Automated prediction of isthmus areas in scar-related atrial tachycardias using artificial intelligence.

INTRODUCTION: Ablation of scar-related reentrant atrial tachycardia (SRRAT) involves identification and ablation of a critical isthmus. A graph convolutional network (GCN) is a machine learning structure that is well-suited to analyze the irregularly-structured data obtained in mapping procedures and may be used to identify potential isthmuses. METHODS: Electroanatomic maps from 29 SRRATs were collected, and custom electrogram features assessing key tissue and wavefront properties were calculated for each point. Isthmuses were labeled off-line. Training data was used to determine the optimal GCN parameters and train the final model. Putative isthmus points were predicted in the training and test populations and grouped into proposed isthmus areas based on density and distance thresholds. The primary outcome was the distance between the centroids of the true and closest proposed isthmus areas. RESULTS: A total of 193 821 points were collected. Thirty isthmuses were detected in 29 tachycardias among 25 patients (median age 65.0, 5 women). The median (IQR) distance between true and the closest proposed isthmus area centroids was 8.2 (3.5, 14.4) mm in the training and 7.3 (2.8, 16.1) mm in the test group. The mean overlap in areas, measured by the Dice coefficient, was 11.5 ± 3.2% in the training group and 13.9 ± 4.6% in the test group. CONCLUSION: A GCN can be trained to identify isthmus areas in SRRATs and may help identify critical ablation targets.

Diabetes Is Predictive of Postoperative Outcomes and Readmission Following Posterior Lumbar Fusion.

STUDY DESIGN: Retrospective cohort study. OBJECTIVE: The present study analyzes complication rates and episode-based costs for patients with and without diabetes mellitus (DM) following posterior lumbar fusion (PLF). METHODS: PLF cases at a single institution from 2008 to 2016 were queried (n = 3226), and demographic and perioperative data were analyzed. Patients with and without the diagnosis of DM were compared using chi-square, Student's t test, and multivariable regression modeling. RESULTS: Patients with diabetes were older (63.10 vs 56.48 years, P < .001) and possessed a greater number of preoperative comorbidities (47.84% of patients had Elixhauser Comorbidity Index >0 vs 42.24%, P < .001) than did patients without diabetes. When controlling for preexisting differences, diabetes remained a significant risk factor for prolonged length of stay (OR = 1.59, 95% CI 1.26-2.01, P < .001), intensive care unit stay (OR = 1.52, 95% CI 1.07-2.17, P = .021), nonhome discharge (OR = 1.86, 95% CI 1.46-2.37, P < .001), 30-day readmission (OR = 2.15, 95% CI 1.28-3.60, P = .004), 90-day readmission (OR = 1.65, 95% CI 1.05-2.59, P = .031), 30-day emergency room visit (OR = 2.15, 95% CI 1.27-3.63, P = .004), and 90-day emergency room visit (OR = 2.27, 95% CI 1.41-3.65, P < .001). Cost modeling controlling for overall comorbidity burden demonstrated that diabetes was associated with a $1709 increase in PLF costs (CI $344-$3074, P = .014). CONCLUSIONS: The present findings indicate a correlation between diabetes and a multitude of postoperative adverse outcomes and increased costs, thus illustrating the substantial medical and financial burdens of diabetes for PLF patients. Future studies should explore preventive measures that may mitigate these downstream effects.

Perioperative Outcomes of Spinal Cord Stimulator Placement in Patients with Complex Regional Pain Syndrome Compared with Patients without Complex Regional Pain Syndrome.

BACKGROUND AND OBJECTIVE: Complex regional pain syndrome (CRPS) is a multifaceted disorder resulting in an abnormal pain response to tissue injury. Among key CRPS features are neurogenic inflammation, maladaptive plasticity, and vasomotor dysfunction, which can result in severe pain and disability. Spinal cord stimulation (SCS) is an efficacious treatment for several chronic pain conditions and may improve pain and life quality in CRPS patients with CRPS. However, little information exists regarding perioperative outcomes of patients with CRPS undergoing surgical implantation of an SCS device. METHODS: Patients were included if they underwent an SCS procedure at our institution between 2008 and 2016 for chronic pain. Cases were excluded if the procedure involved stimulator removal or if it was an outpatient procedure. Multivariate regression assessed the effect of CRPS and other clinical variables on perioperative outcomes. RESULTS: Eighty-one inpatient SCS implantation cases for chronic pain were included, with 9 patients (11.1%) having a CRPS diagnosis. The CRPS cohort received higher mean quantities of intraoperative opioids and had a lower proportion of patients reporting meaningful pain reduction (16.7%) in the 24-hour postoperative setting compared with patients without CRPS (35.9%), although this was not statistically significant. Multivariate regression modeling suggested that CRPS was a significant predictor of increased odds of extended time to the postanesthesia care unit discharge (P = 0.0406) and higher direct costs of hospitalization (P = 0.0326). CONCLUSIONS: Our data suggest that CRPS may pose several unique risks in the perioperative period after inpatient SCS implantation. These findings support the need for future prospective investigations examining risks and outcomes for SCS procedures in this population.

Spatial mapping of collagen content and structure in human intervertebral disk degeneration.

Collagen plays a key structural role in both the annulus fibrosus (AF) and nucleus pulposus (NP) of intervertebral disks (IVDs). Changes in collagen content with degeneration suggest a shift from collagen type II to type I within the NP, and the activation of pro-inflammatory factors is indicative of fibrosis throughout. While IVD degeneration is considered a fibrotic process, an increase in collagen content with degeneration, reflective of fibrosis, has not been demonstrated. Additionally, changes in collagen content and structure in human IVDs with degeneration have not been characterized with high spatial resolution. The collagen content of 23 human lumbar L2/3 or L3/4 IVDs was quantified using second harmonic generation imaging (SHG) and multiple image processing algorithms, and these parameters were correlated with the Rutges histological degeneration grade. In the NP, SHG intensity increased with degeneration grade, suggesting fibrotic collagen deposition. In the AF, the entropy of SHG intensity was reduced with degeneration indicating increased collagen uniformity and suggesting less-organized lamellar structure. Collagen orientation entropy decreased throughout most IVD regions with increasing degeneration grade, further supporting a loss in collagen structural complexity. Overall, SHG imaging enabled visualization and quantification of IVD collagen content and organization with degeneration. There was an observed shift from an initially complex structure to more uniform structure with loss of microstructural elements and increased NP collagen polarity, suggesting fibrotic remodeling.