Risk factors of postoperative infection in head and neck surgery

Auris Nasus Larynx 36 (2009) 457–460 www.elsevier.com/locate/anl

Risk factors of postoperative infection in head and neck surgery Hitomi Ogihara, Kazuhiko Takeuchi *, Yuichi Majima Department of Otorhinolaryngology-Head & Neck Surgery, Mie University Graduate School of Medicine, 2-174 Edobashi, Tsu, Mie 514-8507, Japan Received 7 September 2007; accepted 30 October 2008 Available online 25 December 2008

Abstract Objectives: Patients undergoing head and neck oncological surgery are at great risk of developing complications following surgery. It is important to clarify risk factors contributing to operative wound infections. Methods: Two hundred and nine cases with head and neck tumor (87 female, and 122 male, averaged 58.6 years old) were studied at Mie University Hospital within a 24-month interval, during 2002 and 2003. Results: Twenty-one (10.0%) surgical site infections (SSIs) were recorded. Univariate analysis revealed that the following factors were significantly related to SSI: blood loss, previous chemotherapy, clean-contaminated surgery, tracheotomy, malignant tumor, advanced T-stage, flap reconstruction, and long operative time. Multivariate analysis showed that blood loss, previous chemotherapy, and the type of surgery contributed to SSI. Otherwise, the following factors did not show any significant correlations: age, body mass index, smoking, alcohol intake, diabetes, sex, previous radiotherapy, N-stage or ASA score. Conclusions: In order to minimize postoperative wound infection, clinicians managing the head and neck tumor patients should have a thorough understanding of the risk factors leading to postoperative infections such as blood loss, previous chemotherapy, and cleancontaminated surgery. # 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Surgical site infection; Blood loss; Chemotherapy

1. Introduction Patients undergoing head and neck oncological surgery are at great risk of developing complications following surgery. Postoperative wound infection is a frequent cause of morbidity and prolonged hospitalization [1]. Antibiotic prophylaxis dramatically reduces the risk of postoperative infection [2,3]. In head and neck surgery, there are many clean-contaminated surgical procedures, and the degree of contamination of the wound is also important. Chemoprophylaxis prolonged over 24 h after surgery cannot further reduce wound infection rates in subjects undergoing cleancontaminated major oncological surgery [4]. There is data demonstrating non-efficacy of antibiotics administered in patients undergoing uncontaminated head and neck surgery (parotidectomy, thyroidectomy or submandibular gland * Corresponding author. Tel.: +81 59 232 1111; fax: +81 59 231 5218. E-mail address: [email protected] (K. Takeuchi).

excision) when no infection exists prior to surgery [5]. Besides, preoperative radiation therapy and preoperative chemotherapy is frequently performed for patients with head and neck cancer. These therapies may be associated with a risk of postoperative wound infection [6–10]. As other risk factors, patient characteristics (sex, age, body mass index, tobacco and alcohol history, and so on), disease characteristics (site of tumor, tumor size, and so on) have been studied [6–8,11,12]. Clinicians managing the head and neck tumor patients should also have a thorough understanding of the risk factors leading to postoperative infections and should apply the most basic surgical principles at all times, to minimize infection rates. The purpose of this study was to reveal the risk factors associated with postoperative wound infection in the head and neck tumor in order to reduce wound infections. Many factors are interrelated, so multivariate analysis is used to identify important risk factors.

0385-8146/$ – see front matter # 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.anl.2008.10.005

458

H. Ogihara et al. / Auris Nasus Larynx 36 (2009) 457–460

2. Materials and methods

Table 2 Tumor site.

Records from 209 patients who underwent a head and neck surgical procedure for tumor were examined for postoperative infection. All procedures were performed by the Department of Otorhinolaryngology-Head and Neck Surgery at Mie University Hospital between January 2002 and December 2003. There were 87 women and 122 men. The mean patient age was 58.6 years (range, 9–86). We used cefazolin sodium 2 g/day bid only for 1 day in clean procedures, and for 3 days in clean-contaminated procedures. First, we examined whether they had postoperative infection or not, and the kind of infections. Postoperative infection includes all of the infections which occurred in the postoperative period. Next, 17 variables were recorded for all the patients. Those included age, sex, body mass index (BMI), diabetes, history of smoking and alcohol, American Society of Anesthesiology score (ASA score), benign or malignant of the tumor, T-stage, lymph node metastasis, previous chemotherapy, previous radiotherapy, tracheotomy, clean or clean-contaminated surgery, flap reconstruction, operation time, and the amount of blood loss. SSI was defined as the presence of purulent drainage from the wound or the presence of an orocutaneous or pharyngocutaneous fistula regardless of origin (including flap failure). One or two courses of 5-FU (800 mg/m2) on days 1–5 and cisplatin (80 mg/m2) on day 6 were the standard regimen used for induction chemotherapy. Statistical evaluation, univariate analysis was performed using thex2-test for categorical data and Mann–Whitney test for continuous variables. The level of significance was set at p < 0.05. Multivariate logistic regression analysis was carried out to calculate odd ratios.

Tumor sites

Overall no. of patient

Patient with SSI (%)

Oral cavity Hypopharynx Mesopharynx Larynx Hearing organs Nose and sinus Thyroid gland Parotid gland Parapharyngealspace Neck

30 8 9 12 6 11 31 40 1 20

6 4 2 2 2 1 1 1 1 1

3. Results The overall post-operative infection rate was 14.4% (30 cases), with 21 cases (10.0%) having SSI. Remaining infections include pneumonia (4 cases), upper respiratory tract infection (3 cases), meningitis from spinal drainage tube (1 case), and cholangitis (1 case) (Table 1). Table 2 shows the primary tumor location of cases which had SSI. Oral cavities and pharynx were frequent sites of Table 1 The kind of infections. Infections

No. of patients (%)

SSI Pneumonia Upper respiratory infection Meningitis Cholangitis

21 4 3 1 1

Total

30 (14.4)

(10.0) (1.9) (1.4) (0.48) (0.48)

(20.0) (50.0) (22.2) (16.7) (33.3) (9.1) (3.2) (2.5) (100.0) (5.0)

primary tumors. The cases of thyroid tumor and parotid tumor with SSI were both malignant. Several of the preoperative and operative risk factors [1– 3,6] anticipated to have some association with SSI rate are outlined in Table 3. T-stage, previous chemotherapy, tracheotomy, surgery classification (clean or not), and free flap reconstruction were found to be significantly associated with an increase of SSI. The preoperative patient parameters and operative parameters examined are outlined in Table 4. The amount of blood loss and duration of operative times were associated with SSI. However, age and BMI were found not to be associated with increased infections. In addition, each factor was examined by multivariate logistic regression analysis. Blood loss, previous chemotherapy, and type of surgery were significantly related in SSI (Table 5).

4. Discussion In head and neck surgery, the operation fields are not necessarily clean as in neck dissections, thyroidectomy, and parotidectomy, but sometimes clean-contaminated as in oral cavity, pharynx, larynx, etc., where there are normal flora. Besides, the operation fields can sometimes be contaminated such as in injury wound and the drainage of deep neck abscess. In this study, we analyzed the risk of factors in the surgery of head and neck tumor. Multivariate analysis revealed that blood loss, previous chemotherapy, and type of surgery were especially related in SSI. The patients with head and neck malignant tumors sometimes have preoperative chemotherapy. When large defects of tissues are given by the dissection of the primary tumors, they need more extensive surgery involving the reconstruction by free flap or local flap and so on, which usually lead to increased intraopeartive blood loss. These factors, blood loss and preoperative chemotherapy, are considered to influence local and general states of the patients, and to result in postoperative infection. The half of the cases with free flap reconstruction had surgical site infections. Technical improvement in the reconstruction

H. Ogihara et al. / Auris Nasus Larynx 36 (2009) 457–460

459

Table 3 Parameters and SSI rate (categorical data). Risk factor

Categories

Overall no. of patients

Patients with SSI (%)

p value

Sex

Men Women

122 87

15 (12.3) 6 (6.9)

0.2

Diabetes

Yes No

19 190

3 (15.8) 18 (25.3)

0.38

Smoking

Yes No

91 118

11 (12.1) 10 (8.5)

0.39

Alcohol

Yes No

99 110

13 (13.1) 8 (7.2)

0.16

ASA score

3+4 1+2

8 181

1 (12.5) 20 (11.0)

0.89

Malignant

Yes No

149 60

20 (13.4) 1 (1.7)

0.0105

T-stage

3+4 1+2

19 110

11 (57.9) 35 (31.8)

0.0284

Neck metastasis

Yes No

61 70

7 (11.5) 12 (17.1)

0.36

Previous chemotherapy

Yes No

23 183

7 (30.4) 14 (7.7)

<0.0001

Previous radiotherapy

Yes No

19 188

1 (5.3) 20 (10.6)

0.46

Tracheotomy

Yes No

35 174

10 (28.6) 11 (6.3)

<0.0001

Surgery classification

Clean-contaminated Clean

86 123

16 (18.6) 5 (4.1)

0.0006

Free flap reconstruction

Yes No

18 191

9 (50.0) 12 (6.3)

<0.0001

Table 4 Parameters and SSI rate (continuous data).

Blood loss (ml) Operation time (min) Age (years old) BMI

Over all average

Average with SSI

Average without SSI

p value

275  449 265  229 58.8  16.8 22  3

701  600 510  282 63.0  9.45 23  4

227.2  403.4 238.9  207.5 58.3  17.4 23  3

<0.0001 <0.0001 0.45 0.27

Standard deviation.

procedures is necessary. We found the degree of contamination of the surgery also significantly associated with SSI. Out of the 21 cases with SSI, necrotomy was performed in 6 cases involving reconstruction. In the other 15 cases, the wound was made open and was irrigated with physiological Table 5 Multivariate logistic regression analysis. Factors

Odds ratio (CI 95%)

Blood loss > 700 ml Previous chemotherapy Clean-contaminated surgery Tracheotomy T3 + 4 Free flap reconstruction

4.21 4.06 2.57 2.45 2.21 1.63

CI: Confidence interval.

(0.30–59.5) (0.59–27.9) (0.40–16.6) (0.41–14.5) (0.51–9.50) (0.25–10.6)

saline. Depending on the case, it took 1–4 weeks for SSIs to be resolved. Compared with the reported overall rate of SSI (6.6– 41.8%) [6–9,14], the rate in our series (10.0%) is not so high. Previously, the risk factors of SSI have been analyzed by several investigators [6–8]. In our study, no significance correlation was found between infection and preoperative radiotherapy, but Girod et al. noted significantly higher infection rates in the patients who experienced preoperative radiotherapy [7]. In the present study, the factors related to the complexity of the operation procedures, such as blood loss or operative time, were associated with SSI. In contrast, several general factors, age, sex, BMI, history of smoking and alcohol, diabetes, were not significantly associated with SSI in the present study. Although we did not examine the term of antibiotic prophylaxis, Mauro et al. reported that short-term (1 day)

460

H. Ogihara et al. / Auris Nasus Larynx 36 (2009) 457–460

Table 6 Comparison of univariate analysis results of risk factors of surgical site infection by previous reports. Factors

The present study

Panel [6] (2001)

Coskun [14] (2000)

Previous chemotherapy Tracheostomy Blood loss Operation time Free flap reconstruction Clean or contaminated Tumor stage Prior radiotherapy Nutritional status Alcohol consumption Blood replacement

p < 0.0001 p < 0.0001 p < 0.0001 p < 0.0001 p = 0.0001 p = 0.0006 p = 0.0284

p = 0.008 p = 0.00008

p = 0.006

p < 0.0001

p = 0.002

p < 0.001 p = 0.003 p < 0.0005 p = 0.001

p = 0.044

prophylaxis is as effective as a long-term (3 days) antibiotic regimen, and the incidence of systemic infection could not be reduced by increasing the duration of antibiotic prophylaxis [4]. Prolonged antibiotic courses with a prophylactic purpose are still frequently used in common surgical practice in Japan. Short antibiotic regimes of four doses per 24 h are as effective as prolonged courses regardless of the complexity of the procedure [13]. A combination of antibiotic agents covering aerobic, anaerobic and Gram-negative bacteria is superior to single agents [13]. It would be important to maintain the effective blood concentration of the antibiotics when we start surgery. Finally Table 6 shows the summary of the univariate analysis of risk factors of SSI reported in previous studies [6–8,14]. Many of the factors found in the present study have been pointed out by other authors. The disadvantage of postoperative infection is not only cosmetic disturbance but also prolonged hospitalization and delay of starting postoperative therapy. Furthermore, it has reported that SSI was related to patient’s prognosis [1]. In order to prevent infections, we need to make an effort to reduce these risk factors, and for high risk patients, we should pay more attention to infections. In order to prevent postoperative dead space formation, we use drainage tubes with larger diameter and squeeze them regularly. Since dead spaces are often formed in the supra clavicular fossa, we undermine the skin on the clavicula and bring it down to the fossa. Furthermore, we use antibiotics every 6 h in prolonged operations.

5. Conclusions We examined the factors contributing to postoperative infections for patients with head and neck tumor. In order to minimize postoperative wound infection, clinicians managing the head and neck tumor patients should have a thorough understanding of the risk factors leading to

Girod [7] (1995)

Robbins [8] (1990)

p = 0.005 p = 0.001 p < 0.05 p = 0.01

postoperative infections such as blood loss, previous chemotherapy, and clean-contaminated surgery.

References [1] Grandis JR, Synderman CH, Johnson JT, Yu VL, D’Amico F. Postoperative wound infection. A poor prognostic sign for patients with head and neck cancer. Cancer 1992;70:2166–70. [2] Fraioli R, Johnson JT. Prevention and treatment of postsurgical head and neck infection. Curr Infect Dis Rep 2004;6:172–80. [3] Johnson JT, Myers EN, Thearle PB, Sigler BA, Schramm Jr VL. Antimicrobial prophylaxis for contaminated head and neck surgery. Laryngoscope 1984;94:46–51. [4] Righi M, Manfedi R, Farneti G, Pasquini E, Cenacchi V. Short-term versus long-time antimicrobial prophylaxis in oncologic head and neck surgery. Head Neck 1996;18:399–404. [5] Johnson JT, Wagner RL. Infection following uncontaminated head and neck surgery. Arch Otolaryngol Head Neck Surg 1987;113:368–9. [6] Penel N, Lefebvre D, Fournier C, Sarini J, Kara A, Lefebvre J. Risk factors for wound infection in head and neck cancer surgery: a prospective study. Head Neck 2001;23:447–55. [7] Girod DA, McCulloch TM, Tsue TT, Weymuller EA. Risk factors for complications in clean-contaminated head and neck surgical procedures. Head Neck 1995;17:7–13. [8] Robbins KT, Favrot S, Hanna D, Cole R. Risk of wound infection in patients with head and neck cancer. Head Neck 1990;12:143–8. [9] Penel N, Fournier C, Lefebvre D, Roussel-Delvallez M, Sarini J, Kara A, et al. Previous chemotherapy as a predictor of wound infections in nonmajor head and neck surgery: results of a prospective study. Head Neck 2004;26:513–7. [10] Johnson JT, Bloomer WD. Effect of prior radiotherapy on postsurgical wound infection. Head Neck 1989;11:132–6. [11] Brown BM, Johnson JT, Wagner RL. Etiologic factors in head and neck wound infections. Laryngoscope 1987;97:587–90. [12] Juan PR, Juan CA, Justo RG, Carlos S, Jose AF, Jose AM. Comparison of three prophylactic antibiotic regimens in clean-contaminated head and neck surgery. Head Neck 1997;19:189–93. [13] Simo R, French G. The use of prophylactic antibiotics in head and neck oncological surgery. Curr Opin Otolaryngol Head Neck Surg 2006;14:55–61. [14] Coskun H, Erisen L, Basut O. Factors affecting wound infection rates in head and neck surgery. Otolaryngol Head and Neck Surg 2000;123:328–33.