The article by Kalakouti et al1 from April 2018 underlines the concern about surgical site infections (SSIs) related to colorectal surgery. The authors report an overall 21% rate of SSIs (19% among elective procedures), of which 38% were related to extended-spectrum β-lactamase (ESBL) pathogens.1 These SSI figures are similar to rates observed in a number of Spanish centers reporting data to a regional surveillance program,2 which seems to be improved after a bundle of measures has been implemented.
Our institution, Hospital de Terrassa (Terrassa, Barcelona, Spain), reported a SSI rate of 7% in the patients undergoing elective colorectal surgery during the period of 2015 to 2017 by using a bundle that includes mechanical bowel preparation, oral antibiotic prophylaxis with rifaximin, intravenous antibiotic prophylaxis with amoxicillin-clavulanic acid, and a double ring 360º plastic wound retractor. Twelve SSIs were documented, according to the US Centers for Disease Control and Prevention criteria, in 176 elective patients. Twenty-two germs were isolated, mainly being Enterobacteriaceae; Escherichia coli was the most frequent germ as in the commented paper.1 However, no ESBL strains grew from such cultures, whereas 2 cefamicinase-producing E coli and Klebsiella pneumoniae were observed.
The paper by Kalakouti et al1 raises the unsolved question of preventing SSIs for the highest germ-containing part of the anatomy but only focuses on antibiotic resistance. Even considering the capital importance of such measure, there are other factors related to this topic that are not explained in their article1 (ie, rate of laparoscopic approach, wound protection, oral antibiotic, control of body temperature and glycemia, skin asepsis), which are constitutive of some recommended bundles. But a determinant issue is the high rate of ESBL obtained in the microbiological samples in striking contrast with our institution’s experience. While the ESBL rate in E coli isolated from intra-abdominal infections was reported to be 16.9% for the UK,3 the rate in Spain in a comparable period of time was 9.5% in nosocomial infections.4 Although the samples were not matching, these different rates may indicate a lower incidence of ESBL in our geographical region but could hardly explain the dramatic figures of ESBL shown in the reported paper.1
Perhaps a more aggressive broad-spectrum antibiotic policy would be responsible for the selection of so many ESBL-positive samples. So, a hypothetical intravenous prophylaxis with an antibiotic covering ESBL strains would help in selecting more multiresistant germs, thus generating a worse situation.
Luis Oms, PhD
Hospital de Terrassa (Terrassa, Barcelona, Spain)