Telemedicine home blood pressure measurements and progression of albuminuria in elderly people with diabetes

Abstract

We assessed whether home blood pressure monitoring improved the prediction of progression of albuminuria when added to office measurements and compared it with ambulatory blood pressure monitoring in a multiethnic cohort of older people (n=392) with diabetes mellitus, without macroalbuminuria, participating in the telemedicine arm of the Informatics for Diabetes Education and Telemedicine Study. Albuminuria was assessed by measuring the spot urine albumin:creatinine ratio at baseline and annually for 3 years. The ambulatory sleep:wake systolic blood pressure ratio was categorized as dipping (ratio: < or =0.9), nondipping (ratio: >0.9 to 1.0), and nocturnal rise (ratio: >1.0). In a repeated-measures mixed linear model, after adjustment that included office pulse pressure, home pulse pressure was independently associated with a higher follow-up albumin:creatinine ratio (P=0.001). That association persisted (P=0.01) after adjusting for 24-hour pulse pressure and nocturnal rise, which were also independent predictors (P=0.02 and P=0.03, respectively). Cox proportional hazards models examined the progression of albuminuria (n=74) as defined by cutoff values used by clinicians. After the adjustment for office pulse pressure, the hazards ratio (95% CI) per 10-mm Hg increment of home pulse pressure was 1.34 (range: 1.1 to 1.7; P=0.01). Home pulse pressure was not an independent predictor in the model including ambulatory monitoring data; a nocturnal rise was the only independent predictor (P=0.035). Cox models built separately for home pulse pressure and ambulatory monitoring exhibited similar calibration and discrimination. In conclusion, nocturnal blood pressure elevation was the strongest predictor of worsening albuminuria. Home blood pressure measurements added to office measurements and may constitute an adequate substitute for ambulatory monitoring.