Raimann Jochen G, Wang Yuedong, Mermelstein Ariella, Kotanko Peter, Daugirdas John T
Renal Research Institute, New York, New York, USA.
Department of Statistics and Applied Probability, University of California, Santa Barbara, Santa Barbara, California, USA.
Kidney Int Rep. 2022 Apr 22;7(7):1585-1593. doi: 10.1016/j.ekir.2022.04.079. eCollection 2022 Jul.
One proposed threshold ultrafiltration rate (UFR) of concern in hemodialysis patients is 13 ml/h per kg. We evaluated associations among UFR, postdialysis weight, and mortality to determine whether exceeding such a threshold would result in similar levels of risk for patients of different body weights.
Data were analyzed in this retrospective cohort study for 1 year following dialysis initiation (baseline) and over 2 years of follow-up in incident patients receiving thrice-weekly in-center hemodialysis. Patient-level UFR was averaged over the baseline period. To investigate the joint effect of UFR and postdialysis weight on survival, we fit Cox proportional hazards models using bivariate tensor product spline functions, adjusting for sex, race, age, diabetes, and predialysis serum albumin, phosphorus, and systolic blood pressure (BP). We constructed contour plots of mortality hazard ratios (MHRs) over the entire range of UFR values and postdialysis weights.
In the studied 2542 patients, UFR not scaled to body weight was strongly associated with MHR, whereas postdialysis weight was inversely associated with MHR. MHR crossed 1.5 when unscaled UFR exceeded 1000 ml/h, and this relationship was largely independent of postdialysis weight in the range of 80 to 140 kg. A UFR warning level associated with a lower MHR of 1.3 would be 900 ml/h, whereas the UFR associated with an MHR of 1.0 was patient-size dependent. The MHR when exceeding a UFR threshold of 13 ml/h per kg was dependent on patient weight (MHR = 1.20, 1.45, and >2.0 for a 60, 80, and 100 kg patient, respectively).
UFR thresholds based on unscaled UFR give more uniform risk levels for patients of different sizes than thresholds based on UFR/kg.
血液透析患者中一个备受关注的超滤率阈值是每千克体重13毫升/小时。我们评估了超滤率、透析后体重与死亡率之间的关联，以确定超过该阈值是否会给不同体重的患者带来相似程度的风险。
在这项回顾性队列研究中，分析了开始透析（基线）后1年以及接受每周三次中心血液透析的新发病例患者超过2年随访期的数据。患者水平的超滤率在基线期进行平均。为了研究超滤率和透析后体重对生存的联合影响，我们使用双变量张量积样条函数拟合Cox比例风险模型，并对性别、种族、年龄、糖尿病以及透析前血清白蛋白、磷和收缩压进行了校正。我们构建了超滤率值和透析后体重整个范围内的死亡风险比（MHR）等高线图。
在研究的2542例患者中，未根据体重进行调整的超滤率与MHR密切相关，而透析后体重与MHR呈负相关。当未调整的超滤率超过1000毫升/小时时，MHR超过1.5，并且在80至140千克范围内，这种关系在很大程度上独立于透析后体重。与较低MHR为1.3相关的超滤率警告水平为900毫升/小时，而与MHR为1.0相关的超滤率则取决于患者体型。超过每千克体重13毫升/小时的超滤率阈值时的MHR取决于患者体重（60千克、80千克和100千克患者的MHR分别为1.20、1.45和>2.0）。
基于未调整超滤率的超滤率阈值比基于超滤率/千克的阈值能为不同体型的患者提供更统一的风险水平。
