From Isolated Ultrafiltration to Blood-Temperature-Controlled Feedback: An Odyssey Started by Jonas Bergström

The relationship between blood pressure (BP) and mortality in hemodialysis patients has remained controversial. Some studies suggested that a lower pre- or postdialysis BP was associated with excess mortality, while others showed poorer outcome in patients with uncontrolled hypertension. We conducted a multicenter prospective cohort study to evaluate the impact of hemodialysis-associated hypotension on mortality. We recruited 1244 patients (685 males; mean age, 60 +/- 13 years) who underwent hemodialysis in 28 units during the two-year study period beginning in December 1999. Pre-, intra-, and postdialysis BP, and BP upon standing soon after hemodialysis, were measured in all patients at entry. Logistic regression analysis was used to assess the effect on mortality of pre-, intra-, and postdialysis BP, a fall in BP during hemodialysis, and a fall in BP upon standing soon after hemodialysis. During the study period, 149 patients died. Logistic models identified the lowest intradialysis systolic blood pressure (SBP) and degree of fall in SBP upon standing soon after hemodialysis as significant factors affecting mortality, but not pre- or postdialysis SBP and diastolic BP. The adjusted odds ratio for death was 0.79 (95% CI 0.64-0.98) when the lowest intradialysis SBP was analyzed in increments of 20 mm Hg, and was 0.82 (95% CI 0.67-0.98) when the fall in SBP upon standing soon after hemodialysis was analyzed in increments of 10 mm Hg. These results suggest that intradialysis hypotension and orthostatic hypotension after hemodialysis are significant and independent factors affecting mortality in hemodialysis patients.

Many reports note that the use of cool dialysate has a protective effect on blood pressure during hemodialysis (HD) treatments. However, formal clinical trials in which dialysate temperature is tailored to the body temperature of appropriately selected hypotension-prone patients are lacking. We investigated the effect of thermal control of dialysate on hemodynamic stability in hypotension-prone patients selected from 27 centers in nine European countries. Patients were eligible for the study if they had symptomatic hypotensive episodes in 25% or more of their HD sessions, assessed during a prospective screening phase over 1 month. The study is designed as a randomized crossover trial with two phases and two treatment arms, each phase lasting 4 weeks. We used a device allowing the regulation of thermal balance (Blood Temperature Monitor; Fresenius Medical Care, Bad Homberg, Germany), by which we compared a procedure aimed at preventing any transfer of thermal energy between dialysate and extracorporeal blood (thermoneutral dialysis) with a procedure aimed at keeping body temperature unchanged (isothermic dialysis). One hundred sixteen HD patients were enrolled, and 95 patients completed the study. During thermoneutral dialysis (energy flow rate: DeltaE = -0.22 +/- 0.29 kJ/kg x h), 6 of 12 treatments (median) were complicated by hypotension, whereas during isothermic dialysis (energy flow rate: DeltaE = -0.90 +/- 0.35 kJ/kg x h), the median decreased to 3 of 12 treatments (P < 0.001). Systolic and diastolic blood pressures and heart rate were more stable during the latter procedure. Isothermic dialysis was well tolerated by patients. Results show that active control of body temperature can significantly improve intradialytic tolerance in hypotension-prone patients. Copyright 2002 by the National Kidney Foundation, Inc.

A significant proportion of hemodialysis patients have subnormal body temperature. Dialysis against cool dialysate has been frequently shown to reduce the incidence of symptomatic hypotension (SH), although only one of these reports included patient temperatures. Our hypothesis was that the response to cool or normal temperature dialysis could depend on a patient's baseline temperature. Of 128 patients in two hemodialysis units, 28 had a (mean of 5) baseline temperature less than 36 degrees C and 48 patients had a temperature higher than 36.5 degrees C. A crossover study was performed by dialyzing patients for 10 consecutive treatments with the same dialysate temperature, either 37 degrees C or 35 degrees C. All patients combined had a significant reduction in SH with 35 degrees C dialysate, 11.2% versus 5.5% with 37 degrees C dialysate (P = 0.001). The incidence of SH in euthermic patients was not affected by dialysate temperature. Hypothermic patients dialyzed against 37 degrees C dialysate had the highest incidence of SH, which decreased markedly with 35 degrees C dialysate (15.9% v 3.4%; P = 0.0001). There were no differences in age, duration of dialysis, gender, hemoglobin, urea, creatinine, or volume removed per dialysis between the two groups. In conclusion, subnormal temperature is common in dialysis patients but the etiology is unclear. The hemodynamic protective effect of cool dialysate only occurs in patients with subnormal temperatures. Only the subpopulation of patients with SH and low body temperature should be dialyzed against cool dialysate.