1 and 2 was compared by log-rank test. LVAD, left ventricular assist device. Table 3. Univariate Analysis With a Cox Hazard Model of Risk Factors for Mortality P value OR 95%CI Age 0.049* Age (> 50 years) 0.032* Body surface area Body surface area (< 1.4 m 2 ) Male Ischemic cardiomyopathy Right ventricular assist device 0.009* LV ejection fraction LV ejection fraction (< 20%) LV diastolic diameter LV diastolic diameter (> 65 mm) B-type natriuretic peptide B-type natriuretic peptide (> 1,000 pg/ml) C-reactive protein C-reactive protein (> 10 mg/dl) Serum creatinine Serum creatinine (> 1.2 mg/dl) Serum sodium Serum sodium (< 130 meq/l) Serum total protein Serum total protein (< 6 mg/dl) Serum albumin 0.023* Serum albumin (< 3.0 mg/dl) 0.020* Serum total bilirubin 0.004* Serum total bilirubin (> 5 mg/dl) 0.012* Hemoglobin Hemoglobin (< 10 g/dl) Platelet count Platelet count (< /ml) Activated partial thromboplastin time 0.009* Activated partial thromboplastin time (> 36 s) 0.031* The cut-off level of each categorical variable was decided from the point of view of clinical importance. OR, odds ratio; CI, confidence interval. Other abbreviations see in Table 1. *P<0.05.

5 Risk Stratification for Extracorporeal LVAD 125 Table 4. Multivariate Analysis With a Cox Hazard Model of Risk Factors for Mortality P value OR 95%CI Age 0.013* Right ventricular assist device 0.014* Serum albumin Serum total bilirubin <0.001* Activated partial thromboplastin time 0.040* *P<0.05. Abbreviations see in Table 3. Figure 3. Kaplan-Meier analysis of patients stratified by preoperative serum total bilirubin (Tbil) level (>5 mg/dl, mg/dl and <2.0 mg/dl). Survival among each stratum was compared by log-rank test. *P= LVAD, left ventricular assist device. nificantly different between profiles. Patients Survival The overall survival after TOYOBO VAD implantation was 93.6%, 83.0%, 73.1%, and 67.0% at 1 month, 3 month, 1 year, and 2 years, respectively (Figure 1). The mean survival after TOYOBO VAD implantation was 407.0±314.3 days (range 16 1,132 days). As shown in Figure 2, survival of patients with profile 1 decreased dramatically within the first 3 months (66.3% at 137 days). Patients with profile 2 had a significantly better prognosis (P=0.036) and none of them died within 600 days. Multiple organ failure (42.9%) and sepsis (28.6%) were the main causes of postoperative death after TOYOBO VAD implantation. Notably, all deaths within the first 3 months was from multiple organ failure, except for 1 patient. Univariate and Multivariate Analyses for Death After TOYOBO VAD Implantation Univariate and multivariate analyses of the preoperative risk factors for death after TOYOBO VAD implantation are shown in Tables 3 and 4. In the univariate analysis with a Cox hazard model, we found that the following variables were significantly associated with death: age, serum total protein level, serum albumin level, Tbil level, and APTT. Perioperative need for RVAD was also significantly associated with a poor prognosis. To determine the risk factors associated with death after TOYOBO VAD implantation, we performed a multivariate analysis with a Cox hazard model, enrolling RVAD and total protein, albumin and Tbil levels and APTT. This analysis revealed that age (OR 1.076, 95%CI , P=0.013), RVAD (OR 6.350, 95%CI , P=0.014), Tbil (OR 1.312, 95%CI , P<0.001) and APTT (OR 1.013, 95%CI , P=0.040) were independent risk factors associated with death after TOYOBO VAD implantation. Stratification of Survival by the Level of Tbil Figure 3 depicts the estimated survival curves stratified by the preoperative Tbil level in patients for whom Tbil data were available (>5.0 mg/dl: n=10; mg/dl, n=11; <2.0 mg/dl, n=24). The survival of patients whose serum Tbil level was greater than 5.0 mg/dl was significantly worse compared with the patients whose Tbil was less than 2.0 mg/dl (P=0.008). Comparison of Survival Between LVAD and BiVAD Patients We compared the survival of LVAD and BiVAD patients with Kaplan-Meier analysis (Figure 4) and found that the survival rate of BiVAD patients was significantly decreased in the first 3 month after implantation (50.0% vs. 89.7%: BiVAD vs. LVAD survival rate at the first three month, P=0.003 by log-rank test). Discussion We examined the long-term prognosis and preoperative independent risk factors for death of patients in Japan with a

6 126 SHIGA T et al. Figure 4. Comparison of survival between LVAD and BiVAD patients by Kaplan-Meier analysis. LVAD, left ventricular assist device; BiVAD, biventricular assist device. TOYOBO extracorporeal VAD, and we found that those with cardiogenic shock had a worse prognosis and that patient age and preoperative increases in Tbil were independent risk factors for mortality. Recently, the long-term prognosis for LVAD from the 2 leading institutes in Japan was reported, 20,21 and our results are very consistent with theirs. However, there are several differences in the patients characteristics and analytical methods. The study from the National Cerebral and Cardiovascular Center (NCVC) 21 did not include a preoperative risk analysis, rather a postoperative risk analysis, whereas in the study from Osaka University, 20 the prognosis was analyzed in association with preoperative risk factors, as we did. Both studies included more than 100 patients, which definitely strengthened the data analyses, but varying numbers of patients were enrolled in terms of device type and implantation era. Patients indicated for implantable LVADs usually have stable preoperative hemodynamics compared with those who need TOYOBO VAD implantation. As clearly shown in the NCVC study, 21 management of the TOYOBO VAD has improved remarkably in this century, so it is important to have recent data exclusively from patients with the TOYOBO VAD. Firstly, the average age of our 47 patients implanted with a TOYOBO VAD was 38.6 years, which is much younger than that of LVAD patients included in BTT trials in the USA and Europe. In Japan, LVAD use for DT has not been approved to date, and so all our patients were eligible for heart transplantation at least at the time of LVAD implantation. The Japanese guideline for heart transplantation recommends that recipients are less than 60 years old. 13 In the Japanese studies, the mean age was 40.2 years (Osaka study 20 ) and 33.9 years (a recent era group in the NCVC study 21 ), similar to our study. In contrast, the BTT trials in Western countries included patients over 60 years, and the average age of patients in those trials was usually high (50 55 years). 3,10,12 17 In our study, 74.5% of the patients had a nonischemic cause of HF, which was compatible with the approximately 90% of registered patients in the Japan Organ Transplantation network who have a nonischemic background. The high prevalence of a nonischemic cause is characteristic of Japanese BTT patients, and the Japanese studies included % of nonischemic HF patients. 20,21 In contrast, nonischemic cardiomyopathy was recorded in 51 63% of the BTT patients in the US and European trials. 3,10,12 17 According to a Japanese registry for HF patients (JCARE-CARD), 22 there were less patients with an ischemic background ( 32%) compared with similar registries in Western countries, 23,24 but we consider that the high prevalence of nonischemic cardiomyopathy in Japanese BTT patients is primarily because of the lower age limit for transplantation eligibility. Because of the extreme shortage of donor organs in Japan, the average waiting period after being registered in the Japan Organ Transplantation network is over 800 days and the registered patients are usually on LVAD support during most of that long period. In this study, we observed long-term outcomes in BTT patients with a TOYOBO VAD for over 1 year, which was the standard waiting period for Japanese patients. In contrast, most of the eligible patients can be bridged to transplantation within 1 year after LVAD implantation in the US. For example, in a BTT trial using the HeartMate VE, 14 67% of patients were bridged to transplantation, 4% of devices were removed, and the remaining patients (29%) had died within 1 year. Considering the different circumstances for heart transplantation between Japan and other countries, the results from BTT trials in Western countries are not appropriate references for the long-term prognosis (>1 year) of Japanese patients on LVAD support. On the other hand, outcomes of DT patients can be compared with those of Japanese LVAD patients because of the similarity in the observation period; for example, 1- and 2-year survival rates were ~55% and ~25%, respectively, in the DT trials of the HeartMate VE 18 and XVE, 19 which were both pulsatile as is the TOYOBO VAD. In this study, we observed better 1-year (73.1%) and 2-year (67.0%) survival rates for the TOYOBO VAD, but all patients were in hospital. Japanese studies also report similar results to ours for the long-term prognosis in low-risk patients 20 or in the recent era. 21 The better prognosis in Japanese TOYOBO VAD patients might

7 Risk Stratification for Extracorporeal LVAD 127 be attributable to several reasons: thrombi in the pump are visible because the device is extracorporeal, and anticoagulation therapy can be meticulously adjusted because usage of this device outside of hospitals is prohibited and consequently all patients are in hospital. However, the mean age of patients in those DT trials was years, which was higher than the present patient population. The older population is mainly because DT trials normally include many patients who are ineligible for transplantation because of age. Moreover, the older population in the DT trials inevitably consists of more patients with an ischemic cause of HF (46 66% vs % in BTT), and the patients characteristics are much different from our patients. Consistent with the Osaka study. 20 we found that age is an independent risk factor for death after LVAD implantation, and it is no surprise that our younger patients had a better long-term prognosis than those in the DT trials. In the most recent DT trial for HeartMate II, the 1- and 2-year survival rates were reported to be 68% and 58%, respectively. 4 Considering the mean age of the patients was 62 years in that study, the survival rates were surprisingly good. In a BTT trial for HeartMate II. 3,16 the mean age of the enrolled patients was 50 years, and the 18-month survival was much better (72%) than that of any device previously tested. Because the survivors in the BTT trial included 56% of patients who were successfully bridged to transplantation, the survival rate of 72% was not in fact the long-term prognosis for LVAD support. Nevertheless, it is remarkable to note that more than 20% of eligible patients for transplantation were still on support with the HeartMate II in the USA where the average waiting period before transplantation is approximately 90 days for LVAD patients. This result suggests many patients are comfortable with remaining on the support given by HeartMate II rather than undergoing heart transplantation. HeartMate II is now under clinical trial in Japan, and we anticipate superb outcomes after the introduction of this device. Much information is now available from INTERMACS, which is a large-scale registry for VADs in the US since In this registry, the preoperative condition of patients is classified into 7 profiles: profile 1: cardiogenic shock (eg, in need of PCPS); profile 2: progressively declining with intravenous inotropes; profile 3: stable with intravenous inotropes. Profiles 1 3 are compatible to status 1 in the Japan Organ Transplantation network registry, but TOYOBO VAD is not normally indicated for patients with profile 3 because the patient s quality of life cannot be improved after implantation with an extracorporeal LVAD. Even in the US, 72% of VADs were implanted in patients with profile 1 or 2 (data from INTERMACS). Therefore, it is important to compare the prognosis after LVAD implantation between patients with profile 1 or profile 2. Consistent with the data from the INTERMACS registry, we found that patients with profile 2 had a significantly better prognosis than those with profile 1. Our study demonstrated that death early after implantation accounted for a large part of the overall death rate, suggesting that preoperative factors greatly influenced the survival rate after TOYOBO VAD implantation. Almost all deaths within the 3 months after implantation resulted from multiple organ failure, which occurred only in the patients with profile 1. Severe organ damage caused by poor or collapsed preoperative hemodynamics could not be recovered even after LVAD implantation in many of patients with cardiogenic shock. To explore which factors were responsible for the poor prognosis after TOYOBO VAD implantation, we performed a multivariate analysis, which revealed that age, preoperative Tbil level and APTT were independent risk factors. In particular, the preoperative Tbil level (P<0.001) and age (P=0.013) were strongly associated with mortality after TOYOBO VAD implantation. Many biomarkers are susceptible to concomitant treatment (eg, APTT by heparin, creatinine by dialysis), complicated by infection (eg, C-reactive protein, platelet count), or affected by the patient s nutritional status (eg, albumin, total protein), and thus do not directly reflect hemodynamic collapse. Tbil level may be a marker that is specific for liver damage caused by compromised hemodynamics, albeit a higher Tbil level may not only indicate liver congestion but, more importantly, liver injury caused by low cardiac output. Several previous studies have confirmed this, 7 11 but in the Osaka study multivariate analysis failed to show that the preoperative Tbil level was an independent risk factor for mortality. 20 The discrepancy may be attributable to the fact that half of their patients received different types of LVADs, such as implantable LVADs, or the TOYOBO VAD was inserted in the left atrium. We also observed that perioperative right heart support resulted in extremely poor outcomes after TOYOBO VAD implantation, which was consistent with the results of the Osaka study 20 and data from INTERMACS. 9 Because there were only 8 such cases in our study, we were unable to analyze the preoperative risk factors responsible for the necessity of right heart support. Previous studies showed that preoperative levels of creatinine and Tbil were independent risk factors for right heart support, 11,20 suggesting that biventricular dysfunction causes more severe organ failure. Study Limitations We acknowledge that our study has several limitations. This study was conducted in a single center, and consequently included a limited number of patients. The J-MACS registry has recently started to collect data of VAD patients from multiple centers, and we are looking forward to more extensive analyses based on large numbers of patients. This study was conducted in a retrospective manner, and, from the ethical point of view, during the post-rematch era we could not set a medical treatment group as a control. Nowadays a prospective controlled study can only be feasible when different types of LVADs are compared, such as in the HeartMate II DT trial. 4 We did not include hemodynamic data obtained by right heart catheterization, which might be useful to determine right ventricular function and thus precisely predict the perioperative necessity of right heart support. Conclusions According to an analysis of 47 patients implanted with extracorporeal LVADs, we found that TOYOBO VAD implantation resulted in a better prognosis unless the patient had preoperative hemodynamic collapse. Age and the preoperative Tbil level were independent risk factors for mortality after TOYOBO VAD implantation. Preoperative risk stratification is important because the extracorporeal LVAD demands long-term care for years until a donor heart becomes available. None. Sources of Funding

8 128 SHIGA T et al. None. Disclosures References 1. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG, et al. ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to update the 2001 guidelines for the evaluation and management of heart failure): Developed in collaboration with the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: Endorsed by the Heart Rhythm Society. Circulation 2005; 112: e154 e Taylor DO, Stehlik J, Edwards LB, Aurora P, Christie JD, Dobbels F, et al. Registry of the international society for heart and lung transplantation: Twenty-sixth official adult heart transplant report J Heart Lung Transplant 2009; 28: Miller LW, Pagani FD, Russell SD, John R, Boyle AJ, Aaronson KD, et al. Use of a continuous-flow device in patients awaiting heart transplantation N Engl J Med 2007; 357: Slaughter MS, Rogers JG, Milano CA, Russell SD, Conte JV, Feldman D, et al. Advanced heart failure treated with continuousflow left ventricular assist device. N Engl J Med 2009; 361: Saito S, Matsumiya G, Sakaguchi T, Fujita T, Kuratani T, Ichikawa H, et al. Fifteen-year experience with toyobo paracorporeal left ventricular assist system. J Artif Organs 2009; 12: Nishimura T, Kyo S. High-dose carvedilol therapy for mechanical circulatory assisted patients. J Artif Organs 2010; 13: El-Banayosy A, Arusoglu L, Kizner L, Tenderich G, Boethig D, Minami K, et al. Predictors of survival in patients bridged to transplantation with the Thoratec VAD device: A single-center retrospective study on more than 100 patients. J Heart Lung Transplant 2000; 19: Frazier OH, Rose EA, Oz MC, Dembitsky W, McCarthy P, Radovancevic B, et al. Multicenter clinical evaluation of the Heartmate vented electric left ventricular assist system in patients awaiting heart transplantation. J Thorac Cardiovasc Surg 2001; 122: Holman WL, Kormos RL, Naftel DC, Miller MA, Pagani FD, Blume E, et al. Predictors of death and transplant in patients with a mechanical circulatory support device: A multi-institutional study. J Heart Lung Transplant 2009; 28: Farrar DJ. Preoperative predictors of survival in patients with Thoratec ventricular assist devices as a bridge to heart transplantation [Thoratec ventricular assist device principal investigators]. J Heart Lung Transplant 1994; 13: ; discussion Matthews JC, Koelling TM, Pagani FD, Aaronson KD. The right ventricular failure risk score a pre-operative tool for assessing the risk of right ventricular failure in left ventricular assist device candidates. J Am Coll Cardiol 2008; 51: Korfer R, El-Banayosy A, Arusoglu L, Minami K, Korner MM, Kizner L, et al. Single-center experience with the thoratec ventricular assist device. J Thorac Cardiovasc Surg 2000; 119: McBride LR, Naunheim KS, Fiore AC, Moroney DA, Swartz MT. Clinical experience with 111 Thoratec ventricular assist devices. Ann Thorac Surg 1999; 67: ; discussion Frazier OH, Rose EA, Oz MC, Dembitsky W, McCarthy P, Radovancevic B, et al. Multicenter clinical evaluation of the Heartmate vented electric left ventricular assist system in patients awaiting heart transplantation J Thorac Cardiovasc Surg 2001; 122: Nishinaka T, Schima H, Roethy W, Rajek A, Nojiri C, Wolner E, et al. The Duraheart VAD, a magnetically levitated centrifugal pump: The University of Vienna bridge-to-transplant experience. Circ J 2006; 70: Pagani FD, Miller LW, Russell SD, Aaronson KD, John R, Boyle AJ, et al. Extended mechanical circulatory support with a continuous-flow rotary left ventricular assist device. J Am Coll Cardiol 2009; 54: Morshuis M, El-Banayosy A, Arusoglu L, Koerfer R, Hetzer R, Wieselthaler G, et al. European experience of duraheart magnetically levitated centrifugal left ventricular assist system. Eur J Cardiothorac Surg 2009; 35: ; discussion Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W, et al. Long-term mechanical left ventricular assistance for end-stage heart failure. N Engl J Med 2001; 345: Lietz K, Long JW, Kfoury AG, Slaughter MS, Silver MA, Milano CA, et al. Outcomes of left ventricular assist device implantation as destination therapy in the post-rematch era: Implications for patient selection Circulation 2007; 116: Saito S, Matsumiya G, Sakaguchi T, Miyagawa S, Yoshikawa Y, Yamauchi T, et al. Risk factor analysis of long-term support with left ventricular assist system. Circ J 2010; 74: Sasaoka T, Kato TS, Komamura K, Takahashi A, Nakajima I, Oda N, et al. Improved long-term performance of pulsatile extracorporeal left ventricular assist device. J Cardiol 2010; 56: Tsuchihashi-Makaya M, Hamaguchi S, Kinugawa S, Yokota T, Goto D, Yokoshiki H, et al. Characteristics and outcomes of hospitalized patients with heart failure and reduced vs preserved ejection fraction: Report from the Japanese Cardiac Registry of Heart Failure in Cardiology (JCARE-CARD). Circ J 2009; 73: Lenzen MJ, Scholte op Reimer WJ, Boersma E, Vantrimpont PJ, Follath F, Swedberg K, et al. Differences between patients with a preserved and a depressed left ventricular function: A report from the Euroheart Failure Survey. Eur Heart J 2004; 25: Fonarow GC, Stough WG, Abraham WT, Albert NM, Gheorghiade M, Greenberg BH, et al. Characteristics, treatments, and outcomes of patients with preserved systolic function hospitalized for heart failure: A report from the Optimize-HF registry. J Am Coll Cardiol 2007; 50: