9 February 2022 in Head & Neck
1 October 2021 in International Journal of Pediatric Otorhinolaryngology
24 February 2022 in International Journal of Pediatric Otorhinolaryngology
15 February 2024 in Head & Neck
Background: To develop machine learning (ML) models predicting unplanned readmission and reoperation among patients undergoing free flap reconstruction for head and neck (HN) surgery.Methods: Data were extracted from the 2012–2019 NSQIP database. eXtreme Gradient Boosting (XGBoost) was used to develop ML models predicting 30‐day readmission and reoperation based on demographic and perioperative factors. Models were validated using 2019 data and evaluated.Results: Four‐hundred and sixty‐six (10.7%) of 4333 included patients were readmitted within 30 days of initial surgery. The ML model demonstrated 82% accuracy, 63% sensitivity, 85% specificity, and AUC of 0.78. Nine‐hundred and four (18.3%) of 4931 patients underwent reoperation within 30 days of index surgery. The ML model demonstrated 62% accuracy, 51% sensitivity, 64% specificity, and AUC of 0.58.Conclusion: XGBoost was used to predict 30‐day readmission and reoperation for HN free flap patients. Findings may be used to assist clinicians and patients in shared decision‐making and improve data collection in future database iterations.
1 April 2022 in Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology
28 July 2020 in Vascular and Endovascular Surgery
Vascular closure device (VCD) use following antegrade femoral access may present unique challenges relative to retrograde access. We retrospectively compared safety and efficacy of these devices between antegrade and retrograde patient cohorts undergoing percutaneous intervention. Over a 5-year period, a consecutive series of 107 limbs in 84 patients underwent VCD arteriotomy closure following percutaneous revascularization using an antegrade approach (VCD-A). Device deployment success rates, time to ambulation, and complication rates were compared to a contemporaneous control group of 401 limbs in 305 patients who underwent closure following retrograde access (VCD-R) during revascularization or embolization procedures. Closure was attempted in VCD-A using 53 StarClose, 35 Perclose, and 19 Angio-Seal devices. Hemostasis (without supplemental manual compression) was achieved in 86/107 (80.4%) limbs. Closure was attempted in VCD-R using 215 StarClose, 119 Perclose, and 67 Angio-Seal devices with hemostasis in 357/401 (89.0%) limbs. Device deployment failure occurred in 7/107 (6.5%) of VCD-A and 20/401 (5.0%) of VCD-R (P = .52), independent of specific device type. Femoral pseudoaneurysm developed in 1/107 and 1/401 of VCD-A and VCD-R (P = .31), and minor hematoma developed in 3/107 and 8/401 of the VCD-A and VCD-R (P = .61). Mean time to ambulation was 204.1 minutes in VCD-A and 204.8 minutes in VCD-R (P = .97). Antegrade femoral closure was associated with high rates of technical success and low complications, similar to retrograde closure. Time to ambulation was the same in both groups despite higher heparin doses in the antegrade patients.