Skip main navigation

Association of Pulse Pressure With New-Onset Atrial Fibrillation in Patients With Hypertension and Left Ventricular Hypertrophy

The Losartan Intervention For Endpoint (LIFE) Reduction in Hypertension Study
Originally publishedhttps://doi.org/10.1161/HYPERTENSIONAHA.112.195032Hypertension. 2012;60:347–353

Abstract

Previous studies have found pulse pressure (PP), a marker of arterial stiffness, to be an independent predictor of atrial fibrillation (AF) in general and hypertensive populations. We examined whether PP predicted new-onset AF in comparison with other blood pressure components in the Losartan Intervention For Endpoint reduction in hypertension study, a double-blind, randomized (losartan versus atenolol), parallel-group study, including 9193 patients with hypertension and electrocardiographic left ventricular hypertrophy. In 8810 patients with neither a history of AF nor AF at baseline, Minnesota coding of electrocardiograms confirmed new-onset AF in 353 patients (4.0%) during mean 4.9 years of follow-up. In multivariate Cox regression analyses, baseline and in-treatment PP and baseline and in-treatment systolic blood pressure predicted new-onset AF, independent of baseline age, height, weight, and Framingham Risk Score; sex, race, and treatment allocation; and in-treatment heart rate and Cornell product. PP was the strongest single blood pressure predictor of new-onset AF determined by the decrease in the −2 Log likelihood statistic, in comparison with systolic blood pressure, diastolic blood pressure, and mean arterial pressure. When evaluated in the same model, the predictive effect of systolic and diastolic blood pressures together was similar to that of PP. In this population of patients with hypertension and left ventricular hypertrophy, PP was the strongest single blood pressure predictor of new-onset AF, independent of other risk factors.

Introduction

Atrial fibrillation (AF) is the most prevalent sustained cardiac arrhythmia, and the prevalence is increasing.1 In the Rotterdam study, the prevalence of AF varied from 0.7% in the age group 55 to 59 years to 17.8% in those aged ≥85 years.2 AF incidence increases with age,3 and other risk factors include diabetes, obesity, hypertension, left ventricular hypertrophy (LVH), coronary heart disease, congestive heart failure, valvular heart disease, and increased left atrial size by echocardiography.46 AF is associated with a 4- to 5-fold increased risk of ischemic stroke7,8 and with a nearly doubled cardiovascular mortality risk.9 Prevention of AF is thus of great importance, and hypertension is currently the most prevalent, potentially modifiable risk factor, accounting for ≈14% to 22% of AF cases.4,10,11

Increased pulse pressure (PP), defined as the difference between systolic blood pressure (SBP) and diastolic blood pressure (DBP), is a marker of arterial stiffness.12 Studies have found PP to be an independent predictor of new-onset AF in both general13 and hypertensive14 populations. Mitchell et al13 showed that increased baseline PP was the single blood pressure (BP) component most predictive of AF in 5331 participants (≈23% on antihypertensive treatment; ≈1.2% with electrocardiograpic LVH [ECG-LVH]) during ≈20 years of follow-up in the Framingham Heart Study and indicated that the relation between BP and incident AF is potentially related, specifically, to the pulsatile component of BP as assessed by PP. In a study by Ciaroni et al,14 increased PP (measured by 24-hour ambulatory BP measurement) during antihypertensive treatment was associated with an increased risk of new-onset AF, independent of age, sex, body mass index, and SBP in 597 patients with essential hypertension followed for ≈7 years. A pathophysiological explanation may be that arterial stiffness increases with age, resulting in increased PP and increased pulsatile load on the heart,15 promoting LVH,16 left ventricular diastolic dysfunction,17,18 and increased left atrial size,19 possibly leading to fibrosis and electric remodeling in the left atrium and, eventually, AF. In a study by Goette et al,20 patients with permanent AF had increased amount of atrial fibrosis; however, whether atrial fibrosis induces AF or is a consequence of AF is still unknown.

To our knowledge, the relation between baseline PP and PP during antihypertensive treatment and risk of new-onset AF has not yet been evaluated in high-risk patients with hypertension and ECG-LVH. Therefore, the goals of this prespecified Losartan Intervention For Endpoint (LIFE) reduction in hypertension substudy were to investigate the predictive value of higher baseline and in-treatment brachial PP for new-onset AF in patients with hypertension and LVH and to perform a thorough comparison of the predictive value of PP to that of other BP components such as SBP, DBP, and mean arterial pressure (MAP), using the Framingham study by Mitchell et al as a model.13

Methods

Study Design and Population

The LIFE study21,22 enrolled 9193 patients with essential hypertension (mean sitting brachial BP: 160 to 200 mm Hg systolic, 95 to 115 mm Hg diastolic, or both) and ECG-LVH (determined by Cornell voltage-duration product23,24 and/or Sokolow-Lyon voltage criteria,25) randomized to losartan- versus atenolol-based therapy. (For further details, please see http://hyper.ahajournals.org.) New-onset AF was a prespecified secondary end point. The present analyses included 8810 patients with neither a history of AF nor AF on their baseline ECG. New-onset AF was identified by Minnesota coding of annual in-study ECGs at the core laboratory at Sahlgrenska University Hospital/Östra, Göteborg, Sweden.21,26

Statistical Analyses

Statistical analyses were performed by the investigators using SPSS version 16.0 (SPSS Inc). Data are presented as mean±standard deviation (SD) for continuous variables and as proportions for categorical variables. Brachial PP was calculated as the difference between SBP and DBP. MAP was calculated as DBP plus one third of PP. Baseline characteristics in patients grouped according to quartiles of baseline PP were compared using analysis of variance (ANOVA) for continuous variables and Pearson χ2 statistics for categorical variables. Annual measurements of mean PP, SBP, DBP, and MAP were compared using general linear models to account for the within-subject correlation. The incidence of new-onset AF according to quartiles of baseline PP was illustrated in an unadjusted Kaplan-Meier curve.

In the primary analyses, possible associations between baseline PP or PP during antihypertensive therapy and the risk of developing new-onset AF were analyzed using Cox proportional hazards regression analyses.27,28 Baseline PP was entered as a continuous covariate, and in-treatment PP (baseline and subsequent routine measurements of PP during follow-up) was entered as a time-varying continuous covariate into univariate and multivariate Cox regression models. Additional covariates in the multivariate model were selected based on being significant univariate predictors that continued to be significant predictors in stepwise forward and backward multivariate analyses.

In the secondary analyses, we explored the relations between different BP components (PP, SBP, DBP, and MAP) as baseline and time-varying covariates and new-onset AF using the same multivariate Cox regression model and including a single BP component (PP, SBP, DBP, or MAP) or a combination of BP components (SBP and DBP, or PP and MAP). Hazard ratios (HR) for the incidence of new-onset AF associated with baseline and in-treatment PP, SBP, DBP, and MAP were computed per 1 SD of the baseline mean and per 10 mm Hg increments in BP.29,30 Wald χ2 statistics and P values were calculated. The decrease in the −2 Log likelihood statistic (a measure of model fit with data), caused by adding a single BP component (degrees of freedom [df]=1) or a combination of BP components (df equals the number of covariates added to the model) to the multivariate Cox regression model and χ2 tests, were used to evaluate and compare the relative importance and predictive effects of PP, SBP, DBP, and MAP. In addition, PP was also evaluated as a categorical variable with quartiles of baseline PP in multivariate analyses. Interaction analyses were performed using Cox regression models with 2 and 2 covariates and their cross-products.

Possible correlations between BP components were analyzed using Pearson correlation coefficient. A 2-tailed P<0.05 was required for statistical significance. All study data reside in a database with the authors.

Results

Patient Population and Blood Pressures

In 8810 patients (46% men) at risk of developing new-onset AF, mean baseline PP was 76.5±15.5 mm Hg (74.6±15.6 mm Hg for men and 78.0±15.3 mm Hg for women), with a range of 23.5 to 134.0 mm Hg. Mean age at randomization was 65.9±6.9 years for men and 67.5±7.0 years for women. Elevated PP ≥60 mm Hg at baseline was recorded in 7623 (86.5%) patients. Clinical characteristics according to quartiles of baseline PP (≤67.0 mm Hg, 67.5 to 77.0 mm Hg, 77.5 to 87.0 mm Hg, and ≥87.5 mm Hg) are presented in Tables 1 and online-only Data Supplement S1 (see http://hyper.ahajournals.org).

Table 1. Baseline Characteristics by Quartiles of Baseline PP (n=8810)*

CharacteristicsQ1 (≤67.0 mm Hg) n=2334Q2 (67.5–77.0 mm Hg) n=2189Q3 (77.5–87.0 mm Hg) n=2139Q4 (≥87.5 mm Hg) n=2148P Value
Male sex, n (%)1234 (53)1019 (47)928 (43)833 (39)
Age, y64±766±768±770±6
White race, n (%)2115 (91)2016 (92)1998 (93)2013 (94)
Weight, kg81±1579±1578±1676±14
Height, cm169±10168±9167±9166±9
BMI, kg/m228.2±4.728.0±4.828.2±5.127.7±4.6
History of diabetes, n (%)213 (9)235 (11)296 (14)359 (17)
History of CHD, n (%)268 (12)293 (13)276 (13)322 (15)
SBP, mm Hg159±10171±7179±9189±10
DBP, mm Hg102±699±797±993±10
PP, mm Hg57±872±382±396±7NA
MAP, mm Hg121±6123±7125±9125±9
Heart rate, bpm75±1174±1174±1173±11
Cornell product, mm×msec2738±9102828±10932849±10372851±1021
Total cholesterol, mmol/L5.99±1.136.06±1.106.09±1.126.09±1.13
UACR, mg/mmol5.7±30.75.6±20.77.3±29.19.8±35.5
FRS20±822±923±1025±10

*Values are mean±SD or numbers (n) and percentages.

P<0.01 (analysis of variance).

P<0.01 (Pearson χ2).

PP indicates pulse pressure; y, years; BMI, body mass index; CHD, coronary heart disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure; NA, not applicable; bpm, beats per minute; UACR, urine albumin-creatinine ratio; FRS, Framingham Risk Score.

Mean BP values at baseline and during follow-up are displayed in Figure 1. At baseline, mean SBP was 174.3±14.3 mm Hg, mean DBP was 97.9±8.8 mm Hg, and average MAP was 123.3±8.1 mm Hg. In patients followed for at least 4 years, 41.3% had a reduction in PP ≥15.5 mm Hg (1 SD of the baseline mean), 79.8% had a reduction in SBP ≥14.3 mm Hg (1 SD), 80.2% had a reduction in DBP ≥8.8 mm Hg (1 SD), and 87.8% had a reduction in MAP ≥8.1 mm Hg (1 SD).

Figure 1.

Figure 1. Mean blood pressure during 4.9 years of follow-up. The number of patients at each examination is noted in parentheses. P<0.001 for all blood pressure components (general linear models). BP indicates blood pressure; MAP, mean arterial pressure.

Baseline PP was strongly correlated with SBP (Pearson correlation coefficient [r]=0.83; P<0.001), moderately correlated with DBP (r=−0.41; P<0.001), and more weakly correlated with MAP (r=0.19; P<0.001). Baseline MAP was strongly correlated with SBP (r=0.71; P<0.001) and DBP (r=0.82; P<0.001). There was a relatively weak correlation between baseline SBP and DBP (r=0.17; P<0.001).

Multivariate Cox Regression Analyses

ECG confirmed new-onset AF in 353 (4.0%) of 8810 patients during a mean follow-up of 4.9±0.9 years. Figures 2 and 3 present the incidence of AF by quartiles of baseline PP.

Figure 2.

Figure 2. Incidence of atrial fibrillation according to quartiles of baseline pulse pressure. P<0.001 for the trend across quartiles (Pearson χ2).

Figure 3.

Figure 3. Kaplan-Meier curve of the unadjusted 5-year cumulative atrial fibrillation incidence across quartiles of baseline pulse pressure.

Results of the multivariate Cox regression model examining the predictive effect of baseline and in-treatment PP for new-onset AF are presented in Model 2 of Table 2 and in Table S2 (see http://hyper.ahajournals.org). Baseline PP was associated with a 39% (95% confidence interval [CI], 22% to 58%; P<0.001) increased risk of new-onset AF per 15.5 mm Hg (SD) increase, and in-treatment PP was associated with a 33% (95% CI, 18% to 50%; P<0.001) increased risk of new-onset AF per SD increase in a model adjusting for baseline age, height, weight, and Framingham Risk Score (FRS); sex, race, and a treatment group indicator (atenolol versus losartan), entered as continuous or categorical covariates; and in-treatment heart rate and ECG-LVH by Cornell product, entered as time-varying continuous covariates. Sex was a significant univariate predictor and was included in the multivariate Cox regression model for biological reasons, even though it was not significant in multivariate analyses. Smoking, diabetes, previous myocardial infarction, and body mass index did not predict new-onset AF; however, replacing height and weight with body mass index in the multivariate model did not alter the results. Baseline total cholesterol, potassium, and urine albumin-creatinine ratio were significant univariate predictors and were significant in the multivariate model; however, the model did not change when these covariates were excluded. Cox proportional hazards models for PP in comparison with other BP components are presented in Table 2. All 10 models were adjusted for baseline age, height, weight, and FRS; sex, race and treatment allocation; and in-treatment heart rate and ECG-LVH by Cornell product. When comparing single BP components in parallel multivariate models, adjusting for the same covariates, baseline and in-treatment PP (Models 1 and 2) and baseline and in-treatment SBP (Models 3 and 4), in addition to in-treatment MAP adjusted for baseline MAP (Model 8), were significant independent predictors of new-onset AF. Baseline and in-treatment DBP were not significant predictors (Models 5 and 6). The initial −2 Log likelihood was 5773.6 for the multivariate model, with baseline age, height, weight, and FRS; sex, race, and treatment allocation; and in-treatment heart rate and ECG-LVH by Cornell product. This model was used as a basis to evaluate decrease in −2 Log likelihood when introducing BP measures. Baseline and in-treatment PP (Model 2) were the strongest single component predictors (−2 Log likelihood 5739.6; χ2=34.0; 2 df, P<0.001); however, when entering baseline and in-treatment SBP and DBP into 1 model (Model 9), the model fit was equally good as for the baseline and in-treatment PP model: −2 Log likelihood 5739.4 (χ2=34.2; df=4; P<0.001) compared with 5739.6. The model with baseline and in-treatment SBP alone (Model 4) had a −2 Log likelihood of 5750.2, and adding baseline and in-treatment DBP to the model (Model 9) thus induced a significant improvement (χ2=10.8; df=2; P<0.01). In model 9, baseline and in-treatment SBP and DBP were all significant predictors of new-onset AF; however, the effects of SBP and DBP were opposite. Adding baseline and in-treatment MAP to the model with baseline and in-treatment PP did not change the model fit (−2 Log likelihood 5739.4 for Model 10 and 5739.6 for Model 2), and baseline and in-treatment MAP were not significant predictors in this model. When forcing baseline and in-treatment PP, SBP, and DBP into the same model, the HRs for DBP were not calculated owing to excess colinearity (r≈1.0) with PP and SBP. In the same model, baseline PP had a higher χ2 (Wald score) than baseline SBP (χ2 8.7 versus 0.01), and in-treatment PP was a stronger predictor than in-treatment SBP (χ2 7.2 versus 0.08).

Table 2. Cox Proportional Hazards Models for PP and other BP Components as Independent Predictors of New-Onset AF in Patients With Hypertension and ECG-LVH

Multivariate Model−2 Log Likelihood for ModelBP Components in Model*HR (95% CI) per 10 mm Hg IncreaseHR (95% CI) per 1 SD IncreaseP Value
Model 15762.0Baseline PP1.14 (1.06–1.23)1.23 (1.09–1.38)0.001
Model 25739.6Baseline PP1.24 (1.14–1.34)1.39 (1.22–1.58)<0.001
In-treatment PP1.20 (1.11–1.30)1.33 (1.18–1.50)<0.001
Model 35764.5Baseline SBP1.13 (1.04–1.23)1.20 (1.06–1.34)0.003
Model 45750.2Baseline SBP1.18 (1.08–1.28)1.27 (1.12–1.43)<0.001
In-treatment SBP1.13 (1.06–1.20)1.19 (1.09–1.31)<0.001
Model 55772.7Baseline DBP0.94 (0.84–1.07)0.95 (0.86–1.06)0.35
Model 65772.6Baseline DBP0.95 (0.84–1.08)0.96 (0.85–1.07)0.45
In-treatment DBP1.02 (0.90–1.15)1.02 (0.91–1.13)0.77
Model 75772.6Baseline MAP1.07 (0.94–1.22)1.06 (0.95–1.18)0.33
Model 85767.4Baseline MAP1.11 (0.97–1.27)1.09 (0.97–1.22)0.14
In-treatment MAP1.13 (1.02–1.25)1.10 (1.01–1.20)0.02
Model 95739.4Baseline SBP1.24 (1.14–1.36)1.36 (1.20–1.55)<0.001
In-treatment SBP1.20 (1.11–1.30)1.30 (1.17–1.46)<0.001
Baseline DBP0.81 (0.71–0.93)0.83 (0.74–0.94)0.003
In-treatment DBP0.82 (0.70–0.95)0.84 (0.73–0.95)0.007
Model 105739.4Baseline PP1.24 (1.14–1.35)1.39 (1.22–1.59)<0.001
In-treatment PP1.21 (1.11–1.33)1.35 (1.17–1.55)<0.001
Baseline MAP1.01 (0.88–1.16)1.01 (0.90–1.13)0.93
In-treatment MAP0.98 (0.87–1.11)0.99 (0.89–1.09)0.77

*All models are adjusted for baseline age, height, weight, and Framingham Risk Score; sex, race, and treatment allocation; and in-treatment heart rate and ECG-LVH by Cornell product. One SD of the baseline mean was 15.5 mm Hg for PP, 14.3 mm Hg for SBP, 8.8 mm Hg for DBP, and 8.1 mm Hg for MAP.

PP indicates pulse pressure; BP, blood pressure; AF, atrial fibrillation; ECG-LVH, electrocardiographic left ventricular hypertrophy; HR, hazard ratio; CI, confidence interval; SD, standard deviation; SBP, systolic blood pressure; DBP, diastolic blood pressure; MAP, mean arterial pressure.

PP was also computed as a categorical variable, with quartiles of baseline PP (quartile 4 versus quartiles 1 to 3). When adjusted for baseline age, height, weight, and FRS; sex, race, and treatment allocation; and in-treatment heart rate and ECG-LVH by Cornell product, baseline PP quartile 4 (≥87.5 mm Hg) was associated with a 67% (95% CI, 32% to 211%; P<0.001) higher risk of new-onset AF compared with quartiles 1 to 3. This result was strengthened when we also adjusted for in-treatment PP in the same model (HR, 1.98; 95% CI, 1.55 to 2.52; P<0.001).

There were no significant interactions between baseline or in-treatment PP and other BP components or between baseline or in-treatment PP and baseline age, height, weight, and FRS; sex, race, and treatment allocation; and in-treatment heart rate and ECG-LVH by Cornell product. There were significant interactions between in-treatment heart rate and weight (P=0.03) and in-treatment heart rate and race (P=0.003) in all 10 models (Table 2). In model 9, there were significant interactions between weight and in-treatment SBP (P=0.03) and weight and in-treatment DBP (P=0.01). In model 10, there were significant interactions between age and in-treatment MAP (P=0.02) and weight and in-treatment MAP (P=0.004).

Discussion

In the present study, increased baseline PP and PP during antihypertensive treatment were associated with an increased risk of incident AF, independent of other predictors of AF in this population (ie, baseline age, height, weight, and FRS; sex, race, and treatment allocation; and in-treatment heart rate and ECG-LVH by Cornell product). Baseline PP quartile 4 (≥87.5 mm Hg) was associated with a highly significant increase in risk of developing AF during mean 4.9 years of follow-up compared with quartiles 1 to 3.

In comparison with SBP, DBP, and MAP as single BP components, PP was the strongest predictor of incident AF. When we considered the predictive effect of SBP and DBP together, model fit improved significantly and had the same −2 Log likelihood as the PP model. This is a consequence of the mathematical calculation of PP as the difference between SBP and DBP. When evaluated in the same model, the effects of SBP and DBP were significant but opposite, suggesting that, for a certain value of SBP, lower DBP was associated with an increased risk of new-onset AF. When evaluating PP, SBP, and DBP in the same model, both baseline and in-treatment PP had higher χ2 (Wald score) than SBP. This supports the finding that PP is the strongest single BP measure for predicting incident AF in our study; however, it should be interpreted with caution, considering the high correlations between the BP components in this specific model.

In-treatment MAP was associated with incident AF when adjusted for baseline MAP and the above-mentioned AF risk factors. Entering MAP into the same model as PP did not improve model fit; baseline and in-treatment MAP were not significant, and the HRs of baseline and in-treatment PP were unaltered. Thus, PP predicted incident AF independent of MAP.

AF is associated with increased risk of cardiovascular morbidity and mortality. It is highly important to identify modifiable risk factors, as both men and women have an approximate 25% overall lifetime risk of AF.31 To our knowledge, this is the first study to report a strong, independent association between brachial PP and new-onset AF in patients with hypertension and ECG-LVH. Our results are in agreement with a Framingham Heart Study investigation evaluating PP as a predictor for incident AF in a general population with normal or moderately increased BP.13 Furthermore, Mitchell et al demonstrated that there is a potential weakness of concentrating on SBP alone and ignoring DBP and PP, and our data support this finding. When evaluating the risk of incident AF in a hypertensive population with ECG-LVH, PP should be considered or, alternatively, SBP and DBP together. PP is simple to calculate as the absolute difference between SBP and DBP.

Increased PP, a marker of advanced vascular aging32 and arterial stiffness,12,33 may contribute in the structural and electric remodeling of the myocardium, leading to the development of AF, possibly through increased pulsatile load on the heart and increased left atrial size.19 Studies have shown that reduced distensibility of large arteries parallel cardiac hypertrophy and remodeling in patients with hypertension.34,35 Large artery stiffness may increase the workload on the heart similar to volume overload and, perhaps, represent one of the mechanisms by which hypertension leads to eccentric hypertrophy and left atrial enlargement.35 In a LIFE substudy, there was a significant correlation between baseline brachial PP and left atrial size, independent of age, sex, and body surface area (data not shown).36 Furthermore, there is much evidence for linking brachial PP to microvascular damage in the heart and other target organs, which, again, may lead to increased peripheral resistance and MAP, further increasing arterial stiffness and central PP. Increased central PP may then further damage small arteries and lead to LVH.37 Studies have found brachial PP to be a powerful predictor of cardiovascular morbidity and mortality,3845 and the predictive effect increases with age.4244 The present study evaluated brachial PP and not central PP. Noninvasive central PP has been shown to better predict cardiovascular outcomes than brachial PP and to be closer associated with extent of atherosclerosis (carotid plaque burden and intimal-medial thickness, and vascular mass).46

In conclusion, in patients with hypertension and ECG-LVH in the LIFE study, increased baseline and in-treatment PP were independently associated with increased risk of new-onset AF. PP was (in comparison with SBP, DBP, and MAP) the single BP component with the strongest predictive effect.

Limitations

Patients evaluated in the LIFE study were predominantly white and from Western countries. They had hypertension and ECG-LVH and increased risk of cardiovascular events compared with hypertensive subjects without LVH. The results may not be generalizable to normotensives and hypertensives without LVH. BP was measured with a sphygmomanometer, which is considered less accurate than 24-hour ambulatory BP measurement.43 New-onset AF was a prespecified secondary end point; however, the LIFE study was designed and had statistical power for the primary composite end point, and the HRs for AF require careful interpretation.

Perspectives

In patients with hypertension and ECG-LVH in the LIFE study, increased baseline and in-treatment PP were independently associated with new-onset AF. PP was (in comparison with SBP, DBP and MAP) the single BP component with the strongest predictive effect, supporting the hypothesis that the relation between BP and incident AF is related specifically to the pulsatile component of BP as assessed by PP.13 Furthermore, SBP and DBP together had a predictive effect similar to the predictive effect of PP, reflecting the definition of PP. In-treatment MAP was significantly associated with new-onset AF when adjusted for baseline MAP and the mentioned risk factors; however, the predictive effect was weaker than for PP or for SBP and DBP evaluated together. This result may imply that the association between MAP (the steady component of BP) and AF is weak. When evaluating risk of AF in patients with hypertension and ECG-LVH, both baseline PP and PP during antihypertensive treatment, alternatively SBP and DBP together, should be considered. Furthermore, lowering of PP may prevent new-onset AF in patients with hypertension and LVH; however, this must be further explored in randomized clinical trials.

Sources of Funding

The LIFE study was originally sponsored by Merck & Co, Inc, Whitehouse Station, NJ. This substudy was partially funded by a grant from South-Eastern Norway Regional Health Authority.

Disclosures

Drs Gjesdal, Olsen, Devereux, Kjeldsen, and Wachtell were investigators and Drs Devereux, Ibsen, Kjeldsen, and Dahlöf were steering committee members for the LIFE Study. Drs Dahlöf, Devereux, and Wachtell have received grant support from Merck & Co, Inc, the sponsor for the LIFE Study. Drs Gjesdal, Olsen, Ibsen, Devereux, Okin, Dahlöf, Kjeldsen, and Wachtell have received occasional speaker honoraria from Merck & Co, Inc.

Footnotes

Clinical Trials Registration Information: URL: http://www.clinicaltrials.gov/ct2/show/NCT00338260 (Identifier NCT00338260).

The online-only Data Supplement is available with this article at http://hyper.ahajournals.org/lookup/suppl/doi:10.1161/HYPERTENSIONAHA.112.195032/-/DC1.

Correspondence to Anne Cecilie K. Larstorp, MD,
Department of Cardiology, Oslo University Hospital Ullevål, Postboks 4956 Nydalen, N-0424 Oslo, Norway
. E-mail

References

  • 1. Roger VL, Go AS, Lloyd-Jones DM, Benjamin EJ, Berry JD, Borden WB, Bravata DM, Dai S, Ford ES, Fox CS, Fullerton HJ, Gillespie C, Hailpern SM, Heit JA, Howard VJ, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Makuc DM, Marcus GM, Marelli A, Matchar DB, Moy CS, Mozaffarian D, Mussolino ME, Nichol G, Paynter NP, Soliman EZ, Sorlie PD, Sotoodehnia N, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart disease and stroke statistics–2012 update: a report from the American Heart Association. Circulation. 2012; 125:e2–e220.LinkGoogle Scholar
  • 2. Heeringa J, van der Kuip DA, Hofman A, Kors JA, van Herpen G, Stricker BH, Stijnen T, Lip GY, Witteman JC. Prevalence, incidence and lifetime risk of atrial fibrillation: the Rotterdam study. Eur Heart J. 2006; 27:949–953.CrossrefMedlineGoogle Scholar
  • 3. Khairallah F, Ezzedine R, Ganz LI, London B, Saba S. Epidemiology and determinants of outcome of admissions for atrial fibrillation in the United States from 1996 to 2001. Am J Cardiol. 2004; 94:500–504.CrossrefMedlineGoogle Scholar
  • 4. Benjamin EJ, Levy D, Vaziri SM, D'Agostino RB, Belanger AJ, Wolf PA. Independent risk factors for atrial fibrillation in a population-based cohort. The Framingham Heart Study. JAMA. 1994; 271:840–844.CrossrefMedlineGoogle Scholar
  • 5. Psaty BM, Manolio TA, Kuller LH, Kronmal RA, Cushman M, Fried LP, White R, Furberg CD, Rautaharju PM. Incidence of and risk factors for atrial fibrillation in older adults. Circulation. 1997; 96:2455–2461.LinkGoogle Scholar
  • 6. Wang TJ, Parise H, Levy D, D'Agostino RB, Wolf PA, Vasan RS, Benjamin EJ. Obesity and the risk of new-onset atrial fibrillation. JAMA. 2004; 292:2471–2477.CrossrefMedlineGoogle Scholar
  • 7. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991; 22:983–988.LinkGoogle Scholar
  • 8. Go AS. The epidemiology of atrial fibrillation in elderly persons: the tip of the iceberg. Am J Geriatr Cardiol. 2005; 14:56–61.CrossrefMedlineGoogle Scholar
  • 9. Benjamin EJ, Wolf PA, D'Agostino RB, Silbershatz H, Kannel WB, Levy D. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998; 98:946–952.LinkGoogle Scholar
  • 10. Aksnes TA, Flaa A, Strand A, Kjeldsen SE. Prevention of new-onset atrial fibrillation and its predictors with angiotensin II-receptor blockers in the treatment of hypertension and heart failure. J Hypertens. 2007; 25:15–23.CrossrefMedlineGoogle Scholar
  • 11. Huxley RR, Lopez FL, Folsom AR, Agarwal SK, Loehr LR, Soliman EZ, Maclehose R, Konety S, Alonso A. Absolute and attributable risks of atrial fibrillation in relation to optimal and borderline risk factors. Circulation. 2011; 123:1501–1508.LinkGoogle Scholar
  • 12. Darne B, Girerd X, Safar M, Cambien F, Guize L. Pulsatile versus steady component of blood pressure: a cross-sectional analysis and a prospective analysis on cardiovascular mortality. Hypertension. 1989; 13:392–400.LinkGoogle Scholar
  • 13. Mitchell GF, Vasan RS, Keyes MJ, Parise H, Wang TJ, Larson MG, D'Agostino RB, Kannel WB, Levy D, Benjamin EJ. Pulse pressure and risk of new-onset atrial fibrillation. JAMA. 2007; 297:709–715.CrossrefMedlineGoogle Scholar
  • 14. Ciaroni S, Bloch A, Lemaire MC, Fournet D, Bettoni M. Prognostic value of 24-hour ambulatory blood pressure measurement for the onset of atrial fibrillation in treated patients with essential hypertension. Am J Cardiol. 2004; 94:1566–1569.CrossrefMedlineGoogle Scholar
  • 15. Mitchell GF, Parise H, Benjamin EJ, Larson MG, Keyes MJ, Vita JA, Vasan RS, Levy D. Changes in arterial stiffness and wave reflection with advancing age in healthy men and women: the Framingham Heart Study. Hypertension. 2004; 43:1239–1245.LinkGoogle Scholar
  • 16. Gardin JM, Arnold A, Gottdiener JS, Wong ND, Fried LP, Klopfenstein HS, O'Leary DH, Tracy R, Kronmal R. Left ventricular mass in the elderly: the Cardiovascular Health Study. Hypertension. 1997; 29:1095–1103.LinkGoogle Scholar
  • 17. Leite-Moreira AF, Correia-Pinto J, Gillebert TC. Afterload induced changes in myocardial relaxation: a mechanism for diastolic dysfunction. Cardiovasc Res. 1999; 43:344–353.CrossrefMedlineGoogle Scholar
  • 18. Tsang TS, Gersh BJ, Appleton CP, Tajik AJ, Barnes ME, Bailey KR, Oh JK, Leibson C, Montgomery SC, Seward JB. Left ventricular diastolic dysfunction as a predictor of the first diagnosed nonvalvular atrial fibrillation in 840 elderly men and women. J Am Coll Cardiol. 2002; 40:1636–1644.CrossrefMedlineGoogle Scholar
  • 19. Vaziri SM, Larson MG, Lauer MS, Benjamin EJ, Levy D. Influence of blood pressure on left atrial size. The Framingham Heart Study. Hypertension. 1995; 25:1155–1160.LinkGoogle Scholar
  • 20. Goette A, Staack T, Rocken C, Arndt M, Geller JC, Huth C, Ansorge S, Klein HU, Lendeckel U. Increased expression of extracellular signal-regulated kinase and angiotensin-converting enzyme in human atria during atrial fibrillation. J Am Coll Cardiol. 2000; 35:1669–1677.CrossrefMedlineGoogle Scholar
  • 21. Dahlöf B, Devereux R, de Faire U, Fyhrquist F, Hedner T, Ibsen H, Julius S, Kjeldsen S, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H. The Losartan Intervention For Endpoint reduction (LIFE) in Hypertension study: rationale, design, and methods. The LIFE Study Group. Am J Hypertens. 1997; 10:705–713.CrossrefMedlineGoogle Scholar
  • 22. Dahlöf B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristiansson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H; LIFE Study Group. Cardiovascular morbidity and mortality in the Losartan Intervention For Endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002; 359:995–1003.CrossrefMedlineGoogle Scholar
  • 23. Molloy TJ, Okin PM, Devereux RB, Kligfield P. Electrocardiographic detection of left ventricular hypertrophy by the simple QRS voltage-duration product. J Am Coll Cardiol. 1992; 20:1180–1186.CrossrefMedlineGoogle Scholar
  • 24. Okin PM, Roman MJ, Devereux RB, Kligfield P. Electrocardiographic identification of increased left ventricular mass by simple voltage-duration products. J Am Coll Cardiol. 1995; 25:417–423.CrossrefMedlineGoogle Scholar
  • 25. Sokolow M, Lyon TP. The ventricular complex in left ventricular hypertrophy as obtained by unipolar precordial and limb leads. Am Heart J. 1949; 37:161–186.CrossrefMedlineGoogle Scholar
  • 26. Wachtell K, Lehto M, Gerdts E, Olsen MH, Hornestam B, Dahlöf B, Ibsen H, Julius S, Kjeldsen SE, Lindholm LH, Nieminen MS, Devereux RB. Angiotensin II receptor blockade reduces new-onset atrial fibrillation and subsequent stroke compared to atenolol: the Losartan Intervention For End Point Reduction in Hypertension (LIFE) study. J Am Coll Cardiol. 2005; 45:712–719.CrossrefMedlineGoogle Scholar
  • 27. Okin PM, Devereux RB, Jern S, Kjeldsen SE, Julius S, Nieminen MS, Snapinn S, Harris KE, Aurup P, Edelman JM, Wedel H, Lindholm LH, Dahlöf B; LIFE Study Investigators. Regression of electrocardiographic left ventricular hypertrophy during antihypertensive treatment and the prediction of major cardiovascular events. JAMA. 2004; 292:2343–2349.CrossrefMedlineGoogle Scholar
  • 28. Cox DR. Regression models and life-tables. J R Stat Soc, B. 1972; 34:187–220.Google Scholar
  • 29. Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York, NY: John Wiley & Sons; 1980.Google Scholar
  • 30. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958; 53:457–481.CrossrefGoogle Scholar
  • 31. Lloyd-Jones DM, Wang TJ, Leip EP, Larson MG, Levy D, Vasan RS, D'Agostino RB, Massaro JM, Beiser A, Wolf PA, Benjamin EJ. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Circulation. 2004; 110:1042–1046.LinkGoogle Scholar
  • 32. Aviv A. Hypothesis: pulse pressure and human longevity. Hypertension. 2001; 37:1060–1066.LinkGoogle Scholar
  • 33. Franklin SS, Sutton-Tyrrell K, Belle SH, Weber MA, Kuller LH. The importance of pulsatile components of hypertension in predicting carotid stenosis in older adults. J Hypertens. 1997; 15:1143–1150.CrossrefMedlineGoogle Scholar
  • 34. Bouthier JD, De Luca N, Safar ME, Simon AC. Cardiac hypertrophy and arterial distensibility in essential hypertension. Am Heart J. 1985; 109:1345–1352.CrossrefMedlineGoogle Scholar
  • 35. Boutouyrie P, Laurent S, Girerd X, Benetos A, Lacolley P, Abergel E, Safar M. Common carotid artery stiffness and patterns of left ventricular hypertrophy in hypertensive patients. Hypertension. 1995; 25:651–659.LinkGoogle Scholar
  • 36. Gerdts E, Papademetriou V, Palmieri V, Boman K, Bjornstad H, Wachtell K, Giles TD, Dahlöf B, Devereux RB. Correlates of pulse pressure reduction during antihypertensive treatment (losartan or atenolol) in hypertensive patients with electrocardiographic left ventricular hypertrophy (the LIFE study). Am J Cardiol. 2002; 89:399–402.CrossrefMedlineGoogle Scholar
  • 37. Laurent S, Briet M, Boutouyrie P. Large and small artery cross-talk and recent morbidity-mortality trials in hypertension. Hypertension. 2009; 54:388–392.LinkGoogle Scholar
  • 38. Benetos A, Rudnichi A, Safar M, Guize L. Pulse pressure and cardiovascular mortality in normotensive and hypertensive subjects. Hypertension. 1998; 32:560–564.LinkGoogle Scholar
  • 39. Domanski MJ, Davis BR, Pfeffer MA, Kastantin M, Mitchell GF. Isolated systolic hypertension: prognostic information provided by pulse pressure. Hypertension. 1999; 34:375–380.LinkGoogle Scholar
  • 40. Millar JA, Lever AF. Implications of pulse pressure as a predictor of cardiac risk in patients with hypertension. Hypertension. 2000; 36:907–911.LinkGoogle Scholar
  • 41. 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.LinkGoogle Scholar
  • 42. Vaccarino V, Berger AK, Abramson J, Black HR, Setaro JF, Davey JA, Krumholz HM. Pulse pressure and risk of cardiovascular events in the systolic hypertension in the elderly program. Am J Cardiol. 2001; 88:980–986.CrossrefMedlineGoogle Scholar
  • 43. Glynn RJ, Chae CU, Guralnik JM, Taylor JO, Hennekens CH. Pulse pressure and mortality in older people. Arch Intern Med. 2000; 160:2765–2772.CrossrefMedlineGoogle Scholar
  • 44. Khattar RS, Swales JD, Dore C, Senior R, Lahiri A. Effect of aging on the prognostic significance of ambulatory systolic, diastolic, and pulse pressure in essential hypertension. Circulation. 2001; 104:783–789.LinkGoogle Scholar
  • 45. Fyhrquist F, Dahlöf B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, de Faire U, Ibsen H, Kristianson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Hille DA, Lyle PA, Edelman JM, Snapinn SM, Wedel H; LIFE Study Group. Pulse pressure and effects of losartan or atenolol in patients with hypertension and left ventricular hypertrophy. Hypertension. 2005; 45:580–585.LinkGoogle Scholar
  • 46. Roman MJ, Devereux RB, Kizer JR, Lee ET, Galloway JM, Ali T, Umans JG, Howard BV. Central pressure more strongly relates to vascular disease and outcome than does brachial pressure: the Strong Heart Study. Hypertension. 2007; 50:197–203.LinkGoogle Scholar

Novelty and Significance

What Is New?

  • To our knowledge, this is the first study to report a strong, independent association between baseline pulse pressure and pulse pressure during antihypertensive treatment and new-onset atrial fibrillation in patients with hypertension and left ventricular hypertrophy.

What Is Relevant?

  • In 8810 patients in this randomized (losartan versus atenolol) treatment trial, pulse pressure (the pulsatile component of blood pressure and a marker of arterial stiffness) was the strongest single blood pressure predictor for atrial fibrillation compared with systolic blood pressure, diastolic blood pressure, and mean arterial pressure.

Summary

When evaluating risk of atrial fibrillation in patients with hypertension and left ventricular hypertrophy, both baseline pulse pressure and pulse pressure during antihypertensive treatment should be considered.

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.