ABC | Volume 112, Nº6, June 2019

Original Article Andrade et al. Surgical site infection prevention bundle in cardiac surgery Arq Bras Cardiol. 2019; 112(6):769-774 The epidemiological diagnosis of infections followed the North American Center for Disease Control and Prevention (CDC) criteria. Infections were classified as superficial, deep, or organ/space. The institution has used a protocol for the prevention of SSI since 2003. The bundle of preventive measures at the institution includes the following 6 items to be executed throughout the pre- and postoperative periods: preoperative bath with 2% chlorhexidine, 24 hours before the procedure; hair removal with electric clippers, within 2 hours prior to the start of surgery; maintenance of normothermia, at least 36ºC, during the immediate postoperative period; controlling blood glucose below 200 mg/dl, measured at 6:00 am on the first postoperative day; infusion of surgical antibiotic prophylaxis with anesthetic induction within 60 minutes prior to incision; and additional doses if the procedure lasts more than 4 hours, with a maximum use time of 24-48 hours. The risks scores considered were the ASA classification developed by the American Society of Anesthesiologists and the National Nosocomial Infection Surveillance (NNIS) risk index. The NNIS risk index ranges from 0 to 3, taking the following items into account: contamination potential, procedure duration, and ASA classification. Each item is worth either 0 or 1 points in the score. 24 Data collection made use of the Brazilian Hospital Infection Control Service (SCIH) information system. Additionally, patients’ medical records were reviewed using the Brazilian Medical and Statistical Archive Service (SAME). The medical records were reviewed in the second semester of 2015. Statistical analyses Sample size calculation considered the rates of infection between 2003 and 2012. Considering a number of 900 procedures per year in the hospital and an average SSI rate of 3.23% for the period between 2003 and 2012 and that the complete application of the prevention bundle reduces the infection rate by 60%. 8 a sample of 1.846 medical records was calculated, with an alpha error of 5% and a beta error of 20%. Fisher’s exact test was used for bivariate comparison. Poisson regression was used for multivariate analysis which included variables with p < 0.20 from the bivariate analysis. 9 A significance level of p < 0.05 was considered. The data collected were codified and inserted into a table using the program Microsoft Office Excel 2007, thus creating a databank. Complementary analyses were carried out using the program SPSS, version 18.0. This study was approved by the research ethics committee of the Instituto de Cardiologia of the Fundação Universitária de Cardiologia, on September 17, 2014, under certificate number 4997/14, being accredited by the Brazilian National Commission of Ethics in Research (CONEP), with the Term of Confidentiality for Data Use attached. Results One thousand, eight hundred, and forty-six medical records of patients who underwent major surgical procedures were analyzed, 138 of which were excluded from the study for the following reasons: 23 pediatric patients, 85 deaths or hospitalizations lasting less than 48 hours, and 30 records with incomplete data which did not meet study inclusion criteria. The period studied, thus, included a total of 1,708 major cardiac surgery procedures in 1,708 patients. One hundred and forty-two (8.3%) procedures developed SSI, of which 48.0% (n = 69) were thoracic site infections (13.3% superficial incisional; 24.5% deep incisional; 11.2% organ/space); 40.6% (n = 58) were saphenous vein infections; 7.7% (n = 9) were thoracic site and saphenous vein infections; and 3.0% (n = 4) were endocarditis. In heart transplant procedures, 1 in 4 became infected. The demographic data of patients with and without the presence of SSI are shown in Table 1. The following variables correlated with infection in bivariate analysis: arterial hypertension (p = 0.01), diabetes mellitus (p = 0.001), dyslipidemia (p = 0.05), obesity (p = 0.001), blood glucose level ≥ 200 mg/dl (p = 0.03), public or private hospitalization (p = 0.008), surgical risk index (p = 0.001). The following variables were associated with the diagnosis of SSI in multivariate analysis: surgical risk index, obesity, diabetes mellitus, and blood glucose level (Table 2). Discussion The SSI rate in our study was 8.3%. In developed countries, the SSI rate varies from 1.2% to 5.2%, whereas, in developing countries, it may be as high as 11.8%. Our rate was, therefore, higher than the general SSI rate in developed countries (1.2–5.2%), but lower than the rate in developing countries (11.8%). 3 SSI rates following cardiac surgery in developing countries may vary from 3.5% to 21.0%. 4,5 Diabetes mellitus, blood glucose level, obesity, and surgical risk index are factors associated with SSI, in accordance with the latest World Health Organization report (WHO, 2016), which underlines these factors in relation to risks that affect HAI. 3 SSI risk factors are complex, and their prevention requires the integration of a range of measures, before, during, and after surgery. Prevention is the principal focus of the Institute for Healthcare Improvement (IHI) and the Surgical Care Improvement Project (SCIP) in the USA, both of which recommend a group of preventive measures to be taken. 10 These measures, called a “bundle,” are carried out together to obtain better results than would be obtained by individual application. SSI prevention bundles in cardiac surgery involve the use of prophylactic antibiotics during the immediate pre‑and postoperative period (up to 48 hours following incision); blood glucose level control during the first and second postoperative period; temperature and oxygenation control; decolonization of patients with intra-nasal mupirocin, and preoperative chlorhexidine bath. 3 , 7,10 In our study, adherence to surgical prophylaxis protocol was not associated with a reduction in SSI rates. Studies in surgical procedures indicate that antimicrobial use within 60 minutes before the procedure has been associated with reduced infection rate. 3,11,12 770