William B. Schwartz Division of Nephrology, Tufts Medical Center, Boston, MA.
Biostatistics Research Center, Boston, MA.
Am J Kidney Dis. 2014 Feb;63(2):276-85. doi: 10.1053/j.ajkd.2013.08.007. Epub 2013 Oct 3.
Hemodialysis patients have high mortality rates, potentially reflecting underlying comorbid conditions and ongoing catabolism. Intradialytic oral nutritional supplements may reduce this risk.
Retrospective propensity-matched cohort.
SETTING & PARTICIPANTS: Maintenance hemodialysis patients treated at Dialysis Clinic Inc facilities who were initiated on a nutritional supplement protocol in September to October 2010 were matched using a propensity score to patients at facilities at which the protocol was not used.
Prescription of the protocol, whereby hemodialysis patients with serum albumin levels ≤3.5g/dL would initiate oral protein supplementation during the dialysis procedure. Sensitivity analyses matched on actual supplement intake during the first 3 study months. Covariates included patient and facility characteristics, which were used to develop the propensity scores and adjust multivariable models.
All-cause mortality, ascertained though March 2012.
Of 6,453 eligible patients in 101 eligible hemodialysis facilities, the protocol was prescribed to 2,700, and 1,278 of these were propensity matched to controls. Mean age was 61 ± 15 (SD) years and median dialysis vintage was 34 months. There were 258 deaths among protocol assignees versus 310 among matched controls during a mean follow-up of 14 months. In matched analyses, protocol prescription was associated with a 29% reduction in the hazard of all-cause mortality (HR, 0.71; 95% CI, 0.58-0.86); adjustment had minimal impact on models. In time-dependent models incorporating change in albumin level, protocol status remained significant but was attenuated in models incorporating a 30-day lag. Similar results were seen in sensitivity analyses of 439 patients receiving supplements who were propensity-matched to controls, with 116 deaths among supplement users versus 140 among controls (HR, 0.79; 95% CI, 0.60-1.05), achieving statistical significance in adjusted models.
Observational design, potential residual confounding.
Prescription of an oral nutritional supplement protocol and use of oral protein nutritional supplements during hemodialysis are associated with reduced mortality among in-center maintenance hemodialysis patients, an effect likely not mediated by change in serum albumin levels.
血液透析患者的死亡率较高，这可能反映了潜在的合并症和持续的分解代谢。透内口服营养补充剂可能会降低这种风险。
回顾性倾向匹配队列研究。
2010 年 9 月至 10 月期间在透析诊所接受治疗并开始营养补充方案的维持性血液透析患者，使用倾向评分与未使用该方案的设施中的患者进行匹配。
该方案规定，血清白蛋白水平≤3.5g/dL 的血液透析患者将在透析过程中开始口服蛋白质补充剂。敏感性分析则根据前 3 个月的实际补充摄入量进行匹配。协变量包括患者和设施特征，这些特征用于制定倾向评分并调整多变量模型。
在 101 个符合条件的血液透析设施中的 6453 名合格患者中，该方案规定了 2700 名患者，其中 1278 名患者与对照组进行了倾向匹配。平均年龄为 61±15（标准差）岁，中位透析时间为 34 个月。在平均随访 14 个月期间，方案分配者中有 258 人死亡，对照组中有 310 人死亡。在匹配分析中，方案处方与全因死亡率降低 29%相关（HR，0.71；95%CI，0.58-0.86）；调整对模型的影响很小。在纳入白蛋白水平变化的时间依赖性模型中，方案状态仍然显著，但在纳入 30 天滞后的模型中则减弱。在对接受补充剂的 439 名患者进行的敏感性分析中也观察到了类似的结果，这些患者与对照组进行了倾向匹配，补充剂使用者中有 116 人死亡，对照组中有 140 人死亡（HR，0.79；95%CI，0.60-1.05），在调整后的模型中达到统计学意义。
观察性设计，潜在的残余混杂。
口服营养补充方案的处方和在血液透析过程中使用口服蛋白质营养补充剂与中心维持性血液透析患者的死亡率降低相关，这种效果可能不是通过血清白蛋白水平的变化来介导的。
Zotero 插件