Screening Performance of S100 Calcium-Binding Protein B, Glial Fibrillary Acidic Protein, and Ubiquitin C-Terminal Hydrolase L1 for Intracranial Injury Within Six Hours of Injury and Beyond.

The Scandinavian NeuroTrauma Committee (SNC) guidelines recommend S100 calcium-binding protein B (S100B) as a screening tool for early detection of Traumatic brain injury (TBI) in patients presenting with an initial Glasgow Coma Scale (GCS) of 14-15. The objective of the current study was to compare S100B's diagnostic performance within the recommended 6-h window after injury, compared with glial fibrillary acidic protein (GFAP) and UCH-L1. The secondary outcome of interest was the ability of these biomarkers in detecting traumatic intracranial pathology beyond the 6-h mark. The Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) core database (2014-2017) was queried for data pertaining to all TBI patients with an initial GCS of 14-15 who had a blood sample taken within 6 h of injury in which the levels of S100B, GFAP, and UCH-L1 were measured. As a subgroup analysis, data involving patients with blood samples taken within 6-9 h and 9-12 h were analyzed separately for diagnostic ability. The diagnostic ability of these biomarkers for detecting any intracranial injury was evaluated based on the area under the receiver operating characteristic curve (AUC). Each biomarker's sensitivity, specificity, and accuracy were also reported at the cutoff that maximized Youden's index. A total of 531 TBI patients with GCS 14-15 on admission had a blood sample taken within 6 h, of whom 24.9% (n = 132) had radiologically confirmed intracranial injury. The AUCs of GFAP (0.86, 95% confidence interval [CI]: 0.82-0.90) and UCH-L1 (0.81, 95% CI: 0.76-0.85) were statistically significantly higher than that of S100B (0.74, 95% CI: 0.69-0.79) during this time. There was no statistically significant difference in the predictive ability of S100B when sampled within 6 h, 6-9 h, and 9-12 h of injury, as the p values were >0.05 when comparing the AUCs. Overlapping AUC 95% CI suggests no benefit of a combined GFAP and UCH-L1 screening tool over GFAP during the time periods studied [0.87 (0.83-0.90) vs. 0.86 (0.82-0.90) when sampled within 6 h of injury, 0.83 (0.78-0.88) vs. 0.83 (0.78-0.89) within 6 to 9 h and 0.81 (0.73-0.88) vs. 0.79 (0.72-0.87) within 9-12 h]. Targeted analysis of the CENTER-TBI core database, with focus on the patient category for which biomarker testing is recommended by the SNC guidelines, revealed that GFAP and UCH-L1 perform superior to S100B in predicting CT-positive intracranial lesions within 6 h of injury. GFAP continued to exhibit superior predictive ability to S100B during the time periods studied. S100B displayed relatively unaltered screening performance beyond the diagnostic timeline provided by SNC guidelines. These findings suggest the need for a reevaluation of the current SNC TBI guidelines.

The revised cardiac risk index is associated with morbidity and mortality independent of injury severity in elderly patients with rib fractures.

BACKGROUND: Risk factors for mortality and in-hospital morbidity among geriatric patients with traumatic rib fractures remain unclear. Such patients are often frail and demonstrate a high comorbidity burden. Moreover, outcomes anticipated by current rubrics may reflect the influence of multisystem injury or surgery, and thus not apply to isolated injuries in geriatric patients. We hypothesized that the Revised Cardiac Risk Index (RCRI) may assist in risk-stratifying geriatric patients following rib fracture. METHODS: All geriatric patients (age ≥65 years) with a conservatively managed rib fracture owing to an isolated thoracic injury (thorax AIS ≥1), in the 2013-2019 TQIP database were assessed including demographics and outcomes. The association between the RCRI and in-hospital morbidity as well as mortality was analyzed using Poisson regression models while adjusting for potential confounders. RESULTS: 96,750 geriatric patients sustained rib fractures. Compared to those with RCRI 0, patients with an RCRI score of 1 had a 16% increased risk of in-hospital mortality [adjusted incidence rate ratio (adj-IRR), 95% confidence interval (CI): 1.16 (1.02-1.32), p=0.020]. An RCRI score of 2 [adj-IRR (95% CI): 1.72 (1.44-2.06), p<0.001] or ≥3 [adj-IRR (95% CI): 3.07 (2.31-4.09), p<0.001] was associated with an even greater mortality risk. Those with an increased RCRI also exhibited a higher incidence of myocardial infarction, cardiac arrest, stroke, and acute respiratory distress syndrome. CONCLUSIONS: Geriatric patients with rib fractures and an RCRI ≥1 represent a vulnerable and high-risk group. This index may inform the decision to admit for inpatient care and can also guide patient and family counseling as well as computer-based decision-support.

Predicting 1-Year Mortality after Hip Fracture Surgery: An Evaluation of Multiple Machine Learning Approaches.

Postoperative death within 1 year following hip fracture surgery is reported to be up to 27%. In the current study, we benchmarked the predictive precision and accuracy of the algorithms support vector machine (SVM), naïve Bayes classifier (NB), and random forest classifier (RF) against logistic regression (LR) in predicting 1-year postoperative mortality in hip fracture patients as well as assessed the relative importance of the variables included in the LR model. All adult patients who underwent primary emergency hip fracture surgery in Sweden, between 1 January 2008 and 31 December 2017 were included in the study. Patients with pathological fractures and non-operatively managed hip fractures, as well as those who died within 30 days after surgery, were excluded from the analysis. A LR model with an elastic net regularization were fitted and compared to NB, SVM, and RF. The relative importance of the variables in the LR model was then evaluated using the permutation importance. The LR model including all the variables demonstrated an acceptable predictive ability on both the training and test datasets for predicting one-year postoperative mortality (Area under the curve (AUC) = 0.74 and 0.74 respectively). NB, SVM, and RF tended to over-predict the mortality, particularly NB and SVM algorithms. In contrast, LR only over-predicted mortality when the predicted probability of mortality was larger than 0.7. The LR algorithm outperformed the other three algorithms in predicting 1-year postoperative mortality in hip fracture patients. The most important predictors of 1-year mortality were the presence of a metastatic carcinoma, American Society of Anesthesiologists(ASA) classification, sex, Charlson Comorbidity Index (CCI) ≤ 4, age, dementia, congestive heart failure, hypertension, surgery using pins/screws, and chronic kidney disease.