Effect of End-Stage Renal Failure and Hemodialysis

on Mortality Rates in Implantable

Cardioverter-Defibrillator Recipients
HAITHAM HREYBE, M.D., EATHAR RAZAK, M.D., and SAMIR SABA, M.D.
From the University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania

Background: Most defibrillator (ICD) trials have excluded patients on hemodialysis (HD). It is therefore not
known whether the ICD, when indicated, confers the same mortality benefit to HD and non-HD patients.
Method: HD patients implanted with an ICD from July 2001 to June 2004 were matched by age, gender,
left ventricular ejection fraction (LVEF), and class of heart failure to non-HD ICD recipients.
Results: Forty-six (16 on HD) patients (age = 65 15 yrs, LVEF = 30 14%, 44% in class III-IV HF) were
followed for a mean of 30 16 months (range, 461 months) after ICD implantation. During this period,
12/16 HD versus 9/30 non-HD patients died (P = 0.006). The two-year mortality rates were 54% and 29%
in the HD and non-HD groups, respectively (P = 0.01). After correcting for age, gender, race, LVEF, class
of HF, and ICD indication (primary vs. secondary prevention) in a Cox regression model, HD remained a
significant predictor of the time to death (HR = 2.9, adjusted P = 0.023).
Conclusion: Despite having an ICD, HD patients have approximately a three-fold increase in total mortality and may therefore not extract the same survival benefits from the ICD as their non-HD counterparts.
If duplicated in larger randomized trials, these results may demonstrate the futility of implanting defibrillators in HD patients. (PACE 2007; 30:10911095)
renal failure, dialysis, defibrillators, mortality
Introduction
End-stage renal disease constitutes a major
burden on the health-care system in the United
States with more than 300,000 patients nationwide
being treated with renal replacement therapy.1 Although mortality rates among such patients have
been declining over the past two decades, they remain in excess of 20% during the first year of dialysis, with more than half of the deaths being related
to cardiovascular events.1 Based on the United
States Renal Data System database, cardiac arrests
account for about 60% of deaths in patients with
end-stage renal disease2 with the annual mortality
rates among survivors of cardiac arrests reaching
up to 87%.2
Previous studies3,4 have shown that patients
with renal insufficiency or end-stage renal disease
are at a higher risk for ventricular tachyarrhythmia
and appropriate implantable cardioverter defibrillator (ICD) therapies. Other studies have suggested
a potential survival benefit of the ICD in patients on
hemodialysis therapy for renal failure.5 However,
most secondary68 and primary912 ICD trials have
excluded patients with advanced renal failure and
sources of available information on the benefits of

Address for reprints: Samir Saba, M.D., Chief, Electrophysiology Section, 200 Lothrop Street, UPMC Presbyterian, Suite
B-535, Pittsburgh, PA 15213-2582. Fax: 412-647-7979; e-mail:
sabas@upmc.edu
Received April 10, 2007; revised May 15, 2007; accepted June
8, 2007.

ICD in this patient population are limited to small,

nonrandomized clinical trials.13
In the current study, we hypothesize that patients with end-stage renal disease on hemodialysis (HD) have higher total mortality rates compared
to patients not on HD, despite ICD implantation.
We therefore designed this study to evaluate the
impact of HD on survival among patients with endstage renal disease implanted with an ICD.
Methods
Population
Sixteen patients on HD who underwent ICD
implantation for standard indications at the University of Pittsburgh Medical Center from July
2001 to June 2004 were included in the study. Each
HD patient was matched by age, gender, New York
Heart Association (NYHA) class of heart failure,
and left ventricular ejection fraction (LVEF) to two
non-HD ICD recipients, except for two HD patients
for whom only one control could be found. The
study population therefore consisted of 16 HD and
30 non-HD patients each implanted with an ICD.
Patients electronic medical records and cardiac
electrophysiology clinic records were analyzed for
baseline demographics, clinical characteristics including NYHA class of heart failure, LVEF, underlying heart disease, and indication for ICD implantation.
The primary analysis of this study was to determine the effect of HD on survival among ICD
patients as compared to their non-HD counterparts. Mortality data on patients who were lost to


C 2007, The Authors. Journal compilation 
C 2007, Blackwell Publishing, Inc.

PACE, Vol. 30

September 2007

1091

HREYBE, ET AL.

follow-up were obtained by searching the National

Social Security Death Index.
Statistical Analysis
Continuous variables were expressed as
mean standard deviation and were compared using the Students t-test. Categorical variables were
compared using the Chi-square test. Multivariate
analysis of the independent effect of HD on mortality was performed using a binary logistic regression model. Time-to-event curves during the
follow-up period were calculated by the KaplanMeier method and comparison between the HD
and non-HD groups was performed using the logrank test. Multivariate analysis of the independent
effect of HD on the time to death was performed
using the Cox proportional hazards model. A twotailed P value of <0.05 was considered significant.
All analyses were performed using SPSS version
14.0 (Chicago, IL, USA).
Results
Patients Characteristics
Sixteen patients with end-stage renal disease
on HD underwent ICD implantation at the University of Pittsburgh Medical Center between July
2001 and June 2004. They were matched by age,

gender, LVEF, and NYHA class of heart failure to 30

ICD recipients who were not on HD. Patients were
followed for a mean duration of 30 16 months
(range, 461 months). The demographic and clinical characteristics of HD patients and non-HD controls are presented in Table I. There were no significant differences between the two groups with
respect to age, race, gender, NYHA class of heart
failure, LVEF, or other associated medical conditions such as hyperlipidemia and smoking. Patients on HD were more likely to have diabetes
mellitus (56.3% vs. 26.7%, P = 0.06), but the difference did not reach statistical significance.
Incidence of Death Among Hemodialysis Patients
Patients on HD had higher all-cause mortality rates when compared with non-HD controls.
During the study period, 12/16 patients on HD
(75%) and 9/30 controls (30%) died (P = 0.006)
(Fig. 1). After correcting for age, gender, race, LVEF,
NYHA classification, ICD indication (primary vs.
secondary prevention of sudden cardiac death), diabetes mellitus, and hyperlipidemia in a binary logistic multivariate regression model, patients with
end-stage renal disease on HD remained at a higher
risk of death compared to controls (Hazard Ratio,
HR = 10.3, adjusted P = 0.026).

Table I.
Clinical Characteristics of Hemodialysis and Control Patients

Variable
Age (years)
Gender (% male)
Race (% white)
LVEF (%)
NYHA Class
I
II
III
IV
Diabetes mellitus
Hyperlipidemia
Class I antiarrhythmic drugs
-Blockers
Class III antiarrhythmic drugs
Calcium channel blockers
Smoking
Smoker
Non-smoker
Ex-smoker

1092

Patients on Hemodialysis
(n = 16)

Patients Not on Hemodialysis

(n = 30)

P-value

64.3 14.6
62.5%
75%
31.9 14.9

64.6 14.7
80%
66.7%
29.4 14.0

NS
NS
NS
NS

25%
25%
37.5%
12.5%
56.3%
81.3%
0%
81.3%
37.5%
12.5%

26.7%
33.3%
30%
10%
26.7%
53.3%
0%
70.0%
26.7%
3.3%

NS

18.8%
25%
56.3%

13.3
43.3
43.3

September 2007

0.06
NS
NS
NS
NS
NS
NS

PACE, Vol. 30

MORTALITY IN ICD PATIENTS ON HEMODIALYSIS

Causes of Mortality among Hemodialysis

Patients
During the follow-up period, a total of 21 (12
HD and nine non-HD) patients died. The cause of
mortality was determined in nine of those 21 patients (42.8%) from both study groups (four cases
and five controls). In five of nine patients, the cause
of mortality was due to decompensation of heart
failure and subsequent respiratory arrest. Three
patients died of noncardiovascular causes related
to sepsis, and one patient died intraoperatively
during surgery for kidney transplantation. No patients died of lethal arrhythmias. Details of causes
of death are provided in Table II.

Figure 1. Incidence of death in hemodialysis and nonhemodialysis patients.

Time to Death among Hemodialysis Patients

End-stage renal disease was also found to
be a powerful predictor of time to death. The
two-year mortality was 54% and 29% in the
HD and control groups, respectively (P = 0.01).
Kaplan-Meier curves were constructed to reflect
the time to death over the follow-up period and
are shown in Figure 2. After correcting for age,
gender, race, LVEF, NYHA classification, ICD indication (primary vs. secondary prevention of sudden cardiac death) in a Cox proportional hazards
model, HD remained a significant predictor of the
time to death (HR = 2.9, adjusted P = 0.023).

Discussion
As shown previously,3,4 patients with renal
insufficiency and in particular patients with endstage renal disease on HD are at a high risk for
recurrent ventricular tachyarrhythmia, which is
reflected in a higher incidence of ICD therapies
and a shorter time to first appropriate ICD shock
compared to patients with normal kidney function. The survival benefit of the ICD in HD patients
has been suggested by Herzog et al.,5 but not established, as these patients were excluded from all
major ICD trials.612 Also, the study by Herzog et
al. focused on the value of the ICD for secondary
prevention of death in HD patients, which is in
contrast to the primary prevention population included in our current study.
The implications of implanting HD patients
with defibrillators without proof of benefit are numerous. In HD patients who are prone to vascular
access problems, poor platelet function, and high
risk of infection, the invasive and expensive practice of implanting an ICD may negatively impact
patient outcome, thus the great need for clinical

Table II.
Cause of Death in Hemodialysis and Control Patients

Variable

Figure II. Time to death for hemodialysis and nonhemodialysis patients.

PACE, Vol. 30

Death
Patients with known
cause of death
Heart failure
Sepsis
During kidney
transplantation
Lethal arrhythmia

September 2007

Patients on Patients Not on

Hemodialysis Hemodialysis
(n = 16)
(n = 30)

12

0.006

4
1
2

5
4
1

NS

1
0

0
0

1093

HREYBE, ET AL.

trials that can better define the role of the ICD in

the cardiac patient on HD.
In this study, we show that patients with endstage renal disease on HD remain at a high risk of
death in spite of ICD implantation. Even after correcting for other mortality risk factors, patients on
HD were almost three times more likely to reach
the primary endpoint of death during follow-up.
The causes of these findings may be explained by
competing causes of death in this very sick population. HD patients may be at increased risk for heart
failure secondary to fluid overload.14 They may
also be at increased risk for death from noncardiac
causes. Patients with indwelling dialysis catheters
are more likely to suffer from vascular complications and to acquire nosocomial infections that can
disseminate and lead to sepsis, end-organ failure,
and death.15 In our study, patients did not die of arrhythmias but rather of advanced heart failure and
noncardiac causes. The all-cause mortality was so
elevated, however, that the benefit of the ICD in
this sick population was rendered marginal, if not
completely diluted.
The United States Renal Data System2 can provide comparative data regarding the rates of death
in HD patients in general compared to our HD patients with ICD. The yearly mortality rates for HD
patients around 65 years of age (which is the mean
age in our HD group) in this data system ranges
from 217 to 290 per 1,000 patient-years, for the
years 1980 to 1992. It is safe to assume that during this period of time, most HD patients did not
have an ICD. Despite being implanted with an ICD,
the incidence of deaths in our HD study group
was comparable these national rates at 75% risk of
death over a mean follow-up period of 30 months,

which translates into a mortality of 300 per 1,000

patient-years. This provides yet another support
to the fact that the ICD may not confer protection
against mortality in the dialysis population.
The presence of a large discrepancy in the
prevalence of diabetes mellitus between the two
study groups can be explained by the fact that diabetes mellitus is a known risk factor for developing
renal failure requiring HD therapy. This discrepancy did not reach statistical significance because
of the relatively small size of the current study. It
is highly unlikely that the difference in the prevalence of diabetes mellitus accounted for the higher
mortality rates in the HD group since this risk factor was accounted for in the multivariate binary
logistic regression model.
Our study has some limitations. First, the data
were collected retrospectively and therefore its accuracy depends on the accuracy of the electronic
medical records and clinic charts. Second, the
cause of death among patients who were lost to
follow-up at our institution could not be ascertained. Third, our study did not include a control group of HD patients not implanted with an
ICD.
In summary, our data suggest that HD patients
have increased mortality rates despite the implantation of ICD and extract little, if any, survival benefit from this therapy. These patients have very
high all-cause mortality rates secondary to their
other medical conditions associated with renal
failure and are therefore unlikely to benefit from
an ICD. Further prospective, randomized studies
are needed, however, to establish the potential effect of the defibrillator on the overall mortality in
patients with end-stage renal disease on HD.

References
1. US Renal Data System. YSRDS Annual Data Report. Bethesda, MD,
National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases, April 2000.
2. http://www.usrds.org
3. Hreybe H, Ezzeddine R, Bedi M, Barrington W, Bazaz R, Ganz LI,
Jain S, et al. Renal insufficiency predicts the time to first appropriate defibrillator shock. Am Heart J 2006; 151:852856.
4. Robin J, Weinberg K, Tiongson J, Carnethon M, Reddy M, Ciaccio
C, Quadrini M, et al. Renal dialysis as a risk factor for appropriate
therapies and mortality in implantable cardioverter-defibrillator recipients. Heart Rhythm 2006; 3:11961201.
5. Herzog CA, Li S, Weinhandl ED, Strief JW, Collins AJ, Gilbertson
DT. Survival of dialysis patients after cardiac arrest and the impact
of implantable cardioverter defibrillators. Kidney Int 2005; 68:818
825.
6. The Antiarrhythmics versus Implantable Defibrillators (AVID) Investigators. A comparison of antiarrhythmic-drug therapy with
implantable defibrillators in patients resuscitated from nearfatal ventricular arrhythmias. N Engl J Med 1997; 337:1576
1583.
7. Connolly SJ, Gent M, Roberts RS, Dorian P, Roy D, Sheldon RS, Mitchell LB, et al. Canadian implantable defibrillator
study (CIDS): A randomized trial of the implantable cardioverter
defibrillator against amiodarone. Circulation 2000; 101:1297
1302.

1094

8.

9.

10.

11.

12.

Kuck KH, Cappato R, Siebels J, Ruppel R. Randomized comparison of antiarrhythmic drug therapy with implantable defibrillators in patients resuscitated from cardiac arrest: The Cardiac Arrest Study Hamburg (CASH). Circulation. 2000; 102:748
754.
Buxton AE, Lee KL, DiCarlo L, Gold MR, Greer GS, Prystowsky
EN, OToole MF, et al. Electrophysiologic testing to identify patients with coronary artery disease who are at risk for sudden death.
Multicenter Unsustained Tachycardia Trial Investigators. N Engl J
Med 2000; 342:19371945.
Moss AJ, Hall WJ, Cannom DS, Daubert JP, Higgins SL, Klein H,
Levine JH, et al. Improved survival with an implanted defibrillator
in patients with coronary disease at high risk for ventricular arrhythmia. Multicenter Automatic Defibrillator Implantation Trial
Investigators. N Engl J Med 1996; 335:19331940.
Moss AJ, Zareba W, Hall WJ, Klein H, Wilber DJ, Cannom DS,
Daubert JP, et al. Multicenter Automatic Defibrillator Implantation
Trial II Investigators. Prophylactic implantation of a defibrillator in
patients with myocardial infarction and reduced ejection fraction.
N Engl J Med 2002; 346:877883.
Bardy GH, Lee KL, Mark DB, Poole JE, Packer DL, Boineau R, Domanski M, et al. Sudden Cardiac Death in Heart Failure Trial (SCDHeFT) Investigators. Amiodarone or an implantable cardioverterdefibrillator for congestive heart failure. N Engl J Med 2005;
352:225237.

September 2007

PACE, Vol. 30

MORTALITY IN ICD PATIENTS ON HEMODIALYSIS

13. Wase A, Basit A, Nazir R, Jamal A, Shah S, Khan T, Mohiuddin I,
et al. Impact of chronic kidney disease upon survival among implantable cardioverter-defibrillator recipients. J Intervent Cardiac
Electrophysiol 2004; 11:199204.
14. Cheung AK, Sarnak MJ, Yan G, Berkoben M, Heyka R, Kaufman

PACE, Vol. 30

A, Lewis J, et al. For the HEMO study group. Cardiac diseases in

maintenance hemodialysis patients: Results of the HEMO Study.
Kidney Int 2004; 65:23802389.
15. Vanholder R, Van Biesen W. Incidence of infectious morbidity and
mortality in dialysis patients. Blood Purif 2002; 20:477480.

September 2007

1095