Network meta-analysis of antibiotic prophylaxis for prevention of surgical-site infection after groin hernia surgery

Abstract

Background: First-generation cephalosporins (such as cefazolin) are recommended as antibiotic prophylaxis in groin hernia repair, but other broad-spectrum antibiotics have also been prescribed in clinical practice. This was a systematic review and network meta-analysis to compare the efficacy of different antibiotic classes for prevention of surgical-site infection (SSI) after hernia repair.

Methods: RCTs were identified that compared efficacy of antibiotic prophylaxis on SSI after inguinal or femoral hernia repair from PubMed and Scopus databases up to March 2016. Data were extracted independently by two reviewers. Network meta-analysis was applied to assess treatment efficacy. The probability of being the best antibiotic prophylaxis was estimated using surface under the cumulative ranking curve (SUCRA) analysis.

Results: Fifteen RCTs (5159 patients) met the inclusion criteria. Interventions were first-generation (7 RCTs, 1237 patients) and second-generation (2 RCTs, 532) cephalosporins, β-lactam/β-lactamase inhibitors (6 RCTs, 619) and fluoroquinolones (2 RCTs, 581), with placebo as the most common comparator (14 RCTs, 2190). A network meta-analysis showed that β-lactam/β-lactamase inhibitors and first-generation cephalosporins were significantly superior to placebo, with a pooled risk ratio of 0·44 (95 per cent c.i. 0·25 to 0·75) and 0·62 (0·42 to 0·92) respectively. However, none of the antibiotic classes was significantly different from the others. SUCRA results indicated that β-lactam/β-lactamase inhibitors and first-generation cephalosporins were ranked first and second respectively for best prophylaxis.

Conclusion: β-Lactam/β-lactamase inhibitors followed by first-generation cephalosporins ranked as the most effective SSI prophylaxis for adult patients undergoing groin hernia repair.

Figure 1

Flow diagram showing selection of articles for review. SSI, surgical‐site infection

Figure 2

Network plot of uses of antibiotic prophylaxis for reducing surgical‐site infection after hernioplasty. Nodes are weighted by number of studies; edges are weighted by numbers of included subjects. Numbers of studies/subjects are shown

Figure 3

Rankograms showing use of antibiotic prophylaxis for reducing surgical‐site infection after hernioplasty: a first‐generation cephalosporins, b second‐generation cephalosporins, c β‐lactam/β‐lactamase inhibitors, d fluoroquinolones and e placebo

Figure 4

Predictive interval plots for antibiotic prophylaxis network of groin hernioplasty. Values in parentheses are 95 per cent confidence intervals, followed by 95 per cent predictive intervals that take future uncertainty into account. These are also plotted as bold and extended lines respectively. The dashed line indicates no treatment effect (risk ratio = 1·00)