Key Performance Indicators for Acute Intermittent Renal Replacement Therapy in Critically Ill Patients: A Systematic Review

Introduction: There have been initiatives to identify key performance indicators (KPIs) for continuous renal replacement therapy (CRRT). However, no formal reviews of the evidence for KPIs of intermittent renal replacement therapy (IRRT) have been conducted. This systematic review appraised the evidence for KPIs of IRRT in critically ill patients. Methods: Ovid MEDLINE, Ovid Embase, CINAHL, and Cochrane Library were searched for studies involving KPIs for IRRT. Studies were included if they contained KPIs, occurred in critically ill patients, and were associated with IRRT. Risk of bias was evaluated using the modified Cochrane tool and the Newcastle-Ottawa Quality Assessment Scale. The analysis was primarily descriptive. Each KPI was evaluated for importance, scientific acceptability, usability, and feasibility. Results: Our search yielded 7726 citations, of which 240 fulfilled eligibility criteria. This included 201 (83.8%) observational studies, 28 (11.7%) randomized controlled studies, 4 (1.7%) case series, 4 (1.7%) economic modeling studies, and 3 (1.3%) case-controlled studies. The quality of retrieved studies was variable. A total of 11 different KPIs were mentioned in 673 instances. KPIs were classified as process (n=5, 45.5%), outcome (n=4, 36.3%), and structure (n=2, 18.2%). The most commonly identified KPI was mortality (n=166) followed by solute clearance (n=105), intradialytic hypotension (n=104), renal recovery (n=79), ultrafiltration achieved (n=60), adverse events (n=53), filter/circuit clotting (n=49), length of stay (n=40), hemodynamic stability (n=39), healthcare costs (n=14), and organization structure and leadership (n=4). Across studies, the definitions used for KPIs evaluating similar constructs varied considerably. Conclusion: We identified numerous potential KPIs of IRRT care, characterized by heterogeneous definitions, varying quality of derivation, and limited evaluation. Further study is needed to prioritize a concise inventory of KPIs to measure, benchmark, and integrate their reporting into healthcare systems for IRRT care for critically ill patients.