Development of a machine learning algorithm for predicting in-hospital and 1-year mortality after traumatic spinal cord injury.

BACKGROUND CONTEXT: Current prognostic tools such as the Injury Severity Score (ISS) that predict mortality following trauma do not adequately consider the unique characteristics of traumatic spinal cord injury (tSCI). PURPOSE: Our aim was to develop and validate a prognostic tool that can predict mortality following tSCI. STUDY DESIGN: Retrospective review of a prospective cohort study. PATIENT SAMPLE: Data was collected from 1245 persons with acute tSCI who were enrolled in the Rick Hansen Spinal Cord Injury Registry between 2004 and 2016. OUTCOME MEASURES: In-hospital and 1-year mortality following tSCI. METHODS: Machine learning techniques were used on patient-level data (n=849) to develop the Spinal Cord Injury Risk Score (SCIRS) that can predict mortality based on age, neurological level and completeness of injury, AOSpine classification of spinal column injury morphology, and Abbreviated Injury Scale scores. Validation of the SCIRS was performed by testing its accuracy in an independent validation cohort (n=396) and comparing its performance to the ISS, a measure which is used to predict mortality following general trauma. RESULTS: For 1-year mortality prediction, the values for the Area Under the Receiver Operating Characteristic Curve (AUC) for the development cohort were 0.84 (standard deviation=0.029) for the SCIRS and 0.55 (0.041) for the ISS. For the validation cohort, AUC values were 0.86 (0.051) for the SCIRS and 0.71 (0.074) for the ISS. For in-hospital mortality, AUC values for the development cohort were 0.87 (0.028) and 0.60 (0.050) for the SCIRS and ISS, respectively. For the validation cohort, AUC values were 0.85 (0.054) for the SCIRS and 0.70 (0.079) for the ISS. CONCLUSIONS: The SCIRS can predict in-hospital and 1-year mortality following tSCI more accurately than the ISS. The SCIRS can be used in research to reduce bias in estimating parameters and can help adjust for coefficients during model development. Further validation using larger sample sizes and independent datasets is needed to assess its reliability and to evaluate using it as an assessment tool to guide clinical decision-making and discussions with patients and families.

Ketogenic diet improves forelimb motor function after spinal cord injury in rodents.

High fat, low carbohydrate ketogenic diets (KD) are validated non-pharmacological treatments for some forms of drug-resistant epilepsy. Ketones reduce neuronal excitation and promote neuroprotection. Here, we investigated the efficacy of KD as a treatment for acute cervical spinal cord injury (SCI) in rats. Starting 4 hours following C5 hemi-contusion injury animals were fed either a standard carbohydrate based diet or a KD formulation with lipid to carbohydrate plus protein ratio of 3:1. The forelimb functional recovery was evaluated for 14 weeks, followed by quantitative histopathology. Post-injury 3:1 KD treatment resulted in increased usage and range of motion of the affected forepaw. Furthermore, KD improved pellet retrieval with recovery of wrist and digit movements. Importantly, after returning to a standard diet after 12 weeks of KD treatment, the improved forelimb function remained stable. Histologically, the spinal cords of KD treated animals displayed smaller lesion areas and more grey matter sparing. In addition, KD treatment increased the number of glucose transporter-1 positive blood vessels in the lesion penumbra and monocarboxylate transporter-1 (MCT1) expression. Pharmacological inhibition of MCTs with 4-CIN (α-cyano-4-hydroxycinnamate) prevented the KD-induced neuroprotection after SCI, In conclusion, post-injury KD effectively promotes functional recovery and is neuroprotective after cervical SCI. These beneficial effects require the function of monocarboxylate transporters responsible for ketone uptake and link the observed neuroprotection directly to the function of ketones, which are known to exert neuroprotection by multiple mechanisms. Our data suggest that current clinical nutritional guidelines, which include relatively high carbohydrate contents, should be revisited.

Characterization of a cervical spinal cord hemicontusion injury in mice using the infinite horizon impactor.

The majority of clinical spinal cord injuries (SCIs) are contusive and occur at the cervical level of the spinal cord. Most scientists and clinicians agree that the preclinical evaluation of novel candidate treatments should include testing in a cervical SCI contusion model. Because mice are increasingly used because of the availability of genetically engineered lines, we characterized a novel cervical hemicontusion injury in mice using the Infinite Horizon Spinal Cord Impactor (Precisions Systems & Instrumentation, Lexington, KY). In the current study, C57BL/6 mice received a hemicontusion injury of 75 kilodynes with or without dwell time in an attempt to elicit a sustained moderate-to-severe motor deficit. Hemicontusion injuries without dwell time resulted in sustained deficits of the affected forepaw, as revealed by a 3-fold decrease in usage during rearing, a ∼50% reduction in grooming scores, and retrieval of significantly fewer pellets on the Montoya staircase test. Only minor transient deficits were observed in grasping force. CatWalk analysis revealed reduced paw-print size and swing speed of the affected forelimb. Added dwell time of 15 or 30 sec significantly worsened behavioral outcome, and mice demonstrated minimal ability of grasping, paw usage, and overground locomotion. Besides worsening of behavioral deficits, added dwell time also reduced residual white and gray matter at the epicenter and rostral-caudal to the injury, including on the contralateral side of the spinal cord. Taken together, we developed and characterized a new hemicontusion SCI model in mice that produces sufficient and sustained impairments in gross and skilled forelimb function and produced primarily unilateral functional deficits.