Combination Evaluation of Preoperative Risk Indices Predicts Requirement of Biventricular Assist Device

Transcription

1 Circulation Journal Official Journal of the Japanese Circulation Society ORIGINAL ARTICLE Heart Failure Combination Evaluation of Preoperative Risk Indices Predicts Requirement of Biventricular Assist Device Taro Shiga, MD, PhD; Koichiro Kinugawa, MD, PhD; Teruhiko Imamura, MD; Naoko Kato, PhD; Miyoko Endo; Toshiro Inaba, MD; Hisataka Maki, MD; Masaru Hatano, MD; Atsushi Yao, MD, PhD; Takashi Nishimura, MD, PhD; Yasunobu Hirata, MD, PhD; Shunei Kyo, MD, PhD; Minoru Ono, MD, PhD; Ryozo Nagai, MD, PhD Background: Patients with biventricular assist device (BiVAD) placement have a poor prognosis, but preoperative risk factors for the necessity of BiVAD have not been fully elucidated. Methods and Results: Data from 79 patients who received left ventricular assist device (LVAD) between November 2002 and December 2011 were retrospectively reviewed. Overall, 9 patients (11.4%) required BiVAD, and the survival rate of BiVAD patients was significantly lower than that of LVAD patients (P<0.001). Multivariate analysis for BiVAD requirement showed left ventricular diastolic diameter (LVDd) 62 mm (odds ratio [OR], 10.97; P=0.009) to be significantly associated with BiVAD requirement. Preoperative central venous pressure (CVP)/pulmonary capillary wedge pressure (PCWP) ratio 0.5 (OR, 13.09; P=0.028) was also significantly associated with BiVAD requirement. A new scoring system for predicting BiVAD requirement was created from the combination of CVP/PCWP ratio ( 0.5), body surface area ( 1.4 m 2 ), preoperative continuous hemodiafiltration use, B-type natriuretic peptide ( 1,200 pg/ml) and LVDd ( 62 mm), and this had a significantly larger area under the curve (0.909; P=0.003) than right ventricular stroke work index on receiver operating characteristic analysis. A score >20 using the new scoring method indicated significantly high probability of BiVAD requirement (OR, 16.00; P=0.019). Conclusions: The new scoring method, which includes CVP/PCWP ratio, is a novel risk stratification tool for BiVAD therapy. (Circ J 2012; 76: ) Key Words: Heart failure; Right ventricular dysfunction; Stage D; Transplantation Outcomes after implantation of pulsatile left ventricular assist device (LVAD) appear to have been critically dependent on postoperative right ventricular (RV) function. 1 3 Recent innovations of LVAD pumps have dramatically decreased device-related complications and improved overall prognosis, 4 7 but postoperative RV failure is still a serious concern even in the era of continuous flow pumps. 7,8 Editorial p 2740 Generally speaking, open heart surgery using cardiopulmonary bypass is often associated with postoperative RV dysfunction even if preoperative RV function is apparently normal. Postoperative RV failure usually manifests perioperatively or soon after surgery, but resolves within 1 2 weeks in most cases, but sometimes requires i.v. inotropes for >2 weeks and/or inhalation of nitric oxide for >48 h. In more severe cases of RV failure, mechanical support for RV with extracorporeal membrane oxygenation (ECMO) is necessary, albeit temporary. Furthermore, the implantation of RV assist device (RVAD) is an extreme and prolonged form of mechanical support of RV. The aforedescribed requirements of inotropes to RVAD are in fact the definition of postoperative RV failure, 1 3 which is thought to be a manifestation of pre-existing intrinsic RV impairment during the perioperative period. If patients with LVAD implantation have severe RV failure, biventricular assist device (BiVAD) is occasionally necessary, but those who require BiVAD have an extremely poor prognosis, 9 as we have also reported. 10 In this regard, proper and precise identification of patients at high risk for RV failure is indispensable. Thus far, many investigators have reported preoperative risk factors for RV failure. 1 3,8,9,11 13 For instance, Received February 21, 2012; revised manuscript received June 19, 2012; accepted July 17, 2012; released online August 8, 2012 Time for primary review: 29 days Department of Cardiovascular Medicine (T.S., K.K., T. Imamura, N.K., T. Inaba, H.M., M.H., A.Y., Y.H., R.N.), Department of Organ Transplantation (M.E.), Department of Cardiothoracic Surgery (M.O.), University of Tokyo Hospital, Tokyo; and Department of Therapeutic Strategy for Heart Failure, Graduate School of Medicine, University of Tokyo, Tokyo (T.N., S.K.), Japan 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 2786 SHIGA T et al. Table 1. Baseline Patient Characteristics Patient characteristics (n=79) INTERMACS Profile1 42 (53.2) Age (years) 39.4±13.6 Male 58 (73.4) Etiology of heart failure Fulminant myocarditis 3 (3.8) Dilated cardiomyopathy 51 (64.6) Ischemic cardiomyopathy 14 (17.7) Other causes of heart failure 11 (13.9) Types of LVAD NIPRO-VAD 59 (74.7) EVAHEART 9 (11.4) DuraHeart 7 (8.9) Jarvik (3.8) HeartMate II 1 (1.3) Right ventricular assist device 9 (11.4) PCPS (n) 6 NIPRO-VAD (n) 3 Clinical course after LVAD implantation Death 24 (30.4) Transplantation 24 (30.4) Explantation of LVAD 6 (7.6) Alive with LVAD 25 (31.6) Data given as mean ± SD or n (%). Dilated cardiomyopathy included dilated phase hypertrophic cardiomyopathy in the present study. Other causes of heart failure in 11 patients consisted of 1 case of restrictive cardiomyopathy, 3 of postpartum cardiomyopathy, 2 of valvular heart disease, 2 of adriamycin-induced cardiomyopathy, 2 of congenital heart disease, and 1 of cardiac sarcoidosis. INTERMACS, Interagency Registry for Mechanical Circulatory Support; LVAD, left ventricular assist device; PCPS, percutaneous cardiopulmonary support. RV stroke work index (RVSWI) 3,11,12 and the ratio of central venous pressure to pulmonary capillary wedge pressure (CVP/ PCWP) 8 have been widely used. These indices need an invasive measurement of hemodynamics in severely ill patients, and instead some investigators claim that combinations of preoperative parameters, which are able to be non-invasively obtained, well predict postoperative RV failure We maintain, however, that preoperative risk stratification for the necessity of BiVAD remains to be elucidated, and that there have been very few reports from Japan. 16 The aim of this study was to identify preoperative risk factors for the requirement of BiVAD therapy and to create a novel scoring system (Todai RV failure score: TRV score) for predicting BiVAD therapy. Methods Subjects We retrospectively analyzed 83 consecutive patients with endstage heart failure who were treated with LVAD between November 2002 and December 2011 and followed at the University of Tokyo Hospital. Every patient was considered as eligible for transplantation at least at the time of LVAD implantation. We excluded 4 patients whose data were insufficient to analyze, and finally we could analyze 79 patients according to endpoint (death, explantation of VADs with recovery of native heart, or heart transplantation) or the end of follow-up on VAD support. Patients were all treated with standard medical therapy before LVAD implantation and, if indicated, cardiac resynchronization therapy. LVAD was usually implanted due to cardiogenic shock or progressive decline of end-organ function due to low cardiac output in spite of i.v. inotropic agents. Percutaneous cardiopulmonary support (PCPS), intra-aortic balloon pumping, and/or continuous hemodiafiltration (CHDF) were introduced if necessary. Those who required RVAD in addition to LVAD were assigned to the BiVAD group. In this study, RVAD therapy included an ECMO system and/or NIPRO-VAD inserted in the RV or right atrium (RA). Written informed consent was obtained before VAD implantation. The study protocol was approved by the Ethics Committee of Graduate School of Medicine, University of Tokyo (application number 779 [1]). Hemodynamic data of patients with fulminant myocarditis were available only before PCPS insertion if any, but these data might not reflect preoperative status considering the rapid nature of the clinical course. Therefore, we excluded 3 patients with fulminant myocarditis from the main analysis. As a consequence, we analyzed data from 76 patients. Hemodynamic measurement was performed in 55 patients who underwent right heart catheterization preoperatively. Analyzed data included cardiac index (CI), mean RA pressure (RAP), PCWP, mean pulmonary artery pressure (mpap), systolic pulmonary artery pressure, diastolic pulmonary artery pressure (dpap), CVP, transpulmonary pressure gradient (TPG): mpap-pcwp, CVP/PCWP ratio, and RVSWI. RVSWI was calculated using the following formula: (CI/heart rate) (mpap-rap) In some cases we used dpap and mean CVP as substitutes for PCWP and RAP, respectively. In patients who were treated with PCPS before LVAD implantation, we used data of right heart catheterization before PCPS insertion. RV failure risk score (RVFS) from the University of Michigan was also calculated based on the formula provided previously. 13 Statistical Analysis We used PASW Statistics 18 (SPSS) and JMP version 9.0 (SAS Institute) for statistical analysis. Continuous variables are expressed as mean ± SD and were compared using unpaired t- test. Categorical variables are expressed as percentages and were compared using chi-square test. Kaplan-Meier analysis was used to estimate survival. Survival between each group was compared by log-rank test. Plasma level of B-type natriuretic peptide (BNP) was changed into logarithmic data to fit a normal distribution when compared between 2 groups. Univariate and multivariate analysis with a logistic regression model were performed to calculate adjusted odds ratio (OR) and 95% confidence intervals to assess preoperative risk factors for the requirement of BiVAD therapy. Categorical variables that had a tendency (P<0.10) to associate with BiVAD requirement on univariate analysis were used in multivariate analysis with a stepwise method. To select variables for multivariate analysis, multicollinearity was assessed using Pearson s product-moment correlation coefficients (Table S1). Receiver operating characteristic (ROC) analysis for the requirement of BiVAD therapy was carried out on TRV score, 1/RVSWI, CVP/PCWP ratio and RVFS. Cut-offs of continuous variables were determined using JMP version 9.0. Probability was 2-tailed with P<0.05 regarded as statistically significant.

3 New Scoring for Stratification of BiVAD Requirement 2787 Table 2. Patient Baseline Characteristics and Hemodynamics vs. VAD Type Total (n=76) LVAD (n=70) BiVAD (n=6) INTERMACS Profile 1 39 (51.3) 34 (48.6) 5 (83.3) Age (years) 39.2± ± ± Male 56 (73.7) 52 (74.3) 4 (66.7) BSA (m 2 ) 1.60± ± ± * Ischemic cause of heart failure 14 (18.4) 13 (18.6) 1 (16.7) LVAD support period (days) 526.0± ± ±136.0 <0.001* Systolic blood pressure (mmhg) 88.3± ± ± Heart rate (beats/min) 98.9± ± ± LV diastolic diameter (mm) 68.8± ± ± * LVEF (%) 20.2± ± ± Serum sodium (mmol/l) 132.8± ± ± Serum creatinine (mg/dl) 1.20± ± ± Serum total protein (g/dl) 6.01± ± ± Serum albumin (g/dl) 3.35± ± ± Serum TB (mg/dl) 3.61± ± ± Hemoglobin (g/dl) 10.8± ± ± Platelet count ( 10 4 /μl) 15.1± ± ± APTT (s) 54.2± ± ± C-reactive protein (mg/dl) 7.12± ± ±2.14 <0.001* Logarithmic BNP 6.83± ± ± * Mechanical ventilation 38 (50.0) 34 (48.6) 4 (66.7) PCPS 17 (22.4) 15 (21.4) 2 (33.3) IABP 41 (53.9) 38 (54.3) 3 (50.0) CHDF 13 (17.1) 10 (14.3) 3 (50.0) RVFS 5.91± ± ± Total (n=55) LVAD (n=50) BiVAD (n=5) CVP (mmhg) 9.88± ± ± mpap (mmhg) 31.8± ± ± PCWP (mmhg) 23.1± ± ± TPG (mmhg) 8.76± ± ± Cardiac index (L min 1 m 2 ) 2.00± ± ± CVP/PCWP 0.43± ± ± * RVSWI (g/m 2 ) 6.37± ± ± * *P<0.05 between LVAD patients and BiVAD patients. Patients with fulminant myocarditis were not included. APTT, activated partial thromboplastin time; BSA, body surface area; BiVAD, biventricular assist device; BNP, B-type natriuretic peptide; CHDF, continuous hemodiafiltration; CVP, central venous pressure; IABP, intra-aortic balloon pumping; LVEF, left ventricular ejection fraction; mpap, mean pulmonary artery pressure; PCWP, pulmonary capillary wedge pressure; RVFS, right ventricular failure score; RVSWI, right ventricular stroke work index; TB, total bilirubin; TPG, transpulmonary pressure gradient; VAD, ventricular assist device. Other abbreviations as in Table 1. Results Baseline Patient Characteristics The baseline characteristics of the total 79 patients are listed in Table 1. There were 9 BiVAD patients (11.4%), and all BiVAD patients received NIPRO-VAD in LV. Among the 9 patients with BiVAD, 6 patients received ECMO as well as RVAD. The remaining 3 patients had implantation of NIPRO- VAD in RV. Table 2 summarizes comparison of baseline characteristics and hemodynamic parameters between the LVAD and BiVAD patients. The BiVAD group was characterized by smaller body surface area (BSA), shorter period of LVAD support, smaller left ventricular diastolic diameter (LVDd), lower C-reactive protein and higher logarithmic BNP. RVFS was not significantly different between LVAD and BiVAD groups. With regard to hemodynamic parameters for right heart catheterization, CVP/PCWP ratio was significantly higher in the BiVAD group. Lower RVSWI was also significantly associated with BiVAD group. Comparison of Survival Between LVAD Patients and BiVAD Patients As shown in Figure 1, the survival rate of BiVAD patients was markedly lower than that of LVAD patients (P<0.001, logrank test). Of all 6 deceased in the BiVAD group, 4 patients (66.7%) died due to multiple organ failure within the first 3 months. Another 2 patients died of sepsis and device failure. Univariate and Multivariate Preoperative Risk Factors Univariate analysis of predictors of BiVAD requirement from among baseline characteristics and biochemical indices is given in Table 3. Age <23 years, BSA <1.40 m 2, preoperative use of CHDF, LVDd <62 mm, and higher plasma BNP were significantly associated with BiVAD requirement. Serum total bili-

4 2788 SHIGA T et al. rubin >4.8 mg/dl and plasma BNP >1,200 pg/ml had a tendency to associate with BiVAD requirement. Multivariate analysis showed that LVDd <62 mm was an independent risk factor for BiVAD implantation. Table 3 also lists univariate analysis of preoperative hemodynamic parameters. We found that CVP/PCWP >0.5 and RVSWI <4.0 had a significant association with BiVAD requirement. Because of the close correlation among PCWP, CVP/PCWP ratio and RVSWI (Table S1), multivariate analysis of these hemodynamic parameters was avoided. Figure 1. Comparison of survival between left ventricular assist device (LVAD) patients and biventricular assist device (BiVAD) patients (Kaplan-Meier analysis). New Score for BiVAD Requirement: TRV Score We selected 4 risk factors from among baseline characteristics and biochemical indices on the basis of univariate and multivariate analysis. We included CVP/PCWP ratio as well. Based on the ORs from univariate analysis, a weighted coefficient was assigned to each of them in order to construct the new TRV score (Table 4). To compare the predictability for BiVAD requirement among TRV score, CVP/PCWP ratio, 1/RVSWI Table 3. Preoperative Risk Factors for BiVAD Requirement Univariate Variables (n=76), Baseline patient characteristics and biochemical indices INTERMACS level Profile ( ) Age ( ) Age ( 23 years) 0.023* ( ) BSA 0.028* ( ) BSA ( 1.40 m 2 ) 0.029* ( ) Preoperative use of IABP (+) ( ) Preoperative use of PCPS (+) ( ) Preoperative use of CHDF (+) 0.043* ( ) Preoperative mechanical ventilation (+) ( ) Serum TB ( ) Serum TB ( 4.8 mg/dl) ( ) C-reactive protein ( ) Plasma BNP 0.043* ( ) Plasma BNP ( 1,200 pg/ml) ( ) LV diastolic diameter 0.018* ( ) LV diastolic diameter ( 62 mm) 0.025* ( ) RVFS ( ) Multivariate LV diastolic diameter ( 62 mm) 0.025* ( ) Univariate Variables (n=55), Right heart catheterization CVP ( ) mpap ( ) PCWP ( ) TPG ( ) Cardiac index ( ) CVP/PCWP ratio 0.037* ( ) CVP/PCWP ratio ( 0.5) 0.037* ( ) RVSWI ( ) RVSWI ( 4.0) 0.023* ( ) *P<0.05. CI, confidence interval; OR, odds ratio. Other abbreviations as in Tables 1,2.

5 New Scoring for Stratification of BiVAD Requirement 2789 Table 4. TRV Score and Characteristics Score CVP/PCWP ratio BSA 1.40 m 2 7 CHDF (+) 6 BNP 1,200 pg/ml 8 LVDd 62 mm 13 AUC (95%CI) TRV score ( ) 0.003* CVP/PCWP ratio ( ) 0.021* 1/RVSWI ( ) 0.025* RVFS ( ) Univariate TRV score (n=52) TRV score 0.009* ( ) TRV score ( 20 points) 0.019* ( ) *P<0.05. Maximum possible score is 45 points. LVDd, left ventricular diastolic diameter; TRV score, todai right ventricular failure score. Other abbreviations as in Tables 2,3. and RVFS, we performed ROC analysis (Figure 2). As shown in Table 4, the area under the curve of TRV score was larger than that of CVP/PCWP ratio, 1/RVSWI and RVFS. From this ROC analysis, we determined a cut-off for TRV score of 20 points (specificity/sensitivity 0.800/0.800; Figure 2). Table 4 also lists the results of univariate analysis of TRV score for BiVAD requirement. TRV score >20 points indicated a significantly higher risk of BiVAD requirement. Discussion Figure 2. Area under the curve (AUC) of Todai right ventricular failure score (TRV score; 0.909) compared with other indices on receiver operating characteristic analysis. The AUC of TRV score was the largest in comparison with that of CVP/ PCWP ratio and 1/RVSWI. CVP, central venous pressure; PCWP, pulmonary capillary wedge pressure; RVFS, right ventricular failure score; RVSWI, right ventricular stroke work index. Among the 79 LVAD patients, 11.4% of them required BiVAD therapy, and the survival rate of BiVAD patients was extremely poor. Univariate and multivariate analysis showed that younger age, smaller BSA, preoperative use of CHDF, higher plasma BNP, smaller LV cavity, higher CVP/PCWP ratio and lower RVSWI were significantly associated with BiVAD requirement. We constructed a new scoring system (TRV score) composed of a combination of preoperative risk factors, BSA ( 1.40 m 2 ), CHDF (+), BNP ( 1,200 pg/ml), LVDd ( 62 mm) and CVP/PCWP ratio ( 0.5). ROC analysis showed that TRV score was the best discriminative indicator for BiVAD requirement among RVSWI, CVP/PCWP ratio and RVFS. According to the present data, BiVAD patients comprised 11.4% of all VAD patients, and the prevalence was similar to the number reported from INTERMACS. 9 The prognosis of these BiVAD patients was extremely poor as compared with LVAD patients. Deaths occurred in the first 3 months after BiVAD therapy, and the main cause of death was multiple organ failure. Considering this, it appears that preoperative end-organ dysfunction was too bad to be restored in many of the BiVAD patients. According to the INTERMACS registry, there was a 2 4-fold higher adverse event rate of bleeding and infection in the BiVAD group than in the LVAD group. 9 Sustained liver dysfunction may be a predisposing factor toward fatal bleeding or infection in the postoperative period. A couple of studies have reported that smaller BSA as well as younger age were often associated with BiVAD requirement. 9,11 Of the present 9 patients in the BiVAD group, 3 patients were in their teens, 2 of whom had dilated cardiomyopathy and 1 had large myocardial infarction resulting from Kawasaki disease. Cardiomyopathy of younger onset may have a strong genetic cause that also affects RV equally to LV. Another possibility is that RV in youth may be more susceptible to ischemic damage. Further studies are needed to clarify this point. We found that smaller LV cavity was significantly associated with BiVAD requirement. As shown in the present data,

6 2790 SHIGA T et al. Table 5. Pitfalls of RVSWI as a Marker for RV Failure On admission Preoperative Dilated cardiomyopathy without RV failure (n=1) CVP (mmhg) 2 1 mpap (mmhg) PCWP (mmhg) 17 4 Cardiac index (L min 1 m 2 ) CVP/PCWP ratio RVSWI RVFS 4 4 TRV score 7 7 Data from preoperative right heart catheterization performed just before left ventricular assist device implantation. RV, right ventricle. Other abbreviations as in Tables 2,4. PCWP was lower in BiVAD patients. Severe RV dysfunction decreases preload for LV, which well explains lower PCWP and smaller LV cavity. According to the statistical analysis, BSA and LVDd were not significantly correlated (Table S1). Therefore, smaller LVDd may not necessarily be a consequence of smaller stature. End-organ dysfunction including that of liver or kidney was not only the determinant of prognosis but the predictor for BiVAD requirement. Univariate analysis showed that preoperative CHDF use was significantly associated with higher risk of BiVAD implantation. The importance of preoperative end-organ function has been reported in a number of papers, 1 3,8,9,11 13 but in contrast such a combination of parameters as RVFS, 13 which can be measured non-invasively, failed to show good predictability in the present patients. This led us to include preoperative hemodynamic data into the TRV score system. The CVP/PCWP ratio 8 or RVSWI 3 has been widely used as a marker for preoperative RV dysfunction. We also showed that these 2 parameters predicted BiVAD requirement well in the present study. But we had 1 patient for whom it was difficult to assess preoperative RV function accurately by RVSWI. Table 5 lists the details of this case, involving a patient with dilated cardiomyopathy in whom RVSWI was low even though there was no RV failure. RVSWI is largely dependent on the level of PAP. In this case, low RVSWI may be attributable to the relaxation of pulmonary artery by 100% oxygen. From this experience, we considered that combination evaluation including CVP/PCWP ratio for BiVAD requirement is necessary. TRV score, in contrast, was consistently 7 in the first and second hemodynamic measurement. The idea on which construction of TRV score was based, is that preoperative RV function primarily determines the need for RVAD. The present analysis, however, was dependent on the subject group, which mostly consisted of patients with an extracorporeal pulsatile pump. Recently, Kormos et al reported that rate of RVAD requirement was low (6%) in patients with HeartMate II for LV. 8 Some may argue that lower flow of NIPRO-VAD may have contributed to the higher incidence of RVAD requirement in the present study (12%), but let us note that implantable pulsatile VADs such as HeartMate XVE had a similar BiVAD frequency, 9 regardless of the potential of flow as high as HeartMate II. Therefore, the superiority of HeartMate II is not a matter of flow quantity, but is largely explained by the recent trend of earlier LVAD implantation in less-sick patients. Less-sick patients may have less severe RV involvement, if any. Furthermore, less-sick patients have less impairment of liver dysfunction, which is accompanied by lower risk for perioperative bleeding. Large amount of blood transfusion during the perioperative period often exacerbates RV dysfunction even though RV dysfunction is latent preoperatively. Now, in the era of the continuous flow pump in Japan, 5 our new scoring system can contribute to earlier decisionmaking for LVAD implantation before RV dysfunction and/or end-organ damages become irreversible. Such earlier implantation of continuous flow LVAD may decrease the necessity of RVAD, but a certain number of patients will still require BiVAD because of intrinsic severe RV dysfunction. The TRV score can pick up those patients preoperatively, but even if we know that RV function is very bad, is there anything that can be done for the patient? It seems that all that can be done right now is to implant the best available LVAD and keep our fingers crossed. What else can be prepared in advance? There was a report that patients who were treated with planned BiVAD had better survival compared with patients in whom failed LVAD therapy necessitated BiVAD later. 17 In contrast, Cleveland et al analyzed whether durable LVAD/durable RVAD for BiVAD had superiority of prognosis over durable LVAD/temporary RVAD. Durable LVAD/ durable RVAD represents a deliberate or planned strategy for BiVAD in most cases, whereas durable LVAD/temporary RVAD may be a result of failed LVAD in many cases, that is, unexpected BiVAD therapy. But they did not find any differences in survival between these 2 groups. 9 The 2 findings described here contradict each other. The discrepancy should be clarified in future, but may be explained by the fact that durable RVADs were all pulsatile flow pumps even in the most recent era. As has been well-established in LVAD, pulsatile flow pumps are associated with lower survival rate and poorer quality of life than continuous flow pumps. 7 In Japan, only an extracorporeal pulsatile pump has been available for RVAD, therefore it has not been possible to plan BiVAD therapy in advance. Planned BiVAD becomes a real option only when we are able to use continuous flow implantable pumps as RVADs. Recently, BiVAD systems with 2 continuous flow pumps have been reported, and planned BiVAD therapy with implantable VADs may now be possible. In order to stratify patients suitable for planned BiVAD therapy, and as a result to successfully bridge them to transplantation, our new scoring system can be an efficient tool. Several limitations of this study must be acknowledged. This study was conducted in a single center and therefore was limited in the number of enrolled patients. All data were analyzed in a retrospective manner. In addition, >75% of patients received an extracorporeal pulsatile LVAD in the present study, and all patients with BiVAD were implanted with extracorporeal pulsatile pumps in the LV and RV. TRV score has been mainly constructed on the data from 5 BiVAD patients in whom we were able to analyze hemodynamic results, and the small number of data limit statistical power. Finally, the TRV score should be tested in a prospective manner as to whether it works in the era of continuous flow pumps for LVAD. In conclusion, we have described a new scoring system that consists of CVP/PCWP ratio ( 0.5), BSA ( 1.4 m 2 ), preoperative use of CHDF, BNP ( 1,200 pg/ml) and LVDd ( 62 mm), and which could be an effective tool for preoperative stratification for BiVAD requirement. Acknowledgments The data in this paper will be in part presented at the 76 th Annual Scientific Meeting of the Japanese Circulation Society (Fukuoka, Japan, 2012).

7 New Scoring for Stratification of BiVAD Requirement 2791 This work was supported in part by the FUGAKU trust for medicinal research to K.K., a Grant-in-Aid for the JSPS Postdoctoral Research Fellow from the Japan Society for the Promotion of Science (no ) to N.K., and the Japan Society for the Promotion of Science (JSPS) through its Funding Program for World-Leading Innovative R&D on Science and Technology (FIRST Program) to R.N.. None. Disclosures References 1. Kormos RL, Gasior TA, Kawai A, Pham SM, Murali S, Hattler BG, et al. Transplant candidate s clinical status rather than right ventricular function defines need for univentricular versus biventricular support. J Thorac Cardiovasc Surg 1996; 111: ; discussion Dang NC, Topkara VK, Mercando M, Kay J, Kruger KH, Aboodi MS, et al. Right heart failure after left ventricular assist device implantation in patients with chronic congestive heart failure. J Heart Lung Transplant 2006; 25: Ochiai Y, McCarthy PM, Smedira NG, Banbury MK, Navia JL, Feng J, et al. Predictors of severe right ventricular failure after implantable left ventricular assist device insertion: Analysis of 245 patients. Circulation 2002; 106: I198 I Slaughter MS, Rogers JG, Milano CA, Russell SD, Conte JV, Feldman D, et al. Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009; 361: Kinugawa K. How to treat stage D heart failure? Circ J 2011; 75: Morita S. An old ventricular assist device still working for patients with end-stage heart failure in Japan. Circ J 2011; 75: Kirklin JK, Naftel DC, Kormos RL, Stevenson LW, Pagani FD, Miller MA, et al. The Fourth INTERMACS Annual Report: 4,000 implants and counting. J Heart Lung Transplant 2012; 31: Kormos RL, Teuteberg JJ, Pagani FD, Russell SD, John R, Miller LW, et al. Right ventricular failure in patients with the HeartMate II continuous-flow left ventricular assist device: Incidence, risk factors, and effect on outcomes. J Thorac Cardiovasc Surg 2010; 139: Cleveland JC Jr, Naftel DC, Reece TB, Murray M, Antaki J, Pagani FD, et al. Survival after biventricular assist device implantation: An analysis of the Interagency Registry for Mechanically Assisted Circulatory Support database. J Heart Lung Transplant 2011; 30: Shiga T, Kinugawa K, Hatano M, Yao A, Nishimura T, Endo M, et al. Age and preoperative total bilirubin level can stratify prognosis after extracorporeal pulsatile left ventricular assist device implantation. Circ J 2011; 75: Fitzpatrick JR 3rd, Frederick JR, Hsu VM, Kozin ED, O Hara ML, Howell E, et al. Risk score derived from pre-operative data analysis predicts the need for biventricular mechanical circulatory support. J Heart Lung Transplant 2008; 27: Fukamachi K, McCarthy PM, Smedira NG, Vargo RL, Starling RC, Young JB. Preoperative risk factors for right ventricular failure after implantable left ventricular assist device insertion. Ann Thorac Surg 1999; 68: 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: Kavarana MN, Pessin-Minsley MS, Urtecho J, Catanese KA, Flannery M, Oz MC, et al. Right ventricular dysfunction and organ failure in left ventricular assist device recipients: A continuing problem. Ann Thorac Surg 2002; 73: Santambrogio L, Bianchi T, Fuardo M, Gazzoli F, Veronesi R, Braschi A, et al. Right ventricular failure after left ventricular assist device insertion: Preoperative risk factors. Interact Cardiovasc Thorac Surg 2006; 5: 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 2009; 74: Fitzpatrick JR 3rd, Frederick JR, Hiesinger W, Hsu VM, McCormick RC, Kozin ED, et al. Early planned institution of biventricular mechanical circulatory support results in improved outcomes compared with delayed conversion of a left ventricular assist device to a biventricular assist device. J Thorac Cardiovasc Surg 2009; 137: Saito S, Sakaguchi T, Miyagawa S, Yoshikawa Y, Yamauchi T, Ueno T, et al. Biventricular support using implantable continuous-flow ventricular assist devices. J Heart Lung Transplant 2011; 30: Saito S, Sakaguchi T, Sawa Y. Clinical report of long-term support with dual Jarvik 2000 biventricular assist device. J Heart Lung Transplant 2011; 30: Strueber M, Meyer AL, Malehsa D, Haverich A. Successful use of the HeartWare HVAD rotary blood pump for biventricular support. J Thorac Cardiovasc Surg 2010; 140: Supplementary File 1 Supplementary Files Table S1. Correlation Between Confounding Variables Please find supplementary file(s);