articles
Association Between Pulse Pressure and 30-Month
All-Cause Mortality in Peritoneal Dialysis Patients
Jiung-Hsiun Liu1, Ching-Chu Chen2,3, Shu-Ming Wang1, Che-Yi Chou1, Yao-Lung Liu1, Huey-Liang Kuo1,
Hsin-Hung Lin1, I-Kuan Wang1, Ya-Fei Yang1 and Chiu-Ching Huang1
Background
Pulse pressure (PP) is an independent predictor of cardiovascular
and/or all-cause mortality in patients with underlying cardiovascular
disease. We examined whether PP can be used to predict overall
mortality in peritoneal dialysis (PD) patients.
Methods
We studied 153 PD patients (mean age, 54.5 ± 14.2 years) with
end-stage renal disease. PP was measured monthly for 3 months.
At the time of the third PP measurement, baseline demographic,
clinical, biochemical, and dialysis data were collected. Patients
were stratified into tertiles according to average PP, and the
relationship between blood pressure parameters and all-cause
mortality over a 30-month follow-up was assessed using Cox
regression.
Blood pressure is a powerful risk factor for cardiovascu-
lar disease (CVD) because it influences the arterial wall and
is responsible for various cardiovascular events. In general,
the relationship between blood pressure (systolic and mean
artery blood pressure) and mortality is U-shaped in end-stage
renal disease (ESRD) patients receiving hemodialysis (HD).1,2
Moreover, it is reported that low diastolic blood pressure (DBP)
is also a significant risk factor for overall mortality in ESRD
patients on HD.3 The pulsatile component of blood pressure
is estimated by the pulse pressure (PP), which is mainly influ-
enced by the stiffness of large arteries, and is considered to be
a risk factor for cardiovascular or all-cause mortality, both for
research purpose and as a clinical predictor. Thus, some inves-
tigators have indicated that PP could be a good predictor of
all-cause and cardiovascular mortality in patients with under-
lying hypertension and various CVDs.4
Notably, increased PP is frequently observed in chronic renal
failure patients5,6 and is associated with decreased arterial-wall
1Division of Nephrology, Department of Internal Medicine, China Medical
University Hospital, Taichung, Taiwan; 2Division of Endocrinology and
Metabolism, Department of Internal Medicine, China Medical University Hospital,
Taichung, Taiwan; 3Department of Endocrinology and Metabolism, School
of Chinese Medicine College of Chinese Medicine, China Medical University,
Taichung, Taiwan. Correspondence: Ya-Fei Yang (w0951@yahoo.com.tw)
Received 30 March 2008; first decision 5 August 2008; accepted 20 August 2008;
advance online publication 18 September 2008. doi:10.1038/ajh.2008.286
© 2008 American Journal of Hypertension, Ltd.
1318 December 2008 | VOLUME 21 NUMBER 12 | 1318-1323 | AMERICAN JOURNAL OF HYPERTENSION
nature publishing group
Downloaded from https://academic.oup.com/ajh/article-abstract/21/12/1318/166884 by guest on 01 November 2018
Results
There were 27 deaths; three deaths occurred after the change to
hemodialysis (HD) (subjects died within 3 months after HD) and were
counted as events during survival analysis. The overall 30-month
survival (Kaplan–Meier curves) times were significantly different
among the tertiles of PP (P < 0.05). Increased PP was significantly
associated with overall mortality regardless of adjustment for systolic
blood pressure (SBP) or diastolic blood pressure (DBP).
Conclusions
PP may be the most consistent blood pressure indicator of mortality
risk. All-cause mortality events in PD patients are more related to
pulsatile stress caused by the stiffness of large arteries during systole
(reflected in a rise of PP) than to steady-state stress stemming from
resistance during diastole (reflected in a rise of SBP and DBP).
Am J Hypertens 2008; 21:1318-1323 © 2008 American Journal of Hypertension, Ltd.
compliance.5,7 On the other hand, increased PP is strongly
associated with overall mortality in ESRD patients undergoing
maintenance HD.8,9 Cardiovascular events are the leading cause
of all-cause mortality in ESRD patients regardless of whether
they receive HD or peritoneal dialysis (PD). The main patho-
physiologic mechanism in those patients involves progression
of the atherosclerotic process and loss of arterial compliance,
which leads to increased PP. Furthermore, whether the dialy-
sis modality is a risk factor for CVD is still debatable. Recent
reports indicate that the outcome of patients suffering from
congestive heart failure or coronary disease is worse if they are
PD recipients than if they are HD recipients.10,11 PP is clearly
associated with mortality in HD patients.8,9 However, few
studies have evaluated the association between PP and mortal-
ity in PD patients. It is also unknown whether the association
between PP and mortality differs between PD and HD patients.
Thus, we conducted a single-center observational study in
a cohort of PD patients to determine whether increased PP is
associated with all-cause mortality in such patients.
Methods
Study design and patients. This prospective study included
153 patients undergoing maintenance PD between 1 April 2005
and 30 September 2005 and was conducted in China Medical
University Hospital in Taiwan. The inclusion ­criteria were reg-
ular PD for at least 3 months before enrollment and clinical
Pulse Pressure and Mortality in Peritoneal Dialysis
stability for at least 3 months before entry without infectious
or other active diseases. The only exclusion criterion was a his-
tory of renal transplantation or maintenance of HD >3 months
prior to the study. The underlying kidney diseases included
chronic glomerulonephritis in 87 patients, hypertensive neph-
rosclerosis in 16, diabetic nephropathy in 30, chronic intersti-
tial nephritis in 7, and other diseases in 13. All PD patients
had double-cuff silastic Tenckhoff catheters inserted through
the rectus muscle using standard surgical techniques and were
maintained on the Baxter Solo disconnect system and twin-bag
system. The dialysis regimen was prescribed by each patient’s
nephrologist. Clinical indications were the main reasons for
change in dialysis regimen. This study was approved by the
institutional review board of the hospital, and informed con-
sent was obtained from each patient.
Data collection and clinical information. Clinical information
was obtained by a thorough review of the written and electronic
medical records. Blood pressure was measured in the arm (bra-
chial artery) after 20 min of rest in the supine position with a
standard mercury sphygmomanometer, and the auscultatory
method was used with cuff-size adjustment based on arm cir-
cumference. Systolic blood pressure (SBP) was recorded at the
first appearance of Korotkoff sounds (phase I), and DBP was
recorded at the complete disappearance of sounds (phase V).
Two readings were recorded 10 min apart, and SBP and DBP
were recorded as the average of two separate measurements.
PP was calculated as the difference between SBP and DBP.
Blood pressure was measured during three sequential clini-
cal visits (per month). Thereafter, patients were stratified into
tertiles according to the average PP (mean of three separate
months) at baseline.
The peritoneal equilibration test was performed using a
2.27% dextrose solution instilled into the peritoneal cavity for
a 2-l volume exchange. We obtained a 24-h collection of both
urine and peritoneal effluent for analysis of clearance. Urea
and creatinine clearances were obtained to calculate weekly
Kt/V urea (the gold standard dialysis dose parameter) and
weekly creatinine clearance. The weekly creatinine clearance
value was also normalized to 1.73 m2 of the body surface area.
Patients were excluded if there was a possibility of peritoni-
tis (along with clinical and laboratory manifestations), when
the peritoneal function test was performed. The ratio of the
dialysate to plasma concentration of creatinine after 4 h (D/P
creatinine) was reported. Furthermore, the Kt/V urea and
weekly creatinine clearance values were the measures in this
study of residual renal clearance or PD clearance as appropri-
ate. Pre-existing cardiovascular and/or cerebrovascular disease
(CVD) was defined as a history of angina, coronary arterial
disease, myocardial infarction, abnormal angiographic results,
transient ischemic attack, and/or cerebrovascular accident.
Absence of residual renal function was defined as completely
anuric status.
Endpoint and outcome analyses. Clinical outcomes in this
study included actual patient survival. Patients transferred
AMERICAN JOURNAL OF HYPERTENSION | VOLUME 21 NUMBER 12 | december 2008 1319
articles
to alternative renal replacement therapies (kidney transplan-
tation or HD) were censored at the time of transfer. Also,
data for patients who were lost to follow-up but not trans-
ferred to HD were censored from the survival analysis. If
a patient died within 3 months of transfer to HD, then his
or her health status was considered as the health status at
the time of PD failure. The observation period ended on 31
December 2007. At the end of the follow-up, the status of all
patients was assessed and data on mortality were obtained
for the entire cohort.
Downloaded from https://academic.oup.com/ajh/article-abstract/21/12/1318/166884 by guest on 01 November 2018
Statistical analysis. Qualitative variables are expressed as
absolute numbers or percentages. For quantitative variables,
results are expressed as mean ± s.d. for normally distributed
data, and as median and interquartile range for nonpara-
metric data. Patients were stratified into tertiles according to
their PP at study baseline. Comparisons of the quantitative
characteristics of patients across the different tertiles were
performed using one-way analysis of variance for normally
distributed data and the Kruskal–Wallis test for nonparamet-
ric data. The χ2-test was applied for comparison of qualitative
data. Factors predictive of all-cause mortality were further
determined using multivariable Cox regression models. Any
variable for which there was a significant difference between
PP tertiles (P value < 0.1) was considered a potential con-
founder and was adjusted for in the Cox regression analysis.
The selected variables were included into a Cox proportional
hazards model, and a backward stepwise selection process
was performed—this time at the classical α set at 0.05. The
potential confounding variables (gender, age, and residual
renal function status) were entered into the Cox model,
even if their relationships across tertiles were insignificant.
Survival curves were generated using the Kaplan–Meier tech-
nique and tested using the log-rank test. A P value < 0.05 was
considered as statistically significant. All calculations were
performed with the SPSS software version 10.0 for Windows
(SPSS Institute, Chicago, IL).
40
Median = 56
Interquartile range = (46–71)
N = 153
30
Number of patients
20
10
0
32.5 42.5 52.5 62.5 72.5 82.5 92.5 102.5 112.5
Pulse pressure (mm Hg)
Figure 1 | Frequency distribution of pulse pressure in peritoneal dialysis
patients.
articles Pulse Pressure and Mortality in Peritoneal Dialysis

Table 1 | Comparison of baseline demographic, clinical, biochemical and dialysis indices of study patients stratified according to
tertiles of PP
Pulse pressure in tertiles
Lower (PP < 49 mm Hg) Middle (PP ≥ 49, < 67 mm Hg) Upper (PP ≥ 67 mm Hg)
(n = 51) (n = 51) (n = 51) P
Age (years) 51.8 ± 15.8 54.8 ± 14.2 56.8 ± 12.1 0.20
Gender (male) (%) 17 (33.3) 17 (33.3) 15 (29.4) 0.89
Diabetes mellitus (%) 3 (5.9) 3 (5.9) 24 (47.1) <0.001
Pre-existing CVD (%) 12 (23.5) 11 (21.6) 14 (27.5) 0.65

Downloaded from https://academic.oup.com/ajh/article-abstract/21/12/1318/166884 by guest on 01 November 2018

History of PD (months) 23 (10–37) 30 (10–57) 30 (18–55) 0.26
Systolic pressure (mm Hg) 126 ± 15 145 ± 13 175 ± 25 <0.001
Diastolic pressure (mm Hg) 83 ± 13 90 ± 12 93 ± 18 <0.05
Pulse rate (/min) 73 ± 11 75 ± 8 76 ± 7 0.55
Hematocrit (%) 28.1 ± 3.9 27.7 ± 3.8 26.5 ± 3.6 0.083
Calcium (mg/dl) 9.7 ± 1.2 9.7 ± 0.9 9.6 ± 1.0 0.11
Phosphate (mg/dl) 5.5 ± 1.3 5.2 ±1.5 5.3 ± 1.4 0.57
Cholesterol (mg/dl) 200 ± 47 206 ± 39 205 ± 45 0.59
Triglyceride (mg/dl) 157 (104–222) 185 (126–319) 210 (139–307) <0.05
Albumin (g/dl) 3.8 (3.6–3.9) 3.7 (3.4–3.9) 3.5 (3.2–3.6) <0.001
Fasting sugar (mg/dl) 100 ± 17 109 ± 27 147 ± 107 <0.001
D/P creatinine 0.62 ± 0.10 0.67 ± 0.10 0.69 ± 0.11 <0.01
PD KT/V urea 1.90 ± 0.40 1.92 ± 0.38 1.89 ± 0.36 0.92
Renal KT/V urea 0.06 (0–0.36) 0.05 (0–0.29) 0 (0–0.14) 0.30
PD WCC (l/week/1.73 m2) 43.5 ± 10.3 50.0 ± 12.0 47.3 ± 9.6 <0.05
Renal WCC (l/week/1.73 m2) 2.2 (0–14.6) 2.5 (0–14.7) 0 (0–6.8) 0.39
Residual renal function, yes (%) 27 (52.9) 27 (52.9) 23 (45.1) 0.43
Ultrafiltration volume (ml/day) 1,137 ± 197 944 ± 247 928 ± 188 0.054
Medication
  Statins (%) 6 (11.8) 7 (13.7) 7 (13.7) 0.94
  ACEI or ARB (%) 9 (17.6) 14 (27.5) 23 (45.1) <0.01
  Antiplatelet (%) 3 (5.9) 7 (13.7) 13 (25.5) 0.27
  β-antagonist (%) 14 (27.5) 18 (35.3) 22 (43.1) <0.05
Quantitative data are expressed as mean ± s.d. or median (interquartile range) and qualitative data are expressed as number (%) as appropriate. KT/ V indicate urea clearance normalize
to its volume of distribution.
ACEI, angiotensin-coverting enzyme inhibitor; ARB, angiotensin receptor blocker; CVD, cardiovascular or/and cerebrovascular disease; D/P, dialysate/plasma; PD, peritoneal dialysis;
PP, pulse pressure; WCC, weekly creatinine clearance.

Results
Patient characteristics
The enrolled patients (49 male and 104 female; mean age,
54.5  ± 14.2 years) were on PD therapy for a median (inter-
quartile range) duration of 27 (16–53) months. Of these,
67 (49.7%) had no residual renal function. Cardiovascular
medications included statins, angiotensin receptor blockers/
angiotensin-converting enzyme inhibitors (ARB/ACEI),
antiplatelet regimens (aspirin, dipyridamole, or clopidogrel),
and β-antagonists in 13.1, 30.1, 15.0, and 35.3% of patients,
respectively. The tertile cutoff points for PP were 49 and
67 mm Hg; the distribution of PP at the time of study entry is
shown in Figure 1. The median PP (interquartile range) was
56 mm Hg (range: 46–71 mm Hg). The percentage of patients
with PP below 60 mm Hg, between 60 and 80 mm Hg, between
80 and 100 mm Hg, and over 100 mm Hg, was 58.8, 26.2, 11.1,
and 3.9, respectively. The percentage of patients with resid-
ual renal function across the tertiles was 52.9 (lower), 52.9
­(middle), and 45.1 (upper), respectively (P = 0.43).
Comparisons of the different tertiles of PP
Comparisons across PP tertiles (PP <49 mm Hg (lower ter-
tile), PP between 49 and 67 mm Hg (middle tertile), and
PP ≥67 mm Hg (upper tertile)) are shown in Table 1. A signifi-
cant increase in 4-h D/P creatinine (P < 0.01), serum triglyceride
(P < 0.05), and fasting blood sugar (P < 0.001), and a significant
decrease in serum albumin level (P < 0.001) were observed with
increasing PP. More patients in the upper PP tertile had diabetes
mellitus (P < 0.001), received ACEI/ARB treatment (P < 0.01),
and received β-antagonist (P < 0.05) treatment.

1320 december 2008 | VOLUME 21 NUMBER 12 | AMERICAN JOURNAL OF HYPERTENSION

Pulse Pressure and Mortality in Peritoneal Dialysis articles

Table 2 | Clinical outcome of patients according to the pulse pressure level at study baseline
Pulse pressure in tertiles
Lower (n = 51) Middle (n = 51) Upper (n = 51) P
Outcome
  Died on dialysis (n (%)) 6 (11.8) 7 (13.7) 14 (27.5) <0.05
  Alive on peritoneal dialysis (n (%)) 36 (70.6) 34 (66.7) 27 (52.9) 0.065
  Transferred to hemodialysis (n (%)) 8 (15.7) 8 (15.7) 9 (17.6) 1.00
  Transplanted (n (%)) 0 (0) 1 (2.0) 0 (0) 0.22
  Loss of follow-up (n (%)) 1 (2.0) 1 (2.0) 1 (2.0) 1.00

Downloaded from https://academic.oup.com/ajh/article-abstract/21/12/1318/166884 by guest on 01 November 2018

Cause of death
 Cardiovascular causes 1 (2.0) 3 (5.9) 6 (11.8) <0.05
  Coronary artery disease (n (%)) 0 (0) 1 (2.0) 4 (7.8) <0.05
  Cerebrovascular disease (n (%)) 0 (0) 1 (2.0) 2 (3.9) 0.16
   Sudden death (n (%)) 1 (2.0) 1 (2.0) 0 (0) 0.39
  Noncardiovascular causes (n (%)) 5 (9.8) 5 (9.8) 7 (13.7) 0.53
   Unresolved peritonitis (n (%)) 3 (5.9) 1 (2.0) 2 (3.9) 0.61
   Other infection or sepsis (n (%)) 2 (3.9) 1 (2.0) 4 (7.8) 0.35
   Other miscellaneous causes (n (%)) 0 (0) 3 (5.9) 1 (2.0) 0.54
Percentage may not add up to 100 due to rounding off of decimal places.

1.0
Patient survival and causes of death in relation to PP
During the study period, 28 patients were switched to HD,
one received renal transplantation, and three transferred to 0.8

other PD centers. Of the 27 deaths, three deaths occurred

after a change to HD and were counted as events during
Cumulative survival

0.6
survival analysis. The causes of death (Table 2) were CVDs
(10 patients), peritonitis (six patients), nonperitonitis infec-
tions (seven patients), malignancy (one patient), hollow 0.4
organ rupture (two patients), and acute pancreatitis (one
P < 0.05
patient). The remaining 97 patients were administratively Upper tertile

censored on 31 December 2007. Kaplan–Meier representa- 0.2 Middle tertile

tion of the actual patient survival is shown in Figure 2. The
overall 30-month survival differed significantly among ter-
tiles (P < 0.05). The overall survival was 64.3% for patients in
the upper PP tertile compared with 86.0% for patients in the
middle PP tertile (P < 0.05), and the survival rate was 86.0%
for patients in the lower PP tertile (P < 0.05). The survival
rates of the lower and middle PP tertiles did not differ sig-
nificantly (P = 0.89).
To further examine the relationships between blood pres-
sure parameters and mortality, each parameter was entered
into the adjusted-mortality model as single and paired covari-
ates (Table 3). After backward stepwise selection, the variables
that remained in the Cox proportional hazards model were
found to be diabetes (fasting blood sugar level), serum albu-
min level, and 4-h D/P creatinine. The adjusted hazard ratios
of age, diabetes, serum albumin level, and 4-h D/P creatinine
for all-cause mortality were 1.07 (95% confidence interval
(CI), 0.995–1.006; P = 0.083, per year increase), 1.93 (95% CI,
1.431–3.143; P < 0.001), 0.94 (95% CI, 0.903–0.982; P < 0.05,
per 0.1 g/dl increase), and 1.08 (95% CI, 1.010–1.273; P < 0.05,
per 0.1 unit increase), respectively.
Lower tertile
0.0
0 5 10 15 20 25 30
Duration of follow-up (months)
Figure 2 | Kaplan–Meier plot of all-cause mortality in relation to tertiles of
pulse pressure.
Patients in the upper PP tertile had a 1.71-fold higher risk
for all-cause mortality (95% CI, 1.154 to 3.643; P < 0.05) com-
pared with those in the lower PP tertile. Patients in the middle
PP tertile had no significant increase in the risk for all-cause
mortality (P = 0.23) compared with those in the lower PP ter-
tile. The adjusted hazard ratios associated with a 10-mm Hg
increment in blood pressure in single blood pressure models
were as follows: PP, 1.17 (P < 0.01); SBP, 1.12 (P < 0.05); and
DBP, 0.92 (P < 0.05), respectively (Table  3). Using the dual
model combining SBP and DBP (Model 1), DBP (P < 0.05)
contributed modestly and less than SBP to all-cause mortality;
however, using dual models combining SBP and PP (Model 2)
or PP and DBP (Model 3), SBP and DBP had no effect on all-
cause mortality if PP was included in the model.

AMERICAN JOURNAL OF HYPERTENSION | VOLUME 21 NUMBER 12 | december 2008 1321

articles
Table 3 | Mortality risk associated with single and dual blood
pressure components in multivariable Cox proportional
hazards models for 30-month mortality
AHR (95% CI)a
Single blood pressure
  PP 1.17 (1.04–1.30)
  SBP 1.12 (1.02–1.22)
  DBP 0.92 (0.85–0.98)
Dual blood pressure
 Model 1
   SBP 1.19 (1.13–1.25)
   DBP 0.89 (0.79–0.98)
 Model 2
   PP 1.16 (1.03–1.29)
   SBP 1.02 (0.91–1.12)
 Model 3
   PP 1.20 (1.11–1.29)
   DBP 0.95 (0.83–1.07)
AHR, adjusted hazard ratio; CI, confidence interval; DBP, diastolic blood pressure; PP, pulse
pressure; SBP, systolic blood pressure.
Discussion
The major finding of this study in chronic PD patients was that
baseline PP is independently associated with the incidence
of overall mortality. The greater the PP value, the greater was
the risk of mortality. The clinical outcomes of patients in the
middle PP tertile were not significantly different from that
in the lower PP tertile. This is in contrast to studies in car-
diac patients12,13 and in the HD population8,9,14 showing that
even high-normal PP values are associated with adverse out-
come. The reason for this difference is not clear, but it may be
partly related to the power of the study. This study had only
51 patients in each tertile and may not be adequately powered
to detect a difference in outcome between high-normal and
low-normal PP values.
Widespread cardiac dysfunction is a major problem in ESRD
patients, and CVD remains the most common cause of mor-
tality in those patients. Thus, it is important to improve blood
pressure control and reduce cardiovascular risk in ESRD
patients. Regarding blood pressure control, PP is not generally
recognized as a risk factor in ESRD patients. Physiologically,
blood pressure has two components: a steady component (mean
arterial pressure) and a pulsatile component (PP). PP has two
major components: One is due to the relationship between ven-
tricular ejection and the viscoelastic properties of large arteries
(arterial stiffness), and the other to the indirect effect of arte-
rial wave reflection from the peripheral arteries back to central
conduit arteries.15 Elevated PP is usually attributed to increased
central aortic stiffness. Elevated pressure during systole favors
left ventricular hypertrophy, predisposes to left ventricular fail-
ure, and increases myocardial oxygen demands, whereas lower
pressure throughout diastole has the potential to limit coronary
perfusion and predispose to ischemia.
1322 december 2008 | VOLUME 21 NUMBER 12 | AMERICAN JOURNAL OF HYPERTENSION
Pulse Pressure and Mortality in Peritoneal Dialysis
Increased aortic stiffness results in higher SBP and lower DBP
because of shifts in the pressure–volume relationship in the
aorta, loss of elastic recoil in diastole, and more rapid return of
reflected pressure waves, resulting in pressure increase during
P value
end systole. In individuals with ESRD (a population that has sub-
stantially increased all-cause mortality), aortic pulse wave veloc-
<0.01 ity (a measure of aortic stiffness) may be considered to be an
<0.01 independent mortality predictor.16 Thus, death in those patients
<0.05 may be related more to the pulsatile (i.e., PP) than to the steady-
state (i.e., SBP, DBP, mean arterial pressure) components of blood
pressure, which are associated with small-vessel resistance.
Downloaded from https://academic.oup.com/ajh/article-abstract/21/12/1318/166884 by guest on 01 November 2018
PP is considered an independent risk factor for over-
<0.01
all ­mortality and/or cardiovascular events in several stud-
<0.05
ies in  hypertensive patients,12,17 diabetes patients,18 the
elderly,19–21 men,22 postmenopausal women with coronary
<0.05 heart disease,23 and ESRD patients treated with HD.8,9,14
0.59 However, the association between PP and mortality in PD
patients is still unanswered. In the present study, we demon-
<0.01 strated that PP is an independent predictor for all-cause mor-
tality not only in individuals with ESRD undergoing HD,8,9,14
0.14
but also in those undergoing PD treatment. We showed that
PP is an independent risk for all-cause mortality. Therefore,
we suggest that antihypertensive treatment in PD patients
should aim at reducing not only blood pressure,24 but also PP.
This study has several limitations. First, the number of
patients with PD treatment is modest, and cardiovascular
events were not analyzed. A larger cohort of PD patients is
needed to confirm our finding. Second, examination of the
association between PP and death was only observational,
and confounding variables that were not measured could
have biased the results. Finally, brachial PP is an overestimate
of central PP because of pressure amplification in the upper
extremities; central PP was not measured in this study.
In conclusion, PP provides important predictive value for
mortality events in PD patients. PP may be the most consist-
ent blood pressure indicator of mortality risk. Our results sug-
gest that mortality is related more to pulsatile stress due to the
stiffness of large artery during systole (reflected in a rise of PP)
than to steady-state stress caused by resistance during diastole
(reflected in increased SBP and DBP). Also, further evaluation
is required to ascertain whether lowering PP improves out-
come in these patients.
Disclosure: The authors declared no conflict of interest.
1. 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, Teredesai P.  “U” curve
association of blood pressure and mortality in hemodialysis patients. Kidney Int
1998; 54:561–569.
2. Port FK, Hulbert-Shearon TE, Wolfe RA, Bloembergen WE, Golper TA, Agodoa LY,
Young EW. Pre-dialysis blood pressure and mortality risk in a national sample of
maintenance hemodialysis patients. Am J Kidney Dis 1999; 33:507–517.
3. Iseki K, Miyasato F, Tokuyama K, Nishime K, Uehara 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 1997; 51:1212–1217.
4. Safar ME. Pulse pressure, arterial stiffness, and cardiovascular risk. Curr Opin Cardiol
2000; 15:258–263.
5. London GM, Marchais SJ, Safar ME, Genest AF, Guerin AP, Metivier F, Chedid K,
London AM. Aortic and large artery compliance in end-stage renal failure. Kidney
Int 1990; 37:137–142.
Pulse Pressure and Mortality in Peritoneal Dialysis
6. Tozawa M, Iseki K, Iseki C, Oshiro S, Yamazato M, Higashiuesato Y, Tomiyama N,
Tana T, Ikemiya Y, Takishita S. Evidence for elevated pulse pressure in patients on
chronic hemodialysis: a case-control study. Kidney Int 2002; 62:2195–2201.
7. Covic A, Gusbeth-Tatomir P, Goldsmith DJ. Arterial stiffness in renal patients: an
update. Am J Kidney Dis 2005; 45:965–977.
8. Klassen PS, Lowrie EG, Reddan DN, DeLong ER, Coladonato JA, Szczech LA,
Lazarus JM, Owen WF Jr. Association between pulse pressure and mortality in
patients undergoing maintenance hemodialysis. JAMA 2002; 287:1548–1555.
9. Tozawa M, Iseki K, Iseki C, Takishita S. Pulse pressure and risk of total mortality
and cardiovascular events in patients on chronic hemodialysis. Kidney Int 2002;
61:717–726.
10. Ganesh SK, Hulbert-Shearon T, Port FK, Eagle K, Stack AG. Mortality differences
by dialysis modality among incident ESRD patients with and without coronary
artery disease. J Am Soc Nephrol 2003; 14:415–424.
11. Stack AG, Molony DA, Rahman NS, Dosekun A, Murthy B. Impact of dialysis
modality on survival of new ESRD patients with congestive heart failure in the
United States. Kidney Int 2003; 64:1071–1079.
12. Blacher J, Staessen JA, Girerd X, Gasowski J, Bsc LT, Liu L, Wang JG, Fagard RH.
Safar ME. Pulse pressure not mean pressure determines cardiovascular risk in
older hypertensive patients. Arch Int Med 2000; 160:1085–1089.
13. Mitchell GF, Moyé LA, Braunwald E, Rouleau JL, Bernstein V, Geltman EM,
Flaker GC, Pfeffer MA. Sphygmomanometrically determined pulse pressure is a
powerful independent predictor of recurrent events after myocardial infarction in
patients with impaired left ventricular function. SAVE investigators. Survival and
Ventricular Enlargement. Circulation 1997; 96:4254–4260.
14. Safar ME, Blacher J, Pannier B, Guerin AP, Marchais SJ, Guyonvarc’h PM, London
GM. Central pulse pressure and mortality in end-stage renal disease. Hypertension
2002; 39:735–738.
AMERICAN JOURNAL OF HYPERTENSION | VOLUME 21 NUMBER 12 | december 2008 1323
articles
15. Nichol WW, Nicolini FA, Pepine CJ. Determinants of isolated systolic hypertension
in the elderly. J Hypertens 1992; 10(Suppl 6): S73–S77.
16. Blacher J, Guerin A, Pannier B, Marchais S, Safar M, London G. Impact of aortic
stiffness on survival in end-stage renal failure. Circulation 1999; 99:2434–2439.
17. Benetos A, Rudnichi A, Safar M, Guize L. Pulse pressure and cardiovascular
mortality in normotensive and hypertensive subjects. Hypertension 1998; 32:
560–564.
18. Nakano S, Konishi K, Furuya K, Uehara K, Nishizawa M, Nakagawa A, Kigoshi T,
Uchida K. A prognostic role of mean 24-h pulse pressure level for cardiovascular
events in type 2 diabetic subjects under 60 years of age. Diabetes Care 2005;
28:95–100.
19. Glynn RJ, Chae CU, Guralnik JM, Taylor JO, Hennekens CH. Pulse pressure and
mortality in older people. Arch Intern Med 2000; 160:2765–2772.
20. Chae CU, Pfeffer MA, Glynn RJ, Mitchell GF, Taylor JO, Hennekens CH.
Downloaded from https://academic.oup.com/ajh/article-abstract/21/12/1318/166884 by guest on 01 November 2018
Increased pulse pressure and risk of heart failure in the elderly. JAMA 1999;
281:634–639.
21. Vaccarino V, Holford TR, Krumholz HM. Pulse pressure and risk for myocardial
infarction and heart failure in the elderly. J Am Coll Cardiol 2000; 36:
130–138.
22. Benetos A, Safar M, Rudnichi A, Smulyan H, Richard JL, Ducimetieère P, Guize L.
Pulse pressure: a predictor of long-term cardiovascular mortality in a French male
population. Hypertension 1997; 30:1410–1415.
23. Nair GV, Chaput LA, Vittinghoff E, Herrington DM. Pulse pressure and
cardiovascular events in postmenopausal women with coronary heart disease.
Chest 2005; 127:1498–1506.
24. Davenport A, Cox C, Thuraisingham R. Achieving blood pressure targets during
dialysis improves control but increases intradialytic hypotension. Kidney Int 2008;
73:759–764.