Evaluation of a model for glycemic prediction in critically ill surgical patients.

We evaluated a neural network model for prediction of glucose in critically ill trauma and post-operative cardiothoracic surgical patients. A prospective, feasibility trial evaluating a continuous glucose-monitoring device was performed. After institutional review board approval, clinical data from all consenting surgical intensive care unit patients were converted to an electronic format using novel software. This data was utilized to develop and train a neural network model for real-time prediction of serum glucose concentration implementing a prediction horizon of 75 minutes. Glycemic data from 19 patients were used to "train" the neural network model. Subsequent real-time simulated testing was performed in 5 patients to whom the neural network model was naive. Performance of the model was evaluated by calculating the mean absolute difference percent (MAD%), Clarke Error Grid Analysis, and calculation of the percent of hypoglycemic (≤70 mg/dL), normoglycemic (>70 and <150 mg/dL), and hyperglycemic (≥150 mg/dL) values accurately predicted by the model; 9,405 data points were analyzed. The models successfully predicted trends in glucose in the 5 test patients. Clark Error Grid Analysis indicated that 100.0% of predictions were clinically acceptable with 87.3% and 12.7% of predicted values falling within regions A and B of the error grid respectively. Overall model error (MAD%) was 9.0% with respect to actual continuous glucose modeling data. Our model successfully predicted 96.7% and 53.6% of the normo- and hyperglycemic values respectively. No hypoglycemic events occurred in these patients. Use of neural network models for real-time prediction of glucose in the surgical intensive care unit setting offers healthcare providers potentially useful information which could facilitate optimization of glycemic control, patient safety, and improved care. Similar models can be implemented across a wider scale of biomedical variables to offer real-time optimization, training, and adaptation that increase predictive accuracy and performance of therapies.

Conservative management of esophageal perforation after a fall.

INTRODUCTION: Esophageal perforation in the setting of blunt trauma is rare, and diagnosis can be difficult due to atypical signs and symptoms accompanied by distracting injury. PRESENTATION OF CASE: We present a case of esophageal perforation resulting from a fall from height. Unexplained air in the soft tissues planes posterior to the esophagus as well as subcutaneous emphysema in the absence of a pneumothorax on CT aroused clinical suspicions of an injury to the aerodigestive tract. The patient suffered multiple injuries including bilateral first rib fractures, C6 lamina fractures, C4-C6 spinous process fractures, a C7 right transverse process fracture with associated ligamentous injury and cord contusion, multiple comminuted nasal bone fractures, and a right verterbral artery dissection. Esophageal injury was localized using a gastrograffin esophagram to the cervical esophagus and was most likely secondary to cervical spine fractures. Because there were no clinical signs of sepsis and the esophagram demonstrated a contained rupture, the patient was thought to be a good candidate for a trial of conservative management consisting of broad spectrum intravenous antibiotics, oral care with chlorhexadine gluconate, NPO, and total parenteral nutrition. No cervical spine fixation or procedure was performed during this trial of conservative management. The patient was received another gastrograffin esophagram on hospital day 14 and demonstrated no evidence of contrast extravasation. DISCUSSION: Early diagnosis and control of the infectious source are the cornerstones to successful management of esophageal perforation from all etiologies. Traditionally, esophageal perforation relied on a high index of clinical suspicion for early diagnosis, but the use of CT scan for has proved to be highly effective in diagnosing esophageal perforation especially in patients with atypical presentations. While aggressive surgical infection control is paramount in the majority of esophageal perforations, a select subset of patients can be successfully managed non-operatively. CONCLUSION: In the setting of blunt trauma, esophageal perforation is rare and is associated with a high morbidity. In select patients who do not show any clinical signs of sepsis, contained perforations can heal with non-operative management consisting of broad spectrum antibiotics, strict oral hygiene, NPO, and total parenteral nutrition.

Neural network-based real-time prediction of glucose in patients with insulin-dependent diabetes.

BACKGROUND: Continuous glucose monitoring (CGM) technologies report measurements of interstitial glucose concentration every 5 min. CGM technologies have the potential to be utilized for prediction of prospective glucose concentrations with subsequent optimization of glycemic control. This article outlines a feed-forward neural network model (NNM) utilized for real-time prediction of glucose. METHODS: A feed-forward NNM was designed for real-time prediction of glucose in patients with diabetes implementing a prediction horizon of 75 min. Inputs to the NNM included CGM values, insulin dosages, metered glucose values, nutritional intake, lifestyle, and emotional factors. Performance of the NNM was assessed in 10 patients not included in the model training set. RESULTS: The NNM had a root mean squared error of 43.9 mg/dL and a mean absolute difference percentage of 22.1. The NNM routinely overestimates hypoglycemic extremes, which can be attributed to the limited number of hypoglycemic reactions in the model training set. The model predicts 88.6% of normal glucose concentrations (> 70 and < 180 mg/dL), 72.6% of hyperglycemia (≥ 180 mg/dL), and 2.1% of hypoglycemia (≤ 70 mg/dL). Clarke Error Grid Analysis of model predictions indicated that 92.3% of predictions could be regarded as clinically acceptable and not leading to adverse therapeutic direction. Of these predicted values, 62.3% and 30.0% were located within Zones A and B, respectively, of the error grid. CONCLUSIONS: Real-time prediction of glucose via the proposed NNM may provide a means of intelligent therapeutic guidance and direction.