利用先端統合分析方法整合証據以解決胰臟疾病領域之爭議問題

Abstract

背景與目的：關於胰臟炎與胰臟癌有許多廣受爭論的議題。以氣球擴張術或是括約肌切開術治療膽管結石其術後胰臟炎之相對風險仍存有爭議。糖尿病前期是否會增加胰臟癌的風險仍不清楚，胰臟癌腫瘤切除術後應如何治療亦無定論。為解決上述爭議，本論文藉由整合現有證據以比較膽管結石的各種治療及胰臟癌術後各種輔助性治療，及探討空腹血糖值與胰臟癌風險之劑量-反應關係。 方法：文獻回顧找出相關之研究並評估其品質。利用漏斗圖及Egger與Begg試驗評估潛在的小規模研究偏差。以貝氏網絡統合分析比較膽管結石的各種治療及胰臟癌術後各種輔助性治療，以劑量-反應統合分析探討空腹血糖值與胰臟癌風險之劑量-反應關係。 結果：相較於現行標準治療之括約肌切開術，以超過一分鐘之氣球擴張術治療膽管結石其胰臟炎風險相似但總併發症發生率較低，而短於一分鐘之氣球擴張術有較高之胰臟炎風險。空腹血糖值每上升10 mg/dL，胰臟癌之風險會隨之增加13%。輔助性化學治療可減少胰臟癌切除術後死亡率約三分之一，追加放射治療不能進一步延長存活反而會增加副作用。 結論: 氣球擴張術可作為膽管結石之首選治療。糖尿病前期是胰臟癌的危險因子。輔助性化學治療是胰臟癌切除後最佳的治療方式。

Background and aims: Many controversial issues exist regarding pancreatitis and pancreatic cancer. The relative risk of pancreatitis for treating bile duct stones using balloon dilation versus sphincterotomy is under debate. Second, whether prediabetes is a risk factor for pancreatic cancer remains unclear. Third, opinions on the treatment strategy after surgery for pancreatic cancer are widely divided. This dissertation aims to resolve these controversies by synthesizing current evidence to compare various treatments for bile duct stones and available adjuvant therapies for pancreatic cancer, and to evaluate the dose-response relationship between fasting blood glucose and pancreatic cancer risk. Methods: Eligible studies are identified and assessed for quality. Potential small study bias is evaluated by funnel plots and by tests of Egger and Begg. Bayesian network meta-analysis is used to compare various endoscopic treatments for bile duct stones and adjuvant therapies for pancreatic cancer. The dose-response relation between fasting blood glucose level and risk of pancreatic cancer is evaluated using random-effects linear and non-linear dose-response meta-analysis. Results: Compared with sphincterotomy, treating bile duct stones with balloon dilation for >1 minute has comparable pancreatitis risk and lower overall complications rate, whereas balloon dilation for ≦1 minute has higher pancreatitis risk. The risk of pancreatic cancer rises by 13% per 10 mg/dL increase in fasting blood glucose. Adjuvant chemotherapy reduces mortality by approximately a third compared with observation, whereas additional chemoradiation increases toxicity without further survival benefit. Conclusion: Balloon dilation may be the preferred first-line treatment for bile duct stones. Prediabetes is a risk for pancreatic cancer. Chemotherapy is the optimum adjuvant treatment after resection of pancreatic cancer.