- 學位論文
早期梅毒治療效用之網路統合分析與回溯性世代研究
Effectiveness of early syphilis treatment: A network meta-analysis and retrospective cohort study.
摘要
研究背景 長久以來，梅毒首選藥物為安全又有效的長效benzathine penicillin G (BPG)，然而，許多對此成分過敏者不適用，且此針劑在台灣已漸漸退出市場，在多家醫院缺藥且二線用藥的選擇不甚明確情況下，臨床上還是有其治療需求在，因此找出有力的替代藥物有其重要性。 研究目的 本研究先以文獻回顧比較不同抗生素在早期梅毒的療效，之後將其結果對照病歷回溯世代研究，同時進一步探究哪些病患特性為可能影響最終治癒成果之因子，最終目的為找到臨床上可替代BPG作為早期梅毒治療藥物的證據。 研究方法 首先以early syphilis與penicillin G, amoxicillin, azithromycin, ceftriaxone, doxycycline或tetracycline為關鍵字，搜尋PubMed, Cochrane Central Register of Controlled Trials, Embase, Clinicalkey及ClinicalTrials.gov等資料庫，篩選出符合納入條件之隨機對照試驗(RCT)或觀察性研究(observational study)文章，搜尋時間至2019年5月，沒有使用語言或年份限制，經過量表品質評估後主要以隨機效果模型(random effect model)執行網絡統合分析。其次，由某準醫學中心醫療資訊系統篩選自2010年至2017年止，納入具早期梅毒診斷之年滿20歲，並同時處方前述文獻內提及的抗生素來治療者，經初步排除干擾因子後，主要以RPR titer數值為判定標準，若該數值於給藥後開始下降達4倍以上即歸類為治療成功，追蹤時間最長到18個月，再做進一步分析治癒率及其影響因子與人類免疫不全病毒(HIV)感染次群組分析。 研究結果 經12篇文獻所涵蓋抗生素的網絡統合分析結果顯示penicillin G, azithromycin, ceftriaxone及doxycycline/tetracycline之間，於治療早期梅毒之血清學反應無統計學上顯著差異。在統計分析病歷資料後，發現55位azithromycin和6位doxycycline的血清治癒率分別為 90.9% 和100%，優於31位使用標準療法BPG的67.7% (p值=0.011)，三者在治癒天數上也無顯著差異(p值=0.135)。此外，對患者血清學治癒有幫助的可能因子有較低的年齡層、較高的初始RPR數值與使用azithromycin等；進一步多變量分析後則只有使用BPG相較於azithromycin的治癒機會偏低 (adjusted OR 0.16, 95% CI 0.03-0.71,p值=0.016)。而對於已感染HIV的早期梅毒患者來看，在多變量分析當中沒有發現統計上有意義影響治癒率的因子，符合總體研究結果。 結論 根據本研究文獻回顧結果，BPG與納入的另外三種抗生素於早期梅毒的治療效果沒有明顯差異，但本地研究卻顯示出azithromycin相較於BPG有更好的優勢，因此建議可以在本地臨床上依個別病患特性使用，做為BPG缺藥時期或過敏者之早期梅毒替代療法，唯仍需考慮個別病患之用藥順從性與不同地區之抗生素抗藥性，做出最佳的治療選擇。
關鍵字
早期梅毒 ； 網絡統合分析 ； azithromycin ； benzathine penicillin G ； ceftriaxone ； doxycycline ； tetracycline ； 人類免疫不全病毒並列摘要
Background Benzathine penicillin G (BPG) is known as the first-line choice of early syphilis treatment for decades. However, due to the medical environment changed, BPG has gradually been withdrawn from the market in Taiwan. As a result, we are interested in comparing the effectiveness of BPG with other antibiotic options. Aim To compare the effectiveness of early syphilis treatment by network meta-analysis (NMA). On the other hand, finding factors influence the cure rate through clinical observation. The ultimate goal is to find alternative antibiotics of early syphilis. Method By searching PubMed, Cochrane Central Register of Controlled Trials, Embase, Clinicalkey and ClinicalTrials.gov for randomized controlled trials (RCTs) or observational studies of early syphilis therapies with penicillin G, amoxicillin, azithromycin, ceftriaxone, doxycycline or tetracycline until May 2019 without any limitations of languages or years. We used NMA to compare the outcome of serological response 6 to 12 month after treatment. Furthermore, in the retrospective cohort study, 20-year-old or older patients who treated with these antibiotics from 2010 to 2017 in a would-be medical center were observed. The outcome was observed the serum RPR decline more than four-fold until 18 months. Since many syphilis patients are human immunodeficiency virus (HIV) infected at the same time, we used sub-group analysis and multiple logistic regression to find factors which might influence the cure rate. Result After reviewed 4 RCTs and 8 observational studies and conducting the NMA, we found that there was no statistical difference between antibiotics effectiveness in early syphilis. The retrospective cohort study showed the serological cure rate of 55 azithromycin user was 90.9%, which was statistically higher than 67.7% in 31 patients who used BPG. We also found possible factors affected serological cure rate were age, initial RPR titer, and antibiotics for syphilis. After further multivariable analysis, the only factor was BPG had less chance than azithromycin to cure early syphilis (adjusted OR 0.16, 95% CI 0.03-0.71, p-value=0.016). There was no factor influenced serological cure rate in HIV sub-group multivariable analysis. Conclusion Our research demonstrated azithromycin and doxycycline might substitute for early syphilis treatment while BPG allergy or BPG out of stock. Physicians could offer these options of early syphilis to patients for shared medical decision making.