En bloc resection for primary bone tumours and isolated metastasis are complex surgeries associated with a high rate of adverse events (AEs). The primary objective of this study was to explore the relationship between frailty/sarcopenia and major perioperative AEs following en bloc resection for primary bone tumours or isolated metastases of the spine. Secondary objectives were to report the prevalence and distribution of frailty and sarcopenia, and determine the relationship between these factors and length of stay (LOS), unplanned reoperation, and 1-year postoperative mortality in this population.

This is a retrospective study of prospectively collected data from a single quaternary care referral center consisting of patients undergoing an elective en bloc resection for a primary bone tumour or an isolated spinal metastasis between January 1st, 2009 and February 28th, 2020. Frailty was calculated with the modified frailty index (mFI) and spine tumour frailty index (STFI). Sarcopenia, determined by the total psoas area (TPA) vertebral body (VB) ratio (TPA/VB), was measured at L3 and L4. Regression analysis produced ORs, IRRs, and HRs that quantified the association between frailty/sarcopenia and major perioperative AEs, LOS, unplanned reoperation and 1-year postoperative mortality.

One hundred twelve patients met the inclusion criteria. Using the mFI, five patients (5%) were frail (mFI ³ 0.21), while the STFI identified 21 patients (19%) as frail (STFI ³ 2). The mean CT ratios were 1.45 (SD 0.05) and 1.81 (SD 0.06) at L3 and L4 respectively. Unadjusted analysis demonstrated that sarcopenia and frailty were not significant predictors of major perioperative AEs, LOS or unplanned reoperation. Sarcopenia defined by the CT L3 TPA/VB and CT L4 TPA/VB ratios significantly predicted 1-year mortality (HR of 0.32 per one unit increase, 95% CI 0.11-0.93, p=0.04 vs. HR of 0.28 per one unit increase, 95% CI 0.11-0.69, p=0.01) following unadjusted analysis. Frailty defined by an STFI score ≥ 2 predicted 1-year postoperative mortality (OR of 2.10, 95% CI 1.02-4.30, p=0.04).

The mFI was not predictive of any clinical outcome in patients undergoing en bloc resection for primary bone tumours or isolated metastases of the spine. Sarcopenia defined by the CT L3 TPA/VB and L4 TPA/VB and frailty assessed with the STFI predicted 1-year postoperative mortality on univariate analysis but not major perioperative AEs, LOS or reoperation. Further investigation with a larger cohort is needed to identify the optimal measure for assessing frailty and sarcopenia in this spine population.

Despite advances in treating acute spinal cord injury (SCI), measures to mitigate permanent neurological deficits in affected patients are limited. Augmentation of mean arterial blood pressure (MAP) to promote blood flow and oxygen delivery to the injured cord is one of the only currently available treatment options to potentially improve neurological outcomes after acute spinal cord injury (SCI). However, to optimize such hemodynamic management, clinicians require a method to measure and monitor the physiological effects of these MAP alterations within the injured cord in real-time. To address this unmet clinical need, we developed a series of miniaturized optical sensors and a monitoring system based on multi-wavelength near-infrared spectroscopy (MW-NIRS) technique for direct transdural measurement and continuous monitoring of spinal cord hemodynamics and oxygenation in real-time. We conducted a feasibility study in a porcine model of acute SCI. We also completed two separate animal studies to examine the function of the sensor and validity of collected data in an acute experiment and a seven-day post-injury survival experiment.

In our first animal experiment, nine Yorkshire pigs underwent a weight-drop T10 vertebral level contusion-compression injury and received episodes of ventilatory hypoxia and alterations in MAP. Spinal cord hemodynamics and oxygenation were monitored throughout by a transdural NIRS sensor prototype, as well as an invasive intraparenchymal (IP) sensor as a comparison. In a second experiment, we studied six Yucatan miniature pigs that underwent a T10 injury. Spinal cord oxygenation and hemodynamics parameters were continuously monitored by an improved NIRS sensor over a long period. Episodes of MAP alteration and hypoxia were performed acutely after injury and at two- and seven-days post-injury to simulate the types of hemodynamic changes patients experience after an acute SCI. All NIRS data were collected in real-time, recorded and analyzed in comparison with IP measures.

Noninvasive NIRS parameters of tissue oxygenation were highly correlated with invasive IP measures of tissue oxygenation in both studies. In particular, during periods of hypoxia and MAP alterations, changes of NIRS-derived spinal cord tissue oxygenation percentage were significant and corresponded well with the changes in spinal cord oxygen partial pressures measured by the IP sensors (p < 0.05).

Our studies indicate that a novel optical biosensor developed by our team can monitor real-time changes in spinal cord hemodynamics and oxygenation over the first seven days post-injury and can detect local tissue changes that are reflective of systemic hemodynamic changes. Our implantable spinal cord NIRS sensor is intended to help clinicians by providing real-time information about the effects of hemodynamic management on the injured spinal cord. Hence, our novel NIRS system has the near-term potential to impact clinical care and improve neurologic outcomes in acute SCI. To translate our studies from bench to bedside, we have developed an advanced clinical NIRS sensor that is ready to be implanted in the first cohort of acute SCI patients in 2022.