A network meta-analysis of prophylactic antibiotics for preventing post-cesarean endometritis

ID: 

90

Session: 

Poster session 5

Date: 

Thursday 27 October 2016 - 10:30 to 11:00

Location: 

All authors in correct order:

Nozaki T1, Kashiwabara K1, Shinozaki T1, Oba K1, Matsuyama Y1, Ota E2, Mori R3
1 Department of Biostatistics, School of Public Health, University of Tokyo, Japan
2 Graduate School of Nursing Sciences, Global Health Nursing, St. Luke's International University, Japan
3 Department of Health Policy, National Center for Child Health and Development, Japan
Presenting author and contact person

Presenting author:

Takanari Nozaki

Contact person:

Abstract text
Background: The caesarean section rate is increasing all over the world. Endometritis is the most common infection-related complication following caesarean delivery.

Objectives: To compare the effect of prophylactic antibiotics for preventing post-cesarean endometritis via a network meta-analysis.

Methods: Referring the same eligibility criteria as the Cochrane Review 'Different classes of antibiotics given to women routinely for preventing infection at caesarean section' (Gyte 2014), we included 19 randomized controlled trials comparing penicillins and cephalosporins given to women undergoing caesarean section and excluded trials that had compared antibiotics with placebo or antibiotics within the same class. In this study, six kinds of antibiotics were compared: first generation (C1), second generation (C2), and third generation (C3) for cephalosporins; penicillinase-resistants (P1), anti-Pseudomonas aeruginosas (P2), and extended spectrums (P3) for penicillins. After examining inconsistency between direct and indirect comparisons using the Lu-Ades model, C1 was chosen as the control group and odds ratios (ORs) and 95% confidence intervals (95% CI) for endometritis incidence of other antibiotics (C2, C3, P1, P2 and P3) were estimated in a random-effects model.

Results: A total of 5606 women participated, and experienced a 9.1% incidence of endometritis. The network of trials is shown in Fig 1. Statistically significant inconsistency of a network was not observed (P = 0.67). The observed preventive effects were shown in Fig 2 and ranked as P2 > C2 > P3 > C1 (reference) > C3 > P1. In particular, the effect of P2 was significantly higher than that of C1 (OR 0.58; 95% CI 0.38 to 0.87); similarly, the effect of C2 was significantly higher than that of C1 (OR 0.71; 95% CI 0.54 to 0.93).

Conclusions: The effect of second generation cephalosporins or anti-Pseudomonas aeruginosa penicillin on the prevention of post-caesarean endometritis was significantly higher than that of first generation cephalosporins, which differs from current obstetrical care. This may have a large affect on future decision making in obstetric care.

Attachments: