Bhawna Halwan

A decision support system to facilitate management of patients with acute gastrointestinal bleeding

OBJECTIVE To develop a model to predict the bleeding source and identify the cohort amongst patients with acute gastrointestinal bleeding (GIB) who require urgent intervention, including endoscopy. Patients with acute GIB, an unpredictable event, are most commonly evaluated and managed by non-gastroenterologists. Rapid and consistently reliable risk stratification of patients with acute GIB for urgent endoscopy may potentially improve outcomes amongst such patients by targeting scarce healthcare resources to those who need it the most. DESIGN AND METHODS Using ICD-9 codes for acute GIB, 189 patients with acute GIB and all available data variables required to develop and test models were identified from a hospital medical records database. Data on 122 patients was utilized for development of the model and on 67 patients utilized to perform comparative analysis of the models. Clinical data such as presenting signs and symptoms, demographic data, presence of co-morbidities, laboratory data and corresponding endoscopic diagnosis and outcomes were collected. Clinical data and endoscopic diagnosis collected for each patient was utilized to retrospectively ascertain optimal management for each patient. Clinical presentations and corresponding treatment was utilized as training examples. Eight mathematical models including artificial neural network (ANN), support vector machine (SVM), k-nearest neighbor, linear discriminant analysis (LDA), shrunken centroid (SC), random forest (RF), logistic regression, and boosting were trained and tested. The performance of these models was compared using standard statistical analysis and ROC curves. RESULTS Overall the random forest model best predicted the source, need for resuscitation, and disposition with accuracies of approximately 80% or higher (accuracy for endoscopy was greater than 75%). The area under ROC curve for RF was greater than 0.85, indicating excellent performance by the random forest model. CONCLUSION While most mathematical models are effective as a decision support system for evaluation and management of patients with acute GIB, in our testing, the RF model consistently demonstrated the best performance. Amongst patients presenting with acute GIB, mathematical models may facilitate the identification of the source of GIB, need for intervention and allow optimization of care and healthcare resource allocation; these however require further validation.

Endoscopic treatment of Roux-en-Y gastric bypass-related gastrocutaneous fistulas using a novel biomaterial

BackgroundRoux-en-Y gastric bypass (RYGB) is amongst the commonest surgical intervention for weight loss in obese patients. Gastrocutaneous fistula, which usually occurs along the vertical staple line of the pouch, is amongst its most alarming complications. Medical management comprised of wound drainage, nutritional support, acid suppression, and antibiotics may be ineffective in as many as a third of patients with this complication. We present outcomes after endoscopic application of SurgiSIS®, which is a novel biomaterial for the treatment of this complication.DesignA case series of 25 patients.MethodsTwenty-five patients who had failed conservative medical management of gastrocutaneous fistula after RYGB underwent endoscopic application of SurgiSIS®—an acellular fibrogenic matrix biomaterial to help fistula healing.Main outcome measuresFistula closure as assessed by upper gastrointestinal imaging and endoscopic examination.Results In patients who had failed medical management lasting 4–25 (median, 7) weeks, closure of the fistulous tract was successful after one application in six patients (30%), two applications in 11 patients (55%), and three applications in three patients (15%). There were no procedure-related complications.ConclusionsEndoscopic application of SurgiSIS®—an acellular fibrogenic matrix—is safe and effective for the treatment of gastrocutaneous fistula after RYGB.

Submucosal Injection of 0.4% Hydroxypropyl Methylcellulose Facilitates Endoscopic Mucosal Resection of Early Gastrointestinal Tumors

Background and Aims Submucosal injection of a viscoelastic solution prolongs submucosal lift, thus, facilitating endoscopic mucosal resection. Our objective was to assess the safety and clinical effectiveness of 0.4% hydroxypropyl methylcellulose (HPMC) as a submucosal injectant for endoscopic mucosal resection. Patients and Methods A prospective, open-label, multicenter, phase 2 study was conducted at 2 academic institutions in Brazil. Eligible participants included patients with early gastrointestinal tumors larger than 10 mm. Outcomes evaluated included complete resection rates, volume of HPMC injected, duration of the submucosal cushion as assessed visually, histology of the resected leisons, and complication rates. Results Over a 12-month period, 36 eligible patients with superficial neoplastic lesions (stomach 14, colon 11, rectum 5, esophagus 3, duodenum 3) were prospectively enrolled in the study. The mean size of the resected specimen was 20.4 mm (10 to 60 mm). The mean volume of 0.4% HPMC injected was 10.7 mL (range 4 to 35 mL). The mean duration of the submucosal fluid cushion was 27 minutes (range 9 to 70 min). Complete resection was successfully completed in 89%. Five patients (14%) developed immediate bleeding requiring endoclip and APC application. Esophageal perforation occurred in 1 patient requiring surgical intervention. There were no local or systemic adverse events related to HPMC use over the follow-up period (mean 2.2 mo). Conclusion HPMC solution (0.4%) provides an effective submucosal fluid cushion and is safe for endoscopic resection of early gastrointestinal neoplastic lesions.