Abstract Background Hemodialysis patients undergo blood pressure (BP) measurements during the predialysis, intradialytic and postdialysis (POSBP) periods at the dialysis sessions. However, there is no consensus regarding the time point when POSBP measurement should be performed. Purpose This study investigated the relationship of POSBP measurements performed at different time points with home blood pressure monitoring (HBPM). Methods This prospective study included 98 long-term dialysis patients (67% men; age=59.4 ±17.7 years; 79% using antihypertensive medications). During one week, POSBP measurements were obtained in sitting patients by trained health staff using oscillometric BP monitors lodged in the dialysis machines at all dialysis sessions in three different periods: 1) immediately after the dialysis period ended (POSBP1); 2) after returning the patient's blood (POSBP2) and 3) after hemostasis of the fistula or catheter sealing (POSBP3). In the same week, HBPM was performed using validated devices and was calculated as the mean of BP measurements obtained on six consecutive days, comprising three readings in the morning and in the evening after at least 3 min of rest, before antihypertensive medications were taken, with the patients seated. Abnormal BP was defined as HBPM≥135/85 mmHg. Based on the optimal cut-point values derived from ROC curves analysis, we built cutoffs for POSBP1 (143/79 mmHg), POSBP2 (153/78 mmHg) and POSBP3 (151/80 mmHg) corresponding to HBPM values of 135/85 mmHg. Results The values of POSBP1, POSBP2, POSBP3 and HBPM were 138±21/72±13, 143±24/73±13, 142±22/72±13 and 128±19/76±13 mmHg, respectively. Systolic HBPM showed greater correlation with systolic POSBP3 (r=0.71), followed by POSBP1 (r=0.68), POSBP2 (r=0.67), while diastolic HBPM showed greater correlation with diastolic POSBP2 and POSBP3 (both r=0.56), followed by POSBP1 (r=0.49) (all p<0.05). The POSBP3 cutoff showed the highest association (odds ratio [95% CI]= 8.06 [3.21, 20.27]), accuracy (area under the curve [95% CI]= 0.74 [0.65, 0.83]) and concordance (kappa coefficient=0.47) to identify elevated BP at HBPM in comparison with the POSBP1 and POSBP2 cutoffs (Figure). Conclusion(s) POSBP3 exhibited the strongest correlation with HBPM and had the highest accuracy to detect elevated HBPM in dialysis patients and may be a potentially preferable POSBP measurement in clinical practice.
Read full abstract