Accurate blood pressure measurement is critical to successful clinical trials. Concerns about observer errors have led to the use of automated oscillometric devices without evidence that their performance is similar to that of trained observers. This study compares blood pressures obtained by trained observers and with an oscillometric device (Omron 705CP) to 24-h ambulatory blood pressure monitoring.
We performed a post-hoc analysis of 313 untreated hypertensive patients at the end of the washout phase of a Novartis hypertension trial. Patients had three seated trained observer mercury auscultatory blood pressure measurements followed by 24-h ambulatory blood pressure monitoring. The next day, the ambulatory blood pressure monitoring was removed and three seated readings were obtained with an Omron 705CP. Correlations for systolic blood pressure and diastolic blood pressure were obtained between daytime ambulatory blood pressure monitoring (0900 and 2100) and the two office methods. In addition, we investigated the degree of difference of trained observer and Omron measurements from ambulatory blood pressure monitoring.
For systolic blood pressure, the correlation with ambulatory blood pressure monitoring of the trained observer was significantly better than with that of the Omron 705CP (0.641 vs. 0.555, P=0.01). For diastolic blood pressure values, even greater disparity between the two office method correlations with ambulatory blood pressure monitoring was observed (trained observer=0.593 vs. Omron=0.319, P<0.0001). Both trained observer and Omron readings were consistently higher than ambulatory blood pressure monitoring for systolic blood pressure (P<0.0001) and diastolic blood pressure (P<0.0001). Omron measurements, however, deviated from ambulatory blood pressure monitoring more than those of the trained observer (P<0.0001 for systolic blood pressure and diastolic blood pressure).
For clinical trials using diastolic blood pressure targets, the Omron 705CP cannot replace the auscultatory blood pressure measurements of a trained observer. For systolic blood pressure, the Omron device and the trained observer had similar correlations with ambulatory blood pressure monitoring; however, both methods gave consistently higher systolic blood pressure values. Further study of oscillometric devices should be conducted before universally replacing auscultatory blood pressure determinations by trained observers in clinical trials.
Divisions of aNephrology and Hypertension
bBiostatistics, Mayo Clinic College of Medicine, Rochester, Minnesota
cNovartis Pharmaceutical Corporation, East Hanover, New Jersey, USA
Correspondence and requests for reprints to John W. Graves, MD, FACP, Division of Nephrology and Hypertension, Mayo Clinic School of Medicine, 200 First Street, SW, Rochester, MN 55905, USA
Tel: +1 507 284 8712; fax: +1 507 284 1161; e-mail: Graves.John@mayo.edu
Received 16 August 2005 Revised 17 November 2005 Accepted 18 November 2005