根據感染源再細分抗生素敏感試驗結果對於選用經驗性抗生素的影響

Impact of Cumulative Antibiograms Sub-categorized by Origins of Infection Acquisition on the Selection of Empirical Antimicrobial Therapy

背景/目的：對於治療嚴重感染症，抗生素敏感試驗的結果對於使用經驗性抗生素相當重要，尤其是面對嚴重感染的病人時。而根據不同病人族群做出的抗生素敏感試驗結果可能較能準確的預估致病菌對不同抗生素的敏感度，因而較能幫助適當的經驗性抗生素的選擇與使用。方法：在2012年，針對Staphylococcus aureus、Escherichia coli、Klebsiella pneumoniae、Pseudomonas aeruginosa 及Acinetobacter baumannii的抗生素敏感性結果，根據不同病人族群更加細分及比較。結果：此研究包括1810株非重覆的菌株，P. aeruginosa是對抗生素敏感性最高的菌種，因為除了levofloxacin之外，對其他測試抗生素（ceftazidime, imipenem, meropenem, gentamicin, amikacin）的敏感率都 >80%。而A. baumannii是對抗生素最不敏感的菌種，因為除了tigecycline之外，對其他測試抗生素（ceftazidime, ampicillin-sulbactam, imipenem, gentamicin, amikacin, levofloxacin）的敏感率都 <80%。在不同的病人族群中，一些具有高敏感率的抗生素，如S. aureus對tigecycline、vancomycin、linezolid以及E. coli和K. pneumoniae對carbapenems、amikacin、tigecycline的敏感率差異很小。除了社區性感染（community-acquired infections, CAI）的E. coli及K. pneumoniae對gentamicin及非第一代cephalosporins的敏感率較高（> 80%）之外，其他大部分的敏感率都< 80%。一般而言，社區性感染的致病菌對抗生素的敏感率最高，而安養中心相關感染（nursing home-associated infections, NHAI）致病菌對抗生素的的敏感率最低，甚至低於醫療照護相關感染（healthcare-acquired infections, HCAI）及院內感染（hospital-acquired infections, HAI）的致病菌。HCAI致病菌對抗生素的敏感率稍高於HAI的敏感率。結論：不同病人族群的抗生素敏感率結果並不相同，尤其是針對較具抗藥性的細菌或效果較差的抗生素。如果沒有根據不同病人族群再細分抗生素敏感性結果，可能會導致不適當的經驗性抗生素使用，尤其是對CAI病人可能會過度使用抗生素，而對於NHAI及HCAI的病人，可能會出現使用療效不足抗生素的情形。

Background/purpose: Cumulative antimicrobial susceptibility reports (antibiograms) are important when prescribing empirical antibiotics for severe infections. Specific cumulative antibiograms for isolates from different patient populations may better reflect the actual antimicrobial susceptibility and facilitate the selection of appropriate empirical antibiotics. Methods: Antibiograms of Staphylococcus aureus, Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and Acinetobacter baumannii, reported in 2012, were categorized by the origins of acquisition and compared. Results: A total of 1810 non-duplicate isolates were analyzed. P. aeruginosa was the most susceptible species, with the susceptibility being > 80% for all antibiotics tested (ceftazidime, imipenem, meropenem, gentamicin, amikacin) except levofloxacin. A. baumannii was the most resistant species, because the susceptibility rates were < 80% for all antibiotics tested (ceftazidime, ampicillin-sulbactam, imipenem, gentamicin, amikacin, levofloxacin) except tigecycline. Some antimicrobial agents remained highly effective, such as tigecycline, vancomycin, and linezolid for S. aureus, and carbapenems, amikacin, and tigecycline for E. coli and K. pneumoniae, regardless the origins of infection acquisition. Susceptibilities for the other antibiotics were mostly < 80%, except that isolates of E. coli and K. pneumoniae from community-acquired infections (CAI) both demonstrated high susceptibilities (> 80%) for gentamicin and non-first-generation cephalosporins. In general, CAI isolates were significantly more susceptible, whereas those from nursing home-acquired infections (NHAI) were significantly more resistant, than isolates from healthcare-associated infections (HCAI) and hospital-acquired infections (HAI). Susceptibilities of HCAI isolates were relatively higher than those of HAI isolates. Conclusion: Cumulative antibiograms could vary among the categories, especially for resistant bacteria or antibiotics with less effectiveness. Without the sub-division, inappropriate use of empirical antibiotics, especially overuse in CAI and underuse in NHAI and HCAI, may occur.