The American Journal of Surgery (2014) 207, 476-484
Clinical Science
A randomized trial of antibiotic prophylaxis for the prevention of surgical site infection after open mesh-plug hernia repair Takero Mazaki, M.D., Ph.D.a,*, Kazunari Mado, M.D.a, Hideki Masuda, M.D., Ph.D.a, Motomi Shiono, M.D., Ph.D.a, Noahiro Tochikura, M.Sc.b, Morio Kaburagi, M.Sc.b Departments of aSurgery, and bPharmacy, Nihon University School of Medicine, Tokyo, Japan
KEYWORDS: Inguinal hernia; Mesh-plug; Repair; Surgical-site infection; Antibiotic prophylaxis
Abstract BACKGROUND: The efficacy of antibiotic prophylaxis for the prevention of surgical-site infection (SSI) after open tension-free inguinal hernia repair remains controversial. METHODS: A double-blind, randomized, placebo-controlled trial was conducted. Patients who underwent elective open mesh-plug hernia repair were eligible for randomization. In the antibiotic prophylaxis group, 1.0 g cefazolin was intravenously administrated 30 minutes before the incision. In the placebo group, an equal volume of sterile saline was administered. The primary end point was the incidence of SSI. RESULTS: A total of 200 patients were enrolled. SSI developed in 2 of 100 patients (2%) in the antibiotic prophylaxis group and 13 of 100 patients (13%) in the placebo group, indicating a significant difference between the 2 groups (relative risk ratio, 0.25; 95% confidence interval, 0.070 to 0.92; P 5 .003). Other complications occurred in 23 patients: 7 (7%) in the antibiotic prophylaxis group and 16 (16%) in the placebo group (P 5 .046). CONCLUSIONS: This study indicates that antibiotic prophylaxis is effective for the prevention of SSI after open mesh-plug hernia repair. Ó 2014 Elsevier Inc. All rights reserved.
Surgeons routinely administer antibiotic prophylaxis when a prosthetic device is involved, for example, in cardiothoracic surgery and in hip arthroplasty, because infection can result in critical conditions such as removal of the prosthesis, prolongation of the hospital stay, and an increase in cost. On the other hand, it is questionable whether the routine use of antibiotic prophylaxis is necessary for the prevention of surgical-site infection (SSI) in a The authors declare no conflicts of interest. * Corresponding author. Tel.: 181-3-3972-8111; fax: 181-3-3972-8750. E-mail address: [email protected] Manuscript received October 28, 2012; revised manuscript January 2, 2013 0002-9610/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2013.01.047
simple, clean surgical procedure such as inguinal hernia repair or breast surgery.1 In patients undergoing inguinal hernia repair, 10 randomized controlled trials have been conducted during the past decade2–12 to evaluate the effectiveness of the routine use of antibiotic prophylaxis. Two of these studies2,4 showed that prophylactic antibiotics were effective in reducing infection rates, but 8 of the studies3,5–11 recommended against their use. However, most of the studies had a relatively small number of patients and therefore did not have the statistical power necessary to detect a significant difference. Likewise, 6 meta-analyses failed to show statistically conclusive evidence that prophylactic antibiotics were effective.12–17 This is therefore still a controversial
T. Mazaki et al.
SSI after tension-free inguinal hernia repair
issue. The randomized controlled trials enrolled low-risk patients rather than high-risk patients. On the basis of these results, the European Hernia Society guidelines18 state that there is no indication for the routine use of antibiotic prophylaxis for low-risk adult patients in clinical settings with low infection rates (,5%) but that their use is recommended in the presence of risk factors for SSI, such as recurrence of hernia, advanced age, and immunosuppression. Approximately 160,000 inguinal hernia repairs are performed annually in Japan, and .1 million repairs are performed annually in both the United States and Europe.3 If antibiotic prophylaxis can be avoided in inguinal hernia repairs, we can not only minimize cost but also reduce the risk for allergic side effects and the possible development of bacterial resistance.19 Furthermore, the risk for a clinically significant infection is not large enough to justify the routine use of antibiotic prophylaxis, because SSI develops in only 1% to 4% of patients undergoing inguinal hernia repair.20 In addition, a review of inguinal hernia repairs at our institution, performed with mesh and without mesh, indicated that SSI occurred in approximately 1% of patients (unpublished data). We subsequently found that our surgeons were administering prophylactic antibiotics despite the absence of any evidence of their effectiveness. The above concerns are the main arguments against the routine use of prophylactic antibiotics. We therefore conducted a single-center, randomized, double-blind, placebo-controlled trial to clarify whether antibiotic prophylaxis for primary inguinal hernia surgery reduces morbidity.
Methods Study design The trial was conducted in the Department of General Surgery at the Nihon University School of Medicine from July 2007 to December 2011. The ethics committees at the hospital approved the trial, and all patients provided written informed consent. The trial was registered at ClinicalTrials.gov (NCT00636831).
Inclusion and exclusion criteria Inpatients who were scheduled for elective primary unilateral or bilateral open mesh-plug hernia repair were eligible for the study. Because open inguinal hernia repairs for outpatients were uncommon in Japan (the Japanese Ministry of Health, Labor, and Welfare reported approximately 1.6% of the rate of the surgery in 2011; http://www. e-stat.go.jp/SG1/estat/List.do?lid5000001097372), we excluded outpatients from this study. The exclusion criteria were outpatient status; day surgery; age ,18 years; recurrent hernia; incarcerated or strangulated hernia requiring emergency hernia repair; pregnancy or lactation; earlier history of allergy, sensitivity, or anaphylaxis to b-lactam
477 or cephalosporin antibiotics; antibiotic therapy ,48 hours before surgery; presence of an infection at the time of surgery; cardiac valvular problem; increased risk for infection secondary to a coexisting medical condition; immunosuppression (eg, human immunodeficiency virus infection, malignancy, or chemotherapy); American Society of Anesthesiologists (ASA) grade . IV; and refusal to participate in the study. If patients had diabetes mellitus, we controlled the value of glycosylated hemoglobin at ,6.5%.
Randomization Patients were randomly assigned on admission in a double-blinded manner to either the antibiotic prophylaxis group or the placebo group. All surgeons and other staff members were blinded to randomization and to patients’ details. According to a computer-generated list in blocks of 50 patients, a pharmacist carried out randomization.
Surgical technique A certified surgeon or surgical resident who was blinded to study group assignment performed surgery under general and local anesthesia, and this was standardized among surgeons. The skin was shaved just before surgery and prepared using 10% povidone iodine. All patients underwent elective open mesh-plug hernia repair using a monofilament polypropylene mesh (PerFix Plug; CR Bard, Cranston, RI). The mesh-plug was sutured in place using 2-0 monofilament polypropylene sutures (Polysorb; Covidien, MI). Any subcutaneous suture was not used. The skin was closed with interrupted 2-0 nylon sutures (Alfresa Pharma, Ltd, Tokyo, Japan). No drain was placed. This was standardized among surgeons.
Intervention The antibiotic prophylaxis group received 100 mL sterile saline with 1.0 g cefazolin (Astellas Pharma, Ltd, Tokyo, Japan) by continuous intravenous infusion. An equal volume of sterile saline (Otsuka Pharmaceutical Factory Ltd, Tokushima, Japan) was administrated to the placebo group 30 minutes before the incision.
Follow-up Although a type of wound dressing was not standardized in the hospital, surgeons uniformly removed the dressing and suture 7 to 8 days after surgery (the first follow-up visit) in the outpatient clinic. The second and third follow-up visits were at 1 and 3 months after surgery. All wounds were carefully examined by 2 certified surgeons, who did not perform the operations. According to the most recent criteria from the Centers for Disease Control and Prevention, wound infection was categorized as superficial SSI or deep SSI (DSSI).21 These criteria define superficial SSI as occurring
478 ,30 days after surgery and involving only the skin or the subcutaneous tissue at the incision site. DSSI occurs ,1 year after surgery if a prosthesis is in place and the infection involves deep soft tissues at the incision site. The patients’ demographic data were recorded: body mass index (obesity was defined as .30 kg/m2 according to the World Health Organization classification), type of hernia (direct, indirect, and combined), hernia localization (right, left, and bilateral), coexisting conditions (eg, diabetes mellitus, heart disease, neurologic disorders, pulmonary diseases, and others), ASA grade, surgeon (certified and resident), duration of procedure, length of incision, and length of hospital stay.
The American Journal of Surgery, Vol 207, No 4, April 2014 this analysis, we assessed whether the relative treatment effect was consistent across the spectrum of SSI risk. We performed additional subgroup analysis on the following prespecified characteristics: age, sex, type of hernia, body mass index, location of hernia, grade of surgeon, duration of procedure, ASA grade, and coexisting diseases. All subgroup statistical analyses were evaluated for interaction effects with antibiotic prophylaxis by testing for the significance of a corresponding interaction term in a multiple logistic regression analysis. Data were analyzed using Stata version 12.1 (StataCorp LP, College Station, TX).
Results Statistical analysis Continuous, normally distributed variables are presented as mean 6 SD and skewed variables as medians with interquartile ranges. Student’s t tests and Wilcoxon’s ranksum tests were used to analyze continuous variables among the categories. Categorical variables are presented as absolute numbers and percentages. Comparative analysis of proportions was performed using chi-square or Fisher’s exact tests. The primary and secondary outcomes are reported in terms of relative risk ratios and relative risk reduction. The primary end point was the incidence of SSI during the 3-month period after surgery in the antibiotic prophylaxis and placebo groups. All statistics were based on the intention-to-treat principle. We based the sample size calculation on 2 randomized controlled trials.2,3 The target sample size of 400 gave 80% power, with a P value ,.05, to detect a difference in the occurrence of SSI between the antibiotic prophylaxis and placebo groups, with a 6% loss of patients, with the assumption that the occurrence of SSI in the antibiotic prophylaxis and the placebo group was 1% and 7%, respectively.2,3 We planned 4 interim analyses after every 100 patients had been randomized, with a final 2-sided P value ,.031 indicating statistical significance. This P value was based on the Lan-DeMets method and the O’Brien-Fleming a spending function.22,23 A data and safety monitoring board provided regulatory oversight by reviewing blinded subject data quarterly and conducting the a priori scheduled interim analysis. The secondary end point, which was the incidence of any other complications in the study groups during the 3 months after surgery, was similarly calculated, with a P value ,.05 indicating statistical significance. Adverse events that were potentially attributable to the study antibiotics were also reported. According to a previously proposed framework for evaluating the heterogeneity of treatment effects on the primary end point,24 we performed a post hoc analysis of the patients’ data according to their risk factors for SSI. The risk factors were as follows: age R70 years, body mass index R30 kg/m2, diabetes mellitus, ASA grade R II, procedure duration . 60 minutes, bilateral hernia, and recurrence of hernia.3,13,17,18 Individual patient risk scores were determined by assigning 1 point for the presence of each risk factor. In
Patients Based on the results of the first interim analysis, the data and safety monitoring board recommended a second
Table 1
Baseline and operative characteristics*
Characteristic Age (y) Men Body mass index (kg/m2) Type of hernia Direct Indirect Combined Unknown Hernia characteristics Right Left Bilateral Coexisting conditions† Diabetes mellitus Heart disease Neurologic disease Pulmonary disease Other Mixed American Society of Anesthesiologists health status grade I II III Surgeon Resident Certified surgeon Duration of surgery (min) Length of incision (cm) Length of hospital stay (d)
Antibiotic group (n 5 100)
Placebo group (n 5 100)
69 (57–76) 89 23.0 6 2.8
72 (60–77) 94 22.7 6 3.0
30 67 2 1
26 67 7 0
61 34 5
50 42 8
6 24 1 3 2 14
7 31 2 7 0 21
52 43 5
37 58 5
84 83 16 17 66.3 6 25.4 65.2 6 27.1 5.2 6 0.7 5.1 6 0.7 3 (2–3) 3 (2–3)
*There were no significant differences in the baseline characteristics of the antibiotic and placebo groups. † Some patients were counted twice if they had .1 complication.
T. Mazaki et al.
SSI after tension-free inguinal hernia repair
interim analysis after an additional 100 patients had been enrolled. In December 2011, the board assessed the outcomes of the second interim analysis and recommended that the study be terminated early on the basis of the harm of placebo compared with antibiotic prophylaxis. Thus, according to the prior stopping rule, we stopped the study. A total of 100 patients received antibiotics, and 100 patients received placebo (Fig. 1). Follow-up of all patients for both primary and secondary outcomes was completed.
Study outcomes A total of 33 surgeons (23 residents and 10 certified) participated in this study. Residents performed 83% of all operations, and all infections occurred in operations performed by residents. There was no significant difference in any baseline characteristic between the 2 groups, with the exception of the incidence of SSI and any other complications (Table 1). The primary outcome, SSI, occurred in 15 of 200 patients (7.5%), with 2 of 100 (2%) occurring in the antibiotic prophylaxis group and 13 of 100 (13%) occurring in
Figure 1
479 the placebo group. This corresponded to a relative risk ratio of 0.25 (95% confidence interval, 0.07 to 0.92; P 5 .006) and an absolute risk reduction of 11 percentage points (number needed to treat to prevent 1 episode of SSI, 9) (Supplementary Figure) and a relative risk reduction of 85% (Fig. 2A). At this interim analysis, the Bayesian prediction probability of the effectiveness of antibiotic prophylaxis was 99.9%. A majority of cases of SSI were superficial SSI, and no DSSI was found in either group. One late-onset infection was recognized 1 year 9 months after surgery in the placebo group. However, this case was excluded from the primary end point according to the Centers for Disease Control and Prevention inclusion criteria for surgical wound infections. Secondary outcomes occurred in 23 patients: 7 (7%) in the antibiotic prophylaxis group and 16 (16%) in the placebo group (P 5 .046; Fig. 2B). The secondary end points were follows: 4 hematomas, 1 pneumonia, 2 seromas, and 1 urinary retention in the antibiotic prophylaxis group and 1 acute pancreatitis, 7 hematomas, 1 pneumonia, 1 orchitis, 6 seromas, and 1 urinary tract infection in the placebo group (Table 2). Regarding recurrence of hernia, we
Flow of patients through the trial. PHS 5 Prolene hernia system.
480
The American Journal of Surgery, Vol 207, No 4, April 2014
A
B
Figure 2 Incidence of primary and secondary outcomes. (A) SSI developed in 13 of 100 patients (13%) in the placebo group and 2 of 100 patients (2%) in the antibiotic prophylaxis group (P 5 .003). (B) Other complications developed in 16 of 100 patients (16%) in the placebo group and 7 of 100 patients (7%) in the placebo group (P 5 .046).
documented 2 recurrences (2%) in the antibiotic prophylaxis group and 1 recurrence (1%) in the placebo group (Table 2). There were no intraoperative complications, and neither adverse events nor mortality was seen in either group. The median length of hospital stay was 3 days for each group. History of smoking was not investigated.
Heterogeneity of treatment effects According to the patients’ risk scores, the relative efficacy of antibiotic prophylaxis for the prevention of SSI did not vary significantly. The absolute risk reduction was a number needed to treat of 9 for those with risk scores ,2 (%2 risk factors) compared with a number needed to treat of 9 for those with risk scores .3 (R3 risk factors) (Fig. 3).
No patients required removal of the mesh in either group. All patients recovered completely.
Comments This study indicates that antibiotic prophylaxis is effective for the prevention of SSI after open mesh-plug hernia repair, reducing the incidence of SSI from 13% in the placebo group to 2% in the antibiotic prophylaxis group. The number needed to treat to prevent 1 episode of SSI was 9. Moreover, we found that antibiotic prophylaxis decreased the incidence of other complications. Although this study showed that the incidence of SSI occurred 13% in the placebo group, the incidence is not extremely high compared with previous results. Four randomized controlled trials2,4,9,11 assessed the efficacy of
Exploratory subgroup analysis Table 2
The beneficial effect of antibiotic prophylaxis on the primary outcome was also consistent across the other prespecified subgroups, with the exception of female sex (Fig. 4). Antibiotic prophylaxis was protective against SSI in all subgroups.
Details of the infected patients All infected wounds were diagnosed after hospital discharge. Two patients in the antibiotic prophylaxis group were treated with drainage alone of the infected wound. Eleven patients were treated with oral antibiotics and wound drainage, and the remaining 4 patients treated with only drainage (Table 3). One patient with a bilateral hernia and 14 patients with unilateral hernias had SSIs.
Incidence of postoperative complications
End point Any wound infections Deep infection Superficial infection Any other complications Acute pancreatitis Hematoma formation Orchitis Pneumonia Seroma formation Urinary retention Urinary tract infection Recurrence Mortality
Antibiotic group (n 5 100)
Placebo group (n 5 100)
0 2
0 13
0 4 0 1 2 1 0 2 0
1 7 1 1 6 0 1 1 0
T. Mazaki et al.
SSI after tension-free inguinal hernia repair
481
Figure 3 The relative treatment effect was consistent across the spectrum of SSI risk. An individual patient’s risk score was determined by assigning 1 point for the presence of each risk factor. The risk score (range, 0 to 4) was divided into 2 categories and evaluated using multiple logistic regression. Analysis revealed no evidence that the relative risk reduction associated with antibiotic prophylaxis varied across the risk groups (P 5 .60). The difference in the relative risk reduction between the groups was not significant (P 5 .71 for the Mantel-Haenszel test for homogeneity). The red vertical line shows the overall risk reduction, and the dashed vertical line indicates no relative risk reduction.
prophylactic antibiotic use after open inguinal hernia repair and found that the incidence varied from 8.2% to 12.5% in the placebo group. A large Scottish study25 evaluated the efficacy of telephone interviews in postdischarge surveillance for SSI after an inguinal hernia repair, demonstrated that the incidence varied from 0% to 14.6% among hospitals. Our study included patients with more risk factors (aged .70 years, bilateral hernia, diabetes mellitus, ASA grade III, and operation duration .60 minutes) than other randomized controlled trials. Therefore, we believe that these factors reflect our 13% SSI incidence in the placebo group in this study. Residents operated on 83% of all patients, and all infections occurred in operations performed by residents in this study. We have no idea whether our 7.5% infection incidence reflects a common problem in training centers, at which residents perform procedures. However, 2 randomized controlled trials3,9 including 43% and 89% of patients operated on by residents found that the grade of surgeon was not a significant risk factor for SSI. Also, Taylor et al25 concluded that surgeon’s experience did not influence the incidence of SSI. On the other hand, 13% of patients had bilateral hernias, and 1 of those had SSI in this study. In addition to our study, 3 randomized controlled trials3,9,11 in which 6%, 8%, and 11% of all patients had bilateral hernias found no correlation between a bilateral hernia and the incidence of SSI. A previous report stated that antibiotic prophylaxis should be avoided in low-risk patients, because the incidence of SSI is as low as 2% to 3%, and infections can be treated in a straightforward manner with wound drainage plus antibiotic administration.2 The selective use of prophylactic antibiotics for low-risk patients is a matter of debate for several reasons. First, we underestimate the incidence of SSI. This is primarily because the incidence of SSI is influenced by patient-related risk factors, surgery-related risk factors, and the sanitary conditions of the institution, which are not the same for all patients and all institutions. Indeed, a large study
reported that the incidence of SSI varied from 0% to 14.6%25 among hospitals. This indicates that the true incidence of SSI can be quite high, and we could expect a benefit from antibiotic prophylaxis for all patients. Second, a recent meta-analysis17 showed significant effectiveness of antibiotic prophylaxis (odds ratio, 0.54; 95% confidence interval, 0.37 to 0.81), although most patients included in the metaanalysis were at low risk. This study indicates that antibiotic prophylaxis is significantly effective for low-risk patients. Third, as shown in Figs. 3 and 4, the relative efficacy of antibiotic prophylaxis did not vary according to risk factors such as patient’s risk score, ASA grade, and the presence of coexisting conditions. Consequently, regardless of the incidence of SSI and the presence of low-risk patients, we believe that the routine use of antibiotic prophylaxis provides benefit. One may wonder if patients remain at increased risk for SSI because they are in the hospital, but this explanation is inappropriate. Taylor et al25 reported that the incidence of SSI in inpatients and in outpatients was 5.4% and 5.0%, respectively. Also, the mean incidence of SSI in 5 randomized controlled trials2,8–11 including inpatients was approximately 6.5%; this is almost the same as the result of Taylor et al.25 Therefore, there are no difference in the incidence of SSI between outpatients and inpatients. Our study was not designed to answer the question of the cost-effectiveness of prophylactic antibiotic use. Aufenacker et al,3 using a database from the Netherlands, demonstrated that avoiding antibiotic prophylaxis would save approximately V10 million in the United States and Europe. In contrast, a study conducted to assess postoperative infection-related costs found that annual expenses for infections after inguinal hernia repair ($44,800 per patient) were similar to those for colon surgery ($48,440 per patient).26 These studies show that infection constitutes an important problem in terms of health care costs. We believe that prophylactic antibiotics are effective from an economic point of view. However, the potential benefits of antibiotics were not carefully analyzed for cost-effectiveness.
482
The American Journal of Surgery, Vol 207, No 4, April 2014
Figure 4 Exploratory subgroup analysis. The primary end point was consistent across the prespecified and post hoc subgroups. There was no significant interaction between the antibiotic prophylaxis group and any of these subgroups. The relative risk reduction was not indicated for the certified surgeon and presence of diabetes mellitus groups because SSI was not seen in either category. The red vertical line shows the overall risk reduction, and the dashed vertical line indicates no relative risk reduction. BMI 5 body mass index.
DSSI is thought to be a risk for the recurrence of inguinal hernia, whereas some studies have found that SSI does not increase the recurrence of hernia.4,10,27 Actually, 3 recurrences of hernia in our study did not become infected. According to Aufenacker et al,3 who found that the incidence of DSSI ranged from 0.2% to 0.4%, the rare DSSI had a markedly low recurrence rate of inguinal hernia.
This is a controversial issue. A meta-analysis that looked at the effectiveness of antibiotic prophylaxis for the prevention of DSSI showed no efficacy of antibiotics for the prevention of DSSI (odds ratio, 0.50; 95% confidence interval, 0.12 to 2.09). The meta-analysis, however, did not have the power to detect the efficacy of prophylactic antibiotics, as the power was 10%. These findings therefore should be
T. Mazaki et al. Table 3
SSI after tension-free inguinal hernia repair
483
Details of surgical-site infections in the antibiotic prophylaxis group and placebo group
Group
Days after surgery
Microorganism cultured
Treatment
Outcome
Antibiotic
15 10 11 9
Culture not performed Culture not performed Staphylococcus aureus Enterococcus faecalis, coagulase-negative Staphylococcus Coagulase-negative Staphylococcus, Corynebacterium spp, E faecalis Staphylococcus aureus, coagulase-negative Staphylococcus S aureus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Pseudomonas aeruginosa, S aureus, coagulase-negative Staphylococcus Methicillin-resistant S aureus Culture not performed Coagulase-negative Staphylococcus S aureus S aureus, Corynebacterium spp
Drainage Drainage Antibiotics and drainage Antibiotics and drainage
Recovered Recovered Recovered Recovered
Drainage
Recovered
Antibiotics and drainage
Recovered
Antibiotics and drainage Drainage Drainage Antibiotics and drainage
Recovered Recovered Recovered Recovered
Antibiotics Antibiotics Drainage Antibiotics Antibiotics
Recovered Recovered Recovered Recovered Recovered
Placebo
7 10 9 12 12 12 5 7 8 7 9
interpreted cautiously. Taylor et al25 concluded that DSSI frequently necessitated complete removal of the mesh, indicating that management of DSSI is difficult and eventually required repeat hospital admission.6 Whether the placement of a drain can prevent SSI is controversial. Yerdel et al2 showed that the use of drains increased the risk for SSI, whereas Aufenacker et al3 and Tzovaras et al7 demonstrated no increased risk. These randomized controlled trials gave contradictory results. Guidelines, therefore, recommend that drains should be used as indicated for patients with increased blood loss and coagulopathies.18 Late-onset deep infection occurring years after surgery is a rare complication, with reports of occurrence ranging from 0.03% to 1.4%.28–30 We experienced 1 late-onset infection in the placebo group that occurred 1 year 9 months after the operation. The patient did not require mesh removal and recovered completely with antibiotic administration. A retrospective study showed that the incidence of DSSI did not correlate with the administration of antibiotic prophylaxis, the presence of previous SSI, the type of mesh, or the incidence of recurrence of hernia.29 Although the etiology has not yet been elucidated, this type of infection may likely become increasingly common with the widespread use of prosthetic mesh.29
Conclusions The routine use of antibiotic prophylaxis significantly reduced the incidence of not only SSI but also any other complications, whether patients had high-risk or low-risk factors. Simultaneously, there were no significant interaction terms between antibiotic prophylaxis and any subgroups. We therefore conclude that it is appropriate to use
and drainage and drainage and drainage and drainage
routine antibiotic prophylaxis for the prevention of SSI in patients who undergo in-hospital inguinal hernia repair. Further study is needed to answer questions such as what constitutes a real risk factor, the cost-effectiveness of prophylactic antibiotics, and late-onset infection, among other questions.
Acknowledgment We would like to thank the surgeons, residents, pharmacists, and others at Nihon University Nerima-Hikarigaoka Hospital for their help with this study.
References 1. Platt R, Zalenznik DF, Hopkins CC, et al. Perioperative antibiotic prophylaxis for herniorrhaphy and breast surgery. N Engl J Med 1990; 322:153–60. 2. Yerdel MA, Akin EB, Dolalan S, et al. Effect of single-dose prophylactic ampicillin and sulbactam on wound infection after tension-free inguinal hernia repair with polypropylene mesh: the randomized, double-blind, prospective trial. Ann Surg 2001;233:26–33. 3. Aufenacker TJ, van Geldere D, van Mesdag T, et al. The role of antibiotic prophylaxis in prevention of wound infection after Lichtenstein open mesh repair of primary inguinal hernia: a multicenter doubleblind randomized controlled trial. Ann Surg 2004;240:955–60. 4. Celdra´n A, Frieyro O, de la Pinta JC, et al. The role of antibiotic prophylaxis on wound infection after mesh hernia repair under local anesthesia on an ambulatory basis. Hernia 2004;8:20–2. 5. Oteiza F, Ciga MA, Ortiz H. Profilaxis antibio´tica en la hernioplastia inguinal. Cir Esp 2004;75:69–71. 6. Perez AR, Roxas MF, Hilvano SS. A randomized, double-blind, placebo-controlled trial to determine effectiveness of antibiotic prophylaxis for tension-free mesh herniorrhaphy. J Am Coll Surg 2005; 200:393–7. 7. Tzovaras G, Delikoukos S, Christodoulides G, et al. The role of antibiotic prophylaxis in elective tension-free mesh inguinal hernia repair:
484
8.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18.
results of a single-centre prospective randomised trial. Int J Clin Pract 2007;61:236–9. Jain SK, Jayant M, Norbu C. The role of antibiotic prophylaxis in mesh repair of primary inguinal hernias using prolene hernia system: a randomized prospective double-blind control trial. Trop Doct 2008;38:80–2. Shankar VG, Srinivasan K, Sistla SC, Jagdish S. Prophylactic antibiotics in open mesh repair of inguinal herniada randomized controlled trial. Int J Surg 2010;86:444–7. Ergul Z, Akinci M, Ugurlu C, et al. Prophylactic antibiotic use in elective inguinal hernioplasty in a trauma center. Hernia 2011;16:145–51. Othman I. Prospective randomized evaluation of prophylactic antibiotic usage in patients undergoing tension free inguinal hernioplasty. Hernia 2011;15:309–13. Aufenacker TJ, Koelemay MJ, Gouma DJ, Simons MP. Systematic review and meta-analysis of the effectiveness of antibiotic prophylaxis in prevention of wound infection after mesh repair of abdominal wall hernia. Br J Surg 2006;93:5–10. Sanabria A, Dominguez LC, Valdivieso E, Gomez G. Prophylactic antibiotics for mesh inguinal hernioplasty: a meta-analysis. Ann Surg 2007;245:392–6. Sanchez-Manuel FJ, Lozano-Garcia J, Seco-Gil JL. Antibiotic prophylaxis for hernia repair. Cochrane Database Syst Rev 2007;CD003769. Gravante G, Venditti D, Filingeri V. The role of single-shot antibiotic prophylaxis in inguinal hernia repair: a meta-analysis approach of 4336 patients. Ann Surg 2008;248:496–7. Li JF, Lai DD, Zhang XD, et al. Meta-analysis of the effectiveness of prophylactic antibiotics in the prevention of postoperative complications after tension-free hernioplasty. Can J Surg 2012;55: 27–32. Yin Y, Song T, Liao B, et al. Antibiotic prophylaxis in patients undergoing open mesh repair of inguinal hernia: a meta-analysis. Am Surg 2012;78:359–65. Simons MP, Aufenacker T, Bay-Nielsen M, et al. European Hernia Society guidelines on the treatment of inguinal hernia in adult patients. Hernia 2009;13:343–403.
The American Journal of Surgery, Vol 207, No 4, April 2014 19. Waldvogel FA, Vaudaux PE, Pittet D, Lew PD. Perioperative antibiotic prophylaxis of wound and foreign body infections: microbial factors affecting efficacy. Rev Infect Dis 1991;13:S782–9. 20. Cheek CM, Williams MH, Farndon JR. Trusses in the management of hernia today. Br J Surg 1995;82:1611–3. 21. Mangram AJ, Horan TC, Pearson ML, et al. Hospital Infection Control Practices Advisory Committee. Guideline for prevention of surgical site infection 1999. Infect Control Hosp Epidemiol 1999;20:250–78. 22. DeMets DL, Lan KK. Interim analysis: the alpha spending function approach. Stat Med 1994;13:1341–52. 23. O’Brien PC, Fleming TR. Multiple testing procedure for clinical trials. Biometrics 1979;35:549–56. 24. Kent DM, Rothwell PM, Ioannidis JP, et al. Assessing and reporting heterogeneity in treatment effects in clinical trials: a proposal. Trials 2010;12:85. 25. Taylor EW, Duffy K, Lee K, et al. Surgical site infection after groin hernia repair. Br J Surg 2004;91:105–11. 26. Davey PG, Nathwani D. What is the value of preventing postoperative infections? New Horiz 1998;6:S64–71. 27. Gilbert AI, Felton LL. Infection in inguinal hernia repair considering biomaterials and antibiotics. Surg Gynecol Obstet 1993;177:126–30. 28. Taylor SG, O’Dwyer PJ. Chronic groin sepsis following tension-free inguinal hernioplasty. Br J Surg 1999;86:562–5. 29. Delikoukos S, Tzovaras G, Liakou P, et al. Late-onset deep mesh infection after inguinal hernia repair. Hernia 2007;11:15–7. 30. Rehman S, Khan S, Pervaiz A, Perry EP. Recurrence of inguinal herniae following removal of infected prosthetic meshes: a review of the literature. Hernia 2012;16:123–6.
Appendix Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.amjsurg.2013.01.047.
Clinical Science
A randomized trial of antibiotic prophylaxis for the prevention of surgical site infection after open mesh-plug hernia repair Takero Mazaki, M.D., Ph.D.a,*, Kazunari Mado, M.D.a, Hideki Masuda, M.D., Ph.D.a, Motomi Shiono, M.D., Ph.D.a, Noahiro Tochikura, M.Sc.b, Morio Kaburagi, M.Sc.b Departments of aSurgery, and bPharmacy, Nihon University School of Medicine, Tokyo, Japan
KEYWORDS: Inguinal hernia; Mesh-plug; Repair; Surgical-site infection; Antibiotic prophylaxis
Abstract BACKGROUND: The efficacy of antibiotic prophylaxis for the prevention of surgical-site infection (SSI) after open tension-free inguinal hernia repair remains controversial. METHODS: A double-blind, randomized, placebo-controlled trial was conducted. Patients who underwent elective open mesh-plug hernia repair were eligible for randomization. In the antibiotic prophylaxis group, 1.0 g cefazolin was intravenously administrated 30 minutes before the incision. In the placebo group, an equal volume of sterile saline was administered. The primary end point was the incidence of SSI. RESULTS: A total of 200 patients were enrolled. SSI developed in 2 of 100 patients (2%) in the antibiotic prophylaxis group and 13 of 100 patients (13%) in the placebo group, indicating a significant difference between the 2 groups (relative risk ratio, 0.25; 95% confidence interval, 0.070 to 0.92; P 5 .003). Other complications occurred in 23 patients: 7 (7%) in the antibiotic prophylaxis group and 16 (16%) in the placebo group (P 5 .046). CONCLUSIONS: This study indicates that antibiotic prophylaxis is effective for the prevention of SSI after open mesh-plug hernia repair. Ó 2014 Elsevier Inc. All rights reserved.
Surgeons routinely administer antibiotic prophylaxis when a prosthetic device is involved, for example, in cardiothoracic surgery and in hip arthroplasty, because infection can result in critical conditions such as removal of the prosthesis, prolongation of the hospital stay, and an increase in cost. On the other hand, it is questionable whether the routine use of antibiotic prophylaxis is necessary for the prevention of surgical-site infection (SSI) in a The authors declare no conflicts of interest. * Corresponding author. Tel.: 181-3-3972-8111; fax: 181-3-3972-8750. E-mail address: [email protected] Manuscript received October 28, 2012; revised manuscript January 2, 2013 0002-9610/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjsurg.2013.01.047
simple, clean surgical procedure such as inguinal hernia repair or breast surgery.1 In patients undergoing inguinal hernia repair, 10 randomized controlled trials have been conducted during the past decade2–12 to evaluate the effectiveness of the routine use of antibiotic prophylaxis. Two of these studies2,4 showed that prophylactic antibiotics were effective in reducing infection rates, but 8 of the studies3,5–11 recommended against their use. However, most of the studies had a relatively small number of patients and therefore did not have the statistical power necessary to detect a significant difference. Likewise, 6 meta-analyses failed to show statistically conclusive evidence that prophylactic antibiotics were effective.12–17 This is therefore still a controversial
T. Mazaki et al.
SSI after tension-free inguinal hernia repair
issue. The randomized controlled trials enrolled low-risk patients rather than high-risk patients. On the basis of these results, the European Hernia Society guidelines18 state that there is no indication for the routine use of antibiotic prophylaxis for low-risk adult patients in clinical settings with low infection rates (,5%) but that their use is recommended in the presence of risk factors for SSI, such as recurrence of hernia, advanced age, and immunosuppression. Approximately 160,000 inguinal hernia repairs are performed annually in Japan, and .1 million repairs are performed annually in both the United States and Europe.3 If antibiotic prophylaxis can be avoided in inguinal hernia repairs, we can not only minimize cost but also reduce the risk for allergic side effects and the possible development of bacterial resistance.19 Furthermore, the risk for a clinically significant infection is not large enough to justify the routine use of antibiotic prophylaxis, because SSI develops in only 1% to 4% of patients undergoing inguinal hernia repair.20 In addition, a review of inguinal hernia repairs at our institution, performed with mesh and without mesh, indicated that SSI occurred in approximately 1% of patients (unpublished data). We subsequently found that our surgeons were administering prophylactic antibiotics despite the absence of any evidence of their effectiveness. The above concerns are the main arguments against the routine use of prophylactic antibiotics. We therefore conducted a single-center, randomized, double-blind, placebo-controlled trial to clarify whether antibiotic prophylaxis for primary inguinal hernia surgery reduces morbidity.
Methods Study design The trial was conducted in the Department of General Surgery at the Nihon University School of Medicine from July 2007 to December 2011. The ethics committees at the hospital approved the trial, and all patients provided written informed consent. The trial was registered at ClinicalTrials.gov (NCT00636831).
Inclusion and exclusion criteria Inpatients who were scheduled for elective primary unilateral or bilateral open mesh-plug hernia repair were eligible for the study. Because open inguinal hernia repairs for outpatients were uncommon in Japan (the Japanese Ministry of Health, Labor, and Welfare reported approximately 1.6% of the rate of the surgery in 2011; http://www. e-stat.go.jp/SG1/estat/List.do?lid5000001097372), we excluded outpatients from this study. The exclusion criteria were outpatient status; day surgery; age ,18 years; recurrent hernia; incarcerated or strangulated hernia requiring emergency hernia repair; pregnancy or lactation; earlier history of allergy, sensitivity, or anaphylaxis to b-lactam
477 or cephalosporin antibiotics; antibiotic therapy ,48 hours before surgery; presence of an infection at the time of surgery; cardiac valvular problem; increased risk for infection secondary to a coexisting medical condition; immunosuppression (eg, human immunodeficiency virus infection, malignancy, or chemotherapy); American Society of Anesthesiologists (ASA) grade . IV; and refusal to participate in the study. If patients had diabetes mellitus, we controlled the value of glycosylated hemoglobin at ,6.5%.
Randomization Patients were randomly assigned on admission in a double-blinded manner to either the antibiotic prophylaxis group or the placebo group. All surgeons and other staff members were blinded to randomization and to patients’ details. According to a computer-generated list in blocks of 50 patients, a pharmacist carried out randomization.
Surgical technique A certified surgeon or surgical resident who was blinded to study group assignment performed surgery under general and local anesthesia, and this was standardized among surgeons. The skin was shaved just before surgery and prepared using 10% povidone iodine. All patients underwent elective open mesh-plug hernia repair using a monofilament polypropylene mesh (PerFix Plug; CR Bard, Cranston, RI). The mesh-plug was sutured in place using 2-0 monofilament polypropylene sutures (Polysorb; Covidien, MI). Any subcutaneous suture was not used. The skin was closed with interrupted 2-0 nylon sutures (Alfresa Pharma, Ltd, Tokyo, Japan). No drain was placed. This was standardized among surgeons.
Intervention The antibiotic prophylaxis group received 100 mL sterile saline with 1.0 g cefazolin (Astellas Pharma, Ltd, Tokyo, Japan) by continuous intravenous infusion. An equal volume of sterile saline (Otsuka Pharmaceutical Factory Ltd, Tokushima, Japan) was administrated to the placebo group 30 minutes before the incision.
Follow-up Although a type of wound dressing was not standardized in the hospital, surgeons uniformly removed the dressing and suture 7 to 8 days after surgery (the first follow-up visit) in the outpatient clinic. The second and third follow-up visits were at 1 and 3 months after surgery. All wounds were carefully examined by 2 certified surgeons, who did not perform the operations. According to the most recent criteria from the Centers for Disease Control and Prevention, wound infection was categorized as superficial SSI or deep SSI (DSSI).21 These criteria define superficial SSI as occurring
478 ,30 days after surgery and involving only the skin or the subcutaneous tissue at the incision site. DSSI occurs ,1 year after surgery if a prosthesis is in place and the infection involves deep soft tissues at the incision site. The patients’ demographic data were recorded: body mass index (obesity was defined as .30 kg/m2 according to the World Health Organization classification), type of hernia (direct, indirect, and combined), hernia localization (right, left, and bilateral), coexisting conditions (eg, diabetes mellitus, heart disease, neurologic disorders, pulmonary diseases, and others), ASA grade, surgeon (certified and resident), duration of procedure, length of incision, and length of hospital stay.
The American Journal of Surgery, Vol 207, No 4, April 2014 this analysis, we assessed whether the relative treatment effect was consistent across the spectrum of SSI risk. We performed additional subgroup analysis on the following prespecified characteristics: age, sex, type of hernia, body mass index, location of hernia, grade of surgeon, duration of procedure, ASA grade, and coexisting diseases. All subgroup statistical analyses were evaluated for interaction effects with antibiotic prophylaxis by testing for the significance of a corresponding interaction term in a multiple logistic regression analysis. Data were analyzed using Stata version 12.1 (StataCorp LP, College Station, TX).
Results Statistical analysis Continuous, normally distributed variables are presented as mean 6 SD and skewed variables as medians with interquartile ranges. Student’s t tests and Wilcoxon’s ranksum tests were used to analyze continuous variables among the categories. Categorical variables are presented as absolute numbers and percentages. Comparative analysis of proportions was performed using chi-square or Fisher’s exact tests. The primary and secondary outcomes are reported in terms of relative risk ratios and relative risk reduction. The primary end point was the incidence of SSI during the 3-month period after surgery in the antibiotic prophylaxis and placebo groups. All statistics were based on the intention-to-treat principle. We based the sample size calculation on 2 randomized controlled trials.2,3 The target sample size of 400 gave 80% power, with a P value ,.05, to detect a difference in the occurrence of SSI between the antibiotic prophylaxis and placebo groups, with a 6% loss of patients, with the assumption that the occurrence of SSI in the antibiotic prophylaxis and the placebo group was 1% and 7%, respectively.2,3 We planned 4 interim analyses after every 100 patients had been randomized, with a final 2-sided P value ,.031 indicating statistical significance. This P value was based on the Lan-DeMets method and the O’Brien-Fleming a spending function.22,23 A data and safety monitoring board provided regulatory oversight by reviewing blinded subject data quarterly and conducting the a priori scheduled interim analysis. The secondary end point, which was the incidence of any other complications in the study groups during the 3 months after surgery, was similarly calculated, with a P value ,.05 indicating statistical significance. Adverse events that were potentially attributable to the study antibiotics were also reported. According to a previously proposed framework for evaluating the heterogeneity of treatment effects on the primary end point,24 we performed a post hoc analysis of the patients’ data according to their risk factors for SSI. The risk factors were as follows: age R70 years, body mass index R30 kg/m2, diabetes mellitus, ASA grade R II, procedure duration . 60 minutes, bilateral hernia, and recurrence of hernia.3,13,17,18 Individual patient risk scores were determined by assigning 1 point for the presence of each risk factor. In
Patients Based on the results of the first interim analysis, the data and safety monitoring board recommended a second
Table 1
Baseline and operative characteristics*
Characteristic Age (y) Men Body mass index (kg/m2) Type of hernia Direct Indirect Combined Unknown Hernia characteristics Right Left Bilateral Coexisting conditions† Diabetes mellitus Heart disease Neurologic disease Pulmonary disease Other Mixed American Society of Anesthesiologists health status grade I II III Surgeon Resident Certified surgeon Duration of surgery (min) Length of incision (cm) Length of hospital stay (d)
Antibiotic group (n 5 100)
Placebo group (n 5 100)
69 (57–76) 89 23.0 6 2.8
72 (60–77) 94 22.7 6 3.0
30 67 2 1
26 67 7 0
61 34 5
50 42 8
6 24 1 3 2 14
7 31 2 7 0 21
52 43 5
37 58 5
84 83 16 17 66.3 6 25.4 65.2 6 27.1 5.2 6 0.7 5.1 6 0.7 3 (2–3) 3 (2–3)
*There were no significant differences in the baseline characteristics of the antibiotic and placebo groups. † Some patients were counted twice if they had .1 complication.
T. Mazaki et al.
SSI after tension-free inguinal hernia repair
interim analysis after an additional 100 patients had been enrolled. In December 2011, the board assessed the outcomes of the second interim analysis and recommended that the study be terminated early on the basis of the harm of placebo compared with antibiotic prophylaxis. Thus, according to the prior stopping rule, we stopped the study. A total of 100 patients received antibiotics, and 100 patients received placebo (Fig. 1). Follow-up of all patients for both primary and secondary outcomes was completed.
Study outcomes A total of 33 surgeons (23 residents and 10 certified) participated in this study. Residents performed 83% of all operations, and all infections occurred in operations performed by residents. There was no significant difference in any baseline characteristic between the 2 groups, with the exception of the incidence of SSI and any other complications (Table 1). The primary outcome, SSI, occurred in 15 of 200 patients (7.5%), with 2 of 100 (2%) occurring in the antibiotic prophylaxis group and 13 of 100 (13%) occurring in
Figure 1
479 the placebo group. This corresponded to a relative risk ratio of 0.25 (95% confidence interval, 0.07 to 0.92; P 5 .006) and an absolute risk reduction of 11 percentage points (number needed to treat to prevent 1 episode of SSI, 9) (Supplementary Figure) and a relative risk reduction of 85% (Fig. 2A). At this interim analysis, the Bayesian prediction probability of the effectiveness of antibiotic prophylaxis was 99.9%. A majority of cases of SSI were superficial SSI, and no DSSI was found in either group. One late-onset infection was recognized 1 year 9 months after surgery in the placebo group. However, this case was excluded from the primary end point according to the Centers for Disease Control and Prevention inclusion criteria for surgical wound infections. Secondary outcomes occurred in 23 patients: 7 (7%) in the antibiotic prophylaxis group and 16 (16%) in the placebo group (P 5 .046; Fig. 2B). The secondary end points were follows: 4 hematomas, 1 pneumonia, 2 seromas, and 1 urinary retention in the antibiotic prophylaxis group and 1 acute pancreatitis, 7 hematomas, 1 pneumonia, 1 orchitis, 6 seromas, and 1 urinary tract infection in the placebo group (Table 2). Regarding recurrence of hernia, we
Flow of patients through the trial. PHS 5 Prolene hernia system.
480
The American Journal of Surgery, Vol 207, No 4, April 2014
A
B
Figure 2 Incidence of primary and secondary outcomes. (A) SSI developed in 13 of 100 patients (13%) in the placebo group and 2 of 100 patients (2%) in the antibiotic prophylaxis group (P 5 .003). (B) Other complications developed in 16 of 100 patients (16%) in the placebo group and 7 of 100 patients (7%) in the placebo group (P 5 .046).
documented 2 recurrences (2%) in the antibiotic prophylaxis group and 1 recurrence (1%) in the placebo group (Table 2). There were no intraoperative complications, and neither adverse events nor mortality was seen in either group. The median length of hospital stay was 3 days for each group. History of smoking was not investigated.
Heterogeneity of treatment effects According to the patients’ risk scores, the relative efficacy of antibiotic prophylaxis for the prevention of SSI did not vary significantly. The absolute risk reduction was a number needed to treat of 9 for those with risk scores ,2 (%2 risk factors) compared with a number needed to treat of 9 for those with risk scores .3 (R3 risk factors) (Fig. 3).
No patients required removal of the mesh in either group. All patients recovered completely.
Comments This study indicates that antibiotic prophylaxis is effective for the prevention of SSI after open mesh-plug hernia repair, reducing the incidence of SSI from 13% in the placebo group to 2% in the antibiotic prophylaxis group. The number needed to treat to prevent 1 episode of SSI was 9. Moreover, we found that antibiotic prophylaxis decreased the incidence of other complications. Although this study showed that the incidence of SSI occurred 13% in the placebo group, the incidence is not extremely high compared with previous results. Four randomized controlled trials2,4,9,11 assessed the efficacy of
Exploratory subgroup analysis Table 2
The beneficial effect of antibiotic prophylaxis on the primary outcome was also consistent across the other prespecified subgroups, with the exception of female sex (Fig. 4). Antibiotic prophylaxis was protective against SSI in all subgroups.
Details of the infected patients All infected wounds were diagnosed after hospital discharge. Two patients in the antibiotic prophylaxis group were treated with drainage alone of the infected wound. Eleven patients were treated with oral antibiotics and wound drainage, and the remaining 4 patients treated with only drainage (Table 3). One patient with a bilateral hernia and 14 patients with unilateral hernias had SSIs.
Incidence of postoperative complications
End point Any wound infections Deep infection Superficial infection Any other complications Acute pancreatitis Hematoma formation Orchitis Pneumonia Seroma formation Urinary retention Urinary tract infection Recurrence Mortality
Antibiotic group (n 5 100)
Placebo group (n 5 100)
0 2
0 13
0 4 0 1 2 1 0 2 0
1 7 1 1 6 0 1 1 0
T. Mazaki et al.
SSI after tension-free inguinal hernia repair
481
Figure 3 The relative treatment effect was consistent across the spectrum of SSI risk. An individual patient’s risk score was determined by assigning 1 point for the presence of each risk factor. The risk score (range, 0 to 4) was divided into 2 categories and evaluated using multiple logistic regression. Analysis revealed no evidence that the relative risk reduction associated with antibiotic prophylaxis varied across the risk groups (P 5 .60). The difference in the relative risk reduction between the groups was not significant (P 5 .71 for the Mantel-Haenszel test for homogeneity). The red vertical line shows the overall risk reduction, and the dashed vertical line indicates no relative risk reduction.
prophylactic antibiotic use after open inguinal hernia repair and found that the incidence varied from 8.2% to 12.5% in the placebo group. A large Scottish study25 evaluated the efficacy of telephone interviews in postdischarge surveillance for SSI after an inguinal hernia repair, demonstrated that the incidence varied from 0% to 14.6% among hospitals. Our study included patients with more risk factors (aged .70 years, bilateral hernia, diabetes mellitus, ASA grade III, and operation duration .60 minutes) than other randomized controlled trials. Therefore, we believe that these factors reflect our 13% SSI incidence in the placebo group in this study. Residents operated on 83% of all patients, and all infections occurred in operations performed by residents in this study. We have no idea whether our 7.5% infection incidence reflects a common problem in training centers, at which residents perform procedures. However, 2 randomized controlled trials3,9 including 43% and 89% of patients operated on by residents found that the grade of surgeon was not a significant risk factor for SSI. Also, Taylor et al25 concluded that surgeon’s experience did not influence the incidence of SSI. On the other hand, 13% of patients had bilateral hernias, and 1 of those had SSI in this study. In addition to our study, 3 randomized controlled trials3,9,11 in which 6%, 8%, and 11% of all patients had bilateral hernias found no correlation between a bilateral hernia and the incidence of SSI. A previous report stated that antibiotic prophylaxis should be avoided in low-risk patients, because the incidence of SSI is as low as 2% to 3%, and infections can be treated in a straightforward manner with wound drainage plus antibiotic administration.2 The selective use of prophylactic antibiotics for low-risk patients is a matter of debate for several reasons. First, we underestimate the incidence of SSI. This is primarily because the incidence of SSI is influenced by patient-related risk factors, surgery-related risk factors, and the sanitary conditions of the institution, which are not the same for all patients and all institutions. Indeed, a large study
reported that the incidence of SSI varied from 0% to 14.6%25 among hospitals. This indicates that the true incidence of SSI can be quite high, and we could expect a benefit from antibiotic prophylaxis for all patients. Second, a recent meta-analysis17 showed significant effectiveness of antibiotic prophylaxis (odds ratio, 0.54; 95% confidence interval, 0.37 to 0.81), although most patients included in the metaanalysis were at low risk. This study indicates that antibiotic prophylaxis is significantly effective for low-risk patients. Third, as shown in Figs. 3 and 4, the relative efficacy of antibiotic prophylaxis did not vary according to risk factors such as patient’s risk score, ASA grade, and the presence of coexisting conditions. Consequently, regardless of the incidence of SSI and the presence of low-risk patients, we believe that the routine use of antibiotic prophylaxis provides benefit. One may wonder if patients remain at increased risk for SSI because they are in the hospital, but this explanation is inappropriate. Taylor et al25 reported that the incidence of SSI in inpatients and in outpatients was 5.4% and 5.0%, respectively. Also, the mean incidence of SSI in 5 randomized controlled trials2,8–11 including inpatients was approximately 6.5%; this is almost the same as the result of Taylor et al.25 Therefore, there are no difference in the incidence of SSI between outpatients and inpatients. Our study was not designed to answer the question of the cost-effectiveness of prophylactic antibiotic use. Aufenacker et al,3 using a database from the Netherlands, demonstrated that avoiding antibiotic prophylaxis would save approximately V10 million in the United States and Europe. In contrast, a study conducted to assess postoperative infection-related costs found that annual expenses for infections after inguinal hernia repair ($44,800 per patient) were similar to those for colon surgery ($48,440 per patient).26 These studies show that infection constitutes an important problem in terms of health care costs. We believe that prophylactic antibiotics are effective from an economic point of view. However, the potential benefits of antibiotics were not carefully analyzed for cost-effectiveness.
482
The American Journal of Surgery, Vol 207, No 4, April 2014
Figure 4 Exploratory subgroup analysis. The primary end point was consistent across the prespecified and post hoc subgroups. There was no significant interaction between the antibiotic prophylaxis group and any of these subgroups. The relative risk reduction was not indicated for the certified surgeon and presence of diabetes mellitus groups because SSI was not seen in either category. The red vertical line shows the overall risk reduction, and the dashed vertical line indicates no relative risk reduction. BMI 5 body mass index.
DSSI is thought to be a risk for the recurrence of inguinal hernia, whereas some studies have found that SSI does not increase the recurrence of hernia.4,10,27 Actually, 3 recurrences of hernia in our study did not become infected. According to Aufenacker et al,3 who found that the incidence of DSSI ranged from 0.2% to 0.4%, the rare DSSI had a markedly low recurrence rate of inguinal hernia.
This is a controversial issue. A meta-analysis that looked at the effectiveness of antibiotic prophylaxis for the prevention of DSSI showed no efficacy of antibiotics for the prevention of DSSI (odds ratio, 0.50; 95% confidence interval, 0.12 to 2.09). The meta-analysis, however, did not have the power to detect the efficacy of prophylactic antibiotics, as the power was 10%. These findings therefore should be
T. Mazaki et al. Table 3
SSI after tension-free inguinal hernia repair
483
Details of surgical-site infections in the antibiotic prophylaxis group and placebo group
Group
Days after surgery
Microorganism cultured
Treatment
Outcome
Antibiotic
15 10 11 9
Culture not performed Culture not performed Staphylococcus aureus Enterococcus faecalis, coagulase-negative Staphylococcus Coagulase-negative Staphylococcus, Corynebacterium spp, E faecalis Staphylococcus aureus, coagulase-negative Staphylococcus S aureus Coagulase-negative Staphylococcus Coagulase-negative Staphylococcus Pseudomonas aeruginosa, S aureus, coagulase-negative Staphylococcus Methicillin-resistant S aureus Culture not performed Coagulase-negative Staphylococcus S aureus S aureus, Corynebacterium spp
Drainage Drainage Antibiotics and drainage Antibiotics and drainage
Recovered Recovered Recovered Recovered
Drainage
Recovered
Antibiotics and drainage
Recovered
Antibiotics and drainage Drainage Drainage Antibiotics and drainage
Recovered Recovered Recovered Recovered
Antibiotics Antibiotics Drainage Antibiotics Antibiotics
Recovered Recovered Recovered Recovered Recovered
Placebo
7 10 9 12 12 12 5 7 8 7 9
interpreted cautiously. Taylor et al25 concluded that DSSI frequently necessitated complete removal of the mesh, indicating that management of DSSI is difficult and eventually required repeat hospital admission.6 Whether the placement of a drain can prevent SSI is controversial. Yerdel et al2 showed that the use of drains increased the risk for SSI, whereas Aufenacker et al3 and Tzovaras et al7 demonstrated no increased risk. These randomized controlled trials gave contradictory results. Guidelines, therefore, recommend that drains should be used as indicated for patients with increased blood loss and coagulopathies.18 Late-onset deep infection occurring years after surgery is a rare complication, with reports of occurrence ranging from 0.03% to 1.4%.28–30 We experienced 1 late-onset infection in the placebo group that occurred 1 year 9 months after the operation. The patient did not require mesh removal and recovered completely with antibiotic administration. A retrospective study showed that the incidence of DSSI did not correlate with the administration of antibiotic prophylaxis, the presence of previous SSI, the type of mesh, or the incidence of recurrence of hernia.29 Although the etiology has not yet been elucidated, this type of infection may likely become increasingly common with the widespread use of prosthetic mesh.29
Conclusions The routine use of antibiotic prophylaxis significantly reduced the incidence of not only SSI but also any other complications, whether patients had high-risk or low-risk factors. Simultaneously, there were no significant interaction terms between antibiotic prophylaxis and any subgroups. We therefore conclude that it is appropriate to use
and drainage and drainage and drainage and drainage
routine antibiotic prophylaxis for the prevention of SSI in patients who undergo in-hospital inguinal hernia repair. Further study is needed to answer questions such as what constitutes a real risk factor, the cost-effectiveness of prophylactic antibiotics, and late-onset infection, among other questions.
Acknowledgment We would like to thank the surgeons, residents, pharmacists, and others at Nihon University Nerima-Hikarigaoka Hospital for their help with this study.
References 1. Platt R, Zalenznik DF, Hopkins CC, et al. Perioperative antibiotic prophylaxis for herniorrhaphy and breast surgery. N Engl J Med 1990; 322:153–60. 2. Yerdel MA, Akin EB, Dolalan S, et al. Effect of single-dose prophylactic ampicillin and sulbactam on wound infection after tension-free inguinal hernia repair with polypropylene mesh: the randomized, double-blind, prospective trial. Ann Surg 2001;233:26–33. 3. Aufenacker TJ, van Geldere D, van Mesdag T, et al. The role of antibiotic prophylaxis in prevention of wound infection after Lichtenstein open mesh repair of primary inguinal hernia: a multicenter doubleblind randomized controlled trial. Ann Surg 2004;240:955–60. 4. Celdra´n A, Frieyro O, de la Pinta JC, et al. The role of antibiotic prophylaxis on wound infection after mesh hernia repair under local anesthesia on an ambulatory basis. Hernia 2004;8:20–2. 5. Oteiza F, Ciga MA, Ortiz H. Profilaxis antibio´tica en la hernioplastia inguinal. Cir Esp 2004;75:69–71. 6. Perez AR, Roxas MF, Hilvano SS. A randomized, double-blind, placebo-controlled trial to determine effectiveness of antibiotic prophylaxis for tension-free mesh herniorrhaphy. J Am Coll Surg 2005; 200:393–7. 7. Tzovaras G, Delikoukos S, Christodoulides G, et al. The role of antibiotic prophylaxis in elective tension-free mesh inguinal hernia repair:
484
8.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18.
results of a single-centre prospective randomised trial. Int J Clin Pract 2007;61:236–9. Jain SK, Jayant M, Norbu C. The role of antibiotic prophylaxis in mesh repair of primary inguinal hernias using prolene hernia system: a randomized prospective double-blind control trial. Trop Doct 2008;38:80–2. Shankar VG, Srinivasan K, Sistla SC, Jagdish S. Prophylactic antibiotics in open mesh repair of inguinal herniada randomized controlled trial. Int J Surg 2010;86:444–7. Ergul Z, Akinci M, Ugurlu C, et al. Prophylactic antibiotic use in elective inguinal hernioplasty in a trauma center. Hernia 2011;16:145–51. Othman I. Prospective randomized evaluation of prophylactic antibiotic usage in patients undergoing tension free inguinal hernioplasty. Hernia 2011;15:309–13. Aufenacker TJ, Koelemay MJ, Gouma DJ, Simons MP. Systematic review and meta-analysis of the effectiveness of antibiotic prophylaxis in prevention of wound infection after mesh repair of abdominal wall hernia. Br J Surg 2006;93:5–10. Sanabria A, Dominguez LC, Valdivieso E, Gomez G. Prophylactic antibiotics for mesh inguinal hernioplasty: a meta-analysis. Ann Surg 2007;245:392–6. Sanchez-Manuel FJ, Lozano-Garcia J, Seco-Gil JL. Antibiotic prophylaxis for hernia repair. Cochrane Database Syst Rev 2007;CD003769. Gravante G, Venditti D, Filingeri V. The role of single-shot antibiotic prophylaxis in inguinal hernia repair: a meta-analysis approach of 4336 patients. Ann Surg 2008;248:496–7. Li JF, Lai DD, Zhang XD, et al. Meta-analysis of the effectiveness of prophylactic antibiotics in the prevention of postoperative complications after tension-free hernioplasty. Can J Surg 2012;55: 27–32. Yin Y, Song T, Liao B, et al. Antibiotic prophylaxis in patients undergoing open mesh repair of inguinal hernia: a meta-analysis. Am Surg 2012;78:359–65. Simons MP, Aufenacker T, Bay-Nielsen M, et al. European Hernia Society guidelines on the treatment of inguinal hernia in adult patients. Hernia 2009;13:343–403.
The American Journal of Surgery, Vol 207, No 4, April 2014 19. Waldvogel FA, Vaudaux PE, Pittet D, Lew PD. Perioperative antibiotic prophylaxis of wound and foreign body infections: microbial factors affecting efficacy. Rev Infect Dis 1991;13:S782–9. 20. Cheek CM, Williams MH, Farndon JR. Trusses in the management of hernia today. Br J Surg 1995;82:1611–3. 21. Mangram AJ, Horan TC, Pearson ML, et al. Hospital Infection Control Practices Advisory Committee. Guideline for prevention of surgical site infection 1999. Infect Control Hosp Epidemiol 1999;20:250–78. 22. DeMets DL, Lan KK. Interim analysis: the alpha spending function approach. Stat Med 1994;13:1341–52. 23. O’Brien PC, Fleming TR. Multiple testing procedure for clinical trials. Biometrics 1979;35:549–56. 24. Kent DM, Rothwell PM, Ioannidis JP, et al. Assessing and reporting heterogeneity in treatment effects in clinical trials: a proposal. Trials 2010;12:85. 25. Taylor EW, Duffy K, Lee K, et al. Surgical site infection after groin hernia repair. Br J Surg 2004;91:105–11. 26. Davey PG, Nathwani D. What is the value of preventing postoperative infections? New Horiz 1998;6:S64–71. 27. Gilbert AI, Felton LL. Infection in inguinal hernia repair considering biomaterials and antibiotics. Surg Gynecol Obstet 1993;177:126–30. 28. Taylor SG, O’Dwyer PJ. Chronic groin sepsis following tension-free inguinal hernioplasty. Br J Surg 1999;86:562–5. 29. Delikoukos S, Tzovaras G, Liakou P, et al. Late-onset deep mesh infection after inguinal hernia repair. Hernia 2007;11:15–7. 30. Rehman S, Khan S, Pervaiz A, Perry EP. Recurrence of inguinal herniae following removal of infected prosthetic meshes: a review of the literature. Hernia 2012;16:123–6.
Appendix Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.amjsurg.2013.01.047.
