• Oleksa G Rewa, Pierre-Marc Villeneuve, Philippe Lachance, Dean T Eurich, Henry T Stelfox, GibneyR T NoelRTNDepartment of Critical Care Medicine, Faculty of Medicine and Dentistry, University of Alberta, 8440 112 St. NW, Critical Care Medicine 2-124E Clinical Sciences Building, Edmonton, Alberta, T6G 2B7, Canada., Lisa Hartling, Robin Featherstone, and Sean M Bagshaw.
• Department of Critical Care Medicine, Faculty of Medicine and Dentistry, University of Alberta, 8440 112 St. NW, Critical Care Medicine 2-124E Clinical Sciences Building, Edmonton, Alberta, T6G 2B7, Canada. rewa@ualberta.ca.
• Intensive Care Med. 2017 Jun 1; 43 (6): 750-763.

ObjectivesRenal replacement therapy is increasingly utilized in the intensive care unit (ICU), of which continuous renal replacement therapy (CRRT) is most common. Despite CRRT being a relatively invasive and resource intensive technology, there remains wide practice variation in its application. This systematic review appraised the evidence for quality indicators (QIs) of CRRT care in critically ill patients.DesignA comprehensive search strategy was developed and performed in five citation databases (Medline, Embase, CINAHL, Cochrane Library, and PubMed) and select grey literature sources. Two reviewers independently screened, selected, and extracted data using standardized forms. Each retrieved citation was appraised for quality using the Newcastle-Ottawa Scale (NOS) and Cochrane risk of bias tool. Data were summarized narratively.Measurements And Main ResultsOur search yielded 8374 citations, of which 133 fulfilled eligibility. This included 97 cohort studies, 24 randomized controlled trials, 10 case-control studies, and 2 retrospective medical audits. The quality of retrieved studies was generally good. In total, 18 QIs were identified that were mentioned in 238 instances. Identified QIs were classified as related to structure (n = 4, 22.2 %), care processes (n = 9, 50.0 %), and outcomes (n = 5, 27.8 %). The most commonly mentioned QIs focused on filter lifespan (n = 98), small solute clearance (n = 46), bleeding (n = 30), delivered dose (n = 19), and treatment interruption (n = 5). Across studies, the definitions used for QIs evaluating similar constructs varied considerably. When identified, QIs were most commonly described as important (n = 144, 48.3 %), scientifically acceptable (n = 32, 10.7 %), and useable and/or feasible (n = 17, 5.7 %) by their primary study authors.ConclusionsWe identified numerous potential QIs of CRRT care, characterized by heterogeneous definitions, varying quality of derivation, and limited evaluation. Further study is needed to prioritize a concise inventory of QIs to measure, improve, and benchmark CRRT care for critically ill patients.Systematic Review RegistrationPROSPERO CRD42015015530.

20 keywords...

hide…