Predictors of recovery in moderate to severe traumatic brain injury

Machine learning algorithms for predicting outcomes of traumatic brain injury: A systematic review and meta-analysis

BACKGROUND

Traumatic brain injury (TBI) is a leading cause of death and disability worldwide. The use of machine learning (ML) has emerged as a key advancement in TBI management. This study aimed to identify ML models with demonstrated effectiveness in predicting TBI outcomes.

METHODS

We conducted a systematic review in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis statement. In total, 15 articles were identified using the search strategy. Patient demographics, clinical status, ML outcome variables, and predictive characteristics were extracted. A small meta-analysis of mortality prediction was performed, and a meta-analysis of diagnostic accuracy was conducted for ML algorithms used across multiple studies.

RESULTS

ML algorithms including support vector machine (SVM), artificial neural networks (ANN), random forest, and Naïve Bayes were compared to logistic regression (LR). Thirteen studies found significant improvement in prognostic capability using ML versus LR. The accuracy of the above algorithms was consistently over 80% when predicting mortality and unfavorable outcome measured by Glasgow Outcome Scale. Receiver operating characteristic curves analyzing the sensitivity of ANN, SVM, decision tree, and LR demonstrated consistent findings across studies. Lower admission Glasgow Coma Scale (GCS), older age, elevated serum acid, and abnormal glucose were associated with increased adverse outcomes and had the most significant impact on ML algorithms.

CONCLUSION

ML algorithms were stronger than traditional regression models in predicting adverse outcomes. Admission GCS, age, and serum metabolites all have strong predictive power when used with ML and should be considered important components of TBI risk stratification.

Effects of Respiratory Muscle Training on Baroreflex Sensitivity, Respiratory Function, and Serum Oxidative Stress in Acute Cervical Spinal Cord Injury

Background: respiratory complications are a leading cause of morbidity and mortality in individuals with spinal cord injury (SCI). We examined the effects of respiratory muscle training (RMT) in patients with acute cervical SCI. Methods: this prospective trial enrolled 44 adults with acute cervical SCI, of which twenty received RMT and twenty-four did not receive RMT. Respiratory function, cardiovascular autonomic function, and reactive oxidative species (ROS) were compared. The experimental group received 40-min high-intensity home-based RMT 7 days per week for 10 weeks. The control group received a sham intervention for a similar period. The primary outcomes were the effects of RMT on pulmonary and cardiovascular autonomic function, and ROS production in individuals with acute cervical SCI. Results: significant differences between the two groups in cardiovascular autonomic function and the heart rate response to deep breathing (p = 0.017) were found at the 6-month follow-up. After RMT, the maximal inspiratory pressure (p = 0.042) and thiobarbituric acid-reactive substances (TBARS) (p = 0.006) improved significantly, while there was no significant difference in the maximal expiratory pressure. Significant differences between the two groups in tidal volume (p = 0.005) and the rapid shallow breathing index (p = 0.031) were found at 6 months. Notably, the SF-36 (both the physical (PCS) and mental (MCS) component summaries) in the RMT group had decreased significantly at the 6-month follow-up, whereas the clinical scores did not differ significantly (p = 0.333) after RMT therapy. Conclusions: High-intensity home-based RMT can improve pulmonary function and endurance and reduce breathing difficulties in patients with respiratory muscle weakness after injury. It is recommended for rehabilitation after spinal cord injury.

Evaluation of the effects of pycnogenol (French maritime pine bark extract) supplementation on inflammatory biomarkers and nutritional and clinical status in traumatic brain injury patients in an intensive care unit: A randomized clinical trial protocol

Background

Traumatic brain injury (TBI) is one of the major health and socioeconomic problems in the world. Immune-enhancing enteral formula has been proven to significantly reduce infection rate in TBI patients. One of the ingredients that can be used in immunonutrition formulas to reduce inflammation and oxidative stress is pycnogenol.

Objective

The objective of this work is to survey the effect of pycnogenol on the clinical, nutritional, and inflammatory status of TBI patients.

Methods

This is a double-blind, randomized controlled trial. Block randomization will be used. An intervention group will receive pycnogenol supplementation of 150 mg for 10 days and a control group will receive a placebo for the same duration. Inflammatory status (IL-6, IL- 1β, C-reactive protein) and oxidative stress status (malondialdehyde, total antioxidant capacity), at the baseline, at the 5th day, and at the end of the study (10th day) will be measured. Clinical and nutritional status will be assessed three times during the intervention. The Sequential Organ Failure Assessment (SOFA) questionnaire for assessment of organ failure will be filled out every other day. The mortality rate will be calculated within 28 days of the start of the intervention. Weight, body mass index, and body composition will be measured. All analyses will be conducted by an initially assigned study arm in an intention-to-treat analysis.

Discussion

We expect that supplementation of 150 mg pycnogenol for 10 days will improve clinical and nutritional status and reduce the inflammation and oxidative stress of the TBI patients.

Trial registration

This trial is registered at clinicaltrials.gov (ref: NCT03777683) at 12/13/2018.