Use of automated blood pressure measurements in clinical trials and registration studies: data from the VALTOP Study

Auscultatory measurement of office blood pressure (BP) by mercury sphygmomanometers (AuscBPM) is still the gold standard in clinical trials and registration studies for antihypertensive drugs. The increasing availability of accurate automated oscillometric BP measuring devices has offered new perspectives in this field, although their usefulness in drug studies has not been systematically tested yet.

During the course of Valsartan 320 mg EU Registration Study we used an electronic automated oscillometric BP measuring device (eBPM) as an alternative to conventional AuscBPM. Altogether 3776 patients were randomized into a double-blinded actively controlled parallel group study in 303 centers, and 54 422 BP readings were recorded by the validated, electronic, automated oscillometric device OMRON 705IT with digital printouts. Terminal digit preference and preference at therapeutic cutoff points were evaluated. The data were compared with the results of an earlier valsartan study similar in design but based on conventional AuscBPM. Furthermore, based on a simulation, four strategies for automated BP measurement with varying number of office readings (3-5) were analyzed to define an optimal method to collect BP at office visits.

eBPM eliminated terminal digit preference and dramatically reduced preferences for therapeutic cutoff points as compared with earlier valsartan trials with conventional AuscBPM. However, even with eBPM a minor bias with the therapeutic cutoff value was observed probably because of an observer bias during data documentation. The within-patient variability of three measurements sequentially taken at each visit was similar to other strategies including more measurements.

On the basis of our data, we suggest that eBPM is a suitable alternative to AuscBPM in clinical trials and registration studies, and may carry specific advantages. Automatic data transfer of recorded values to electronic patient files may further minimize observer bias. Manufacturers should consider such findings for the development of professional devices.