We performed a retrospective cohort study to evaluate risk factors for SSIs. We reviewed the electronic medical records of patients diagnosed with HNC from January 2006 to June 2010 at Hallym University Kangdong Sacred Heart Hospital, a 700-bed secondary-level university hospital at which the IIsong Head-and-Neck Cancer Center is located. Patients diagnosed with HNC underwent radical neck dissection involving the upper digestive or respiratory tract were included. In addition, only patients aged > 20 years who were undergoing their first operations were included. Patients undergoing thyroid gland surgery with or without lymph node dissection were excluded. All operations were performed by surgeons with more than 5 years of experience in head and neck major oncological surgery or reconstruction. All patients included in the study received prophylactic antibiotics, and all surgical sites were disinfected with povidone iodine before incision. All patients received routine postoperative care, and surgeons or infectious disease specialists diagnosed SSIs. The data collected included age, sex, history of smoking and alcohol consumption, underlying disease, primary site of tumor and TNM stage, ASA score, National Nosocomial Infections Surveillance System (NNIS) risk index, preoperative albumin level (mg/mL), BMI, blood loss during surgery (mL), blood transfusion during surgery, operation time, preoperative chemotherapy, preoperative radiotherapy, duration of hospitalization before operation, causative microorganisms, and prophylactic and therapeutic antibiotics. Severity of comorbid conditions was calculated using the Charlson comorbidity index [18]. This study was approved by the Institutional Review Board of Hallym University Kangdong Sacred Heart Hospital (Seoul, South Korea) as required by the local hospital policy at the time of the study.

According to the Centers for Disease Control and Prevention’s NNIS and the criteria laid out by Horan et al. [19] and Johnson et al. [20], a SSI was defined by the occurrence of at least one of the following within 30 days of surgery: purulent drainage from the incision, spontaneous dehiscence or deliberate opening of the incisions with signs or symptoms of infection (pain, tenderness, localized swelling, redness, or heat), an isolated organism from the incision, purulent discharge from drainage, or an abscess without evidence of clinical anastomotic leakage. According to the guidelines, SSIs are classified as being either incisional or organ/space [21]. Incisional SSIs are further divided into those involving only skin and subcutaneous tissue (superficial incisional SSI) and those involving deeper soft tissue of the incision (deep incisional SSI). Organ/space SSIs involve any part of the anatomy other than incised body wall layers that were opened or manipulated during an operation.

Microorganisms were identified by the VITEK 2 automated system (bioMerieux Inc., Hazelwood, MO, USA) in a microbiology laboratory. Antimicrobial susceptibility testing was performed using the GNI (gram-negative identification) card GPI (gram-positive identification) card in a VITEK 2 automated system according to the recommendations of the Clinical and Laboratory Standards Institute.

The chi-square or Fisher exact tests were used to compare categorical variables. Continuous variables were compared using the Student t test or Mann-Whitney test. To identify independent risk factors for infection, a stepwise forward logistic regression analysis was used to control for the effects of confounding variables. All risk factors with a p < 0.05 at the univariate level were included in the multivariate logistic regression.

All variables with a value of p < 0.05 were retained in the final model. Interactions between variables were not introduced into the models. Odds ratios and their 95% confidence intervals were calculated. All tests of significance were two-tailed, and those with a p value of < 0.05 were considered significant. SPSS version 18 (SPSS Inc., Chicago, IL, USA) was used for these analyses.

During the study period, a total of 370 patients with HNC were included in this study. Out of the total of 370 patients who underwent major HNC surgery, 73 (19.7%) had SSIs. The vast majority of SSIs were in the form of deep incisional SSIs (40/73, 54.8%), followed by superficial incisional SSIs (18/73, 24.7%) and organ/space SSIs (15/73, 20.5%). The median time from surgery to diagnosis of SSI was 12 days (range, 1 to 29). Of the 73 SSIs, 52 SSIs (71.2%) developed within 14 days.

The mean age (± standard deviation) of the study population was 59.02 ± 13.09 years, and 290 patients (78.4%) were male (Table 1). The most common underlying disease was cardiovascular disease (35.4%, 131/370). The most common primary site of tumors was the hypopharynx (179 cases, 48.4%), followed by the oral cavity (114 cases, 30.8%), salivary gland (34 cases, 9.2%), nasopharynx/sinus (29 cases, 7.8%), neck (4 cases, 1.1%), and other (4 cases, 1.1%). The median duration of hospital stay was longer in patients with SSIs than in those without SSIs (78.0 [interquartile range (IQR), 40 to 119] vs. 41.0 [IQR, 24.0 to 87.5], p < 0.001). Table 1 shows the demographic characteristics of the patients and the incidence of SSIs. SSIs were found to be more frequent in men than in women (p = 0.001), and patients who were older (p = 0.019), smokers (p = 0.005), or consumers of alcohol (p = 0.041) were more likely to develop SSIs than their counterparts.

The association of SSIs with preoperative variables is summarized in Table 1. Univariate analysis showed that cardiovascular disease (p = 0.006) and neurologic disease (p = 0.022) were associated with SSIs.

Table 2 summarizes the association of SSIs with perioperative variables. In the univariate analysis, an ASA score ≥ 3 (p = 0.005), NNIS risk index greater than 2 (p < 0.001), and operation time ≥ 6 hours (p = 0.007) were associated with SSIs. The amount of bleeding was greater in patients with SSIs than in those without SSIs (900 mL [IQR, 700 to 1,200] vs. 700 mL [IQR, 500 to 1,000], p < 0.001). We determined a cutoff point for developing SSIs. Using a receiver operating characteristic curve, we determined blood loss during the operation of more than 560 mL as the optimal cutoff point. Moreover, the amount of blood transfusion was higher in patients with SSIs than without SSIs, but this finding was without statistical significance (p = 0.051).

Multivariate analysis revealed that being male, having a long operation time (over 6 hours), underlying cardiovascular disease, and blood loss during the operation of more than 560 mL were independent risk factors for SSIs (Table 3).

A microbiological analysis was performed in all 73 patients with SSIs, and 60 patients showed positive culture results (Table 4). Twenty-nine patients had polymicrobial infections (29/60, 48.3%). A total of 92 isolates from 60 patients with SSIs included 47 gram-positive organisms (51.5%), 41 gram-negative organisms (43.5%), and 5 Candida species (5.4%). The most frequently isolated pathogen was Staphylococcus aureus (30/92, 32.6%), followed by Klebsiella pneumoniae (13/92, 14.1%), Pseudomonas aeruginosa (11/92, 12.0%), and Enterococcus species (11/92, 2.0%). Twenty-eight of the S. aureus isolates were methicillin-resistant S. aureus (MRSA) (93.3%). The resistance rates to cefotaxime and ciprofloxacin among the K. pneumoniae isolates were 30.8% (4/13) and 53.8% (7/13), respectively. The most common organisms associated with superficial incisional and deep incisional SSIs were S. aureus, followed by K. pneumoniae and P. aeruginosa. The distribution of pathogens showed a similar pattern in the two types of SSIs. S. aureus was the most common pathogen associated with organ/space SSIs, as expected. Enterobacter species were more commonly associated with organ/space SSIs (4/8, 50.0%) than with incisional SSIs.

Among the patients with SSIs, nine patients (12.33%) had bacteremia secondary to SSI. Most patients (82.2%, 60/73) improved within 7 days. However, 25 patients (34.3%) required surgical drainage, and one patient died as a result of complications of SSI. Vancomycin and piperacillin/tazobactam were widely used. Among the patients with SSIs, 29 patients (39.7%) received combination therapy with vancomycin plus piperacillin/tazobactam.