Prevalence and clinical characteristics of isolated-office and true resistant hypertension determined by ambulatory blood pressure monitoring.

Hypertension is defined as resistant to treatment when a therapeutic plan including ≥3 hypertension medications failed to sufficiently lower systolic (SBP) and diastolic (DBP) blood pressures (BPs). Most individuals, including those under hypertension therapy, show a "white-coat" effect that could cause an overestimation of their real BP. The prevalence and clinical characteristics of "white-coat" or isolated-office resistant hypertension (RH) has always been evaluated by comparing clinic BP values with either daytime home BP measurements or the awake BP mean obtained from ambulatory monitoring (ABPM), therefore including patients with either normal or elevated asleep BP mean. Here, we investigated the impact of including asleep BP mean as a requirement for the definition of hypertension on the prevalence, clinical characteristics, and estimated cardiovascular (CVD) risk of isolated-office RH. This cross-sectional study evaluated 3042 patients treated with ≥3 hypertension medications and evaluated by 48-h ABPM (1707 men/1335 women), 64.2 ± 11.6 (mean ± SD) yrs of age, enrolled in the Hygia Project. Among the participants, 522 (17.2%) had true isolated-office RH (elevated clinic BP and controlled awake and asleep ambulatory BPs while treated with 3 hypertension medications), 260 (8.6%) had false isolated-office RH (elevated clinic BP, controlled awake SBP/DBP means, but elevated asleep SBP or DBP mean while treated with 3 hypertension medications), and the remaining 2260 (74.3%) had true RH (elevated awake or asleep SBP/DBP means while treated with 3 medications, or any patient treated with ≥4 medications). Patients with false, relative to those with true, isolated-office RH had higher prevalence of microalbuminuria and chronic kidney disease (CKD), significantly higher albumin/creatinine ratio (p < .001), significantly higher 48-h SBP/DBP means by 9.6/5.3 mm Hg (p < .001), significantly lower sleep-time relative SBP and DBP decline (p < .001), and significantly greater prevalence of a non-dipper BP profile (96.9% vs. 38.9%; p < .001). Additionally, the prevalence of the riser BP pattern, which is associated with highest CVD risk, was much greater, 40.4% vs. 5.0% (p < .001), among patients with false isolated-office RH. The estimated hazard ratio of CVD events, using a fully adjusted model including the significant confounding variables of sex, age, diabetes, chronic kidney disease, asleep SBP mean, and sleep-time relative SBP decline, was significantly greater for patients with false compared with those with true isolated-office RH (2.13 [95% confidence interval: 1.95-2.32]; p < .001). Patients with false isolated-office hypertension and true RH, however, were equivalent for the prevalence of obstructive sleep apnea, metabolic syndrome, obesity, diabetes, microalbuminuria, and chronic kidney disease, and they had an equivalent estimated hazard ratio of CVD events (1.04 [95% confidence interval: .97-1.12]; p = .265). Our findings document a significantly elevated prevalence of a blunted nighttime BP decline in patients here categorized as either false isolated-office RH and true RH, jointly accounting for 82.8% of the studied sample. Previous reports of much lower prevalence of true RH plus a nonsignificant increased CVD risk of this condition compared with isolated-office RH are misleading by disregarding asleep BP mean for classification. Our results further indicate that classification of RH patients into categories of isolated-office RH, masked RH, and true RH cannot be based on the comparison of clinic BP with either daytime home BP measurements or awake BP mean from ABPM, as so far customary in the available literature, totally disregarding the highly significant prognostic value of nighttime BP. Accordingly, ABPM should be regarded as a clinical requirement for proper diagnosis of true RH.

Circadian pattern of ambulatory blood pressure in hypertensive patients with and without type 2 diabetes.

There is strong association between diabetes and increased risk of end-organ damage, stroke, and cardiovascular disease (CVD) morbidity and mortality. Non-dipping (<10% decline in the asleep relative to awake blood pressure [BP] mean), as determined by ambulatory BP monitoring (ABPM), is frequent in diabetes and consistently associated with increased CVD risk. The reported prevalence of non-dipping in diabetes is highly variable, probably due to differences in the study groups (normotensive subjects, untreated hypertensives, treated hypertensives), relatively small sample sizes, reliance only on a single, low-reproducibility, 24-h ABPM evaluation per participant, and definition of daytime and nighttime periods by arbitrary selected fixed clock-hour spans. Accordingly, we evaluated the influence of diabetes on the circadian BP pattern by 48-h ABPM (rather than for 24 h to increase reproducibility of results) during which participants maintained a diary listing times of going to bed at night and awakening in the morning. This cross-sectional study involved 12 765 hypertensive patients (6797 men/5968 women), 58.1 ± 14.1 (mean ± SD) yrs of age, enrolled in the Hygia Project, designed to evaluate prospectively CVD risk by ABPM in primary care centers of northwest Spain. Among the participants, 2954 (1799 men/1155 women) had type 2 diabetes. At the time of study, 525/3314 patients with/without diabetes were untreated for hypertension, and the remaining 2429/6497 patients with/without diabetes were treated. Hypertension was defined as awake systolic (SBP)/diastolic (DBP) BP mean ≥135/85 mm Hg, or asleep SBP/DBP mean ≥120/70 mm Hg, or BP-lowering treatment. Hypertensive patients with than without diabetes were more likely to be men and of older age, have diagnoses of microalbuminuria, proteinuria, chronic kidney disease, obstructive sleep apnea, metabolic syndrome, and/or obesity, plus higher glucose, creatinine, uric acid, and triglycerides, but lower cholesterol and estimated glomerular filtration rate. In patients with diabetes, ambulatory SBP was significantly elevated (p < .001), mainly during the hours of nighttime sleep and initial hours after morning awakening, independent of presence/absence of BP-lowering treatment. Ambulatory DBP, however, was significantly higher (p < .001) in patients without diabetes, mainly during the daytime. Differing trends for SBP and DBP between groups resulted in large differences in ambulatory pulse pressure (PP), it being significantly greater (p < .001) throughout the entire 24 h in patients with diabetes, even after correcting for age. Prevalence of non-dipping was significantly higher in patients with than without diabetes (62.1% vs. 45.9%; p < .001). Largest difference between groups was in the prevalence of the riser BP pattern, i.e., asleep SBP mean greater than awake SBP mean (19.9% vs. 8.1% in patients with and without diabetes, respectively; p < .001). Elevated asleep SBP mean was the major basis for the diagnosis of hypertension and/or inadequate BP control among patients with diabetes; thus, among the uncontrolled hypertensive patients with diabetes, 89.2% had nocturnal hypertension. Our findings document significantly elevated prevalence of a blunted nocturnal BP decline in hypertensive patients with diabetes. Most important, prevalence of the riser BP pattern, associated with highest CVD risk among all possible BP patterns, was more than twice as prevalent in diabetes. Patients with diabetes also presented significantly elevated ambulatory PP, reflecting increased arterial stiffness and enhanced CVD risk. These collective findings indicate that diabetes should be included among the clinical conditions for which ABPM is recommended for proper CVD risk assessment.