Article Text
PDF
Abstract
Objective To investigate the risk of major adverse cardiac and cerebrovascular events (MACCEs) and all-cause death of patients with controlled hypertension and suggest the benefits of physical activity in their prognosis.
Methods People aged 40–69 years from the prospective UK Biobank cohort (UKB, n=220 026) and the retrospective Korean National Health Insurance Service cohort (KNHIS, n=3 593 202) were included in this observational cohort study, excluding those with previous cerebrocardiovascular diseases or hypertension without treatment. The study groups were stratified into normotension, controlled hypertension (patients with hypertension with systolic blood pressure <140 mm Hg and diastolic blood pressure <90 mm Hg) and uncontrolled hypertension groups. The outcomes were MACCEs and all-cause mortality, analysed by Cox regression analysis.
Results We included 161 405/18 844/39 777 and 3 122 890/383 828/86 484 individuals with normotension/controlled hypertension/uncontrolled hypertension state from the UKB and KNHIS cohorts, respectively. The controlled hypertension group showed significantly higher risk of MACCEs (UKB: adjusted HR 1.73 (95% CI 1.55 to 1.92); KNHIS: 1.46 (95% CI 1.43 to 1.49)) and all-cause mortality (UKB: adjusted HR 1.28 (95% CI 1.18 to 1.39); KNHIS: 1.29 (95% CI 1.26 to 1.32)) than individuals with normotension. The controlled hypertension group not involved in any moderate or moderate-to-vigorous physical activity showed high risk of adverse outcomes, which was comparable with or even higher than the risk of patients with uncontrolled hypertension who were engaged in physical activity.
Conclusions Controlled hypertension is associated with residual risks of adverse outcomes. Clinicians may encourage physical activity for patients with controlled hypertension, not being reassured by their achieved target blood pressure values.
- hypertension
- coronary artery disease
- epidemiology
Data availability statement
The data that support the findings of this study are available from the UK Biobank (application no. 53799) and the NHIS of the Republic of Korea (NHIS-2018-1-247). The study data will be made available by organisations upon reasonable request.
Statistics from Altmetric.com
Introduction
Hypertension is one of the most common chronic comorbidities in current medicine and is associated with critically increased risk of major adverse cardiac and cerebrovascular events (MACCEs) and death.1 2 Current guidelines suggest robust target blood pressure (BP) values that can be achieved through lifestyle modifications and use of antihypertensive medications.3–5
Although there is substantial evidence on the benefits of controlling hypertension, whether a ‘controlled hypertension state’, a normal BP value in patients with hypertension, is a benign condition has rarely been investigated. Patients with hypertension exhibiting few normal BP values may still have a higher risk of adverse outcomes than individuals with normotension.6 Previous studies suggested that treated hypertension should not be misconceived as a low-risk state, demonstrating that history of hypertension medication use itself is a significant risk factor for adverse outcomes.7 8 Furthermore, if a controlled hypertension state is associated with residual risks of adverse outcomes, an investigation to seek measures to further improve the prognosis of such patients with hypertension with normal BP values is necessary. Particularly, whether physical activity, one of the main lifestyle modification strategies for patients with hypertension,9 should be emphasised even after reaching target BP levels should be addressed. This is important because the benefit of physical activity may not be limited to reducing obvious BP levels and can be present regardless of one’s metabolic health.10 11 In addition, distinct benefits of physical activity in controlling BP and reducing cardiovascular risk may be present independent of pharmacological interventions.12
In this study, we first assessed the risk of MACCEs and all-cause mortality in people with diverse hypertension statuses, including normotension, controlled hypertension and uncontrolled hypertension, in two middle-aged general population cohorts. We additionally investigated the risk of adverse outcomes stratified by the degree of physical activity. Our main goal was to suggest that those with controlled hypertension should be recognised for their residual risks for adverse outcomes and that physical activity may be encouraged for patients with hypertension even if their BP values seem normal.
Methods
Study setting and population
The study performed similar analyses in the two population-based cohorts, one from the UK, including predominantly European individuals, and one from the Republic of Korea, consisting mostly of Asian people (online supplemental method). The two countries are similar in that a nationwide health insurance service is provided for all citizens.
Supplemental material
The details of the UK Biobank have been described previously,13 and the cohort included >500 000 people of ages 40–69 from 2006 to 2010, with initial clinicodemographic, laboratory and lifestyle assessments with prospective follow-ups for various clinical endpoints.
The details of the Korean National Health Insurance Service (NHIS) database have been explained previously.14 In Korea, a free-of-charge general health screening is performed for over 10 million people by the NHIS, including clinicodemographic assessment, laboratory tests, evaluation of lifestyle and linkage to a nationwide claims database. We retrospectively constructed the Korean NHIS cohort, including those with available BP values in ≥2 sequential health examinations and the same baseline age range as the UK Biobank from 2009 to 2011.
Exclusion criteria
For the two cohorts, similar exclusion criteria were applied (figure 1). First, we excluded individuals with history of MACCEs or relevant cerebrocardiovascular diseases. Second, we excluded those without two complete measurements of BP values as they were implemented to determine the hypertension status. In the Korean NHIS cohort, as two separate examinations were implemented to assess BP exposure, those with transient hypertension status were also excluded. Third, we excluded those with missing information in the covariates for further multivariable investigations. Last, those with yet-untreated hypertension or hypertension without an identified history of antihypertensive medication were excluded, as the group was outside of the study aim and whether they received any medications after the health screenings was unknown. The details of the applied exclusion criteria in each cohort are described in the online supplemental method.
Study population. BP, blood pressure; HTN, hypertension; MACCE, major adverse cardiac and cerebrovascular events; NHIS, National Health Insurance Service.
Ascertainment of exposure
The main exposure was the following three statuses of hypertension: normotension, controlled hypertension and uncontrolled hypertension. The normotensive state was defined as a subject not receiving antihypertensive medication and with systolic BP <140 mm Hg and diastolic BP <90 mm Hg.5 Controlled hypertension was determined when a subject was within the normal BP range with the use of an antihypertensive medication. The uncontrolled state was defined as a subject receiving antihypertensive medication with systolic BP ≥140 mm Hg or diastolic BP ≥90 mm Hg. Two sensitivity analyses were performed: one with redefined exposures regarding high BP values as systolic BP ≥130 mm Hg or diastolic BP ≥80 mm Hg following the US guideline4; and one excluding those with potential hypotension (systolic BP <90 mm Hg or diastolic BP <60 mm Hg).
As the two cohorts had distinct advantages with regard to the BP data, we implemented BP values differently, aiming to compensate for the definition of each other. In the UK Biobank cohort, the advantage of the BP data was that two sequential measurements in the initial visit were performed, decreasing the possibility of incorrect measurement. Therefore, the BP value of a subject was determined by the mean value of the two readings.5 In the Korean NHIS cohort, the advantage of the BP data was that consecutive separate health screenings were performed annually or biennially, decreasing the effect of transient changes in BP status. Thus, both BP values in the first two separate examinations must meet the same criteria described above to be considered one of the hypertension statuses, and those with transient changes were excluded.
Secondary exposure was self-reported frequency of moderate or moderate-to-vigorous physical activity, categorised as none, 1–3 days per week or ≥4 days per week. In the UK Biobank cohort, frequencies of moderate-degree physical activity were collected, and in the Korean NHIS cohort frequencies of moderate and vigorous physical activity were collected, respectively, according to data availability and because of the differences in the questionnaires used (online supplemental method). As sensitivity analysis, first, we limited our analysis to individuals with normotension with optimal range BP (systolic BP <120 mm Hg and diastolic BP <80 mm Hg). Next, considering the possible measurement bias due to using the yet validated questionnaires in the primary analysis, we implemented the International Physical Activity Questionnaire (IPAQ) physical activity category, available in the UK Biobank, as the secondary exposure.15 Third, we tested the association between objectively estimated time of moderate-to-vigorous physical activity from accelerometer results and the risk of the study outcomes, which were available in the UK Biobank.16 Although the accelerometry results were collected after the baseline visit, we implemented the information as the proxy of the objective degree of moderate-to-vigorous physical activity as baseline. The details regarding the sensitivity analysis are described in the online supplemental method.
Ascertainment of study outcomes
The first study outcome was incident MACCEs. MACCEs in the UK Biobank included myocardial infarction, cardiac death and ischaemic stroke. However, myocardial infarction, revascularisation and acute ischaemic stroke were outcomes in the Korean NHIS cohort, as direct causes of death were not available.17 The second study outcome was all-cause mortality. Follow-up was initiated from the initial visit dates in the UK Biobank cohort and from the second examination dates in the Korean NHIS cohort after the exposure assessment was done with the first two examinations. The details are described in the online supplemental method.
Ascertainment of covariates
We collected the following clinical information: age, sex, smoking status, body mass index, waist circumference, baseline diabetes mellitus, cancer history, serum creatinine values, estimated glomerular filtration rate, high-density lipoprotein, low-density lipoprotein, triglyceride levels and usage of antihypertensive medications. The details are described in the online supplemental method.
Statistical analysis
We calculated the incidence rates of the study outcome. We performed Cox regression analysis to compare the study outcomes according to exposure. The other details of the statistical analysis are described in the online supplemental method.
Results
Study population
Among the available 502 506 UK Biobank cohort subjects, 320 026 individuals remained for the final analysis, and the most substantial proportion of exclusion was patients with hypertension but without any antihypertensive medications at baseline visit (n=120 649) (figure 1). Within the Korean NHIS cohort, 5 751 190 subjects were screened and 3 593 202 were included in the final data set. The most substantial proportion of exclusion was people having transient hypertension status that was not consistent between the first two inclusion examinations or those who had any missing BP values (n=1 117 730).
Baseline characteristics
In the UK Biobank cohort, 161 405 (73%), 18 844 (9%) and 39 777 (18%) individuals were included in the normotension, controlled hypertension and uncontrolled hypertension groups (table 1). The uncontrolled hypertension group had the oldest age range and the highest systolic BP values. Obesity, central obesity and diabetes were mostly prevalent in patients with hypertension. Approximately 40% and 50% of the study groups engaged in moderate-degree physical activity for 1–3 days per week and ≥4 days per week, respectively.
Baseline characteristics of the study population according to hypertension status
In the Korean NHIS cohort, 3 122 890 (87%), 383 828 (11%) and 86 484 (2%) had normotension, controlled hypertension and uncontrolled hypertension status for the two successive health examinations. The Korean NHIS cohort had a much lower proportion of obesity and central obesity than the UK Biobank cohort. On the other hand, people in the Korean NHIS cohort participated more infrequently in physical activity, as more than 40% of all the study groups engaged in zero days of moderate-to-vigorous physical activity per week.
Prognosis according to various hypertension statuses
In the UK Biobank cohort, the controlled hypertension group had a significantly worse prognosis than the normotensive group (table 2 and figure 2). Namely, the controlled hypertension group had a higher risk of MACCEs (adjusted HR 1.73 (95% CI 1.55 to 1.92)) and a higher risk of all-cause mortality (adjusted HR 1.28 (95% CI 1.18 to 1.39)) than the normotensive group in the fully adjusted model. The uncontrolled hypertension group had a higher risk of MACCEs than the controlled hypertension group but a similar risk of all-cause death compared with the controlled hypertension group.
Survival curves showing the cumulative survival probability from the study outcomes. HTN, hypertension; MACCE, major adverse cardiac and cerebrovascular events; NHIS National Health Insurance Service.
Association between frequency of moderate or moderate-to-vigorous physical activity and the risk of study outcome in subgroups divided by HTN status. The adjusted HR and CI were calculated in the multivariable Cox regression models adjusted for age, sex, body mass index, waist circumference, smoking history (non-smoker, ex-smoker, current smoker), underlying diabetes mellitus, cancer, laboratory results for estimated glomerular filtration rate, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol and triglycerides. The incidence rate was presented in per 1000 person-years unit. HTN, hypertension; MACCE, major adverse cardiac and cerebrovascular event; M-PA, moderate physical activity; MV-PA, moderate-to-vigorous physical activity; NHIS, National Health Insurance Service.
Risk of major adverse cardiac and cerebrovascular events and all-cause mortality according to various controlled states of hypertension
Risk of major adverse cardiac and cerebrovascular events and all-cause mortality according to frequencies of moderate or moderate-to-vigorous physical activity
In the Korean NHIS cohort, the results were similar, and the controlled hypertension group had a higher risk of MACCEs (adjusted HR 1.46 (95% CI 1.43 to 1.49)) and a higher risk of all-cause mortality (adjusted HR 1.29 (95% CI 1.26 to 1.32)) than the normotensive group. Unlike the UK Biobank cohort, the uncontrolled hypertension group had a higher risk of both MACCE and all-cause mortality than the controlled hypertension group.
These results were similarly identified when high BP was determined by applying the criteria of systolic BP ≥130 mm Hg or diastolic BP ≥80 mm Hg (online supplemental table 1) or after excluding those with possible hypotension (systolic BP <90 mm Hg or diastolic BP <60 mm Hg) (online supplemental table 2).
In addition, when the study outcomes were compared between the controlled hypertension group and the uncontrolled group accounting for antihypertensive medication types and numbers (online supplemental table 3), the controlled group had a significantly lower risk of MACCEs and all-cause mortality than the uncontrolled hypertension group in the Korean NHIS cohort. In the UK Biobank cohort, the risk of MACCE was also lower in the controlled hypertension group; however, the risk of all-cause mortality was not significantly different between the two groups.
Prognosis according to physical activity
The baseline characteristics according to the frequency of moderate or moderate-to-vigorous physical activity are shown in online supplemental table 4. We identified that those who were engaged in physical activity had significantly lower risk of adverse outcomes than those who were not in both cohorts (table 3, online supplemental table 5 and online supplemental table 6). The magnitude of the associations was larger for the all-cause mortality outcome than for MACCEs, but a dose-dependent response from frequencies of physical activity was not prominent.
We additionally stratified the risk of MACCEs and death in the three hypertension status groups according to frequencies of physical activity (figure 3 and online supplemental figure 1). In the UK Biobank cohort, there were no significant interactions between frequencies of physical activity and hypertension status for adverse outcomes (interaction p=0.73 with MACCEs and p=0.60 with all-cause mortality). The controlled hypertension group not engaged in moderate physical activity showed the highest risk for MACCEs and all-cause death. Further, patients with hypertension engaged in any frequency of moderate physical activity had comparable or even lower risk of all-cause mortality than persons with normotension not engaging in any moderate physical activity.
In the Korean NHIS cohort, interaction effect was again absent for the MACCE outcome (interaction p=0.22) but was present for all-cause death (interaction p=0.003). The results were similar in the Korean NHIS cohort, and the uncontrolled hypertension group not performing any moderate-to-vigorous physical activity showed the highest hazards for MACCEs and all-cause death, while the normotensive group showed the lowest risk. On the other hand, the hypertension group without any moderate-to-vigorous physical activity showed similar or even higher hazards for both MACCEs and all-cause death risks than those engaged in physical activity in the uncontrolled hypertension group. The magnitude of the association sizes of moderate-to-vigorous physical activity for all-cause death outcome was larger in the normotension group.
In additional sensitivity analysis including only individuals with normotension with optimal BP (systolic BP <120 mm Hg and diastolic BP <80 mm Hg), we again identified that being engaged in physical activity was associated with lower risk of adverse outcomes (online supplemental table 7). Further, when we used the physical activity measured by IPAQ in the UK Biobank as the exposure, those with moderate or high IPAQ physical activity showed significantly lower risk of MACCE in the controlled hypertension group than the low physical activity group (online supplemental table 8). For all-cause mortality, both moderate and high physical activity categories were associated with significantly lower risk of mortality in all three subgroups when compared with low physical activity category. Next, we implemented the accelerometer-determined time of moderate-to-vigorous physical activity, as a proxy of baseline physical activity, although the information was collected after the baseline visits. An hour increment of moderate-to-vigorous physical activity was significantly associated with lower hazards for MACCE (adjusted HR 0.93 (95% CI 0.87 to 0.99)) and risk of death (adjusted HR 0.89 (95% CI 0.80 to 0.99)), respectively, in the controlled hypertension group (online supplemental table 9). The possible benefits of longer time of objectively measured moderate-to-vigorous physical activity were also present in the normotension and uncontrolled hypertension groups.
Discussion
In this study investigating two large-scale population-based cohorts, individuals with controlled hypertension showed a significantly higher risk of MACCEs and all-cause mortality than persons with normotension. Moderate or moderate-to-vigorous physical activity was associated with better prognosis regardless of hypertension status, and the controlled hypertension group not engaged in such physical activity showed a particularly high risk of adverse outcomes.
As evidence is firm on the benefits of hypertension control, current guidelines suggest clear-cut target BP levels aiding clinicians to easily assess and manage hypertension in daily practice.4 5 However, residual risks for adverse outcomes in patients with hypertension who achieved normal BP values have rarely been investigated in a large-scale population-based cohort. According to our results, as patients with controlled hypertension showed generally better prognosis than those with uncontrolled hypertension, efforts to reach target BP values should remain the primary goal of hypertension management. What this study additionally emphasises is that patients with apparently controlled hypertension, although they may be considered to have successfully reached their targets, still had significantly higher risk of MACCEs and all-cause mortality than persons with normotension. A possible hypothesis for the residual risks in the controlled hypertension state is that previous damage caused by increased BP or related metabolic syndrome before reaching the controlled state may not be completely reversed.17 Regarding clinical aspects, as BP fluctuates with diverse patterns, few measurements of BP levels during clinical visits do not secure a well-controlled BP throughout the day.18 Therefore, attention to the cumulative effect of previous metabolic factors and thorough monitoring of non-office BP may be necessary to robustly evaluate certain patients with apparently controlled hypertension for their risk of adverse outcome.
Current guidelines suggest moderate or moderate-to-vigorous physical activity as one of the lifestyle modification strategies for patients with hypertension.4 5 On the other hand, evidence showing whether physical activity should be equally emphasised even for patients with controlled hypertension is relatively scarce. Our findings suggest that reducing average BP values may not be the sole benefit of exercising, and the benefits of physical activity may be present in those who already achieved normal BP values or even in those without hypertension.11 Simply, physical activity may prevent the development of hypertension or worsening of metabolic health, further reducing the risk of MACCEs or death, even in those with controlled hypertension or normotension.10 11 Furthermore, physical activity may independently promote cardiovascular health from visible metabolic parameters.19 Thus, this study supports that physical activity should be promoted throughout the healthcare system and society for the general population.20–22
There are several limitations that should be considered in our study. First, as the definitions of several variables differed, particularly for the MACCE outcome as cardiac deaths were not collected in the Korean NHIS cohort, directly comparing the reported numbers between the two cohorts may yield inappropriate interpretation. Second, as the study is observational, additional population-scale trials encouraging exercising are necessary to conclude its benefit. Third, there are other important lifestyle factors (eg, diet or alcohol) that warrant additional assessment for their benefits in cardiovascular health or survival of the general population. Fourth, the similar adjusted death risk in the controlled hypertension group and the uncontrolled hypertension group in the UK Biobank cohort could be a biased result from implementing single-visit BP measurements to determine the exposure, and the evidence indicating that reducing BP improves survival is firm, which should be heeded. Also, as one’s hypertension status may change over time, a study including a longer exposure assessment period or a time-varying status of hypertension is warranted. Last, as some diseased individuals may be less likely to be engaged in physical activity, the possible reverse causation in the physical activity assessment should be noted.
In conclusion, controlled hypertension based on few BP measurements is associated with higher risk of MACCEs and all-cause mortality compared with normotension. The importance of moderate or moderate-to-vigorous physical activity may be consistently emphasised, even for patients with hypertension who have reached target BP levels or for individuals with normotension.
Key messages
What is already known on this subject?
Reducing blood pressure to target levels in people with hypertension is crucial to reduce the risk of major adverse cardiac and cerebrovascular events (MACCEs) or all-cause mortality.
Moderate or moderate-to-vigorous physical activity reduces blood pressure and benefits cardiovascular health.
What might this study add?
In this large-scale observational study including two population-based cohorts, individuals with controlled hypertension had a higher risk of MACCEs and all-cause mortality when compared with individuals with normotension even if their blood pressure values were in target levels.
Moderate or moderate-to-vigorous physical activity was associated with lower risk of MACCEs and death in individuals with controlled hypertension and even in those with normotension or uncontrolled hypertension.
How might this impact on clinical practice?
Clinicians may encourage physical activity for general population, even for patients with controlled hypertension, not being reassured by their achieved target blood pressure values.
Data availability statement
The data that support the findings of this study are available from the UK Biobank (application no. 53799) and the NHIS of the Republic of Korea (NHIS-2018-1-247). The study data will be made available by organisations upon reasonable request.
Ethics statements
Ethics approval
The study was performed in accordance with the Declaration of Helsinki. The investigation for the UK Biobank cohort was approved by the institutional review boards of Seoul National University Hospital (E-1910-044-1067) and by the UK Biobank organisation (application no. 53799). The investigation for the Korean National Health Insurance Service (NHIS) cohort was approved by the institutional review boards (E-2001-038-1092) and by the NHIS (NHIS-2018-1-247). The requirement for informed consent was waived as the study investigated anonymous databases.
Acknowledgments
We thank the UK Biobank consortium and the NHIS of Korea for providing the data for this study.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
SP and KH contributed equally.
Contributors The corresponding author attests that all listed authors meet the authorship criteria and that no others meeting the criteria have been omitted. SeP, KH, SaP, HL, KWJ and DKK contributed to the conception and design of the study. SL, YK, YL, MWK, YCK, SSH, JPL, KWJ, CSL, YSK and DKK provided statistical advice and interpreted the data. SeP, KH and SaP performed the main statistical analysis, assisted by SL. HL, JPL, KWJ, CSL, YSK and DKK provided advice regarding data interpretation. YCK, SSH, HL, JPL, KWJ, CSL and YSK provided material support during the study. SeP, KH and DKK had full access to all data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. All authors participated in drafting the manuscript. All authors reviewed the manuscript and approved the final version to be published.
Funding This work was supported by a grant from the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare, Republic of Korea (HI17C0530). The funder played no role in the performance of the study, and the study was performed independently by the authors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.