Postoperative antibiotic prophylaxis in clean-contaminated head and neck oncologic surgery: a retrospective cohort study

Transcription

1 DOI /s HEAD AND NECK Postoperative antibiotic prophylaxis in clean-contaminated head and neck oncologic surgery: a retrospective cohort study C.-J. Busch 1 R. Knecht 1 A. Münscher 1 J. Matern 1 C. Dalchow 1 B. B. Lörincz 1 Received: 30 June 2015 / Accepted: 7 December 2015 Springer-Verlag Berlin Heidelberg 2015 Abstract Antibiotic prophylaxis is commonly used in head and neck oncologic surgery, due to the clean-contaminated nature of these procedures. There is a wide variety in the use of prophylactic antibiotics regarding the duration of application and the choice of agent. The purpose of this study was to determine whether short-term or long-term antibiotic prophylaxis has an impact on the development of head and neck surgical wound infection (SWI). Retrospective chart review was carried out in 418 clean-contaminated head and neck surgical oncology cases at our department. More than 50 variables including tumour type and stage, type of surgical treatment, comorbidities, duration and choice of antibiotic prophylaxis, and the incidence of SWI were analysed. Following descriptive data analysis, Chi square test by Pearson and Fisher s exact test were used for statistical evaluation. Fifty-eight of the 418 patients (13.9 %) developed SWI. Patients with advanced disease and tracheotomy showed a & C.-J. Busch cjbusch@uke.de R. Knecht r.knecht@uke.de A. Münscher a.muenscher@uke.de J. Matern julia.matern@gmail.com C. Dalchow c.dalchow@uke.de B. B. Lörincz b.loerincz@uke.de 1 Department of Otorhinolaryngology, Head and Neck Surgery and Oncology, University Medical Center Hamburg- Eppendorf, Martinistr. 52, Hamburg, Germany significantly higher rate of SWI than those with early stage disease and without tracheotomy (p = and p = , respectively). However, there was no significant difference between the SWI rates in the short term and long term treatment groups (14.6 and 13.2 %, respectively; p = 0.689). Diabetes and body weight were not found to be risk factors for SWI. Long-term antibiotic prophylaxis was not associated with a decrease in SWI in the entire cohort of patients undergoing clean-contaminated major head and neck oncologic surgery. Our data confirmed the extent of surgery and tracheotomy as being risk factors for postoperative SWI. Keywords Postoperative antibiotic prophylaxis Head and neck squamous cell carcinoma (HNSCC) Head and neck oncologic surgery Clean-contaminated surgery Surgical wound infection (SWI) Introduction Surgical wound infection (SWI) is one of the most common postoperative complications. It may affect hospitalisation time and possibly results in additional interventions. Despite the advances in surgical and anaesthesia procedures and improvements in perioperative care, such adverse events, compromising the outcomes of patients undergoing head and neck surgery, are well known. After urinary tract infection (23 40 %) and respiratory complication (17 23 %), surgical wound infection is the third most common type of nosocomial infections [1, 2]. Patients undergoing major head and neck oncologic surgery are at great risk of developing complications since they often have significant co-morbidities, resulting in increased anaesthetic and surgical risks [3]. Additionally,

2 there is little doubt that most head and neck procedures include clean-contaminated surgical sites, exposed to bacterial contamination through the communicating mucosal barriers and increasing the risk of wound infection [4]. Prophylactic antibiotics are commonly used in patients undergoing head and neck surgery [5]. However, the choice of the specific antimicrobial agents and the duration of the treatment are controversial and the incidence of the reported wound infections (40 63 %) is still high [6, 7]. Currently, there is no scientific evidence for the use of prophylactic antibiotics in clean or clean-contaminated head and neck surgery [8]. The development of a SWI often results in prolonged hospital stay, additional surgical interventions, delayed adjuvant treatment and, altogether, in a higher cumulative burden for the patient as well as for the health care system. A number of patient-related and treatment-related risk factors have been identified, possibly leading to SWI (Table 1). Performing a neck dissection in head and neck cancer patients can also increase the risk of SWI [9, 10]. The reasons for this may be the more significant tissue trauma, blood loss, number of surgical drains and longer operation time [3]. SWIs in head and neck cancer patients typically show multi-microbial colonisation [11]. Usually mainly Grampositive, but also Gram-negative aerobic bacteria cultures can be isolated in % of the cases [4, 11, 12]. To reduce the rate of SWI, pre-, peri- and postoperative measures are to be combined [13]. Besides reduction of patient-related risk factors, keeping bacterial contamination of the surgical site to a minimum is the most important prophylaxis. It is assumed that with an optimised preoperative surgical setting and preparation of the patient, SWI rates can be reduced significantly [14, 15]. The aim of this study was to assess the value of the current use of postoperative prophylactic antibiotics in our Table 1 Risk factors of surgical wound infections, according to Mangram et al. [33] Patient-related factors Advanced age Malnutrition Obesity Smoking Diabetes Colonization with S. aureus Coexistent infections at a remote body site Long preoperative hospital stay Altered immune response Previous radiotherapy Procedure-related factors Prolonged operation time Inadequate skin antisepsis Surgical drains Surgical technique Tissue trauma Operating room ventilation Insufficient antimicrobial prophylaxis Foreign material in surgical site Re-operation Duration of surgical scrub department. On this large cohort of surgically treated head and neck cancer patients, we compared the short-term versus long-term administration of antibiotic substances as well as assessed the role of risk factors leading to wound infections. Results of this study may lead to further prospective studies investigating the outcomes and the cost effectiveness of different prophylactic antibiotic treatment regimens. The role of antibiotic prophylaxis Antibiotic prophylaxis is usually recommended in two scenarios: either in case of increased risk for SWI, e.g. clean-contaminated or contaminated surgical site, or if the potential development of SWI would lead to an disproportionately significant loss for the patient, e.g. to removal of the implanted foreign material or to free flap failure [16, 17]. Starting time and duration of the antibiotic treatment are to be considered carefully, along with the right choice of the antimicrobial agent [18]. The standard procedure is to administer prophylactic antibiotics within 2 h prior to surgery [19]. The frequency of the antimicrobial prophylaxis should correspond to the time interval that ensures adequate concentrations at the incision site during the period of potential infection, keeping the risk of adverse effects, development of resistance and the associated costs at the minimum in the same time [20]. Inappropriate use of postoperative antibiotics still occurs in the daily routine. Personal decisions of the surgeons are not always based on established guidelines to meet safety criteria. A common problem in this setting is the use of prophylactic antibiotics for a longer period of time than recommended [21]. Patients and methods A retrospective chart review of all patients having undergone clean-contaminated major head and neck oncologic surgery of the upper aerodigestive tract in the past 4 years at our tertiary academic center was carried out. A total of 418 patients were included in this study. Most cases were primary surgeries. In 41 of the 418 patients (9.8 %), locoregional recurrences were treated. Patients with salvage surgery were also included: 35 of the 418 patients (8.4 %) received radiotherapy and 32 of the 418 patients (7.7 %) received chemotherapy prior to their surgery. More than 50 variables such as tumour type and stage, type of surgical treatment (i.e. partial or total laryngectomy, partial or total pharyngectomy, tracheotomy, oral cavity resection, etc.), comorbidities, preoperative radiotherapy and/or chemotherapy, body weight/bmi, duration

3 and type of antibiotic prophylaxis, and the incidence of surgical wound infection (SWI) were collected and analysed. SWI was defined according to the North American Centers for Disease Control (CDC) classification upon presenting with wound dehiscence, purulent drainage, abscess formation, or cellulitis [22]. Features such as erythema, skin induration in the wound and/or contact sensitivity alone were not accepted as SWI. Fistula formation was recorded in case of obvious clinical observation or radiological detection. Pneumonia was registered after a positive chest X-ray and/or microbiology culture. As for the cut-off point in the duration of the antibiotic treatment, we have chosen B7 or[7 days of antibiotic prophylaxis. Patients who received 7 or fewer days of prophylactic treatment were allocated to the short-term group, while patients with at least 8 days of treatment or more, were allocated to the long-term group. Subgroup analysis regarding diabetes, body mass index (BMI, classification according to the WHO, 2004), and further postoperative complications were also performed. Due to the retrospective character of this study, a potential selection bias should be taken into account concerning the antibiotic treatment time. Data collection and statistical analysis were performed using Microsoft Office Excel and SPSS (version 22) for Windows. Our analysis was restricted to individuals with a complete set of data on all variables required for a particular analysis. Following descriptive data analysis, we used the chi square test by Pearson as well as Fisher s exact test to statistically evaluate our data and to compare the two groups. Only p values below 0.05 were considered as statistically significant (values written bold in the tables). Multivariate analysis was performed with logistic regression. Patient characteristics A total of 418 patients were included in this study, with a mean age of 62 years and a 95 % standard deviation (±SD) of 12 years. There were 318 (76.1 %) male and 100 (23.9 %) female patients. Mean body mass index (BMI) ± SD was ± 4.4 kg/m 2 with approximately half of the patients presenting with normal weight (n = 208, 49.8 %), 40 patients (9.6 %) were underweight, and 132 patients (31.6 %) as well as 38 patients (9.1 %) showing preobesity and obesity, respectively. In this setting, the main indications for clean-contaminated major head and neck oncologic surgery was laryngeal cancer (n = 137, 32.8 %) and oropharyngeal cancer (n = 100, 23.9 %), followed by oral cavity cancer (n = 67, 16 %). Procedures for the treatment of laryngeal cancer included total laryngectomy (n = 40, 9.6 %), partial laryngectomy (n = 40, 9.6 %), and endolaryngeal chordectomy (n = 39, 9.3 %). In cases of the latter, only those patients undergoing neck dissection and/or tracheotomy were included in this study. More than half of the patients (n = 192, 53 %) had advanced disease with UICC Stage IV. Only eight patients (2.2 %) presented with distant metastasis. Thirty-five (8.4 %) patients received radiotherapy and 32 patients (7.7 %) received chemotherapy prior to their surgery. Altogether, 197 patients (47.1 %) received temporary tracheotomy either before (59 of the 197 patients, with no communication between the tracheotomy site and the main operative site) or during (138 of the 197 patients, with possible communication between the two sites) the oncologic resection or reconstruction. Antibiotics Since the given antibiotic prophylaxis was subject to individual decision of the surgeon in charge, the choice of antibiotic substances, their dose, their daily distribution and the duration of their administration showed very heterogenic patterns. Most commonly cefazolin, clindamycin, cefuroxime, ampicillin/sulbactam, and metronidazole were applied postoperatively. A subgroup of 269 patients (64.4 %) received a single antibiotic agent during the entire treatment period. Another 118 (28.2 %) and 31 patients (7.4 %) were given a second and third antibiotic substance over time, respectively. Results Surgical wound infection Altogether, 58 of our 418 patients (13.9 %) developed SWI. Patients with advanced stage disease had significantly more frequently SWI than those with early stage disease (p = 0.012). In the former group, 19.8 % of patients had Stage IV disease. Further, patients with tracheotomy showed a higher SWI rate than those without (p = ), irrespective of the timing of their tracheotomy (i.e., prior to or concurrently with their oncologic surgery) and of the possible communication between the tracheotomy site and the oncologic site. The rate of SWI in patients with tracheotomy was threefold higher than that in those without tracheotomy (19.3 vs. 6.2 %). There was no significant difference between the short-term and long-term treated groups with regards to SWI, as their infection rate was 14.6 and 13.2 %, respectively (p = 0.689). Other patient-specific risk factors for SWI were positive nodal stage, the extent of surgical resection and the

4 Table 2 The effect of patient-specific risk factors on SWI Risk factor No. of patients (n) No. of infected wounds (n) Rate of SWI (%) p value Age B65 years Age [65 years Underweight Normal weight Preobesity/obesity No diabetes Diabetes No preop. irradiation Preop. irradiation No preop. chemotherapy Preop. chemotherapy Early stage disease (St. I, II) Advanced disease (St. III, IV) N0 nodal stadium N-positive nodal stadium Minor surgery Major surgery No neck dissection Concurrent neck dissection No recurrence Recurrent disease No tracheotomy Tracheotomy Non-smoker Nicotine abuse No alcohol Alcohol abuse No liver cirrhosis Liver cirrhosis present No COPD COPD Bold values indicate statistical significance (p B 0.05) COPD chronic obstructive pulmonary disease Table 3 Multivariate analysis of selected risk factors for surgical wound infection Risk factor OR (95 % CI) p value N-positive nodal stadium 1.84 ( ) Tracheotomy 3.40 ( ) Neck dissection 1.15 ( ) Advanced stage disease 0.61 ( ) Major surgery 2.54 ( ) Bold values indicate statistical significance (p B 0.05) OR odds ratio, CI confidence interval addition of a neck dissection (Table 2). With these five significant risk factors, a multivariate analysis was performed, which verified the significance of tracheotomy and that of the extent of surgical resection (Table 3). Patients who received major surgery (e.g. operating time [3 h, laryngectomy, pharyngectomy, reconstructive surgery, etc.) and a tracheotomy had a and 3.4-fold higher risk, respectively, for developing SWI. Out of the 54 patients with diabetes, 7 (13 %) developed SWI. This rate was 14 % in the non-diabetic subpopulation (n = 51) of the same cohort. Within the diabetic group, further subgroups with short-term versus long-term antibiotic prophylaxis were compared to each other, with regards to their SWI rates. Thirty diabetic patients (55.6 %) were in the short-term subgroup, and 24 patients were (44.4 %) in the long-term subgroup. The rate of SWI in the former subgroup was 16.7 % (n = 5) versus 8.3 % (n = 2) in the latter. There was no significant difference between the number of patients with and without tracheotomy in the diabetic and in the non-diabetic patient groups. With a

5 Table 4 Body mass index (BMI) subgroups breakdown p value at 0.443, the difference was not significant, thus there was no reduction in the SWI rates after a longer period of antibiotic treatment in diabetic patients, compared to the short term treated diabetic subgroup. According to their BMI, the patients were divided into three subgroups (Table 4). There was no correlation between the patients weight and the risk of wound infection (p = 0.834). Regarding the length of antibiotic prophylaxis, there was also no significant difference in the rates of SWI between short term and long term treated patients among these three subgroups (all p [ 0.05). Further, there was no significant difference between patients with and without preoperative radiotherapy in their SWI rates. Fistula formation A pharyngocutaneous fistula was developed in 28 of the 418 patients (6.7 %), diagnosed after 10 ± 4 days postoperatively. The risk of fistula formation was higher in patients with tracheotomy (p = 0.004) and in patients having undergone major surgical procedures (p = 0.007; e.g. pharyngectomy, laryngectomy, free flap reconstruction, etc.). In a multivariate analysis, they were exposed to a and 2.49-fold higher risk of developing a fistula, respectively (Table 5). The risk of fistula formation did not correlate with preoperative radiotherapy, diabetes and BMI, nor with the duration of antibiotic prophylaxis. Pneumonia Underweight Normal weight Preobesity/obesity BMI B C25.00 n (%) 40 (9.6) 208 (49.8) 170 (40.6) Ø ± SD 16.9 ± ± ± 3.2 Median SD standard deviation In the entire cohort, a total of ten patients (2.4 %) developed pneumonia. Patients with tracheotomy (p = 0.04) and liver cirrhosis (p = 0.021) were exposed to a higher risk of pneumonia. All of these ten patients received only shortterm prophylactic antibiotic treatment. None of the patients in the long-term group suffered from pneumonia (p = ). Only one patient without tracheotomy developed pneumonia, the other nine pneumonia patients underwent tracheotomy as part of their oncologic head and neck procedure. The incidence of pneumonia was significantly higher in patients undergoing tracheotomy simultaneously with their oncologic surgery (p = 0.04). Table 5 Multivariate analysis of selected risk factors for fistula formation Risk factor OR (95 % CI) p value N-positive nodal 1.94 ( ) stadium Tracheotomy 3.35 ( ) Neck dissection 1.08 ( ) Advanced stage disease 0.55 ( ) Major surgery 2.49 ( ) Preop. chemotherapy 0.71 ( ) COPD 1.37 ( ) Poor dental status 1.19 ( ) Bold values indicate statistical significance (p B 0.05) OR odds ratio, CI confidence interval, COPD chronic obstructive pulmonary disease Discussion SWI rates vary a lot among different surgical interventions. Our retrospective study investigated the incidence of SWI in clean-contaminated head and neck cancer surgery, which often leads to higher SWI rates than sterile procedures do [23]. Without antibiotic prophylaxis, the incidence of SWI in clean-contaminated wounds varies between 24 and 87 % in the literature [24]. In our patient cohort, the overall SWI rate was 13.9 %, irrespective of the duration of the antibiotic treatment. The infection was diagnosed after 8.2 days (mean value) postoperatively. Comparing our results to similar studies in the literature, our SWI rate of 13.9 % is rather low. SWI rates are commonly reported between 20.1 and 50.5 % in the literature [3, 7, 25], with some exceptions at the low end between 3.1 and 11 % [5, 26]. Since study designs are very diverse, it is difficult to compare all these results with each other. Tumour stage is reported to have an impact on the risk of SWI. Several studies showed a positive correlation between advanced tumour stage and SWI in univariate analysis [3, 23, 25]. However, we did not find tumour stage to have a significant impact on SWI in our multivariate analysis, similarly to Liu et al. [24]. Advanced tumour stage often requires larger resections and prolonged surgical procedures. More extensive tissue trauma and larger wound surface may contribute to increased contamination, and consequently to a higher risk of surgical wound infection. Major procedures with large resections and prolonged operating times, often associated with a temporary tracheotomy performed during the same surgery, represent a significant independent risk factor for developing SWI. Robbins et al. [27] also demonstrated tracheotomy and longer operation time as risk factors for SWI. The infection rate of 8 % associated with procedures not longer than 2 h

6 increases up to 32.9 % in those lasting more than 8 h. Similarly, inclusion of a tracheotomy also increases the SWI rate from 8 up to 27 %. Several other groups have published comparable data in this regard [7, 23, 25, 28]. Further, foreign bodies, e.g. tracheal cannulas may be colonised by bacteria forming biofilm with an antibiotic resistency profile well adapted to the hospital environment [7]. A prolonged surgery usually correlates with major resections and reflects the complexity of the procedure, where preoperatively given antibiotics and surgical lavage may no longer be effective by the end of the procedure [28]. However, in contrast to our results, there are other publications demonstrating that tracheotomy and operating time did not have an impact on their SWI rates [11, 29]. Our results showed no difference between the SWI rates of the short-term (1 7 days) and long-term (C8) antibiotictreated groups (p = 0.689), similar to several other studies. However, definition of short term and long term in this regard varies a lot in the literature [5, 30]. Considering the distribution of risk factors for SWI between the two groups, patients in the long term treated group had significantly more risk factors than those in the short term treated group. However, this difference did not lead to a higher SWI rate in the long-term group. This imbalance is obviously a bias that arises from the retrospective character of this study. A similar effect was described by Sepehr et al. [31]. Although their SWI rates did not differ between the two groups either, there may have been a benefit from the prolonged antibiotic treatment in the long-term group, compensating for the unequal distribution of risk factors between the two groups. To clarify this hypothesis, prospective studies should be performed with stratification of the mentioned risk factors. In the literature, reported data regarding diabetes and low BMI as potential risk factors for SWI are controversial [3, 25]. In our study, we did not observe reduced SWI rates due to prolonged antibiotic treatment in diabetic patients. There was no correlation among the patients weight, the risk of SWI, and the duration of antibiotic prophylaxis either. Fistula formation did not correlate either with diabetes or with low BMI. Additional surgical procedures (e.g. tracheotomy) and the extent of surgery (major procedures) seem to have more impact on the incidence of complications, such as fistula formation or developing pneumonia, than the duration of the prophylactic antibiotic treatment. This is concordant with the results of a meta-analysis of postlaryngectomy pharyngocutaneous fistula formation [32]. However, 90 % of our patients with pneumonia did have a tracheotomy and they all were in the short-term antibiotic-treated group. Thus, high-risk patients undergoing major procedures including a tracheotomy might benefit from long-term prophylactic antibiotic treatment in terms of minimising the risk of pneumonia. Conclusion To date, this study presents the largest series of its kind, reporting on the effects of antibiotic prophylaxis in surgically treated head and neck cancer patients undergoing clean-contaminated procedures. Our data showed that long-term antibiotic prophylaxis is not associated with a further decrease in SWI rates in the entire cohort of patients. Furthermore, diabetes and low BMI did not increase the SWI rates. In either of these subgroups, prolonged administration of prophylactic antibiotics was not associated with lower infection rates. However, our data did confirm the extent of surgical procedure and tracheotomy as being risk factors for postoperative SWI. Due to the retrospective nature of this study, we may only assume that patients with a high risk of SWI might benefit from prolonged prophylactic antibiotic treatment. To verify this, a prospective study is required to focus on these subgroups with matched controls or risk factor stratification, in order to determine whether short-term antibiotic prophylaxis is still a valid option for patients with higher risk for SWI. The present retrospective analysis may only conclude that long-term antibiotic prophylaxis was not associated with further decreased SWI rates in patients undergoing clean-contaminated head and neck oncologic surgery and, therefore, prophylactic antibiotic treatment should not exceed 7 days in such patients without increased risk for SWI. References 1. Steinbrecher E et al (2002) Surveillance of postoperative wound infections: reference data of the Hospital Infection Surveillance System (KISS). Chirurg 73: Emmerson AM et al (1996) The Second National Prevalence Survey of infection in hospitals overview of the results. J Hosp Infect 32: Fusconi M et al (2006) Clean-contaminated neck surgery: risk of infection by intrinsic and extrinsic factors. Arch Otolaryngol Head Neck Surg 132: Clayman GL et al (1993) Bacteriologic profile of surgical infection after antibiotic prophylaxis. Head Neck 15: Righi M et al (1996) Short-term versus long-term antimicrobial prophylaxis in oncologic head and neck surgery. Head Neck 18: Agra IM et al (2003) Postoperative complications after en bloc salvage surgery for head and neck cancer. Arch Otolaryngol Head Neck Surg 129: Penel N et al (2005) Multivariate analysis of risk factors for wound infection in head and neck squamous cell carcinoma surgery with opening of mucosa. Study of 260 surgical procedures. Oral Oncol 41: Kreutzer K et al (2014) Current evidence regarding prophylactic antibiotics in head and neck and maxillofacial surgery. BioMed Res Int 2014:879437

7 9. de Melo GM et al (2001) Risk factors for postoperative complications in oral cancer and their prognostic implications. Arch Otolaryngol Head Neck Surg 127: Galli J et al (2005) Postlaryngectomy pharyngocutaneous fistula: incidence, predisposing factors, and therapy. Otolaryngol Head Neck Surg 133: Lotfi CJ et al (2008) Risk factors for surgical-site infections in head and neck cancer surgery. Otolaryngol Head Neck Surg 138: Rodrigo JP et al (2004) Efficacy of piperacillin-tazobactam in the treatment of surgical wound infection after clean-contaminated head and neck oncologic surgery. Head Neck 26: Simo R, French G (2006) The use of prophylactic antibiotics in head and neck oncological surgery. Curr Opin Otolaryngol Head Neck Surg 14: McConkey SJ et al (1999) Results of a comprehensive infection control program for reducing surgical-site infections in coronary artery bypass surgery. Infect Control Hosp Epidemiol 20: Harbarth S et al (2003) The preventable proportion of nosocomial infections: an overview of published reports. J Hosp Infect 54: quiz Kirby JP, Mazuski JE (2009) Prevention of surgical site infection. Surg Clin N Am 89: Woods RK, Dellinger EP (1998) Current guidelines for antibiotic prophylaxis of surgical wounds. Am Fam Physician 57: Barie PS (2002) Surgical site infections: epidemiology and prevention. Surg Infect (Larchmt) 3(Suppl 1):S9 S Classen DC et al (1992) The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection. N Engl J Med 326: Fennessy BG et al (2007) Antimicrobial prophylaxis in otorhinolaryngology/head and neck surgery. Clin Otolaryngol 32: Codina C et al (1999) Perioperative antibiotic prophylaxis in Spanish hospitals: results of a questionnaire survey. Hospital Pharmacy Antimicrobial Prophylaxis Study Group. Infect Control Hosp Epidemiol 20: Leaper DJ (2010) Surgical-site infection. Br J Surg 97: Coskun H et al (2000) Factors affecting wound infection rates in head and neck surgery. Otolaryngol Head Neck Surg : Weber RS, Callender DL (1992) Antibiotic prophylaxis in cleancontaminated head and neck oncologic surgery. Ann Otol Rhinol Laryngol Suppl 155: Liu SA et al (2007) Risk factors for wound infection after surgery in primary oral cavity cancer patients. Laryngoscope 117: Ganly I et al (2009) Analysis of postoperative complications of open partial laryngectomy. Head Neck 31: Robbins KT et al (1990) Risk of wound infection in patients with head and neck cancer. Head Neck 12: Schwartz SR et al (2004) Predictors of wound complications after laryngectomy: a study of over 2000 patients. Otolaryngol Head Neck Surg 131: Girod DA et al (1995) Risk factors for complications in cleancontaminated head and neck surgical procedures. Head Neck 17: Velanovich V (1991) A meta-analysis of prophylactic antibiotics in head and neck surgery. Plast Reconstr Surg 87: discussion Sepehr A et al (2009) Antibiotics in head and neck surgery in the setting of malnutrition, tracheotomy, and diabetes. Laryngoscope 119: Paydarfar JA, Birkmeyer NJ (2006) Complications in head and neck surgery: a meta-analysis of postlaryngectomy pharyngocutaneous fistula. Arch Otolaryngol Head Neck Surg 132: Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR (1999) Guideline for prevention of surgical site infection, Hospital infection control practices advisory committee. Infect Control Hosp Epidemiol 20(4): ; quiz