Malka, Edmond S.. Ambulatory blood pressure measurements: reproducibility and predictive capability for serious adverse health outcomes. Retrieved from https://doi.org/doi:10.7282/T3CV4H2R
DescriptionBackground: Hypertension is present in approximately 25% of all adults in the US aged 20 to 74 years according to the National Health and Nutrition Examination Survey (NHANES 1999-2002). Ambulatory Blood Pressure (ABP) monitoring data contains far more information than is captured by a single office measurement and this additional information may be useful for diagnosis and treatment decisions. Objectives: There are three primary aims. The first aim is to evaluate the association between mean ABP values and various measures of ABP variability. The second aim is to assess the reproducibility of ABP measures of variability in a sample of patients treated in a specialty hypertension and nephrology practice. The third aim is to evaluate the association between ABP measures of BP variability and all cause mortality. Methods: The study sample consisted of subjects with ABP data collected at a specialty hypertension and nephrology practice as part of routine clinical assessments. The practice focuses primarily on the treatment of hypertension in adults and the clinical population is highly enriched with hypertensive subjects. The data cover all ABP sessions from 1984 to 2004 which met the data completeness criteria. There were 2015 subjects with a total of 2292 ABP monitor sessions that met these inclusion/exclusion criteria. Of these, 200 subjects had more than one ABP session. Aim one was evaluated by Pearson correlation coefficients. Aim two was evaluated by intracluster correlation coefficients using GEE methods and by the Bland-Altman method for assessment of reproducibility. Aim three was evaluated by Cox proportional hazards analyses. Results: For aim one, standard deviations (SD) as a measure of BP variability were highly correlated with their respective mean values, whereas coefficients of variations (CV) did not. Day-night differences, a measure of diurnal variation, were correlated with mean systolic (SBP), diastolic (DBP) and pulse (PP) pressure. For aim two, SD and CV showed good reproducibility over time. The same was true for day-night measures of diurnal variation. However, dipping status, defined as a greater than 10% drop in ABP from daytime to nighttime was not reproducible over time. For aim three, SD and CV for SBP, DBP and mean arterial pressure (MAP) was associated with an increased risk of all cause mortality. The same was true for mid-BP and PP. Increased SD and CV for heart rate (HR) were generally associated with a reduced risk of all cause mortality. Results were similar for an endpoint of combined cause specific mortality due to cardiovascular, cerebrovascular or renal disease. Conclusions: ABP measures of variability generally showed good reproducibility with the exception of dipping status, defined as day-night drop of 10% or more. Further study is warranted on the use of this common 10% cutoff value utilized to define dippers in the literature. Associations between ABP measures of variability and their respective mean values were consistent with the published literature, as was the association between these measures of variability and all cause mortality.