Prevalence and clinical correlates of ambulatory blood pressure phenotypes in a Saudi hypertensive population

Abstract

International Guidelines recommend ambulatory blood pressure monitoring (ABPM) for the management of hypertension. ABPM phenotypes predict outcomes independent of office blood pressure (BP). The authors explored the prevalence and clinical correlates of ABPM phenotypes and relationship with office BP in Saudi patients (n = 428, mean age 53.5 ± 14.6, 55% male) referred to a Specialist Hypertension clinic in Riyadh, Saudi Arabia. ABPM phenotypes included sustained normotension (27%), masked hypertension, MHT(32%), sustained hypertension, SHT(52%), and white coat hypertension(2.6%). MHT was more prevalent using asleep than 24‐hours (26.4% vs 12.9%, P < .01) or awake BP (26.4% vs 8.5%, P < .001) and observed in 85% of pre‐hypertensive patients. Isolated nocturnal hypertension was more prevalent in MHT vs SHT (70% vs 30%, P < .001). Office BP overestimated control rates compared with ABPM (48% vs 12.9%, P < .001). Our study shows that one in three Saudi patients will be managed inappropriately if office BP alone was relied upon for management of hypertension.

1. INTRODUCTION

The accurate measurement of blood pressure (BP) is of vital importance to prevent either unnecessary treatment of normotensive individuals or underestimation of uncontrolled hypertension (HTN) in those receiving anti‐hypertensive treatment. Out‐of‐office BP measurement is a Class I recommendation with 24‐hours ambulatory blood pressure monitoring (ABPM) advocated as the gold standard for accurate diagnosis and treatment of hypertension (HTN). ¹ , ² ABPM phenotypes, especially masked and nocturnal HTN carry prognostic significance, independent of office BP. ³ HTN constitutes a major cardiovascular (CV) risk factor in the Middle East (ME) with prevalence estimates between 15% and 60% in the general population. ⁴ The recent PURE study reported prevalence of HTN and pre‐hypertension in Saudia Arabia at 36% and 40%, respectively. ⁵ Despite International guideline recommendations, there is paucity of data on ABPM in Saudi Arabia. The aim of this study was to explore the prevalence and determinants of ABPM phenotypes and relationship with office BP in a Saudi hypertensive population.

2. METHODS

We performed a cross‐sectional study in adult patients (n = 435) referred to a Specialist Hypertension Clinic, at King Abdul Aziz Cardiac Center, Riyadh, Saudia Arabia who underwent ABPM at the index visit. Patients with incomplete ABPM data (n = 15) were excluded, leaving 428 patients for analyses. Office BP measurements were performed in duplicate (Omron Model, HEM 705‐CP; Omron Corp.). ABPM was performed over 24 hours using oscillometric device (Spacelabs 90207, Spacelabs, Snowqualmie, Washington, USA) according to International protocols, ⁶ with readings every 20 minutes from 6 AM to 10 PM and 30 minutes from 10 PM to 6 AM. A patient diary was kept to define sleep/awake periods. An adequate study had at least 70% valid readings with a minimum of 21 during daytime and seven during nighttime. ⁶

2.1. Office blood pressure definition

Hypertension was diagnosed as office systolic BP >140 or diastolic BP >90, mm Hg and further categorized into optimal, normal, high‐normal, and HTN. ⁷ In treated patients, BP control was defined as systolic BP <140 and diastolic BP <90, mm Hg.

2.2. Ambulatory blood pressure definitions

Hypertension phenotypes were stratified into white coat hypertension (WHT); office systolic BP >140 or diastolic BP >90, mm Hg with awake systolic BP <135 and diastolic BP <85, mm Hg; MHT, (systolic BP <140 and diastolic BP <90, mm Hg with elevated awake BP ≥135/85 mm Hg (definition‐1), 24‐hours ambulatory BP ≥130/80 mm Hg (definition‐2) or asleep BP ≥ 120/70 mm Hg (definition‐3); sustained hypertension (SHT), office systolic BP >140 and/or diastolic BP >90, mm Hg and awake BP ≥ 135/85 and 24‐hours BP ≥ 130/80 mm Hg and sustained normotension(SNT) as office systolic BP ≤140 or diastolic BP ≤90 and awake BP <135/85 and 24‐hours BP <130/80. ⁶ Corresponding terms used in treated patients were white coat effect (WCE), masked uncontrolled hypertension (MUCH), sustained uncontrolled hypertension(SUHT), and controlled hypertension (CHT), respectively. INH was defined as asleep systolic BP ≥120 and/or diastolic BP ≥70 mm Hg and daytime BP <135/85 mm Hg. Nocturnal BP dipping was categorized as described previously. ⁸ Information on clinical characteristics was gathered from electronic medical records (BestCare^®).

Results were analyzed using JMP, version 13Pro (SAS for Windows, Cary, NC, USA). Data are expressed as means ± SD for continuous and percentages for categorical data. Differences between means were analyzed using chi‐square test for categorical and one‐way analysis of variance (ANOVA) for continuous data, followed by Tukey‐Kramer test for multiple comparisons when ANOVA demonstrated significant heterogeneity. Independent determinants of ABPM phenotypes were explored using multivariate nominal logistic regression. Graphs are plotted using Prism version 8.4.1 (GraphPad Software, San Diego, CA, USA). The study had institutional ethics committee permission, and procedures followed were in accordance with the principles of the Declaration of Helsinki.

3. RESULTS

Of 435 patients who underwent initial assessment, 428 patients had complete ABPM recordings (mean age 53.5 ± 0.70, 55% males, 78% treated). While there was no significant difference in age, treated individuals were more likely to be male (P < .01) heavier (P < .05), dyslipidemic (P < .0001), diabetic (P < .0001) with a history of CV (P < .001), and chronic kidney disease (P < .001). Compared with untreated, those on treatment had higher total and low‐density lipoprotein cholesterol (P < .01), fasting blood sugar (P < .0001), HbA₁C (P < .01), and serum creatinine (P < .0001) with significantly reduced high‐density lipoprotein (HDL) cholesterol (P < .01), and eGFR (P <.0001) (Table S1).

The prevalence of office BP categories was overall: 9.9%, 10.3%, 18.8%, and 61% with 13%, 11%, 26%, and 50% in untreated and 9.5%, 10.7%, 17.6%, and 62% in treated patients for optimal, normal, high‐normal, and hypertension, respectively. Relationship between office BP phenotypes and clinical characteristics is presented in Table S2. The prevalence of INH was significantly higher in normal and high‐normal BP (P < .001) with no difference in dipping. While BMI and serum creatinine increased and eGFR and HDL cholesterol decreased across office BP categories, there was no significant difference in age, gender, or CV risk factors.

Overall, the prevalence of ambulatory HTN was as follows: 24‐hours, 64%; awake, 52%; and asleep, 85%. Treated patients had a higher prevalence of 24‐hr ( 68% vs 49%, P < .001), awake (56% vs 39%, P < .01), and asleep HTN (86.5% vs 78%, P < .05) vs untreated. INH, observed in 21%, overall was more prevalent in untreated vs treated patients (29% vs 18.7%, P < .05). Table 1 shows the cohort stratified by ABPM phenotypes. The overall prevalence was 27%, 32%, 52% and 2.6%, and 17.7%, 30%, 49%, and 3% in untreated patients for SNT, MHT, SHT, and WHT, respectively. In treated patients, the prevalence was 9.6%, 26%, 62%, and 2.4% for CHT MUCH, SUHT, and WCE, respectively. The prevalence of MHT was significantly higher using asleep compared with 24‐hours (26.4% vs 12.9%, P < .01) or awake BP (26.4% vs 8.5%, P < .001). INH was more prevalent in MHT compared with SHT (67.7% vs 32.3%, P < .0001) irrespective of treatment. Some 86% exhibited blunted nighttime dipping, 50% of which comprised reverse dipping. SHT and MHT were associated with non‐dipping, compared with SNT and WHT (Table 1). In multivariate analysis, predictors of MHT included chronic kidney disease (P < .01), age (P < .05), and obesity (P < .05); SHT was associated with age (P < .001), obesity (P < .01), dyslipidemia (P < .05), and chronic kidney disease (P < .05) while WHT was associated with age only (P < .05). INH emerged as an independent predictor of MHT (P < .0001) when forced into the model replacing CKD.

TABLE 1.

Clinical characteristics of patients categorized according to ABPM phenotypes as defined in European Society of Hypertension 2013 Guidelines (n = 428, Mean ± SD)

	Group I(SNT) (n = 48)	Group II( MHT) (n = 115)	Group III (SHT) (n = 249)	Group IV(WHT) (n = 11)	P‐ value
Age (years)	57.9 ± 12	54.2 ± 14.3	52.6 ± 14.6	48.8 ± 23	.08*
Gender (M)%	11	26.2	55.8	2.6	.9
BMI (kg/m2)	28 ± 7II,III	30 ± 5I,IV	31 ± 6I,IV	28 ± 5II,III	<.001*
Smoking (%)	12.2	34.7	51	2	.60
Type 2 diabetes (%)	10.7	24	62	3.2	.60
Dyslipidemia (%)	8.9III	26.6	62	2.3III	<.05**
Previous CVD (%)	25.7	25.7	62	2.7	.71
CKD (%)	6.9	13.9	13.4	6.7	<.05**
No. of anti‐HTN drugs	1.4 ± 0.24III	1.9 ± 0.15	2.6 ± 0.1I	1.9 ± 0.5	<.0001*
S. creat (μmol/L)	67.5 ± 31III	73 ± 32	86 ± 34I,IV	68 ± 26III	<.01*
eGFR (mL/min)	98.5 ± 27III	92 ± 23III	83.2 ± 25I,II	94 ± 31	<.001*
T chol (mmol/L)	4.5 ± 1.4	4.4 ± 1.35	4.4 ± 1	4.6 ± 01.7	.90
HDL chol (mmol/L)	1.16 ± 0.3	1.14 ± 0.3	1.07 ± 0.3	1.13 ± 0.3	.21
LDL chol (mmol/L)	2.7 ± 0.86	2.6 ± 0.9	2.7 ± 1	2.4 ± 0.86	.64
TG (mmol/L)	1.4 ± 0.9	1.45 ± 1	1.6 ± 0.8	1.5 ± 0.76	.45
FBS (mmol/L)	5.8 ± 1.7	7.4 ± 6	7.7 ± 6	6.8 ± 2	.48
HbA1C (%)	6.4 ± 1.2	6.6 ± 1.7	6.9 ± 1.7	6.6 ± 1.3	.31
Office SBP (mm Hg)	120 ± 10II,III,IV	128.5 ± 9I,III,IV	155 ± 15I,II	148.5 ± 13I,II	<.0001*
Office DBP (mm Hg)	70.3 ± 8.4 II, III	76.7 ± 9.5 I, III	87.7 ± 13.5I, II	81 ± 10	<.0001*
24‐h SBP (mm Hg)	113 ± 7II,III	127.6 ± 8.7I,III	141 ± 14I,II	119 ± 4	<.0001*
24‐h DBP (mm Hg)	65.3 ± 6.7II,III	74.3 ± 8I,III	78.5 ± 12I,II,IV	67 ± 6.7III	<.0001*
Awake SBP (mm Hg)	115.5 ± 9 II,III	128 ± 11 I,III	142 ± 14.5 I,II	122.4 ± 5.6	<.0001*
Awake DBP (mm Hg)	66.8 ± 8II,III	73.8 ± 12I,III	78.8 ± 13.5I,II	69.4 ± 8.4	<.0001*
Asleep SBP (mm Hg)	109.6 ± 6.7II,III	127.4 ± 10I,III,IV	140.2 ± 14.5I,II,IV	113.4 ± 5.7III	<.001*
Asleep DBP (mm Hg)	62.3 ± 5.2II,III	74 ± 8.4I,III	78.5 ± 12.5I,II,IV	62.6 ± 6III	<.001*
INH(%)	0	67.7	32.3	0	<.0001**
Dipping (%)	4.6 ± 7.4II,III	‐0.06 ± 8.2	0.86 ± 8	7.2 ± 5.5III	<.001*

Abbreviations: ABPM, ambulatory blood pressure monitoring; BMI, body mass index; CKD, chronic kidney disease; CVD, cardiovascular disease; DPB, diastolic blood pressure; eGFR, estimated glomerular filtration rate; FBS, fasting blood sugar; HDL‐chol, high‐density lipoprotein cholesterol; HTN, hypertension; HTN, hypertension; INH, isolated nocturnal hypertension; LDL‐chol, low‐density lipoprotein cholesterol; S. creat, serum creatinine; SBP, systolic blood pressure; SD, standard deviation; T.Chol, total cholesterol; TG, triglycerides.

All values are shown as mean ± standard deviation or as a case number.

^* P‐value is based on one‐way ANOVA test.

^** P‐value is based on Pearson's chi‐square test. Groups with significant differences according to the Tukey‐Kramer test are depicted in superscript roman.

The prevalence of MHT increased through office BP categories: optimal, 44%; 67.4%, normal; and high‐normal, 82% (P < .0001) with opposite trend for SNT. Figure 1 shows the relationship between ABPM and office BP phenotypes categorized by treatment demonstrating high‐normal BP category with the highest prevalence of MHT. BP control rates were significantly lower using ABPM (12% vs 48%, P < .001) compared with office BP.

FIGURE 1.

Ambulatory blood pressure phenotypes across office blood pressure categories in untreated (Top) and treated patients (Bottom) (n = 428)

4. DISCUSSION

To the best of our knowledge, this is the first study in an Arab population to describe ABPM phenotypes. Nocturnal HTN emerged as the most prevalent phenotype in our population. Whereas office BP underestimated SHT in untreated, it overestimated CHT in treated individuals. The high prevalence of MHT observed in high‐normal office BP was largely attributed to INH. Compared with SUHT, individuals with MUCH received suboptimal treatment. The prevalence of WHT was low, irrespective of treatment. Finally, BP control rates using ABPM were lower compared with office BP criteria.

We observed MHT in 32% of our cohort, associated with high‐normal office BP, CKD, older age, and obesity, similar to SHT. While our findings confirm previous studies, ⁹ they are at odds with the only ME study showing a prevalence of MHT at 11%, associated with older age only. ¹⁰ MHT, but not SHT, was also associated with nocturnal HTN and INH. Although suboptimal prescribing in MUCH is ascribed to many factors, ⁹ therapeutic inertia may play an important role as these individuals present with high‐normal office BP, as observed in this study. These findings reiterate MHT to be a high‐risk ABPM phenotype comparable to SHT, with added risk of INH ¹¹ and suboptimal treatment.

The higher prevalence of nocturnal HTN, INH, and non‐dipping ⁸ compared with others, ³ , ¹¹ including recent studies from ME, ¹² may indicate the Arab population to have a different risk profile than other ethnicities. The low prevalence of WHT in our study is comparable with some ¹³ but not others, ⁶ including a study from ME with prevalence of 17%. ¹⁰ Recent evidence also suggests that the incremental risk of events with WHT may only be slightly higher than normotension. ¹⁴

With limited availability of ABPM in Saudi Arabia, home BP monitoring may be an alternative with newer cost‐effective devices providing nighttime BP recordings. While it is superior to office BP in the diagnosis of HTN, whether ABPM‐guided treatment would confer an outcome advantage is being investigated in the ongoing MASTER trial. ¹⁵

Our study has certain limitations. It has limited generalizability being single‐center study. The phenotypes were identified using a single ABPM with no data on hypertension‐mediated organ damage or outcome.

5. CONCLUSIONS

We have shown for the first time in a Saudi hypertensive population that one in three individuals will be treated inappropriately if office BP alone was relied upon. Identifying WHT, almost as benign as normotension and MHT, with almost similar prognosis as SHT can be accomplished only by ABPM. Indeed, future research comparing diagnostic accuracy and prognostic significance of ABPM to office BP is of vital importance in Saudi Arabia and the wider Middle East.

CONFLICT OF INTEREST

None.

AUHTOR CONTRIBUTIONS

AM conceptualized the hypothesis, analysed and interpreted the data, drafted the manuscript, and revised the manuscript critically for important intellectual content. RA, HAS, MA, AA, AAF, and SAH designed the work and revised the manuscript critically for important intellectual content. MAZ revised the manuscript critically for important intellectual content. All authors have read and approved the final version of the manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the paper are appropriately investigated and resolved.

Supporting information

Tab S1

Tab S2

Mahmud A, Alahaideb R, Alshammary H, et al. Prevalence and clinical correlates of ambulatory blood pressure phenotypes in a Saudi hypertensive population. J. Clin. Hypertens. 2020;22:2372–2376. 10.1111/jch.14059