An assessment of blood pressure measurement in a diabetic clinic using random-zero, semi-automated, and 24-hour monitoring.

We have undertaken a randomized, observer-blinded comparison of three different methods of assessing blood pressure in 40 outpatients with hypertension complicating diabetes mellitus: the Hawksley random zero sphygmomanometer (RZS) by two observers using a dual headed stethoscope; the semi-automated Dinamap monitor, and 24-h ambulatory blood pressure monitoring (ABPM) using a Spacelabs 90207. The techniques were compared by plotting the difference against average of readings obtained by combinations of two techniques (RZS observer 1 vs 2; RZS (mean observer 1 and 2) vs Dinamap; RZS (mean observer 1 and 2) vs daytime ABPM; and Dinamap vs daytime ABPM). There was good agreement of readings with RZS, observers 1 and 2 and RZS with Dinamap and daytime ABPM. When the three methods were used to classify blood pressure control according to BDA criteria, it was found that they produced equivalent results and although mean systolic pressures appeared lower with ABPM this did not reach statistical significance. Overall control of systolic blood pressure was classified as unsatisfactory in 82% of patients, diastolic blood pressure was unsatisfactory in 55%. Fifty-five percent were determined to be nocturnal 'non-dippers' by ABPM. They were older: median age 62.9 (range 35.9-83.5) years compared to 48.2 (32.4-70.0) years for 'dippers' (p<0.05). Thirteen of the 18 patients who did dip overnight had creatinines within the normal laboratory range, whereas only 6 of 22 'non-dippers' had normal creatinines (median 91 (56-768) micromol l(-1) for 'dippers' vs 166 (76-479) micromol l(-1) for 'non-dippers', p < 0.001). Nineteen of the 29 males studied were 'non-dippers' compared with only 4 out of 11 females (p < 0.001). There was no association between dipper status, duration or type of diabetes or presence of proteinuria.