Preoperative frailty to predict short-term outcomes in patients with cancer.

810 Background: Frailty is associated with increased postoperative complications. Despite multiple evidence-based modalities to assess frailty, surgeons often make subjective assessments of patients’ ability to tolerate surgery. Identification of those at risk of poor outcomes is particularly important in Surgical Oncology, as major resections may worsen debility and functional status, limiting the patients' ability to receive further therapy. The Risk Analysis Index (RAI-C) is a validated tool to assess frailty in surgical patients. It assigns points for any cancer diagnosis, potentially overestimating frailty in Surgical Oncology patients, and has not been validated in this subgroup. Methods: We performed a prospective, surgeon-blinded study to correlate results of a preoperatively administered frailty survey to predict short- and long-term outcomes in patients with histologically proven malignancy who underwent surgery. The RAI-C, Patient Health Questionnaire (PHQ9), and General Anxiety Disorder (GAD) surveys were administered to assess for frailty, depression, and anxiety, respectively; C-reactive protein (CRP), pre-albumin, albumin, and hematocrit were obtained. Outcomes, including readmission, major complications, ICU admission, disposition (home vs rehabilitation/nursing facility), and 1-year survival were compared between the different groups as defined by survey and lab results. Results: One hundred ninety-one patients were included, of which 135 patients were considered frail based on the predefined value (RAI-C ≥21). Frail patients were more likely to be readmitted within 30 days (14% vs 2%, P=0.01). No difference was found in length of stay, 90-day readmission, ICU admission, major complications, or disposition. Only 15.8% of patients met criteria for anxiety and 11% met criteria for depression. Measured outcomes were similar in patients defined as anxious or depressed. Patients with abnormal pre-albumin levels were more likely to require ICU admissions (31% vs 8%, P <0.005), postoperative intubation (12% vs 0.02%, P=0.02), and to be discharged to a facility other than home (19% vs 10%, P=0.01). Similar trends were noted for patients with abnormal albumin levels who also had longer lengths of stay (mean 9.32 vs 7.88 days, P=0.007). There was an association of improved 1-year survival in patients with normal CRP, pre-albumin, albumin, and hematocrit levels. Frailty as determined by RAI-C survey was not associated with decreased 1-year survival. Conclusions: The RAI-C is a validated tool that can be used for prescreening of Surgical Oncology patients at risk of postoperative readmission, however the score did not correlate with other important outcomes including major morbidity, discharge disposition, and 1-year survival. Results from the survey, as well as preoperative laboratory results can be used in screening patients who may benefit from preoperative optimization and increased follow-up.