Comparison Of Three Nutritional Risk Screening Methods For Predicting Mortality In Maintenance Hemodialysis Patients

Background:The prevalence of Protein-energy wasting(PEW)varies between 18%and 75%in hemodialysis patients.Since PEW is highly prevalent and strongly associated with increased risk of mortality,it is important to detect PEW accurately and easily.There has been a great consensus that the key first step in the evaluation of nutritional status is to identify "at risk" status by using the validated screening tools.Although there are many methods to evaluate the nutritional risk in maintenance hemodialysis(MHD)patients,a simple and efficient nutritional risk screening method has not yet been determined.The SGA,which was recommended by the KDOQI(Kidney Disease Outcomes Quality Initiative,KDOQI)and European guideline,was verified as a validated diagnostic tool for assessing nutritional status in MHD patients.However,the SGA requires a subjective assessment by a well trained staff,which is time-consuming and cumbersome.And there is no uniform standard for the evaluation of some indicators in the SGA.Besides,the 2007 European Best Practices Guideline states that SGA is only suitable for assessing patients with severe malnutrition in MHD.Later,a more comprehensive tool,the malnutrition inflammation score(MIS),was proposed by Kalantar-Zadeh et al.The MIS has been validated as a better nutritional indicator than SGA and was proved to correlate with various nutritional variables,inflammation and mortality.However,the MIS is based on SGA and,therefore,also requires a subjective assessment.As such,the objective screening nutrition dialysis(OSND)was developed as an alternative for the MIS in the assessment of the nutritional risk of hemodialysis patients.In fact,there are some simple nutritional risk screening methods can also be used to assess the nutritional risk of MHD patients.The geriatric nutritional risk index(GNRI)was recommended as the simplest and most accurate risk index for use with hemodialysis patients.Among the above screening tools,the MIS has been validated as a better nutritional indicator than the SGA in a previous study.Moreover,the predictive validity of the MIS was reported to be superior to the GNRI in a recent study.More importantly,although the OSND has been validated as an effective nutritional screening tool,the predictive effects on mortality risk between the OSND and the MIS or the GNRI among MHD patients have never been fully compared.Of note,all the nutritional risk screening methods(the MIS,the OSND and the GNRI)were developed for use with Western populations.Whether they can be applied to the Chinese MHD population needs further verification.Although there have been studies validating some nutritional risk screening methods(except OSND)in Chinese MHD patients,the population in their studies were relatively small and limited,most of them were mainly elderly population.Besides,the follow-up time of these studies were short.Therefore,China still lacks effective clinical studies to validate the efficacy of these nutritional risk screening methods in MHD patients.Objective:The objective of this study was to compare the effect of different nutritional risk screening methods on predicting the risk of mortality in Chinese maintenance hemodialysis(MHD)patients with a large cohort and find the suitable nutritional risk screening method in Chinese MHD patients.Research Methods:This study was a multicenter,prospective cohort study conducted from January 2014 to December 2015 in eight outpatient dialysis centers,a cohort of 1,025 MHD patients were enrolled at baseline.Patients’ general demographic information,habits of diet and lifestyle,history of disease,use of medications,dialysis related parameters,physical examination,blood sample,and the MIS,the OSND,and the GNRI scores were collected at baseline.The primary outcomes of this study were all-cause mortality and cardiovascular(CV)mortality.Cox proportional hazards models were used to estimate the HRs and 95%CI for the risk of all-cause mortality and CV mortality associated with three nutritional risk screening methods Harrell’s C statistic and time-dependent receiver Operating Characteristic(ROC)curve were derived to examine the discrimination between three tools and mortality.Results:The mean age of this study population was 53.9 years old.Men and women account for 57.4%and 42.6%of the total population respectively.The median follow-up time of this study was 28.1 months.All-cause and CV mortality occurred in 226(22%)and 136(13.3%)participants during the follow-up,respectively.The results of the multiple Cox analysis showed that the MIS(per 1 standard deviation(SD)increase,HR=1.35,95%CI:1.18-1.55),the OSND(per 1 standard SD decrease,HR=1.24,95%CI:1.09-1.42)and the GNRI(per 1 standard SD decrease,HR=1.26,95%CI:1.10-1.43)were correlated with all-cause mortality.Similar results were found in CV mortality.In the stratified analysis,None of these variables,including age(<60 vs.≥60 years),sex,diabetes,history of cardiovascular disease and dialysis age(<24.5 vs.≥24.5 months),significantly modified the associations between the three nutritional risk screening methods and the risk of all-cause mortality(P for interaction>0.05 for all comparisons).More importantly,The Harrell’s C statistics of the MIS and GNRI were significantly higher than the OSND(MIS vs.OSND:P=0.001;GNRI vs.OSND:P=0.045),and there was no significant difference between the MIS and the GNRI in the values of the Harrell’s C statistic(P=0.182).Similar results were founded in CV mortality.Conclusion:Each of the three nutritional risk screening methods was significantly associated with an increased risk of all-causes and CV mortality.The mortality predictability of the MIS appears similar to the GNRI and greater than the OSND.And the process of assessing the GNRI is more simple than the MIS,which indicates the GNRI is more suitable for clinic using.