Telemonitoring of home blood pressure: diagnosis and treatment

2017

Januar

Nikolai Hoffmann-Petersen

• PhD-afhandling

Background
Hypertension is one of the leading cause of death and disability worldwide. Hypertension is a greater risk for cardiovascular disease than smoking or obesity, and about half of the cases of ischaemic heart disease and stroke are associated with high blood pressure (BP). Treatment of hypertension is closely related to reduction in cardiovascular morbidity and mortality.
Reliable BP measurements are important. Home BP measurements are better correlated to end organ damage than office blood pressure (OBP). Telemedical home blood pressure monitoring (TBPM) represents an improvement in the quality of data by eliminating reporting bias. Whether TBPM as the only add-on intervention can improve control of hypertension is unknown. Cardiovascular risk gradually increases from normotension, over WCH and MH to persistent hypertension. After years of focus on BP how well are we doing in Denmark? How prevalent is white coat hypertension (WCH), masked hypertension (MH) and
persistent hypertension? Blood pressure monitors are clinically validated before recommended for use. But monitors have not been compared head-to-head. Thus it is not known whether clinical relevant differences exist between different validated devices.

Hypotheses
1. TBPM is superior to OBP measurement and comparable to daytime ambulatory blood pressure (ABPM). (Study I)
2. Measuring TBPM as well as OBP in citizens of Holstebro, Denmark, aged 55-64 years will give a precise estimate of the prevalence of a) hypertension, b) insufficiently treated hypertension, c) WCH, and d) MH. (Study II)
3. Blood pressure lowering treatment based on TBPM lowers daytime ABPM more effectively than usual care by general practitioners. (Study III)
4. Commonly used, clinically validated BP monitors fulfil the accuracy and precision limits set by the manufacturers when tested in a technical validation setup using oscillometric
waveform simulators. (Study IV)

Methods
Study I. Observational study among 112 patients attending the renal outpatient clinic at Holstebro Hospital. OBP was measured with an A&D 767PlusBT monitor. TBPM was undertaken for 4 days with three measures three times daily, followed by a 24-ABPM.
Study II. Population study among the citizens of Holstebro, Denmark, aged 55-64 years. OBP and TBPM were measured and a questionnaire filled in.
Study III. Randomised, controlled, un-blinded 3-month trial. Patients with TBPM above limits from the preceding prevalence study, confirmed by AMBP, were recruited. The patients were randomised to either antihypertensive treatment monitored by TBPM (intervention group) or conventional blood pressure monitoring (usual care). The usual GP treated all his or hers patients. In the intervention group, 3-day TBPM was done every second week. Mean BP of days 2 to 3 was transmitted to the GPs’ electronic patient record.
Study IV. Firstly, 30 previously used BP monitors from one manufacturer were tested on two simulators. Secondly, five previously used A&D 767PlusBT and five previously used Omron 57 705IT monitors were tested on one simulator. Thirdly, 10 new and unused A&D 767PlusBT and 10 new and unused Omron 705IT were tested on one simulator.

Main results and conclusions
In Study I, we found that TBPM was lower than daytime ABPM. The observed difference could be due to less discomfort associated with TBPM, or simply a systematic difference because of use of different algorithms for estimating the systolic and diastolic BP. Our findings support the current Danish recommendation that BP be measured at home BP for 3 days.
Study II is the first large-scale prevalence study to eliminate completely reporting bias by using TBPM. The prevalence of hypertension, MH and WCH was found to be highly dependent on the threshold set for the diagnosis of hypertension. The prevalence of hypertension in the age group 55-64 years was 50%. Only half of patients with an awareness of hypertension had controlled hypertension.
One fourth had either WCH or MH, which means that one in four patients who have their BP measured in the office will be misclassified. In patients with an awareness of hypertension one in three patients will be misclassified. Only when measuring extremely high or low OBP were we de facto able to rule out MH or WCH. Our findings emphasise the relevance of more frequent use of out-of-office BP measurements in the treatment and diagnosis of hypertension.
Study III was designed to create a technological platform for collaboration between GPs, a municipal preventive centre and the hospital. We showed that the development of a fully integrated telemedical system is possible. Patients using TBPM and controls achieved a significant, but equal reduction in systolic as well as diastolic BP. There was no difference in the number of patients reaching the BP goals. An equal number of patients in the TBPM and the control group reached the BP targets, but only one in five patients reached the targets. Receiving a TBPM report every second week during a 3-month period did not improve the
treatment of hypertension.
Study IV showed that with use of Omron 705IT, high accuracy and good precision were obtained for both diastolic and systolic BP values. No significant difference was found between previously used and newer devices. On the other hand we found that the A&D 767PlusBT monitor systematically measured too low systolic BP values but exhibited good precision. The accuracy limits of ± 3 mmHg, set by the A&D 767PlusBT manufacturer, were exceeded. Previously used monitors showed lower accuracy than newer ones. It is appropriate to develop a protocol for testing monitors using oscillometric waveform simulators. This could
serve as a guideline for later post-clinical calibration of oscillometric waveform simulators.
Our studies have confirmed, that hypertension still presents us with a number of serious challenges. Hypertension is a condition with a very high prevalence. Only a fraction of the patients is sufficiently treated.

Universitetsklinik for nyresygdomme og hypertension, Regionshospitalet Holstebro

Aarhus Universitet

• Ja