Impact of Renal Function on Survival After Cardiac Resynchronization Therapy

Impact of Renal Function on Survival After Cardiac Resynchronization Therapy J. Alvin Kpaeyeh, Jr, MD, Laura Divoky, MD, J. Madison Hyer, MS, David D. Daly, Jr, MD, Anbukarasi Maran, MD, Ashley Waring, MD, and Michael R. Gold, MD, PhD* Chronic kidney disease (CKD) is associated with worse survival in patients with heart disease including those with implantable devices. Cardiac resynchronization therapy (CRT) can potentially improve renal function. To assess the relation between the change in renal function and survival with CRT, 238 patients undergoing initial CRT with defibrillator implantation between 2002 and 2011 were followed. The primary end point was all-cause mortality. The estimated glomerular filtration rate (eGFR), before implantation and 6 – 3 months after CRT was calculated. Patients were grouped at baseline into mild (stage I/II) or advanced (stage III/IV) CKD. Patients with end-stage renal disease were excluded. The mean follow-up time was 4.3 years. Multivariate analysis of baseline clinical characteristics showed that only renal function predicted the change in eGFR over the first 6 months of CRT. In the subgroup with mild CKD, eGFR decreased (78.5 – 17.3 to 67.8 – 26.8 p <0.001), whereas eGFR did not change in the subgroup with advanced CKD (45.6 – 11.1 to 46.8 – 17.0, p [ 0.46). Patients with advanced CKD had higher mortality than those with mild CKD (p <0.002). In both subgroups, an increase in eGFR was associated with improved survival (hazard ratio [ 0.79, p <0.001). In conclusion, baseline renal function and the subsequent change in eGFR are associated with long-term survival with CRT. Ó 2017 Elsevier Inc. All rights reserved. (Am J Cardiol 2017;120:262e266) Cardiac resynchronization therapy (CRT) is a wellestablished therapy for patients with heart failure (HF), a reduced left ventricular ejection fraction, and QRS prolongation.1,2 However, significant proportions of patients do not improve with CRT and are classified as nonresponders. Understanding the factors that contribute to CRT response is important to optimize outcomes. In this regard, chronic kidney disease (CKD) is commonly present in patients with HF and is also associated with an increased risk for sudden cardiac death.2 Approximately, half of patients with systolic HF also have some CKD.3,4 Unfortunately, most of the landmark trials have excluded patients with advanced renal failure and do not evaluate the change in renal function in patients after cardioverter defibrillator (CRT-D) implantation.2 Observational studies of CRT or implantable CRT-Ds, in patients with end-stage renal disease on dialysis, show very high mortality, raising question about the benefit of device therapy in this cohort.5 However, the high risk of mortality is not restricted to such patients, as registries that include subjects with a full spectrum of CKD show a progressive effect of the severity of CKD on mortality.6 Although CKD is usually considered a progressive process, the hemodynamic improvement associated with CRT could potentially improve renal function. Accordingly, the

Division of Cardiology, Medical University of South Carolina, Charleston, South Carolina. Manuscript received December 31, 2016; revised manuscript received and accepted April 18, 2017. This study was funded in part by the South Carolina Clinical and Translational Research Institute (SCTR Award UL1 TR001450). See page 266 for disclosure information. *Corresponding author: Tel: (843) 876-4760; fax: (843) 876-4990. E-mail address: [email protected] (M.R. Gold). 0002-9149/17/$ - see front matter Ó 2017 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2017.04.017

present study was designed to assess the impact of CKD and the change in renal dysfunction on long-term survival with CRT. Methods This was a retrospective cohort study from a prospective database of patients undergoing initial CRT implantation with implantable CRT-D from 2002 to 2011 at the Medical University of South Carolina. Patients were included if they had paired measurements of serum creatinine preoperatively and 6
3 months postoperatively (Table 1). Patients with end-stage renal disease (stage V) at baseline were excluded from analysis because most of these patients were on dialysis, and therefore, assessment of temporal changes in renal function wound not be as relevant. This study was approved by the institutional review board, and all patients gave informed consent for device implantation. The primary end point for this study was all-cause mortality. Date of death was determined using hospital records and the United States Social Security Death Index. Baseline demographic data were ascertained from the medical record. HF severity was based on the New York Heart Association functional class. For each patient, serum creatinine, weight, height, age and race were used to calculate the estimated glomerular filtration rate (eGFR ml/min/1.73 m2) using the Modified Diet in Renal Dysfunction equation.7 The eGFR was then used to categorize patients into mild CKD (stage 1/2 [eGFR  60 ml/min/1.73 m2] or advanced CKD [stage 3/4 (eGFR 15 to 59 ml/min/1.73 m2]). Stage 5 CKD (eGFR<15 ml/ min/1.73 m2 or on dialysis) subjects were excluded. Descriptive statistics are presented for all demographic, exposure, and co-morbidities in the form of frequency (%) www.ajconline.org

Heart Failure/Renal Function and Survival With CRT

263

Table 1 Baseline patient characteristics Variable Age (years) Women Body Surface Area Body Mass Index LVEF NYHA III HF Ischemic heart disease QRS Duration LBBB Hypertension Diabetes mellitus Atrial fibrillation ACE/ARB Beta blocker Amiodarone

Total (n¼238)

Mild CKD (n¼126)

Advanced CKD (n¼112)

p-value

65.9
10.4 63 (26.5%) 2.0
0.3 28.9
7.5 25.3
7.0 196 (82.4%) 135 (56.7%) 155.6
23.4 138 (58.2%) 153 (64.3%) 86 (36.1%) 23 (9.7%) 197 (82.8%) 209 (87.8%) 22 (9.2%)

64.3
10.1 34 (27%) 2.0
0.3 29.2
8.4 24.8
7.3 99 (78.6%) 73 (57.9%) 155.4
24.9 73 (57.9%) 77 (61.1%) 42 (33.3%) 8 (6.4%) 109 (86.5%) 114 (90.5%) 10 (7.9%)

67.6
10.4 29 (25.9%) 2.0
0.2 28.5
6.3 25.9
6.5 97 (86.6%) 62 (55.4%) 155.8
21.6 65 (58.6%) 76 (67.9%) 44 (39.3%) 15 (13.5%) 88 (78.6%) 95 (84.8%) 12 (10.7%)

0.014 0.85 0.53 0.59 0.20 0.11 0.69 0.90 0.93 0.28 0.34 0.066 0.11 0.18 0.46

The full cohort, as well as subgroups with mild (stages 1 to 2) and advanced CKD (stages 3 to 4) are shown. All numbers are presented as n (%) and mean
SD for categorical and continuous data, respectively. The p values are the comparisons between subgroups.

and mean
SD for discrete and continuous variables, respectively. Differences between baseline CKD stages were assessed using the chi-square and analysis of variance tests for discrete and continuous variables, respectively. eGFR measurements are presented as mean and standard error. Differences between preoperative and postoperative eGFR were assessed using paired t tests with the aforementioned model. Linear regression was performed to identify predictors of change in eGFR. Variables associated with change in eGFR as a main effect were entered into a full model upon which a backward stepwise approach was used to identify all predictors independently associated with change in eGFR. To assess the association between CKD stage and mortality before the investigation of covariates, a time-to-event analysis was performed using Kaplan-Meier estimator and a log-rank test. To investigate the relation of change in eGFR and mortality, a Cox proportional hazards model was constructed, adjusting for all relevant covariates.8 Time ¼ 0 was defined as the date of device implantation. For patients with no death documented, survival time was censored at the last visit or patient contact by electronic medical record review. An a level of 0.05 was considered statistically significant, and p values were not adjusted for multiple comparisons. Results There were 238 patients included in this study. Subjects were followed for up to 12 years with a mean follow-up time of 4.3
3.2 years. This cohort was primarily male, and the mean age was 65.9
10.4 years. A majority of patients had underlying coronary artery disease and advanced (New York Heart Association III/IV) HF. The mean QRS duration was 155.6
23.4, and a majority had a left bundle branch block. As expected, clinical characteristics differed between CKD stages as listed in Table 1. Compared with mild CKD, subjects with advanced CKD tended to be older, have a higher incidence of hypertension, diabetes, and atrial fibrillation and less likely to be treated

Table 2 Comparing change in eGFR by baseline renal disease stage BL CKD Stage

Pre eGFR

Post eGFR

Difference*

p*

I/II III/IV

78.5 (1.3) 45.6 (1.4)

67.8 (2.0) 46.8 (2.1)

10.8 (1.8) 1.4 (1.9)

<0.001 0.46

All numbers are presented as least-square means (standard error). * Controlled for follow-up time.

with an angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs). Over the first 6 months of CRT, there was a significant decrease in eGFR (p <0.001) in patients with mild CKD (stage I/II), whereas there was no significant change in renal function in patients with advanced CKD (p ¼ 0.46, Table 2). In the mild CKD cohort, the eGFR was 90.4
18.0 ml/min at follow-up in the subgroup with improved renal function, whereas it was 61.4
25.4 ml/min in the subgroup with worsened renal function. In the advanced CKD group, the eGFR at follow-up was 57.2
16.8 versus 38.2
11.5 ml/min among the subgroups with improved versus worsened renal function, respectively. Thus, the advanced renal failure group with improved renal function had very similar eGFR to the mild renal failure patients who had worsened renal function with CRT. Multivariate analysis showed that as main effects, baseline eGFR, ischemic heart disease, and ACE/ARB use were associated with change in eGFR; higher eGFR at baseline and ischemic heart disease were associated with a reduction in eGFR (3.2
0.6 and 5.5
2.5 units, respectively) whereas ACE/ARB was associated with an increase in eGFR (7.6
3.3). In a model controlling for baseline eGFR, neither the origin of HF (i.e., ischemic) nor ACE/ARB was significantly associated with change in eGFR. To assess this further, forest plots were constructed to evaluate the association of baseline characteristics with changes in eGFR. These results are shown in Figure 1. The only baseline

264

The American Journal of Cardiology (www.ajconline.org)

Figure 1. Forest plot of the association of clinical factors on the change in eGFR. LCL ¼ lower confidence limit; OR ¼ odds ratio; UCL ¼ upper confidence limit.

Figure 2. Kaplan-Meier curves, comparing baseline renal function and mortality with CRT. Advanced CKD subjects had higher mortality compared with mild CKD (p ¼ 0.034).

clinical factor that showed a statistically significant interaction with regard to change in eGFR was the baseline (i.e., preoperative) renal function. Both the baseline renal function and the change in eGFR were associated with changes in survival. Specifically, long-term survival was reduced in the subgroup with advanced baseline CKD (Figure 2). With regard to the change in eGFR with CRT, those patients with an improvement of renal function had better survival compared with those with worsening renal function (per 10 ml/min/1.73 m2 change, corrected hazard ratio 0.79, 95% CI 0.70 to 0.90, p <0.001). These results are listed in Table 3. The 5-year survival for the mild CKD cohort with improved renal function was 85% compared with 76% for

the mild CKD cohort with worsening renal function. The 5-year survival for the advanced CKD cohort with improved versus worsening renal function was 79% versus 60%, respectively (Figure 3). Thus, the advanced renal failure group with improved renal function had similar long-term survival to the mild renal failure patients who had worsened renal function with CRT (Figure 3). As noted previously, the eGFR was also similar in these 2 subgroups. These results provide further support that renal function with CRT is an important determinant of outcomes. Multivariate analysis showed that several other comorbidities were also independently associated with mortality including age, advanced HF class, ischemic cardiomyopathy, QRS duration, and amiodarone use (Table 3).

Heart Failure/Renal Function and Survival With CRT

265

Table 3 Results of multivariable analysis of eGFR and clinical factors on mortality Parameter Change in eGFR BL CKD Class Age Ischemic QRS Duration Advanced Heart Failure Amiodarone

Comparison

Hazard Ratio

95% CI

p

Per 10 ml/min/1.73m2 I/II vs. III/IV Per 10 years Ischemic vs. Non-ischemic Per 1 ms NYHA I/II vs III/IV Amiodarone vs. Not

0.79 0.44 1.28 1.72 0.99 3.36 2.03

0.70-0.90 0.26-0.75 1.01-1.61 1.03-2.89 0.98-1.00 1.33-8.46 1.08-3.83

< 0.001 0.002 0.040 0.040 0.014 0.010 0.028

Repercussion is likely outweighed by their mortality benefit after CRT-D implantation. CI ¼ confidence interval; HF ¼ heart failure; ms ¼ milliseconds.

Figure 3. Kaplan-Meier curves for the mild and advanced CKD at baseline. The subjects with an increase in eGFR at follow-up and those with a decreased eGFR at follow-up are shown separately for each CKD group.

Discussion The primary results of this study were that both the severity of baseline CKD and change in renal function with CRT-D are strongly associated with long-term survival. CRT-D patients with advanced CKD had a higher mortality than those with mild CKD. The subgroup with mild renal dysfunction had a significant worsening of renal function, whereas patients with advanced renal dysfunction had no significant change in eGFR. Regardless of the baseline stage of renal function, an improvement in renal function over the first 6 months of CRT was associated with decreased mortality. An important consideration in the recommendation of CRT with implantable cardiac defibrillator therapy is the prognosis of the patient.1 Accordingly, understanding the preoperative co-morbidities that impact survival is

important. In this regard, we show that renal function is an important predictor of survival as has been noted in several of the large landmark device trials.9,10 The effect of baseline renal function on survival with CRT-D has been studied previously with conflicting results. Lin et al11 showed that patients with eGFR < 60 ml/min/ 1.73 m2 had an increased mortality at 3-year follow-up after CRT-D implantation. Similar results were reported by Adelstein et al,5 who found that at 2 years, patients with eGFR > 30 ml/min/1.73 m2 had better survival than those with eGFR between 30 and 60 ml/min/1.73 m2 and patients with an eGFR < 30 ml/min/1.73 m2. These findings are consistent with several other small, retrospective studies.12e14 In contrast, neither the Resynchronization/Defibrillation for Ambulatory Heart Failure nor the Resynchronization Reserves Remodeling in Systolic Left

266

The American Journal of Cardiology (www.ajconline.org)

Ventricular Dysfunction trials found an independent effect of baseline renal function on survival, but advanced CKD was excluded from those studies.2,15,16 It is noteworthy that our study included follow-up to 12 years with a mean of 4.3 years providing prognostic data on more long-term therapy with CRT than reported previously. Importantly, inspection of the long-term mortality curves (Figure 2) indicate that the excess mortality associated with advanced CKD persists throughout the duration of follow-up. We observed that the change in renal function was dependent in part on the severity of baseline CKD. Similarly, Adelstein et al5 observed that patients with stage II or less renal dysfunction had a significant decrease in their renal function. Fung et al17 also found that patients with a decrease in eGFR with CRT also had better preimplantation renal function. In contrast, Lin et al11 found that subjects without significant baseline CKD had no significant change eGFR with CRT. The impact of baseline renal function on the change in GFR has also been studied previously with mixed results. Singal et al14 reported that stage 4 CKD had the highest rates of renal function improvement and that an improvement in renal function was associated with improved eventfree survival in this subgroup. Adelstein5 observed that an improvement in renal function was associated with better survival regardless of baseline renal function. Given the importance of the change in renal function on survival, identifying which subjects will have worsened eGFR with CRT is important for predicting prognosis preoperatively. However, we were unable to identify any clinical characteristics other than baseline renal function associated with a change in eGFR with CRT. This suggests that patients with worse baseline renal function may be more sensitive to the improved hemodynamic function with CRT to offset the progressive deleterious effect of HF and other co-morbidities on CKD. However, the proinflammatory and prothrombotic state of progressive renal failure may be responsible for the higher mortality even when accounting for the differences in baseline characteristics. This study should be interpreted in light of certain methodological limitations. This was a retrospective analysis. Consecutive patients from a prospectively constructed database were analyzed in an effort to mitigate any bias of a retrospective analysis. All-cause mortality was the end point of this study, but other important responses to CRT such as reverse remodeling, HF hospitalizations, percent biventricular pacing, or functional status were not assessed. Finally, we cannot exclude that other unmeasured variables may have affected renal function.

2.

3.

4.

5.

6. 7.

8.

9.

10.

11. 12.

13.

14.

15.

Disclosures The authors have no conflicts of interest to disclose. 1. Epstein AE, DiMarco JP, Ellenbogen KA, Estes NA 3rd, Freedman RA, Gettes LS, Gillinov AM, Gregoratos G, Hammill SC, Hayes DL, Hlatky MA, Newby LK, Page RL, Schoenfeld MH, Silka MJ, Stevenson LW, Sweeney MO; American College of Cardiology Foundation, American Heart Association Task Force on Practice Guidelines, Heart Rhythm Society. 2012 ACCF/AHA/HRS focused update incorporated into the ACCF/AHA/HRS 2008 guidelines for

16.

17.

device-based therapy of cardiac rhythm abnormalities: a report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. Circulation 2013;127:e283ee352. Cannizzaro LA, Piccini JP, Patel UD, Hernandez AF. Device therapy in heart failure patients with chronic kidney disease. J Am Coll Cardiol 2011;58:889e896. Harnett JD, Foley RN, Kent GM, Barre PE, Murray D, Parfrey PS. Congestive heart failure in dialysis patients: prevalence, incidence, prognosis and risk factors. Kidney Int 1995;47:884e890. Smith GL, Lichtman JH, Bracken MB, Shlipak MG, Phillips CO, DiCapua P, Krumholz HM. Renal impairment and outcomes in heart failure: systematic review and meta-analysis. J Am Coll Cardiol 2006;47:1987e1996. Adelstein EC, Shalaby A, Saba S. Response to cardiac resynchronization therapy in patients with heart failure and renal insufficiency. Pacing Clin Electrophysiol 2010;33:850e859. Jha V, Garcia-Garcia G, Iseki K, Li Z, Naicker S, Plattner B, Saran R, Wang AY, Yang CW. Chronic kidney disease: global dimension and perspectives. Lancet 2013;382:260e272. Levey AS, Coresh J, Greene T, Marsh J, Stevens LA, Kusek JW, VanLente F. Chronic Kidney Disease Epidemiology Collaboration, Expressing the modification of diet in renal disease study equation for estimating glomerular filtration rate with standardized serum creatinine values. Clin Chem 2007;53:766e772. Daly DD Jr, Maran A, Hyer JM, Funke F, Waring A, Cuoco FA, Sturdivant JL, Leman RB, Gold MR. The effect of chronic kidney disease on mortality with cardiac resynchronization therapy. Pacing Clin Electrophysiol 2016;39:863e869. Saxon LA, Bristow MR, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, Feldman AM, Galle E, Ecklund F. Predictors of sudden cardiac death and appropriate shock in the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) trial. Circulation 2006;114:2766e2772. Bristow MR, Saxon LA, Boehmer J, Krueger S, Kass DA, De Marco T, Carson P, DiCarlo L, DeMets D, White BG, DeVries DW, Feldman AM; Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) Investigators. Cardiacresynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med 2004;350:2140e2150. Lin G, Gersh BJ, Greene EL, Redfield MM, Hayes DL, Brady PA. Renal function and mortality following cardiac resynchronization therapy. Eur Heart J 2011;32:184e190. Bai R, Di Biase L, Elayi C, Ching CK, Barrett C, Philipps K, Lim P, Patel D, Callahan T, Martin DO, Arruda M, Schweikert RA, Saliba WI, Wilkoff B, Natale A. Mortality of heart failure patients after cardiac resynchronization therapy: identification of predictors. J Cardiovasc Electrophysiol 2008;19:1259e1265. Hosoda J, Ishikawa T, Matsushita K, Matsumoto K, Kimura Y, Miyamoto M, Ogawa H, Takamura T, Sugano T, Ishigami T, Uchino K, Kimura K, Umemura S. Impact of renal insufficiency on long-term clinical outcome in patients with heart failure treated by cardiac resynchronization therapy. J Cardiol 2012;60:301e305. Singal G, Upadhyay GA, Borgquist R, Friedman DJ, Chatterjee NA, Kandala J, Park MY, Orencole M, Dec GW, Picard MH, Singh JP, Mela T. Renal response in patients with chronic kidney disease predicts outcome following cardiac resynchronization therapy. Pacing Clin Electrophysiol 2015;38:1192e1200. Tang AS, Wells GA, Talajic M, Arnold MO, Sheldon R, Connolly S, Hohnloser SH, Nichol G, Birnie DH, Sapp JL, Yee R, Healey JS, Rouleau JL; Resynchronization-Defibrillation for Ambulatory Heart Failure Trial Investigators. Cardiac-resynchronization therapy for mildto-moderate heart failure. N Engl J Med 2010;363:2385e2395. Linde C, Abraham WT, Gold MR, St John Sutton M, Ghio S, Daubert C; REVERSE (REsynchronization reVErses Remodeling in Systolic left vEntricular dysfunction) Study Group. Randomized trial of cardiac resynchronization in mildly symptomatic heart failure patients and in asymptomatic patients with left ventricular dysfunction and previous heart failure symptoms. J Am Coll Cardiol 2008;52: 1834e1843. Fung JW, Szeto CC, Chan JY, Zhang Q, Chan HC, Yip GW, Yu CM. Prognostic value of renal function in patients with cardiac resynchronization therapy. Int J Cardiol 2007;122:10e16.