Differences in time-to-fusion based on "absence of peri-graft radiolucency" and "trabecular bone bridging" criteria after transforaminal lumbar interbody fusion in patients with low and normal bone density.

OBJECTIVE: To investigate the difference in time-to-fusion between two sets of interbody fusion criteria (absence of peri-graft radiolucency vs. trabecular bone bridging), and to determine the effect of osteoporosis on time-to-fusion. MATERIALS AND METHODS: This retrospective study enrolled 79 patients treated for degenerative disease with one-level transforaminal lumbar interbody fusion from February 2012 to December 2018, and who had both pre- and post-operative CTs. Patients were divided into osteoporosis, osteopenia, and normal groups based on L1 vertebral body attenuation values in pre-operative CT with cutoff of 90 Hounsfield units (HU) and 120 HU. The osteoporosis, osteopenia, and normal groups included 36 patients (mean age: 69.9 years; 8 men and 28 women), 18 patients (mean age: 62.6 years; 7 men and 11 women), and 25 patients (mean age: 56.6 years; 15 men and 10 women), respectively. Fusion was assessed annually on post-operative CT images using absence of peri-graft radiolucency and trabecular bone bridging criteria. Time-to-fusion was estimated using the Kaplan-Meier method, and differences between the groups were examined using the log-rank test. Cox proportional hazards regression was performed. RESULTS: Time-to-fusion took significantly longer in the osteoporosis group in both fusion criteria (0.5 years in normal vs. 2 years in osteopenia vs. 3 years in osteoporosis for absence of peri-graft radiolucency; p = 0.003, and 3 years vs. 4 years vs. 5 years for trabecular bone bridging; p = 0.001). Only osteoporosis grouping was independent risk factor for slow trabecular bone fusion (hazard ratio:0.339; p = 0.003). CONCLUSION: The median time to fusion was significantly longer when using trabecular bone bridging criteria than absence of peri-graft radiolucency criteria.

Psoas muscle measurement as a predictor of recurrent lumbar disc herniation: A retrospective blind study.

AbstractParaspinal (erector spinae and multifidus) and psoas muscles contribute to spinal stability, but no study has yet examined the relationship between muscle mass and recurrent lumbar disc herniation (rLDH). The purpose of this study was to investigate the effect of psoas and paraspinal muscle mass on recurrent Lumbar disc herniation (LDH). This retrospective study included 49 patients with LDH (22 men, 27 women; mean age: 59.9 years; range 32-80) who underwent discectomy and partial laminectomy without fusion and underwent both pre- and postoperative magnetic resonance imaging. The presence of rLDH was determined using medical records and postoperative magnetic resonance imagings. Patients were divided into an rLDH group (26 patients) and a without-rLDH group (23 patients). Clinical characteristics, segmental motion, and paraspinal and psoas muscle mass were compared between the groups. Using ImageJ software, the cross-sectional area (CSA), lean muscle mass (LMM), and skeletal muscle index (SMI) were measured on T2 axial preoperative magnetic resonance images at L2-L3, L3-L4, and L4-L5 disc levels to represent muscle mass. Univariate and multivariate logistic regression analyses were performed. In the rLDH group, patients were younger (52.6 years vs 68.2 years; P = .001), segmental instability was more common (50.0% vs 4.3%; P = .001), and the CSA, LMM, CSASMI, and LMMSMI of psoas muscles were larger (5851.59 mm2 vs 4264.93 mm2, 5456.59 mm2 vs 4044.77 mm2, 18.77 cm2/m2 vs 13.86 cm2/m2, and 17.52 cm2/m2 vs 12.98 cm2/m2; P < .01 for all 4 variables). On multivariate logistic regression, age and segmental instability were independent risk factors for rLDH (odds ratio 0.886 and 18.527; P = .01 and P = .02, respectively). In middle-aged and elderly patients with lumbar disc herniation, relatively younger age, segmental instability, and greater psoas muscle mass may be risk factors for recurrence.

Development of Machine Learning Models Predicting Estimated Blood Loss during Liver Transplant Surgery.

The incidence of major hemorrhage and transfusion during liver transplantation has decreased significantly over the past decade, but major bleeding remains a common expectation. Massive intraoperative hemorrhage during liver transplantation can lead to mortality or reoperation. This study aimed to develop machine learning models for the prediction of massive hemorrhage and a scoring system which is applicable to new patients. Data were retrospectively collected from patients aged >18 years who had undergone liver transplantation. These data included emergency information, donor information, demographic data, preoperative laboratory data, the etiology of hepatic failure, the Model for End-stage Liver Disease (MELD) score, surgical history, antiplatelet therapy, continuous renal replacement therapy (CRRT), the preoperative dose of vasopressor, and the estimated blood loss (EBL) during surgery. The logistic regression model was one of the best-performing machine learning models. The most important factors for the prediction of massive hemorrhage were the disease etiology, activated partial thromboplastin time (aPTT), operation duration, body temperature, MELD score, mean arterial pressure, serum creatinine, and pulse pressure. The risk-scoring system was developed using the odds ratios of these factors from the logistic model. The risk-scoring system showed good prediction performance and calibration (AUROC: 0.775, AUPR: 0.753).

Explainable machine learning model to predict refeeding hypophosphatemia.

BACKGROUND & AIMS: Refeeding syndrome (RFS) is a disease that occurs when feeding is restarted and metabolism changes from catabolic to anabolic status. RFS can manifest variously, ranging from asymptomatic to fatal, therefore it may easily be overlooked. RFS prediction using explainable machine learning can improve diagnosis and treatment. Our study aimed to propose a machine learning model for RFS prediction, specifically refeeding hypophosphatemia, to evaluate its performance compared with conventional regression models, and to explain the machine learning classification through Shapley additive explanations (SHAP) values. METHODS: A retrospective study was conducted including 806 patients, with 2 or more days of nothing-by-mouth prescription, and with phosphate (P) level measurements within 5 days of refeeding were selected. We divided the patients into hypophosphatemia (n = 367) and non-hypophosphatemia groups (n = 439) at a P level of 0.8 mmol/L. Among the features examined within 48 h after admission, we reviewed laboratory test results and electronic medical records. Logistic, Lasso, and ridge regressions were used as conventional models, and performances were compared with our extreme gradient boosting (XGBoost) machine learning model using the area under the receiver operating characteristic curve. Our model was explained using the SHAP value. RESULTS: The areas under the curve were 0.950 (95% confidence interval: 0.924-0.975) for our XGBoost machine learning model and surpassed the performance of conventional regression models; 0.760 (0.707-0.813) for logistic regression, 0.751 (0.694-0.807) for Lasso regression, and 0.758 (0.701-0.809) for ridge regression. According to the SHAP values in the order of importance, low initial P, recent weight loss, high creatinine, diabetes mellitus with insulin use, low haemoglobin A1c, furosemide use, intensive care unit admission, blood urea nitrogen level of 19-65, parenteral nutrition, magnesium below or above the normal range, low potassium, and older age were features to predict refeeding hypophosphatemia. CONCLUSIONS: The machine learning model for predicting RFS has a substantially higher effectiveness than conventional regression methods. Creating an accurate risk assessment tool based on machine learning for early identification of patients at risk for RFS can enable careful nutrition management planning and monitoring in the intensive care unit, towards reducing the incidence of RFS-related morbidity and mortality.