A Body Size–Adjusted Maximum Ultrafiltration Rate Warning Level Is Not Equitable for Larger Patients

It is now well established that higher ultrafiltration (UF) rates during a hemodialysis session are associated with a higher propensity to intradialytic hypotension, cardiac stunning, and mortality (¹,²). These associations have been derived from observational dataset reports, with all of their known shortcomings. There have been no randomized trials examining this issue and very few interventional trials. On the basis of observational data, guidelines and quality assurance metrics have been proposed, suggesting a maximum allowable UF rate, such as 13 ml/kg per hour, although it has been recognized that the higher mortality risk associated with UF rate might be more or less continuous. Daugirdas and Schneditz (³) have argued that, on the basis of physiologic measurements such as those related to blood volume, the UF rate might be better scaled to body surface area than to body weight. Flythe et al. (⁴) have responded to this suggestion by pointing out that mortality risk is higher with higher UF rates in both men and women of different body sizes and that this is true whether body size is expressed as postdialysis weight, body mass index, or body surface area (Table 1).

Table 1. - Relative mortality risk associated with exceeding ultrafiltration rate targets on the basis of postdialysis body weight, body mass index, or body surface area

	Women			Men
Body Size Measure	n	Percentile	Hazard Ratio (95% Confidence Interval)	n	Percentile	Hazard Ratio (95% Confidence Interval)
Postdialysis weight	10,633	<20 (≤55.6 kg)	1.20 (1.13 to 1.27)	12,983	<20 (≤65.8 kg)	1.10 (1.04 to 1.17)
Postdialysis weight	10,708	20–39 (55.7–65.1 kg)	1.13 (1.06 to 1.21)	13,074	20–39 (65.9–74.5 kg)	1.18 (1.11 to 1.26)
Postdialysis weight	10,642	40–59 (65.2–75.7 kg)	1.32 (1.22 to 1.43)	12,998	40–59 (74.6–84.0 kg)	1.24 (1.14 to 1.34)
Postdialysis weight	10,699	60–79 (75.8–90.7 kg)	1.31 (1.18 to 1.45)	13,055	60–79 (84.1–98.5 kg)	1.17 (1.06 to 1.30)
Postdialysis weight	10,623	≥80 (≥90.8 kg)	1.28 (1.08 to 1.51)	12,976	≥80 (≥98.6 kg)	1.41 (1.22 to 1.63)
Body mass index	10,643	<20 (≤21.7 kg/m2)	1.20 (1.13 to 1.28)	12,992	<20 (≤21.9 kg/m2)	1.13 (1.06 to 1.19)
Body mass index	10,651	20–39 (21.8–25.1 kg/m2)	1.14 (1.07 to 1.22)	13,006	20–39 (22.0–24.6 kg/m2)	1.12 (1.05 to 1.20)
Body mass index	10,648	40–59 (25.2–28.9 kg/m2)	1.36 (1.26 to 1.47)	13,000	40–59 (24.7–27.4 kg/m2)	1.25 (1.15 to 1.35)
Body mass index	10,643	60–79 (29.0–34.4 kg/m2)	1.29 (1.17 to 1.42)	12,998	60–79 (27.5–31.6 kg/m2)	1.24 (1.13 to 1.37)
Body mass index	10,643	≥80 (≥34.5 kg/m2)	1.37 (1.18 to 1.59)	12,991	≥80 (>31.7 kg/m2)	1.36 (1.19 to 1.56)
Body surface area	10,642	<20 (≤1.56 m2)	1.18 (1.11 to 1.25)	12,999	<20 (≤1.77 m2)	1.10 (1.04 to 1.17)
Body surface area	10,656	20–39 (1.57–1.69 m2)	1.22 (1.14 to 1.30)	13,009	20–39 (1.78–1.89 m2)	1.23 (1.15 to 1.31)
Body surface area	10,657	40–59 (1.70–1.81 m2)	1.29 (1.19 to 1.39)	12,992	40–59 (1.90–2.01 m2)	1.26 (1.17 to 1.36)
Body surface area	10,639	60–79 (1.82–1.97 m2)	1.29 (1.17 to 1.43)	13,000	60–79 (2.02–2.18 m2)	1.20 (1.09 to 1.33)
Body surface area	10,648	≥80 (≥1.98 m2)	1.38 (1.18 to 1.60)	12,999	≥80 (≥2.19 m2)	1.33 (1.15 to 1.53)

Fine and Gray proportional subdistribution hazards regression models with kidney transplantation and dialysis modality change treated as competing risks were used to estimate the ultrafiltration (UF) rate and all-cause mortality association comparing mean UF rates >13 ml/h per kilogram with those ≤13 ml/h per kilogram across strata of sex and body weight; mean UF rates >37 ml/h per kilogram per meter² with those ≤37 ml/h kilogram per meter² across strata of sex and body mass index (BMI); and mean UF rates >500 ml/h per meter² with those ≤500 ml/h per meter² across strata of sex and body surface area (BSA). Models were adjusted for age (continuous), race (Black patient versus non-Black), ethnicity (Hispanic patient versus non-Hispanic), time on dialysis (1–2, 3–4, or ≥5 versus <1 year), vascular access (graft or fistula versus catheter), history of heart failure (yes versus no), history of cardiovascular disease (yes versus no), history of diabetes (yes versus no), albumin (3.1–3.5, 3.6–4.0, or >4.0 versus ≤3.0 g/dl), creatinine (continuous; 4.1–5.0, 5.1–6.0, or >6.0 versus ≤4.0 mg/dl), hemoglobin (10.0–11.9 or ≥12.0 versus <10.0 g/dl), urea reduction ratio (continuous), predialysis systolic BP stimulating activity (131–150, 151–170, or >170 versus ≤130 mm Hg), and missed treatments (≥3 versus <3). Postdialysis weight was used to calculate normalized UF rates for weight, BMI, and BSA. In the overall study population, a UF rate of 13 ml/h per kilogram corresponded to the 80th percentile when UF rate was scaled to body weight (kilograms). Thus, analogous 80th percentile thresholds for UF rates scaled to BMI and BSA were evaluated, where the 80th percentile of UF rate scaled to BMI=37 ml/h per kilogram per meter² and the 80th percentile of UF rate scaled to BSA=500 ml/h per meter². Modified from ref. 4, with permission.

I would argue that the data by Flythe et al. (⁴) suggest that a UF rate warning value on the basis of any of these body size measures is not optimal. If one considers their data in Table 1 and averages the mean relative mortality risk in the lowest two quintiles of body weight versus the highest two quintiles, one finds that the change in mortality risk is higher in the heavier patients when the warning level UF rate is exceeded. For example, for body weight scaling where the warning level UF rate was 13 ml/h per kilogram, for smaller women, the mean relative risk of exceeding this level is 1.16, compared with 1.29 for larger women (highest two quintiles of body weight averaged). This means that by the time the UF cutoff is reached in larger women, the relative mortality risk compared with smaller women exceeding the warning level is about 129/116 or 1.11. For men, the situation is similar. For smaller men, again averaging the risk for the two lowest weight quintiles in Table 1, a relative mortality risk of 1.14 is observed when the 13–ml/h per kilogram UF warning level is exceeded. However, in the highest two weight quintiles of men, exceeding the weight-scaled UF warning level is associated with a relative mortality risk of 1.29, very similar to the magnitude of higher risk in the two largest weight quintiles of women. Again, the relative risk of exceeding the UF warning value in heavier versus lighter men is 129/114=1.13. What does this mean? It means that UF rate is not flagged in larger patients until their UF-associated incremental mortality risk is 11%–13% higher compared with the risk in lighter patients exceeding the same weight-based UF warning level.

The data of Flythe et al. (⁴) for body surface area (Table 1) are similar to those for body weight. Here, Flythe et al. (⁴) used a cutoff UF rate value on the basis of body surface area of 500 ml/h per meter². Their data for body surface area–scaled UF rate show a similar pattern of body size inequity. In women, in the two smaller versus two larger quintiles of body surface area, the relative mortality risk of exceeding the UF warning value is 1.20 for the smaller women versus 1.33 for the larger women. The higher mortality risk in larger versus smaller women is about 133/120=1.11, very similar to that when UF was scaled to body weight. For men, again, the data are similar. The magnitude of higher mortality risk was 1.165 for men in the two smaller quintiles of body surface area versus 1.265 for men in the two larger surface area quintiles for a “size inequity” factor of 1.265/1.1265=1.09. A similar size bias can be seen for the body mass index quintile rows (Table 1).

One potential solution to the problem of size inequity in UF warning levels is to consider using the “raw” UF (i.e., simply in terms of milliliter per hour). In a preliminary report published as an American Society of Nephrology abstract in 2014, Larkin et al. looked at the incremental risk of exceeding a UF level of 800 ml/h in different size quintiles of men and women and found, somewhat counterintuitively, that the risk of exceeding this simple UF warning level was similar in both sexes, regardless of body size. No one to date has proposed a fixed UF rate (independent of body size) warning level, perhaps because we are so used to scaling most parameters in dialysis to body size. Of course, at some extremes of body size, especially on the small end, the need for some sort of scaling becomes inevitable. However, before peremptorily discarding a simple size-independent value of maximum UF rate as a quality metric, we need to repeat, replicate, and expand on the preliminary data from Larkin et al. If the data do show a more uniform incremental mortality rate risk in patients across a range of body sizes who exceed a size-independent maximum value of UF rate, then perhaps a rate not scaled to body size (for adult patients) would be a preferred metric Alternatively, one could develop a statistically derived UF rate warning level that incorporated a correction for body size but where the incremental risk of exceeding such a value would be similar across sex and body size. At present, the initial data from Flythe et al. (⁴) suggest that setting as a warning level, a UF rate divided by either body weight or body surface area may be suboptimal in larger patients, as a proper UF rate warning level for such larger patients might optimally be set at a lower value.

Disclosures

J.T. Daugirdas reports consultancy agreements with Fresenius Medical Care.

Funding

None.