Biomarkers associated with mortality in patients undergoing dialysis

Abstract

SUMMARY Background Survival for patients commencing renal replacement therapy is around 90% in the first year and 83% at two years after starting dialysis. The factors that appear to predict mortality are comorbidity and frailty associated with kidney disease, glomerular filtration rate, age and biochemical factors. Objectives To analyse the condition of patients starting renal replacement therapy, based on biomarkers commonly used in clinical practice and their association with mortality, measured 6 and 12 months after initiating replacement therapy. Methods A one‐year prospective follow‐up study with 189 patients. Sociodemographic variables, aetiology of renal disease, comorbidities, prior nephrology service monitoring, prior renal transplantation and biochemical parameters at the time of initiating replacement therapy were analysed. Results The overall percentage of death during the one‐year follow‐up was 6.87%, with 64% of deaths occurring during the first six months. The only variable independently associated with mortality was low albumin levels. Conclusion Although most patients in this centre are monitored by a nephrologist prior to starting replacement therapy, many nevertheless fail to achieve the biochemical targets recommended. One such parameter is albumin, which proved at the start of replacement therapy to be an independent predictor of mortality. Findings of this study show the need to intervene on certain biochemical parameters during the pre‐dialysis stage and at the start of dialysis, in order to improve survival in this group of patients.

Early Mortality Rates After Commencement of Maintenance Hemodialysis: A Systematic Review and Meta‐Analysis

“…The 32 studies included in this review represent a combined population of 1 083 264 new dialysis starters and 283 277 person‐years of observation for early mortality. The characteristics of these studies are shown in Table .…”

“…The characteristics of these studies are shown in Table . Studies originating in North America accounted for the majority of observations (12 studies, population: 831559), while the remaining observations originated in Europe (10 studies, population: 93 815 patients), Asia (four studies, combined population of 46 421 patients), Australia‐New Zealand (one study; population: 23 658), Africa (three studies, population: population of 530 patients), and South America (one study, 395 participants). There was also one multinational study , involving developed countries with 86 886 participants.…”

“…The summary of quality assessment of studies is presented in Appendix S1. Overall, 13 studies (combined population 967 282) , were judged to be at low risk of bias (high quality) and 19 (combined population 115 982) at high risk of bias (low quality). The main contributory factor for being assessed at high risk of bias included reliance on registry data where the completion of entries in the registry where uncertain , restricting participation to selected groups such as the elderly and diabetic subjects , studies with high probability of acute cases being included in the cohort , eligibility criteria of study participants or the source population unclear , method of outcome ascertainment not fully described , exclusion of eligible patients , and possible selective reporting of outcomes with high numbers of censoring events other than all‐cause mortality .…”

“…The number of patients who died in the early period, after the commencement of HD ranged between 3.9% and 30.1% (see Table ). The calculated early morality rates ranged between 11.0 and 122.0 per 100 person‐years. The weighted average of early mortality rates in all the included studies was 30.4 (95% CI 29.8–31.0) per 100 person‐years.…”

“…Three studies recruited their first patient in the 1970s , three in the 1980s , nine in the 1990s , and 16 in the period after 2000s . One study did not specify when their first patient was recruited .…”