PREDICTION OF AMBULATORY HYPERTENSION WITH CLINIC AND HOME BLOOD PRESSURE RECORDINGS IN PATIENTS RECEIVING LONG-TERM PERITONEAL DIALYSIS

Abstract

Objective: Contrary to patients on chronic hemodialysis, those receiving long-term peritoneal dialysis (PD) routinely measure and record their blood pressure (BP) at home. The accuracy of these routine home BP recordings relative to standardized clinic BP measurements in detecting hypertension confirmed by the “gold-standard” method of 24-hour ambulatory BP monitoring (ABPM) remains elusive. Design and method: A total of 56 consecutive patients on long-term PD underwent 24-hour ABPM with the oscillometric device Mobil-O-Graph (IEM, Germany). Triplicate clinic BP measurements under standardized conditions were obtained with a self-inflating automatic monitor (HEM 705 CP, Omron Healthcare). Morning home BP recordings routinely taken by the patients themselves the week prior to their study visit were also recorded and averaged. Results: In correlation analysis, both standardized clinic (r = 0.723, p < 0.001) and routine weekly-averaged morning home systolic BP (SBP) (r = 0.739, p < 0.001) exhibited strong positive associations with 24-hour ambulatory SBP. The area under the curve of receiver operating characteristic (ROC) curve for standardized clinic and routine home SBP to detect systolic ambulatory hypertension was 0.834 (95% CI: 0.725–0.939) and 0.832 (95% CI: 0.726–0.939), respectively. In Bland-Altman analysis, standardized clinic SBP overestimated 24-hour SBP by 6.62 mmHg with 95% limits of agreement ranging from −18.8 to 32.0 mmHg. Similarly, routine home SBP overestimated 24-hour SBP by 7.31 mmHg, again with wide 95% limits of agreement (−16.2 to 30.8 mmHg). Conclusions: This study suggests that home BP recordings routinely taken by patients on long-term PD do not offer additive value to that of standardized clinic BP in predicting ambulatory hypertension. Standardized protocols for home BP monitoring should be applied in order to improve hypertension management in the PD population.