| Home | E-Submission | Sitemap | Contact Us
Korean J Intern Med > Volume 30(5); 2015 > Article
Shin, Park, Kim, Ihm, Kim, Kim, Pyun, Kim, Choi, and Kim: Discordance between ambulatory versus clinic blood pressure according to global cardiovascular risk group

Abstract

Background/Aims:

The detection of white coat hypertension (WCH), treated normalized hypertension, and masked hypertension (MH) is important to improve the effectiveness of hypertension management. However, whether global cardiovascular risk (GCR) profile has any effect on the discordance between ambulatory blood pressure (ABP) and clinic blood pressure (CBP) is unknown.

Methods:

Data from 1,916 subjects, taken from the Korean Multicenter Registry for ABP monitoring, were grouped according to diagnostic and therapeutic thresholds for CBP and ABP (140/90 and 135/85 mmHg, respectively). GCR was assessed using European Society of Hypertension 2007 guidelines.

Results:

The mean subject age was 54.1 ± 14.9 years, and 48.9% of patients were female. The discordancy rate between ABP and CBP in the untreated and treated patients was 32.5% and 26.5%, respectively (p = 0.02). The prevalence of WCH or treated normalized hypertension and MH was 14.4% and 16.0%, respectively. Discordance between ABP and CBP was lower in the very high added-risk group compared to the moderate added-risk group (odds ratio [OR], 0.649; 95% confidence interval [CI], 0.487 to 0.863; p = 0.003). The prevalence of WCH or treated normalized hypertension was also lower in the very high added-risk group (OR, 0.451; 95% CI, 0.311 to 0.655).

Conclusions:

Discordance between ABP and CBP was observed more frequently in untreated subjects than in treated subjects, and less frequently in the very high added-risk group, which was due mainly to the lower prevalence of WCH or treated normalized hypertension.

INTRODUCTION

Arterial blood pressure (BP) is regulated by neuronal, vascular, and renal mechanisms and can vary substantially according to temperature, physical activity, emotional stress, food intake, and the use of drugs including alcohol and caffeine. BP measured by ambulatory blood pressure monitoring (ABPM) is generally considered to be more effective at predicting cardiovascular events than the measurement in a clinic [1]. Studies have confirmed that for both treated and untreated hypertensive patients, ambulatory blood pressure (ABP) is a more powerful predictor of cardiovascular events than clinic blood pressure (CBP) [2-4].
In clinical practice, however, the most important application of ABPM is to detect white coat hypertension (WCH) and masked hypertension (MH). Diagnostic thresholds based on cardiovascular outcomes have been reported in many studies [2,3,5-7]. In the recently published United Kingdom National Institute for Clinical Excellence (NICE) guidelines, the routine use of ABPM for the diagnosis of hypertension was recommended [8], and therapeutic thresholds for ABPM were also suggested. The current thresholds for diagnosis and treatment are around 135/85 mmHg, and this same threshold is applied for all spectrums of global cardiovascular risk (GCR) groups determined by a risk stratification process [8].
The most important rationale for the routine use of ABPM in the NICE guidelines is to reduce the cost incurred by unnecessary treatment of WCH. However, the downstream cost driven by cardiovascular events resulting from neglected treatment of MH was not considered. Unlike WCH, the higher the GCR of a hypertensive subject, the greater the effectiveness of prevention related to the detection of MH.
The role of ABPM in high-risk patients might be different from that in low-risk patients [9]. In addition, the clinical implication of a discordant diagnosis; i.e., WCH or MH diagnosed by ABPM, can differ among GCR groups. Considering the absolute risk, detection of WCH and/or MH may be more cost-effective in the higher GCR group [10]. However, there have been few reports regarding the prevalence of such discordant diagnoses according to GCR group, or the clinical implications of the risk stratification process in terms of the clinical usefulness of ABPM.
In the present cross-sectional study, the association between GCR profile and ABP CBP discordance was investigated in a multicenter clinical cohort in Korea.

METHODS

Subjects

A total of 2,215 patients were enrolled in this study from 1 August 2009 to 31 December 2010. Of these patients, data from 1,916 subjects were obtained from the Korean ABPM Registry for Evaluation of the Prognostic Threshold in Hypertension (Kor-ABP) cohort study organized by the Korean Society of Hypertension. Twenty-seven referral hospitals participated in this study, 24 of which were affiliated with one of 18 medical schools, one was a veterans’ hospital, one hospital was affiliated with a Catholic foundation, and one was a medical insurance referral hospital. All investigators were cardiologists involved mainly in the field of hypertension and clinical cardiology.
A total of 2,215 patients that had undergone ABPM for the evaluation of high BP were included in this study; however, 299 patients with incomplete data, and/or a lack of informed consent were excluded. The minimum data required were demographic information, a clinical questionnaire, CBP, medical information from hospital records, and the raw ABPM data files.

Clinical and laboratory variables

The following demographic informations were included: age, gender, height, weight, abdominal circumference, smoking status, alcohol intake, extent of physical exercise, family history of hypertension and premature cardiovascular death, and past medical history including the occurrence of hypertension, diabetes, hyperlipidemia, stroke, myocardial infarction, heart failure, and cancer.
Informations taken from medical records included the presence of diabetes, hyperlipidemia, and cardiovascular diseases, as well as the time of diagnosis, mode of treatment, and prescribed medications for any of the above conditions.
Data from laboratory tests, where available, included complete blood cell count, chemistry, lipid profile, electrolytes, liver function, fasting glucose, fasting insulin, high-sensitivity C-reactive protein test, microalbuminuria, left ventricular hypertrophy measured by electrocardiogram voltage, chest X-ray, echocardiography, carotid ultrasonography, and pulse wave velocity.

CBP and ABPM data

CBP was measured using an A&D UA-767 (A&D Co., Ltd., Tokyo, Japan), which passed European Society of Hypertension (ESH) International Protocols at all institutes. CBP is defined as the average BP of two measurements taken 1 minute apart, with 5 minutes of rest before the first measurement.
ABPM data were gathered as raw data files from the website, or typed manually when raw data files were not available. The ABPM raw data were regarded as valid only when at least 14 awake BP readings, taken from 8:00 AM to 9:00 PM, were available after omitting erroneous readings according to the following criteria: (1) systolic BP > 250 or < 70 mmHg; (2) diastolic BP > 150 or < 40 mmHg; and (3) pulse pressure > 150 or < 20 mmHg [11].

Definition of risk groups

The 10-year cardiovascular event rate for the GCR groups was defined according to the 2003 ESH-European Society of Cardiology (ESC) guidelines for the management of arterial hypertension [12]. Subjects were assigned to GCR groups according to the 2007 guidelines for the management of arterial hypertension: the task force for the management of arterial hypertension of the ESH and the ESC [13]. The risk group was determined by adding the various risk factors; i.e., age over 55 years for males or over 65 years for females; dyslipidemia; fasting blood glucose between 102 and 125 md/dL; obesity, defined by abdominal circumference ³ 90 cm for males and ³ 80 cm for females, or body mass index (BMI) ≥ 25 kg/m2; smoking habit; and a family history of cardiovascular disease before the age of 55 years for males and 65 years for females. Dyslipidemia was defined by a history of dyslipidemia, the use of statins or other drugs for lipid abnormality, or diagnosis or data from medical records including either abnormal high density lipoprotein (HDL) or triglyceride levels, or total cholesterol ³ 190 mg/dL. Diabetes mellitus was defined by past history or current diagnosis [13]. Patients with three or more of the following: impaired fasting blood glucose without diabetes mellitus, abdominal obesity, triglyceride levels ³ 150 mg/dL, low HDL levels, BP ³ 130/85 mmHg, or a history of hypertension were considered to have metabolic syndrome [14].
GCR profiles were classified as groups with average, low added, moderate added, high added, and very high added risk [13]. Treated subjects with clinic systolic BP below 140 mmHg and diastolic BP below 90 mmHg were classified as grade 1 hypertensive.

Definition of thresholds for controlled hypertension

Controlled hypertension by CBP was defined as systolic BP below 140 mmHg and diastolic BP below 90 mmHg. Controlled hypertension by ABP was defined as a daytime systolic BP below 135 mmHg and a daytime diastolic BP below 85 mmHg. Subjects with controlled hypertension by ABP and uncontrolled hypertension by CBP were described as having ‘treated normalized hypertension’ [15]. The study protocol was approved by the Clinical Research Ethics Committees of all hospitals involved in the study.

Statistical analyses

All data are expressed as mean ± standard deviation. The statistical significance of the differences in the mean values was evaluated using analysis of variance. Chi-square and Fisher exact tests were used to determine the statistical significance of differences between groups. The various diagnostic categories were evaluated between CBP and ABP using McNemar test. Reliability analyses for inter-rater reliability and internal consistency using the κ test and Cronbach’s α were performed to determine the consistency of the two BP measurement methods.
Multiple logistic regression analysis was performed to examine the association between GCR group and the various BP categories and with incorrect categorization, by adjusting age, gender, clinic systolic BP, and antihypertensive medication. The group with moderate added risk was used as the reference point. Statistical significance was defined as a 95% confidence interval (CI) and a p < 0.05. All data processing and analyses were performed using SPSS version 20.0 (IBM Co., Armonk, NY, USA).

RESULTS

General characteristics of the study subjects

The age of the participants was 54.1 ± 14.9 years (n = 1,916), BMI was 24.7 ± 3.4 kg/m2, and 48.9% of patients were female. The clinic systolic BP was 142.9 ± 20.5 mmHg, and the clinic diastolic BP was 88.5 ± 14.6 mmHg. The clinic heart rate was 75.8 ± 13.8 beats per minute. ABPM devices used in this study were A&D (50.3%), Tonoport V (GE Medical Systems, Freiburg, Germany) and other GE devices (38.5%), Del Mar (7.6%; Avionics, Irvine, CA, USA), Spacelabs (3.5%; Spacelabs Medical, Issaquah, WA, USA), and Mobil O graph (0.1%; I.E.M., Stolberg, Germany). The indications for ABPM were the diagnosis of hypertension (59.3%), assessment of the efficacy of anti-hypertensive treatment (38.0%), symptoms of hypotension (1.6%), symptoms of autonomic dysfunction (0.8%), and diagnosis of pregnancy (0.3%). The period from midnight to 5:00 AM was defined as night. Daytime systolic BP was 136.3 ± 16.8 mmHg, and daytime diastolic BP was 85.7 ± 12.0 mmHg. Nighttime systolic and diastolic BPs were 127.6 ± 25.7 and 79.0 ± 13.5 mmHg, respectively. Heart rate was 73.3 ± 11.5 beats per minute in the day and 64.7 ± 4.7 beats per minute at night. The quality of sleep was good in 23.8% of patients, fair in 26.2%, bad in 28.4%, and very bad in 13.1% of patients. Measurement intervals were 15 (49.8%), 30 (40.8%), and 60 minutes (0.3%) in the daytime, and 15 (0.5%), 30 (77.2%), and 60 minutes (22.1 %) at night.
Among the study population, 14.3% of the subject were current smokers, 36.5% of the subject were current drinkers, 67.4% of the subject exercised less than three times per week, and 48.3%, 19.4%, and 3.2% of patients had family histories of hypertension, diabetes mellitus, and premature cardiac death, respectively.
Diabetes mellitus and dyslipidemia were present in 11.6% and 65.7% of patients, respectively. Coronary artery disease was present in 7.9% of patients and 8.7% of patients had a stroke previously. Metabolic syndrome could be assessed in only 910 subjects due to a lack of data, but 34.3% of these patients had metabolic syndrome without diabetes mellitus.
As shown in Table 1, the prevalences of optimal BP, normal BP, high normal BP, grade 1 hypertension, grade 2 hypertension, and grade 3 hypertension were 3.0%, 5.6%, 6.4%, 53.6%, 20.7%, and 10.7%, respectively. Only 7.0% (n = 136) of subjects had no cardiovascular risk factors, and only 3.5% of subjects had grade 1 or hypertension without risk factor.
As shown in Tables 1 and 2, 34.1% of patients were treated with antihypertensive medication. As shown in Table 2, the treated subjects were older, more obese, and had higher CBP and lower HDL levels. The prevalence of drinking and smoking, as well as cholesterol levels, were reduced in the treated group; however, a previous history of cardiovascular disease and a higher GCR profile were more frequent.

Global cardiovascular risk profile

As shown in Table 3, the percentage of patients with average risk, low added risk, moderate added risk, high added risk, and very high added risk was 1.5%, 8.9%, 32.1%, 32.7%, and 24.8%, respectively. Most of the parameters, with the exception of heart rate, varied as a function of risk category. The risk factors for cardiovascular disease increased as GCR increased.

Discordance between ABP and CBP in all subjects

For all subjects, WCH or treated normalized hypertension and MH were observed in 14.4% and 16.1% of patients, respectively. Concordant categories below and above diagnostic or therapeutic thresholds were observed in 23.7% and 45.8% of patients, respectively. Kappa was 0.360 (p < 0.001) with a 95% CI (0.316 to 0.403), and Cronbach’s α was 0.519 (p < 0.001) with a 95% CI (0.474 to 0.560). The distribution of these BP categories was significantly different between untreated and treated groups (p = 0.021), as shown in Fig. 1.

Discordance between ABP and CBP in untreated subjects

Of the 1,262 subjects (61.5%) not on antihypertensive medication, the mean age was 52.1 ± 15.3 years and 48.0% were female. Clinic systolic BP was 141.6 ± 19.7 mmHg, and clinic diastolic BP was 88.2 ± 14.1 mmHg. Daytime systolic and diastolic BPs were 135.4 ± 15.4 and 86.4 ± 11.9 mmHg, respectively, which were lower than clinic BPs (p < 0.001 for both systolic and diastolic BPs) (Table 2). The categorization of hypertension by clinic and daytime BP readings is shown in Fig. 1A. Kappa was 0.323 (p < 0.001) with a 95% CI (0.270 to 0.375), and Cronbach’s α was 0.499 (p < 0.001) with a 95% CI (0.441 to 0.552). The number of subjects with WCH was 188 (14.9%), and 222 subjects (17.6%) had MH, while 298 subjects (23.6%) were normotensive, and 554 subjects (43.9%) had sustained hypertension.

Discordance between ABP and CBP in treated hypertensive patients

The age of the 654 subjects (37.5%) on antihypertensive medication was 58.0 ± 13.2 years, and 50.5% were female. Clinic systolic BP was 144.6 ± 21.7 mmHg, and diastolic BP was 88.5 ± 15.2 mmHg. Daytime systolic and diastolic BPs were 136.3 ± 15.9 and 84.3 ± 11.4 mmHg, respectively, which were lower than the CBPs (p < 0.001 for both) (Table 2). Kappa was 0.434 (p < 0.001) with a 95% CI (0.363 to 0.504), and Cronbach’s α was 0.556 (p < 0.001) with a 95% CI (0.482 to 0.619). A total of 88 subjects (13.5%) showed a ‘treated normalized hypertension,’ and 85 subjects (13.0%) showed a reverse white-coat effect, or ‘masked’ uncontrolled hypertension. According to both CBP and ABP, 157 subjects (24.0%) were normotensive, and 324 subjects (49.5%) were hypertensive as shown in Fig. 1B.

Discordance between ABP and CBP according to GCR group

As shown in Table 4, the discordancy rate between ABP and CBP were different among the GCR groups. Kappa and Cronbach’s α coefficients in the very high added-risk group were significantly higher than those in the other three groups (Table 4). As shown in Table 5, in a multiple logistic regression analysis adjusted by age, gender, and antihypertensive medication status, the risk of WCH was significantly lower in the very high added-risk group than in the moderate added-risk group. However, the risk of MH was significantly higher in the low added-risk group than in the moderate added-risk group. The odds ratio for discordant diagnosis was significantly lower in the very high added-risk group compared to the moderate added-risk group when adjusted for age, gender, clinic SBP, and antihypertensive medication (Table 5).

DISCUSSION

In this study, the discrepancy rate between CBP and ABP was shown to be affected by patient GCR profiles. About one third of hypertensive patients would have been misdiagnosed had BP been measured using CBP alone. The likelihood of observing MH was higher in the low added and average risk groups compared to the moderate risk group. However, the likelihood of observing WCH was lower in the very high added-risk group. Overall, discordance decreased as the GCR profile of a patient increased. However, the finding that MH did not decrease in the high or very high added-risk group suggests that ABPM should still be performed in the very high added-risk group to prevent cardiovascular events or to reduce downstream costs incurred by treating MH. In this study, the moderate added-risk group was set as the reference group for the logistic regression analysis because, from a clinical viewpoint, most of the average or low added-risk subjects were likely to be seen in the general population, not in the clinic.
Considering the confidence interval in the κ statistics, the agreement between ABP and CBP was moderate and significantly higher in the very high added-risk group compared to the other groups. The reliability of the ABPM, as measured by the Cronbach’s α statistic, was acceptable only in the very high added-risk group. These findings were unexpected since BP variability could be higher in the very high added-risk group, which would make CBP unreliable. However, the emotional stabilization that occurs from multiple clinic visits may be an explanation for the findings in this group. Further studies are needed to investigate the precise role of ABPM in the very high added-risk group despite better agreement between ABP and CBP.
The discordance between ABP and CBP was higher in untreated subjects than in treated subjects. Compared to a previous study in Korea, the prevalence of MH was higher [16], whereas it was lower than that in a Spanish study [17]. These differences may be due to the use of different diagnostic criteria, and the use of daytime versus 24-hour BP measurements [18]. Considering the markedly higher prevalence of MH in the general population, a lower GCR profile or patient selection bias could affect the prevalence of MH [19].
Taking the above findings together, the use of ABPM it is expected to reduce cardiovascular events and downstream cost by detecting and treating MH, and by preventing the treatment of WCH and treated normalized hypertension. Considering the higher risk in the high added-risk group, the role of ABPM seems to be more important even though the prevalence of MH is comparable or lower than in the other GCR groups. Such an implication seems to be consistent with a previous study showing that cardiac damage was more frequent in MH than in WCH [20]. Therefore, ABPM could be the first step in detecting and managing MH in clinical practice. In one previous study the prevalence of WCH was reported to be around 15% to 50% [21], which is higher than that in the present study. Despite this, two other studies have reported similar WCH prevalences [19,22], and only one study has reported a lower prevalence (5.7%) [16]. The term ‘treated normalized hypertension’ has recently been adopted to describe patients with a CBP ≥ 140/90 mmHg and an ABP below < 135/85 mmHg, which, if only the CBP was considered, could be mistaken for uncontrolled BP [15]. The lower prevalence of treated normalized hypertension in the higher GCR groups, and concern regarding overtreatment or treatment-induced hypotension, should not result in inaction by physicians treating uncontrolled hypertension. However, in resistant hypertension, the white-coat effect is reported to be associated with myocardial ischemia, making it useful to perform ABPM before prescribing four or more antihypertensive drugs, or before an invasive procedure [23-25].
The clinical implication of the present study is the importance of ABPM in high-risk hypertension patients, not because of the prevalence of WCH, but because of the combination of sustained MH prevalence and higher attributable risk. In clinical practice, ABPM, as well as risk stratification, is helpful to prevent cardiovascular events and to reduce downstream costs incurred by undertreated hypertension.

Study limitations

The present study had several limitations. GCR was assessed mainly by anthropometry, questionnaires, and medical records from a registry database. Shortcomings included the lack of more detailed biochemical data, and the lack of target organ damage evaluation. The GCR may therefore have been underestimated or somewhat different from those in the guidelines [13]. However, because information from the questionnaires and the medical records was available for most subjects, this investigation may be more representative than smaller-scale studies with more detailed evaluation.
In this study, about 50% of patients were in groups with high or very high added risk, which may be higher than in other large-scale ABPM registries, such as the Spanish ABPM registry, because the subjects in this study were registered at referral hospitals rather than at primary clinics. Therefore, the ability to generalize the results of this study may be limited, especially in a primary care setting.
In conclusion, discordance between ABP and CBP was observed more frequently in untreated subjects than in treated subjects. Discordance was lower in the very high added-risk group compared to the lower risk groups, and this was due to decreased WCH or treated normalized hypertension, not decreased MH.

KEY MESSAGE

1. About 30% of the subjects visiting hypertension clinic are misdiagnosed without using ambulatory blood pressure monitoring regardless of the antihypetensive medication status.
2. Masked hypertension or masked uncontrolled hypertension are not decreased in the patient with high or very high risk profile.
3. Ambulatory blood pressure monitoring can be important tool to decrease the downstream cost of the misdiagnosis in hypertension management especially in high risk patients.

Conflict of interest

No potential conflict of interest relevant to this article was reported.

Acknowledgments

This study was supported and organized by the Korean Society of Hypertension and the Working Group on Blood Pressure monitoring in the Korean Society of Hypertension.
Principal Investigator: Soon Kil Kim
Investigators: Sang Hong Baek, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea; Sang Hyun Ihm, Yoon Seok Choi, Yeouido St. Mary’s Hospital, College of Medicine, The Catholic University of Korea; Byung-Ryul Cho, Kangwon National University Hospital; Woo-Shik Kim, Il Suk Sohn, Kyung Hee University School of Medicine; Jin-Won Jeong, Eun Mi Lee, Wonkwang University School of Medicine; Jang Hyun Cho, Saint Carollo Hospital; Wan Joo Shim, Korea University Anam Hospital; Young-Kwon Kim, Dongguk University Ilsan Hospital; Jin Ok Jeong, Chungnam National University School of Medicine; Sung Ha Park, Se-Joong Rim, Yonsei University College of Medicine; Soon Kil Kim, Jinho Shin, Hanyang University College of Medicine; Joong Hwa Jeong, Chosun University School of Medicine; Myeong-Chan Cho, Chungbuk National University Hospital; Ju Han Kim, Gye Hoon Kim, Chonnam National University Hospital; Gwang-il I Kim, Seoul National University College of Medicine; Dae-Gyun Park, Hallym University Kangdong Sacred Heart Hospital; Sahng Lee, Eulji University School of Medicine; Wook Bum Pyun, Ewha Womans University School of Medicine; Wan Kim, Gwangju Veterans Hospital; Se Jeong Yoon, National Health Insurance Corporation Ilsan Hospital; Seok Ku Choi, Inje University College of Medicine; Won Ho Kim, Chonbuk National University Medical School.

Figure 1.
Prevalence of hypertension categories according to antihypertensive medication status. (A) The left upper category indicates masked hypertension and the right lower category indicates white-coat hypertension. (B) The left upper category indicates masked uncontrolled hypertension and the right lower category indicates treated normalized hypertension. The distribution was different between the treated and untreated subjects.
kjim-30-5-610f1.tif
Table 1.
Distribution of subjects according to total cardiovascular risk profile
Variable Blood pressure, mmHg
Optimal (SBP < 120 and DBP < 80) Normal (SBP 120−129 or DBP 80−84) High normal (SBP 130−139 or DBP 85−89) Grade 1 HTN (SBP 140−159 or DBP 90−99) Grade 2 HTN (SBP 160−179 or DBP 100−109) Grade 3 HTN (SBP ≥ 180 or DBP ≥ 110)
No other risk factors 4 (0.2) 15 (0.8) 10 (0.5) 57 (3.0) 30 (1.6) 20 (1.0)
1−2 risk factors 26 (1.4) 40 (2.1) 48 (2.5) 376 (19.6) 159 (8.3) 80 (4.2)
3 or more risk factors, MS, OD, or diabetes 13 (0.7) 36 (1.9) 46 (2.4) 413 (21.6) 148 (7.7) 85 (4.4)
Established CV or renal disease 15 (0.8) 17 (0.9) 18 (0.9) 181 (9.4) 59 (3.1) 20 (1.0)
Total 58 (3.0) 108 (5.6) 122 (6.4) 1,027 (53.6) 396 (20.7) 205 (10.7)
Antihypertensive medication - - - 427 (22.2) 148 (7.7) 79 (4.1)

Values are presented as number (%). All treated hypertensive patients with blood pressure below grade 1 hypertension were categorized as grade 1 hypertension.

SBP, systolic blood pressure; DBP, diastolic blood pressure; HTN, hypertension; MS, metabolic syndrome; OD, organ damage; CV, cardiovascular.

Table 2.
Characteristics of the subjects according to antihypertensive medication treatment status
Characteristic Untreated (n = 1,262) Treated (n = 654) p value
Age, yr 52.1 ± 15.3 58.0 ± 13.2 < 0.001
Female sex, % 48.0 50.5 0.95
Body mass index, kg/m2 24.5 ± 3.4 25.1 ± 3.5 0.001
Clinic SBP, mmHg 141.6 ± 19.7 144.6 ± 21.7 0.002
Clinic DBP, mmHg 88.2 ± 14.1 88.5 ± 15.2 0.052
Daytime SBP, mmHg 135.4 ± 15.4 136.3 ± 15.9 0.441
Daytime DBP, mmHg 86.4 ± 11.9 84.3 ± 11.4 0.055
Fasting blood glucose, mg/dL 104.5 ± 28.5 109.3 ± 37.1 0.073
Total cholesterol, mg/dL 188.6 ± 40.2 181.8 ± 44.2 0.005
HDL, mg/dL 49.1 ± 13.5 46.6 ± 11.7 0.007
Triglyceride, mg/dL 139.8 ± 99.9 146.1 ± 119.3 0.144
Smoking, % 17.1 10.7 < 0.001
Drinking, % 39.5 17.2 0.004
Cardiovascular diseases history, % 11.4 25.2 < 0.001
Global cardiovascular risk groups, % < 0.001
 Average added risk 4.5 0
 Low added risk 18.1 2.9
 Moderate added risk 41.4 40.8
 High added risk 16.1 22.0
 Very high added risk 19.7 34.2

Values are presented as mean ± SD unless other indicated. p values were adjusted for age.

SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL, high density lipoprotein.

Table 3.
Clinical characteristics according to total cardiovascular risk group
Characteristic Average risk group (n = 29) Low added-risk group (n = 171) Moderate added-risk group (n = 614) High added-risk group (n = 627) Very high added-risk group (n = 475) p value
Age, yr 30.8 ± 12.0 48.2 ± 16.5 50.9 ± 14.8 57.1 ± 12.7 57.6 ± 14.4 < 0.001
Female sex, % 38 53 52 49 44 0.043
Clinic SBP, mmHg 123.1 ± 9.4 127.2 ± 15.2 142.1 ± 16.8 142.5 ± 17.2 151.1 ± 26.0 < 0.001
Clinic DBP, mmHg 75.5 ± 10.1 79.5 ± 10.2 88.4 ± 11.7 88.0 ± 12.5 93.2 ± 19.2 < 0.001
Heart rate, bpm 78.0 ± 16.4 76.5 ± 13.5 76.9 ± 14.1 75.1 ± 12.8 74.5 ± 14.2 0.041
Height, cm 170.6 ± 9.1 162.7 ± 8.9 163.2 ± 9.0 162.3 ± 9.6 162.6 ± 9.7 0.458
Weight, cm 61.4 ± 9.6 60.8 ± 10.7 65.0 ± 12.0 67.5 ± 12.0 66.7 ± 12.9 < 0.001
Body mass index, kg/m2 21.0 ± 2.1 22.8 ± 2.9 24.3 ± 3.3 25.5 ± 3.4 25.1 ± 3.6 < 0.001
Daytime SBP, mmHg 125.0 ± 11.3 129.5 ± 14.3 135.5 ± 16.8 136.3 ± 15.0 140.4 ± 19.0 < 0.001
Daytime DBP, mmHg 78.3 ± 9.99 83.6 ± 10.8 86.8 ± 11.8 84.8 ± 11.5 86.6 ± 12.9 < 0.001
Daytime heart rate, bpm 73.1 ± 7.6 73.7 ± 8.6 73.3 ± 9.6 73.7 ± 10.2 72.6 ± 11.3 0.541
Nocturnal SBP, mmHg 112.8 ± 12.1 120.8 ± 17.7 127.8 ± 35.4 126.7 ± 17.3 131.9 ± 22.1 < 0.001
Nocturnal DBP, mmHg 69.1 ± 7.9 76.1 ± 14.1 79.4 ± 13.6 78.7 ± 12.9 80.5 ± 14.1 < 0.001
Nocturnal heart rate, bpm 60.0 ± 8.7 64.3 ± 9.5 64.5 ± 10.9 65.1 ± 11.0 65.1 ± 11.6 0.138
24-Hour SBP, mmHg 121.5 ± 11.7 126.3 ± 14.3 132.4 ± 16.0 133.3 ± 14.5 137.9 ± 18.7 < 0.001
24-Hour DBP, mmHg 75.9 ± 7.0 80.7 ± 11.7 84.1 ± 11.7 83.2 ± 10.8 84.7 ± 12.6 < 0.001
24-Hour heart rate, bpm 69.7 ± 6.9 70.9 ± 8.2 70.6 ± 9.2 71.1 ± 9.9 70.3 ± 10.9 0.739
Antihypertensive medication, % - 2.9 29.2 39.4 46.9 < 0.001
Diabetes mellitus history, % - - 1.1 20.6 18.5 < 0.001
Dyslipidemia history, % - 3.5 21.1 49.1 26.3 < 0.001
Smoking, % - 3.2 28.1 41.0 27.7 < 0.001
Obesity, % - 16.4 34.4 56.1 47.2 < 0.001

Values are presented as mean ± SD unless other indicated.

SBP, systolic blood pressure; DBP, diastolic blood pressure.

Table 4.
Distribution of blood pressure subtypes according to total cardiovascular risk group
Variable Average or low added risk (n = 200) Moderate added risk (n = 614) High added risk (n = 627) Very high added risk (n = 475) p value
Normotension, % 51.3 19.6 21.4 21.2
White-coat or treated normalized HTN, % 5.5 20.1 15.4 9.7
Masked HTN, % 30.7 13.0 16.7 13.1
Sustained HTN, % 13.1 48.3 46.9 56.1 < 0.001
κa 0.216 (0.09–0.333) 0.291 (0.214–0.367) 0.314 (0.237–0.390) 0.480 (0.395–0.564)
Cronbach’s αa 0.386 (0.188–0.535) 0.428 (0.329–0.512) 0.481 (0.393–0.556) 0.630 (0.557–0.691)

HTN, hypertension,

a 95% confidence interval in the parenthesis.

Table 5.
Odds ratios of the discordant classification of blood pressure subtypes according to total cardiovascular risk group
Variable Average or low added risk (n = 200) Moderate added risk (n = 614) High added risk (n = 627) Very high added risk (n = 475)
White-coat or treated normalized HTN 0.203 (0.104–0.397) Reference 0.752 (0.556–1.018) 0.451 (0.311–0.655)
Masked HTNa 2.963 (2.000–4.389) Reference 1.247 (0.903–1.724) 0.965 (0.668–1.392)
Discordanceb 0.956 (0.671–1.361) Reference 0.945 (0.740–1.208) 0.649 (0.487–0.863)

Discordance was defined as white-coat hypertension, treated normalized hypertension, or masked hypertension.

HTN, hypertension.

a Adjusted for age, gender, and the status of antihypertensive medication.

b Adjusted for age, gender, clinic systolic blood pressure, and the status of antihypertensive medication.

REFERENCES

1. Rothwell PM. Limitations of the usual blood-pressure hypothesis and importance of variability, instability, and episodic hypertension. Lancet 2010;375:938–948.
crossref pmid
2. Boggia J, Li Y, Thijs L, et al. Prognostic accuracy of day versus night ambulatory blood pressure: a cohort study. Lancet 2007;370:1219–1229.
crossref pmid
3. Dawes MG, Coats AJ, Juszczak E. Daytime ambulatory systolic blood pressure is more effective at predicting mortality than clinic blood pressure. Blood Press Monit 2006;11:111–118.
crossref pmid
4. Dolan E, Stanton A, Thijs L, et al. Superiority of ambulatory over clinic blood pressure measurement in predicting mortality: the Dublin outcome study. Hypertension 2005;46:156–161.
crossref pmid
5. Mesquita-Bastos J, Bertoquini S, Polonia J. Cardiovascular prognostic value of ambulatory blood pressure monitoring in a Portuguese hypertensive population followed up for 8.2 years. Blood Press Monit 2010;15:240–246.
crossref pmid
6. Kikuya M, Hansen TW, Thijs L, et al. Diagnostic thresholds for ambulatory blood pressure monitoring based on 10-year cardiovascular risk. Circulation 2007;115:2145–2152.
crossref pmid
7. Thijs L, Hansen TW, Kikuya M, et al. The International Database of Ambulatory Blood Pressure in relation to Cardiovascular Outcome (IDACO): protocol and research perspectives. Blood Press Monit 2007;12:255–262.
crossref pmid
8. National Clinical Guideline Centre. Hypertension: The Clinical Management of Primary Hypertension in Adults: Update of Clinical Guidelines. London: NICE (National Institute of Clinical Excellence), 2011.

9. Taylor PJ. Evidence for key recommendation in NICE guidance on hypertension is poor. BMJ 2011;343:d6494.
crossref pmid
10. Verdecchia P, Angeli F, Cavallini C. Ambulatory blood pressure for cardiovascular risk stratification. Circulation 2007;115:2091–2093.
crossref pmid
11. Staessen J, Fagard R, Lijnen P, Thijs L, van Hoof R, Amery A. Ambulatory blood pressure monitoring in clinical trials. J Hypertens Suppl 1991;9:S13–S19.
crossref
12. European Society of Hypertension-European Society of Cardiology Guidelines Committee. 2003 European Society of Hypertension-European Society of Cardiology guidelines for the management of arterial hypertension. J Hypertens 2003;21:1011–1053.
pmid
13. Mancia G, De Backer G, Dominiczak A, et al. 2007 Guidelines for the management of arterial hypertension: the task force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens 2007;25:1105–1187.
crossref pmid
14. Park SH, Yoon JS, Won KC, Lee HW. Usefulness of glycated hemoglobin as diagnostic criteria for metabolic syndrome. J Korean Med Sci 2012;27:1057–1061.
crossref pmid pmc
15. Franklin SS, Thijs L, Hansen TW, et al. Significance of white-coat hypertension in older persons with isolated systolic hypertension: a meta-analysis using the International Database on Ambulatory Blood Pressure Monitoring in Relation to Cardiovascular Outcomes population. Hypertension 2012;59:564–571.
crossref pmid pmc
16. Hwang ES, Choi KJ, Kang DH, et al. Prevalence, predictive factor, and clinical significance of white-coat hypertension and masked hypertension in Korean hypertensive patients. Korean J Intern Med 2007;22:256–262.
crossref pmid pmc
17. Segura J, Banegas JR, Ruilope LM. Usefulness of ambulatory blood pressure monitoring (ABPM) in daily clinical practice: data from the Spanish ABPM registry. Clin Exp Pharmacol Physiol 2014;41:30–36.
crossref pmid
18. Banegas JR, Ruilope LM, de la Sierra A, et al. High prevalence of masked uncontrolled hypertension in people with treated hypertension. Eur Heart J 2014;35:3304–3312.
crossref pmid
19. Bobrie G, Clerson P, Menard J, Postel-Vinay N, Chatellier G, Plouin PF. Masked hypertension: a systematic review. J Hypertens 2008;26:1715–1725.
crossref pmid
20. Yoon HJ, Ahn Y, Park JB, et al. Are metabolic risk factors and target organ damage more frequent in masked hypertension than in white coat hypertension? Clin Exp Hypertens 2010;32:480–485.
crossref pmid
21. Celis H, Fagard RH. White-coat hypertension: a clinical review. Eur J Intern Med 2004;15:348–357.
crossref pmid
22. Park SJ, Park JB, Choi DJ, et al. Detection of masked hypertension and the ‘mask effect’ in patients with well-controlled office blood pressure. Circ J 2011;75:357–365.
crossref pmid
23. Modolo R, Ruggeri Barbaro N, de Faria AP, et al. The white-coat effect is an independent predictor of myocardial ischemia in resistant hypertension. Blood Press 2014;23:276–280.
crossref pmid
24. Myers MG. Pseudoresistant hypertension attributed to white-coat effect. Hypertension 2012;59:532–533.
crossref pmid
25. Vongpatanasin W. Resistant hypertension: a review of diagnosis and management. JAMA 2014;311:2216–2224.
crossref pmid
Editorial Office
101-2501, Lotte Castle President, 109 Mapo-daero, Mapo-gu, Seoul 04146, Korea
Tel: +82-2-2271-6792   Fax: +82-2-790-0993    E-mail: kaim@kams.or.kr
Copyright © 2018 The Korean Association of Internal Medicine. All rights reserved.
About |  Browse Articles |  Current Issue |  For Authors and Reviewers
powerd by m2community