115 PROTEIN PROVISION AND EXERCISE THERAPY TO IMPROVE OUTCOMES IN ICU-SURVIVORS 7 Exclusion criteria were: (I) background articles, systematic reviews, conference abstracts, case studies, diagnostic accuracy studies, in vitro, paediatric, and animal studies, (II) studies that did not report protein target or amount of protein delivery, (III) RCTs reporting mortality and no other relevant outcomes for this review. 7.2.3 Search strategy The database search was limited from 1 January 2014 to adhere to the American Society for Parenteral and Enteral Nutrition (ASPEN) guidelines for adult critically ill patients5 and increasing awareness of the condition PICS around this period.2 PubMed, Embase, and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched using Medical Subject Headings (MeSH) terms, and truncation of keywords regarding patients, interventions, and study designs related to our research question. Keywords included for example: critically ill, intensive care unit, mechanically ventilated, protein, amino acid, nutritional support, enteral nutrition, parenteral nutrition, physical therapy, exercise therapy, mobilisation, rehabilitation and clinical study. The PubMed’s Clinical Study Category search filter (Therapy, Sensitive/Broad) based on the work of Haynes RB et al38 was applied to increase the sensitivity of the search. Records from nutritional guidelines5,39 and systematic reviews25,26 were screened, without the restriction of publication year. Detailed search strategies are available in the Appendix 2; supplement 1. 7.2.4 Data collection Screening and assessment of eligibility were performed independently by two reviewers. Decisions were supported by using checklists, deliberating the results until consensus was reached. In case of disagreement, a third reviewer was consulted if necessary. TheCochrane checklist “study features for including NRSI”36 was used for the decision regarding the eligibility of NRSI, and studies were excluded when overall judgment was “no”. The level of inter-observer agreement was determined in each stage with Cohen’s Kappa (к).40 Values of к were interpreted as follows: <0.20 poor, 0.21-0.40 fair, 0.41-0.60 moderate, 0.61-0.80 good and ≥0.81 excellent agreement. All references were stored in EndNote software (X9.3.3, Bld13966) and exported to the Rayyan Qatar Computing Research Institute tool (http://rayyan.qcri.org/) after removing duplicates. 7.2.5 Data extraction Data items were independently collected using a standardised data extraction form, with relevant adjustments made after piloting. Results were compared to
RkJQdWJsaXNoZXIy MjY0ODMw