Hyperglycemia as a risk factor for surgical site infections in patients undergoing mastectomy

Hyperglycemia as a risk factor for surgical site infections in patients undergoing mastectomy Diana Vilar-Compte, MD, MsC,a Ine´s A´lvarez de Iturbe, MD,a Alexandra Martı´n-Onraet, MD,a Maricruz Pe´rez-Amador, MD,b Claudia Sa´nchez-Herna´ndez, MD,a and Patricia Volkow MDa Mexico City, Mexico

Background: The aim of this study was to evaluate the association between perioperative hyperglycemia and surgical site infections (SSIs) in patients undergoing mastectomy. Methods: In this nested case-control study, patients undergoing mastectomy from May 2004 to June 2006, at the National Cancer Institute (INCan), Mexico, were included. Five blood glucose values were obtained for each patient. Patients were followed prospectively by direct observation for at least 30 days. Results: A total of 260 patients were included. Patient characteristics were similar in cases and controls. Cases were older (.50 years) (P 5 .001) and were more frequently treated with concomitant chemoradiation (P , .0001) than controls. Fifty cases (23.8%) developed an SSI. At least 1 measurement of blood glucose value $ 150 mg/dL increased the risk for SSI (odds ratio [OR] 5 3.05; 95% confidence interval [CI] 5 1.5 to 6.3; P 5 .006). Variables associated with SSI after logistic regression analysis included age .50 years (OR 5 3.7; 95% CI 5 1.5 to 9.2; P 5 .005), preoperative concomitant chemoradiation treatment (OR 5 2.8; 95% CI 5 1.4 to 5.8, P 5 .0004), and any blood glucose value $ 150 mg/dL (OR 5 2.9, 95% CI 5 1.2 to 6.2; P 5 .02). Conclusion: Postoperative SSI was a very frequent complication in this cohort. Our data indicate that higher blood glucose values are an independent risk factor for postoperative SSI. Preoperative concomitant chemoradiation and age . 50 years were also risk factors for developing an SSI in patients undergoing mastectomy. (Am J Infect Control 2008;36:192-8.)

Surgical site infections (SSIs) remain one of the most common causes of morbidity in surgical patients despite advances in surgical practice and antibiotic prophylaxis. Mastectomies have traditionally been considered clean procedures; most studies suggest that the infection rate in clean surgery is 3% or lower.1-3 According to the National Nosocomial Infections Surveillance System, the SSI rate for mastectomies during the last decade ranged between 2.07% and 3.9%;2 other series have reported higher infection rates, however.4-7 In cardiovascular surgery, hyperglycemia has been associated with adverse outcomes in acutely ill adult patients, and its close control has been shown to improve mortality and morbidity in various settings.8 Hyperglycemia is a growing problem in modern

medicine; evidence from observational studies indicates that 1/3 of hospitalized patients with hyperglycemia have mean blood glucose values of . 200 mg/dL during their stay.9 In patients with trauma, a single blood glucose value of $ 150 mg/dL is associated with worse outcome.10 The risk of SSI is correlated with the degree of glucose elevation in patients with and without a history of diabetes; strict glucose control has proven to decrease the rate of SSIs.11,12 In surgical intensive care units (ICUs), strict glycemic control has proven to be safe and effective in reducing the incidence of nosocomial infections.13 The present study was undertaken to investigate the relationship between hyperglycemia and SSI in breast cancer patients undergoing mastectomy.

METHODS From the Department of Infectious Diseases,a and Department of Anesthesia,b National Cancer Institute, Mexico City, Mexico. Address correspondence to Diana Vilar-Compte, MD, MsC, Departamento de Infectologı´a, Instituto Nacional de Cancerologı´a, Av San Fernando 22, Col Seccio´n XVI, 14080 Me´xico, DF, Me´xico. E-mail: [email protected].

0196-6553/$34.00 Copyright ª 2008 by the Association for Professionals in Infection Control and Epidemiology, Inc. doi:10.1016/j.ajic.2007.06.003

192

Background and interventions The study was conducted at the Breast Tumor Department of the National Cancer Institute (INCan) in Mexico City, a 150-bed teaching and referral hospital for adult patients with cancer. The study was approved by the INCan’s Institutional Review Board. All patients with breast cancer scheduled for a mastectomy were invited to participate between May 2004 and June 2006, and all patients who chose to participate in the study provided written informed consent.

Vilar-Compte et al

Patient management and surveillance Five glucose values were obtained for each patient: fasting blood glucose at the time of surgery scheduling (taken from the patient’s chart), at anesthesia induction, intraoperatively, at recovery room arrival, and 24 hours postsurgery, by means of capillary blood measurement (Accu-Chek Sensor Comfort, Roche Diagnostics, Basel, Switzerland). Perioperative hyperglycemia was defined as any intraoperative and/or 24-hour postsurgical glucose value $ 150 mg/dL. All patients were operated on by the same surgical team and managed following the standard INCan care protocol. Each patient was routinely evaluated by the surgeon preoperatively, and surgical procedures were performed according to accepted techniques. Perioperative antibiotic prophylaxis (PAP) was prescribed by the surgeon if deemed necessary (according to our guidelines, not prescribing PAP for breast surgery is accepted) within 2 hours of surgery; cefuroxime (1.5 g IV) or cephalotin (1 g IV) were accepted antibiotics for prophylaxis, and clyndamicin was an option for those patients allergic to penicillin. All patients received general balanced anesthesia. Second-level axillary node dissection was performed in modified radical mastectomies, and third-level dissection was performed in radical mastectomies. A Biovac (Biomertrix, Israel) was inserted at the chest anterior wall, and in patients undergoing axillary node dissection another Biovac was inserted at the axilla. Wounds were closed by primary intention and covered by regular gauze for 24 hours, after which the drapes were removed, and wounds were washed with soap and water during the daily shower. At hospital discharge, patients were instructed to do the same every 24 hours and to cover the incision with regular gauze. They were also instructed to measure the amount of fluid collected by their drains every 12 to 24 hours through evacuation of the drain bulb. During the postoperative period, patients were followed with daily chart review, microbiology reports, and rounds at the bedside while in the hospital. Hyperglycemia was managed only in diabetic patients with a regular insulin nomogram according to each patient’s blood glucose value. All patients resumed their usual blood glucose–lowering medications at discharge. After hospital discharge, a team physician conducted direct observation once a week along with the surgeon until the final surgical follow-up at the outpatient wound care clinic.7 Data on wound aspect and the amount and characteristics of drainage material was collected at each visit on a standardized form. The Biovac was removed when the patient’s average daily discharge was # 30 mL. All patients were followed for at least 30 days at the clinic. A culture was obtained when infection was

April 2008

193

suspected due to pain or tenderness, localized swelling, redness or heat, purulent drainage from the superficial or deep incision, fever $ 388C, or the surgeon’s suspicion of an infected wound. SSI was classified using Centers for Disease Control (CDC) definitions for surgical infection.14

Statistical analysis All data collected were introduced into a database (Paradox 9; Corel, Ottawa, Ontario, Canada). Variables studied included age, body mass index (BMI), hospital stay (preoperative and postoperative), type of mastectomy, smoking, diabetes mellitus, high blood pressure, previous chemothrapy and/or radiation therapy, American Society of Anesthesiologists score, prophylactic and postoperative antibiotics, immediate breast reconstruction, and other wound complications, such as flap necrosis, dehiscence, hematoma, and seroma formation. We conducted a case-control analysis, with patients with SSI as cases and patients undergoing mastectomy who were infection-free by the 30th postoperative day as controls. The cases and controls were all selected from the same population (patients undergoing mastectomy between May 2004 and June 2006) and followed by the same means. For analysis, SSI per 100 surgeries was calculated. Student’s t and x2 tests were used as appropriate, using SPSS software (SPSS Inc, Chicago, IL). To test the association between SSIs and possible risk factors, odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using Epi Info 6 (CDC, Atlanta, GA). Logistic regression analysis was conducted using SPSS. A P value , .05 was considered statistically significant.

RESULTS In the 260 patients, a total of 50 SSIs were registered, 37 (74%) superficial incisional and 13 (26%) deep incisional infections. Six patients developed 2 SSIs. Mean time from surgery to SSI diagnosis was 20.9 days (range, 4 to 42 days); 2 patients developed flap necrosis and dehiscence before the presenting infection and were diagnosed on the 49th and 71st postoperative days. Of the 50 SSIs, 36 (72%) were cultured. The bacteria isolated included Staphylococcus aureus (n 5 8), Escherichia coli (n 5 5), Pseudomonas aeruginosa (n 5 4), Staphylococcus epidermidis (n 5 3), Enterobacter cloacae (n 5 2), Serratia spp (n 5 1), and other bacteria (n 5 8). Five cultures were negative. Tumor, node, and metastasis (TNM) staging for patients at the time of surgery was as follows: in situ, 6 (2.34%); early stage of cancer (I and IIA), 79 (30.8%); locally advanced (IIB–IIIC), 134 (52.3%); metastatic (IV), 16 (6.2%); phylodes, 8 (3.1%), and Paget, 2 (0.78%).

194

Vol. 36 No. 3

Nine patients (3.5%) underwent biopsy elsewhere for which the pathology reports were unavailable and thus staging was incomplete at the time of surgery. Table 1 gives the characteristics of the patients studied. Surgery-associated variables were similar in the cases and controls except for the higher prevalence of more radical procedures in the case group (Table 1). Mean amounts of bleeding (274.5 6 144.7 mL vs 246 6 189.1 mL; P 5 not significant [NS]) and duration of surgery (163.6 6 65.5 min vs 156.1 6 95.6 min; P 5 NS) were similar in the cases and controls. PAP was prescribed to 229 of the 260 patients (88.1%). As shown in Table 1, PAP was prescribed slightly more often in cases (94%) than in controls (86.7%) (P 5 NS). Cefuroxime was the most frequently used antibiotic for prophylaxis, prescribed in 43 of the cases (91.5%) and 180 of the controls (98.9%) (P 5 NS). Cephalotin was prescribed for 3 cases (6.3%), and ciprofloxacin (which is not an appropriate prophylactic antibiotic) was prescribed for 1 case (2.0%) and 2 controls (1.1%). Overall, correct selection of PAP was achieved in 91% of cases and 90% of controls (P 5 NS). Timeliness of antibiotic prophylaxis was also similar in the 2 groups (70% vs 72%; P 5 NS) (Table 1). The prevalences of diabetes and high blood pressure were higher in the infected patients, as shown in Table 1, but these disorders were not associated with an increased risk of SSI (P 5 NS). Overweight (BMI . 27.3) was as common in cases as in controls (P 5 NS). The cases were older than controls (P 5 .002), and locally advanced breast cancer was more frequent in the cases (P 5 .0006). The mean times of pectoral and axillary drain removal were 11.68 days and 19.72 days, respectively, in cases and 12.36 and 14.38 days, respectively, in controls. Although the axillary drains were kept in place longer in the infected group, no statistically significant differences were observed when compared with controls. Stepwise logistic regression analysis found associations between SSI and age .50 years (OR 5 3.7 years; 95% CI 5 1.5 to 9.2; P 5 .005), preoperative concomitant chemoradiation treatment (OR 5 2.8; 95% CI 5 1.4 to 5.8; P 5 .0004), and any blood glucose values $ 150 mg/dL (OR 5 2.9; 95% CI 5 1.2 to 6.2; P 5 .02). Other wound-related complications were also monitored by the prospective surveillance program; flap necrosis and dehiscence occurred in 13.4% and 14.2% of patients, respectively. Hyperglycemia did not correlate with increased risk of developing dehiscence or flap necrosis in any case.

DISCUSSION Hyperglycemia is frequently encountered during hospitalization, and several studies have noted an

Vilar-Compte et al

association between hyperglycemia and poor hospital outcomes.8-13 In cardiac surgery, postoperative blood glucose levels are a significant predictor of infection, and studies on patients in ICUs have also exhibited similar results;10,12,13,15-17 these observations are not confined to patients with diabetes. Several observational studies on patients undergoing cardiovascular surgery have identified postoperative hyperglycemia as an independent risk factor for developing infections in patients with and without a history of diabetes and have correlated the risk of infection with the degree of glucose elevation.12,16,17 Intraoperative hyperglycemia is associated with an increased risk of adverse postoperative events, including death, after accounting for postoperative glucose levels.16 Our study indicates that in patients undergoing mastectomy, elevated blood glucose values during surgery and/or the immediate postoperative period correlate with increased risk of SSI (Fig 1). In this cohort, any perioperative glucose value $ 150 mg/dL increased the risk of SSI (OR 5 2.9; 95% CI 5 1.2 to 6.8; P 5 .02), and, as reported by Gandhi et al,16 the likelihood of experiencing an event increases with increasing glucose values (Fig 2). It is noteworthy that values . 175 mg/dL were more weakly associated with SSI, possibly related to the small number of patients in this stratum. In the patients studied herein (both cases and controls), mean preoperative and postoperative glucose values were lower compared with all patients undergoing cardiovascular surgery, as was the prevalence of diabetes.16,17 Although diabetes prevalence was lower, we postulate that acute response to stress was diminished in these patients, possibly explained by the fact that . 50% of these patients had received chemotherapy or concomitant chemoradiation before surgery. The frequency of diabetes in this group of patients was consistent with the prevalence reported in the general female Mexican population.18 Most of the risk factors for SSI development observed in this and other reports are not alterable; gender, race, obesity, neoadjuvant chemotherapy, and even duration of surgery are fixed obstacles to overcome.7,17,19-21 Hyperglycemia in the perioperative period is distinctly unique in this regard, and may be possible to control as long as it is identified. In this cohort, older age and preoperative chemoradiation therapy were strong risk factors for SSI, as was hyperglycemia $ 150 mg/dL. The mechanisms by which acute hyperglycemia increases the risk of postoperative infection are unclear. In vitro studies have shown that patients with diabetes mellitus may have an impaired immune response, and that abnormalities in immune function can partially be reversed by better glucose control.17,21-23 It also has been demonstrated that the stress of surgery and anesthesia results in increased secretion

Vilar-Compte et al

April 2008

195

Table 1. Patient characteristics Variable Age in years (n 5 256) BMI (n 5 242) Smoking (n 5 260) Diabetes (n 5 260) High blood pressure (n 5 260) Previous treatment (n 5 260) None Chemotherapy Chemoradiation Type of surgery (n 5 260) Radical mastectomy Modified radical mastectomy Other Duration of surgery (min) (n 5 258) Bleeding (mL) (n 5 249) Immediate breast reconstruction (n 5 260) Perioperative antimicrobial prophylaxis (n 5 229) Appropriate antibiotic selection Timeliness of perioperative antimicrobial prophylaxis

Cases (n 5 50)

Controls (n 5 210)

54.91 6 11.36 29.01 6 5.54 3 (6.0%) 13 (26%) 13 (26%)

47.85 6 12.73 27.7 6 5.24 11 (5.2%) 30 (14%) 37 (17%)

9 (18%) 6 (12%) 35 (70%)

103 (49%) 53 (25%) 54 (25%)

10 (20%) 36 (72%) 4 (8%) 165 6 65.5

8 (4%) 127 (60%) 75 (36%) 156.1 6 95.6

274.5 6 144.7 5 (10%)

246.4 6 189.1 21 (10%)

NS NS

Received PAP (n 5 47)

Received PAP (n 5 182)

NS

43 (91%) 30 (70%)

163 (90%) 131 (72%)

NS

P value .002 NS NS .04 NS

, .0001

, .0001 NS

Continuous variables are expressed as mean 6 standard deviation; nominal data are expressed as number and percentage.

Fig 1. Mean blood glucose in cases and controls. of counterregulatory hormones (catecholamines, cortisol, glucagon, and growth hormone) and excessive release of inflammatory cytokines, leading to various alterations in carbohydrate metabolism, including insulin resistance, increased hepatic glucose production, impaired peripheral glucose utilization, and relative insulin deficiency.24,25 Several studies have observed an increased risk of SSIs in patients with diabetes undergoing surgery.16,17,26-29 In the group of patients studied, diabetes did not correlate with SSI, although the prevalence of diabetes was higher in the cases compared with controls (26% vs 14.3%; P 5 NS) (Table 2). The explanation for this finding is difficult to establish but may be related to the low number of cases with diabetes. It is

Fig 2. Risk of SSI by stratified glucose values. noteworthy that only 1 patient (0.38%) had evidence of previously undiagnosed diabetes, an extremely low number compared with other reports; nevertheless, we were unable to measure glycosylated hemoglobin. Determining whether or not hyperglycemia has a causal association with development of SSI is tempting, but only a few clinical trials in ICUs have demonstrated a benefit of aggressive glycemic control in acutely ill patients.10,13,30 In cardiothoracic surgery patients, several observational studies have reported a positive relationship between hyperglycemia and SSI.11-13,16,17,20

196

Vilar-Compte et al

Vol. 36 No. 3

Table 2. Risk factor analysis for SSI according to variables studied Variable Age .50 years (n 5 256) Overweight (BMI .27.3) (n 5 242) Locally advanced breast cancer (n 5 256) Diabetes (n 5 260) High blood pressure (n 5 260) Previous chemoradiation (n 5 260) Radical mastectomy versus other mastectomies (n 5 260) Any glucose value $ 150 mg/dL (n 5 260)

Cases, n (%)

Controls, n (%)

Crude OR (95% CI)

Adjusted OR (95% CI)

33 (66.0) 37 (74.0) 37 (74.0) 13 (26.0) 13 (26.0) 35 (70.0) 10 (20.0)

81 (38.0) 137 (65.2) 97 (46.1) 30 (14.3) 37 (17.6) 54 (25.7) 7 (3.3)

2.9 (1.4 to 5.7) 1.45 (0.75 to 2.8) 4.52 (1.70 to 12.7) 2.1 (1.0 to 4.7) 1.64 (0.75 to 3.6) 7.4 (3.1 to 18.0) 16 (4.39 to 60.9)

3.7 (1.5 to 9.2) 1.7 (0.9 to 2.9) 0.92 (0.35 to 2.7) 1.1 (0.34 to 3.0) 0.61 (0.2 to 1.8) 2.8 (1.4 to 5.8) 1.0 (0.99 to 1.01)

35 (70.0)

91 (43.3)

3.1 (1.5 to 6.3)

2.9 (1.2 to 6.2)

Based on Hill’s criteria of causation,31 we believe that hyperglycemia and SSI fulfill the most important causation premises. Our study and other observational studies11-13,16,17,20 have demonstrated strength of association, a temporal relationship, and consistency. The dose-response relationship is always desirable when analyzing causation; although the risk of developing an SSI increased with increasing glucose values, the trend toward a greater risk of infection decreased at a level of 175 mg/dL. This fact is difficult to explain but may be related to the low number of patients in this strata. In a future study, we should be able to prove this premise. For some epidemiologists, experimental evidence provides a strong criterion when studying causation, and it could be argued that this is missing when studying hyperglycemia and SSI. Notwithstanding this shortcoming, however, Furnary et al15 have demonstrated a strong benefit from attaining euglycemia in patients undergoing cardiothoracic surgery; extrapolation from experimental models and studies in ICUs also support these findings.22,23,31,32 The American Diabetes Association currently recommends blood glucose targets of 80 to 111 mg/dL for patients in the ICU and a preprandial glucose level of 90 to 130 mg/dL (midpoint, 110 mg/dL) and a postprandial or random glucose levels , 180 mg/dL in non–critical care settings.33 Given all of this evidence, attempting to achieve low glucose values in patients undergoing surgery appears to be worthwhile. The benefits of these preventive measures are attractive and should be taken seriously by all health care providers. The National Surgical Prevention Program in support of the National Surgical Infection Prevention Project implemented by the CDC and Medicaid Services recommends monitoring and controlling hyperglycemia in surgical patients to improve surgical outcomes and maintain low rates of SSI.34 Although our study was an observational study, and as such is prone to more frequent bias and confounding, selection bias was probably low, because cases and controls were selected from the same population and studied and managed under the same care protocol, and the nested design always diminishes the risk

of information bias. Our findings, along with the current body of knowledge, indicate that control of hyperglycemia during the perioperative period is a suitable preventive measure to achieve lower SSI rates. In this series, concomitant preoperative chemoradiation (OR 5 2.8; 95% CI 5 1.4 to 5.8; P 5 .004) and age . 50 years (OR 5 3.7; 95% CI 5 1.5 to 9.2; P 5 .005) were also risk factors for developing an infection. These findings are consistent with our previous reports.7,35 Preoperative chemoradiation has been shown to increase the risk of infection and other wound complications.35-38 Radiation therapy has been associated with increased fibrosis and decreased vascularity, which can lead to tissue necrosis and ulceration. In our patients, this preoperative treatment was common, because 52.3% had locally advanced breast cancer. Besides radiation, these patients often needed more ample resections for cancer control, which also seems to increase the risk of postoperative wound complications. Advanced age has consistently been found to be a risk factor for SSIs. In our patients, age . 50 years was associated with increased risk of SSI, in contrast to other reports citing age . 60 years as a risk factor.1,39 A possible explanation for this finding could be related to an increased prevalence of diabetes and hypertension in patients age . 50 years. Although in our study, both of these disorders were more frequent in the cases than in the controls, neither was independently associated with SSI (Table 2). Perioperative antibiotic prophylaxis was used in 229 of the 260 patients (88.0%) in the present study. Although PAP is recommended in breast surgery, 2 clinical trials have shown no benefit of PAP in this type of surgery.40,41 In our guidelines, cefuroxime and cephalotin are accepted prophylactic antibiotics for patients undergoing mastectomy; not prescribing PAP is permitted in this group of patients. In our experience with more than 2500 breast surgeries, we have found no difference in SSI rates or other important outcomes associated with PAP (data not published). As shown in Table 1, the correct selection of PAP and the timeliness of prophylaxis were very similar in the cases and controls.

Vilar-Compte et al

In this study, as in others reported by our group,7,35,42 the prevalence of isolated Gram-negative bacteria was slightly higher than that of Staphylococci, which is unusual in patients undergoing breast cancer surgery. This finding could be explained by the fact that most of our patients have the drains in situ for long periods (10 to 14 days), develop the infection during the second or third postoperative weeks, and have more flap necrosis; some of them do not even have running water at home. In our institution, E. coli has been the most common bacteria isolated from wounds over the last 15 years. The results previously reported for patients with locally advanced breast cancer are probably a common scenario for this group of cancer patients. Although in developed countries, patients with locally advanced breast cancer are not very common, in many countries in Latin America, Asia, and Africa, where the prevalence of breast cancer is increasing, its is being detected more often at advanced stages. In patients with advanced breast cancer, the rate of surgical complications is higher, and the isolated bacteria may differ from those traditionally found in breast surgery. In conclusion, our data indicate that elevated blood glucose value is an independent risk factor for postoperative SSI in patients undergoing mastectomy. As reported previously, preoperative concomitant chemoradiation and age . 50 years also were risk factors for developing an SSI in these patients. Perioperative hyperglycemia is a modifiable risk factor, and thus represents a potential target for reducing the risk of SSI in breast cancer surgery. Given the current scenario of increasing prevalence of overweight, diabetes, and carbohydrate intolerance along with other comorbid conditions in daily medical practice, strict glycemic control during the perioperative period should be taken seriously. References 1. Cruse PJE, Foord R. The epidemiology of wound infection: a 10-year prospective study of 62,939 wounds. Surg Clin North Am 1980;60: 27-40. 2. Gaynes RP, Culver DH, Horan TC, Edwards JR, Richards C, Tolson JS. National Nosocomial Infections Surveillance System. Surgical site infection (SSI) rates in the United States, 1992–1998: The National Nosocomial Infections Surveillance System Basic SSI Risk Index. Clin Infect Dis 2001;33(Suppl 2):S69-76. 3. Barber GR, Miransky J, Brown AE, Coit DG, Lewis FM, Thaler HT, et al. Direct observations of surgical wound infections at a comprehensive cancer center. Arch Surg 1995;130:1042-7. 4. Broadwater JR, Edwards MJ, Klugen C, Hortobagyi GN, Ames FC, Balch CM. Mastectomy following preoperative chemotherapy: strict operative criteria control operative morbidity. Ann Surg 1991;213:126-9. 5. Canavase G, Catturich A, Vecchio C, Gipponi M, Tomei D, Sertoli MR, et al. Surgical complications related to peri-operative adjuvant chemotherapy in breast cancer: results of a prospective, controlled randomized clinical trial. Eur J Surg Oncol 1997;23:10-2.

April 2008

197

6. Furey PC, Macgillivray DC, Castiglione CL, Allen L. Wound complications in patients receiving adjuvant chemotherapy after mastectomy and immediate breast reconstruction for breast cancer. J Surg Oncol 1994;55:194-7. 7. Vilar-Compte D, J.B., Robles-Vidal C, Volkow P. Surgical site infections in breast surgery: case-control study. World J Surg 2004;28:242-6. 8. Krinsley J. Perioperative glucose control. Curr Opin Anaesthesiol 2006;19:111-6. 9. Umpierrez G, Maynard G. Glycemic chaos (not glycemic control) still the rule for the inpatient care. J Hosp Med 2006;3(3):141-3. 10. Collier B, Dı´az J Jr, Forbes R, Morris J Jr, May A, Guy J, et al. The impact of a normoglycemic management protocol on clinical outcomes in the trauma intensive care unit. JPEN J Parenter Enteral Nutr 2005;29:353-8. 11. Furnary AP, Wu Y, Bookin SO. Effect of hyperglycemia and continuous intravenous insulin infusions on outcomes of cardiac surgical procedures: the Portland Diabetic Project. Endocr Pract 2004;10(Suppl 2):21-33. 12. Latham R, Lancaster AD, Covington JE, Pirolo JS, Thomas CS Jr. The association of diabetes and glucose control with surgical-site infections among cardiothoracic surgery patients. Infect Control Hosp Epidemiol 2001;22:607-12. 13. Grey NJ, Perdrizet GA. Reduction of nosocomial infections in the surgical intensive-care unit by strict glycemic control. Endocr Pract 2004; 10(Suppl 2):46-52. 14. Horan TC, Gaynes RP, Martone WJ, William RJ, Emori TG. CDC definitions of nosocomial surgical site infections, 1992: a modification of CDC definitions of surgical wound infections. Infect Control Hosp Epidemiol 1992;13:606-8. 15. Furnary AP, Wu Y. Clinical effects of hyperglycemia in the cardiac surgery population: the Portland Diabetic Project. Endocr Pract 2006; 12(Suppl 3):22-6. 16. Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, Williams BA, et al. Intraoperative hyperglycemia and perioperative outcomes in cardiac surgery patients. Mayo Clin Proc 2005;80:862-6. 17. Viesendorp TM, Morelis QJ, DeVries JH, Legemate DA, Hoekstra JBL. Early postoperative glucose levels are an independent risk factor for infection after peripheral vascular surgery: a retrospective study. Eur J Vasc Endovasc Surg 2004;28:520-5. 18. Encuesta Nacional de Salud. 2000. La salud de los adultos (2). Instituto Nacional de Salud Pu´blica, Secrtaria de Salud, Me´xico. Available at http://www.insp.mx/ensa/ensa_tomo2.pdf. Accessed January 10, 2007. 19. Ruvalcaba-Limo´n E, Robles-Vidal C, Poitevin-Chaco´n A, Cha´vez-MacGregor M, Gamboa-Vignolle C, Vilar-Compte D. Wound complications after breast cancer surgery in patients treated with pre-operative concomitant chemo-radiation: a case-control analysis. Breast Cancer Res Treat 2006;95:147-52. ´ lvarez C, Revuelta JM, Gonza´20. Ferna´ndez-Ayala M, Nan DN, Farinas-A lez-Macı´as J, Farinas MC. Surgical site infection during hospitalization and after discharge in patients who have undergone cardiac surgery. Infect Control Hosp Epidemiol 2006;27:85-8. 21. Geerlings SE, Hopelman AI. Immune dysfunction in patients with diabetes mellitus (DM). FEMS Immunol Med Microbiol 1999;26:259-65. 22. Weeker F, Giulietti AP, Machalaki M, Coopmans W, Van Herck E, Mathieu C, et al. Metabolic, endocrine, and immune effects of stress hyperglycemia in rabbits model of prolonged critical illness. Endocrinology 2003;144:5329-38. 23. Esposito K, Nappo F, Marfella R, Giugliano G, Giugliano F, Ciotola M, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation 2002;106:2067-72. 24. Smiley DD, Umpierrez G. Perioperative glucose control in the diabetic or nondiabetic patient. Southern Med J 2006;99:580-7. 25. Umpierrez GE, Kitabchi AE. ICU care for patients with diabetes. Curr Opin Endocrinol Diabetes 2004;11:75-81. 26. Arabshahi KS, Koohpayezade J. Investigation of risk factors for surgical wound infection among teaching hospitals in Teheran. Int Wound J 2006;3:59-62.

198

Vol. 36 No. 3

27. Fiorio M, Marvaso A, Vigano F, Marchetti F. Incidence of surgical site infections in general surgery in Italy. Infection 2006;34:310-4. 28. Erman T, Demirhindi H, Gocer AI, Tuna M, Ildan F, Boyar B. Risk factors for surgical site infections in neurosurgery patients with antibiotic prophylaxis. Surg Neurol 2005;63:107-12. 29. Vilar-Compte D, Mohar A, Sandoval S, De la Rosa M, Gordillo P, Volkow P. Surgical site infections at the National Cancer Institute in Mexico: a case-control study. Am J Infect Control 2000;28:14-20. 30. Van den Berghe G, Wouters P, Weekers F, Vewaest C, Bruynincks M, Shaetz M, et al. Intensive insulin therapy in critically ill patients. N Engl J Med 2001;345:1352-67. 31. Hill AB. The environment and disease: association or causation? Proc R Soc Med 1965;58:295-300. 32. Krinsley JS. Effects of an intensive glucose management protocol on the mortality of critically ill adult patients. Mayo Clin Proc 2004;79:992-1000. 33. Garber AJ, Moghissi ES, Bransome ED Jr. American College of Endocrinology Position Statement on Inpatient Diabetes and Metabolic Control. Endocrin Pract 2004;10:77-82. 34. Dellinger EP, Hausmann SM, Bratzler DW, Johnson RM, Daniel DM, Bunt KM, et al. Hospitals collaborate to decrease surgical site infections. Am J Surg 2005;190:16-7. 35. Ruvalcaba-Limo´n E, Robles-Vidal C, Poitevin-Chaco´n A, Cha´vez-MacGregor M, Gamboa-Vignolle C, Vilar-Compte D. Complications alter breast cancer surgery in patients treated with concomitant preoperative chemoradiation: a case-control analysis. Breast Cancer Res Treat 2006;95:147-52.

Vilar-Compte et al 36. Badr el Din A, Coibion M, Guenier CP, Nogaret JM, Lorent I, Van Houtte P, et al. Local postoperative morbidity following pre-operative irradiation in locally advanced breast cancer. Eur J Surg Oncol 1989;15: 486-9. 37. Sauter ER, Eisenberg BL, Hoffman JP, Ottery FD, Borass MC, Goldstein LJ, et al. Postmastectomy morbidity after combination preoperative irradiation and chemotherapy for locally advanced breast cancer. World J Surg 1993;17:237-41. 38. Huang E, McNeese MD, Strom EA, Perkins GH, Katz A, Hortobagyi GN, et al. Locoregional treatment outcomes for inoperable anthracycline-resistant breast cancer. Int J Radiat Oncol Biol Phys 2000;53: 1225-33. 39. Barber GR, Miransky J, Brown AE, Coit DG, Lewis FM, Thaler HT, et al. Direct observations of surgical wound infections at a comprehensive cancer center. Arch Surg 1995;130:1042-7. 40. Wagman LD, Tegmeier B, Beatty JD, Kloth DD, Kokal WA, Riihimaki DU, et al. A prospective, randomized double-blind study of the use of antibiotics at the time of mastectomy. Surg Gynecol Obstet 1990; 170:12-6. 41. Gupta R, Sinnett D, Carpenter R, Preece PE, Royle GT. Antibiotic prophylaxis for post-operative wound infection in clean elective breast surgery. Eur J Surg Oncol 2000;26:363-6. 42. Vilar-Compte D, Rolda´n-Marı´n R, Robles-Vidal C, Volkow P. Surgical site infection (SSI) rates among patients who underwent mastectomy after the introduction of SSI prevention policies. Infect Control Hosp Epidemiol 2006;27:829-34.