« Back

Among patients receiving hemodialysis, at-home, ambulatory BP readings predict survival better than readings collected in dialysis unit, study reveals

Wednesday, March 17 2010 | Comments

---

Among patients receiving long-term hemodialysis, blood pressure (BP) recordings collected in the dialysis unit do not predict mortality as well as at-home or ambulatory BP readings do, according to a recent analysis.

The analysis compared the prognostic values of in-unit and out-of-unit BP readings in 326 patients who were treated at 1 of the 4 dialysis units staffed by nephrology faculty from Indiana University School of Medicine. The participants used a self-inflating automatic oscillometric device to collect BP readings at home 3 times daily for 1 week. In addition, ambulatory BP measurements were collected during an interdialytic interval for 44 hours. These measurements were compared with pre- and postdialysis measurements collected in the dialysis unit during the 2 weeks that spanned the period prior to and following the ambulatory BP recording.

During a median follow-up of 29 months, 102 patients (31%) died.

The adjusted hazard ratios (HRs) for mortality with increasing quartiles of ambulatory BP were (relative to the bottom quartile) 2.51 (95% CI, 1.27-4.95; P=.008), 3.43 (95% CI, 1.73-6.79; P<.0001), and 2.62 (95% CI, 1.33-5.17; P=.005). The HRs for increasing quartiles of home BP were 2.15 (95% CI, 1.13-4.11; P=.02), 1.7 (95% CI, 0.88-3.29; P=.113), and 1.44 (95% CI, 0.72-2.9; P=.306).

Systolic BP readings recorded before and after dialysis were not significantly associated with mortality (P=.17 and P=.997, respectively). There were no associations between diastolic BP and mortality, regardless of how it was measured.

Three models were created to compare the prognostic information provided by in-unit versus out-of-unit BP readings. The first model considered all 4 BP measurements (predialysis, postdialysis, home, and ambulatory), the second model considered only in-unit BP measurements, and the third model considered only home and ambulatory BP measurements. The model containing in-unit BP measurements only was inferior to the model that contained all 4 BP measurements, while the model containing home and ambulatory BP measurements only was similar to the model that contained all 4 BP measurements.

The data suggested a W-shaped association existed between home BP readings and mortality, with a similar but less pronounced association between ambulatory BP readings and mortality. Mortality rates seemed to be minimal with a home systolic BP of 120 mm Hg to 130 mm Hg and an ambulatory systolic BP of 110 mm Hg to 120 mm Hg.

Dr. Rajiv Agarwal, the author of the study, described the W-shaped association as an unexpected finding. A possible explanation for this, he suggested, is that elevated home or ambulatory BP readings prompted treatment, which subsequently led to improvement in survival.

Dr. Agarwal added that the current study differs from larger epidemiological studies, which have reported higher mortality with lower BP values, because patients who were recently hospitalized or sick (ie, those more prone to hypotension) were excluded from the current analysis. (Hypertension 2010;55:762-768.)

Print  |  E-mail