Deep Learning Prediction of Cervical Spine Surgery Revision Outcomes Using Standard Laboratory and Operative Variables.

BACKGROUND: Cervical spine procedures represent a major proportion of all spine surgery. Mitigating the revision rate following cervical procedures requires careful patient selection. While complication risk has successfully been predicted, revision risk has proven more challenging. This is likely due to the absence of granular variables in claims databases. The objective of this study was to develop a state-of-the-art model of revision prediction of cervical spine surgery using laboratory and operative variables. METHODS: Using the Stanford Research Repository, patients undergoing a cervical spine procedure between 2016 and 2022 were identified (N = 3151), and recent laboratory values were collected. Patients were classified into separate cohorts by revision outcome and time frame. Machine and deep learning models were trained to predict each revision outcome from laboratory and operative variables. RESULTS: Red blood cell count, hemoglobin, hematocrit, mean corpuscular hemoglobin concentration, red blood cell distribution width, platelet count, carbon dioxide, anion gap, and calcium all were significantly associated with ≥1 revision cohorts. For the prediction of 3-month revision, the deep neural network achieved an area under the receiver operating characteristic curve of 0.833. The model demonstrated increased performance for anterior versus posterior and arthrodesis versus decompression procedures. CONCLUSIONS: Our deep learning approach successfully predicted 3-month revision outcomes from demographic variables, standard laboratory values, and operative variables in a cervical spine surgery cohort. This work used standard laboratory values and operative codes as meaningful predictive variables for revision outcome prediction. The increased performance on certain procedures evidences the need for careful development and validation of one-size-fits-all risk scores for spine procedures.

Sigma-1 receptor expression in a subpopulation of lumbar spinal cord microglia in response to peripheral nerve injury.

Sigma-1 Receptor has been shown to localize to sites of peripheral nerve injury and back pain. Radioligand probes have been developed to localize Sigma-1 Receptor and thus image pain source. However, in non-pain conditions, Sigma-1 Receptor expression has also been demonstrated in the central nervous system and dorsal root ganglion. This work aimed to study Sigma-1 Receptor expression in a microglial cell population in the lumbar spine following peripheral nerve injury. A publicly available transcriptomic dataset of 102,691 L4/5 mouse microglial cells from a sciatic-sural nerve spared nerve injury model and 93,027 age and sex matched cells from a sham model was used. At each of three time points-postoperative day 3, postoperative day 14, and postoperative month 5-gene expression data was recorded for both spared nerve injury and Sham cell groups. For all cells, 27,998 genes were sequenced. All cells were clustered into 12 distinct subclusters and gene set enrichment pathway analysis was performed. For both the spared nerve injury and Sham groups, Sigma-1 Receptor expression significantly decreased at each time point following surgery. At the 5-month postoperative time point, only one of twelve subclusters showed significantly increased Sigma-1 Receptor expression in spared nerve injury cells as compared to Sham cells (p = 0.0064). Pathway analysis of this cluster showed a significantly increased expression of the inflammatory response pathway in the spared nerve injury cells relative to Sham cells at the 5-month time point (p = 6.74e-05). A distinct subcluster of L4/5 microglia was identified which overexpress Sigma-1 Receptor following peripheral nerve injury consistent with neuropathic pain inflammatory response functioning. This indicates that upregulated Sigma-1 Receptor in the central nervous system characterizes post-acute peripheral nerve injury and may be further developed for clinical use in the differentiation between low back pain secondary to peripheral nerve injury and low back pain not associated with peripheral nerve injury in cases where the pain cannot be localized.