Which frailty scales for patients with adult spinal deformity are feasible and adequate? A systematic review

Background Context: Frailty as a concept is not yet fully understood, and is not the same as comorbidity. It is associated with an increased risk of adverse events and mortality after surgery, which makes its preoperative assessment significant. Despite its relevance, it still remains unclear which scales are appropriate for use in patients with spinal pathology. Purpose: To evaluate the feasibility and measurement properties of frailty scales for spine patients, specifically with adult spinal deformity (ASD), and to propose adequate scales for primary triage to prevent surgery in too frail patients and for preoperative assessment to modify patients’ condition and surgical plans. Study Design/Setting: Systematic review. Methods: Systematic search was performed between 2010 and 2021 including terms relating to spinal disorders, frailty scales, and methodological quality. Characteristics of the studies and frailty scales and data describing relation to treatment outcomes were extracted. The risk of bias was determined with the QAREL score. Results: Of the 1993 references found, 88 original studies were included and 23 scales were identified. No prospective interventional study was found where the preoperative frailty assessment was implemented. Predictive value of scales for surgical outcomes varied, dependent on spinal disorders, type of surgeries, patients’ age and frailty at baseline, and outcomes. Seventeen studies reported measurement properties of eight scales but these studies were not free of bias. In 30 ASD studies, ASD-Frailty Index (ASD-FI, n=14) and 11-item modified Frailty Index (mFI-11, n=11) were most frequently used. These scales were mainly studied in registry studies including young adult population, and carry a risk of sample bias and make their validity in elderly population unclear. ASD-FI covers multidisciplinary concepts of frailty with 40 items but its feasibility in clinical practice is questionable due to its length. The Risk Analysis Index, another multidisciplinary scale with 14 items, has been implemented for preoperative assessment in other surgical domains and was proven to be feasible and effective in interventional prospective studies. The FRAIL is a simple questionnaire with five items and its predictive value was confirmed in prospective cohort studies in which only elderly patients were included. Conclusions: No adequate scale was identified in terms of methodological quality and feasibility for daily practice. Careful attention should be paid when choosing an adequate scale, which depends on the setting of interest (eg triage or preoperative work-up). We recommend to further study a simple and predictive scale such as FRAIL for primary triage and a comprehensive and feasible scale such as Risk Analysis Index for preoperative assessment for patients undergoing spine surgery, as their adequacy has been shown in other medical domains.