Comparison between unattended automated office blood pressure and conventional office blood pressure under the environment of health checkup among Japanese general population

Abstract

Unattended automated office blood pressure (AOBP) measurement has been endorsed as the preferred in‐office measurement modality in recent Canadian and American clinical practice guidelines. However, the difference between AOBP and conventional office blood pressure (CBP) under the environment of a health checkup remains unclear. We aimed to identify the clinical significance of AOBP as compared to CBP under the environment of a health checkup. There were 491 participants (333 females, mean age of 62.5 years) who were at least 20 years old, including 179 participants who were previously diagnosed with hypertension. Mean AOBPs were 131.8 ± 20.9/76.6 ± 11.7 mm Hg, and CBPs were 135.6 ± 21.6/77.3 ± 11.5 mm Hg. There was a difference of 3.9 mm Hg in systolic blood pressure (SBP) and 0.8 mm Hg in diastolic BP between AOBP and CBP. In all participants, SBP and pulse pressure, as well as the white coat effect (WCE), increased with age. The cutoff value used was 140/90 mm Hg for CBP and 135/85 mm Hg for AOBP, and the prevalence of WCE and masked hypertension effect (MHE) was 12.4% and 14.1%, respectively. Even in a health checkup environment of the general population, there was a difference between the AOBP and CBP, and the WCE was observed more strongly in the elderly with a history of hypertension, suggesting that a combination of AOBP with CBP may be useful in detecting WCE and MHE in all clinical scenarios including health checkups, and help solve the “hypertension paradox” not only in Japan but in all over the world.

1. INTRODUCTION

Accurate evaluation and optimal management of hypertension are important to prevent hypertension‐related target organ damage. ¹ , ² Office blood pressure (BP) has traditionally been measured by health care professionals. However, these BP measurements often produce inaccurate readings that can lead to disease misclassification and inappropriate treatment decisions. Furthermore, the white coat effect (WCE) is a well‐documented limitation of these measurements. ³ The diagnosis of hypertension involves a 24‐hour ambulatory blood pressure monitoring (ABPM) and home BP rather than conventional office BP (CBP). ⁴ , ⁵ Although ABPM is recommended by current guidelines as part of the routine assessment for diagnosing and managing hypertension, ⁴ , ⁵ ABPM can cause disruption to daily activities and relies on patient compliance. Therefore, ABPM remains impractical for repeated or regular BP monitoring. In contrast, automated office BP (AOBP) is gaining recognition as an alternative way of measuring BP in the clinic. ⁶ , ⁷ Sphygmomanometer that measures AOBP allows multiple BP readings to be recorded with the patient at rest and the operator absent from the examination room, which presumably minimizes WCE. AOBP readings correlated more closely with those of ABPM than CBP readings. ⁸ , ⁹ Thus, recent Canadian and American clinical practice guidelines ¹⁰ , ¹¹ endorse AOBP measurements as the preferred in‐office measurement modality. However, the difference between AOBP and CBP under the environment of a health checkup has not been clarified. Therefore, this study aimed to compare the AOBP measurements with CBP measurements and to detect WCE and masked hypertension effect (MHE) based on AOBP and CBP for the general population under the environment of a health checkup.

2. METHOD

2.1. Study design and population

This retrospective cross‐sectional study used data acquired during health promotion events in Asahikawa and Furano, Hokkaido, Japan, from September 2017 to June 2018. Participants were recruited by advertising the study through a public relations magazine. We enrolled 491 participants aged at least 20 years after excluding younger patients and/or those who declined to participate in this study.

We acquired written consent for the secondary use of data from each participant. In detail, participants were asked “I admit to utilizing of my data for academic research purposes.” with “Yes/No” options. This protocol is in line with the Japanese Ethical Guidelines for Epidemiological Research, which regulates that informed consent is not necessarily required for observational studies utilizing the existing data. ¹² All procedures of this study received approval from the Asahikawa Medical University Ethics Committee (No. 20 037).

2.2. BP measurements

To compare the AOBP measurements and CBP measurements of the general population under the environment of a health checkup, participants’ AOBP and CBP were measured as follows. First, AOBP measurements were taken with the Omron HEM‐907 (Omron Healthcare, Kyoto, Japan), which was also used in the Systolic Blood Pressure Intervention Trial (SPRINT). ¹³ Participants were seated comfortably for 3 minutes, and then their preferred (more commonly used) upper arm was used. Research staff trained to program an Omron HEM‐907 were advised to wait 3 minutes before recording two readings at 1‐minute intervals. After the device was activated, the research staff left the dedicated examination room, which had partitions between each participant, leaving the participants alone. The measurements were automatically performed, and completing questionnaires and talking were prohibited throughout this process. After measurement was complete, research staff checked the AOBP measurements.

Next, CBP was measured in the dedicated area, in accordance with the Japanese Society of Hypertension Guidelines for the Management of Hypertension published in 2014 (JSH 2014). ⁴ While maintaining the arm‐cuff position at the level of the heart in a seated position, two readings were recorded by doctors wearing white coats at 1‐minute intervals after a 2‐minute rest using an electronic sphygmomanometer (Terumo Elemano ES‐H55), which deflates automatically by a single oscillometric measurement after manual inflation. This instrument has passed a clinical validation study conducted in the general population using a recognized standard protocol. ¹⁴ We compared the average measurement value of systolic BP (SBP), diastolic BP (DBP), and pulse rate (PR) between AOBP and CBP by age and with/without history of hypertension.

In this study, we defined hypertension status using conventional office SBP (CSBP) and automated office SBP (AOSBP) using mean SBPs (CSBP < 140 mm Hg, AOSBP < 135 mm Hg) as follows: MHE, CSBP < 140 mm Hg and AOSBP ≥ 135 mm Hg; WCE, CSBP ≥ 140 mm Hg and AOSBP < 135 mm Hg; poorly controlled hypertension, CSBP ≥ 140 mm Hg and AOSBP ≥ 135 mm Hg; and well‐controlled BP, CSBP < 140 mm Hg and AOSBP < 135 mm Hg according to previous report ¹⁵ with minor modifications.

2.3. Other measurements

We used the self‐reported standard questionnaire to obtain information on each patient's medical history and related conditions including sex, age, body mass index (BMI), smoking and drinking status, diabetes, dyslipidemia, past history of stroke, heart disease and kidney disease, and exercise habits. Smoking status was evaluated by positive answers to the question “Have you smoked until now?” For exercise habits, two questions were asked: “Are you in the habit of exercising to sweat lightly for over 30 minutes each time, two times weekly, for over a year?” and “In your daily life, do you walk or do any equivalent amount of physical activity for more than one hour a day?”

2.4. Statistical analyses

We used IBM SPSS Advanced Statistical Version 26.0 (SPSS) for database management and statistical analyses. The statistical significance was α < 0.05 in 2‐sided tests. All data are expressed as the mean (SD) unless otherwise stated. For the AOBP and CBP values, all of the readings were averaged for the analysis. Variable differences between the two groups with and without history of hypertension were analyzed using Student's t test, Mann‐Whitney test, or Fisher's exact test. Pearson's correlation coefficients were calculated to determine the correlation between each BP value because BPs can be treated as a normal distribution. The differences of SBP, DBP, and PR were calculated as the CBP minus the AOBP and were compared among each other, dividing the participants with and without history of hypertension into tertile groups based on age.

3. RESULTS

3.1. Subject characteristics

A total of 495 participants attended the health promotion events, and four participants were excluded due to the age criteria. The clinical characteristics of the 491 eligible participants (158 male and 323 female) are shown in Table 1. The median age was 68 years, including the ages of 179 participants (36.5%) who had previously diagnosed hypertension. The average BMI was 23.1 kg/m². The distribution of participants by age and with and without history of hypertension is shown in Figure 1. Majority of participants were older than 60 years old, with 70‐79 years old being the most prevalent age category of those with hypertension and 60‐69 years old of those without. The median age, mean BMI, prevalence of diabetes mellitus, dyslipidemia, past histories of stroke, heart disease, and kidney disease were significantly higher in participants with a history of hypertension than in those without. In contrast, prevalence of smoking, drinking, and exercise habits was not different between the participants with and without history of hypertension.

TABLE 1.

Characteristics of enrolled participants

	With history of hypertension	Without history of hypertension	Total
N	179	312	491
Age (median [IQI]) (years)	71 [67‐75]*	62 [43‐70]	68 [55‐74]
Sex (male/female)	68/111	90/222	158/323
Body mass index (kg/m2)	24.5 ± 3.5*	22.3 ± 3.5	23.1 ± 3.7
Prevalence of diabetes mellitus (n (%))	26 (14.5)*	6 (1.9)	32 (6.5)
Dyslipidemia (n (%))	73 (40.8)*	41 (13.1)	114 (23.2)
Past history of stroke (n (%))	17 (9.5)*	1 (0.3)	18 (3.7)
Heart disease (n (%))	20 (11.2)*	10 (3.2)	30 (6.1)
Kidney disease (n (%))	15 (8.4)*	3 (1.0)	18 (3.7)
Prevalence of smoking (n (%))	26 (14.5)	41 (13.5)	67 (13.6)
Drinking (n (%))	86 (48.0)	170 (54.5)	256 (52.1)
Exercise habits (n (%))	77 (43.0)	127 (40.7)	204 (41.5)
CBP
SBP (mm Hg)	146.8 ± 19.5*	129.2 ± 20.1	135.6 ± 21.6
DBP (mm Hg)	80.6 ± 10.8*	75.5 ± 11.6	77.3 ± 11.5
PR (/min)	73.4 ± 12.2	72.6 ± 11.5	72.9 ± 11.7
AOBP
SBP (mm Hg)	140.0 ± 19.3*	127.0 ± 20.4	131.8 ± 20.9
DBP (mm Hg)	78.9 ± 11.3*	75.2 ± 11.8	76.6 ± 11.7
PR (/min)	73.4 ± 10.9*	72.8 ± 11.8	73.0 ± 11.5
Difference between CBP and AOBP
SBP (mm Hg)	6.9 ± 23.4 a	2.1 ± 17.7	3.9 ± 20.1
DBP (mm Hg)	1.6 ± 13.1	0.3 ± 10.9	0.8 ± 11.8
PR (/min)	−0.2 ± 10.7	0.01 ± 10.8	−0.1 ± 10.8

Abbreviations: AOBP, unattended automated office blood pressure; CBP, conventional office blood pressure; DBP, diastolic blood pressure; PR, pulse rate; SBP, systolic blood pressure.

^a P = .053 vs without history of hypertension.

^* P < .01 vs without history of hypertension

FIGURE 1.

Age distribution of 491 participants with/without history of hypertension

3.2. Comparison of mean BP readings

The mean BP readings of AOBP and office BP are shown in Table 1 and Figure 2. On the AOBP readings, mean SBP, DBP, and PR were 131.8 mm Hg, 76.6 mm Hg, and 73.0 bpm, respectively. On the CBP readings, mean SBP, DBP, and PR were 135.6 mm Hg, 77.3 mm Hg, and 72.9 bpm, respectively. In all participants, SBP increased with age in both measurements, and pulse pressure increased from 60 years of age or older (Figure 2). The mean CSBP by age was greater than each of the AOSBP values, which were noted to be caused by the WCE, and the prevalence of this effect was observed to increase with age (Figure 2).

FIGURE 2.

Mean systolic blood pressure (SBP), diastolic blood pressure (DBP), and pulse rate (PR) of conventional office blood pressure (CBP) (A) and unattended automated office blood pressure (AOBP) measurement (B) according to age

Next, we subtracted AOBP from CBP and the differences of each of the readings are shown in Table 1 and Figure 3. In all participants, there was a difference of 3.9 mm Hg in SBP and 0.8 mm Hg in DBP, whereas no significant difference was observed between those of male and female participants. In participants with history of hypertension, a difference of 6.9 mm Hg in SBP and 1.6 mm Hg in DBP was observed, whereas 2.1 mm Hg in SBP and 0.3 mm Hg in DBP in those without history of hypertension. When we divided the participants with and without history of hypertension into tertile groups based on age, there were no significant differences seen in SBP, DBP, and PR between the groups in the participants with history of hypertension (Figure 3A), whereas the difference of SBP was significantly greater in the second and third tertile than the first, suggesting that aging might have a greater impact on WCE in the general population without history of hypertension (Figure 3B).

FIGURE 3.

Mean difference of systolic blood pressure (SBP), diastolic blood pressure (DBP) and pulse rate (PR), and unattended automated office blood pressure (AOBP) subtracted from conventional office blood pressure (CBP) measurement according to age tertile in participants with history of hypertension (A) and without (B). *P < .05 vs first tertile

3.3. Detection of WCE and MHE

When we defined hypertension status as the cutoff values of CSBP < 140 mm Hg and AOSBP < 135 mm Hg, the prevalence of WCE and MHE was 12.4% and 14.1%, respectively, in all participants, whereas that of WCE, MHE, and poorly controlled hypertension was 21%, 16%, and 42%, respectively, in the participants with history of hypertension (Figure 4A). In contrast, those of WCE, MHE, and poorly controlled hypertension was 7%, 13%, and 20%, respectively, in the participants without history of hypertension (Figure 4B).

FIGURE 4.

Distributions of blood pressure (BP) status based on unattended automated office blood pressure (AOBP) from conventional office blood pressure (CBP) measurement using cutoff values of 140 mm Hg for systolic CBP and 135 mm Hg for systolic AOBP in participants with history of hypertension (A) and without (B)

4. DISCUSSION

To the best of our knowledge, this is the first study to identify the clinical significance of AOBP as compared to CBP under the environment of a health checkup. In all participants, AOBP was lower than CBP by an average of 3.9/0.8 mm Hg. Furthermore, the prevalence of WCE and MHE was 15.9% and 9.0%, respectively, suggesting that a combination of AOBP and CBP may be useful in detecting WCE and MHE under the environment of a health checkup and help solve the “hypertension paradox” not only in Japan but also in all over the world.

Based on the findings from NIPPON DATA, the number of hypertensives in Japan is estimated to be 43 million as of 2017. ¹⁶ Of these individuals, 31 million are estimated to be poorly controlled (140/90 mm Hg or higher), 14 million (33%) are estimated to be unaware of hypertension, 4.5 million (11%) are estimated to remain untreated despite awareness of the disease, and 12.5 million (29%) are estimated to be poorly controlled despite ongoing treatment and this status is called the “hypertension paradox.” ¹ , ⁵ Furthermore, it is well known that cardiovascular risk in WCE appears to be dependent on the presence of coexisting risk factors, whereas patients with MHE are at increased risk of target organ damage and cardiovascular events. ³ , ¹⁷ In the present study, the prevalence of WCE and MHE was 12.4% and 14.1%, respectively, reporting similar results to IDACO (International Database on Ambulatory Blood Pressure Monitoring in Relation to Cardiovascular Outcomes), which included random population samples from Asia, Europe, and South America, where white coat and masked hypertension frequencies ranged from 6.3% to 12.5% and from 9.7% to 19.6%, respectively, in untreated participants. ¹⁸ Notably, the prevalence of WCE, MHE, and poorly controlled hypertension was 21%, 16%, and 42%, respectively, in participants with history of hypertension. Furthermore, even in the participants without history of hypertension, the prevalence of WCE, MHE, and poorly controlled hypertension was 7%, 13%, and 20%, respectively, suggesting that combination of AOBP with CBP may be useful in detecting WCE and MHE under the environment of a health checkup.

In SPRINT, ¹³ AOBP was measured using the automated Omron HEM‐907 cuff‐oscillometric monitor with a 5‐minute rest period before the first measurement and a 1‐minute interval in between measurements. However, only 38 of the 88 clinics complied with the definition of AOBP, and staff attended measurement during the rest time and/or measurement at the other 50 clinics. Therefore, actual BP measurements were almost identical on cross‐classification of the attended and unattended groups in SPRINT. Recent systematic review and meta‐analyses reported that the agreement between AOBP and daytime ABPM was significantly better than CBP. A mean difference of 14.5 mm Hg or 10.48 mm Hg for systolic CBP as compared with AOBP was observed, whereas systolic daytime ABPM and AOBP were similar, with a mean difference of 0.3 mm Hg or 1.85 mm Hg. ⁸ , ⁹ In the present study, the mean difference between the CSBP and AOSBP was larger in older than in younger participants in all groups, implying that the fundamental differences between CBP and AOBP can be explained in part by the WCE in a sense. Although the unattended AOBP is measured at a quiet space, the remaining portion of the WCE which can be called the clinic effect remains, since the circumstances are quite different from taking the BP at home. Therefore, AOBP measurements might be a useful alternative to CBP measurements as previously reported. ¹⁹ , ²⁰

There are some limitations to the present study. First, the cross‐sectional and retrospective design of the analysis limits causal inferences for the associations found. We were unable to assess the long‐term trends or reproducibility in the differences between CBP and AOBP. Second, in our study, the CBP was measured in each participant by doctors wearing white coats using automatic oscillometric devices, whereas in health checkup settings, CBP might be measured by non‐physician health care workers. Thus, the procedures to measure CBP in this study may cause overestimation of the WCE. Furthermore, most CBP measurements in an office setting might be determined by the auscultatory method using an aneroid device. Third, we have measured CBP and AOBP by different oscillometric devices and have not compared BP measurements between both devices. Therefore, the BP measurements might be slightly different between both devices, although both devices have passed clinical validation studies in the general population, which followed recognized standard protocols. ¹⁴ , ²¹ All BP measurements in all clinical settings, including health checkups, may have to be performed by oscillometric AOBP and validated devices. Finally, the study population was relatively small. This was mainly because of some important problems in measuring AOBP, such as space and time. The consumption of office space and long measurement time reduces the feasibility of employing unattended AOBP measurements in clinical practice. Since AOBP reading in this study requires >10 minutes per person, AOBP would not be feasible for application in a routine health checkup. Further studies on a large number of patients aimed at investigating whether AOBP can accurately detect WCE and MHE under the environment of a health checkup are warranted.

Finally, to the best of our knowledge, this is the first study to compare the AOBP and CBP under the environment of a health checkup. The mean difference between the CSBP and AOSBP was larger in older than in younger participants and in participants with a history of hypertension than those without. An important issue in Asia is not just the high prevalence of hypertension, particularly in some countries, but also the low level of awareness and treatment rates in many regions. ²² Therefore, in all clinical scenarios including health checkups, the combination of AOBP and CBP measurements is very useful for detecting WCE and MHE and may help in solving the “hypertension paradox” in all over the world.

CONFLICT OF INTEREST

None.

AUTHOR CONTRIBUTIONS

H. Sakuma, N. Nakagawa, and N. Hasebe contributed to conception, design, and preparation of the first draft. N. Nakagawa has conducted the statistical analyses. All authors have read and given final approval of the version to be published.

Sakuma H, Nakagawa N, Horiuchi K, et al. Comparison between unattended automated office blood pressure and conventional office blood pressure under the environment of health checkup among Japanese general population. J Clin Hypertens. 2020;22:1800–1806. 10.1111/jch.14008