Abstract
This study aimed to quantitatively evaluate the predictive value of brachial pulse pressure and cardiovascular or all‐cause mortality in the general population based on prospective observational studies by conducting a meta‐analysis. Only prospective observational studies investigating baseline brachial pulse pressure and cardiovascular or all‐cause mortality risk were selected from PubMed and Embase databases until July 2013. Fourteen studies involving 510,456 participants were analyzed. Pooled risk ratio (RR) of cardiovascular and all‐cause mortality for the highest vs lowest brachial pulse pressure category was 1.80 (95% confidence interval [CI], 1.49–2.17) and 1.32 (95% CI, 1.23–1.41), respectively. Pooled RR of cardiovascular and all‐cause mortality per 10 mm Hg pulse pressure increment was 1.13 (95% CI, 1.10–1.17) and 1.09 (95% CI, 1.07–1.11), respectively. Wide brachial pulse pressure is associated with greater risk of cardiovascular and all‐cause mortality. However, more well‐designed studies specifically on age and sex are needed to further confirm these findings.
Cardiovascular disease (CVD) is the leading cause of mortality and a primary contributor to the burden of disease worldwide.1 Hypertension is one of the modifiable risk factors for CVD mortality and accounts for up to 30% of deaths in the world.2 Brachial pulse pressure (PP) is defined as the difference between systolic blood pressure (SBP) and diastolic blood pressure (DBP) at the brachial level.
Increasing attention has focused on the role of different blood pressure (BP) components as predictors for CVD and death.3, 4 Some evidence suggests that wide PP is a risk factor for cardiovascular (CV) or all‐cause mortality.5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 However, conflicting reports still exist.17, 18, 19 These inconsistent results could be partly explained by PP simply being associated with mortality, not a causative factor. PP is higher after age 50 mainly because the trend of DBP gets lower and SBP gets higher with age.20, 21 Moreover, there is a lack of accurate information concerning the real risk of mortality in the various study populations. Apart from PP, other BP components such as SBP, DBP, and mean arterial pressure (MAP) were also used to predict mortality in the same studies.6, 8, 9, 10, 12, 14, 16, 17
Large artery stiffness contributes to wide PP.22 Wide PP is linked to higher levels of the inflammatory state.23 These findings suggest that both wide PP and inflammatory state share the common risk factor for CVD and thereby increase the risk of mortality.
To the best of our knowledge, no previous meta‐analysis has been performed to estimate the magnitude between brachial PP and CV or all‐cause mortality risk in the general population. Therefore, we conducted this meta‐analysis using prospective observational studies to quantitatively assess the baseline brachial PP and risk of CV or all‐cause mortality in the general population.
Materials and Methods
Search Strategy
This study was carried out according to the checklist of the Meta‐Analysis of Observational Studies in Epidemiology24 and followed by the PRISMA rules. We performed searches of PubMed (1966 to July 2013) and Embase (1980 to July 2013) for relevant prospective observational studies. Only papers published in English language were considered. Studies were to assess baseline brachial PP and subsequent CV or all‐cause mortality events in the studied populations. MeSH terms used to identify papers included “pulse pressure,” “blood pressure,” “cardiovascular mortality,” “all‐cause mortality,” “total mortality,” and “death” or “mortality” and included observational, prospective, and follow‐up studies. In addition, we also retrieved the reference lists of the selected papers.
Study Selection
Studies were included in the meta‐analysis if they had: (1) prospective design; (2) a follow‐up duration of at least 5 years; (3) reported risk estimates for the baseline brachial PP and CV or all‐cause mortality in the general populations; and (4) provided at least age‐adjusted risk ratio (RR) or hazard ratio (HR) with 95% confidence intervals (CIs) for brachial PP comparing the highest with the lowest category and/or as a continuous variable (per 10 mm Hg brachial PP increments). Studies were excluded when: (1) the study design was cross‐sectional or retrospective; (2) unadjusted RR or HR was reported; (3) the study did not select the lowest brachial PP category as the reference; and (4) the patients were from a particular occupation or high‐risk populations.
The outcomes were CV and all‐cause mortality. CV and all‐cause mortality were defined according to the death certificate underlying cause of death coded according to the International Classification of Diseases, Ninth Revision (ICD‐9).
Data Extraction and Quality Assessment
Two reviewers (Y.J.S. and Z.J.L.) independently extracted the data from each study. The following items were extracted from each study: first author's surname; year of publication; geographic origin of study; design; duration of follow‐up; the number of patients, sex, and age; the data with the most fully adjusted RR or HR; the number of deaths; and statistical adjustments for confounding factors. The quality of each study obtained from the literature search was evaluated by the same two reviewers according to the meta‐analysis of Observational Studies in Epidemiology (MOOSE) reporting guidelines.24, 25
Data Synthesis and Analysis
Data analyses were conducted using multivariate‐adjusted RR and 95% CIs. We pooled the separate RR estimates for the different items comparing the highest brachial PP category vs the lowest category from each study. For continuous risk estimates, we pooled RR of per 10 mm Hg PP increment. Studies reporting the different scales for continuous values were recalculated into per 10 mm Hg increment (RR per 10 mm Hg=exp ((ln (RRSD)/SD) ×10).
Heterogeneity due to effect sizes across studies was assessed with the Q statistic and I 2 statistic. If P>.10 or I 2<50% were taken as indicators of no significance, heterogeneity was assessed using a fixed effects model. If P≤.10 or I 2>50% were taken as indicators of significant, heterogeneity of outcomes was assessed using the random‐effects model.26 Begg's rank correlation test27 and Egger's linear regression test28 were used to evaluate the publication bias, with P<.10 indicating statistical significance. All analyses were performed with STATA statistical software (version 12.0; StataCorp, College Station, TX). Statistical significance was defined as P<.05.
Results
Study Selection and Baseline Characteristics
After application of the initial search strategy, a total of 837 potential papers were identified and reviewed. After screening the abstracts and titles, we excluded 801 studies because they were reviews, animal studies, or not relevant to our analysis. Thirty‐six full‐text studies were retrieved for detailed evaluation. Fifteen studies appeared to meet the inclusion criteria. However, one study15 reporting the results in 62.5 mm Hg brachial PP vs the rest was excluded. Finally, 14 studies5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 16, 17, 18, 19 involving 510,456 participants met our inclusion criteria. A detailed step of the study selection is presented in Figure 1.
Figure 1.
Flow chart of study selection process for meta‐analysis.
All of the PPs were reported to be measured by conventional brachial sphygmomanometry. Most of the studies were conducted in Europe. Detailed characteristics of the included trials are shown in Table 1. The qualities of the included studies are shown in Table 2.
Table 1.
Summary of Clinical Studies Included in the Meta‐Analysis
| Study/Year | Country | Design | Patients (% Women) | Age/Range Mean (SD) | Comparison | Outcome Assessment | RR or HR (95% CI)/Events | Follow‐Up, y | Adjustment for Covariates |
|---|---|---|---|---|---|---|---|---|---|
| Benetos et al17/1998 | France | Prospective cohort study | 19,955 (37.3) | 40–69 | Per 10 mm Hg increment | Mortality records of INSEE. ICD‐8 code (until 1978) and ICD‐9 code (after 1978) |
1.00 (0.86–1.17) NT+F+total (333) 1.07 (0.99–1.14) HT+F+total (378) 1.07 (0.97–1.18) NT+M+total (910) 1.07 (1.04–1.11) HT+M+total (1514) 0.85 (0.60–1.21) NT+F+CVD (69) 1.00 (0.91–1.11) HT+F+CVD (160) 1.20 (1.01–1.44) NT+M+CVD (249) 1.09 (1.03–1.14) HT+M+CVD (605) |
19.5 | Age, mean values of BMI, blood pressure, and total cholesterol |
| Antikainen et al5/2000 | Finland | Prospective cohort study | 9603 (54.9) | 45–64 | Per 10 mm Hg increment[Link] | Death certificates. ICD‐9 code 390–458 |
1.094 (1.062–1.138) M+total (1142) 1.116 (1.072–1.161)+F+total (622) 1.149 (1.094–1.195) +M+CVD (655) 1.195 (1.127–1.268) +F+CVD (313) |
15 | Age, BMI, serum cholesterol, and smoking |
| Schram et al18/2002 | The Netherlands | Prospective cohort study | 2484 (54.4) | 50–74 | Per 10 mm Hg increment | Medical record ICD‐9 code 390–459; code 798 |
0.98 (0.85–1.13) CVD (116) no DB 1.27 (1.00–1.61) CVD (34) DB 1.02 (0.92–1.12) total (265) no DB 1.12 (0.93–1.34) total (65) DB |
8.8 8.6 |
Age, sex, mean arterial pressure |
| Strandberg et al19/2002 | Finland | Prospective cohort study | 3267 (0) | 37.3 (4.2) | Per 10 mm Hg increment | Death certificates. ICD‐9 code 410–459 |
CVD (325) 0.99 (0.90–1.10) |
27.0 | Age, BMI, serum cholesterol, SBP, DBP, and mean arterial pressure |
| Panagiotakos et al16/2005 | Greece | Prospective, population‐based cohort study | 12,763 (0) | 40–59 | Per 10 mm Hg increment | Medical record. ICD‐8 code |
CVD (2445) 1.22 (1.10–1.34) |
20.3 | Age, BMI, height, total cholesterol, smoking habits, and physical activity status |
| Bowman et al12/2006 | United States | Prospective cohort study | 53,163 (0) | 39–85 | Per 10 mm Hg increment | Death certificates. ICD‐9 code 390–459 |
CVD (459) 1.25 (1.14–1.36) |
5.7 | Age, cigarette smoking, BMI, DB, alcohol intake, exercise, aspirin use, and multivitamin use |
| Miura et al8/2001 | Japan | Prospective study | 28,360 (27.9) | 18–74 | Per 10 mm Hg increment[Link] |
National Death Index records or death certificates. ICD‐8 code 400–445.9 |
CVD (1807) M 1.13 (1.03–1.25) 18–39 y 1.10 (1.06–1.15) 40–59 y 1.10 (1.04–1.17) 60–74 y CVD (760) F 1.13 (1.06–1.20) 40–59 y 1.08 (1.00–1.16) 60–74 y |
25 | Age, cholesterol, cigarettes per day, BMI, BMI, ECG abnormality, race, and education |
| Van Trijp et al10/2005 | The Netherlands | Prospective cohort study | 7813 (100) | 49–66 | Per 10 mm Hg increment[Link] | Municipal authorities ICD‐9 code 390–459. |
CVD (313) 1.22 (1.14–1.31) |
13.1 | Age, BMI, DB, smoking, SBP, DBP, MAP, and use of blood pressure–lowering medication |
| Hadaegh et al14/2012 | Iran | Prospective study | 5991 (54.8) | ≥30 y | Per 10 mm Hg increment[Link] | Medical files |
Total (157) 1.55 (1.24 – 1.93) <60 y 1.17 (1.04 – 1.32) >60 y CVD (67) 1.23 (1.05 – 1.44) |
8.7 | Age, sex, smoking status, DB, intervention, and BMI |
| Glynn et al6/2000 | United States | Prospective study | 9431 (NP) | 65–102 | Highest tertile vs lowest tertile ≥77 vs <53 mm Hg | Death certificates. ICD‐9 code 401–459 |
Total (4528) 1.34 (1.23–1.46) CVD (2304) 1.57 (1.39–1.77) |
10.6 | Age, sex, and stratified site |
| Fang et al7/2000 | United States | Prospective study | 7346 (NP) | 25–74 |
Highest tertile vs lowest tertile ≥49 vs <37 mm Hg among men; Highest tertile vs lowest tertile ≥48 vs <35 mm Hg among women Per 10 mm Hg increase |
Medical records and death certificates. ICD‐9 code 401–459 |
Total (1443) 1.20 (1.06–1.49) M 1.25 (1.11–1.60) F CVD (577) 1.45 (1.12–2.01) Male 1.52 (1.18–2.11) Female 2.06 (1.31–3.24)<55 y 1.08 (0.86–1.34)≥55 y |
17.4 |
Age‐/race‐adjusted For CVD: Age, race, BMI, cholesterol, smoking status, and history of cardiovascular disease |
| Domanski et al9/2002 | United States | Prospective study | 342,815 (0) | 35–57 | Highest tertile vs lowest tertile ≥51 vs <39 mm Hg among 35–44 y; Highest tertile vs lowest tertile ≥55 vs <40 mm Hg among 45–57 y | National Death Index and Social Security file. ICD‐9: 393–459, ICD‐10: I00–I99 |
CVD (5440) 1.45 (1.35–1.56) 35–44 y CVD (20,281) 2.01 (1.93–2.10) 45–57 y |
22 | Age, race cholesterol, and number of cigarettes smoked per day |
| Garcia‐Palmieri et al11/2005 | United States | Prospective study | 9824 (0) | 35–79 | Highest quintile vs lowest quintile ≥57 vs <38 mm Hg | Death certificates and medical records |
CVD (650) 3.33 (2.59–4.39) |
12 | Age, education, smoking status, high blood cholesterol status, DB, physical activity, and residual of mean arterial pressure |
| Lorenzo et al13/2009 | United States | Prospective cohort study | 3632 (NP) | 25–64 | Highest tertile vs lowest tertile ≥46 vs ≤37 mm Hg | Death certificates. ICD‐9 code 401–405, 410–414, 420–429, 430–439, 440–447. |
Total (548) 1.86 (1.27–2.74) CVD (261) 2.17 (1.05–4.50) |
15.2 | Age, sex, ethnic origin, education, BMI, current cigarette smoking, and total cholesterol concentration |
Abbreviations: BMI, body mass index; CHD, coronary heart disease; CVD, cardiovascular disease; DB, diabetes mellitus; DBP, diastolic blood pressure; F, female; HR, hazard ratio; HT, hypertensive; ICD, International Classification of Diseases; INSEE, Institut National de Statistiques et d'Etudes conomiques; M, male; MAP, mean arterial pressure; NP, not provide; NT, normotensive; RR, relative risk; SBP, systolic blood pressure. aRR from re‐calculation.
Table 2.
Quality Assessment of Studies Included in the Meta‐Analysis
| Study/Year | Clear Inclusion and Exclusion Criteria | Document the Loss to Follow‐Up Rate | Clear Definition of Outcome | Sufficient Duration Follow‐Up (>10 y) | Control of Confounding | Appropriate Statistics |
|---|---|---|---|---|---|---|
| Benetos et al17/1998 | Yes | Yes | Yes | Yes | Yes | Yes |
| Antikainen et al5/2000 | Yes | Yes | Yes | Yes | Yes | Yes |
| Schram et al18/2002 | Yes | Yes | Yes | No | Yes | Yes |
| Strandberg et al19/2002 | Yes | Yes | Yes | Yes | Yes | Yes |
| Panagiotakos et al16/2005 | Yes | Yes | No | Yes | Yes | Yes |
| Bowman et al12/2006 | Yes | No | Yes | No | Yes | Yes |
| Miura et al8/2001 | Yes | Yes | Yes | Yes | Yes | Yes |
| Van Trijp et al10/2005 | Yes | Yes | Yes | Yes | Yes | Yes |
| Hadaegh et al14/2012 | Yes | Yes | No | No | Yes | Yes |
| Glynn et al6/2000 | Yes | Yes | Yes | Yes | Yes | Yes |
| Fang et al7/2000 | Yes | No | Yes | Yes | Yes | Yes |
| Domanski et al9/2002 | Yes | Yes | Yes | Yes | Yes | Yes |
| Garcia‐Palmieri et al11/2005 | Yes | Yes | No | Yes | Yes | Yes |
| Lorenzo et al13/2009 | Yes | Yes | Yes | Yes | Yes | Yes |
CV Mortality
Five studies6, 7, 9, 11, 13 reported CV mortality in the highest brachial PP category vs the lowest category. The total number of participants included in this meta‐analysis was 373,048, with 29,513 reported CV mortality events. As is shown in Figure 2, elevated brachial PP was associated with an increase in CV mortality in a random‐effects model comparing the highest brachial PP category vs the lowest category group (RR, 1.80; 95% CI, 1.49–2.17). Substantial heterogeneity was shown (I 2=93.3%; P=.000). No significant publication bias was observed according to the Begg's rank correlation test (P=1.000) or Egger's linear regression test (P=.742).
Figure 2.
Relative risk (RRs) and 95% confidence interval (CIs) of elevated pulse pressure and cardiovascular mortality comparing the highest pulse pressure with the lowest category group in a random‐effects model.
Ten studies5, 7, 8, 10, 12, 14, 16, 17, 18, 19 reported CV mortality per 10 mm Hg brachial PP increment. The total number of participants included in this meta‐analysis was 150,745, with 8934 reported CV mortality events. As is shown in Figure 3, the pooled adjusted RR of CV mortality per 10 mm Hg brachial PP increments was 1.13 (95% CI, 1.10–1.17) in a random‐effects model, with obvious heterogeneity (I 2=58.3%; P=.001). No significant publication bias was observed according to the Begg's rank correlation test (P=.770) or Egger's linear regression test (P=.471).
Figure 3.
Relative risk (RRs) and 95% confidence interval (CIs) of elevated pulse pressure and cardiovascular mortality per 10 mm Hg pulse pressure increment in a random‐effects model.
All‐Cause Mortality
Three studies6, 7, 13 reported the highest brachial PP category vs the lowest category and risk of all‐cause mortality. The total number of participants included in this meta‐analysis was 20,409, with 6519 reported all‐cause mortality events. As is shown in Figure 4A, elevated brachial PP was associated with an increase in all‐cause mortality in a fixed‐effect model comparing the highest brachial PP category vs the lowest category group (RR, 1.32; 95% CI, 1.23–1.41). Substantial heterogeneity was not shown (I 2=36.4%; P=.194).
Figure 4.
Relative risk (RRs) and 95% confidence interval (CIs) of elevated pulse pressure and all‐cause mortality comparing the highest pulse pressure with the lowest category group (A) and per 10 mm Hg pulse pressure increment (B) in a fixed‐effects model.
Four studies5, 14, 17, 18 reported per 10 mm Hg brachial PP increment and the risk of all‐cause mortality. The total number of participants included in this meta‐analysis was 38,033, with 5386 reported all‐cause mortality events. As is shown in Figure 4B, the pooled adjusted RR of all‐cause mortality per 10 mm Hg PP increments was 1.09 (95% CI, 1.07–1.11) in a fixed‐effects model, with moderate heterogeneity (I 2=47.9%; P=.045).
Discussion
Findings from this meta‐analysis indicate that wide brachial PP is associated with a significantly increased risk of CV and all‐cause mortality. Individuals with wide brachial PP have an 80% and 32% increased risk of CV and all‐cause mortality, respectively, comparing the highest brachial pulse pressure category with the lowest category. The robustness of brachial PP as a risk factor was reinforced by the clear relationship of brachial PP as a continuous risk estimates (per 10 mm Hg increment) with CV and all‐cause mortality, independent of all other confounding. For each 10 mm Hg increment in brachial PP, the risk of CV mortality and all‐cause mortality was 1.13 and 1.09, respectively. These findings suggest that wide PP increases greater risk of mortality.
PP is higher after age 50 due to the tendency of SBP to increase and DBP to decrease with age, which is attributed to arterial stiffness and pulse wave reflection.21, 29 Aging is associated with a loss of elasticity of the aorta and major arterial conduits. However, only a small number of studies provided RRs stratified by age, therefore this question requires further investigation. The relatively lower risk of mortality events for older adults compared with younger patients in two studies8, 14 could be interpreted in many ways. The first reason is that antihypertensive treatment may decrease the risk of mortality events. The second reason could be attributed to the management of the coexistence of other comorbidities in the older adults resulting in a lower risk of mortality events. Another reason for the relatively lower risk of mortality events for the older adults is that many older adults with CVD may have died before the primary study was performed. Therefore, the population included in the study population might not have been a true representation of the population that was being studied.
Many studies that did not meet the inclusion criteria for the meta‐analysis also addressed the association between wide brachial PP and risk of mortality events. In the Cardiovascular Study in the Elderly30 involving 3282 patients aged 65 years and older, coronary mortality in women was predicted by brachial PP (1.01 excess risk per 1 mm Hg increment) and was significantly higher in the third than in the first tertile of PP (RR 2.90). A meta‐analysis investigating the prognostic value of PP in the European Working Party on Hypertension in the Elderly (EWPHE) showed a comparable predictive effect of PP on coronary events and stroke after adjustment for mean BP.31 Another well‐designed meta‐analysis demonstrated that central PP determined by central hemodynamics has a marginally but not significantly (P=.057) better predictive ability when compared with peripheral PP for prediction of CV events and all‐cause mortality.32
PP is dependent on stroke volume and arterial wall elastic properties. Wide brachial PP can contribute to mortality events through a variety of mechanisms. First, wide brachial PP can be the result of arterial stiffening that leads to increased systemic load and subsequent risk of CV death.9, 33 In addition, elevated PP was associated with a higher risk of inflammation‐dependent atherosclerotic CVD.17, 34 Furthermore, wide PP was both a cause and a consequence of atherosclerosis.23, 35
Wide brachial PP could be used to predict mortality risk in studied populations. However, there is no evidence to support that wide PP is superior to SBP in the prediction of mortality in younger8 or middle‐aged populations.10, 16 MAP measurement may be complicated in the clinical practice, and it was no better than SBP in risk prediction.14, 36 It is still uncertain whether PP is superior to SBP, DBP, and MAP in the prediction of subsequent mortality risk in various studied population. Therefore, all BP components as predictors are warranted.
Strengths and Limitations
Although we only included prospective observational studies in this analysis, there were still several weaknesses in the current study. First, use of published data, including the absence of standardization in study design, duration, outcome definition, characteristics of study populations, and adjustment for variables across studies is an inherent limitation. Individual studies did not adjust for potential risk factors in a consistent way or lack of adjustment for important confounding such as MAP might have resulted in a slight overestimation of the risk estimate. Second, brachial PP measurement was obtained at a single clinic visit in most of the studies, which might have misclassified the estimated risk of mortality. In addition, there was no standardization of BP measurement technique and it was not known whether the individuals measuring the BP were trained and demonstrated to be competent in BP measurement in selected studies. Moreover, the use of baseline PP instead of PP during follow‐up to estimate RR might be a possible basis of the analysis. Third, another consideration in interpreting the association between PP and mortality is the other BP components, particularly mean pressure. Most of the included studies did not adjust MAP or other BP parameters, which could have affected the results. Finally, only articles published in English were selected. Most of the studies were from European and American populations. Additional studies conducted in Asian and African populations are needed to generalize the findings.
Study Strengths
In spite of the above limitations, our study has the following strengths. A major strength of this meta‐analysis is based on a large number of prospective cohort studies with a long follow‐up duration (>5 years) and large sample sizes. On the whole, the quality of the studies in the meta‐analysis was good. These well‐designed prospective studies might have minimized the selection bias of the included publications than a retrospective study.
Conclusions
Our meta‐analysis provides evidence that wide brachial PP is associated with an increased risk of CV or all‐cause mortality. However, whether the increased potential risk is an independent risk factor requires additional well‐designed studies specifically for age and sex to further confirm our findings. Screening brachial PP may help to identify patients who are at increased risk for mortality events and therefore aid in early therapeutic decision‐making.
Disclosures
The authors have no conflicts of interest or funding sources to disclose.
J Clin Hypertens (Greenwich). 2014;16:678–685. © 2014 Wiley Periodicals, Inc.
References
- 1. Lopez AD, Mathers CD, Ezzati M, et al. Global and regional burden of disease and risk factors, 2001: systematic analysis of population health data. Lancet. 2006;367:1747–1757. [DOI] [PubMed] [Google Scholar]
- 2. Lawes CM, Vander Hoorn S, Law MR, et al. Blood pressure and the global burden of disease 2000. Part II: estimates of attributable burden. J Hypertens. 2006;24:423–430. [DOI] [PubMed] [Google Scholar]
- 3. Dart AM, Kingwell BA. Pulse pressure–a review of mechanisms and clinical relevance. J Am Coll Cardiol. 2001;37:975–984. [DOI] [PubMed] [Google Scholar]
- 4. Franklin SS. Pulse pressure as a risk factor. Clin Exp Hypertens. 2004;26:645–652. [DOI] [PubMed] [Google Scholar]
- 5. Antikainen RL, Jousilahti P, Vanhanen H, et al. Excess mortality associated with increased pulse pressure among middle‐aged men and women is explained by high systolic blood pressure. J Hypertens. 2000;18:417–423. [DOI] [PubMed] [Google Scholar]
- 6. Glynn RJ, Chae CU, Guralnik JM, et al. Pulse pressure and mortality in older people. Arch Intern Med. 2000;160:2765–2772. [DOI] [PubMed] [Google Scholar]
- 7. Fang J, Madhavan S, Alderman MH. Pulse pressure: a predictor of cardiovascular mortality among young normotensive subjects. Blood Press. 2000;9:260–266. [DOI] [PubMed] [Google Scholar]
- 8. Miura K, Dyer AR, Greenland P, et al. 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. [DOI] [PubMed] [Google Scholar]
- 9. Domanski M, Mitchell G, Pfeffer M, et al. Pulse pressure and cardiovascular disease‐related mortality: follow‐up study of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA. 2002;287:2677–2683. [DOI] [PubMed] [Google Scholar]
- 10. van Trijp TM, Grobbee DE, Peeters PH, et al. Average blood pressure and cardiovascular disease‐related mortality in middle‐aged women. Am J Hypertens. 2005;18:197–201. [DOI] [PubMed] [Google Scholar]
- 11. Garcia‐Palmieri MR, Crespo CJ, Mc Gee D, et al. Wide pulse pressure is an independent predictor of cardiovascular mortality in Puerto Rican men. Nutr Metab Cardiovasc Dis. 2005;15:71–78. [DOI] [PubMed] [Google Scholar]
- 12. Bowman TS, Sesso HD, Gaziano JM. Effect of age on blood pressure parameters and risk of cardiovascular death in men. Am J Hypertens. 2006;19:47–52. [DOI] [PubMed] [Google Scholar]
- 13. Lorenzo C, Aung K, Stern MP, et al. Pulse pressure, prehypertension, and mortality: the San Antonio heart study. Am J Hypertens. 2009;22:1219–1226. [DOI] [PubMed] [Google Scholar]
- 14. Hadaegh F, Shafiee G, Hatami M, et al. Systolic and diastolic blood pressure, mean arterial pressure and pulse pressure for prediction of cardiovascular events and mortality in a Middle Eastern population. Blood Press. 2012;21:12–18. [DOI] [PubMed] [Google Scholar]
- 15. Weiss A, Boaz M, Beloosesky Y, et al. Pulse pressure predicts mortality in elderly patients. J Gen Intern Med. 2009;24:893–896. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Panagiotakos DB, Kromhout D, Menotti A, et al. The relation between pulse pressure and cardiovascular mortality in 12,763 middle‐aged men from various parts of the world: a 25‐year follow‐up of the seven countries study. Arch Intern Med. 2005;165:2142–2147. [DOI] [PubMed] [Google Scholar]
- 17. Benetos A, Rudnichi A, Safar M, et al. Pulse pressure and cardiovascular mortality in normotensive and hypertensive subjects. Hypertension. 1998;32:560–564. [DOI] [PubMed] [Google Scholar]
- 18. Schram MT, Kostense PJ, Van Dijk RA, et al. Diabetes, pulse pressure and cardiovascular mortality: the Hoorn study. J Hypertens. 2002;20:1743–1751. [DOI] [PubMed] [Google Scholar]
- 19. Strandberg TE, Salomaa VV, Vanhanen HT, et al. Isolated diastolic hypertension, pulse pressure, and mean arterial pressure as predictors of mortality during a follow‐up of up to 32 years. J Hypertens. 2002;20:399–404. [DOI] [PubMed] [Google Scholar]
- 20. Safer ME. Pulse pressure in essential hypertension: clinical and therapeutical implications. J Hypertens. 1989;7:769–776. [DOI] [PubMed] [Google Scholar]
- 21. Mitchell GF. Pulse pressure, arterial compliance and cardiovascular morbidity and mortality. Curr Opin Nephrol Hypertens. 1999;8:335–342. [DOI] [PubMed] [Google Scholar]
- 22. James MA, Watt PA, Potter JF, et al. Pulse pressure and resistance artery structure in the elderly. Hypertension. 1995;26:301–306. [DOI] [PubMed] [Google Scholar]
- 23. Abramson JL, Vaccarino V. Pulse pressure and inflammatory process in atherosclerosis. Adv Cardiol. 2007;44:223–233. [DOI] [PubMed] [Google Scholar]
- 24. Stroup DF, Berlin JA, Morton SC, et al. Meta‐analysis of observational studies in epidemiology: a proposal for reporting. Meta‐analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283:2008–2012. [DOI] [PubMed] [Google Scholar]
- 25. Thompson S, Ekelund U, Jebb S, et al. A proposed method of bias adjustment for meta‐analyses of published observational studies. Int J Epidemiol. 2011;40:765–777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta‐analyses. BMJ. 2003;327:557–560. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. 1994;50:1088–1101. [PubMed] [Google Scholar]
- 28. Egger M, Davey Smith G, Schneider M, et al. Bias in meta‐analysis detected by a simple, graphical test. BMJ. 1997;315:629–634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Franklin SS, Gustin W, Wong ND, et al. Hemodynamic patterns of age‐related changes in blood pressure. The Framingham heart study. Circulation. 1997;96:308–315. [DOI] [PubMed] [Google Scholar]
- 30. Casiglia E, Tikhonoff V, Mazza A, et al. Pulse pressure and coronary mortality in elderly men and women from general population. J Hum Hypertens. 2002;16:611–620. [DOI] [PubMed] [Google Scholar]
- 31. Blacher J, Staessen JA, Girerd X, et al. Pulse pressure not mean pressure determines cardiovascular risk in older hypertensive patients. Arch Intern Med. 2000;160:1085–1089. [DOI] [PubMed] [Google Scholar]
- 32. Vlachopoulos C, Aznaouridis K, O'Rourke MF, et al. Prediction of cardiovascular events and all‐cause mortality with central haemodynamics: a systematic review and meta‐analysis. Eur Heart J. 2010;31:1865–1871. [DOI] [PubMed] [Google Scholar]
- 33. Benetos A, Safar M, Rudnichi A, et al. Pulse pressure: a predictor of long‐term cardiovascular mortality in a French male population. Hypertension. 1997;30:1410–1415. [DOI] [PubMed] [Google Scholar]
- 34. Franklin SS, Khan SA, Wong ND, et al. Is pulse pressure useful in predicting risk for coronary heart disease? The Framingham heart study. Circulation. 1999;100:354–360. [DOI] [PubMed] [Google Scholar]
- 35. Malone AF, Reddan DN. Pulse pressure. Why is it important? Perit Dial Int. 2010;30:265–268. [DOI] [PubMed] [Google Scholar]
- 36. Mosley WJ, Greenland P, Garside DB, Lloyd‐Jones DM. Predictive utility of pulse pressure and other blood pressure measures for cardiovascular outcomes. Hypertension. 2007;49:1256–1264. [DOI] [PubMed] [Google Scholar]