Yeongkeun Kwon1,2, Jin-Won Kwon3, Jane Ha2,4, Dohyang Kim5, Jaehyeong Cho6, Soo Min Jeon3, Shin-Hoo Park1,2, Jinseub Hwang5, Nam Hoon Kim7, Sungsoo Park8,9. 1. Division of Foregut Surgery, Korea University College of Medicine, Seoul, Korea. 2. Centre for Obesity and Metabolic Diseases, Korea University Anam Hospital, Seoul, Korea. 3. BK21 FOUR Community-Based Intelligent Novel Drug Discovery Education Unit, College of Pharmacy and Research Institute of Pharmaceutical Sciences, Kyungpook National University, Daegu, Korea. 4. Department of Medicine, Korea University College of Medicine, Seoul, Korea. 5. Department of Statistics, Daegu University, Gyeongbuk, Korea. 6. Department of Biomedical Sciences, Ajou University Graduate School of Medicine, Suwon, Korea. 7. Division of Endocrinology and Metabolism, Department of Internal Medicine, Korea University College of Medicine, Seoul, Korea. 8. Division of Foregut Surgery, Korea University College of Medicine, Seoul, Korea. kugspss@korea.ac.kr. 9. Centre for Obesity and Metabolic Diseases, Korea University Anam Hospital, Seoul, Korea. kugspss@korea.ac.kr.
Abstract
BACKGROUND: Although type 2 diabetes (T2D) remission after gastric cancer surgery has been reported, little is known about the predictors of postoperative T2D remission. METHODS: This study used data from a nationwide cohort provided by the National Health Insurance Service in Korea. We developed a diabetes prediction (DP) score, which predicted postoperative T2D remissions using a logistic regression model based on preoperative variables. We applied machine-learning algorithms [random forest, XGboost, and least absolute shrinkage and selection operator (LASSO) regression] and compared their predictive performances with those of the DP score. RESULTS: The DP score comprised five parameters: baseline body mass index (< 25 or ≥ 25 kg/m2), surgical procedures (subtotal or total gastrectomy), age (< 65 or ≥ 65 years), fasting plasma glucose levels (≤ 130 or > 130 mg/dL), and antidiabetic medications (combination therapy including sulfonylureas, combination therapy not including sulfonylureas, single sulfonylurea, or single non-sulfonylurea]). The DP score showed a clinically useful predictive performance for T2D remission at 3 years after surgery [training cohort: area under the receiver operating characteristics (AUROC) 0.73, 95% confidence interval (CI), 0.71-0.75; validation cohort: AUROC 0.72, 95% CI 0.69-0.75], which was comparable to that of the machine-learning models (random forest: AUROC 0.71, 95% CI 0.68-0.74; XGboost: AUROC 0.70, 95% CI 0.67-0.73; LASSO regression: AUROC 0.75, 95% CI 0.73-0.78 in the validation cohort). It also predicted the T2D remission at 6 and 9 years after surgery. CONCLUSIONS: The DP score is a useful scoring system for predicting T2D remission after gastric cancer surgery.
BACKGROUND: Although type 2 diabetes (T2D) remission after gastric cancer surgery has been reported, little is known about the predictors of postoperative T2D remission. METHODS: This study used data from a nationwide cohort provided by the National Health Insurance Service in Korea. We developed a diabetes prediction (DP) score, which predicted postoperative T2D remissions using a logistic regression model based on preoperative variables. We applied machine-learning algorithms [random forest, XGboost, and least absolute shrinkage and selection operator (LASSO) regression] and compared their predictive performances with those of the DP score. RESULTS: The DP score comprised five parameters: baseline body mass index (< 25 or ≥ 25 kg/m2), surgical procedures (subtotal or total gastrectomy), age (< 65 or ≥ 65 years), fasting plasma glucose levels (≤ 130 or > 130 mg/dL), and antidiabetic medications (combination therapy including sulfonylureas, combination therapy not including sulfonylureas, single sulfonylurea, or single non-sulfonylurea]). The DP score showed a clinically useful predictive performance for T2D remission at 3 years after surgery [training cohort: area under the receiver operating characteristics (AUROC) 0.73, 95% confidence interval (CI), 0.71-0.75; validation cohort: AUROC 0.72, 95% CI 0.69-0.75], which was comparable to that of the machine-learning models (random forest: AUROC 0.71, 95% CI 0.68-0.74; XGboost: AUROC 0.70, 95% CI 0.67-0.73; LASSO regression: AUROC 0.75, 95% CI 0.73-0.78 in the validation cohort). It also predicted the T2D remission at 6 and 9 years after surgery. CONCLUSIONS: The DP score is a useful scoring system for predicting T2D remission after gastric cancer surgery.
Authors: Enrique Lanzarini; Attila Csendes; Hans Lembach; Juan Molina; Luis Gutiérrez; Johanna Silva Journal: World J Surg Date: 2010-09 Impact factor: 3.352
Authors: Emmanuel E Zervos; Steven C Agle; Alex J Warren; Christina G Lang; Timothy L Fitzgerald; Moahad Dar; Michael F Rotondo; Walter J Pories Journal: J Am Coll Surg Date: 2010-05 Impact factor: 6.113
Authors: Bethany B Barone; Hsin-Chieh Yeh; Claire F Snyder; Kimberly S Peairs; Kelly B Stein; Rachel L Derr; Antonio C Wolff; Frederick L Brancati Journal: JAMA Date: 2008-12-17 Impact factor: 56.272
Authors: L B Koppert; M L G Janssen-Heijnen; M W J Louwman; V E P P Lemmens; B P L Wijnhoven; H W Tilanus; J W W Coebergh Journal: Eur J Gastroenterol Hepatol Date: 2004-07 Impact factor: 2.566
Authors: Yvette R B M van Gestel; Valery E P P Lemmens; Ignace H J T de Hingh; Jessie Steevens; Harm J T Rutten; Grard A P Nieuwenhuijzen; Ronald M van Dam; Peter D Siersema Journal: Ann Surg Oncol Date: 2012-09-18 Impact factor: 5.344
Authors: Craig J Currie; Chris D Poole; Sara Jenkins-Jones; Edwin A M Gale; Jeffrey A Johnson; Christopher Ll Morgan Journal: Diabetes Care Date: 2012-01-20 Impact factor: 19.112