Interdialytic Weight Gain, Compliance With Dialysis Regimen,

and Age Are Independent Predictors of Blood Pressure

in Hemodialysis Patients
Mahboob Rahman, MD, Pingfu Fu, MS, Ashwini R. Sehgal, MD, and Michael C. Smith, MD

● Hypertension is a common problem in patients undergoing chronic hemodialysis. The purpose of this study is to
identify the clinical and demographic factors independently associated with blood pressure in this population. Data
collected for the Dialysis Morbidity and Mortality Study Wave 1 by the US Renal Data System were analyzed. The
mean predialysis blood pressure for this cohort of 5,369 patients was 149/79 mm Hg. Sixty-three percent of the
patients were hypertensive; 27%, 25%, and 11% had stages 1, 2, and 3 hypertension, respectively. Young age, black
race, male sex, diabetes as cause of end-stage renal disease, erythropoietin therapy, and smoking were associated
with higher blood pressure in the univariate analysis. Patients skipping or shortening one or more dialysis
treatments had higher blood pressure. The presence of congestive heart failure and coronary heart disease was
associated with lower blood pressure. On multivariate analysis, high interdialytic weight gain, noncompliance with
dialysis regimen, and younger age were independent predictors of higher blood pressure. In summary, hyperten-
sion is common and poorly controlled in patients undergoing chronic hemodialysis. Greater interdialytic weight
gain and noncompliance with dialysis regimen are independently associated with higher blood pressure, and
advancing age is associated with lower blood pressure levels in this population. Therapeutic regimens emphasiz-
ing reduction of interdialytic weight gain and improved compliance with the dialysis regimen need to be evaluated
for improving the management of hypertension. The effect of age and other comorbid conditions, particularly
cardiovascular disease, must be considered while studying the relationship between blood pressure and mortality
in patients undergoing chronic hemodialysis.
娀 2000 by the National Kidney Foundation, Inc.
INDEX WORDS: Hemodialysis; hypertension; interdialytic weight gain; compliance.

cular disease,12 a better understanding of the

H YPERTENSION is a common and difficult
clinical problem in patients undergoing
chronic hemodialysis.1-6 In recent years, there
has been increasing interest in defining the role
of hypertension as a risk factor for cardiovascu-
lar disease in this population, developing tech-
niques to improve blood pressure control, and
assessing their effect on cardiovascular morbid-
ity and mortality.7 Several authorities have em-
phasized the urgent need to conduct prospective
trials to provide guidelines for optimal manage-
ment of hypertension in these high-risk pa-
tients.8,9
Our approach has been to identify the barriers
to adequate blood pressure control in this popula-
tion.10,11 We reasoned that identification of fac-
tors independently associated with blood pres-
sure is a necessary first step in designing strategies
to overcome these barriers. These studies will
then provide a rational basis for therapeutic inter-
ventions in prospective clinical trials.
In addition, the relationship between blood
pressure and mortality in patients undergoing
chronic hemodialysis is complex in the sense
that low, rather than high, blood pressure has
been consistently associated with increased mor-
tality.12,13 Although this has conventionally been
attributed to the presence of coexistent cardiovas-
effect of comorbid conditions on blood pressure
in these patients is needed to define the role of
hypertension as an independent risk factor for
morbidity and mortality.
The purpose of this study is to identify the
clinical and demographic factors independently
associated with blood pressure in a large random
From the Divisions of Hypertension and Nephrology,
Department of Medicine, Case Western Reserve University/
University Hospitals of Cleveland, Cleveland, OH; the Divi-
sion of Nephrology, Department of Medicine, Case Western
Reserve University/Metrohealth Medical Center, Cleveland,
OH; and the Department of Epidemiology and Biostatistics,
Case Western Reserve University, Cleveland, OH.
Received April 29, 1999; accepted in revised form August
20, 1999.
The data reported here have been supplied by the US
Renal Data System. The interpretation and reporting of
these data are the responsibility of the authors and in no way
should be seen as an offıcial policy or interpretation of the
US government.
Address reprint requests to Mahboob Rahman, MD, Divi-
sions of Hypertension and Nephrology, Department of Medi-
cine, Case Western Reserve University/University Hospitals
of Cleveland, 11100 Euclid Ave, Cleveland, OH 44106.
E-mail: mxr9@po.cwru.edu
娀 2000 by the National Kidney Foundation, Inc.
0272-6386/00/3502-0027$3.00/0

American Journal of Kidney Diseases, Vol 35, No 2 (February), 2000: pp 257-265 257
258
sample of patients undergoing chronic hemodi-
alysis.
METHODS
Data collected for the Dialysis Morbidity and Mortality
Study (DMMS) Wave 1 by the US Renal Data System
(USRDS) were used for this study. The DMMS was an
observational study in which demographic, comorbidity,
laboratory, treatment, socioeconomic, and insurance data
were collected for a large random sample of US dialysis
patients using the patients’ dialysis records.14 A sample of
dialysis units (n ⫽ 550) was randomly selected for Wave I of
the DMMS from the Master List of Medicare-Approved
Dialysis Facilities as of December 31, 1993. The sample of
patients for Wave 1 was selected from a national census of
hemodialysis patients as of December 31, 1993. Patients
were excluded if they were aged younger than 15 years, in
training for self-care treatment, or undergoing continuous
ambulatory peritoneal dialysis, home hemodialysis, or other
dialysis on December 31, 1993. From the 550 selected
facilities, 6,300 in-center hemodialysis patients were se-
lected for inclusion onto the study. Data were abstracted
from patient medical records by dialysis unit personnel, and
completed questionnaires were sent to the USRDS. Several
measures were taken to confirm the validity of data abstrac-
tion, described in the USRDS Researchers Guide.14
Predialysis blood pressure was abstracted by the clinic
personnel as the average of the last three values from the
patient’s chart in the last week of 1993. Patient compliance
with dialysis regimen was measured in two ways: the num-
ber of dialysis treatments skipped between December 1 and
23, 1993, and the number of treatments shortened by 10
minutes or greater in December 1993, not including skipped
treatments. Interdialytic weight gain was calculated from the
predialysis and postdialysis weights collected for six consecu-
tive treatments. Comorbid conditions were defined as those
present within 10 years of the study start date, determined by
personnel at the site. We defined coronary heart disease as
the presence of a prior history of coronary heart disease
and/or coronary artery disease, angina, myocardial infarc-
tion, coronary artery bypass surgery, coronary angioplasty,
abnormal coronary angiography, or cardiac arrest.
The primary outcome variables of interest were predialy-
sis systolic and diastolic blood pressures and mean arterial
pressure. Mean arterial pressure was defined as follows:
Mean arterial pressure
⫽ ([systolic blood pressure ⫺ diastolic blood pressure]/3)
⫹ diastolic blood pressure
Student’s t-test, chi-squared test, and analysis of variance
with adjustment for multiple comparisons were used for
univariate analysis. Pearson’s correlation coefficient was
used to study the relationship between two continuous
variables. P less than 0.05 is considered significant for all
tests. Variables found to be significant at the level of P less
than 0.1 in the univariate analysis or considered clinically
significant were further examined by multiple linear regres-
sion (PROC GLM in SAS; SAS Institute, Cary, NC) to
adjust for the effect of multiple covariates and determine
RAHMAN ET AL
independent predictors of mean arterial pressure. SAS soft-
ware version 6.12 was used for all data analyses.
RESULTS
The descriptive characteristics of the study
population (n ⫽ 5,369) are listed in Table 1. Age,
race, sex, and cause of end-stage renal disease
(ESRD) in this cohort are similar to those in the
overall US hemodialysis population.15
The mean predialysis blood pressure for the
entire cohort was 149/79 mm Hg (Table 2). As
defined by the Sixth Joint National Committee
Classification of Blood Pressure,16 1,958 patients
(37%) were normotensive. Sixty-three percent
were hypertensive; 27%, 25%, and 11% had
stages 1, 2, and 3 hypertension, respectively.
Blood pressure was highest in the youngest
patients (in the first quartile of age) and declined
with each quartile of advancing age (Table 3).
The oldest patients (in the highest quartile of
age) had the lowest systolic and diastolic pres-
sures.
Black patients had higher blood pressure (Table
4) and were less likely to be in the normotensive
group compared with whites (34% versus 39%;
P ⬍ 0.05). Men had a higher diastolic blood
pressure compared with women (80 ⫾ 12 versus
78.1 ⫾ 12.4 mm Hg; P ⬍ 0.05). Patients receiv-
ing erythropoietin therapy had higher systolic
blood pressure compared with those not receiv-
ing erythropoietin (150.4 ⫾ 22.3 versus 145.9 ⫾
21.8 mm Hg; P ⬍ 0.05). Twenty-two percent of
patients were smokers and had significantly
higher blood pressure than nonsmokers (Table
Table 1. Descriptive Characteristics of the Study
Population
Age (y) 59.7 ⫾ 15.8
Race
White 2,713 (52)
Black 2,077 (40)
Hispanic 298 (6)
Other 193 (2)
Sex (% men) 2,729 (51)
Cause of ESRD
Diabetes 1,840 (34)
Hypertension 1,666 (31)
Chronic glomerulonephritis 1,065 (20)
Polycystic kidney disease 193 (4)
Other 596 (11)
Years of dialysis 3.34 ⫾ 3.42
NOTE. N ⫽ 5,369. Values are expressed as ⫾ SD or
number (percent).
HYPERTENSION IN HEMODIALYSIS PATIENTS 259

Table 2. Prevalence of Hypertension crit, albumin level, or length of dialysis treat-

in the Study Population ment among the four groups (Table 5).
Systolic Diastolic The effect of compliance with dialysis treat-
Blood Blood ments on blood pressure is listed in Table 6.
No. of Pressure Pressure Patients with stage 3 hypertension were more
Patients (mm Hg) (mm Hg)
likely to skip treatments (0.99 ⫾ 2.2 treatments
All patients 5,369 149 ⫾ 22.2 79 ⫾ 12.2 over a 3-week period versus 0.73 ⫾ 1.9, 0.75 ⫾
Normotensive 1.8, and 0.67 ⫾ 1.6 treatments in normotensive,
(⬍140/90 stage 1, and stage 2 hypertensive patients, respec-
mm Hg) 1,958 (37) 127 ⫾ 12.0 70.9 ⫾ 9.5
tively; P ⬍ 0.05) and shorten more dialysis
JNC stage I (140-
159/90-99 treatments (0.29 ⫾ 1.02 versus 0.18 ⫾ 0.79,
mm Hg) 1,445 (27) 149 ⫾ 4.8 78 ⫾ 9.3 0.19 ⫾ 0.87, and 0.21 ⫾ 0.82 treatments in
JNC stage II normotensive, stage 1, and stage 2 hypertensive
(160-179/ patients, respectively; P ⬍ 0.05). Conversely,
100-109
mm Hg) 1,338 (25) 165 ⫾ 6.4 84.6 ⫾ 10.9
patients who skipped or shortened one or more
JNC stage III dialysis treatments had significantly higher dia-
(ⱖ180/ⱖ110 stolic pressures (Table 6).
mm Hg) 600 (11) 187 ⫾ 11.3 94 ⫾ 14.19 Forty-three percent of the study population
(2,175 patients) had a clinical diagnosis of con-
NOTE. Values are expressed as mean ⫾ SD or number
(percent).
gestive heart failure and had lower diastolic
Abbreviation: JNC, Sixth Report of the Joint National pressures than those without congestive heart
Committee on Prevention, Detection, Evaluation, and Treat- failure (77.8 ⫾ 13.1 versus 80.0 ⫾ 12.6 mm Hg;
ment of High Blood Pressure, 1997. P ⬍ 0.05). Similarly, patients with coronary
heart disease (n ⫽ 1,645; 31% of the study
population) had lower diastolic blood pressures
4). Patients with diabetes as the cause of ESRD than patients without coronary heart disease
had the highest systolic but lowest diastolic blood (76.5 ⫾ 12.2 versus 80.5 ⫾ 13.4 mm Hg; P ⬍
pressure compared with other causes of ESRD 0.05). The presence of left ventricular hypertro-
(Table 4). phy was associated with a higher systolic blood
Normotensive patients received dialysis for a pressure when assessed both by echocardiogra-
longer period of time than hypertensive patients phy and electrocardiography (Table 7).
(3.5 ⫾ 3.76 versus 3.16 ⫾ 3 years in patients Interdialytic weight gain was greatest in pa-
with stage 2 hypertension; P ⬍ 0.05). There were tients with stage 3 hypertension (2.98 ⫾ 2.1
no clinically significant differences in hemato- versus 2.63 ⫾ 2.14, 2.75 ⫾ 3.56, and 2.82 ⫾
1.91 kg in the normotensive, stage 1, and stage 2
Table 3. Age and Blood Pressure hypertensive groups, respectively), although this
difference did not achieve statistical significance.
Systolic Blood Diastolic Blood The relationship between interdialytic weight
No. of Pressure Pressure
Age Patients (mm Hg) (mm Hg)
gain and blood pressure stratified by level of
blood pressure is listed in Table 8. Greater inter-
First quartile dialytic weight gains were associated with higher
(⬍48.5 y) 1,387 150.42 ⫾ 22.3 85.45 ⫾ 13.7* mean arterial pressures in patients with stage 3
Second quartile hypertension, but not in normotensive patients or
(48.5-62.4 y) 1,394 152.14 ⫾ 22.6 80.6 ⫾ 12.3*
Third quartile
those with less severe hypertension. Similarly,
(62.4-71.7 y) 1,397 149.82 ⫾ 21.5 76.9 ⫾ 11.5* there was a significant correlation between blood
Fourth quartile pressure and interdialytic weight gain in patients
(⬎71.7 y) 1,390 146.27 ⫾ 21.9† 73.4 ⫾ 10.6* with stage 3 hypertension (r ⫽ 0.16 and 0.14 for
systolic and diastolic blood pressures, respec-
NOTE. Values are expressed as mean ⫾ SD.
*P ⬍ 0.05, diastolic blood pressure in each quartile.
tively; P ⬍ 0.05) and not in the other groups (r ⫽
†P ⬍ 0.05, systolic blood pressure fourth quartile versus 0.02 and 0.07 in the normotensive group; r ⫽
each quartile. 0.01 and 0.07 in stage 1 hypertension; r ⫽
260 RAHMAN ET AL

Table 4. Effect of Race, Sex, Erythropoetin Use, Smoking, and Cause of ESRD on Blood Pressure

No. of Systolic BP Diastolic BP

Patients (mm Hg) P (mm Hg) P

Race
White 2,697 148 ⫾ 22.9 0.0001 77.0 ⫾ 12.6 0.0001
Black 2,066 151 ⫾ 21.7 82 ⫾ 13.0
Sex
Men 2,729 149.2 ⫾ 22 0.13 80 ⫾ 12 0.0001
Women 2,649 150.1 ⫾ 22.4 78.1 ⫾ 12.4
Erythropoietin use
Yes 4,338 150.4 ⫾ 22.27 0.0001 79.25 ⫾ 12.98 0.32
No 749 145.9 ⫾ 21.76 78.77 ⫾ 12.15
Smoking
Smokers 1,093 151.2 ⫾ 22.63 0.02 81.88 ⫾ 13.8 0.0001
Nonsmokers 3,793 149.5 ⫾ 22.26 78.47 ⫾ 12.07
Cause of ESRD
Diabetes 1,840 153.4 ⫾ 22* 78.2 ⫾ 12.36†
Hypertension 1,666 150.4 ⫾ 22.8‡ 79.9 ⫾ 13
Chronic glomerulonephritis 1,065 146.7 ⫾ 21.7§ 80 ⫾ 13
Polycystic kidney disease 193 146.7 ⫾ 22 81.3 ⫾ 12.6
Other 596 142.8 ⫾ 22.8 77.6 ⫾ 13.6㛳

NOTE. Values are expressed as mean ⫾ SD.

*P ⬍ 0.05, diabetes versus hypertension, chronic glomerulonephritis, polycystic kidney disease, and other.
†P ⬍ 0.05, diabetes versus hypertension, chronic glomerulonephritis, and polycystic kidney disease.
‡P ⬍ 0.05, hypertension versus chronic glomerulonephritis and other.
§P ⬍ 0.05, chronic glomerulonephritis versus other.
㛳P ⬍ 0.05, other versus hypertension, chronic glomerulonephritis, and polycystic kidney disease.

–0.002 and 0.01 in stage 2 hypertension for
systolic and diastolic pressures, respectively;
P ⫽ not significant).
Results of the multivariate analysis are listed
in Table 9. Compliance with dialysis treatment,
age, and interdialytic weight gain were indepen-
dent predictors of mean arterial pressure after
adjustment for other covariates.
The presence of congestive heart failure had a
significant influence on the relationship between
blood pressure and other predictor variables. In
patients with congestive heart failure, there was
no significant change in mean arterial pressure
with increasing interdialytic weight gain, whereas
in patients without congestive heart failure, in-
creasing interdialytic weight gain was associated
with higher blood pressure (Table 10). Similarly,
the effect of poor compliance with dialysis treat-
ment on blood pressure was more marked in
patients with congestive heart failure. Shortening
one or more dialysis treatments in a 1-month
period was associated with higher blood pressure
in patients with congestive heart failure (mean
arterial pressure, 104.12 ⫾ 15.58 versus
100.82 ⫾ 14.38 mm Hg in patients with no
shortened dialysis treatments in the 1-month pe-

Table 5. Descriptive Features of the Study Population Stratified by Level of Blood Pressure

Stage 1 Stage 2 Stage 3

Normotensive Hypertension Hypertension Hypertension
(n ⫽ 1,958) (n ⫽ 1,445) (n ⫽ 1,338) (n ⫽ 600)

Weight (kg) 68 ⫾ 17.8 70 ⫾ 18.6 71 ⫾ 17.8* 69.5 ⫾ 21.8

Years of dialysis 3.5 ⫾ 3.76 3.25 ⫾ 3.35 3.16 ⫾ 3* 3.46 ⫾ 3.2
Hematocrit (%) 30.9 30.4 30.0 29.9
Albumin (mg/dL) 3.6 ⫾ 0.4 3.7 ⫾ 0.4 3.7 ⫾ 0.4 3.7 ⫾ 0.4
Duration of dialysis (min) 192.3 ⫾ 29.85 191.8 ⫾ 31.4 194.5 ⫾ 29.8 193.3 ⫾ 30.2

NOTE. Values are expressed as ⫾ SD unless noted otherwise.

*P ⬍ 0.05, stage 2 hypertension versus normotensive.
HYPERTENSION IN HEMODIALYSIS PATIENTS 261

Table 6. Effect of Compliance With Dialysis Regimen on Blood Pressure

No. of Systolic Blood Diastolic Blood

Patients Pressure (mm Hg) P Pressure (mm Hg) P

No. of skipped dialysis treatments during

December 1-23, 1993
0 4,545 149.47 ⫾ 22.08 0.02 78.65 ⫾ 12.66 0.0001
ⱖ1 497 151.87 ⫾ 23.01 83.11 ⫾ 13.91
No. of shortened dialysis treatments
during December 1993
0 3,693 149.34 ⫾ 22.08 0.12 78.42 ⫾ 12.5 0.0001
ⱖ1 1,335 150.45 ⫾ 22.52 80.71 ⫾ 13.55

NOTE. Values are expressed as mean ⫾ SD.

riod; P ⫽ 0.0001), but not in patients without going chronic hemodialysis.1-6 At the present
congestive heart failure (mean arterial pressure, time, there is little consensus regarding optimal
102.87 ⫾ 13.87 versus 103.81 ⫾ 14.4 mm Hg; management strategies in hypertensive hemodi-
P ⫽ not significant). However, the presence of alysis patients. To design and implement effec-
congestive heart failure was not an independent tive therapeutic regimens, it is important to iden-
predictor of blood pressure in the multivariate tify the clinical and demographic factors that
analysis (Table 9). contribute to higher blood pressure in these
patients. In previous studies in a single hemodi-
DISCUSSION alysis facility, we showed that excessive interdia-
It is well recognized that hypertension is com- lytic weight gain and submaximal antihyperten-
mon and often poorly controlled in patients under- sive drug therapy were associated with poor
blood pressure control.11 The aim of the present
Table 7. Association Between Coexistent study is to determine the factors associated with
Cardiovascular Disease and Blood Pressure high blood pressure in a large random sample of
maintenance hemodialysis patients. The DMMS
Systolic Diastolic
Blood Blood
study offered a unique opportunity to examine
No. of Pressure Pressure this issue because of the large sample size and
Patients (mm Hg) (mm Hg) the strength of its sampling and data collection
techniques.
Congestive
heart failure The prevalence of uncontrolled hypertension
Present 2,175 149.4 ⫾ 22.0 77.8 ⫾ 13.1*
Absent 2,911 149.4 ⫾ 21.7 80.0 ⫾ 12.6 Table 8. Interdialytic Weight Gain and Blood Pressure
Coronary heart Stratified by Blood Pressure Levels
disease
Present 1,645 149.4 ⫾ 22.9 76.5 ⫾ 12.29* Interdialytic Weight Gain (kg)
Absent 3,639 149.72 ⫾ 22.2 80.5 ⫾ 13.4
Left ventricular ⬍2 2-4 ⬎4
hypertrophy
by echocar- Normotensive
diography (n ⫽ 670) 88.5 ⫾ 9.1 90.6 ⫾ 8.8 90.3 ⫾ 9.9
Present 1,034 151.6 ⫾ 22.7* 79.59 ⫾ 13.4 JNC stage 1
Absent 2,849 148.9 ⫾ 21.6 79.07 ⫾ 12.7 (n ⫽ 490) 102 ⫾ 6.6 102.8 ⫾ 6.5 103 ⫾ 6.1
Left ventricular JNC stage 2
hypertrophy (n ⫽ 447) 111 ⫾ 8.2 112.8 ⫾ 7.5 111.2 ⫾ 6.5
by ECG JNC stage 3
Present 1,480 151.98 ⫾ 22.5* 79.3 ⫾ 13.3 (n ⫽ 216) 123.8 ⫾ 9.8 125.6 ⫾ 9.8 128.9 ⫾ 12.3*
Absent 3,160 148.5 ⫾ 21.7 79.0 ⫾ 12.7
NOTE. Mean arterial pressure (mmHg). Values are ex-
NOTE. Values are expressed as mean ⫾ SD. pressed as mean ⫾ SD.
Abbreviation: ECG, electrocardiogram. *P ⬍ 0.05, interdialytic weight gain greater than 4 kg
*P ⬍ 0.05, present versus absent. versus less than 2 kg.
262 RAHMAN ET AL

Table 9. Multiple Linear Regression Analysis With Several common demographic factors were
Mean Arterial Pressure as Dependent Variable found to affect blood pressure in this study.
SE of Black patients had higher blood pressures com-
Parameter Estimate Estimate T P pared with other racial/ethnic groups. Although
not unexpected from our knowledge of epidemio-
Intercept 107.66 4.10 26.20 0.0001
logical characteristics of essential hypertension
Age ⫺0.17 0.03 ⫺5.75 0.0001
Race in blacks,19 this study confirms that a similar
Black 2.91 2.08 1.40 0.16 difference exists in patients undergoing chronic
White 0.01 2.02 0.01 0.99 hemodialysis. Male sex, smoking, and erythropoi-
Sex (men) 1.23 0.90 1.37 0.17 etin use were, as expected, associated with higher
Cause of ESRD
Diabetes ⫺0.82 2.21 ⫺0.37 0.7
blood pressure.
Hypertension 0.91 2.2 0.41 0.67 There were significant differences in blood
Chronic glomeru- pressure depending on the underlying cause of
lonephritis ⫺1.24 2.2 ⫺0.56 0.57 ESRD. Patients with diabetes as the cause of
Years on dialysis ⫺0.11 0.12 ⫺0.9 0.36 ESRD had the highest systolic blood pressure
Erythropoietin use 0.68 1.24 0.55 0.58
Congestive heart
and pulse pressure compared with other disease
failure 0.49 0.92 0.53 0.59 categories. It is intriguing to speculate whether
Coronary heart dis- these differences in blood pressure contribute to
ease ⫺0.84 0.96 ⫺0.88 0.38 the overall poorer prognosis of patients with
Interdialytic weight diabetes undergoing chronic hemodialysis.20
gain 0.51 0.18 2.81 0.0051
Compliance with There are increasing data showing the deleteri-
dialysis (ⱖ1 ous effects of noncompliance with dialysis regi-
treatment men on overall morbidity and mortality.21 Our
skipped) 4.3 1.4 2.95 0.0033 study confirms that noncompliant behavior, such
Smoking 0.48 1.03 0.46 0.64
as shortening or skipping dialysis treatments, is
Weight 0.03 0.02 1.24 0.21
associated with higher blood pressure. This may
be caused by inadequate ultrafiltration with con-
in the study population was 63%, which is simi- tinued volume overload resulting in hyperten-
lar to data that we and other investigators re- sion or may serve as a marker that the same
ported.11,17,18 Although the true prevalence of patients may be noncompliant with antihyperten-
hypertension is undoubtedly underestimated be- sive medication or dietary salt restriction. The
cause of lack of data regarding the use of antihy- efficacy of multidisciplinary programs aimed at
pertensive drug therapy, these data confirm that improving compliance with the dialysis regimen
hypertension continues to be a challenging prob- in improving blood pressure control needs to be
lem in the care of patients undergoing chronic evaluated.
hemodialysis. A vexing issue facing clinicians and research-
ers in this area is the relationship between hyper-
Table 10. Relationship Between Blood Pressure and tension and mortality and whether the conven-
Interdialytic Weight Gain Stratified by the Presence of tional wisdom of associating high blood pressure
Congestive Heart Failure with increased mortality is applicable to this
Patients With Patients Without population. Whereas a large volume of data sup-
Interdialytic Congestive Congestive ports the contention that hypertension is an inde-
Weight Heart Failure Heart Failure pendent risk factor for morbidity and mortality in
Gain (kg) (n ⫽ 782) (n ⫽ 943)
hemodialysis patients,12,22-31 several recent stud-
⬍2 99.86 ⫾ 15.02 101.69 ⫾ 13.86 ies have shown little or no association between
2-4 102.81 ⫾ 14.76 104.88 ⫾ 14.12* high blood pressure levels and adverse out-
⬎4 102.8 ⫾ 15.09 106.98 ⫾ 15.34* comes.32,33 A relationship has been consistently
shown between low blood pressure and mortal-
NOTE. Mean arterial pressure (mm Hg). Values are
expressed as mean ⫾ SD.
ity.12,13 The conventional explanation for this
*P ⬍ 0.05 versus interdialytic weight gain greater than observation has been that low blood pressures
2 kg. are caused by coexistent cardiovascular disease,
HYPERTENSION IN HEMODIALYSIS PATIENTS
particularly congestive heart failure, which is
associated with very high mortality. However,
other than the series from Foley et al,12 there are
limited data that this is the case. Our study
provides two important pieces of data to under-
stand this apparent paradox. First, advancing age
was strongly and independently associated with
lower blood pressure. Second, patients with con-
gestive heart failure and coronary heart disease
had lower blood pressure in the univariate analy-
sis, although after adjustment for other covari-
ates, this did not remain statistically significant.
Because both older age and congestive heart
failure are independent risk factors for increased
mortality in hemodialysis patients, it is possible
that the studies showing the association between
low blood pressure and high mortality may re-
flect the effect of age and coexistent cardiovascu-
lar disease, masking the true relationship be-
tween blood pressure and mortality. Although
statistical adjustment may minimize these differ-
ences, it will be important in future studies,
particularly in prospective clinical trials, to stratify
patients based on cardiac function and age to
determine the true impact of treatment of hyper-
tension on morbidity and mortality in these pa-
tients.
The role of interdialytic weight gain and vol-
ume overload in pathophysiological characteris-
tics of hypertension in dialysis patients is contro-
versial. Many studies using 24-hour ambulatory
blood pressure monitors have shown no correla-
tion between interdialytic weight gain and blood
pressure.28,34,35 Salem and Davis,36 in an epide-
miological study of change in blood pressure
over a year’s time, showed no relationship be-
tween interdialytic weight gain and blood pres-
sure. However, Dionisio et al37 showed a strong
correlation between total-body water, measured
by bioelectrical impedance, and mean 24-hour
blood pressure. We previously showed that inter-
dialytic weight gain was greater in patients with
uncontrolled hypertension compared with normo-
tensive patients. In addition, hypertensive pa-
tients were likely to have a greater increment in
blood pressure with a similar interdialytic weight
gain than normotensive patients.11
The present study strongly supports the conten-
tion that interdialytic weight gain is an important
factor affecting blood pressure in hemodialysis
patients. Patients with stage 3 hypertension had
263
the greatest interdialytic weight gain, although
this did not reach statistical significance. In the
multivariate analysis, interdialytic weight gain
was an independent predictor of mean arterial
pressure after adjustment for other covariates.
Furthermore, in a striking similarity to our previ-
ous study, there was a significant, albeit low,
positive correlation between blood pressure and
interdialytic weight gain in patients with severe
hypertension, but not in normotensive patients.
This suggests that with increasing interdialytic
weight gain, hypertensive patients have a greater
increment in blood pressure than normotensive
patients. The reasons underlying this differential
response are unclear but may relate to ventricular
compliance, arterial distensibility,38 impaired va-
sodilatation, or other unidentified factors. The
implication of this finding is that education and
counseling to minimize interdialytic weight gain
should be a component of the therapeutic regi-
men to improve the management of hypertension
in these patients. Although rational from a patho-
physiological and now epidemiological point of
view, prospective studies are still lacking to
confirm that reduction of interdialytic weight
gain is an effective therapeutic tool in hyperten-
sive hemodialysis patients.
Congestive heart failure is associated with
high mortality in patients undergoing chronic
hemodialysis.39 The prevalence of congestive
heart failure in this population was 40%, similar
to the 31% prevalence shown in the Canadian
study by Harnett et al.40 Whereas there are cer-
tainly limitations in a clinical diagnosis of conges-
tive heart failure without echocardiographic data,
it serves as a marker for increased risk for mor-
bidity and mortality. In our study, the presence of
congestive heart failure had a significant effect
on the relationships between blood pressure and
other variables of interest. For example, patients
with congestive heart failure did not show the
significant increase in blood pressure with in-
creased interdialytic weight gain seen in patients
without congestive heart failure. We can specu-
late that less steep Frank-Starling curves in pa-
tients with congestive heart failure result in
smaller elevations of stroke volume and blood
pressure compared with patients with preserved
ventricular function when faced with a similar
interdialytic weight gain.
Our study has several strengths. The large
264
sample size drawn from many dialysis units
across the United States makes this the largest
epidemiological and descriptive study of the con-
trol of hypertension and the factors that influence
blood pressure in US hemodialysis patients. The
characteristics of this cohort are similar to the
hemodialysis population nationwide, facilitating
the ability to generalize about these findings. In
addition, the USRDS has taken several measures
to enhance the quality of data collection and
validity of data.14
There are, however, important limitations to
this study. Perhaps the biggest limitation is the
lack of information about antihypertensive drug
use. The segment of the population prescribed
antihypertensive medication that has normal
blood pressure cannot be identified in this data-
base. Therefore, the true prevalence of hyperten-
sion is clearly underestimated in this study. In
addition, the effect of antihypertensive drug
therapy on the various parameters measured in
this study cannot be estimated. We11 and other
investigators17 have shown that use of antihyper-
tensive drug therapy ranges between 58% and
82% of patients undergoing hemodialysis. There-
fore, it is quite likely that the majority of patients
in this large representative national sample were
prescribed antihypertensive drug therapy. How-
ever, we are unable to confirm this because of
lack of data regarding use of antihypertensive
drugs in this data set. In addition, blood pressure
values were abstracted from patients’ charts, and
therefore the reliability of the readings and stan-
dardization of measurement cannot be shown.
Conversely, these represent real-world data that
clinicians use, making it meaningful to draw
conclusions about daily practice. Last, data col-
lection was performed during the month of De-
cember, traditionally the holiday season, which
may influence some measures used in this study.
Our study adds to the epidemiological data
promoting our understanding of the complexities
of hypertension in hemodialysis patients. How-
ever, it is clear that a prospective clinical trial
will be needed, with careful patient selection and
targeted interventions to improve blood pressure
management, to determine their effect on morbid-
ity and mortality.
In summary, this study confirms that hyperten-
sion is common and poorly controlled in most
patients undergoing chronic hemodialysis. Exces-
RAHMAN ET AL
sive interdialytic weight gain and noncompli-
ance with dialysis regimen are associated with
higher blood pressure, and advancing age is
associated with lower blood pressure levels in
this population.
REFERENCES
1. Tozawa M, Iseki K, Fukiyama K: Hypertension in
dialysis patients: A cross-sectional analysis. Jpn J Nephrol
3:129-135, 1997
2. Ritz E, Strumpf C, Katz F, Wing AJ, Quellhorst E:
Hypertension and cardiovascular risk factors in hemodia-
lyzed diabetic patients. Hypertension 7:II118-II124, 1985
3. Quarello F, Piccoli GB, Magistroni P, Cavalli PL,
Salomone M, Boero R, Funaro L, Marciello A, Ferro M,
Piccoli G: Arterial hypertension and mortality in dialysis
patients. RPDT Working Group. Contrib Nephrol 119:141-
146, 1996
4. Salem MM: Hypertension in the dialysis population:
No easy answers. Int J Artif Organs 19:693-694, 1996
5. Raine A, Margreiter R, Brunner F: Report on manage-
ment of renal failure in Europe, XXII, 1991. Nephrol Dial
Transplant 2:7-35, 1992
6. Ang KS, Simon P, Cam G: Ambulatory blood pressure
during the interdialytic period in uremic patients treated by
hemodialysis. Kidney Int 35:238A, 1989
7. Levey AS, Beto JA, Coronado BE, Eknoyan G, Foley
RN, Kasiske BL, Klag MJ, Mailloux LU, Manske CL,
Meyer KB, Parfrey PS, Pfeffer MA, Wenger NK, Wilson
PWF, Wright JT Jr: Controlling the epidemic of cardiovascu-
lar disease in chronic renal disease: What do we know?
What do we need to learn? Where do we go from here? Am J
Kidney Dis 32:853-906, 1998
8. Salem MM: Hypertension in hemodialysis patients?
High time for answers. Am J Kidney Dis 33:592-594, 1999
9. Scribner BH: Reducing mortality in dialysis patients:
Why does hypertension continue to be overlooked? Semin
Dial 10:250-251, 1997
10. Ogundipe A, Dixit A, Donley V, Hanslik T, Rahman
M, Weigel K, Smith MC: Prevalence and causes of inad-
equate blood pressure control in 490 hemodialysis patients. J
Am Soc Nephrol 8:320A, 1997 (abstr)
11. Rahman M, Dixit A, Donley V, Hanslik T, Lacson E,
Ogundipe A, Smith MC: Factors associated with inadequate
blood pressure control in hypertensive hemodialysis pa-
tients. Am J Kidney Dis 33:498-506, 1999
12. Foley RN, Parfrey PS, Harnett JD, Kent GM, Murray
DC, Barre PE: Impact of hypertension on cardiomyopathy,
morbidity and mortality in end-stage renal disease. Kidney
Int 49:1379-1385, 1996
13. Iseki K, Miyasato F, Tokuyama K, Nishime K, Ue-
hara H, Shiohira Y, Sunagawa H, Yoshihara K, Yoshi S,
Toma S, Kowatari T, Wake T, Oura T, Fukiyama K: Low
diastolic blood pressure, hypoalbuminemia, and risk of death
in a cohort of chronic hemodialysis patients. Kidney Int
51:1212-1217, 1997
14. US Renal Data System: Researchers Guide to the
USRDS Database. Bethesda, MD, The National Institutes of
Health, National Institute of Diabetes and Digestive and
Kidney Diseases, 1997
HYPERTENSION IN HEMODIALYSIS PATIENTS
15. Excerpts from the US Renal Data System 1997 An-
nual Report: V. Patient mortality and survival. Am J Kidney
Dis 30:S86-S106, 1997 (suppl 1)
16. The Sixth Report of the Joint National Committee on
Prevention, Detection, Evaluation and Treatment of High
Blood Pressure. National Institutes of Health, Bethesda,
MD, publication 98-4080, 1997
17. Salem MM: Hypertension in the hemodialysis popu-
lation: A survey of 649 patients. Am J Kidney Dis 26:461-
468, 1995
18. Cheigh JS, Milite C, Sullivan JF, Rubin AL, Stenzel
KH: Hypertension is not adequately controlled in hemodialy-
sis patients. Am J Kidney Dis 19:453-459, 1992
19. Rahman M, Douglas JG, Wright JT: Pathophysiology
and treatment implications of hypertension in the African-
American population. Endocrinol Metab Clin North Am
26:125-144, 1997
20. Excerpts from the US Renal Data System 1997 An-
nual Data Report. Causes of death. Am J Kidney Dis
30:S107-S117, 1997 (suppl 1)
21. Leggat JEJ, Orzol SM, Hulbert-Shearon TE, Golper
TA, Jones CA, Held PJ, Port FK: Noncompliance in hemodi-
alysis: Predictors and survival analysis. Am J Kidney Dis
32:139-145, 1998
22. Tomita J, Kimura G, Inoue T, Inenaga T, Sanai T,
Kawano Y, Nakamura S, Baba S, Matsuoka H, Omae T: Role
of systolic blood pressure in determining prognosis of hemo-
dialyzed patients. Am J Kidney Dis 25:405-412, 1995
23. Kimura G, Tomita J, Nakamura S, Uzu T, Inenaga T:
Interaction between hypertension and other cardiovascular
risk factors in survival of hemodialyzed patients. Am J
Hypertens 9:1006-1012, 1996
24. DeGoulet P, Legrain M, Reach I, Aime F, Devries C,
Rojas P, Jacobs C: Mortality risk factors in patients treated
with chronic hemodialysis. Nephron 31:103-110, 1982
25. Fernandez JM, Carbonnel ME, Mazzuchi N: Simulta-
neous analysis of morbidity and mortality factors in chronic
hemodialysis patients. Kidney Int 41:1029-1034, 1992
26. Neff MS, Esier AR, Slifkin RF, Feduska NJ, Salvati-
erra O: Patients surviving 10 years of hemodialysis. Am J
Med 74:996-1004, 1983
27. Vincenti F, Amend WJ, Abele J, Feduska NJ, Salvati-
erra O: The role of hypertension in hemodialysis-associated
atherosclerosis. Am J Med 68:363-369, 1980
28. Conion PJ, Walshe JJ, Heinle SK, Minda S, Krucoff
M, Schwab SJ: Predialysis systolic blood pressure correlates
strongly with mean 24-hour systolic blood pressure and left
ventricular mass in stable hemodialysis patients. J Am Soc
Nephrol 7:2658-2663, 1996
265
29. Harnett JD, Kent GM, Barre PE, Taylor R, Parfrey
PS: Risk factors for the development of left ventricular
hypertrophy in a prospectively followed cohort of dialysis
patients. J Am Soc Nephrol 4:1486-1490, 1994
30. Cannella G, Paoletti E, Delfino R, Peloso G, Molinari
S, Traverso GB: Regression of left ventricular hypertrophy
in hypertensive dialyzed uremic patients on long-term anti-
hypertensive therapy. Kidney Int 44:881-886, 1993
31. Charra B, Calemard M, Laurent G: Importance of
treatment time and blood pressure control in achieving
long-term survival on dialysis. Am J Nephrol 16:35-44,
1996
32. Port FK, Hulbert-Shearon TE, Wolfe RA, Bloember-
gen WE, Agodoa LYC, Young EW: Predialysis blood pres-
sure and mortality risk in a national sample of maintenance
hemodialysis patients. Am J Kidney Dis 33:507-517, 1999
33. Zager PG, Nikolic J, Brown RH, Campbell MA, Hunt
WC, Peterson D, Van Stone J, Levey A, Meyer KB, Klag
MJ, Johnson HK, Clark E, Sadler JH, Teredasai P: ‘‘U’’
curve association of blood pressure and mortality in hemodi-
alysis patients. Kidney Int 54:561-569, 1998
34. Rodby RA, Vonesh EF, Korbet SM: Blood pressures
in hemodialysis and peritoneal dialysis using ambulatory
blood pressure monitoring. Am J Kidney Dis 23:401-411,
1994
35. Coomer RW, Schulman G, Breyer JA, Shyr Y: Ambu-
latory blood pressure monitoring in dialysis patients and
estimation of mean interdialytic blood pressure. Am J Kid-
ney Dis 29:678-684, 1997
36. Salem MM, Davis M: Effects of one year of hemodi-
alysis on weight and blood pressure in 434 patients. Artif
Organs 21:402-404, 1997
37. Dionisio P, Valenti M, Bergia R, Caramello E, Strami-
gnoni E, Berto IM, Pellerey M, Bajardi P: Influence of the
hydration state on blood pressure values in a group of
patients on regular maintenance hemodialysis. Blood Purif
15:25-33, 1997
38. Luik AJ, Gladziwa U, Kooman JP, van Hooff JP, de
Leeuw PW, van Bortel LM, Leunissen KM: Influence of
interdialytic weight gain on blood pressure in hemodialysis
patients. Blood Purif 12:259-266, 1994
39. Parfrey PS, Foley RN, Harnett JD, Kent GM, Murray
DC, Barre PE: Outcome and risk factors for left ventricular
disorders in chronic uremia. Nephrol Dial Transplant 11:
1277-1285, 1996
40. 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
47:884-890, 1995