Renal dysfunction (RD) is common among patients with advanced heart failure (HF). It is often caused by the hemodynamic perturbations and potentially could improve after restoration of normal hemodynamics. No definitive tests are available to reliably predict the reversibility of RD in HF patients.
The prognosis for renal function is especially important in patients with end-stage HF who are considered for advanced therapy, such as heart transplantation (HTx) and left ventricular assist device (LVAD). Cardiac transplant recipients have an increased risk of postoperative renal impairment after HTx,1 which is higher than those for other forms of cardiac surgery.2 Subsequently, there is continuous risk related to nephrotoxic effect of calcineurin inhibitor immunosuppression. Based on the fact that 39% of heart transplant recipients with glomerular filtration rate (GFR) < 40 ml/min required the postoperative dialysis and early mortality in these patient group was 17%, the International Society of Heart and Lung Transplantation Guidelines for the care of cardiac transplant candidates considered the presence of irreversible RD reflected by GFR < 40 ml/min as a contraindication to heart transplant.3 This has resulted in a sizeable increase in combined heart and kidney transplantation (CHKTx) in the past decade, with 350 CHKTx procedures recorded every year in the Registry of the International Society for Heart and Lung Transplantation (ISHLT).4 The selection of patients for CHKTx was based on criteria derived from limited experience5,6 in the era, when inotropic infusion with hemodynamic optimization and intra-aortic balloon pump insertion were the only tools for hemodynamic optimization. Therefore, in patients without significant intrinsic renal disease, RD remains a controversial indication for CHKTx.7
The use of LVAD in patients with advanced HF improves functional capacity, quality of life, and survival.8,9 The data suggest that the effect of LVAD support on renal function is complex, and patient selection criteria for LVAD implantation remain one of the most challenging issues. First, poor baseline RD is associated with increased overall mortality and low bridge-to-transplant rates.10,11 Second, RD occurs frequently after LVAD implantation12 and often requires renal replacement therapy.13 Therefore, severe pre-LVAD RD, defined by the need for hemodialysis or serum creatinine bigger than 2.5–3 mg/dl, is considered a relative contraindication for LVAD implantation.14 However, several studies demonstrate that restoration of cardiac output provided by the pulsatile15 and continuous-flow LVAD11,16 improves renal function in advanced HF patients.
The aim of this study was to determine serial post-LVAD changes in renal function, to identify the prevalence and predictors of severe RD and dialysis treatment after LVAD implantation, and to assess the effect of severe RD on outcomes after LVAD implantation.
The protocol was approved by the University of Nebraska Medical Center institutional review board.
The study was a retrospective evaluation of prospectively gathered information of 165 advanced HF patients who underwent LVAD implantation at the University of Nebraska Medical Center between January 2009 and September 2014.
Renal function was assessed before LVAD implantation (the best value during the index hospitalization was determined as baseline renal function) and during routine follow-up visits at 1, 3, and 6 months and 1 year after LVAD implantation. Based on GFR before LVAD implantation, the total cohort was divided into two groups: 1) baseline GFR (bGFR) ≤ 40 group (n = 30), included patients with bGFR ≤ 40 ml/min/1.73 m2 despite hemodynamic optimization with intravenous inotropes, vasodilator agents, and temporary hemodynamic support with intra-aortic balloon pressure (IABP) or extracorporeal membrane oxygenator (ECMO); 2) bGFR > 40 group (n = 135), included patients with bGFR > 40 ml/min/1.73 m2. GFR was calculated by the abbreviated Modification of Diet in Renal Disease equation: 186 × (serum creatinine [mg/dl])−1.154 × (age [years])−0.203 × 0.742 (if female).17,18 Patients with RD secondary to chronic intrinsic or structural renal disease (bGFR ≤ 40 along with small [<10 cm] echogenic kidneys on ultrasound study or proteinuria > 2 g/day) and on chronic renal replacement treatment (RRT) were not selected for LVAD. Preoperative worsening in kidney function was defined as a rise in the serum creatinine level >0.3 mg/dl, despite medical optimization during the index hospitalization before LVAD implantation.19
The model for end-stage liver disease-XI (MELD-XI) score was defined by the formula: 5.11 X Ln (bilirubin) + 11.76 X Ln (creatinine) + 9.44.20 Given that international normalized ratio (INR) is not used in its calculation, MELD-XI remains accurate even in patients receiving oral anticoagulation.20 According to the pre-LVAD MELD-XI score, patients were dichotomized into those with MELD-XI <17 and MELD-XI ≥17.21
Patient demographic and clinical data were collected at the time of pre-LVAD evaluation, when all patients were already receiving maximal medical support. All echocardiograms were performed at University of Nebraska Medical Center with a standardized protocol.22 Indications for RRT in the form of continuous venovenous hemofiltration dialysis or hemodialysis included oliguria (urine output < 400 ml/day) unresponsive to the diuretic therapy with a serum creatinine level > 2.0 mg/dl, severe acidemia (pH < 7.30), and volume overload. Post-LVAD right HF was defined by the need for nitric oxide or intravenous inotrope therapy for >14 consecutive days to increase the cardiac index >2 L/min/m2. Other post-LVAD complications and death were recorded up to July 1, 2015.
Data are displayed as means ± SD or count and percentage. Variables with skewed distribution are reported as medians with first and third quartiles in parenthesis. Analysis to compare demographic and clinical data between the groups was performed using Wilcoxon rank sum test for continuous data and the χ2 and Fisher’s test. The change of GFR from 1 time point to another was tested by analysis of covariance with bGFR as a covariant. Univariate logistic regression was used to identify preoperative predictors of GFR ≤ 40 ml/min/1.73 m2 at 3 months after LVAD implantation. The final multivariate model included univariate significant variables for GFR ≤ 40 ml/min/1.73 m2 at 3 months after LVAD implantation (p < 0.05), and because of the high correlation between MELD score and bGFR, two models were used: the first model included aldosterone antagonists, tricuspid regurgitation (TR) grade ≥2, TV repair, and bGFR; the second model included aldosterone antagonists, TR grade ≥2, TV repair, and MELD-XI score. All statistical tests were two-sided, and a p < 0.05 was considered to be statistically significant.
Pre-LVAD GFR ≤ 40 ml/min/1.73 m2 presented in 30 (28%) patients (Table 1). The bGFR ≤ 40 patients were older, had a higher rate of ischemic cardiomyopathy, and more often were initially assigned to LVAD as a destination treatment. The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) score, the use of inotropes, and IABP before LVAD implantation were similar; however, a higher number of bGFR ≤ 40 patients were bridged to LVAD with ECMO support. The bGFR ≤ 40 patients had significantly higher MELD-XI score. There were no differences between the groups in kidney size and proteinuria.
Exercise capacity assessed by 6 min’ walk was significantly worse in the bGFR ≤ 40 group. The resting hemodynamics did not differ between the groups.
Echocardiographically, there were no differences between the groups in the cardiac remodeling indexes, left or right ventricular systolic function, and mitral valve disease. E/E′ was higher in the bGFR ≤ 40 group. More patients in the bGFR ≤ 40 group had moderate and severe TR. Overall, 18 patients required temporary right ventricular assist device (RVAD) usage: 11 patients underwent intraoperative RVAD implantation and 8 patients underwent delayed right ventricular mechanical support 2.6 ± 1.4 days after LVAD implantation. There were significantly more bGFR ≤ 40 group’s patients who required temporary RVAD support. The incidence of tricuspid valve repair, mitral valve repair, and coronary artery bypass graft during LVAD implantation was similar in both groups.
Cardiopulmonary bypass time and intraoperative blood product use did not differ between the bGFR ≤ 40 and the GFR > 40 groups (Table 2). Nevertheless, the patients with pre-LVAD MELD-XI score ≥ 17 had significantly longer cardiopulmonary bypass time (134 ± 51 min vs. 108 ± 49 min, p = 0.01) and required more perioperative blood product transfusions (red blood cells units: 5.7 ± 4.9 vs. 2.9 ± 3.1, p = 0.04; fresh frozen plasma units: 5.6 ± 4.1 vs. 3.6 ± 2.9, p = 0.02; platelets units: 3.3 ± 2.5 vs. 2.1 ± 1.7, p = 0.02).
In the overall patient population, GFR increased by 1 month after LVAD implantation, and then gradually declined but did not differ from the pre-LVAD level at 1 year follow-up. Similar patterns were found in the GFR > 40 group. In the GFR ≤ 40 group, GFR increased significantly at 1 month and 3 months, then plateaued, and remained higher than the pre-LVAD level at 1 year post-LVAD (Table 3; Figure 1).
There was an improvement in stages of RD by 3 months after LVAD implantation in the bGFR ≤ 40 group (Figure 2): 3 months GFR ≥ 60 and 3 months GFR 40–60 ml/min/1.73 m2 was found in 10 (33%) and 7 (23%) patients, respectively. However, 8 (27%) patients remained with 3 months GFR ≤ 40 ml/min/1.73 m2, 1 patient underwent CHKTx, and 4 (14%) patients had died by 3 months after LVAD in this group.
In the bGFR > 40 group at 3 months after LVAD implantation, 77 (57%) and 31 (23%) patients presented with GFR ≥ 60 and GFR 40–60 ml/min/1.73 m2, respectively. However, 10 (7%) patients had deterioration in kidney function to GFR ≤ 40 ml/min/1.73 m2, 4 (3%) patients underwent HTx, and 12 (9%) patients had died.
Post-LVAD RRT was used in 6 (20%) and 9 (7%) of the bGFR ≤ 40 group and of the bGFR > 40 patients, respectively (p = 0.02), for median 11.5 (interquartile range [IQR], 3–43) days. The patients requiring post-LVAD RRT received more perioperative red blood cell transfusions (6.6 ± 4.0 vs. 3.4 ± 3.1, p = 0.05). Overall, among patients requiring RRT, seven died, one underwent CHKTx, and one continued chronic dialysis while awaiting CHKTx; six patients were weaned from dialysis, but two left with GFR ≤ 40 and four improved renal function to GFR > 40. Median survival of patients on post-LVAD RRT was only 26 (IQR, 16–84) days. Post-LVAD survival rates at 1, 6, and 12 months for dialysis versus nondialysis groups of patients were 53%, 44%, and 39% versus 99%, 91%, and 82%, respectively (p < 0.0001, log-rank test) (Figure 3).
By 3 months after LVAD implantation, 20 (14%) patients had presented with severe RD (3 months GFR ≤ 40). An older age, lack of aldosterone antagonists treatment, TR grade ≥2, tricuspid valve repair during LVAD implantation, pre-LVAD MELD-XI ≥ 17, and pre-LVAD bGFR ≤ 40 were univariate predicators for severe RD at 3 months after LVAD implantation (Table 4). After a multivariate analysis, only the grade ≥ 2 TR (OR, 3.4; 95% confidence interval [CI], 1.23–10.28; p = 0.02) and MELD-XI score ≥ 17 (OR, 4.2; 95% CI, 1.45–12.24; p = 0.01) reached statistical significance.
Post-LVAD outcomes and adverse events are presented in Table 5. The duration of follow-up was 10.5 (IQR, 5.7–15.5) months from LVAD implantation and was similar between the groups. There was no difference in the length of index hospitalization between the GFR ≤ 40 and GFR > 40 groups (Table 5), and it was still true when only survivors were included in the analysis (22 [14–23] vs. 19 [15–27]; p = 0.75).
The prevalence of post-LVAD right HF was higher in bGFR ≤ 40 group. There was no difference in other LVAD complications.
Similar prevalence of stroke, LVAD thrombosis, and infections was observed in patients with 3 months GFR ≤ 40 and GFR > 40 ml/min/1.73 m2; however, the prevalence of gastrointestinal (GI) bleeding was higher in patients with 3 months GFR ≤ 40 ml/min/1.73 m2 (7 [35%] vs. 16 [12%]; p = 0.003).
Eight bGFR ≤ 40 and 54 bGFR > 40 patients underwent HTx. All eight transplanted bGFR ≤ 40 patients maintained good renal function after heart transplant with GFR 57.2 ± 17.2 ml/min/1.73 m2 at post-transplant hospital discharge.
This study demonstrated the following: 1) 56% of patients with advanced HF and severe pre-LVAD RD had durable improvement in renal function with LVAD support; 2) despite having a decline from the peak GFR at 1 month during 1 year follow-up, the kidney function did not decline and remained similar to the pre-LVAD value in patients without severe RD; 3) pre-LVAD MELD-XI ≥ 17 score and TR ≥ grade 2 can predict poor renal function on LVAD support; and 4) extremely poor survival because of early mortality among patients requiring RRT after LVAD implantation.
Severe chronic RD in patients with advanced HF indicates end-organ compromise and is associated with poor prognosis.23 The etiology of RD and the degree of reversibility remain unclear in many patients,24 and the ISHLT guidelines considered the presence of RD reflected by GFR < 40 ml/min as a relative contraindication to isolated heart transplant.3 This was defined in an era when infusion of inotropes and vasodilators and intra-aortic balloon pump insertion were the only tools for pretransplant hemodynamic optimization.5,6 With the introduction of LVAD support, RD may become more modifiable contraindication11; therefore, the ascertainment of the pre-LVAD predictors of RD and its reversibility in patients with LVAD support held particular interest in our study.
Overall, 28% of our LVAD patients had severe RD with GFR ≤ 40 ml/min/min2 before LVAD implantation. These patients had more advanced vascular damage, more prominent biventricular HF, higher prevalence of TR, and worse functional capacity. Not surprisingly, they had more end-organ damage, including liver dysfunction. The improvement in renal function in 56% of patients with GFR ≤ 40 ml/min/min2, however, suggests a significant degree of reversibility of RD with LVAD support even in this high-risk patient population. This improvement was quite durable and continued up to 12 months after LVAD implantation. Moreover, in our study 10 (33%) patients have become isolated heart transplant candidates and 8 (27%) underwent successful HTx with good post-transplant renal outcome.
Our data showed that despite having a decline from their peak GFR at 1 month, the long-term kidney function remained similar to the pre-LVAD value over 12 month period of time in the patient without significant RD before LVAD. The pattern of initial postsurgical improvement and subsequent decline in renal function post-LVAD patients is in concordance with the previous observations16,25 and multicenter INTERMACS registry analysis.26 Based on our results and the studies comparing nonpulsatile and pulsatile-flow LVADs,27 however, it seems unlikely that the decline in renal function can be attributed to adverse effects of continuous LVAD flow. Indeed, we found the same trend of initial postsurgical improvement and subsequent decline in renal function after heart transplant surgery in our institution (Eugenia Raichlin, University of Nebraska Medical Center, unpublished data). Postsurgical muscle mass loss and hence lower serum creatinine in the early postsurgical period can overestimate the true GFR calculated at 1 month postsurgery. In addition, resuming treatment with rennin-angiotensin pathway inhibitors could affect the later decline in GFR. The higher prevalence of GI bleeding in patients with kidney dysfunction after LVAD implantation may further contribute to worsening kidney function after LVAD implantation.
Moderate and severe pre-LVAD TR was a predictor of post-LVAD RD in our study, underlining the role of systemic congestion in RD in patients with advanced HF28 even on LVAD support. It was previously demonstrated that tricuspid valve surgery during LVAD implantation in patients with severe pre-LVAD TR promotes reverse remodeling of the right ventricle.29 In our study, tricuspid valve repair during LVAD implantation still was a univariate predictor of severe RD on LVAD support. Indeed, mean right atrial pressure remained significantly higher after TV repair,29 and this may cause the hemodynamic impairment and cardiorenal compromise.28 Moreover, in line with previous studies,30,31 our data showed that severe pre-LVAD RD was associated with right ventricular failure post-LVAD implantation. Thus, this study confirms the importance of the interrelationship between right ventricular hemodynamics and kidney dysfunction in the LVAD patients.
The role of increased MELD-XI score has been recently evaluated for its utility in predicting outcomes in post-LVAD patients. Although creatinine represents a major determinant of the MELD-XI score, in our study a MELD-XI score ≥ 17 was a stronger predictor of severe RD on LVAD support than GFR ≤ 40 ml/min/min2. This suggests that a MELD-XI score is not simply a surrogate for RD but rather a marker of the multiorgan dysfunction in advanced HF and more precisely reflects the patient’s clinical condition. Furthermore, similar to prior studies,21,32 our results demonstrate that elevated MELD-XI score is associated with increased requirement of perioperative blood products and longer time on cardiopulmonary bypass. Our data also showed a significantly higher red blood cell transfusion rate in patients who required RRT after LVAD implantation. Taken together, our study suggests that elevated MELD-XI score may be predictive of postoperative renal failure in the setting of high transfusion requirements and can help identify appropriate candidates for LVAD support.
In our study, 20% of patients with bGFR ≤ 40 ml/min/1.73 m2 and 7% of patients with bGFR > 40 ml/min/1.73 m2 required RRT after LVAD implantation. Topkara et al.33 reported poor survival of post-LVAD RRT patients with a 1 year mortality rate of 56.8%. Our study showed extremely poor survival because of early mortality: 47% at 1 month, and median survival time for the post-LVAD RRT patients in our study was only 26 days. Interestingly, patients who experienced clinical stabilization after the LVAD implantation had subsequent survival similar to nondialyzed patients (Figure 3).
This study was a retrospective analysis of prospectively collected data. Our bGFR ≤ 40 patients were carefully evaluated and those with significant renal failure because of intrinsic renal disease were not selected for LVAD; therefore, this finding cannot be interpolated to HF patients with severe renal failure.
This study demonstrated that carefully selected patients with advanced HF and GFR ≤ 40 ml/min/1.73 m2 secondary to cardiorenal syndrome can benefit from LVAD support. In the patient without severe RD at baseline, kidney function did not decline and remained similar to the pre-LVAD value during 1 year follow-up. Significant preexisting TR and advanced end organ dysfunction reflected by a MELD score ≥ 17 can identify patients at increased risk for poor kidney function after LVAD implantation. Additional research is needed to refine patient selection for advanced HF treatment with LVAD in this critically ill patient population.
1. Jokinen JJ, Tikkanen J, Kukkonen S, et al.: Natural course and risk factors for impaired renal function during the first year after heart transplantation. J Heart Lung Transplant 2010.29: 633640.
2. Martinelli SM, Patel UD, Phillips-Bute BG, et al.: Trends in cardiac surgery-associated acute renal failure in the United States: A disproportionate increase after heart transplantation. Ren Fail 2009.31: 633640.
3. Mehra MR, Kobashigawa J, Starling R, et al.: Listing criteria for heart transplantation: International Society for Heart and Lung Transplantation guidelines for the care of cardiac transplant candidates—2006. J Heart Lung Transplant 2006.25: 10241042.
4. Stehlik J, Edwards LB, Kucheryavaya AY, et al.: The Registry of the International Society for Heart and Lung Transplantation: Twenty-seventh official adult heart transplant report—2010. J Heart Lung Transplant 2010.29: 10891103.
5. Leeser DB, Jeevanandam V, Furukawa S, et al.: Simultaneous heart and kidney transplantation in patients with end-stage heart and renal failure. Am J Transplant 2001.1: 8992.
6. Bergler-Klein J, Pirich C, Laufer G, et al.: The long-term effect of simultaneous heart and kidney transplantation on native renal function. Transplantation 2001.71: 15971600.
7. Raichlin E, Kushwaha SS, Daly RC, et al.: Combined heart and kidney transplantation provides an excellent survival and decreases risk of cardiac cellular rejection and coronary allograft vasculopathy. Transplant Proc 2011.43: 18711876.
8. Abshire M, Dennison Himmelfarb CR, Russell SD: Functional status in left ventricular assist device
-supported patients: A literature review. J Card Fail 2014.20: 973983.
9. Rose EA, Gelijns AC, Moskowitz AJ, et al.; Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) Study Group: Long-term use of a left ventricular assist device
for end-stage heart failure. N Engl J Med 2001.345: 14351443.
10. Butler J, Geisberg C, Howser R, et al.: Relationship between renal function and left ventricular assist device
use. Ann Thorac Surg 2006.81: 17451751.
11. Sandner SE, Zimpfer D, Zrunek P, et al.: Renal function and outcome after continuous flow left ventricular assist device
implantation. Ann Thorac Surg 2009.87: 10721078.
12. Filsoufi F, Rahmanian PB, Castillo JG, Chikwe J, Carpentier A, Adams DH: Early and late outcomes of cardiac surgery in patients with moderate to severe preoperative renal dysfunction
without dialysis. Interact Cardiovasc Thorac Surg 2008.7: 9095.
13. Alba AC, Rao V, Ivanov J, Ross HJ, Delgado DH: Predictors of acute renal dysfunction
after ventricular assist device placement. J Card Fail 2009.15: 874881.
14. Lietz K, Miller LW: Patient selection for left-ventricular assist devices. Curr Opin Cardiol 2009.24: 246251.
15. Frazier OH: Mechanical circulatory support: New advances, new pumps, new ideas. Semin Thorac Cardiovasc Surg 2002.14: 178186.
16. Hasin T, Topilsky Y, Schirger JA, et al.: Changes in renal function after implantation of continuous-flow left ventricular assist devices. J Am Coll Cardiol 2012.59: 2636.
17. McBride LR, Naunheim KS, Fiore AC, et al.: Risk analysis in patients bridged to transplantation. Ann Thorac Surg 2001.71: 18391844.
18. O’Meara E, Chong KS, Gardner RS, Jardine AG, Neilly JB, McDonagh TA: The Modification of Diet in Renal Disease (MDRD) equations provide valid estimations of glomerular filtration rates in patients with advanced heart failure. Eur J Heart Fail 2006.8: 6367.
19. Gottlieb SS, Abraham W, Butler J, et al.: The prognostic importance of different definitions of worsening renal function in congestive heart failure. J Card Fail 2002.8: 136141.
20. Heuman DM, Mihas AA, Habib A, et al.: MELD-XI: A rational approach to “sickest first” liver transplantation in cirrhotic patients requiring anticoagulant therapy. Liver Transpl 2007.13: 3037.
21. Matthews JC, Pagani FD, Haft JW, Koelling TM, Naftel DC, Aaronson KD: Model for end-stage liver disease score predicts left ventricular assist device
operative transfusion requirements, morbidity, and mortality. Circulation 2010.121: 214220.
22. Lang RM, Mor-Avi V: Clinical utility of contrast-enhanced echocardiography. Clin Cardiol 2006.29(9 suppl 1): I15I25.
23. Odim J, Wheat J, Laks H, et al.: Peri-operative renal function and outcome after orthotopic heart transplantation. J Heart Lung Transplant 2006.25: 162166.
24. Russo MJ, Rana A, Chen JM, et al.: Pretransplantation patient characteristics and survival following combined heart and kidney transplantation: An analysis of the United Network for Organ Sharing Database. Arch Surg 2009.144: 241246.
25. Singh M, Shullo M, Kormos RL, et al.: Impact of renal function before mechanical circulatory support on posttransplant renal outcomes. Ann Thorac Surg 2011.91: 13481354.
26. Brisco MA, Kimmel SE, Coca SG, et al.: Prevalence and prognostic importance of changes in renal function after mechanical circulatory support. Circ Heart Fail 2014.7: 6875.
27. Sandner SE, Zimpfer D, Zrunek P, et al.: Renal function and outcome after continuous flow left ventricular assist device
implantation. Ann Thorac Surg 2009.87: 10721078.
28. Mullens W, Abrahams Z, Francis GS, et al.: Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol 2009.53: 589596.
29. Maltais S, Topilsky Y, Tchantchaleishvili V, et al.: Surgical treatment of tricuspid valve insufficiency promotes early reverse remodeling in patients with axial-flow left ventricular assist devices. J Thorac Cardiovasc Surg 2012.143: 13701376.
30. Kormos RL, Teuteberg JJ, Pagani FD, et al.; HeartMate II Clinical Investigators: 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: 13161324.
31. 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: 21632172.
32. Deo SV, Daly RC, Altarabsheh SE, et al.: Predictive value of the model for end-stage liver disease score in patients undergoing left ventricular assist device
implantation. ASAIO J 2013.59: 5762.
33. Topkara VK, Dang NC, Barili F, et al.: Predictors and outcomes of continuous veno-venous hemodialysis use after implantation of a left ventricular assist device
. J Heart Lung Transplant 2006.25: 404408.