Effects of supine blood pressure on interpretation of standing up test in 500 patients with diabetes mellitus

Abstract

The relationship between fall in blood pressure (BP) on standing and supine BP before standing up was studied in 75 healthy controls and in 500 patients with diabetes mellitus (DM) using conventional BP measurements. The influences of physiological (sex, age, height) and DM-related factors (type, duration, HbA 1c-level, use of insulin, oral antidiabetic and anti-hypertensive medication) on BP-fall were assessed. The effects of using a fixed abnormality threshold and a new supine BP-related abnormality definition on interpretation of the test were determined. Highly significant relationships of BP-fall with supine BP were found in control and DM subjects. Slopes did not differ between these groups. Slopes for systolic BP-fall were steeper in type 1 than in type 2 DM patients. A forward stepwise regression procedure revealed that supine BP (explaining 24% of variance) and HbA Ic (explaining 1%) had significant influences on systolic BP-fall. Diastolic supine BP explained 14% of diastolic BP-fall, age 3%, and sex 2%. Only supine BP was thus of practical relevance in explaining BP-fall. Taking supine BP into consideration affected test results: of 74 subjects with an abnormal conventional systolic BP-fall, 10 (13.5%) had been misclassified according to the new method, and 4 additional patients had been misclassified as normal. Classification changes were much larger for diastolic BP-fall (63% of conventionally abnormal results were reclassified as normal), but the range of diastolic BP is smaller than for systolic BP, meaning that the measurement error interferes with its clinical utility. BP-fall was higher in type 2 than in type 1 diabetics, and higher in those using anti-hypertensives than in those who did not. These effects were due to differing supine BPs, showing that taking supine BP into account can prevent incorrect conclusions regarding abnormality of the standing up test in DM. The validity of this test can be improved by accounting for supine BP.