Identifying risk factors for poor multidimensional recovery after major surgery: A systematic review.

Unlabelled: Traditional risk factors used for predicting poor postoperative recovery have focused on postoperative complications, adverse symptoms (nausea, pain), length of hospital stay, and patient quality of life. Despite these being traditional performance indicators of patient postoperative "status," they may not fully define the multidimensional nature of patient recovery. The definition of postoperative recovery is thus evolving to include patient-reported outcomes that are important to the patient. Previous reviews have focused on risk factors for the above traditional outcomes after major surgery. Yet, there remains a need for further study of risk factors predicting multidimensional patient-focused recovery, and investigation beyond the immediate postoperative period after patients are discharged from the hospital. This review aimed to appraise the current literature identifying risk factors for multidimensional patient recovery.

Methods: A systematic review without meta-analysis was performed to qualitatively summarize preoperative risk factors for multidimensional recovery 4-6 weeks after major surgery (PROSPERO, CRD42022321626). We reviewed three electronic databases between January 2012 and April 2022. The primary outcome was risk factors for multidimensional recovery at 4-6 weeks. A GRADE quality appraisal and a risk of bias assessment were completed.

Results: In total, 5150 studies were identified, after which 1506 duplicates were removed. After the primary and secondary screening, nine articles were included in the final review. Interrater agreements between the two assessors for the primary and secondary screening process were 86% (k = 0.47) and 94% (k = 0.70), respectively. Factors associated with poor recovery were found to include ASA grade, recovery tool baseline score, physical function, number of co-morbidities, previous surgery, and psychological well-being. Mixed results were reported for age, BMI, and preoperative pain. Due to the observational nature, heterogeneity, multiple definitions of recovery, and moderate risk of bias of the primary studies, the quality of evidence was rated from very low to low.

Conclusion: Our review found that there were few studies assessing preoperative risk factors as predictors for poor postoperative multidimensional recovery. This confirms the need for higher quality studies assessing risk for poor recovery, ideally with a consistent and multi-dimensional definition of recovery.