Secondary Logo

Journal Logo

Institutional members access full text with Ovid®

Peixoto Aldo J.; Gray, Trina A.; Crowley, Susan T.
Blood Pressure Monitoring: October 1999


Ambulatory blood pressure monitoring (ABPM) has been used increasingly in assessing dialysis patients. However, devices have not been validated formally for this population. Formal device validation is important in order to guarantee adequate performance for populations with special characteristics, such as patients undergoing hemodialysis.


To achieve formal validation of the SpaceLabs 90207 ambulatory blood pressure device (SLD) using a modified British Hypertension Society protocol.


Eighty-five hemodialysis patients were studied, generating 255 readings. Readings were obtained with patients supine (in a dialysis chair) over 5–10min during hemodialysis. Simultaneous, same-arm readings were obtained through the use of a T connector to a calibrated mercury manometer. The mean differences between readings obtained by observers and those obtained by the device were calculated. Limits of agreement between observers and SLD were determined and plotted according to the method of Bland and Altman. For grading of performance, we determined the number of readings for which the readings obtained by the device were within 5, 10, and 15mmHg of manometer readings. Final gradings were ascribed according to British Hypertension Society criteria.


The mean blood pressure (±SD) was 141/76± 31/15mmHg (observers) and 141/77 ± 27/15 mmHg (SLD), and the mean (±SD) difference between observers and device (observer-device) was −0.5± 7.5mmHg for systolic blood pressure (SBP) and −0.2± 5.2 mmHg for diastolic blood pressure (DBP). The device was less accurate in extreme ranges of SBP. In fact, there was a positive correlation between average [(observer+ device)/2] and difference (observer - device) for SBP (r= 0.54, P<0.0001), showing that underestimation in higher ranges, and overestimation in lower ranges of blood pressure occurred for SBP. For SBP, 53% of readings were within 5 mmHg of those obtained by the observers, 85% were within 10 mmHg, and 97% within 15 mmHg. For DBP, 78% were within 5 mmHg, 96% within 10 mmHg, and 98% within 15 mmHg. These observations conferred on the device grade C for SBP and grade B for DBP. The type of vascular access and the presence of non-functioning arteriovenous grafts and fistulas in the ipsilateral arm did not alter these results significantly.


These data validate the use of this device for hemodialysis patients. However, caution should be exercised in the evaluation of upper and lower ranges of SBP.

Copyright © 1999 Wolters Kluwer Health, Inc. All rights reserved.