Sarcopenia as a Prognostic Factor for 90-Day and Overall Mortality in Patients Undergoing Spine Surgery for Metastatic Tumors: A Multicenter Retrospective Cohort Study

BACKGROUND Novel methods in predicting survival in patients with spinal metastases may help guide clinical decision-making and stratify treatments regarding surgery vs palliative care. OBJECTIVE To evaluate whether the frailty/sarcopenia paradigm is predictive of survival and morbidity in patients undergoing surgery for spinal metastasis. METHODS A total of 271 patients from 4 tertiary care centers who had undergone surgery for spinal metastasis were identified. Frailty/sarcopenia was defined by psoas muscle size. Survival hazard ratios were calculated using multivariate analysis, with variables from demographic, functional, oncological, and surgical factors. Secondary outcomes included improvement of neurological function and postoperative morbidity. RESULTS Patients in the smallest psoas tertile had shorter overall survival compared to the middle and largest tertile. Psoas size (PS) predicted overall mortality more strongly than Tokuhashi score, Tomita score, and Karnofsky Performance Status (KPS). PS predicted 90-d mortality more strongly than Tokuhashi score, Tomita score, and KPS. Patients with a larger PS were more likely to have an improvement in deficit compared to the middle tertile. PS was not predictive of 30-d morbidity. CONCLUSION In patients undergoing surgery for spine metastases, PS as a surrogate for frailty/sarcopenia predicts 90-d and overall mortality, independent of demographic, functional, oncological, and surgical characteristics. The frailty/sarcopenia paradigm is a stronger predictor of survival at these time points than other standards. PS can be used in clinical decision-making to select which patients with metastatic spine tumors are appropriate surgical candidates.

CMET-15. SARCOPENIA INDEPENDENTLY AND STRONGLY PREDICTS SURVIVAL IN PATIENTS UNDERGOING SPINE SURGERY FOR METASTATIC TUMORS

INTRODUCTION Predicting survival and surgical morbidity in patients with spinal metastases would help guide clinical decision making and stratify treatments between surgical intervention and palliative care. This multi-center retrospective cohort study evaluates whether the frailty/sarcopenia paradigm, as measured by psoas size, is predictive of survival in patients undergoing surgery for spinal metastasis. METHODS 271 patients from four institutes who had undergone surgery for spinal metastasis were identified. Morphometric measurements were taken of the psoas muscle at the L4 vertebral level < 200d from surgery. Mortality hazard ratios were calculated using multivariate analysis, with variables included from past medical history, type and extent of tumor spread, type and intensity of surgery, and postoperative chemotherapy or radiation. RESULTS Psoas size was predictive of overall mortality; patients in the smallest tertile had shorter overall survival compared to the middle (OR 0.52, p< 0.001) and largest tertile (OR 0.45, p< 0.001). Psoas size predicted overall mortality more strongly than Tokuhashi score (OR 0.91, p= 0.010), Tomita score (OR 1.07, p= 0.04), and KPS (OR 0.99, p= 0.58). Psoas size was also predictive of 90-day survival; patients in the smallest tertile had shorter 90-day survival compared to the middle (OR 0.24, p= 0.003) and largest tertile (OR 0.16, p= 0.001). Psoas size predicted 90-day mortality more strongly than Tokuhashi score (OR 0.73, p= 0.002), Tomita score (OR 1.00, p= 0.92), and KPS (OR 0.98, p= 0.39). CONCLUSION In patients undergoing surgery for spine metastases, psoas size as a surrogate for frailty/sarcopenia predicts 90-day and overall mortality, independent of demographical, functional, oncological, and surgical characteristics. The sarcopenia/frailty paradigm is a stronger predictor of survival at these time points than the Tokuhashi score, Tomita score, and KPS. Psoas size can be used in clinical decision-making to select which patients with metastatic spine tumors are appropriate surgical candidates.

Percutaneous Vertebroplasty Relieves Pain in Cervical Spine Metastases

Percutaneous vertebroplasty (PVP) has been shown to release spinal pain and stabilize the vertebral body. PVP is suggested as an alternative treatment in spinal metastasis. Although cervical metastases is less prevalent than thoracic and lumbar spine, PVP procedure in cervical vertebrae remains technical challenging. We retrospectively analyzed the data from patients (n=9) who underwent PVP using anterolateral approach to treat severe neck pain and restricted cervical mobility from metastatic disease. Patients were rated using modified Tokuhashi score and Tomita score before the procedure. Visual analog scale (VAS), neck disability index (NDI), analgesic use, and imaging (X-ray or CT) were evaluated before PVP and 3 days, 3 months, and 6 months after PVP. All patients were in late stage of cancer evaluated using modified Tokuhashi and Tomita score. The cement leakage rate was 63.6% (14 of the 22 vertebrae) with no severe complications. VAS, NDI, and analgesic use were significantly decreased 3 days after the procedure and remained at low level until 6 months of follow-up. Our result suggested PVP effectively released the pain from patients with cervical metastasis. The results warrant further clinical investigation.

Survival score scales of patients operated with spinal metastases: retrospective application in a Brazilian population

ABSTRACT Spinal cord epidural metastasis (SEM) is a common complication of systemic cancer. Predicting these patient’s survival is a key factor to select the proper treatment modality, but the three most used score scales to predict their survival (Tokuhashi revised score, Tomita score and Bauer modified score) were designed in single institutions and their reliability to predict correctly the patient’s survival were first tested only in those specific populations. This prognostication issue is addressed in this article, evaluating retrospectively the survival of 17 patients with SEM from a Brazilian general hospital with these score scales. Our results show that the actual survival of those patients were worse than the predicted of all three score scales, suggesting that differences between the different populations might have affected their reliability and alert that their usage as a major factor to select the most appropriate treatment have to be done with caution.