Early Renal Replacement Therapy Versus Standard Care in the ICU: A Systematic Review, Meta-Analysis, and Cost Analysis. (April 2019)

• Record Type: Journal Article
• Authors: Chaudhuri, Dipayan; Herritt, Brent; Heyland, Daren; Gagnon, Louis-Philippe; Thavorn, Kednapa; Kobewka, Daniel; Kyeremanteng, Kwadwo
• Is Part Of: Journal of intensive care medicine. Volume 34:Number 4(2019)
• Page Start: 323
• Page End: 329
• Publication Date: 2019-04
• Subjects: early dialysis -- cost analysis -- systematic review; Critical care medicine -- Periodicals; Critical Care -- Periodicals; Soins intensifs -- Périodiques; Soins intensifs; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.02805
• DOI: 10.1177/0885066617698635
• Languages: English
• ISSNs: 0885-0666

Abstract

Objective: Renal replacement therapy (RRT) is the treatment of choice for severe acute kidney injury, but there are no firm guidelines as to the time of initiation of RRT in the critically ill. The primary objective of this study is to determine 1-month mortality rates of early versus late dialysis in critical care. As secondary end points, we provide a cost analysis of early versus late RRT initiation, intensive care unit (ICU) length of stay (LOS), hospital LOS, and number of patients on dialysis at day 60 postrandomization. Data Sources: We identified all randomized controlled trials (RCTs) through EMLINE and MEDBASE that examined adult patients admitted to critical care who were randomized to receiving early dialysis versus standard of care. Study Selection: Inclusion criteria: (1) RCTs conducted after the year 2000, (2) the population evaluated had to be adults admitted to ICU, (3) the intervention had to be early RRT versus standard care, and (4) outcomes had to measure patient mortality. Data Extraction: Two independent investigators reviewed search results and identified appropriate studies. Information was extracted using standardized case report forms. Data Synthesis: Overall, 7 RCTs were included with a total of 1400 patients. Early RRT showed no survival benefit when compared to standard treatment (odds ratio [OR], 0.90 95% confidence interval [95% CI] 0.70-1.15, P = .39). There was no significant difference in length of hospital stay in patients with early RRT (−1.55 days [95% CI −4.75 to 1.65, P = .34]), in length of ICU stay (−0.79 days [95% CI −2.09 to 0.52], P = .24), or proportion of patients on dialysis at day 60 (OR 0.93 [95% CI 0.62 to 1.43], P = .79). Per patient, there is likely a small increase in costs (<US$1000) owing to increased total dialysis. Conclusion: Across all measured domains, there is no clear benefit to early RRT. Moreover, this intervention may result in increased costs and exposes patients to an invasive therapy with potential harm.