摘要
抗生素治療在現代醫療中，尤其是重症照護，是不可或缺的一環。然而，我們一方面必須使用有效且積極的廣效性抗生素來治療嚴重感染的重症病患，另一方面，又必須從群體的觀點來省思因爲過度使用抗生素所帶來的抗藥性問題及其對後代的長遠影響。因此，在持續研發新抗生素的同時，使用一些策略來減緩抗生素的抗藥性有其迫切的需要，這其中能讓抗生素治療最佳化的方法之一就是降階治療。目前較確定可以運用抗生素降階治療的疾病包括呼吸器相關肺炎及敗血症。降階治療的作法分爲兩個階段。第一階段是針對特定病人給與立即且廣效性的抗生素以便改善臨床結果，第二階段則是爲了減少抗藥性與醫療花費進行降階治療。然而，要進行降階治療必須具備三項前提：好的臨床反應、非難治之病原與適當的評估指標。進行降階的時機大約在治療後第三天，也就是在臨床觀察至少兩天，同時有培養的結果可供參考後進行。至於抗生素的策略則是由廣效性改爲窄效性，由組合的治療改爲單一的治療，並且使用最短而有效的療程。目前抗生素降階治療的原則已經在某些加護病房中有效地實行，但是若要大規模地將降階治療運用於各種不同的臨床狀況，卻需要更多研究與證據來支持。總結而言，當臨床醫師希望病人得到最佳結果，卻又必須面臨盡量減少抗藥性產生的兩難局面時，抗生素的降階治療，應該是重症病人經驗抗生素使用一個可行的策略。
並列摘要
Antibiotic therapy is indispensable in modern medicine, especially in critical care. However, we not only must use effective and aggressive broad-spectrum antibiotics to treat severe infections of critically ill patients, but also should reflect the drug resistance problem due to overuse of antibiotics and its long term influences on our successors from global point of view. Therefore, while continuing development of new antibiotics, it is urgently necessary to use some strategies to minimize antibiotic resistance. De-escalation therapy is one of these strategies to optimize antibiotic treatment. At present time, the diseases which can be applied to by de-escalation therapy of antibiotics include ventilator-associated pneumonia and sepsis. The method of de-escalation therapy can be divided into two stages. The first stage is to give immediate and broad-spectrum antibiotics to certain groups of patients in order to improve clinical outcomes. The second stage is to de-escalate treatment for reduction of drug resistance and medical cost. However, there are three premises before implementation of de-escalation therapy: good clinical response, no ”difficult to treat” bugs and appropriate indicators for evaluation. The timing to de-escalate is about on the third day post treatment, that is, after clinical observation for at least two days when culture results available. As regards to antibiotic strategy, broad-spectrum antibiotics are changed to narrow-spectrum ones, combination therapy to single therapy, and the duration of antimicrobial treatment should be limited to the shortest effective course of therapy. Nowadays de-escalation therapy of antibiotics has been implemented effectively in some intensive care units. But if we want to apply de-escalation therapy to different clinical situations in a large scale, we need more researches and evidence to support it. In conclusion, when physicians hope their patients to get the best results, but also must face the dilemma of reducing emergence of drug-resistance, de-escalation therapy of antibiotics ought to be a feasible strategy of empirical antibiotic treatment in critically ill patients.