Prevalence of resistant hypertension in 1810 patients followed up in a specialized outpatient clinic and its association with the metabolic syndrome

Abstract

We aimed to assess the prevalence of resistant hypertension (RH) in patients attending hypertension outpatient clinics and to identify risk factors for RH. We studied the medical records of the last visit of all patients (n = 1810; 40.4% males, age 56.5 ± 13.5 years) who attended at least once our hypertension outpatient clinic during the last decade. RH was defined as blood pressure (BP) > 140/90 mmHg in patients without diabetes or chronic kidney disease (or BP > 130/80 mmHg in patients with the latter diseases) despite treatment with full doses of three antihypertensive agents from different classes or controlled BP on four or more different antihypertensive agents. The prevalence of RH was 12.3%, whereas 22.2% of the patients had well-controlled hypertension and 65.5% had uncontrolled hypertension but were on less than three antihypertensive agents. Independent predictors of RH were age (risk ratio, RR = 1.08, 95% confidence interval, CI 1.05–1.12, p < 0.001), body mass index (RR = 1.06, 95% CI 1.00–1.13, p < 0.05) and the presence of the metabolic syndrome (MetS) (RR = 2.01, 95% CI 1.03–3.91, p < 0.05). Conclusions. RH is frequent in patients followed up in hypertension outpatient clinics. In addition to age and obesity, MetS appears to be associated with increased risk for RH. Clarification of the mechanisms underpinning the association between MetS and hypertension might reduce the prevalence of RH.

Key Words::

• Chronic kidney disease
• metabolic syndrome
• obesity
• resistant hypertension
• type 2 diabetes mellitus

Introduction

Resistant hypertension (RH) is defined as office blood pressure (BP) that remains above goal (i.e. > 140/90 mmHg in subjects without diabetes or chronic kidney disease (CKD) or > 130/80 mmHg in patients with the latter diseases) despite treatment with optimal doses of three different classes of antihypertensive agents, including a diuretic (1). Patients whose BP is controlled but who are receiving four or more different antihypertensive agents are also considered to have RH (1). Patients with RH appear to be at higher risk for cardiovascular events than patients with controlled hypertension (2).

There are limited data on the prevalence of RH (3,4). A recent US study based on data from the National Health and Nutrition Examination Survey (NHANES) reported that 8.9% of all hypertensive patients and 12.8% of treated hypertensive patients have RH (3). In a study from Spain, the prevalence of RH among treated hypertensive patients followed up in hypertension outpatient clinics was 14.8% (4).

Several risk factors for RH have been identified, including older age, obesity, type 2 diabetes mellitus (T2DM), CKD, black race, female sex and high baseline BP (1,3,5–8). Excessive dietary sodium intake, heavy alcohol consumption, treatment with BP-increasing agents (mainly non-steroidal inflammatory agents) and secondary causes of hypertension (particularly primary aldosteronism and obstructive sleep apnea) are also associated with RH (1,3,6–9). However, the independent contribution of these risk factors to the development of RH is unclear (3).

The aim of the present study was to assess the prevalence of RH in a large cohort of patients attending hypertension outpatient clinics and to identify risk factors for RH.

Methods

We studied the medical records of the last visit of all patients who attended at least once the hypertension outpatient clinic of our department during the last decade (2002–2011) (n = 1810; 40.4% males, age 56.5 ± 13.5 years).

Office BP measurements were performed according to current guidelines (10). Patients remained seated for 5 min before measurements, with their arm supported at heart level. Caffeine and alcohol consumption, cigarette smoking and exercise were discouraged on the day of examination. A calibrated mercury sphygmomanometer with an appropriately sized cuff was used. A minimum of two measurements were performed in each patient and the average was recorded.

In all patients, weight, height and waist circumference were measured. Body weight was measured with an analog scale and in light clothing; height was measured barefoot with a stadiometer. Body mass index (BMI) was calculated by dividing weight (in kg) by height squared (in m). The waist circumference was measured at the smallest circumference at the level of the umbilicus.

Blood samples were collected in the morning after an overnight fast. Serum glucose, total cholesterol, high-density lipoprotein cholesterol, triglycerides and creatinine levels were determined. Low-density lipoprotein cholesterol levels were calculated using Friedewald's formula (11). Glomerular filtration rate (GFR) was estimated using the Modification of Diet in Renal Disease (MDRD) equation (12).

Coexisting conditions, including T2DM, coronary heart disease (CHD), stroke, family history of hypertension and smoking were recorded. CKD was defined as eGFR < 60 ml/min/1.73 m². Diagnosis of the metabolic syndrome (MetS) was based on the definition proposed by the American Heart Association/National Heart, Lung and Blood Institute (13).

Patients were divided into three groups according to their office BP: (i) patients with RH (i.e. with BP > 140/90 mmHg in the absence of diabetes or CKD (or with BP > 130/80 mmHg in patients with the latter diseases) despite treatment with optimal doses of three different classes of antihypertensive agents, including a diuretic, or with controlled BP on four or more different antihypertensive agents); (ii) patients with uncontrolled hypertension (i.e. with BP higher than the above limits but while receiving less than three different antihypertensive agents); and (iii) patients with well-controlled hypertension (i.e. with BP lower than the above limits).

Statistical analysis

All data were analyzed using the statistical package SPSS (version 17.0; SPSS, Chicago, IL, USA). Data are presented as mean and standard deviation. Differences in categorical variables between groups were assessed with the chi-square test. Differences in continuous variables between groups were assessed with one-way analysis of variance and post hoc tests were carried out with the Bonferroni test. Independent predictors of RH were assessed with stepwise logistic regression analysis including all variables that were significantly different between groups in univariate analysis. In all cases, a two-tailed p < 0.05 was considered significant.

Results

The prevalence of RH was 12.3%. In addition, 22.2% of the patients had well-controlled hypertension and 65.5% of the patients had uncontrolled hypertension but were on less than three antihypertensive agents. Only 2.5% of the study population was not receiving any antihypertensive medication.

The characteristics of patients with resistant, well-controlled and uncontrolled hypertension are shown in Tables I and II. Patients with RH had higher systolic (SBP) and diastolic BP (DBP) than patients with well-controlled hypertension. SBP did not differ between patients with resistant and uncontrolled hypertension but the former had higher DBP than the latter.

Table I. Demographic characteristics of the study population [data are percentages except age and package-years (mean ±SD)].

	Patients with well-controlled hypertension (n = 402)	Patients with uncontrolled hypertension (n = 1,185)	Patients with resistant hypertension (n = 223)	p
Age (years)	55.3 ± 14.4	55.5 ± 13.4	63.8 ± 10.4	< 0.001*
Males	37.8	41.9	36.7	NS
Smoking	23.7	25.3	26.0	NS
Package-years	16 ± 22	11 ± 20	18 ± 22	NS
Type 2 diabetes mellitus	5.1	7.6	15.9	< 0.001
Metabolic syndrome	12.5	16.1	26.4	< 0.001
Normal weight/overweight/obese	23.4/38.3/38.3	15.1/41.9/43.0	6.6/28.6/64.8	< 0.001
Family history of hypertension	22.3	21.9	21.6	NS
Coronary heart disease	6.4	6.8	11.5	< 0.05
Chronic kidney disease	8.3	20.4	23.3	< 0.001
Stroke	4.0	4.3	6.3	NS

NS, not significant. *Post hoc pair-wise comparisons: resistant vs well-controlled, p < 0.001; resistant vs uncontrolled, p < 0.001; uncontrolled vs well-controlled, p = NS.

Table II. Clinical and laboratory characteristics of the study population (data are mean± SD).

	Patients with well-controlled hypertension (n = 402)	Patients with uncontrolled hypertension (n = 1185)	Patients with resistant hypertension (n = 223)	p (overall)	p (for post hoc pair-wise comparisons)
					Resistant vs well-controlled	Resistant vs uncontrolled	Uncontrolled vs well-controlled
SBP (mmHg)	123 ± 9	154 ± 19	156 ± 23	< 0.001	< 0.001	NS	< 0.001
DBP (mmHg)	76 ± 8	93 ± 12	89 ± 14	< 0.001	< 0.001	< 0.001	< 0.001
Heart rate	73 ± 9	76 ± 11	75 ± 11	< 0.001	NS	NS	< 0.001
BMI (kg/m^2)	29.2 ± 6.1	30.1 ± 5.5	31.9 ± 4.9	< 0.005	0.001	< 0.05	NS
Waist (cm)	99 ± 12	100 ± 13	104 ± 15	NS	NA	NA	NA
LDL-C (mg/dl)	141 ± 50	152 ± 57	147 ± 60	NS	NA	NA	NA
HDL-C (mg/dl)	52 ± 13	51 ± 14	50 ± 15	NS	NA	NA	NA
TG (mg/dl)	141 ± 75	157 ± 87	159 ± 75	NS	NA	NA	NA
Glucose (mg/dl)	101 ± 24	106 ± 33	120 ± 38	< 0.001	< 0.001	< 0.001	NS
eGFR (ml/min/1.73 m^2)	79.9 ± 20.4	79.4 ± 24.6	76.4 ± 24.2	NS	NA	NA	NA

NS, not significant; NA, not applicable; SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglycerides; eGFR, estimated glomerular filtration rate.

Regarding risk factors for RH, patients with RH were older and had higher BMI and waist circumference than patients with both well-controlled and uncontrolled hypertension. The prevalence of T2DM, MetS, obesity, CKD and CHD was also higher in patients with RH than in the two other groups. In multivariate analysis, independent predictors of RH were age (risk ratio, RR = 1.08, 95% confidence interval, CI 1.05–1.12, p < 0.001), BMI (RR = 1.06, 95% CI 1.00–1.13, p < 0.05) and the presence of MetS (RR = 2.01, 95% CI 1.03–3.91, p < 0.05).

Discussion

In our population, the prevalence of RH was 12.3%, i.e. higher than the rate reported in a recent US study and lower than the rate reported in a Spanish study (8.9% and 14.8%, respectively) (3,4). This discrepancy probably results from differences in the study population; the US study assessed the prevalence of RH in the general US population, the Spanish study evaluated patients with ambulatory BP measurements who were followed up both in primary care and in specialized centers, whereas we assessed patients attending a specialized hypertension outpatient clinic (3,4). It should also be mentioned that both the US and Spanish studies used the 140/90 mmHg BP cut-off for defining RH in patients with T2DM or CKD instead of the recommended 130/80 mmHg cut-off (3,4). In addition, 14.4% of the patients in the US study considered to have RH were not receiving diuretics (3). In contrast, we applied a strict definition of RH; we used the 130/80 mmHg cut-off for patients with T2DM or CKD and all patients had to be treated with a diuretic to be considered to have RH. Moreover, the Spanish study included only treated hypertensive patients (4), whereas we evaluated both treated and untreated patients.

A novel finding of the present study is the independent association between MetS and RH. This observation might have important implications because MetS is common, currently affecting 20–30% of the adult population worldwide, and its prevalence continues to increase, primarily driven by the rising prevalence of obesity (14–16). In addition, several studies showed that MetS is associated with increased risk for cardiovascular disease (CVD) (17,18). Therefore, the rising rates of MetS might result in higher rates of RH and CVD events. The association between MetS and RH might be partly due to the higher prevalence of well-known risk factors for RH in patients with MetS, primarily obesity, T2DM and CKD (19,20). However, in our study, MetS independently predicted RH after adjusting for these risk factors. Therefore, additional abnormalities present in MetS might also contribute to the higher prevalence of RH in these patients. It is possible that insulin resistance, oxidative stress, endothelial dysfunction, the altered regulation of adipokines, and the proinflammatory and procoagulant milieu characterizing MetS might be involved (19,21,22). On the other hand, multiple neuroendocrine mediators, including neuropeptide Y and alpha-melanocyte stimulating hormone, have been implicated in the association between obesity and hypertension and might also play a role in the pathogenesis of hypertension in obese patients with MetS (23,24).

In addition to MetS, older age and higher BMI were also associated with higher risk for RH. These findings are in agreement with previous studies (3,5,6). The prevalence of other previously reported risk factors for RH, including T2DM, CKD and CHD was also higher in patients with RH (3,5,6). However, these factors were not independently associated with RH in multivariate analysis. Since patients with MetS have a higher prevalence of these comorbidities (18,19), it is possible that when MetS is added in the multivariate model, the association of T2DM, CKD and CHD with RH is no longer significant.

There are some limitations in the present study. First, we evaluated patients attending a specialized outpatient clinic and our results might not be applicable to the general hypertensive population. Second, we studied an exclusively Caucasian population and we cannot evaluate race differences in the prevalence of RH, which have been reported in previous studies (3). Finally, and similarly to previous studies (3), we did not assess adherence to medication and the possibility of the white-coat effect, and this might have resulted in overestimation of the prevalence of RH. On the other hand, the reported prevalence of RH might likely be an underestimate, because a proportion of patients with uncontrolled hypertension while using less than three antihypertensive agents might remain uncontrolled even after the administration of three agents.

In conclusion, RH is quite frequent in patients followed up in hypertension outpatient clinics. In addition to age and obesity, MetS appears to be associated with increased risk for RH. Given the aging of the population and the current epidemic of obesity, the prevalence of RH is expected to increase. Therefore, lifestyle interventions, appropriate antihypertensive pharmacotherapy and improvements in adherence to treatment are needed to reach BP goals (25) and prevent CVD events in hypertensive patients.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.