Blood pressure, frailty status, and all-cause mortality in elderly hypertensives; The Nambu Cohort Study

Hypertension is a major risk factor for cerebrovascular disease, and antihypertensive therapy reduces events and mortality. Prior guidelines (JSH 2014) recommended BP <140/90 mmHg for ages 65–74 and <150/90 mmHg for ≥75 years, while updated JSH 2019 set a target of <140/90 mmHg for those ≥75 years based on randomized trials. However, the optimal BP target in older adults with frailty is debated due to limited evidence and lack of clear frailty diagnostic criteria. Frailty, a state of increased vulnerability with reduced physiologic reserve, is linked to impaired activities of daily living, increased hospitalization, and worse life expectancy. Frail individuals often have more comorbidities, somewhat lower traditional cardiovascular risk factor levels, and lower BP levels that have been associated with poor prognosis. This study investigated how baseline systolic BP and frailty status relate to subsequent all-cause mortality among older hypertensive outpatients in the Nambu Cohort Study.

Background literature indicates a log-linear association between BP levels and cardiovascular mortality with a flatter slope in older adults, suggesting lower relative contribution of BP to events at advanced age. Trials informed guideline targets for older adults (JSH 2019). Frailty is characterized by decreased reserves and increased vulnerability; it is associated with higher comorbidity burden, reduced BP, lipids, and body weight compared to non-frail peers, and lower BP in frailty has been linked to worse outcomes. Evidence guiding antihypertensive therapy specifically in frail patients is limited, and standardized diagnostic criteria for frailty have been lacking, contributing to uncertainty in optimal BP targets for frail hypertensives.

Design and setting: Prospective outpatient cohort (Nambu Cohort Study) in the southern area of Okinawa, Japan, initiated in 2017. Population: Outpatients aged ≥65 years who could complete a frailty interview were consecutively registered. Exclusions: Patients with a Kihon Checklist (KCL) physical function section score of 5 (worst) and those without diagnosed hypertension. Final sample: 535 hypertensive patients allocated into four groups by initial systolic BP (<140 or ≥140 mmHg) and frailty status (frailty vs non-frailty [robust + pre-frail]): <140/Frailty (n=114), <140/Non-frailty (n=165), ≥140/Frailty (n=84), ≥140/Non-frailty (n=172). Measurements: Height and weight measured without shoes; BMI calculated kg/m². BP measured once, seated, after ~10 minutes rest, with an automatic monitor (HBP-9020, Omron) on either arm. Fasting venous blood samples analyzed for serum creatinine, lipid profile (total, LDL, HDL cholesterol), triglycerides, fasting glucose (hexokinase/G6PD), HbA1c (HPLC), complete blood count. Handgrip strength measured using a Smedley-type dynamometer. Frailty assessment: Kihon Checklist (25-item yes/no) administered by trained staff; frailty categorized as robust (0–3), pre-frail (4–7), frail (≥8). For analyses, robust and pre-frail were grouped as non-frail. Outcome and follow-up: All-cause mortality over a median follow-up of 41 (34–43) months; mortality rates calculated per 100 patient-years. Statistical analysis: Continuous variables summarized as median (IQR), compared by Kruskal–Wallis; categorical variables as percentages, compared by chi-square; normality assessed by Kolmogorov–Smirnov. Cox proportional hazards models estimated hazard ratios (HRs) with 95% confidence intervals (CIs) for combinations of systolic BP and frailty status, adjusting for age, sex, diabetes, dyslipidemia, obesity, chronic kidney disease, stroke, coronary artery disease, heart failure, and use of calcium channel blockers and renin–angiotensin system inhibitors. Interaction between systolic BP and frailty tested with an interaction term. Two-sided tests; p<0.05 significant. Ethics: Approved by Social Medical Corporation Yuaikai, Okinawa, Japan (H30R009); conducted per the Declaration of Helsinki.

• Cohort characteristics: 535 older hypertensive patients; median age 78 (71–84) years; 51% men; frailty prevalence 37% (robust 29%, pre-frail 34%). Median follow-up 41 (34–43) months. Deaths: 49.
• Baseline differences: Within the same systolic BP category, frail patients were older, more often female, and had higher prevalence of heart failure, stroke, peripheral arterial disease, and atrial fibrillation than non-frail patients. Frailty was associated with lower BP, BMI, handgrip strength, eGFR, serum lipids, hemoglobin, and hematocrit; uric acid, fasting glucose, HbA1c, and WBC counts were similar between frail and non-frail.
• Primary analysis (adjusted Cox models comparing to <140 mmHg/Non-frailty reference): • ≥140 mmHg/Non-frailty: HR 3.19 (95% CI 1.12–11.40) • <140 mmHg/Frailty: HR 4.72 (95% CI 1.67–16.90) • ≥140 mmHg/Frailty: HR 3.56 (95% CI 1.16–13.40)
• Independent effects model: Frailty vs non-frail associated with doubled risk of all-cause mortality (HR 2.00, 95% CI 1.01–4.05; p=0.0467). Systolic BP ≥140 mmHg vs <140 mmHg was not significantly associated with mortality (HR 1.17, 95% CI 0.65–2.11; p=0.6045).
• Mortality rates: Lowest in <140 mmHg/Non-frailty, followed by ≥140 mmHg/Non-frailty; highest in frail patients regardless of BP. Among frail patients, lower systolic BP was associated with higher mortality than higher systolic BP.
• Subgroup ≥75 years: Similar trends observed; interaction between systolic BP and frailty showed marginal significance (p=0.085).

Findings indicate that frailty is a dominant determinant of all-cause mortality in older hypertensive outpatients. Compared with non-frail patients with systolic BP <140 mmHg, frail patients exhibited substantially higher mortality irrespective of BP category, and non-frail patients with systolic BP ≥140 mmHg also had increased risk, suggesting that BP control below 140 mmHg is beneficial primarily among non-frail individuals. The lack of a significant independent association between systolic BP ≥140 mmHg and mortality after adjustment, alongside higher mortality at lower BP within frail patients, underscores the complex interplay between BP and physiologic reserve in advanced age. Biological mechanisms related to frailty, including inflammaging and reduced muscle mass and hemodynamic stability, may contribute to lower measured levels of traditional cardiovascular risk factors (BP, lipids, body weight) despite higher comorbidity burden. These results support considering frailty status when individualizing BP targets and antihypertensive therapy intensity in older adults.

In this prospective cohort of older hypertensive outpatients, frailty was associated with markedly higher all-cause mortality regardless of systolic BP, whereas non-frail patients with systolic BP <140 mmHg had the lowest mortality. Frailty may serve as a clinical marker to identify older hypertensive patients more or less likely to benefit from intensive BP control. The study supports incorporating frailty assessment into treatment decision-making for elderly hypertensives.