Seventy-five percent of adults aged >75 years have hypertension.1-3 According to the Joint National Commission 8 (JNC 8), the recommended target blood pressure (BP) is < 150/80 mm Hg for adults aged > 60 years.4 In 2016 the Systolic Blood Pressure Intervention Trial (SPRINT) suggested that more aggressive BP control with a goal of < 120/80 mm Hg reduced rates of cardiovascular disease and lowered the risk of death in adults aged > 50 years with hypertension.5 It is anticipated that as a result of the landmark SPRINT results, clinicians may attempt to treat hypertension more intensely in older patients with an increased risk of adverse consequences if BPs are not appropriately measured.
There is a standardized protocol for BP measurement, but these recommendations typically are not followed in routine office visits.6,7 Some studies have noted that home BP measurement may be more accurate than office measurement.8 However, clinicians may not always trust the accuracy of home BP readings, and many patients are not adherent with home measurement. As a result, physicians usually manage hypertension in older patients based on office readings, though it is likely that most office measurements do not follow protocol on proper measurement. Office measurements have been noted to be inaccurate with high likelihood of overestimating or underestimating BP control.9
Office BP measurements demonstrate poor correlation with home measurements and have not been shown to be as good of a predictor for target organ damage or long-term cardiovascular outcomes compared with that of home measurements.10,11 Although there have been studies comparing home and office BP measurements and comparing office and ambulatory BP measurement, no literature has been found that reports on the difference between routine office and standardized measurement of BP.9,12-14
This study seeks to identify the magnitude of difference among BP measured according to a standardized protocol, routine clinical, and home BP. The authors hypothesized that there would be a significant, clinically relevant difference among the 3 BP measurement methods, especially between the routine office and standardized office measurements. This study has implications for implementing intensive treatment of hypertension based on office measurements.
Participants included 30 male veterans aged > 65 years who were actively participating in the Gerofit program at the VA Greater Los Angeles Healthcare System (VAGLAHS). The Gerofit program is a model clinical demonstration exercise and health promotion program targeting older and veterans at risk for falls or institutionalization. Gerofit was established in 1987 at the Durham VA Health System and successfully implemented in 2014 at VAGLAHS. Supervised exercise is offered 3 times per week and consists of individually tailored exercises aimed at reducing functional deficits that are identified and monitored by an initial and quarterly functional assessment. Blood pressures are checked routinely once a week as a part of the program. Gerofit was reviewed and approved by the institutional review board at VAGLAHS as a quality improvement/quality assurance project.
Routine office and standardized protocol measurements were obtained by a single CasMED 740 (Branford, CT) automated BP machine and were conducted separately on different days. The CasMED 740 machine was not otherwise calibrated; however, a one-time correlation was performed between the CasMED 740 and the home BP monitor for each participant, when it was brought to VAGLAHS. Two measurements were made with the CasMed 740 automated BP machine on the arm that gave the higher BP reading throughout the standardized and routine protocol. Two subsequent measurements were made with the participant’s home automated BP cuff. Averages for the CasMED 740 and the home BP monitoring device were compared and assessed for significance by paired t test. No rest was scheduled prior to the first measurement, but there was a 1-minute rest after each subsequent measurement.
Mean values (SD) were used for participant characteristics and mean values (standard error [SE]) were used for BP measurements. Data were analyzed using Microsoft Excel (Redmond, WA) and GraphPad Prism version 7.03 (San Diego, CA). T tests were used for analysis of home BP measurements due to low sample size. Values of P < .05 were considered to be statistically significant.
Routine office protocol. Automated BP was measured to mimic routine office visits. Upon arrival, participants sat down, and the BP cuff was placed around their arm. Any rest before a measurement was incidental and not intentionally structured. Appropriate cuff size was determined by visual estimation of arm circumference. Only 1 measurement was made unless BP was > 150/90 mm Hg, in which case a repeat measurement was made after 2 to 4 minutes of rest. The BP was then determined based on the average of 2 or more readings. The BPs were recorded by hand in a weekly log. Participants had at least 12 weeks of BP readings measured by the routine method, and these BPs were averaged over 12 weeks to yield their average routine measured BP.
Standardized protocol. Automated BP was measured according to the 2015 USPSTF Guidelines and Look AHEAD trial protocol.7,15 A participant’s arm circumference was measured, and appropriate cuff size was determined. The participant rested quietly in a chair for at least 5 minutes with feet flat on the floor and back supported. The cuff was snugly placed 2 to 3 cm above the antecubital fossa, and the arm was supported at the level of the right atrium during the measurement. Blood pressure was determined using the mean of 4 automated cuff readings, 2 on each arm, taken 1 minute apart. Participants did not necessarily have their BP measured by the standardized method immediately following the routine method but all measurements were performed during the same 12-week time period.
Home blood pressure protocol. Participants were given instructions according to the American Heart Association (AHA) recommendations for measuring home BP. Patients were instructed to use a calibrated, automated arm BP cuff. Home BP machines were not provided in advance, and each individual’s BP machine was not calibrated. They also were instructed to rest at least 5 minutes before measuring their BP. The mean home BP was determined by the cumulative average of 3 readings in the morning and evening, taken 1 minute between each reading, for a total of 6 readings/d. Participants recorded home BPs for 2 weeks before submitting their readings. Each participant affirmed clear understanding of how to measure BP by correctly demonstrating placement of the cuff 1 time under supervision.