The correlation between blood pressure and kidney function decline in older people: a registry-based cohort study

Abstract

Objectives: To examine the relation between static and dynamic blood pressure (BP) measurements and the evolution of kidney function in older people, adjusted for the presence of multimorbidity.

Design: Retrospective cohort study during a 10-year time interval (2002-2012) in three age strata of patients aged 60 and older.

Setting: Primary care registration network with 97 general practitioners working in 55 practices regularly submitting collected patient data.

Participants: All patients with at least one BP measurement in 2002 and at least four serum creatine measurements after 2002 (n=8636). A modified Charlson Comorbidity Index (mCCI) at baseline was registered. Change in systolic and diastolic BP (DBP) and pulse pressure (PP) from 2002 onwards was calculated. The relation between kidney function evolution and baseline BP and change in BP was examined using linear and logistic regression analysis.

Main outcome measures: The slope of the estimated glomerular filtration rate (eGFR, MDRD, Modification of Diet in Renal Disease equation) was calculated by the ordinal least square method. A rapid annual decline of kidney function was defined as ≥ 3 L/min/1.73 m(2)/year.

Results: Rapid annual decline of kidney function occurred in 1130 patients (13.1%). High baseline systolic BP (SBP) and PP predicted kidney function decline in participants aged 60-79 years. No correlation between baseline BP and kidney function decline was found in participants aged 80 years and older. An annual decline of ≥ 1 mm Hg in SBP and PP was a strong risk factor for a rapid annual kidney function decline in all age strata, independent of baseline BP and mCCI. A decline in DBP as also a strong independent predictor in participants aged 60-79 years.

Conclusions: The present study identified a decline in BP over time as a strong risk factor for kidney function decline in all age strata, adjusted for mCCI and baseline kidney function and BP.

Keywords: EPIDEMIOLOGY; GERIATRIC MEDICINE.

Figure 1

Prevalence of rapid annual decline in kidney function according to different blood pressure measurements.

Figure 2

Baseline blood pressure (BP) as a predictor of rapid annual decline in kidney function (≥3 mL/min/1.73 m²/year) (logistic regression) (n=8636). Adjusted for gender, age, baseline Charlson Comorbidity Index, baseline cardiovascular medication, time between the first and last estimated glomerular filtration rate (eGFR) measurement after 2002, baseline eGFR.

Figure 3

Prevalence (%) of change in systolic and diastolic blood pressures and pulse pressure in different age strata.

Figure 4

Correlation between change in blood pressure (BP) and a rapid annual decline in kidney function (≥3 mL/min/1.73 m²/year) (logistic regression) (n=7283). Adjusted for gender, age at baseline, baseline measurements (systolic or diastolic BP or pulse pressure), baseline Charlson Comorbidity Index (CCI), baseline cardiovascular medication, time between the first and last estimated glomerular filtration rate (eGFR) measurement after 2002, baseline eGFR. ^µAdjusted for interaction term ‘systolic BP change × baseline CCI’ (adjusted OR of the interaction term 1.2 (95% CI 1.0 to 1.3). *Adjusted for interaction term ‘pulse pressure change×baseline pulse pressure’ (adjusted OR of the interaction term 1.3 (95% CI 1.1 to 1.5).