当前位置: 首页 > 期刊 > 《交互式心脏血管和胸部手术》 > 2006年第4期 > 正文
编号:11132268
Right ventricular failure after left ventricular assist device insertion: preoperative risk factors
http://www.100md.com 《交互式心脏血管和胸部手术》
     a Department of Anesthesiology and Critical Care – Policlinico San Matteo, via Donatello 7, 20020 Villa Cortese (MI), Pavia, Italy

    b Department of Cardiac Surgery – Policlinico San Matteo, Pavia, Italy

    Abstract

    Right ventricular failure after left ventricular assist device placement is the major concern on weaning from cardiopulmonary bypass and it is one of the most serious complications in the postoperative period. This complication has a poor prognosis and is generally unpredictable. The identification of pre-operative risk factor for this serious complication is incomplete yet. In order to determine pre-operative risk for severe right ventricular failure after left ventricular assist device support we analyzed preoperative hemodynamics, laboratory data and characteristics of 48 patients who received Novacor (World Heart Corp., Ottawa, ON, Canada). We compared the data from the patients who developed right ventricular failure and the patients who did not. Right ventricular failure occurred in 16% of the patients. There was no significant difference between the groups in demographic characteristics. We identified as preoperative risk factors the pre-operative low mean pulmonary artery and the impairment of hepatic and renal function on laboratory data. Our results confirm in part the findings of the few previous studies. This information may be useful for the patient selection for isolated left ventricular assist device implantation, but other studies are necessary before establishing criteria for patient selection for univentricular support universally accepted.

    Key Words: Cardiomyopathy; Heart-assist device; Heart failure

    1. Introduction

    Left ventricular assist device (LVAD) support has been used as a bridge to heart transplant and as an alternative to heart transplantation [1–3]. There are several potential complications during LVAD support including right ventricular (RV) failure, bleeding, air embolism, multiorgan failure, thromboembolism, infection and device failure. The ability of the RV to provide sufficient output to fill the LVAD is the major determinant of the correct LVAD functioning and the patients' survival. The RV failure will develop in approximately 25% of patients receiving LVAD support [4–6]. The major concern on weaning from cardiopulmonary bypass is that RV failure usually occurs at variable degrees and that these patients are invariably in a state of vasodilatory shock [7,8]. Abnormalities of RV function have been attributed to primary abnormalities of the RV myocardium, excessive load imposed on the RV, or obstruction to RV inflow. Recent studies have also suggested that left ventricular (LV) function may significantly affect RV function. This ventricular interdependence is defined as the forces that are transmitted from one ventricle to the other through the myocardium and pericardium, independent from neural, humoral, or circulatory effects. Ventricular interdependence is a consequence of the close anatomic association between the ventricles. The evidence of diastolic ventricular interaction has been observed in post-mortem and in isolated heart preparation: the volume/pressure in one ventricle can directly influence the ones in the other ventricle. Many studies [9] showed a systolic ventricular interdependence: right ventricular systolic pressure and pulmonary artery blood flow present both RV and LV components. The LVAD support induces the LV unloading and a decreasing afterload, increasing compliance and decreasing contractility of RV. With full LVAD support LV end-diastolic volume decreases, while right RV end-diastolic volume increases with a significant leftward septal shift. Global RV contractility is impaired with leftward septal shifting, but RV myocardial efficiency is maintained through a decrease in RV afterload and an increase in RV preload. In preexisting pathologic condition the RV response is qualitatively the same, but the decrease in LV systolic function decreases LV assistance to RV function.

    The aim of this study is to identify preoperative risk factors for RV failure development after LVAD support among the preoperative hemodynamics and characteristics of patients who received Novacor (World Heart Inc., Ottawa, Ontario, Canada) and then establishing criteria for the patient selection.

    2. Materials and methods

    From November 1992 to January 2005, Novacor LVADs (World Heart Inc., Ottawa, Ontario, Canada) were implanted in 54 patients at San Matteo Hospital (Pavia, Italy). The pump uses a dual pusher plates mechanism with a maximum stroke of 65 ml. The LVAD was implanted as bridge to transplant in 51 patients, destination therapy in 2 patients and bridge to recovery in 1 patient. The 2 patients selected for destination therapy presented formal contraindication to transplantation: age 65 years and colon cancer. The bridge to recovery was attempted in a female patient who was myocarditis affected. There were 48 males and 6 females with a mean age of population of 51.4 years (range 29–68 years). The underlying diseases were dilated cardiomyopathy in 34 patients, ischemic cardiomyopathy in 13 patients, myocarditis in 1 patient. The four cases of post-cardiothomy heart failure and the two rejections of heart transplantation have been excluded because, for these pathologies, biventricular assist device is more often needed. All the patients presented with advanced heart failure: dependence on intravenous inotropic and vasodilatory therapy (43 patients), intraaortic balloon pump support (13 patients) and centrifugal pump support (3 patients). In every patient we performed ECG, chest radiography, cardiac biomarkers, coronarography, echocardiography and right heart catheterization for the right heart failure diagnosis; the evidence of right heart failure is an exclusion criteria for left ventricular assist device insertion.

    To identify the pre-operative risk factors for RV failure during LVAD support, we divided the remaining 48 patients into two groups: the ‘RV failure’ group presented, after LVAD assistance, systolic artery pressure <80 mmHg, cardiac output <2 l/min, acute massive tricuspid regurgitation at transesophageal echocardiography, evidence of RV dysfunction at the direct observation on the operative field, requirement of intravenous inotropes, nitric oxide inhalation, and right side centrifugal pump for cadiopulmonary bypass weaning; the same criteria were used by the Cleveland Clinic Foundation group [6,7] and in the Columbia Presbyterian interim experience [10]. The management of RV failure was performed through accurate fluid loading on the basis of the pulmonary artery catheterization data and the transesophageal echocardiography monitoring, inotropic and vasopressor support with epinephrine, dopamine, dobutamine, enoximone, norepinephrine in various associations according to the circumstances. We also commonly used inhaled nitric oxide as selective pulmonary vasodilator to decrease RV afterload. In spite of this kind of support, the patients in RV failure group needed right side centrifugal pump for cadiopulmonary by-pass weaning.

    Pre-operative hemodynamic parameters were measured in the week preceding the intervention after the introduction of the therapy to optimize RV function. We used rapid-response thermistor pulmonary artery catheter (Swan-Ganz catheter, Edwards Lifesciences LLC, Irvine, CA) to obtain the mean pulmonary arterial pressure (PAP), the right atrial pressure (RAP), the cardiac output, RV ejection fraction (EF), the pulmonary wedge capillary pressure. Then we derived transpulmonary gradient; cardiac index, pulmonary vascular resistance, pulmonary vascular resistance index and right ventricular stroke index. Also many clinical factors, reported in literature as predictors of RV failure, were evaluated: preoperative fever, preoperative laboratory tests for liver and renal function, quantitative bleeding amount. We express the data as the mean±standard deviation; we performed statistical analysis by 2 test or unpaired Student t-test. Difference was considered significant at the level of P<0.05.

    Multivariate analysis has not been performed for the relative number of the patients, and because, on the basis of the univariate analysis, only two variables would enter into the multivariable logistic regression model.

    3. Results

    Table 1 summarizes patients' pre-operative characteristics and laboratory data for each group. The two groups did not show any significant difference about the diagnosis of the underling disease. The need of a pre-operative intraaortic balloon pump or extracorporeal membrane oxygenation support is not a risk factor for RV failure development, but pre-operative mechanical ventilation shows a significant correlation with RV failure, but the patients are very few, so it is hard to draw out a firm conclusion. Among the laboratory tests for renal and liver function, creatinine, blood urea nitrogen, aspartate aminotrasferase and alanine aminotransferase are significantly different between the two groups (Figs. 1 and 2).

    Table 2 summarizes preoperative hemodynamic variables for each group. A significant difference is appreciated in mean pulmonary artery pressure (Fig. 3).

    The incidence of reoperation for bleeding is not significantly higher in the RV failure group. The survival rate after heart transplant was 85% in the patients group that did not develop RV failure and 25% in the other group.

    4. Discussion

    Heart failure represents a major public health issue in the Western World. Heart transplantation is the therapy for end stage heart failure, but the main impediment is the discrepancy between the number of donors and recipients. In our experience LVAD was used, for the main part, as bridge to transplantation in patients who could not wait for a donor only on medical therapy. The quality of life on assistance was reasonable and many patients were discharged from the hospital. The aim of this retrospective study is the identification of the pre-operative risk factors for severe RV failure after LVAD insertion to improve patient selection for isolated LVAD insertion. We analyzed pre-operative demographic, hemodynamic and laboratory data and the intraoperative variables reported as risk factors in the previous studies. The literature does not report a firm conclusion, so it is still difficult to define universal criteria for patient selection for univentricular or biventricular assist device implantation.

    Considering the demographic variables and the diagnosis of the underling disease, our patients did not present any significant differences; Ochai and associates [7] found significant difference in female gender, result not confirmed after adjusting for body surface area, and in the non-ischemic etiology as a predictor of RV failure after LVAD support, in accordance with Thoratec experience [11], which reported that the need of biventricular support was more frequent in dilated cardiomyopathy rather than coronary artery disease.

    Pulmonary hypertension with elevated PVR has been retained as a contraindication for LVAD use of high risk RV failure development [3], but recent studies indicated that when RV contractility is severely impaired in presence of a high PVR, RV is not able to elevate PAP [4–7]. Our data confirm the latest hypothesis: mean PAP was >42 mmHg in the 92% of the patients in the ‘no RV failure’ group and in none of the patients in the ‘RV failure’ group. The two groups, however, did not significantly differ for RV EF, but this finding also suggests that the RV contractility was lower in the RV failure group. If the RV contractility was the same in the two groups RV EF should have been higher in the patients with lower PAP (‘RV failure’ group), due to afterload dependence of the EF. Among the hemodynamic variables, Levin and associates [12] identified also a high value of RAP as a risk factor for RV failure after LVAD insertion. In our analysis, RAP was not significantly different between the two groups.

    The RV afterload seemed to be less important than the RV contractility for the RV function during LVAD support (no significant difference in the TPG, PVR or PVRI).

    Among the laboratory tests for liver and renal function creatinine, aspartate aminotransferase, alanine aminotransferase, blood urea nitrogen showed a significant difference between the two groups. These data may testify a chronic preoperative RV failure state at a higher degree with venous congestion in the patients at high risk for RV failure after LVAD support, as reported in a previous study [6]. In the light of this observation, our pre-operative diagnostics patients affected with heart failure may underestimate the presence of RV failure.

    Other studies [13] reported fever without infection, perioperative blood transfusions, and increased need for preimplant inotropic support as a risk factor for RV failure, that did not reach significant difference in our current study. Pagani et al. [14] reported that the incidence of RV failure after LVAD insertion was higher in patients supported with ECMO before LVAD. The authors attributed this finding to a lung injury exacerbated by ECMO or may reflect the overall increased severity of illness. The Pittsburgh group [15] and Thoratec [11] experience confirmed this finding. In our series ECMO was applied only in three cases; we found significant difference in the need of mechanical ventilation in the pre-operative period, but there were only 5 cases. Unfortunately, our experience with levosimendan is limited only to the latest two cases, and so we cannot draw our conclusion about its use in supporting failing right ventricle.

    In our series, RV failure complicated the 16% of LVAD insertions confirming the observation of the previous studies. RV failure is one of the most serious complications of the LVAD support, because the survival to heart transplant was significantly lower than the RV failure group (25%) when compared with the other group (85%). Both the study of Frazier and associates [3] and the one of Goldstein et al. [5] reported significantly negative correlation with the outcome. For this reason, the prediction of RV failure is very important for the selection of patients [19]. Although several studies have attempted to identify risk factors for RV failure after LVAD support, few data are available to predict RV failure before the operation, and more data are needed.

    5. Conclusions

    The RV failure incidence after LVAD support was 16%, and it had negative correlation with the outcome. Preoperative low PAP was a significant risk factor for RV failure development. Additional risk factors included laboratory tests deposing for chronic congestive RV failure in the preoperative screening (creatinine, aspartate aminotransferase, alanine aminotransferase and blood urea nitrogen). This information may be useful for the patient selection for isolated LVAD implantation, but other studies are necessary before establishing criteria for patient selection for univentricular support are universally accepted.

    References

    Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W, Long JW, Ascheim DD, Tierrney AR, Levitan RG, Watson JT, Meier P, Ronan NS, Shapiro PA, Lazar RM, Miller LW, Gupta L, Fraier OH, Desvigne-Nickens P, Poirier VL. Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) study group. Long-term use of a ventricular assist device for end-stage heart failure. N Engl J Med 2001; 345:1435–1443.

    Frazier OH, Rose EA, McCarthy P, Burton NA, Tector A, Levin H, Kayne HL, Poirier VL, Dasse KA. Improved mortality and rehabilitation of transplant cadidates treated with long-term implantable left ventricular assist system. Ann Surg 1995; 222:327–336.

    El-Banayosy A, Deng M, Loisance DY, Vetter H, Gronda E, Loebe M, Vigano M. The European experience of Novacor left ventricular asssit (LVAS) therapy as a bridge to transplant: a retrospective multi-centre study. Eur J Cardiothorac Surg 1999; 15:835–841.

    Goldstein DJ, Seldomridge JA, Chen JM, Catanese KA, DeRosa CM, Weinberg AD, Smith CR, Rose EA, Levin HR, Oz MC. Use of aprotinin in LVAD recipients reduces blood loss, blood use, and perioperative mortality. Ann Thorac Surg 1995; 59:1063–1067.

    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; 68:2181–2184.

    Ochiai Y, McCarthy PM, Smedira NG, Banbury MK, Navia J, Feng J, Asu AP, Yeager ML, Buda T, Hoercher KJ, Howard MW, Takagaki M, Doi K, Fukamachi K. Predictors of severe right ventricular failure after implantable left ventricular assist device insertion: analysis of 245 patients. Circulation 2002; 106:suppl I, I198–I202.

    Kavarana MN, Pessin-Minsley MS, Urtheco J, Catanese KA, Flannery M, Oz MC, Naka Y. Right ventricular dysfunction and organ failure in left ventricular assist device recipients: a continuing problem. Ann Thorac Surg 2002; 73:745–750.

    Argenziano M, Choudhri AF, Oz MC, Rose EA, Smith CR, Landry DW. A prospective randomized trial of arginine vasopressin in the treatment of vasodilatory shock after left ventricular assist device placement. Circulation 1997; 96II:286–290.

    Santamore WP, Laman AG Jr. Left ventricular contributions to right ventricular systolic function during LVAD support. Ann Thorac Surg 1996; 61:350–356.

    Sun BC, Catanese KA, Spanier TB, Flannery MR, Gardocki MT, Marcus LS, Levin HR, Rose EA, Oz MC. Hundred long-term implantable left ventricular assist devices: the Columbia Presbyterian interim experience. Ann Thorac Surg 1999; 68:688–694.

    Farrar DJ, Hill JD, Pennington DG, McBride L, Holman W, Kormos R, Esmore D, Gray L, Seifert P, Schoettle G. Pre-operative and postoperative comparison of patients with univentricular and biventricular support with thoratec ventricular assist device as bridge to cardiac transplantation. J Thorac Cardiovasc Surg 1997; 113:202–209.

    Levin HR, Burkhoff D, Chen J. Right ventricular performance is the major preoperative determinant of right heart failure in patients with chronic heart failure who rreceive a left ventricular assist device – abstract. J Am Coll Cardiol 1994;.

    Kormos RL, Gasior TA, Kawai A, Pham SM, Murali S, Hattler BG, Griffith BP. Transplant cadidate's clinical status rather than right ventricular function defines need of univentricular versus biventricular support. J Thorac Cradiovasc Surg 1996; 111:725–734.

    Pagani FD, Lynch W, Swaniker F, Dyke DB, Bartlett R, Koelling T, Moscucci M, Deeb GM, Bolling S, Monaghan H, Aaronson KD. Extracorporeal life support to left ventricular assist device bridge to heart transplant: a strategy to optimize survival and resource utilization. Circulation 1999; 100:II206–II210.

    Oz MC, Argenziano M, Catanese KA, Gardocki MT, Goldstein DJ, Ashton RC, Gelijns AC, Rose EA, Levin HR. Bridge experience with long-term implantable left ventricular assist device. Circulation 1997; 95:1844–1852.(Luisa Santambrogioa, Tizi)