背景與目的：鑑於本院在2001年1月曾經對冠狀動脈繞道手術的預防性抗生素作一介入性的分析研究，發現不論是使用amoxicillin/clavulanic acid或cefazolin，二者術後的感染率差異不大，其平均療程為17±3天。因此本研究再次作一後續的追蹤，以了解近年來預防性抗生素的使用及術後感染率的情形，並針對不合理之處作一檢討以改善用藥的品質和所節省的醫療費用。
方法：本研究採回溯性的收集本院台北院區自2002年~2003年兩年間所有曾接受冠狀動脈繞道手術的病患，並逐一記錄病患的層級變數。
結果：總共有261例病患且預防性抗生素皆使用cefazolin，平均療程已比2001年的平均療程降低為6.9天。研究結果也顯示延長使用預防性抗生素並無法降低感染率（p>0.05），而手術的複雜度與術後感染發生率有顯著的影響（p = 0.01）。
結論：延長使用預防性抗生素並不能降低感染率，卻會增加醫療成本的支出。未來應可加強預防性抗生素的處方醫令管控功能，及依照病患的臨床狀況評估之後方可續用之。
 

Background and purpose: We tested a intervention program and analysis in-hospital use of antimicrobial prophylaxis for coronary artery bypass surgery during 2001. The surgical site infection rate was revealed no difference between amoxicillin / clavulanic acid and cefazolin. However, the mean course of prophylaxis were 17±3 days. The aim of this follow-up study was to realize the intervention performance. 
Methods: Patients who received coronary artery bypass surgery during the years 2002 to 2003 were retrospected selected in the study. 
Resluts: A total of 261 consecutive patients was included and cefazolin was prescribed as antibiotic prophylaxis. Mean prophylaxis course was reduced to 6.9days. Study result revealed prolong-used prophylactic antibiotics cannot reduce infection rate (p>0.05)，and the complication of surgery was the most significant predictor of surgical site infection (p=0.01)。
Conclusion: These findings indicate prolong-used prophylactic antibiotic cannot reduce infection rate but may increase the related cost.