This article aims to compare the importance of systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), and pulse pressure (PP) as risk factors for stroke and ischemic heart disease and to assess whether the patterns are consistent by age and gender. Cox proportional-hazards regression, adjusted for cholesterol and smoking, was used to assess the associations of the 4 BP indices with stroke and ischemic heart disease by age and gender. The relative importance of individual indices was assessed with a hazard ratios for a 1-SD change in BP and by likelihood-ratio χ2 tests. The influence of >1 BP index in the Cox model was also estimated. The analyses demonstrated similar associations of SBP, DBP, and MAP with both fatal stroke and ischemic heart diseases, which were stronger than those of PP. Both SBP and MAP tended to be more important in the regression model than DBP or PP. In Cox models including DBP, addition of SBP improved the goodness of fit at all ages and for both genders. However, in Cox models including SBP, addition of DBP typically resulted in little incremental benefit over and above that of SBP alone. These data suggest that if time or resources are highly constrained, such as in much-needed epidemiologic surveys in developing countries, very little is lost from only measuring SBP.
Blood pressure (BP) is an important risk factor for cardiovascular disease, and it is a key component of any formula that predicts cardiovascular risk.1–4 However, uncertainties remain over which BP indices, systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), or pulse pressure (PP), are the most important risk factors for a cardiovascular event.
In the past, greater emphasis was always placed on DBP than SBP, because elevation of the former variable was thought to confer greater risk for cardiovascular disease than elevated SBP.5,6 Data from the Framingham study, published >30 years ago, demonstrated that this was not the case,7,8 and most guidelines now include both indices.9,10 More recently, attention has also focused on the role of MAP and PP in cardiovascular disease,11–15 and several analyses using different methodologies have explored the relative importance of all 4 BP indices. The majority of analyses have demonstrated that the association between SBP and cardiovascular disease is stronger than that of DBP in most age and gender groups.11–14 16–18 Neither MAP nor PP has been shown to be conclusively superior to SBP.11,12,14,15,18,19 Data from a large number of cohort studies have demonstrated that the relative magnitude of the association between SBP and DBP and cardiovascular disease differs by age.20,21 Several analyses have also suggested that the relative importance of BP indices may change with age and that DBP could be more important in those aged <50 years,14,18,22 but this finding is not consistent.13,16,17 Most analyses have not compared the BP indices by gender, and most have been based on North American or European populations.
Data from the Asia Pacific Cohort Studies Collaboration (APCSC) presented an ideal opportunity to compare all 4 BP indices for both fatal stroke and ischemic heart disease (IHD). This dataset has considerable follow-up data, which allowed analyses to assess whether the importance of BP indices varied by age, gender, and cardiovascular end point. Analyses of data from Asia, Australia, and New Zealand determined whether patterns in these populations were similar to those previously documented in predominantly European and North American studies. Secondary analyses were able to determine the influence of a combination of BP indices. This information will help clarify the relative importance of BP indices, and it also has the potential to contribute to better risk prediction.
Methods
Study Sample
The methods of study identification for the APCSC have been reported elsewhere.23 In brief, prospective cohort studies were eligible for inclusion if they included study populations from the Asia Pacific region with at least 5000 person-years of follow-up recorded or planned. They had to have recorded data on date of birth or age, gender, BP at baseline, and age at death. Additional data sought included date of the baseline survey, total blood cholesterol, diabetes, height, weight, smoking habit, and any data on repeated measures of risk factors. Outcome data included stroke and IHD.23 Data from a total of 37 cohort studies have been collated to date.
Statistical Methods
Primary analyses assessed the relation of individual BP indices to cardiovascular end points. MAP was calculated as 2/3 DBP+1/3 SBP, and PP as SBP−DBP. The analyses were performed on “one-off” measures (ie, taken at one point in time only) of BP, which are appropriate in a clinical setting. They were not adjusted for regression dilution bias.24 Because several different cohort studies were included, potential statistical heterogeneity was taken into account by stratifying by cohort. Additional effect modification was assessed with the use of statistical interaction terms for age and gender. When there was evidence of effect modification, analyses were stratified into the relevant subgroups. Cox proportional-hazards analyses25 were then used to regress time until first event against baseline BP with individual participant data collected in all cohorts. Analyses were undertaken for total (fatal and nonfatal) and fatal events separately and were also performed separately for 2 end points: stroke and IHD. Fatal events were defined as death occurring within 28 days after the stroke or IHD event; additional data on disease end points has been published elsewhere.23 Each BP index was considered separately as a continuous variable in the Cox regression. Regression coefficients and SEs were calculated with and without adjusting for cholesterol and smoking. Serum total cholesterol was fitted as a continuous variable (mmol/L), and smoking status was categorical (never vs ever smoked).
To compare the association of different BP indices with a particular end point, hazard ratios and 95% confidence intervals (CIs) were estimated for a 1-SD difference in each index. This is a “standardized” comparison of hazard ratios and is necessary because each BP index is measured on a different scale. The likelihood ratio χ2 was used as a measure of the improvement of goodness of fit,26 or “informativeness,” between a model containing each BP index (and the confounders smoking and cholesterol) compared with a model that only contained the confounders but no BP index (the “base model”) by age and gender.
Because some of the 37 cohorts in the collaboration included different age and gender subgroups, only a subset of cohorts is included in this report. This was done because one of the aims of the current analysis was to assess whether the relative importance of BP indices varied by age and gender. Comparisons of BP indices by using likelihood-ratio χ2 across age and gender subgroups are valid only if the regression models for males, females, and different age groups are nested within each other. This means that they must be based on the same participants, and thus, restriction of studies was necessary. Inclusion of studies that did not contribute to all age-gender subgroups would have made age and gender comparisons invalid, because it would not have been possible to determine whether patterns with age and gender were real or artifactual due to missing data. In total, a subset of 16 cohort studies included participants of both genders with age groups ranging from <50 years to ≥70 years and with data on BP, cholesterol, and smoking status. Participants with complete data from these 16 studies were used for the purposes of these analyses.
Secondary analyses consisted of including 2 BP indices in the Cox proportional-hazards model to assess the relation simultaneously with adjustment for each other. The likelihood-ratio χ2 test between the model containing a single BP index and the model containing >1 index was used to assess whether the additional index significantly improved the adequacy of the model. A significant likelihood-ratio χ2 indicates that the regression coefficient of the additional index is significant compared with zero;26,27 ie, the index provides significantly more information. If the 2 BP indices are highly correlated, then collinearity might become a problem. When considering the joint effects of SBP and DBP, we assessed collinearity by inspecting the changes in the coefficient and SE of SBP caused by the addition of DBP and vice versa. Large changes in either the parameter estimate or SE are indicative of collinearity.28
Results
Study Sample
Data from 16 cohort studies, which included 94 147 participants and >799 000 person-years of follow-up, were included in these analyses (Table 1). There were 3 cohorts from mainland China (13% of participants), 8 from Japan (17%), 1 from Taiwan (6%), 3 from Australia (61%), and 1 from New Zealand (3%). The mean age of participants at baseline was 52 years, and 54% of participants were female. During a mean follow-up of 8.5 years, a total of 1814 strokes (1120 fatal) and 1677 IHD events (1312 fatal) were recorded.
TABLE 1. Characteristics of the 16 APCSC Cohorts Used for Analyses to Compare BP Indices*
Region and Study Name
Start Year
Sample Size
Follow-Up, y
Female, %
Age at Baseline, y
Mean SBP
Mean DBP
Stroke Events
IHD Events
Mean
Range
Total
Fatal
Total
Fatal
SBP indicates systolic blood pressure; DBP, diastolic blood pressure; IHD, ischemic heart disease. Total person years of follow-up: 799 366.
*Only those APCSC studies and participants that had data on blood pressure, cholesterol, and smoking (defined as never and ever) and contained participants aged <50 and >70 years were included in these analyses.
**Weighted average, weighting by person years of follow-up.
Mainland China
Huashan
1990
1801
2.8
52
53
35–75
126
76
16
6
3
3
Seven Cities Cohorts
1987
6362
5.7
57
58
35–91
134
83
220
123
61
61
Shanghai Factory Workers
1972
3966
13.9
5
50
30–78
126
79
138
138
50
50
Japan
Aito Town
1980
1116
15.3
59
51
20–73
137
80
20
20
10
10
Hisayama
1961
1548
19.1
56
56
40–92
135
78
315
171
88
52
Konan
1987
1220
6.3
55
52
20–96
130
78
13
13
3
3
Miyama
1988
413
6.6
62
59
40–80
129
77
0
0
0
0
Ohasama
1992
1907
4.2
65
58
27–91
127
72
42
5
4
4
Saitama
1986
3615
9.6
62
55
20–94
135
80
55
55
24
24
Shibata
1977
2328
16.0
58
57
40–89
131
78
203
203
66
66
Shigaraki Town
1991
3722
3.9
59
57
29–95
132
78
13
13
3
3
Taiwan
CVDFACTS
1988
5540
6.4
56
47
20–92
118
75
29
29
12
12
Australia
Busselton
1966
7344
20.3
52
45
20–94
138
77
636
251
970
665
Melbourne
1990
41 141
5.7
59
55
27–75
138
77
32
32
161
161
Perth
1979
9729
12.5
47
45
20–90
130
81
59
59
186
186
New Zealand
Fletcher Challenge
1992
2395
4.8
22
40
20–88
124
77
23
2
36
12
Total or overall
94 147
8.5
54
52
20–96
134
78
1814
1120
1677
1312
Weighted average**
52
51
26–86
134
78
Investigating a Single BP Index
When the interaction terms for age and gender were introduced into the model, there were significant (P<0.05) differences between some of these subgroups, suggesting that there was an effect modification by age and gender. This effect modification was also evident in previous BP analyses of all APCSC cohorts.21 Therefore, all further analyses were performed by age (classified as <50, 50 to 59, 60 to 69, and ≥70 years) and gender. Analyses were carried out separately by country, and there were no consistent differences between countries. Country-specific results were also consistent with analyses of the overall group, which are presented in this article. (For example, in 60- to 69-year-old women, the association between BP and stroke was stronger for both SBP and MAP compared with DBP and PP across all countries.) In addition, separate analyses were performed that excluded participants who had a history of cardiovascular disease (for those studies that provided this information) and on several large APCSC cohorts excluded from the main analyses, and similar patterns were evident to those presented later. The difference between adjusted hazard ratios and those that were unadjusted for cholesterol and smoking was negligible, so adjusted analyses are presented.
Stroke
Analyses that were adjusted for cholesterol and smoking demonstrated that the association between each BP index and fatal stroke events was log-linear and positive and that the association was significant for most age and gender subgroups (Figure 1). The association between any given BP index and stroke was stronger in the youngest age groups; ie, the association was attenuated with age for males and females. The associations were similar in magnitude for all 4 BP indices within each subgroup; however, most frequently, the hazard ratios were highest for MAP and SBP, but the absolute differences were small. These patterns were also evident when analyses were performed on total stroke events.
Figure 1. Relation between BP and fatal stroke events by age and gender. Hazard ratios for fatal stroke events, adjusted for cholesterol and smoking, have been calculated for a 1-SD difference in BP for each index. Data are stratified by age (<50, 50 to 59, 60 to 69, and ≥70 years) and gender. Solid squares are larger where there were more events, because their size is proportional to the inverse variance, and horizontal lines represent 95% CIs. Likelihood-ratio χ2 statistics for fatal stroke events have been calculated by comparing a regression model including a single BP index (and confounders cholesterol and smoking) to the base model (which included cholesterol and smoking but no BP index).
The results of each BP index compared with the base model by age and gender subgroups are shown in Figure 1. Analyses demonstrated that in both males and females, SBP tended to be more informative than DBP (the only exception was in males aged <50 years). MAP ranked similarly to SBP in most age groups, and one or the other of these indices was most informative in all of the age-gender subgroups for fatal stroke events and total stroke events, the only exception being males aged ≥70 years. In contrast, PP was typically less important than SBP.
Ischemic Heart Disease
Adjusted analyses demonstrated that in all age groups, the association between each of the BP indices and fatal IHD events was log-linear and positive, but it was not statistically significant for all subgroups (Figure 2). Overall, the hazard ratios were of similar magnitude for DBP, SBP, and MAP but tended to be less for PP. This was consistent when analyses were performed on total events. A comparison of the information provided by each BP index in the regression model by age and gender was undertaken as before (Figure 2). Analyses demonstrated that SBP and MAP were more informative than DBP and PP in most subgroups.
Figure 2. Relation between BP and fatal IHD events by age and gender. Conventions are as described in the legend to Figure 1.
Joint Effects of BP Indices
In secondary analyses, SBP and DBP were simultaneously included in the Cox models to assess whether there was additional information provided for stroke or IHD compared with a model including either SBP or DBP alone. The hazard ratios and SEs for SBP and DBP were similar, regardless of the order in which these indices were included in the model, which is indicative of no major collinearity being present. MAP or PP was not included in models with SBP or DBP, because the former indices are already calculated from the latter.
In most age and gender subgroups, there was a relatively small change in the likelihood-ratio χ2 statistic for the SBP and DBP model compared with the SBP-alone model (“adding DBP to a model with SBP”). Most of these changes were not significant (Table 2). This indicates that DBP provided no significant incremental contribution over and above the information provided by SBP alone. The results for younger males (<50 years) were less clear. Analyses of total stroke and IHD events produced similar results. In contrast, comparison of the combined SBP and DBP model to the DBP-only model (“adding SBP to a model with DBP”) produced significant (P<0.05) changes in the likelihood-ratio χ2 statistic for most age-gender subgroups. This indicates a significant contribution of SBP over and above that for DBP in these subgroups.
TABLE 2. Increases in Goodness of Fit From Adding BP Indices to the Regression Model, by Age, Gender, and Fatal Outcome
Change in Model by Age Group
Males
Females
Likelihood Ratio χ2
P
Likelihood Ratio χ2
P
This table shows increases in goodness of fit from adding DBP to a model including SBP (and also age, gender, cholesterol, and smoking) and vice versa. The greater the likelihood ratio χ2, the greater the increase in goodness of fit or “informativeness” with the additional blood pressure index. The P value tests the statistical significance of the change.
Stroke
<50 years
Adding DBP to a model with SBP
24.1
<0.001
0.01
0.9
Adding SBP to a model with DBP
3.6
0.06
13.4
<0.001
50–59 years
Adding DBP to a model with SBP
2.3
0.1
0.1
0.8
Adding SBP to a model with DBP
36.9
<0.001
12.9
<0.001
60–69 years
Adding DBP to a model with SBP
2.2
0.1
0.8
0.4
Adding SBP to a model with DBP
19.9
<0.001
16.2
<0.001
70+ years
Adding DBP to a model with SBP
1.1
0.3
0.1
0.8
Adding SBP to a model with DBP
12.7
<0.001
5.8
0.01
Ischemic heart disease
<50 years
Adding DBP to a model with SBP
7.8
0.005
0.03
0.9
Adding SBP to a model with DBP
4.8
0.03
16.1
<0.001
50–59 years
Adding DBP to a model with SBP
2.6
0.1
3.6
0.06
Adding SBP to a model with DBP
6.5
0.01
5.2
0.02
60–69 years
Adding DBP to a model with SBP
2.1
0.2
2.1
0.2
Adding SBP to a model with DBP
25.6
<0.001
20.5
<0.001
70+ years
Adding DBP to a model with SBP
0.1
0.7
1.1
0.3
Adding SBP to a model with DBP
1.0
0.3
3.0
0.08
Discussion
These analyses clarify the impact of different BP indices as cardiovascular risk factors in the Asia Pacific region. Whereas the hazard ratios were of similar magnitude for all indices, the associations were typically stronger for SBP, DBP, and MAP than for PP, particularly in those aged >50 years. Cox models including a single BP index indicated that SBP or MAP provided more information than those including DBP or PP. Analyses also demonstrated broadly similar patterns for males and females, total and fatal events, and stroke and IHD end points. There was, however, attenuation of the association between all BP indices and cardiovascular disease end points with age. In those >50 years of age, the joint effect of SBP and DBP in the Cox model typically resulted in little incremental benefit over and above that of SBP alone. The converse was not true; inclusion of SBP in the Cox model did typically provide additional information over and above that provided by DBP alone.
The finding that overall SBP appears to be statistically more important than DBP in this “Asia Pacific” population is consistent with results from analyses of European and North American populations.11–13,16,17 Several analyses have concluded that DBP is superior to SBP in those aged <50 years.14,15,18,22 This was not completely borne out in the current study, but the results suggest that SBP and DBP may be of similar importance in the youngest age groups. Many authors have suggested that PP is an important predictor of cardiovascular disease,11,12,19,29 possibly more so than other BP measures. However, previous analyses frequently compared the hazard ratio for a 10–mm Hg change in each BP index.11,12,19,29 This comparison is dubious, because each index has a different scale; therefore, a 10 mm Hg change in one index is not comparable to a 10 mm Hg change in another. The current analyses standardized the comparison of hazard ratios by using a 1-SD difference in BP, which equated to ≈22 mm Hg SBP and 12 mm Hg DBP. In addition, several of the earlier analyses only included MAP and PP but did not include a comparison between PP and SBP.11,19,29 Our results are consistent with another recent analysis that used a similar methodology, which also concluded that the associations of SBP and MAP to cardiovascular disease were stronger than those of DBP and PP.13 Although we demonstrated that in most subgroups there was little incremental benefit of including both SBP and DBP in Cox regression models, recent analyses of Multiple Risk Factor Intervention Trial (MRFIT) data suggested that cardiovascular disease risk assessment was improved by considering both SBP and DBP, not just SBP or DBP alone.15 However, the current analyses were still broadly compatible, because the MRFIT sample only included men aged 35 to 57 years at baseline, and our analyses demonstrated improved goodness of fit with both indices in some younger subgroups.
There are some data limitations that could have influenced the current results, such as the limited number of events in some of the age-gender subgroups. However, even in the age group aged <50 years, there were 128 fatal IHD events (of 269 events) and 88 fatal strokes (of 227 events). The inclusion of a wider sample of studies from APCSC could have limited the comparability of the age-gender subgroups, because different cohort studies would have contributed to different subgroups. Analyses were performed with and without adjustment for cholesterol and smoking. However, because of the different ways that the cohorts had recorded smoking status, this had to be reduced to a dichotomous variable in these analyses. Data were not available from all cohort studies on diabetic status, electrocardiographic changes, and history of treatment for hypertension. When the current analyses were limited to participants without a history of cardiovascular disease (for those cohort studies that had provided this information), consistent results were obtained. In addition, adjustment for the included risk factors had negligible impact on hazard ratios, and there is no reason to believe that information on additional risk factors would have substantially changed the results.
These analyses were conducted on a large sample with >1800 stroke and 1600 IHD events, more than many previous analyses. Unlike several previous analyses, there was also a high proportion of females. The current analyses were performed on 1-off measures of BP that were not corrected for regression dilution bias,24 but which are more commonly available than “usual” BP. Correction for regression dilution would have led to stronger associations between all BP indices and end points but would not have changed the relative comparisons of SBP and DBP.21 Estimates of the size of the association between BP and cardiovascular disease from all 37 studies included in APCSC have been published elsewhere.21 In the age groups <60, 60 to 69, and ≥70 years, a 10–mm Hg lower usual SBP was associated with 54% (95% CI, 53% to 56%), 36% (34% to 38%), and 25% (22% to 28%) lower total stroke risk and 46% (43% to 49%), 24% (21% to 28%), and 16% (13% to 20%) lower total IHD risk, respectively.
Overall, these analyses have demonstrated that SBP and MAP were statistically the most important BP indices, but DBP is still a clinically important measurement and should not be discarded. In a clinical setting, use of SBP will be more practical and relevant than MAP, because the former is easier to measure and interpret.
Perspectives
In summary, this study suggests that although the hazard ratios for all 4 BP indices are similar, SBP is statistically more important than DBP as a risk factor for cardiovascular disease end points in Asia Pacific populations. The patterns were less clear for the youngest age groups; however, less of the burden of cardiovascular disease occurs in these age groups. This may, in part, reflect the fact that prevalent disease and arteriosclerosis selectively reduce DBP.30,31 Excessive emphasis on PP should be avoided, because there is no evidence that it has a stronger association with disease than SBP. In contrast, MAP often appeared as important as SBP. However, MAP is likely to have less clinical application than SBP and may be less advantageous overall. SBP is relatively simple to measure owing to a more distinct auscultatory end point than DBP, and MAP requires measurement of both SBP and DBP and a calculation. The use and interpretation of MAP is also unlikely to be practical, because few guidelines exist for this index. Because there is no definite evidence that MAP or PP was clearly superior to DBP and SBP, the 2 former measures do not need to be incorporated into future guidelines.32 Most clinicians think in terms of SBP and DBP, and these 2 measures remain important and clinically useful. However, these data suggest that if time or resources are highly constrained, such as in much-needed epidemiologic surveys in developing countries, very little is lost from measuring only SBP.
Appendix: Asia Pacific Cohort Studies Collaboration
Writing Committee
Carlene M.M. Lawes, Derrick A. Bennett, Varsha Parag, Mark Woodward, Gary Whitlock, Tai Hing Lam, Il Suh, and Anthony Rodgers
Statistical Analyses
V. Parag, D.A. Bennett, S. Vander Hoorn, M. Woodward, and F. Barzi
Executive Committee
D.A. Bennett, D.F. Gu, H. Horibe, S. MacMahon, W.H. Pan, A. Rodgers, I. Suh, H. Ueshima, M. Woodward, Z.L. Wu, and X.H. Zhang
Participating Studies and Principal Collaborators
Aito Town: A. Okayama, H. Ueshima, H. Maegawa; Akabane: N. Aoki, M. Nakamura, N. Kubo, T. Yamada; Anzhen02: Z.S. Wu; Anzhen: C.H. Yao, Z.S. Wu; Beijing Steelworkers: L.S. Liu, J.X. Xie; Busselton: M.W. Knuiman; Canberra-Queanbeyan: H. Christensen; Capital Iron and Steel Company: X.G. Wu; CISCH: J. Zhou, X.H. Yu; Civil Service Workers: A. Tamakoshi; East Beijing: Z.L. Wu, L.Q. Chen, G.L. Shan; Fangshan: D.F. Gu, X.F. Duan; Fletcher Challenge: S. MacMahon, R. Norton, G. Whitlock, R. Jackson; Hisayama: M. Fujishima, Y. Kiyohara, H. Iwamoto; Hong Kong: J. Woo, S. Ho; Huashan: Z. Hong, M.S. Huang, B. Zhou; Kinmen: J.L. Fuh; Konan: H. Ueshima, Y. Kita, S.R. Choudhury; KMIC: I. Suh, S.H. Jee, I.S. Kim; Melbourne: G. Giles; Miyama: T. Hashimoto, K. Sakata; Ohasama: Y. Imai, T. Ohkubo, A. Hozawa; Perth: K. Jamrozik, M. Hobbs, R. Broadhurst; Saitama: K. Nakachi; Seven Cities Cohorts: X.H. Fang, S.C. Li, Q.D. Yang; Shanghai FactoryWorkers:Z.M. Chen; Shibata: H. Tanaka; Shigaraki Town: Y. Kita, A. Nozaki, H. Ueshima; Shirakawa: H. Horibe, Y. Matsutani, M. Kagaya; Singapore Heart: K. Hughes, J. Lee; Singapore NHS92: D. Heng; Six Cohorts: B.F. Zhou, H.Y. Zhang; Tanno/Soubetsu: K. Shimamoto, S. Saitoh; Tianjin: Z.Z. Li, H.Y. Zhang; CVDFACTS: W.H. Pan; Xi’an: Y. He, T.H. Lam; Yunnan: S.X. Yao. (The underlined studies provided data used in this article.)
Acknowledgments
The Asia Pacific Cohort Studies Collaboration has been supported by grants from the Health Research Council of New Zealand and National Health and the Medical Research Council of Australia. We thank Sarah Lewington for helpful comments on an earlier version of this report.
Footnote
The Writing Committee and members are listed in the Appendix.
References
1.
Shaper AG, Pocock SJ, Phillips AN, Walker M. Identifying men at high risk of heart attacks: strategy for use in general practice. BMJ Clin Res Ed. 1986; 293: 474–479.
Lloyd-Jones DM, Evans JC, Larson MG, O’Donnell CJ, Levy D. Differential impact of systolic and diastolic blood pressure level on JNC-VI staging: Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 1999; 34: 381–385.
Kannel WB. Historic perspectives on the relative contributions of diastolic and systolic blood pressure elevation to cardiovascular risk profile. Am Heart J. 1999; 138 (pt 2): 205–210.
Kannel WB, Wolf PA, Verter J, McNamara PM. Epidemiologic assessment of the role of blood pressure in stroke: the Framingham study. JAMA. 1970; 214: 301–310.
Kannel WB, Gordon T, Schwartz MJ. Systolic versus diastolic blood pressure and risk of coronary heart disease: the Framingham study. Am J Cardiol. 1971; 27: 335–346.
The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med. 1997; 157: 2413–2446.
Guidelines Subcommittee. World Health Organization–International Society of Hypertension guidelines for the management of hypertension. J Hypertens. 1999;1999: 17: 151–183.
Mitchell GF, Moye 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.
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: 354–360.
Miura K, Dyer AR, Greenland P, Daviglus ML, Hill M, Liu K, Garside DB, Stamler J. Pulse pressure compared with other blood pressure indexes in the prediction of 25-year cardiovascular and all-cause mortality rates. The Chicago Heart Association Detection Project in Industry Study. Hypertension. 2001; 38: 232–237.
Sesso HD, Stampfer MJ, Rosner B, Hennekens CH, Gaziano JM, Manson JE, Glynn RJ. Systolic and diastolic blood pressure, pulse pressure, and mean arterial pressure as predictors of cardiovascular disease risk in men. Hypertension. 2000; 36: 801–807.
Domanski M, Mitchell G, Pfeffer M, Neaton JD, Norman J, Svendsen K, Grimm R, Cohen J, Stamler J. Pulse pressure and cardiovascular disease-related mortality: follow-up study of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 2002; 287: 2677–2683.
Franklin SS, Larson MG, Khan SA, Wong ND, Leip EP, Kannel WB, Levy D. Does the relation of blood pressure to coronary heart disease risk change with aging? The Framingham Heart Study. Circulation. 2001; 103: 1245–1249.
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: 1903–1913.
Tverdal A. Systolic and diastolic blood pressures as predictors of coronary heart disease in middle aged Norwegian men. BMJ Clin Res Ed. 1987; 294: 671–673.
Asia Pacific Cohort Studies Collaboration. Determinants of cardiovascular disease in the Asia Pacific region: protocol for a collaborative overview of cohort studies. CVD Prev. 1999; 2: 281–289.
Psaty BM, Furberg CD, Kuller LH, Cushman M, Savage PJ, Levine D, O’Leary DH, Bryan RN, Anderson M, Lumley T. Association between blood pressure level and the risk of myocardial infarction, stroke, and total mortality: the cardiovascular health study. Arch Int Med. 2001; 161: 1183–1192.
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.
Correspondence to Carlene Lawes, Clinical Trials Research Unit, University of Auckland, Private Bag 92019, Auckland, New Zealand. E-mail [email protected]
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
Chong Hu,
Ziqi Yu,
Changli Wei,
Guotai Sheng,
Jianyong Chen,
Yang Zou,
Evaluating the relative importance of different blood pressure indices in screening for NAFLD: a survey report based on a health examination population, Frontiers in Cardiovascular Medicine, 11, (2024).https://doi.org/10.3389/fcvm.2024.1338156
Knowledge of Home Blood Pressure Monitoring among Medical Students in Malaysia, Malaysian Journal of Medicine and Health Sciences, 19, s17, (41-47), (2023).https://doi.org/10.47836/mjmhs.19.s17.6
What did we learn from the International Databases on Ambulatory and Home Blood Pressure in Relation to Cardiovascular Outcome?, Hypertension Research, 46, 4, (934-949), (2023).https://doi.org/10.1038/s41440-023-01191-4
Blood pressure levels within normotensive range are independently associated with increased risk of arterial stiffness in adults without hypertension or prehypertension, Nutrition, Metabolism and Cardiovascular Diseases, 33, 12, (2363-2371), (2023).https://doi.org/10.1016/j.numecd.2023.08.009
Mean systolic blood pressure above the control threshold in people with treated uncontrolled hypertension: a pooled, cross-sectional analysis of 55 national health surveys, eClinicalMedicine, 57, (101833), (2023).https://doi.org/10.1016/j.eclinm.2023.101833
Relationship of Four Blood Pressure Indexes to Subclinical Cerebrovascular Diseases Assessed by Brain MRI in General Japanese Men, Journal of Atherosclerosis and Thrombosis, 29, 2, (174-187), (2022).https://doi.org/10.5551/jat.58537
Blood Pressure Indices and Cardiovascular Disease in the Asia Pacific Region
Hypertension
Vol. 42
No. 1
Purchase access to this journal for 24 hours
Hypertension
Vol. 42
No. 1
Restore your content access
Note: This functionality works only for purchases done as a guest. If you already have an account, log in to access the content to which you are entitled.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.