Mittelkötter U
Klinik für Allgemeine Chirurgei der Universität Münster, Germany.
J Chemother. 2001 Nov;13 Spec No 1(1):27-34. doi: 10.1179/joc.2001.13.Supplement-2.27.
The objective of the present multicenter observational study was to assess postoperative infections as a function of risk factors and antibiotic prophylaxis under everyday clinical conditions. 2513 patients from 114 centers in Germany who received infection prophylaxis prior to elective colonic resection were included in the study between 1st September 1996 and 30th September 1997. In the descriptive analysis of the study it was noted that 36.1% of the patients had received no prophylaxis with metronidazole despite the fact that the study protocol recommended the use of this drug in preoperative antibiotic combinations. The present analysis therefore considers the influence of metronidazole on the postoperative infection rate. To exclude any bias due to intergroup differences in risk profile, the groups with and without metronidazole were subjected to a matched-pair analysis. Matching parameters were: duration of operation, blood loss, age, diabetes mellitus, hepatic, renal, or chronic airways disease, immunosuppressive therapy, and rectal resection. This led to the formation of 800 pairs that were matched with respect to these parameters. The 800 pairs were then stratified into the following treatment groups: Group 1: long-acting cephalosporin (ceftriaxone) with or without metronidazole (n = 2 x 491); Group 2: short-acting cephalosporins with or without metronidazole (n = 2 x 133); Group 3: broad-spectrum penicillins with or without metronidazole (n = 2 x 176). In all three treatment groups combination therapy with metronidazole was found to be significantly superior. Postoperative infection rates with and without metronidazole were 9.4% and 18.7% respectively in Group 1, 12.0% and 25.6% respectively in Group 2, and 19.9% and 29.0% respectively in Group 3. The lowest infection rate was thus achieved by means of preoperative infection prophylaxis with ceftriaxone plus metronidazole. Thus, preoperative administration of metronidazole in addition to a long-acting beta-lactam antibiotic is strongly advised in elective colon surgery, as absence of antibiotic cover against anaerobic colonic flora leads to a significantly higher postoperative infection rate.
本多中心观察性研究的目的是在日常临床条件下,评估术后感染与风险因素及抗生素预防之间的关系。1996年9月1日至1997年9月30日期间,德国114个中心的2513例在择期结肠切除术前接受感染预防的患者被纳入研究。在该研究的描述性分析中发现,尽管研究方案建议在术前抗生素联合使用中使用甲硝唑,但36.1%的患者未接受甲硝唑预防。因此,本分析考虑了甲硝唑对术后感染率的影响。为排除风险特征组间差异导致的任何偏倚,对使用和未使用甲硝唑的组进行配对分析。配对参数为:手术时间、失血量、年龄、糖尿病、肝、肾或慢性气道疾病、免疫抑制治疗和直肠切除术。这导致形成了800对在这些参数方面匹配的对子。然后将这800对分为以下治疗组:第1组:长效头孢菌素(头孢曲松)加或不加甲硝唑(n = 2×491);第2组:短效头孢菌素加或不加甲硝唑(n = 2×133);第3组:广谱青霉素加或不加甲硝唑(n = 2×176)。在所有三个治疗组中,发现甲硝唑联合治疗明显更优。第1组中使用和未使用甲硝唑的术后感染率分别为9.4%和18.7%,第2组分别为12.0%和25.6%,第3组分别为19.9%和29.0%。因此,通过术前使用头孢曲松加甲硝唑进行感染预防可实现最低感染率。因此,在择期结肠手术中,强烈建议除长效β-内酰胺抗生素外,术前给予甲硝唑,因为缺乏针对结肠厌氧菌群的抗生素覆盖会导致术后感染率显著更高。