Abstract
Automated office blood pressure measurement eliminates the white coat effect and is associated with awake ambulatory blood pressure. This study examined whether automated office blood pressure values at lower limits were comparable to those of awake and mean 24‐hour ambulatory blood pressure. A total of 552 patients were included in the study, involving 293 (53.1%) men and 259 (46.9%) women, with a mean age 55.0 ± 12.5, of whom 36% were treated for hypertension. Both systolic and diastolic automated office blood pressures exhibited lower values compared to awake ambulatory blood pressure among 254 individuals with systolic automated office blood pressure <130 mm Hg (119 ± 8 mm Hg vs 125 ± 11 mm Hg, P < .0001 and 75 ± 9 mm Hg vs 79 ± 9 mm Hg, P < .0001 for systolic and diastolic BPs, respectively). Furthermore, the comparison of systolic automated office blood pressure to the mean 24‐hour ambulatory blood pressure levels also showed lower values (119 ± 8 vs 121 ± 10, P = .007), whereas the diastolic automated office blood pressure measurements were similar to 24‐hour ambulatory blood pressure values. Our findings show that when automated office blood pressure readings express values <130/80 mm Hg in repeated office visits, further investigation should be performed only when masked hypertension is suspected; otherwise, higher automated office blood pressure values could be used for the diagnosis of uncontrolled hypertension, especially in individuals with organ damage.
Keywords: agreement, automated office blood pressure, cutoff values, hypertension, mean ambulatory blood pressure values
1. INTRODUCTION
International guidelines recommend out‐of‐office blood pressure (BP) measurements, ambulatory BP (ABP), or home BP (HBP) for diagnosis of hypertension rather than conventional office BP. 1 , 2 This preference was based on evidence that out‐of‐office BP monitoring is a better predictor of hypertension and associated cardiovascular risk than office. Research studies have shown that office BP measurements, taken in accordance with the established guidelines, provide lower BP values compared with conventional office readings. Specifically, conventional office readings are 10/7 mm Hg higher than “research‐quality” readings and are less accurate since they are subject to the white coat effect (WCE). 3 Automated office blood pressure (AOBP) taken with the patient seated alone in the examination room reduces or eliminates the WCE and produces values up to 5‐15 mm Hg lower than those obtained by conventional office BP and are thus closely related to awake ABP. 2 , 3 It is also known that at a target systolic BP < 130 mm Hg, all office measurements, AOBP, attended oscillometric and manual performed according to guidelines appear to be devoid of a WCE. 4 Moreover, when systolic AOBP performed without a preceding rest was < 130 mm Hg, similar to awake ABP values were obtained. 5 , 6 In this context, it has been shown that at the lower part of the normal BP range, the mean awake ABP becomes progressively higher than the AOBP in both treated and untreated individuals. 7 , 8 Since BP thresholds and targets for treatment in the lower part of the normal range are equally important, we conducted this study to verify whether AOBP represents a more valuable approximation to mean 24‐hour ABP than awake ABP.
2. METHODS
2.1. Study participants
We evaluated all patients referred to the Hypertension and Cardiovascular Disease Prevention Center by their family physicians for suspected hypertension. Both treated and untreated patients were considered for inclusion in the study. The inclusion criteria were individuals aged ≥18 years who had never taken or who had not received antihypertensive medication for at least the previous 6 months and patients on stable antihypertensive medication for at least 4 weeks. Hypertension was defined as an average office reading of systolic BP (SBP) ≥140 mm Hg or diastolic BP (DBP) ≥90 mm Hg on three consecutive visits, based on the semi‐automated oscillometric technique. Exclusion criteria were secondary hypertension, severe arrhythmia, gross proteinuria (protein >1 g/dL per 24 hour), renal failure with estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m2 or heart failure (HF) with ejection fraction <30%, mental disorders and severe non‐cardiovascular disease that limits survival. Written informed consent was obtained from all patients, and the study was approved by the Scientific Board of the hospital.
2.2. Study procedures
Patients were subsequently monitored by AOBP and ABP techniques. Using a fully automated Omron‐HEM 907 XL sphygmomanometer, 9 the average of 3 readings at 1‐minute intervals after a 5‐minute resting period in the examination room was obtained. Appropriate cuff size was used for all patients, and measurements were completed by a trained nurse. Participants were alone during the 5‐minute resting period and during the 3 measurements. AOBP was recorded in the morning (08:3‐10:30 am) on a routine working day prior to 24‐hour ABP monitoring. Twenty‐four‐hour ABP was then monitored using the validated Microlife Watch BPO3 device. 10 Measurements were performed at 20‐minute intervals for 24hours, and study participants were instructed to remain still with the forearm extended during each BP reading. Awake and nighttime periods were defined according to the patients’ diaries (awake and asleep periods). Seven patients whose ABP recordings yielded less than 70% usable BP readings were excluded. All valid awake and nighttime ABP readings were averaged to provide a single awake and nighttime ABP value per study participant.
2.3. Statistical analysis
Continuous variables are presented as mean ± standard deviation (SD). The paired sample t test was used to determine the mean difference between AOBP, awake ABP, and mean 24‐hour ABP, among patients with systolic AOBP < 130 mm Hg and ≥130 mm Hg, for the systolic and diastolic BP, respectively. The intraclass correlation coefficients (b) and Pearson's r for all sets of systolic BP measurements were also calculated. Mean systolic AOBP and mean systolic 24‐hour ABP among patients with mean systolic AOBP < 130 and ≥ 130 mm Hg were compared using Bland‐Altman plots. Bias defined as the mean difference of the measurements and 95% limits of agreement is reported. Finally, a box plot of systolic BP (AOBP, awake ABP, and mean 24‐hour ABP) among patients with mean systolic AOBP < 130 and ≥130 mm Hg was also generated. We used IBM SPSS version 22.
3. RESULTS
A total of 552 patients were included in the study, involving 293 (53.1%) men and 259 (46.9%) women, with a mean age 55.0 ± 12.5, of whom 35.9% were treated for hypertension. Their clinical characteristics and their BP are shown in Table 1.
TABLE 1.
Study population (N = 552)
| Variables | Mean value ± SD |
|---|---|
| Age, y | 52.0 ± 12.6 |
| Systolic AOBP with 5 min of rest, mm Hg | 132.8 ± 16.8 |
| Diastolic AOBP with 5 min of rest, mm Hg | 82.0 ± 12.5 |
| Systolic awake ABP, mm Hg | 131.6 ± 13.5 |
| Diastolic awake ABP, mm Hg | 82.6 ± 10.8 |
| Systolic 24h mean BP, mm Hg | 127.3 ± 12.9 |
| Diastolic 24h mean BP, mm Hg | 78.7 ± 10.2 |
| N (%) | |
|---|---|
| Gender | |
| Males | 293 (53.1%) |
| Females | 259 (46.9%) |
| Treated hypertensives | 198 (35.9%) |
| Untreated hypertensives | 354 (64.1%) |
Abbreviations: ABP, ambulatory blood pressure; AOBP, automated office blood pressure; SD, standard deviation.
When AOBP values <130/80 mm Hg were used as a threshold, awake ABP expressed higher BP readings. Higher ABP values were also observed when systolic AOBP < 130 mm Hg levels were compared against systolic 24‐hour ABP, while diastolic AOBP values <80 mm Hg yielded similar values to those of diastolic 24hour ABP (Table 2).
TABLE 2.
AOBP, awake ABP, and mean 24‐hour ABP, among patients with systolic AOBP < 130 mm Hg (N = 254)
| STUDY population (N = 254) | AOBP | Awake ABP | P value |
|---|---|---|---|
| Mean ± SD‡ systolic BP, mm Hg | 119 ± 8 | 125 ± 11 | <.001 |
| Mean ± SD diastolic BP, mm Hg | 75 ± 9 | 79 ± 9 | <.001 |
| AOBP | Mean 24‐h ABP | ||
|---|---|---|---|
| Mean ± SD systolic BP, mm Hg | 119 ± 8 | 121 ± 10 | .007 |
| Mean ± SD diastolic BP, mm Hg | 75 ± 9 | 75.0 ± 9 | .368 |
Abbreviations: ABP, ambulatory blood pressure; AOBP, automated office blood pressure; BP, blood pressure; SD, standard deviation.
Statistically significant P < .05 are indicated in bold.
In contrast, when systolic and diastolic AOBP values were ≥130/80 mm Hg lower, awake ABP and mean 24‐hour ABP were observed (Table 3).
TABLE 3.
AOBP (unattended with 5 min of preceding rest), awake ABP, and mean 24‐hour ABP, among patients with systolic AOBP ≥ 130 mm Hg (N = 298)
| Study population (N = 298) | AOBP | Awake ABP | P value |
|---|---|---|---|
| Mean ± SD‡ systolic BP, mm Hg | 144 ± 13 | 137 ± 13 | <.001 |
| Mean ± SD diastolic BP, mm Hg | 89 ± 12 | 86 ± 11 | <.001 |
| AOBP | Mean 24‐hour ABP | ||
|---|---|---|---|
| Mean ± SD systolic BP, mm Hg | 144 ± 13 | 133 ± 13 | <.001 |
| Mean ± SD diastolic BP, mm Hg | 89 ± 12 | 82 ± 10 | <.001 |
Abbreviations: ABP, ambulatory blood pressure; AOBP, automated office blood pressure; SD, standard deviation; BP, blood pressure.
The intraclass correlation coefficient (ICC) for mean systolic AOBP and awake ABP was 0.75 (P < .001; 95% confidence interval, 0.70‐0.78). Similarly, high coefficients were found when mean systolic AOBP was either < or ≥130 mm Hg. Likewise, similar results were obtained from the comparison of mean systolic AOBP and mean systolic 24‐hour ABP. The ICCs for all sets of systolic BP measurements are shown in Table 4.
TABLE 4.
Intraclass correlation coefficients (b) for all sets of systolic BP measurements
| BP measurements | Intraclass correlation (b) | P value | 95% CI |
|---|---|---|---|
| Mean AOBP‐awake ABP | 0.75 | <.001 | 0.70‐0.78 |
| Mean AOBP‐awake ABP (group 1) * | 0.56 | <.001 | 0.44‐0.66 |
| Mean AOBP‐awake ABP (group 2) ** | 0.69 | <.001 | 0.60‐0.74 |
| Mean AOB‐mean 24‐hour ABP | 0.75 | <.001 | 0.70‐0.79 |
| Mean AOBP‐mean 24‐hour ABP (group 1) * | 0.57 | <.001 | 0.44‐0.66 |
| Mean AOBP‐mean 24‐hour ABP (group 2) ** | 0.69 | <.001 | 0.61‐0.75 |
Group 1: In patients with mean systolic AOBP < 130 mm Hg. Group 2: In patients with mean systolic AOBP ≥ 130 mm Hg.
Corresponds to Group 1.
Corresponds to Group 2.
The correlation coefficients (Pearson's r) for all sets of systolic BP measurements are shown in Table 5. Results revealed a strong correlation between both mean AOBP with awake ABP and mean 24‐hour ABP, and this correlation remained high for patients with mean systolic AOBP either < or ≥130 mm Hg.
TABLE 5.
Pearson's r for all sets of systolic BP measurements
| BP measurements | Pearson's r | P value |
|---|---|---|
| Mean AOBP‐awake ABP | 0.612 | <.001 |
| Mean AOBP‐awake ABP (group 1) * | 0.405 | <.001 |
| Mean AOBP‐awake ABP (group 2) ** | 0.521 | <.001 |
| Mean AOBP‐mean 24‐hour ABP | 0.626 | <.001 |
| Mean AOBP‐mean 24‐hour ABP (group 1) * | 0.401 | <.001 |
| Mean AOBP‐mean 24‐hour ABP (group 2) ** | 0.530 | <.001 |
Group 1: In patients with mean systolic AOBP < 130 mmHg. Group 2: In patients with mean systolic AOBP ≥ 130 mmHg.
Corresponds to Group 1.
Corresponds to Group 2.
Bland‐Altman plots for the comparison of systolic AOBP with both awake and mean 24‐hour ABP among patients with mean systolic AOBP < 130 mm Hg and mean systolic AOBP ≥ 130 mm Hg are given in Figures 1 and 2, respectively. A positive bias and equal to 11.7 mm Hg (95% CI 10.3‐13.1 mm Hg) for the comparison of mean systolic AOBP vs systolic awake ABP among those with mean systolic AOBP ≥ 130 mm Hg (Figure 1A). Similarly, the bias remained positive and equal to (95% CI 5.1‐8.9 mm Hg) was observed for the mean systolic AOBP vs mean systolic 24‐hour ABP among patients with mean systolic AOBP ≥ 130 mm Hg (Figure 1B).
FIGURE 1.
A, Bland‐Altman plot comparing mean systolic AOBP (mm Hg) vs mean systolic 24‐hour ABP (mm Hg) in patients with mean systolic AOBP ≥ 130 mm Hg (N = 298). Solid lines, mean bias; dashed lines, 95% limits of agreement. B, Bland‐Altman plot comparing mean systolic AOBP (mm Hg) vs systolic awake ABP (mm Hg) in patients with mean systolic AOBP ≥ 130 mm Hg (N = 298). Solid lines, mean bias; dashed lines, 95% limits of agreement
FIGURE 2.
A, Bland‐Altman plot comparing mean systolic AOBP (mm Hg) vs mean systolic 24‐hour ABP (mm Hg) in patients with mean systolic AOBP < 130 mm Hg (N = 254). Solid lines, mean bias; dashed lines, 95% limits of agreement. B, Bland‐Altman plot comparing mean systolic AOBP (mm Hg) vs systolic awake ABP (mm Hg) in patients with mean systolic AOBP < 130 mm Hg (N = 254). Solid lines, mean bias; dashed lines, 95% limits of agreement
In contrast, a negative bias of −1.7 mm Hg (95% CI −2.9 to −0.5 mm Hg) for the comparison between mean systolic AOBP and systolic awake ABP among patients with mean systolic AOBP < 130 mm Hg (Figure 2A). The bias remained negative, and equal −5.6 mm Hg (95% CI −7.3 to −4.6 mm Hg) was observed for the mean systolic AOBP vs the mean systolic 24‐hour ABP among patients with mean systolic AOBP < 130 mm Hg (Figure 2B).
The Bland‐Altman plots comparing mean systolic AOBP with both awake ABP and mean systolic 24‐hour ABP among all participants are depicted in Figure 3A,B, respectively. In both plots, the bias remained positive (1.2 mm Hg with 95% CI 0.1‐2.4 mm Hg vs 5.5 mm Hg with 95% CI 4.4‐6.6 mm Hg for the comparison between mean systolic AOBP vs awake ABP and the comparison between mean systolic AOBP vs mean 24‐hour ABP, respectively); this is likely due to the higher number of individuals with systolic AOBP ≥ 130 mm Hg compared to those with AOBP < 130 mm Hg (298 vs 254 individuals, respectively).
FIGURE 3.
A, Bland‐Altman plot comparing mean systolic AOBP (mm Hg) vs mean systolic 24‐hour ABP (mm Hg) among all patients (N = 552). Solid lines, mean bias; dashed lines, 95% limits of agreement. B, Bland‐Altman plot comparing mean systolic AOBP (mm Hg) vs systolic awake ABP (mm Hg) among all patients (N = 552). Solid lines, mean bias; dashed lines, 95% limits of agreement
4. DISCUSSION
This study compared systolic AOBP readings <130 mm Hg with awake and 24‐hour ABP. Our findings showed that in patients with a systolic AOBP < 130 mm Hg, mean AOBP values were 6 mm Hg lower compared to the mean awake and 2 mm Hg lower compared to 24‐hour ABP. In contrast, in those individuals with a systolic AOBP ≥ 130 mm Hg, mean AOBP was 7 and 11 mm Hg higher than the mean awake ABP and 24‐hour ABP, respectively.
Comparison of OBP at the lower normal range of <130 mm Hg and awake ABP has previously been obtained in the SPRINT, 11 ACCORD (Action to Control Cardiovascular Risk in Diabetes) 12 and SPS3 (Secondary Prevention of Small Subcortical Strokes) studies. 13 These trials concluded that at this low normal OBP range office BP values were lower than those of the corresponding awake ABP, a finding confirmed by Myers obtaining data from 57 referrals for 24‐hour ABP compared with systolic AOBP < 130 mm Hg. 14 In the SPRINT ABPM sub‐study, 15 mean AOBP was 120 ± 13/66 ± 11 mm Hg compared with a mean awake ABP of 127 ± 12/72 ± 9 mm Hg. Interestingly, it should be emphasized that at a target systolic AOBP < 130 mm Hg, AOBP readings are lower than awake ABP, tending to become closer to the mean 24‐hour ABP, with a Bland‐Altman plot expressing a lower mean difference for mean systolic AOBP against mean systolic 24‐hour ABP and this may be due to the inclusion of the nocturnal BP readings. In contrast, in our subgroup of patients with a mean systolic AOBP ≥ 130 mm Hg, both mean systolic awake and 24‐hour ABP values seem to be lower than the systolic AOBP. This is also clearly demonstrated by Parati et al suggesting that clinic BP values in the higher range of distribution are higher than ABP values. 16 Moreover, as we previously noted, the marked WCE in the range of ≥130 mm Hg is not the case in the subgroup of patients with a systolic AOBP < 130 mm Hg.
AOBP has been preferred by the new AHA statement for use in clinical practice and is recognized by the ESC/ESH Guidelines as the technique that improves the reproducibility of BP measurement in the doctor's office. 2 , 17 Furthermore, AOBP is recommended by the Canadian Hypertension Education Program as the favored technique for office BP measurement. 18 Our novel finding further strengthens the implementation of AOBP in clinical practice since at this low cutoff‐point AOBP provides similar values not only to awake but also to mean 24‐hour ABP. A limitation to our study is that it involved a European population of patients attending a hospital clinic.
The AOBP technique should be used initially in every patient. If AOBP is <130/80 mm Hg in repeated office visits, ABP should be performed only when masked hypertension is suspected. When AOBP readings are higher than this threshold, especially in individuals with organ damage, measurements could be accepted for the diagnosis of uncontrolled hypertension.
AOBP produces values lower than the conventional office BP and similar to awake ABP. Our data showed that regardless of their treatment status patients with mean AOBP values <130/80 mm Hg have a lower AOBP value than the corresponding mean 24‐hour ABP. These findings advocate that AOBP can primarily be used for hypertension screening, and at these lower BP readings, ABP should be used when diagnosis of hypertension is suspected. At higher BP levels, especially in individuals with organ damage, AOBP can identify patients with hypertension. This new evidence could lead AOBP to a greater acceptance in clinical practice and in guidelines.
CONFLICT OF INTEREST
None.
ACKNOWLEDGMENTS
None.
Andreadis EA, Geladari CV, Angelopoulos ET. Automated office blood pressure is in agreement with awake and mean 24‐hour ambulatory blood pressure at the lower blood pressure range. J Clin Hypertens. 2020;22:1177–1183. 10.1111/jch.13927
REFERENCES
- 1. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13–e115. [DOI] [PubMed] [Google Scholar]
- 2. Williams B, Mancia G, Spiering W, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. J Hypertens. 2018;36(10):1953‐2041. [DOI] [PubMed] [Google Scholar]
- 3. Myers MG. The great myth of office blood pressure measurement. J Hypertens. 2012;30:1894‐1898. [DOI] [PubMed] [Google Scholar]
- 4. Myers MG, Kaczorwski J. Office blood pressure is lower than awake ambulatory blood pressure at lower targets for treatment. J Clin Hypertens. 2017;19:1210‐1213. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Colella T, Tahsinul A, Gatto H, Oh P, Myers MG. Antecedent rest may not be necessary for automated office blood pressure at lower treatment targets. J Clin Hypertens. 2018;20:1160‐1164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Andreadis EA, Geladari CV, Angelopoulos ET. Automated office blood pressure measurements obtained with and without preceding rest are associated with awake ambulatory blood pressure. J Clin Hypertens. 2020;22:32‐38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Myers MG. A proposed algorithm for diagnosing hypertension using automated office blood pressure measurement. J Hypertens. 2010;28:703‐708. [DOI] [PubMed] [Google Scholar]
- 8. Myers MG, Godwin M, Dawes M, et al. Conventional versus automated measurement of blood pressure in primary care patients with systolic hypertension randomized parallel design controlled trial. BMJ. 2011;342:d286. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ostchega Y, Nwankwo T, Sorlie PD, Wolz M, Zipf G. Assessing the Validity of the OmronHEM‐907XL oscillometric blood pressure measurement device in a national survey environment. J Clin Hypertens. 2010;12:22‐28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Stergiou GS, Giovas PP, Gkinos CP, Patouras JD. Validation of the microlife WatchBP home device for self home blood pressure measurement according to the international protocol. Blood Press Monit. 2007;12:185‐188. [DOI] [PubMed] [Google Scholar]
- 11. The SPRINT Research Group . A randomized trial of intensive versus standard blood pressure control. N Engl J Med. 2015;375:2103‐2116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. ACCORD Study Group . Effects of intensive blood pressure control in type 2 diabetes mellitus. N Engl J Med. 2010;362:1575‐1585. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. The SPS3 Study Group . Blood‐pressure targets in patients with recent lacunar stroke: the SPS3 randomized trial. Lancet. 2013;382:507‐515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Myers MG, Matangi M, Kaczorowski J. Comparison of awake ambulatory blood pressure and automated blood pressure using linear regression analysis in untreated patients in routine clinical practice. J Clin Hypertens. 2018;20:1696‐1702. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Drawz PE, Pajewski NM, Bates JT, et al. Effect of intensive versus standard clinic‐based hypertension management on ambulatory blood pressure. Results from the SPRINT Ambulatory Blood Pressure Study. Hypertens. 2017;69:42‐50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Parati G, Ochoa JE, Bilo G, Zanchetti A. SPRINT blood pressure Sprinting back to Smirk’s basal blood pressure? Hypertension. 2017;69:15‐19. [DOI] [PubMed] [Google Scholar]
- 17. Muntner P, Shimbo D, Carey RM, et al. Measurement of blood pressure in humans: a scientific statement from the American Heart Association. Hypertension. 2019;73:e35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Leung AA, Nerenberg K, Daskalopoulou SS, et al. CHEP Guidelines Task Force. Canadian Hypertension Education Program Guidelines for blood pressure measurement, diagnosis, assessment of risk, prevention, and treatment of hypertension. Can J Cardiol. 2016;2016(32):569‐588. [DOI] [PubMed] [Google Scholar]