Accuracy of Peridialytic, Intradialytic, and Scheduled Interdialytic Recordings in Detecting Elevated Ambulatory Blood Pressure in Hemodialysis Patients

Abstract

Current recommendations suggest the use of ambulatory blood pressure monitoring (ABPM) as the gold standard for hypertension diagnosis and management in hemodialysis patients. This study assesses the accuracy of peridialytic, intradialytic, and scheduled interdialytic recordings in detecting abnormally elevated 44-hour interdialytic blood pressure (BP). Diagnostic test study. 242 Greek hemodialysis patients who successfully underwent ABPM. Ambulatory BP was used as the reference method to evaluate the accuracy of the following BP metrics: predialysis and postdialysis BP, intradialytic BP, intradialytic plus pre/postdialysis BP, and scheduled interdialytic BP (on an off-dialysis day at 8:00 am, 8:00 pm, and their average). 44-hour ambulatory systolic BP/diastolic BP (SBP/DBP)≥ 130/80 mm Hg. The 44-hour SBP/DBP levels differed significantly from predialysis and postdialysis BP but showed no or minor differences compared with the other BP metrics. Bland-Altman plots showed an absence of systematic bias for all metrics but large between-method difference and wider 95% limits of agreement for predialysis and postdialysis BP compared with intradialytic, intradialytic plus pre/postdialysis, and averaged scheduled interdialytic BP. The sensitivity/specificity and κ-statistic for diagnosing 44-hour SBP≥ 130 mm Hg were low for predialysis (86.5%/38.6%, κ-statistic = 0.27) and postdialysis BP (63.1%/73.3%, κ-statistic = 0.35), but better for intradialytic BP (77.3%/76.2%, κ-statistic = 0.53), intradialytic plus pre/postdialysis BP (76.6%/72.3%, κ-statistic = 0.49), and scheduled interdialytic BP (87.9%/77.2%, κ-statistic = 0.66). In receiver operating characteristic (ROC) analyses, the areas under the curve (AUC) of predialysis SBP (AUC = 0.723) and postdialysis SBP (AUC = 0.746) were significantly lower than that of intradialytic SBP (AUC = 0.850), intradialytic plus pre/postdialysis SBP (AUC = 0.850), and scheduled interdialytic SBP (AUC = 0.917) (z test, P< 0.001 for all pairwise comparisons). Similar observations were made for DBP. Typical home BP data were not obtained, and no assessment was obtained of the reproducibility of the examined metrics over time. Intradialytic, intradialytic plus pre/postdialysis, and scheduled interdialytic BP measurements were more accurate in detecting elevated 44-hour BP than predialysis and postdialysis BP. Averaged intradialytic BP recordings or scheduled readings at the off-dialysis day appear to be promising approaches to the diagnosis of elevated BP in hemodialysis.