Sternal Wound Infections in Pediatric Congenital Cardiac Surgery: A Survey of Incidence and Preventative Practice

Transcription

1 Sternal Wound Infections in Pediatric Congenital Cardiac Surgery: A Survey of Incidence and Preventative Practice Cathy S. Woodward, DNP, RN, Minnette Son, MD, John Calhoon, MD, Joel Michalek, PhD, and S. Adil Husain, MD Departments of Pediatrics, Division of Critical Care, Cardiothoracic Surgery, and Epidemiology and Biostatistics, University of Texas Health Science Center San Antonio, San Antonio, Texas Background. Guidelines exist for prevention of sternal wound infections (SWI) in adults. There are no guidelines for pediatric patients and limited reports on SWI incidence. The purpose of this study was to determine the incidence of, and preventative practice regarding pediatric SWIs with a long-term aim to develop best practice guidelines. Methods. Eighty-nine congenital heart programs were sent a 31 question on-line survey regarding pediatric SWI. Results. Thirty eight (43%) of the 89 programs responded. They reported 8,774 pediatric congenital procedures with a mean SWI rate of 1.53% (range, 0 to 9.09). Mean yearly volume was 237 operations (range, 50 to 720). Neither program size nor delayed sternal closure was associated with increased incidence of SWI. Variations in preoperative measures, antibiotic regimens, and wound care did not statistically impact incidence of SWI. Programs with protocols to monitor and control blood glucose levels postoperatively had statistically lower infection rates (1.04 vs 2.35, p 0.03), and those that sent mediastinal cultures at time of delayed sternal closure reported lower infection rates (1.34 vs 1.74, p 0.051). Conclusions. This report provides a multiinstitutional SWI incidence from pediatric programs of 1.53%. Despite variations in clinical practice between programs, this survey revealed two strategies resulting in reduced SWIs; protocol-based management of glucose levels and mediastinal wound cultures sent at time of closure. Pediatric programs do not consistently follow adult preventative guidelines. Multicenter randomized research is needed to formulate preventative guidelines to reduce the incidence of pediatric SWI. (Ann Thorac Surg 2011;91: ) 2011 by The Society of Thoracic Surgeons Sternal wound infections (SWI) are a costly complication of cardiac surgery. Increased morbidity, mortality, and financial cost of extended length of stay and additional surgical procedures can be expected for children who develop wound infections post cardiac surgery. The average increased cost for children at one children s hospital who had surgical site infections was approximately $28,000 per child more than those who did not develop an infection [1]. The incidence of SWI in adults is well reported and varies between 0.5 to 7.4% [2 4]. A limited number of recent studies report an incidence of pediatric SWIs between 0.2% and 4.8% [5 7]. Comparisons between pediatric programs are difficult because the single institution reports vary on which level of SWIs are included; superficial incisional, deep incisional, or organ-space. In addition, preventative measures are not addressed. Current practice by cardiothoracic programs to prevent SWIs in children is unknown. An assumption is most programs are either using adult guidelines or developing program-based protocols, as there are no specifically pediatric guidelines [8, 9]. Adult guidelines are based on adult risk factors such as obesity, diabetes, and smoking, which Accepted for publication Oct 12, Address correspondence to Dr Woodward, 2430 Enfield Grove, San Antonio, TX 78231; woodwardc@uthscsa.edu. differ from risk factors for SWI in children [10 12]. The incidence of SWI in children has not been clearly reported. A study of nosocomial infections in pediatric intensive care units, utilizing indices from the National Nosocomial Infections Surveillance System (now known as the National Healthcare Safety Network [NHSN]), reported only organspace mediastinitis sternal wound infection incidence, thus possibly underestimating their total incidence of infection [13]. The efforts to reduce SWIs in children are less well studied and we have found no published guidelines for the prevention of this serious postoperative complication. The purpose of this research was to determine the incidence of SWIs in children undergoing cardiac surgery in the United States and to survey current practice to prevent this life threatening complication. The long-term aim was to develop best practice guidelines to prevent SWIs in children undergoing cardiac surgery. Material and Methods This research was approved by the Investigational Review Board at the University of Texas Health Science Center San Antonio. An online survey was sent to the chief surgeons for 89 pediatric cardiothoracic programs in the United States. Follow-up survey requests were 2011 by The Society of Thoracic Surgeons /$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 800 WOODWARD ET AL Ann Thorac Surg PEDIATRIC STERNAL WOUND INFECTIONS 2011;91: sent online and then mailed surveys were sent to nonresponders. The 31 survey questions addressed preoperative skin preparation, intraoperative antibiotic prophylaxis, and postoperative wound care and antibiotic prophylaxis. The chief surgeons were asked to provide their total number of SWIs including superficial incisional (involving the skin or subcutaneous tissue), deep incisional (fascial and muscle layers), and organ spacemediastinitis (involving the sternum and medastinum) for children less than 18 years of age. For the purposes of our study, the definition of a SWI was any sternal surgical site infection which developed after surgery to correct or palliate a child with congenital heart disease. Contributing programs also provided the number of pediatric cardiac operations requiring sternotomy and the number of patients who had delayed sternal closure for the calendar year Statistical Analysis The SAS software package 9.1 (version 9.1; SAS Institute Inc, Cary, NC) was used for analysis. Descriptive statistics were calculated for each item in the survey. We used the mean and standard deviation to summarize the continuous measures such as total number of sternal surgical site infections. Percentages were used to summarize the categoric measures. The univariate relationship between infection rate and each prevention variable was examined by independent t test and a one-way analysis of variance F test was used to compare the difference in infection rate for dichotomous or multilevel prevention variables, respectively. If the distribution of infection rate was skewed, a nonparametric Mann-Whitney U test or Kruskal-Wallis H test were used for dichotomous or multilevel prevention variables, respectively. A p value of 0.05 or less was considered to indicate statistical significance. Results Thirty eight of the 89 programs responded for a 43% response rate. One data from one program were not included because they did not provide data on their infection rate, numbers of operations, or number of delayed sternal closures. The data collected represent a total of 8,774 pediatric cardiac surgery patients. The mean SWI incidence rate was 1.53 per 100 cases with a range of 0 to 9.9. Programs varied in volume of patients at each institution with a range of 50 to 720 pediatric cardiac operations performed in The 20 responding programs reporting less than 200 sternotomies per year had a mean infection rate of 1.4 per 100 and the remaining 17 programs with over 200 sternotomies per year had a mean infection rate of 1.7 per 100. Size of program did not affect infection rates (p 0.142) (Fig 1). When infection rates were stratified by their percentage of delayed sternal closures no difference in infection rate was noted (Fig 2). More specifically, programs which reported a greater than or less than the mean 1.5% incidence of SWI had no statistically significant differences in regard to their respective incidence of delayed sternal closures. Fig 1. Number of operations stratified by infection rate. The approximate means of the infection rate and the number of operations performed are indicated with vertical and horizontal dashed straight lines, respectively. The correlation between the number of operations and infection rate is 0.03 (p 0.84). Simple linear regression gives the following equation: Infection rate * (number of surgeries performed) (represented in the figure as the thick black line). The R 2 for this equation is The preoperative measures to prevent SWIs addressed the use of preoperative baths or showers, the routine use of steroids, and screening of nares for carriage of Staphylococcus aureus (S. aureus) (Table 1). A majority of programs did not prescribe preoperative antiseptic baths or showers but those who did were more likely to use chlorhexidine, with no significant difference in infection rate. In nearly 60% of programs steroids were routinely prescribed prior to patients being placed on cardiopulmonary bypass, with no significant difference in infection rate (p 0.12). Most programs (76%) did not perform a screening of nares for carriage of S. aureus with no difference in the rate of infection. Survey questions which reviewed intraoperative measures to reduce SWIs included which skin antiseptics were used to prepare the operative site and the use and timing of prophylactic antibiotics. The choices of skin antiseptic in the operating room in most programs was iodophor over chlorhexidine (54% vs 34%) but the choice did not significantly affect the mean infection rate (1.3 vs 2.7, p 0.22). In 97% of programs, prophylactic antibiotics were administered one hour before skin incision. The antibiotic most often used was cefazolin (78%) with 78% of programs using a single antibiotic preoperatively. Those programs in which more than one antibiotic was used did not differ significantly in their infection rate when compared with those in which one preoperative antibiotic was used (1.57 vs 1.52, p 0.45). Postoperative measures to reduce SWIs included choice of postoperative antibiotics, duration of postoperative antibiotic use, dressing applied to patients who left

3 Ann Thorac Surg WOODWARD ET AL 2011;91: PEDIATRIC STERNAL WOUND INFECTIONS 801 was cefazolin (76%). Twenty seven programs (73%) discontinued antibiotics based on time; 24, 48, or 72 hours (41%, 30%, and 3%), after surgery. In the remaining 10 programs antibiotics were continued until removal of central lines and or chest tubes with no significant difference in infection rate between the two groups (1.62 vs 1.29, p 0.97). Most delayed sternal closures were performed in the intensive care unit (90%) rather than returning the patient to the operating room. Sending a wound culture to laboratory at the time of delayed sternal closure approached significance for reduced infection rate and the use of a protocol to monitor and control blood glucose levels after surgery did result in a statistically significant difference in infection rate (Table 3). Fig 2. Percentage of infections stratified to number of delayed sternal closures. The approximate mean of the infection rate is indicated with a horizontal dashed straight line. The correlation between the percentage of operations with a delayed sternal closure and infection rate is 0.05 (p 0.76). Simple linear regression gives the following equation: Infection rate * (percentage of delayed sternal closures) (represented in the figure as the thick black line). The R 2 for this equation is the operating room with closed sternums, choice of location for delayed sternum closure procedure, the routine use of mediastinal cultures sent at time of chest closure, and the use of a protocol for monitoring and controlling blood glucose levels after surgery. Table 2 contains the varied responses to questions concerning postoperative dressings for closed sternums and the choice of antibiotics for patients leaving the operating room with open sternums. The postoperative antibiotic most often prescribed for patients with closed sternums Comment This is a reported survey of pediatric cardiothoracic surgery programs to determine the incidence of SWIs in children and efforts to reduce SWIs post pediatric cardiac surgery. The reported incidence of SWIs for this study was 1.5% (range, 0.0 to 9.09), which is lower than the 1.8% pooled mean rate of infections for cardiac operative procedures in the high-risk category reported by the NHSN [14]. Comparing infection rates in children with rates in adults is difficult using this index. The risk index category as determined by the NHSN model is based on variables in adults which have not been validated in children, and have been challenged by a recent study [15]. A risk index based on pediatric risk factors would provide a better way to compare individual program infection rates and provide a benchmark for quality improvement measures. The CDC (Centers for Disease Control and Prevention) guideline for the prevention of surgical site infections states all surgical site infection prevention measures effective in adult surgical care are indicated in pediatric surgical care [8, p.250]. The expectation for pediatric cardiothoracic programs would be that preoperative, intraoperative, and postoperative interventions to prevent SWIs in infants and children would follow the current CDC guide- Table 1. Preoperative Measures to Reduce Sternal Wound Infections and Incidence of Infections Incidence Programs N (%) Mean (SD) p Value a Is preoperative bathing or showering with skin antiseptics prescribed for children? Yes 18 (47.37%) 1.37 (1.09) No 20 (52.5%) 1.67 (2.1) Children are routinely given steroids before being placed on cardiopulmonary bypass? Yes 22 (59.46%) 1.86 (1.96) No 15 (40.5%) 1.06 (1.1) Children undergo preoperative screening of their nares for carriage of S. aureus? Yes 9 (24.32%) 1.14 (0.82) No 28 (75.68%) 1.66 (1.89) a Kruskal-Wallis Test.

4 802 WOODWARD ET AL Ann Thorac Surg PEDIATRIC STERNAL WOUND INFECTIONS 2011;91: Table 2. Postoperative Measures to Reduce Sternal Wound Infections and Incidence of Infection Incidence Programs N (%) Mean (SD) p Value a For patients leaving the operating room with STERNUM CLOSED, the sterile dressing most often applied is Skin or tissue adhesive; eg, dermabond 8 (21.62%) 1.28 (1.21) Sterile gauze only 2 (5.41%) 1.17 (1.65) Steri-strips only 2 (5.41%) 2.47 (0.93) Tissue adhesive AND sterile gauze dressing 5 (13.51%) 1.62 (1.57) Steri-strips AND sterile gauze dressing 16 (43.24%) 1.46 (2.13) Other dressing (please specify) 4 (10.81%) 1.93 (1.68) The postoperative antibiotic(s) most routinely used on children with OPEN STERNUM is(are) 1st generation cephalosporin; eg, cefazolin 15 (40.54%) 1.71 (1.29) 2nd generation cephalosporin; eg, cefuroxime 3 (8.11%) 0.78 (1.35) 3rd generation cephalosporin; eg, cefotaxime, 2 (5.41%) 0.45 (0.64) ceftriaxone, ceftazadime Vancomycin only 4 (10.81%) 1.23 (0.7) Vancomycin and 1st generation cephalosporin 1 (2.7%) 0.28 (.) Vancomycin and 2nd generation cephalosporin 4 (10.81%) 3.92 (3.5) Vancomycin and gentamycin 5 (13.51%) 0.58 (0.5) Other (please specify) 3 (8.11%) 1.34 (1.21) a Kruskal-Wallis test. lines. Pediatric programs responding to this survey did not consistently follow adult guidelines. The preoperative prevention recommendations include hand hygiene and preoperative hair removal, which were not addressed in this study. The study did address preoperative skin antisepsis, screening for nasal carriage of S. aureus, nasal decolonization, and administration of steroids prior to cardiopulmonary bypass. Fifty-three percent of responding programs did not prescribe preoperative bathing or showering with skin antiseptics and there was no significant increase in the mean infection rate between those who do with those who do not use this intervention (1.37 vs 1.67, p 0.78). A Cochrane review [16] of preoperative bathing with skin antiseptics to prevent surgical site infections in adults provided no clear evidence of the benefit of bathing with chlorhexidine over other wash products. For adults, to shower or to bathe with an antiseptic agent is strongly recommended yet most pediatric cardiothoracic programs do not adhere to this recommendation. Screening of nares for carriage of S. aureus was not routinely performed in 76% of pediatric cardiothoracic programs and there was no significant difference in infection rates. Eighty-nine percent did not prescribe topical antibiotic ointment to eradicate S. aureus in nares. The 1999 adult guidelines of the Hospital Infection Control Practices Advisory Committee [8] do not recommend routine screening for S. aureus or the application of mupirocin to the nares to reduce surgical site infections. However, the 2008 guidelines by the Association for Professionals in Infection Control and Epidemiology [9] do recommend nasal decolonization with mupirocin due Table 3. Postoperative SWI Prevention Methods and Incidence of Infection Incidence Programs N (%) Mean (SD) p Value a For children with delayed sternal closure mediastinal cultures are routinely sent at the time of chest closures Yes 19 (51.35%) 1.34 (2.1) No 18 (48.65%) 1.74 (1.14) Does your program have a protocol for monitoring and controlling blood glucose levels in children after their cardiac surgery? Yes 23 (62.16%) 1.04 (0.95) No 14 (37.84%) 2.35 (2.3) a Kruskal-Wallis test. SWI sternal wound infection.

5 Ann Thorac Surg WOODWARD ET AL 2011;91: PEDIATRIC STERNAL WOUND INFECTIONS 803 to the risk of S. aureus as a cause for SWIs, and because it is relatively easy to implement. A randomized, controlled study in pediatrics is lacking on this subject; however, a case report of six pediatric cases of S. aureus sternal wound infections in children whose nares were colonized with S. aureus concluded that preoperative decolonization for children might be warranted [17]. Approximately 60% of the responders to this study reported administering steroids to children before being placed on bypass. There was no significant increase in infections in programs in which steroids are used preoperatively before cardiopulmonary bypass. The risk of nosocomial infection after perioperative steroid use has yet to be determined. Most pediatric cardiothoracic programs are complying with the current recommendations for antimicrobial prophylaxis and the timing of administration prior to incisions. Ninety-seven percent of programs reported the administration of antibiotics one hour prior to incision. The 2008 guidelines recommend a single antibiotic, cefazolin or cefuroxime, for phophylaxis [9]. Ninety-three percent of the survey respondents reported compliance with this recommendation. Vancomycin was the most commonly used second antibiotic but only 8 (22%) of programs used a second antibiotic prior to skin incision. Prophylactic antibiotic coverage should be discontinued within 48 hours of surgery [9, 18]. For children who are returned from surgery with sternum closed, the majority (73%) of programs comply with this guideline. The infection rate for those who did not comply was not statistically significantly different. Nineteen percent continued antibiotics until either central lines and (or) chest tubes were removed. A study of prophylactic antibiotic use for longer than 48 hours in over 2,600 adults after coronary artery bypass graft [19] failed to demonstrate a decreased infection rate while increasing the risk of acquired antibiotic resistance. Maher and colleagues [20] retrospectively reviewed over 3,400 pediatric cardiac patients and found a decrease in SWIs when antibiotics were continued until mediastinal or thoracostomy tubes were removed. In the Maher and colleague study comparing three antibiotic prophylaxis protocols, the infection rate for patients whose antibiotics were continued until removal of thoracostomy tubes was 1.67 versus 6.58 (p 0.05) for those whose antibiotics were discontinued at 48 hours postoperatively [20]. Each of the three protocols which were reviewed in the Maher and colleagues study were used for at least one year over a period of six years. For programs responding to the survey there was no statistically significant difference in the incidence of infection when antibiotics were stopped based on time versus the presence of thoracostomy tubes or mediastinal drains. The timing and use of postoperative antibiotic prophylaxis to prevent infection in children is not well defined and needs further study. The 1999 postoperative incision care guidelines [8] recommend a sterile dressing for 24 to 48 hours and any dressing changes should be done using a sterile technique [8]. The most common choice was adhesive skin closure strips with a sterile gauze dressing (43%). In 46% of programs the sterile technique to change these dressings was not used. Despite not following national guidelines, there was no significant difference in infection rates between types of dressings or whether a sterile technique was used. Programs that had a protocol to monitor and control blood glucose postoperatively had a significantly lower mean infection rate than programs that did not utilize a protocol (1.04 vs 2.35, p 0.03). In a recent retrospective case-control study [21], 24-hour peak blood glucose greater than 130 mg/dl was found to be a multivariate predictor for mediastinitis after pediatric cardiac surgery. This supports findings from similar studies in adults [22]. A prospective, randomized controlled study of tight glycemic control in a pediatric intensive care unit found targeting blood glucose control to normal fasting concentrations improved short-term outcome; however, 25% of the patients in the intensive glucose control group experienced hypoglycemia [23]. Seventy-five percent of the patients in that study had a diagnostic category of cardiac surgery for congenital heart defects. Though discussion of protocols and efforts to maintain glycemic control were beyond the scope of this survey, it presents an opportunity for study. A randomized, prospective study of glycemic control in pediatric cardiac surgery patients as a method to reduce the risk of mediastinitis is warranted. Another response which approached significance for lower mean infection rates was found in programs in which mediastinal cultures were sent to laboratory at the time of chest closure in those patients delayed sternal closure (1.34 vs 1.74, p 0.051). There are no recommendations for routine culture of wounds as a method to prevent sternal wound infections. In our survey we did not request information about interventions for those patients with positive culture results. Perhaps those programs in which cultures are sent at the time of chest closure would initiate empiric antibiotics in the face of a positive culture with no clinical evidence of infection. A positive culture result might also raise an index of suspicion with closer monitoring of sternal wounds for infection postclosure. More study of this intervention to reduce SWIs is needed. This study reporting on a large sample from pediatric thoracic surgery programs regarding their attempts to reduce sternal wound infections is limited by its survey design. The errors, which result from nonresponse, are a concerning source of bias. The nature of the bias of the responding programs is unknown but may have been related to their programs interest in reducing infection rates and may not reflect other pediatric surgery programs. Nonresponders may have simply lacked the time or interest to complete the survey. The choice of using the internet or mail may have influenced some program directors to complete the survey. No attempts were made to contact the program directors directly by phone which might have led to an improved response rate. Despite the limitations, the data reflect the practice of a large number of programs throughout the United States. Employing larger databases such as that provided by the STS (Society of Thoracic Surgeons) allows for more thorough data gathering and analysis for evaluating best practice guidelines. However, the STS database does not

6 804 WOODWARD ET AL Ann Thorac Surg PEDIATRIC STERNAL WOUND INFECTIONS 2011;91: contain the specific variables studied in this survey regarding protocols for the prevention of SWI. This may be a direction for database leadership to pursue to allow for a more efficient investigation of similar clinical queries and the development of best practice guidelines. This report provides multiinstitutional SWI incidence from pediatric congenital cardiac surgery programs and identifies practice to reduce SWIs. The incidence of pediatric SWI was 1.53%, slightly lower than the 1.8% reported by the NHSN. There are disparities in clinical practice between programs with no clear indication of what works to reduce SWIs in children. An interesting finding is pediatric programs surveyed are not consistently following components of national adult guidelines on preoperative showering and nasal decontamination for the prevention of surgical site infections, with no increased rate of infections noted. Glucose monitoring and protocol-based management, along with mediastinal wound cultures sent at time of chest closure, were associated with lower infection rates. Reporting preventative strategies to a national organization such as the STS database may help directors to develop SWI prevention protocols in their programs. Multicenter randomized controlled research is needed to formulate preventative guidelines to reduce the incidence of pediatric SWI. The authors wish to thank the following Institutions (pediatric cardiothoracic programs) that contributed data for this study: Akron Children s Hospital, Alfred DuPont Hospital for Children, Arkansas Children s Hospital, Arnold Palmer Hospital for Children, Children s Healthcare of Atlanta, Children s Hospital at Oklahoma University, Children s Hospital Boston, Children s Hospital of Michigan, Children s Hospital of New York, Children s Hospital of Orange County, Children s Hospital of Philadelphia, Children s Hospital Omaha, Children s Hospital Peoria, Children s Memorial Hospital Chicago, Christus Santa Rosa Children s Hospital, Cincinnati Children s Hospital, Duke Children s Hospital, Helen DeVos Children s Hospital, Johns Hopkins Children s Center, Kentucky Children s Hospital, Kosair Children s Hospital, Loma Linda University Children s Hospital, Maria Fareri Children s Hospital, Mary Bridge Children s Hospital, Methodist Children s Hospital, Monroe Carell Jr. Children s Hospital at Vanderbilt, North Carolina Children s Hospital, Penn State Hershey Children s Hospital, Riley Hospital for Children, Schneider Children s Hospital, Seattle Children s Hospital, St. Christopher Hospital for Children, St. Louis Children s Hospital, Texas Children s Hospital, University of Maryland Hospital for Children, West Virginia University Children s Hospital, Wolfson Children s Hospital, and Yale-New Haven Children s Hospital. References 1. Sparling KW, Ryckman FC, Schoettker PJ, et al. Financial impact of failing to prevent surgical site infections. Qual Manage Health Care 2007;16: Salehi Omran AS, Karimi A, Ahmadi SH, et al. Superficial and deep sternal wound infection after more than 9000 coronary artery bypass graft: incidence, risk factors and mortality. BMC Infect Dis 2007;7: Sharma M, Derriel-Cass D, Baran J Jr. Sternal surgical-site infection following coronary artery bypass graft: prevalence, microbiology, and complications during a 42-month period. Infect Control Hosp Epidemiol 2004;25: Schimmer C, Reents W, Berneder S, et al. Prevention of sternal dehiscence and infection in high-risk patients: a prospective randomized multicenter trial. Ann Thorac Surg 2008;86: Al-Sehly A, Robinson J, Lee B, et al. Pediatric poststernotomy mediastinitis. Ann Thorac Surg 2005;80: Tortoriello T, Friedman J, McKenzie D. Mediastinitis after pediatric cardiac surgery: a 15-year experience at a single institution. Ann Thorac Surg 2003;76: Holzmann-Pazgal G, Hopkins-Broyles D, Recktenwald A, et al. Case-control study of pediatric cardiothoracic surgical site infections. Infect Control Hosp Epidemiol 2008;29: Mangram AC, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999;20: Kohut K. Guide for the prevention of mediastinitis surgical site infections following cardiac surgery. Association for Professionals in Infection Control and Epidemiology Available at: APICEliminationGuides/Mediastinitis_elimination_guide.pdf. Accessed October 1, Mehta P, Cunningham C, Colella C, Alferis G, Weiner L. Risk factors for sternal wound and other infections in pediatric cardiac surgery patients. Pediatr Infect Dis J 2000;19: Cayci C, Russo M, Cheema F, et al. Risk analysis of deep sternal wound infections and their impact on long-term survival: a propensity analysis. Ann Plast Surg 2008;61: Kagen J, Lautenbach E, Bilker W. Risk factors for mediastinitis following median sternotomy in children. Pediatr Infect Dis J 2007;26: Kagen J, Bilker W, Lautenbach E. Risk adjustment for surgical site infection after median sternotomy in children. Infect Control Hosp Epidemiol 2007;28: Edwards J, Peterson K, Andrus M, Dudeck M, Pollock D, Horan T. National Healthcare Safety Network (NHSN) Report, data summary for 2006 through 2007, issued November Am J Infect Control 2008;36: Casanova J, Herruzo R, Diez J. Risk factors for surgical site infection in children. Infect Control Hosp Epidemiol 2006;27: Webster J, Osborne S. Preoperative bathing or showering with skin antiseptics to prevent surgical site infection. Cochrane Database Syst Rev 2007;2:CD Ruef C, Fanconi S, Nadal D. Sternal wound infection after heart operations in pediatric patients associated with nasal carriage of staphylococcus aureus. J Thorac Cardiovasc Surg 1996;112: Edwards FH, Engelman RM, Houck P, Shahian D, Bridges C; Society of Thoracic Surgeons. The Society of Thoracic Surgeons practice guideline series: antibiotic prophylaxis in cardiac surgery, part 1: duration. Ann Thorac Surg 2006;81: Harbarth S, Samore H, Lichtenberg D, Carmeli Y. Prolonged antibiotic prophylaxis after cardiovascular surgery and its effect on surgical site infections and antimicrobial resistance. Circulation 2000;101: Maher K, VanDerElzen K, Bove E, Mosca R, Chenoweth C, Kulik T. A retrospective review of three antibiotic prophylaxis regimens for pediatric cardiac surgical patients. Ann Thorac Surg 2002;74: Ghafoori F, Twite M, Friesen R. Postoperative hyperglycemia is associated with mediastinitis following pediatric cardiac surgery. Pediatric Anaesth 2008;18: Swenne CL, Lindholm C, Borowiec J, Schnell AE, Carlsson M. Peri-operative glucose control and development of surgical wound infections in patients undergoing coronary artery bypass graft. J Hosp Infect 2005;61: Vlasselaers D, Milants I, Desmet L, et al. Intensive insulin therapy for patients in paediatric intensive care: a prospective, randomised controlled study. Lancet 2009;373: