Nighttime ambulatory pulse pressure predicts cardiovascular and all‐cause mortality among middle‐aged participants in the 21‐year follow‐up

Abstract Office pulse pressure (PP) is a predictor for cardiovascular (CV) events and mortality. Our aim was to evaluate ambulatory PP as a long‐term risk factor in a random cohort of middle‐aged participants. The Opera study took place in years 1991–1993, with a 24‐h ambulatory blood pressure measurement (ABPM) performed to 900 participants. The end‐points were non‐fatal and fatal CV events, and deaths of all‐causes. Follow‐up period, until the first event or until the end of the year 2014, was 21.1 years (mean). Of 900 participants, 22.6% died (29.6% of men/15.6% of women, p<.001). A CV event was experienced by 208 participants (23.1%), 68.3% of them were male (p<.001). High nighttime ambulatory PP predicted independently CV mortality (hazard ratio [HR] 2.60; 95% confidence interval [CI 95%] 1.08–6.31, p=.034) and all‐cause mortality in the whole population (HR 1.72; Cl 95% 1.06–2.78, p=.028). In males, both 24‐h PP and nighttime PP associated with CV mortality and all‐cause mortality (24‐h PP HR for CV mortality 2.98; CI 95% 1.11–8.04, p=.031 and all‐cause mortality HR 2.40; CI 95% 1.32–4.37, p=.004). Accordingly, nighttime PP; HR for CV mortality 3.13; CI 95% 1.14–8.56, p=.026, and for all‐cause mortality HR 2.26; CI 95% 1.29–3.96, p=.004. Cox regression analyses were adjusted by sex, CV risk factors, and appropriate ambulatory mean systolic BP. In our study, high ambulatory nighttime PP was detected as a long‐term risk factor for CV and all‐cause mortality in middle‐aged individuals.


INTRODUCTION
Pulse pressure (PP) is defined as the difference between systolic blood pressure (SBP) and diastolic blood pressure (DBP) and it is considered as a surrogate marker for arterial stiffening (viite). PP increases with ageing, as SBP increases, and DBP decreases. 1,2 High office PP is a known independent risk factor for cardiovascular (CV) events and CV mortality, 2-7 especially in individuals over 60 years of age, 2,7 but also in the middle-aged participants. [4][5][6]8 Increased office PP is also associated with all-cause mortality in a middle-aged male population, 4 and in young adults. 9 A rising trend in PP over a period of time was discovered to increase the risk for all-cause mortality in a large population study with a wide age group. 10 In the very elderly however, PP may not be a CV risk factor. 11 There is evidence that ambulatory PP could be an even better method in predicting the risk of CV events and mortality than office PP, 8,12 or ambulatory systolic blood pressure (BP). 8,13 Increased ambulatory PP has been found as a CV risk factor in different specific patient groups: it predicts CV events in patients with peripheral artery disease, 14 CV mortality in hemodialysis patients, 15 and also a variety of adverse CV outcomes and all-cause mortality in participants with hypertension. 16 In a population study, daytime and 24-h PP had a predictive value in the general population, whereas the nighttime PP was a risk factor for men. 17 In a large study consisting of 11 different populations, all individuals within the highest 24-h PP distribution had an elevated risk for CV events, and those over 60 years had the higher allcause mortality than those below. 18 In some studies, however, there has been controversy over significance of ambulatory PP as a predictor for CV and all-cause mortality. 19 In the present study, our aim was to investigate ambulatory PP (24-h, daytime, and nighttime) as a predictor for long-term cardiovascular events and all-cause mortality in a middle-aged population as a whole, and also separately in both sexes, because sex differences have been reported earlier. 17,20 To our knowledge, population based studies targeting ambulatory PP as a predictive factor for CV events or mortality with such a long follow-up period as ours do not exist. and 1993, and ABPM was recorded in 903 participants. In our analyses we included those, whose ABPM was available, a total of 900 participants. Three participants were excluded because of missing nighttime values of ABPM ( Figure 1). Further details of the baseline study can be found elsewhere. 22 Our study was approved by the Ethics Committee of the Faculty of Medicine, University of Oulu, and was conducted by the principles of the Declaration of Helsinki. All study participants gave an informed consent.

Blood pressure measurements
The office BP was measured by a specially trained registered nurse, automatically excluded from the analysis if systolic BP (SBP) was less than 70 or more than 250 mmHg, diastolic BP (DBP) less than 40 or more than 150 mmHg, and heart rate less than 40 or more than 150 beats per minute. Less than 3% of the BP readings were rejected as artifacts based on these criteria. 26 Three recordings were excluded because of missing nighttime variables. Pulse pressure was calculated as a difference between systolic BP and diastolic BP.

RESULTS
Originally 1045 participants attended the OPERA study. All those (n=900, 50.4% women, and 49.6% men) who underwent a qualified 24h ABPM recording in the baseline study were included in the analyses.
Only three recordings were disqualified on the basis of missing nighttime readings.
Regarding CV events, those who experienced a CV event in comparison with those who were event-free, the differences in above mentioned variables were quite similar (Table 1). However, total cholesterol and LDL levels were higher in CV event group, in which also antihypertensive medication was more frequent. There were differences in incidence of CV mortality, CV events, and total mortality between the PP tertiles, and the linear associations were significant. All systolic and diastolic BP measurements, office, daytime and nighttime ambulatory, and pulse pressure, differed between the PP tertile groups. Prevalence of hypertension and diabetes increased by tertiles, but there were no statistically significant differences in prevalence of CAD nor previous stroke. In laboratory measurements, there were significant differences in fasting glucose, insulin, and triglycerides levels, and in estimated GFR between the tertiles. Prevalence of antihypertensive medication increased towards the highest PP tertile. There were statistically significant differences in thiazide and calcium blocker use, but not in other agents.

Pulse pressure and cardiovascular mortality
Pulse pressure as a predictor for CV mortality was analyzed with Cox regression by 24-h, daytime, and nighttime ambulatory PP tertiles and by sex, as shown in Table 3

Pulse pressure and cardiovascular events
We assessed the association of 24-h, daytime, and nighttime ambula-   In the studies by Khattar and Staessen 12,28 the association between high 24-h PP and total mortality was also seen in the whole study group, whereas in the current OPERA study only among men. Arterial stiffness is a complex entity, which is especially affected by aging and sex hormones. 27,30 Decreasing levels of estrogen after menopause is likely to enhance arterial stiffening. In a large community-based population, disappearance of the PP amplification between carotid and brachial arteries was a stronger predictor for all-cause and CV mortality in women than in men, and more prominently in postmenopausal women. 31 The majority of our female participants were presumably premenopausal or early menopausal in the beginning of the study.

Pulse pressure and all-cause mortality
Therefore, it may be assumed that the arterial stiffness in women participants was lower than in younger women but still higher in average compared with the male study population of the same age. This may partly explain the gender difference, as none of the components of ambulatory PP were associated in CV events or mortality in women in the present study. The most probable explanation for this may be the fewer events and deaths among women compared with men, and the younger age range than above-mentioned studies. 12,28 The reason why nighttime PP is an independent predictor for total mortality remains unknown. Several morbidities, such as hypertension and diabetes, may potentially increase PP and mortality. In our study, both hypertension and diabetes were associated with increased PP but controlling these diseases in the multivariate analysis did not change the results. In the present study the day-and nighttimes were fixed to certain hours and no data about participants sleeping habits were acquired. Another major concern is the lacking data on sleep apnoea, as sleep apnoea may affect the nighttime BP values. 32 The practise of blood pressure treatment has developed towards combination medication since the beginning of our study, and the prevalence of reninangiotensin-acting agents has increased substantially over the years.
One limitation is also that study participants were caucasian only, therefore our results may not be applicable to other populations. However, the initial data with ABPM and a follow-up of 21-years is almost unique in the scientific world.
In conclusion, nighttime ambulatory pulse pressure showed a significant and independent association with CV and all-cause mortality in a random cohort of middle-aged normotensive and hypertensive participants in the long-term follow-up. Also 24-h PP was detected as a prognostic factor in male participants. Wide PP, especially nighttime PP, may identify individuals with an increased risk for mortality, and therefore ambulatory BPM should be in use, together with office BP measurements.