Journal of Human Hypertension (2011) 25, 444450

& 2011 Macmillan Publishers Limited All rights reserved 0950-9240/11

www.nature.com/jhh

ORIGINAL ARTICLE

Ambulatory pulse pressure as a novel

predictor for long-term prognosis in
essential hypertensive patients
Y-T Kao1,7, C-C Huang2,3,4,6,7, H-B Leu3,4,5, T-C Wu3,4,5, P-H Huang3,4,5, S-J Lin2,3,4,5
and J-W Chen2,3,4,6
1

Division of Cardiology, Department of Internal Medicine, Taipei Medical University Hospital, Taipei,
Taiwan, ROC; 2Department of Medical Research and Education, Taipei Veterans General Hospital, Taipei,
Taiwan, ROC; 3Division of Cardiology, Department of Medicine, Taipei Veterans General Hospital, Taipei,
Taiwan, ROC; 4Cardiovascular Research Center, National Yang-Ming University, Taipei, Taiwan, ROC;
5
Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan, ROC and 6Institute of
Pharmacology, National Yang-Ming University, Taipei, Taiwan, ROC

The prognostic value of ambulatory blood pressure

(BP) monitoring for long-term prognosis varies in recent
studies. The study aimed to investigate the role of
ambulatory BP parameters in mortality and cardiovascular (CV) events in hypertensive patients. A series
of 412 participants (59.34.0 years) who received
ambulatory BP monitoring for their fluctuated BP, either
untreated or treated since 1995, were enroled. The
mortality and CV events were obtained by follow-up
and linked to the National Death Registry in Taiwan.
There were 233 untreated and 179 treated patients.
The latter were older with more comorbidity when
compared with the former. After follow-up for 8.51.7
years, both ambulatory systolic BP and pulse pressure

(PP) could predict all-cause mortality, non-CV mortality,

CV disease and stroke after adjusting for baseline
covariates. However, only ambulatory PP could predict
CV mortality and coronary heart disease. Ambulatory
PP is better than ambulatory systolic BP, particularly in
prediction of all-cause mortality. There was no predictive value of office BP in any outcome. In conclusion,
ambulatory PP is a good predictor for long-term outcomes in hypertensive patients. The parameters of
ambulatory rather than office BP could be applied for
risk stratification either before or under antihypertensive
treatment.
Journal of Human Hypertension (2011) 25, 444450;
doi:10.1038/jhh.2010.80; published online 5 August 2010

Keywords: ambulatory blood pressure monitoring; cardiovascular events; mortality; pulse pressure

Introduction
The relation between blood pressure (BP) and the
risk of cardiovascular disease (CVD) is direct, graded
and continuous over a wide range.1 Although BP
measurement at the clinic is currently the standard
of reference, there is considerable debate as to
the most appropriate method of assessing BP in
different clinical settings. Clinic BP measurements
may be inappropriate for a number of reasons
including inaccuracies in measurement technique
and artificial increases in BP produced by whitecoat effects.2 Many studies have confirmed superiority of 24-h ambulatory BP monitoring (ABPM)
over office or clinic BP measurement in predicting
Correspondence: Professor J-W Chen, Department of Medical
Research and Education, Taipei Veterans General Hospital,
No. 201, Section 2, Shih-Pai Road, Taipei 112, Taiwan, ROC.
E-mail: jwchen@vghtpe.gov.tw
7
These authors contributed equally to this work.
Received 1 February 2010; revised 24 April 2010; accepted 21
June 2010; published online 5 August 2010

hypertension-induced organ damage or clinical

outcome.36
To what extent ABPM provides an independent
predictive value beyond office BP is of clinical
importance, as ABPM is potentially more burdensome and associated with higher costs than office or
clinic BP measurements. Previous study mainly
focused on ambulatory systolic BP (SBP)3,4,6,7 or
diastolic BP (DBP).3,4 Variation in the daynight BP
pattern (the dipping status) has also been claimed as
an important predictor of both target-organ damage
and CV events.4,6,8 However, the roles of other
parameters derived from ABPM, including 24-h
pulse pressure (PP), daytime PP and nighttime PP,
are not fully defined. Further, previous studies were
mainly conducted in Europe3,4,6 and Japan.5 Both
environmental and genetic factors9 of hypertension
may be different among different ethic populations.
Given the fact that hypertension is one of the major
leading causes of CV events and death in Chinese,
this study was conducted to systemically assess the
prognostic value of these ABPM parameters on CV

24-h pulse pressure and prognosis

Y-T Kao et al
445

morbidity and mortality in an ethnic Chinese

population in Taiwan.

Methods
Study population

A total of 412 consecutive patients were enroled into

the study. They had undergone 24-h ABPM for
fluctuation of BP (SBP X140 mm Hg or DBP
X90 mm Hg or both, either at home or at two or
more consecutive clinical visits) either with or
without antihypertensive treatment in the Taipei
Veterans General Hospital between October 1995
and September 1999. Patients were defined as
untreated and treated according to whether antihypertensive medication had been taken during
ABPM examination. Hypertensive subjects were
defined as those with SBP X140 mm Hg, DBP
X90 mm Hg or prescribed at least one antihypertensive agent. Diabetes mellitus (DM) subjects were
defined according to the criteria of the American
Diabetes Association.10 Hyperlipidemia subjects
were defined according to the criteria of National
Cholesterol Education Program Adult Treatment
Panel III.11 The study protocol was approved by
the ethics committee of the Taipei Veterans General
Hospital. This study was conducted in accordance
with the principles of the Declaration of Helsinki
and Title 45, US Code of Federal Regulations,
Part 46, Protection of Human Subjects, revised on
13 November 2001, effective since 13 December
2001.
BP measurement

Office BP was measured by a well-trained nurse or

assistant with a validated automatic digital BP
monitor (Omron HEM-705CP; OMRON Healthcare,
Bannockburn, IL, USA)12 in the morning hours after
sitting for 15 min in a quiet room. Three consecutive
BP measurements were carried out each time at
5-min intervals. Office BPs were recorded as the
average value of the last two recordings. ABPM was
performed using SpaceLabs 90217 devices (SpaceLabs Inc, Wokingham, Berkshire, UK).13 All patients
were fitted with the device between 0800 and 1000
hours. The device was programmed to record BP
every 15 min from 0600 to 2200 hours and every
30 min from 2200 to 0600 hours. In this analysis,
daytime BP was the average BP from 0600 to 2200
hours and nighttime BP was the average BP from
2200 to 0600 hours. The PP was calculated by SBP
minus DBP.
Follow-up

All of the patients were followed up in our outpatient clinics on regular basis, such as every 13
months. Information about clinical events was
obtained from the hospital charts or telephone
interview. Outcome variables were (1) all-cause

mortality; (2) CV mortality, including all fatal CV

events; (3) non-CV mortality; (4) CVD, including
sudden death, fatal or nonfatal myocardial infarction (MI) and fatal or nonfatal stroke; (5) coronary
heart disease (CHD), including sudden death and
fatal or nonfatal MI; and (6) fatal or nonfatal stoke.
The date and causes of mortality were also collected
by linking our database with the National Death
Registry through a unique, life-long personal identification number given to every Taiwan citizen.
Subjects not appearing on the National Death
Registry on 31 December 2007 were considered
surviving. The National Death Registry database
registers valid information based on the certified
death certificates. The death certificates were coded
according to the International Classification of
Disease, Ninth Revision (ICD-9). The ICD-9 codes
used for CV mortality were 390459. The accuracy
of cause-of-death coding in Taiwans National Death
Registry database has been validated.14

Data analysis

Statistical analysis was performed using SPSS software (version 15.0, SPSS Inc, Chicago, IL, USA). All
data were expressed as frequency (percentage) or
meanstandard deviation or median with interquartile ranges. Parametric continuous data between
untreated and treated groups were compared by
unpaired Students test, and nonparametric data
used the MannWhitney test. Categorical data
between untreated and treated groups were compared with w2-test, with Yates correction or Fishers
exact test as appropriate. The association between
each office and ambulatory BP parameters and
outcome events was tested using Cox proportional
hazards regression models. Each office and ambulatory BP parameter was adjusted for baseline variables: age, gender, smoking, concomitant medications
(antihypertensive and lipid-lowering agents), DM,
previous MI, previous stroke, congestive heart failure
and hyperlipidemia. In further models, the office BP
and ambulatory BP, and ambulatory SBP and PP were
included into the same model for competition.
Statistical significance was inferred at a two-sided
P-value of o0.05.

Results
Patient characteristics at baseline in 412 participants

We enroled a total of 412 consecutive patients

(mean age 59.314.0 years, 61.2% male) who had
undergone 24-h ABPM because of unstable or
unsatisfactorily controlled BP either with or without
antihypertensive treatment between October 1995
and September 1999 in the Taipei Veterans General
Hospital (Table 1). There were 233 patients (56.6%)
who were untreated (the untreated group), whereas
another 179 (43.4%) were already under treatment
(the treated group). Compared with the untreated
Journal of Human Hypertension

24-h pulse pressure and prognosis

Y-T Kao et al
446

Table 1 Basic characteristics

All (N 412)

Untreated (N 233)

Treated (N 179)

P-value

Baseline characteristics
Age (years)
Male (n (%))
Bodyweight (kg)
Height (cm)
BMI (kg m2)
Smoking (n (%))

59.314.0
252 (61.2%)
67.511.9
161.78.4
25.83.6
84 (20.4%)

57.014.3
145 (62.2%)
67.311.8
162.08.2
25.63.6
52 (22.3%)

62.313.0
107 (59.8%)
67.812.0
161.38.6
26.03.6
32 (17.9%)

o0.001
0.612
0.668
0.404
0.205
0.267

Office BP (mm Hg)

SBP
DBP
PP

148.521.6
88.113.9
60.417.9

147.721.4
89.313.8
58.417.2

149.521.9
86.713.8
62.818.6

0.426
0.061
0.017

Ambulatory BP (mm Hg)

24-h SBP
24-h DBP
24-h PP
Daytime SBP
Daytime DBP
Daytime PP
Nighttime SBP
Nighttime DBP
Nighttime PP

134.614.1
82.711.2
51.911.2
136.614.1
84.611.4
52.011.2
128.616.5
77.311.8
51.312.2

134.713.7
84.710.8
50.09.3
137.013.7
86.710.9
50.39.2
128.115.6
78.911.5
49.210.3

134.414.6
80.211.2
54.312.9
136.114.7
81.911.6
54.213.0
129.417.7
75.311.9
54.113.9

0.826
o0.001
o0.001
0.530
o0.001
o0.001
0.431
0.002
o0.001

Concomitant medication (n (%))

Antihypertensives
ACEI or ARB
b-Blockers
CCB
Diuretics
Lipid-lowering drugs

179 (43.4%)
65 (15.8%)
95 (23.1)
117 (28.4%)
32 (7.8%)
53 (13.2%)

35 (15.4%)

179 (100.0%)
65 (36.3%)
95 (53.1%)
117 (65.4%)
32 (17.9%)
18 (10.3%)

0.137

Comorbidity (n (%))
DM
Previous MI
Previous stroke
CHF
Hyperlipidemia

47 (11.5%)
7 (1.7%)
25 (6.1%)
32 (7.8%)
185 (45.0%)

20 (8.6%)
2 (0.9%)
14 (6.0%)
9 (3.9%)
106 (45.7%)

27 (15.2%)
5 (2.8%)
11 (6.1%)
23 (12.8%)
79 (44.1%)

0.039
0.247
0.963
0.001
0.753

Abbreviations: ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMI, body mass index; BP, blood pressure;
CCB, calcium channel blocker; CHF, congestive heart failure; DBP, diastolic blood pressure; DM, diabetes mellitus; MI, myocardial infarction;
PP, pulse pressure; SBP, systolic blood pressure.

patients, patients in the treated group were older

(Po0.001) with higher office PP (P 0.017).
The prevalence of DM (P 0.039) and congestive
heart failure (P 0.001) were also higher in
the treated group. The 24-h DBP (Po0.001), daytime
DBP (Po0.001) and nighttime DBP (P 0.002)
were lower, and the 24-h PP (Po0.001), daytime
PP (Po0.001) and nighttime PP (Po0.001) were
higher in the treated group than in the untreated
group.

Among the ambulatory PP parameters, 24-h PP

could predict all the clinical outcomes including allcause, CV and non-CV mortality, CVD, CHD and
stroke. Both daytime and nighttime PP could predict
all-cause, CV and non-CV morbidity and CVD.
Daytime PP could also predict CHD (Table 2).
On the other hand, 24 h, daytime and nighttime
SBP could predict all-cause mortality, CVD and
stroke. Besides, although 24 h and nighttime SBP
could also predict non-CV mortality, daytime SBP
could further predict CHD (Table 2).

Prognostic values of ambulatory BP parameters in 412

participants

Ambulatory BP parameters versus office BP parameters

After 8.51.7 years of follow-up, there were 45

deaths (12 CV mortality and 33 non-CV mortality)
and 42 CVD (17 CHD and 25 strokes). Table 2
showed the adjusted hazard ratios (HRs) for the
outcome associated with one standard deviation
increase in office and ambulatory BP parameters. All
the office BP parameters failed to predict any
clinical outcome.

Table 3 showed the adjusted HRs for the outcome

when office BP and ambulatory BP were included in
the same model. It was shown that 24-h PP, rather
than office PP, could predict all the CV and non-CV
mortality and events. Besides, 24-h SBP, rather than
office SBP, could also predict the all-cause mortality,
CVD and stroke. Thus, ambulatory PP and SBP are
much more prognostic than office PP and SBP.

Journal of Human Hypertension

24-h pulse pressure and prognosis

Y-T Kao et al
447

Table 2 Adjusted hazard ratios for outcome with office, daytime, nighttime and 24-h blood pressure

No. of events
Office BP (1 s.d.)
Office SBP
Office DBP
Office PP

All-cause mortality

CV mortality

NCV mortality

CVD

CHD

Stroke

45

12

33

42

17

25

1.05 (0.771.43)
0.69 (0.471.02)
1.31 (0.951.80)

Ambulatory BP
24 h
24-h SBP (1 s.d.)
24-h DBP (1 s.d.)
24-h PP (1 s.d.)
Daytime
Daytime SBP (1 s.d.)
Daytime DBP (1 s.d.)
Daytime PP (1 s.d.)
Nighttime
Nighttime SBP (1 s.d.)
Nighttime DBP (1 s.d.)
Nighttime PP (1 s.d.)

1.07 (0.581.97) 1.09 (0.751.58) 1.06 (0.761.46) 1.44 (0.882.35) 0.85 (0.551.31)
0.70 (0.301.62) 0.69 (0.451.07) 0.77 (0.521.13) 0.87 (0.461.63) 0.72 (0.431.18)
1.28 (0.692.36) 1.42 (0.962.10) 1.25 (0.901.74) 1.73 (1.032.90)* 0.99 (0.641.55)

1.49 (1.112.00)w 1.67 (0.913.06) 1.44 (1.012.04)* 1.63 (1.202.22)w 1.54 (0.952.50) 1.74 (1.142.63)w
0.89 (0.601.34) 0.73 (0.291.79) 0.97 (0.611.54) 1.18 (0.801.74) 0.91 (0.491.67) 1.45 (0.872.40)
1.77 (1.332.35)z 2.03 (1.163.56)* 1.67 (1.182.38)w 1.70 (1.242.31)w 1.89 (1.173.04)w 1.57 (1.022.40)*
1.41 (1.041.90)* 1.70 (0.933.11) 1.31 (0.921.87) 1.69 (1.232.33)w 1.74 (1.072.83)* 1.68 (1.082.60)*
0.81 (0.541.22) 0.73 (0.301.76) 0.85 (0.531.35) 1.19 (0.801.76) 0.98 (0.531.81) 1.38 (0.832.28)
1.72 (1.292.28)z 2.07 (1.183.62)* 1.59 (1.122.26)* 1.71 (1.252.35)w 2.03 (1.273.24)w 1.51 (0.972.34)
1.56 (1.192.04)w 1.34 (0.732.46) 1.68 (1.222.30)w 1.36 (1.011.81)* 1.10 (0.661.83) 1.53 (1.062.22)*
1.17 (0.821.67) 0.69 (0.301.58) 1.39 (0.942.06) 1.14 (0.801.61) 0.75 (0.411.38) 1.44 (0.942.22)
1.72 (1.312.28)z 1.75 (1.003.05)* 1.73 (1.232.45)w 1.43 (1.051.94)* 1.43 (0.872.33) 1.39 (0.932.08)

Abbreviations: BP, blood pressure; CHD, coronary heart disease; CV, cardiovascular; CVD, cardiovascular disease; DBP, diastolic blood pressure;
NCV, noncardiovascular; PP, pulse pressure; SBP, systolic blood pressure.
Data are hazard ratios (95% confidence intervals) for each 1 s.d. higher blood pressure. Hazard ratios were also adjusted for baseline
characteristics including age, gender, smoking, concomitant medications (antihypertensive and lipid-lowering drugs), diabetes mellitus, previous
myocardial infarction, previous stroke, congestive heart failure and hyperlipidemia.
Significance of hazard ratios: *Po0.05; wPo0.01; zPo0.001.

Table 3 Adjusted hazard ratios for outcome with ambulatory systolic BP and PP
All-cause mortality

CV mortality

NCV mortality

CVD

CHD

Stroke

Office SBP versus 24-h SBP (1 s.d.)

Office SBP
0.93 (0.671.30)
24-h SBP
1.55 (1.132.12)w

0.98 (0.511.90)
1.86 (0.993.48)

0.97 (0.651.45)
1.44 (0.982.11)

0.93 (0.661.30)
1.69 (1.222.34)w

1.33 (0.792.23)
1.43 (0.842.42)

0.71 (0.451.12)
1.95 (1.273.00)w

Office PP versus 24-h PP (1 s.d.)

Office PP
1.03 (0.721.47)
24-h PP
1.79 (1.302.47)z

1.01 (0.521.99)
2.32 (1.234.38)w

1.18 (0.761.83)
1.55 (1.032.33)*

1.03 (0.721.48)
1.74 (1.242.45)w

1.44 (0.822.53)
1.77 (1.043.02)*

0.81 (0.501.31)
1.76 (1.102.80)*

24-h SBP versus PP (1 s.d.)

24-h SBP
0.81 (0.481.36)
24-h PP
2.12 (1.243.64)w

0.65 (0.202.13)
2.87 (0.958.70)

0.90 (0.501.63)
1.83 (0.983.42)

1.26 (0.782.04)
1.41 (0.862.31)

0.88 (0.411.91)
2.09 (0.954.63)

1.65 (0.883.12)
1.07 (0.562.03)

Daytime SBP versus PP (1 s.d.)

Daytime SBP
0.73 (0.431.22)
Daytime PP
2.25 (1.323.81)w

0.66 (0.212.10)
2.87 (0.988.40)

0.78 (0.441.40)
1.96 (1.063.64)*

1.31 (0.802.14)
1.40 (0.862.27)

0.99 (0.452.17)
2.04 (0.944.45)

1.59 (0.852.99)
1.08 (0.582.02)

Nighttime SBP versus PP (1 s.d.)

Nighttime SBP
1.04 (0.631.72)
Nighttime PP
1.67 (0.982.84)

0.49 (0.151.66)
3.13 (0.999.91)

1.35 (0.772.38)
1.32 (0.722.43)

1.12 (0.691.81)
1.30 (0.782.16)

0.64 (0.281.47)
2.07 (0.884.85)

1.59 (0.882.90)
0.95 (0.501.79)

Abbreviations: BP, blood pressure; CHD, coronary heart disease; CV, cardiovascular; CVD, cardiovascular disease; NCV, noncardiovascular;
PP, pulse pressure; SBP, systolic blood pressure.
Data are hazard ratios (95% confidence intervals) for each 1 s.d. higher blood pressure. Hazard ratios were also adjusted for baseline
characteristics including age, gender, smoking, concomitant medications (antihypertensive and lipid-lowering drugs), diabetes mellitus, previous
myocardial infarction, previous stroke, congestive heart failure and hyperlipidemia.
Significance of hazard ratios: *Po0.05; wPo0.01; zPo0.001.

Ambulatory SBP versus ambulatory PP

When both ambulatory SBP and PP were included in

the same model, 24-h PP and daytime PP could still
predict the all-cause mortality (HR, 2.12 and 2.25,
respectively). Interestingly, daytime PP, rather than

daytime SBP, also predicted the non-CV mortality

(HR, 1.96; 95% CI, 1.063.64; P 0.021) (Table 3).
Thus, compared with ambulatory SBP, ambulatory
PP could be a more powerful prognostic predictor,
especially in all-cause and non-CV mortality.
Journal of Human Hypertension

24-h pulse pressure and prognosis

Y-T Kao et al
448

Discussion
The main findings of this study are that ABPM
parameters have more predictive value in long-term
prognosis than office BP in both treated and
untreated hypertensive patients. Further, ambulatory PP parameters seem more predictive than
ambulatory SBP parameters. To our knowledge, this
could be the first study to show the significant
prognostic impacts of ambulatory PP parameters,
even superior to ambulatory SBP, on both all-cause
and CV mortality in clinical hypertension. Besides,
24-h ambulatory PP may be the single most
prognostic parameter, as it could predict all the
long-term CV and non-CV clinical outcomes in
hypertensive patients.
Our findings are partially in line with the very
first report that ABPM is a better predictor of
morbidity than in-office BP measurement.15 Given
the potential inaccuracies in measurement technique and artificial increases in BP produced by
white-coat effects,2 office or clinic BP measurement
could be inferior to ABPM in predicting hypertension-induced organ damage or clinical outcome.36
In the current study, office BP parameters including
SBP, DBP and even PP failed to predict any clinical
outcome. On the other hand, ambulatory BP parameters such as 24 h, daytime and nighttime PP
predicted not only CVD and mortality but also nonCV and all-cause mortality, suggesting their broad
prognostic impacts in hypertensive patients.
Previous European studies have shown that
ambulatory BP, over and above the conventional
BP, could predict the CV risk and mortality in
untreated hypertensives.4,16 Besides, ABPM also
offers more accurate prognostic information of CV
outcomes than office readings in treated hypertensive subjects.3,16 In this study, although treated
patients were older with more comorbidity than
the untreated patients, ABPM parameters could
still predict the mortality and CVD in both groups.
In treated hypertensives, the results remained
unchanged after adjusting for concomitant antihypertensive medication.
Regarding the fact that which BP parameter is
most prognostic, the answers were varied among
different studies. Several epidemiological studies
reported that PP is a useful predictor for CV
morbidity and mortality, especially in old people.17,18 However, recent cohort studies including
older people showed that the relationship of PP to
mortality from total CVD and CHD was less strong
than those of other BP indexes.19,20 In a metaanalysis, PP was found to be less informative in
the prediction of CHD and stroke mortality than SBP
or DBP.1 Recent studies also showed the complex
associations of PP with all-cause and CV mortality,
depending on age, SBP and DBP, and discouraged
the use of PP for diagnostic or therapeutic decisions.21 However, the PP in these studies was mainly
derived from office BP. In this study, although office

Journal of Human Hypertension

PP could only predict CHD, ambulatory PP could

predict all the clinical outcomes including all-cause,
CV and non-CV mortality and CV events. The
predictive role of 24-h ambulatory PP was even
better than other BP indexes. As a result, 24-h
ambulatory PP could be the single most informative
predictor for long-term prognosis of hypertension.
When concerning about daytime and nighttime
ambulatory BP parameters, nighttime BP seemed to
be a stronger predictor of outcome than daytime
BP.4,6 The BP parameters in these studies included
SBP and DBP but not PP. In this study, both daytime
and nighttime SBP could predict all-cause and nonCV mortality, but not CV mortality. More importantly, 24-h and daytime PP, rather than nighttime
PP, could predict all-cause mortality even adjusted
for ambulatory SBP. Given the similar prognostic
impacts of 24-h and daytime PP, the measurement of
daytime PP might be a convenient way for clinical
risk stratification. Further large-scale studies are
indicated to validate the current findings in different ethnic and geographic cohorts.
Previous ABPM studies were mainly conducted in
Europe3,4,6 and Japan.5 There was one large-scale
ABPM study composed of six population-based
cohorts from three continents.8 Recently, some
international meta-analyses for ABPM predictive
value were published7,8,22 in addition to singlenation population studies.4,23 However, given the
particular clinical importance of hypertension in
Chinese population, our study may not only validate
but also provide some novel rationales to clinical
use of ABPM in ethnic Chinese hypertensives.
Besides, previous studies have shown the prognostic significance of ABPM in hypertensive patients
with CVD24 or DM.25 In this study, most of the
patients were free from CVD before receiving ABPM.
They had low prevalence of previous MI, previous
stroke and congestive heart failure. Their prevalence
of type II DM was 11.4%, which was similar to the
general prevalence of DM in Taiwanese population.26 Therefore, the current findings may reflect
the real-world situation and should be validated in
other specific or complicated hypertensive cohorts if
indicated.
There were some limitations in this study. First,
although our patients were with relatively low risk
and less comorbidity, the potential effects of the
comorbidity on the current results cannot be
completely excluded. However, controlling these
variables in the multivariate analysis did not change
the results. Second, in patients with treated hypertension, the impact of different treatment regimens
was modest, which was controlled in the multivariate analysis. Finally, the use of CV drugs,
including antihypertensives and lipid-lowering
treatment, has been adjusted during analysis in this
study. However, detailed information about atrial
fibrillation, the use of anticoagulants and biochemistry data were not available, although some of
them might also contribute to CV events. Further

24-h pulse pressure and prognosis

Y-T Kao et al
449

study focus on the interaction between ABPM

parameters and other risk profiles may be considered if indicated.
In conclusion, ABPM parameters provide more
individualized risk stratification and prediction of
clinical outcomes than office BP in both treated and
untreated hypertensive patients. Ambulatory PP
rather than other parameters predicted the future
mortality and CV events, which may be applied for
universal risk stratification and consequent clinical
management in hypertensive patients before and
even during antihypertensive treatment.
What is known about topic
K Blood pressure is associated with the risk of cardiovascular
disease.
K Ambulatory blood pressure monitoring is superior to office
or clinic blood pressure in predicting hypertensioninduced organ damage or clinical outcome.
What this study adds
K Ambulatory pulse pressure is a good predictor for long-term
outcomes in hypertensive patients, no matter in treated or
untreated patients.
K The predictive value of ambulatory pulse pressure seems
superior to ambulatory systolic blood pressure.

Conflict of interest
The authors declare no conflict of interest.

Acknowledgements
This work was supported by grants including V95A011, V97C1-125 and V98A-015 from the Taipei
Veterans General Hospital, Taiwan, ROC.

References
1 Lewington S, Clarke R, Qizilbash N, Peto R, Collins R,
Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a
meta-analysis of individual data for one million
adults in 61 prospective studies. Lancet 2002; 360:
19031913.
2 Pickering TG, Hall JE, Appel LJ, Falkner BE, Graves J,
Hill MN, et al., Subcommittee of Professional and
Public Education of the American Heart Association
Council on High Blood Pressure Research. Recommendations for blood pressure measurement in humans
and experimental animals: Part 1: blood pressure
measurement in humans: a statement for professionals
from the Subcommittee of Professional and Public
Education of the American Heart Association Council
on High Blood Pressure Research. Hypertension 2005;
45: 142161.
3 Clement DL, De Buyzere ML, De Bacquer DA, de
Leeuw PW, Duprez DA, Fagard RH, et al., Office
versus Ambulatory Pressure Study Investigators. Prognostic value of ambulatory blood-pressure recordings
in patients with treated hypertension. N Engl J Med
2003; 348: 24072415.

4 Dolan E, Stanton A, Thijs L, Hinedi K, Atkins N,

McClory S et al. Superiority of ambulatory over clinic
blood pressure measurement in predicting mortality:
the Dublin outcome study. Hypertension 2005; 46:
156161.
5 Kikuya M, Ohkubo T, Asayama K, Metoki H, Obara T,
Saito S et al. Ambulatory blood pressure and 10-year
risk of cardiovascular and noncardiovascular mortality: the Ohasama study. Hypertension 2005; 45:
240245.
6 Sega R, Facchetti R, Bombelli M, Cesana G, Corrao G,
Grassi G et al. Prognostic value of ambulatory and
home blood pressures compared with office blood
pressure in the general population: follow-up results
from the Pressioni Arteriose Monitorate e Loro
Associazioni (PAMELA) study. Circulation 2005; 111:
17771783.
7 Conen D, Bamberg F. Noninvasive 24-h ambulatory
blood pressure and cardiovascular disease: a systematic review and meta-analysis. J Hypertens 2008; 26:
12901299.
8 Boggia J, Li Y, Thijs L, Hansen TW, Kikuya M,
Bjorklund-Bodegard K, et al., International Database
on Ambulatory Blood Pressure Monitoring in Relation
to Cardiovascular Outcomes (IDACO) Investigators.
Prognostic accuracy of day versus night ambulatory
blood pressure: a cohort study. Lancet 2007; 370:
12191229.
9 Yang HC, Liang YJ, Wu YL, Chung CM, Chiang KM, Ho
HY et al. Genome-wide association study of youngonset hypertension in the Han Chinese population of
Taiwan. PLoS One 2009; 4: e5459.
10 American Diabetes Association. Diagnosis and classification of diabetes mellitus. Diabetes Care 2004;
27(Suppl 1): S5S10.
11 National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of
High Blood Cholesterol in Adults (Adult Treatment
Panel III). Third Report of the National Cholesterol
Education Program (NCEP) Expert Panel on Detection,
Evaluation, and Treatment of High Blood Cholesterol
in Adults (Adult Treatment Panel III) final report.
Circulation 2002; 106: 31433421.
12 OBrien E, Mee F, Atkins N. An accurate automated
device for home blood pressure measurement at
last! The Omron HEM-705C. J Hypertens 1994; 12:
13171318.
13 OBrien E, Mee F, Atkins N, OMalley K. Accuracy of
the SpaceLabs 90207 determined by the British
Hypertension Society protocol. J Hypertens 1991; 9:
573574.
14 Lu TH, Lee MC, Chou MC. Accuracy of cause-of-death
coding in Taiwan: types of miscoding and effects on
mortality statistics. Int J Epidemiol 2000; 29: 336343.
15 Perloff D, Sokolow M, Cowan R. The prognostic
value of ambulatory blood pressures. JAMA 1983;
249: 27922798.
16 Staessen JA, Thijs L, Fagard R, OBrien ET, Clement D,
de Leeuw PW et al. Predicting cardiovascular risk
using conventional vs ambulatory blood pressure in
older patients with systolic hypertension. Systolic
Hypertension in Europe Trial Investigators. JAMA
1999; 282: 539546.
17 Franklin SS, Khan SA, Wong ND, Larson MG, Levy D.
Is pulse pressure useful in predicting risk for coronary
heart Disease? The Framingham heart study. Circulation 1999; 100: 354360.
Journal of Human Hypertension

24-h pulse pressure and prognosis

Y-T Kao et al
450

18 Staessen JA, Thijs L, OBrien ET, Bulpitt CJ, de Leeuw

PW, Fagard RH et al. Syst-Eur Trial Investigators.
Ambulatory pulse pressure as predictor of outcome in
older patients with systolic hypertension. Am J
Hypertens 2002; 15: 835843.
19 Miura K, Dyer AR, Greenland P, Daviglus ML, Hill M,
Liu K, et al., Chicago Heart Association. Pulse
pressure compared with other blood pressure indexes
in the prediction of 25-year cardiovascular and allcause mortality rates: the Chicago Heart Association
Detection Project in Industry Study. Hypertension
2001; 38: 232237.
20 Lawes CM, Bennett DA, Parag V, Woodward M,
Whitlock G, Lam TH, et al., Asia Pacific Cohort
Studies Collaboration. Blood pressure indices and
cardiovascular disease in the Asia Pacific region: a
pooled analysis. Hypertension 2003; 42: 6975.
21 Pastor-Barriuso R, Banegas JR, Damian J, Appel LJ,
Guallar E. Systolic blood pressure, diastolic blood
pressure, and pulse pressure: an evaluation of their joint
effect on mortality. Ann Intern Med 2003; 139: 731739.
22 Hansen TW, Kikuya M, Thijs L, Bjorklund-Bodegard K,
Kuznetsova T, Ohkubo T, et al., IDACO Investigators.
Prognostic superiority of daytime ambulatory over

Journal of Human Hypertension

23

24

25

26

conventional blood pressure in four populations: a

meta-analysis of 7030 individuals. J Hypertens 2007;
25: 15541564.
Inoue R, Ohkubo T, Kikuya M, Metoki H, Asayama K,
Obara T et al. Predicting stroke using 4 ambulatory
blood pressure monitoring-derived blood pressure
indices: the Ohasama Study. Hypertension 2006; 48:
877882.
Fagard RH, Thijs L, Staessen JA, Clement DL, De
Buyzere ML, De Bacquer DA. Prognostic significance
of ambulatory blood pressure in hypertensive patients
with history of cardiovascular disease. Blood Press
Monit 2008; 13: 325332.
Eguchi K, Pickering TG, Hoshide S, Ishikawa J,
Ishikawa S, Schwartz JE et al. Ambulatory blood
pressure is a better marker than clinic blood pressure
in predicting cardiovascular events in patients with/
without type 2 diabetes. Am J Hypertens 2008; 21:
443450.
Pan WH, Yeh WT, Chang HY, Hwu CM, Ho LT.
Prevalence and awareness of diabetes and mean fasting
glucose by age, sex, and region: results from the
Nutrition and Health Survey in Taiwan, 19931996.
Diabet Med 2003; 20: 182185.

Copyright of Journal of Human Hypertension is the property of Nature Publishing Group and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.