Catherine Cluver, Natalia Novikova, David OA Eriksson, Kevin Bengtsson, Göran K Lingman

Genital Chlamydia trachomatis (C.trachomatis) infection may lead to pregnancy complications such as miscarriage, preterm labour, low birthweight, preterm rupture of membranes, increased perinatal mortality, postpartum endometritis, chlamydial conjunctivitis and C.trachomatis pneumonia.This review supersedes a previous review on this topic. To establish the most efficacious and best-tolerated therapy for treatment of genital chlamydial infection in preventing maternal infection and adverse neonatal outcomes. We searched the Cochrane Pregnancy and Childbirth Group's Trials Register, ClinicalTrials.gov, the WHO International Clinical Trials Registry Platform (ICTRP) (26 June 2017) and reference lists of retrieved studies. Randomised controlled trials (RCTs) as well as studies published in abstract form assessing interventions for treating genital C.trachomatis infection in pregnancy. Cluster-RCTs were also eligible for inclusion but none were identified. Quasi-randomised trials and trials using cross-over design are not eligible for inclusion in this review. Two review authors independently assessed studies for inclusion, assessed trial quality and extracted the data using the agreed form. Data were checked for accuracy. Evidence was assessed using the GRADE approach. We included 15 trials (involving 1754 women) although our meta-analyses were based on fewer numbers of studies/women. All of the included studies were undertaken in North America from 1982 to 2001. Two studies were low risk of bias in all domains, all other studies had varying risk of bias. Four other studies were excluded and one study is ongoing.Eight comparisons were included in this review; three compared antibiotic (erythromycin, clindamycin, amoxicillin) versus placebo; five compared an antibiotic versus another antibiotic (erythromycin, clindamycin, amoxicillin, azithromycin). No study reported different antibiotic regimens. Microbiological cure (primary outcome) Antibiotics versus placebo: Erythromycin (average risk ratio (RR) 2.64, 95% confidence interval (CI) 1.60 to 4.38; two trials, 495 women; I(2) = 68%; moderate-certainty evidence), and clindamycin (RR 4.08, 95% CI 2.35 to 7.08; one trial, 85 women;low-certainty evidence) were associated with improved microbiological cure compared to a placebo control. In one very small trial comparing amoxicillin and placebo, the results were unclear, but the evidence was graded very low (RR 2.00, 95% CI 0.59 to 6.79; 15 women). One antibiotic versus another antibiotic: Amoxicillin made little or no difference in microbiological cure in comparison to erythromycin (RR 0.97, 95% CI 0.93 to 1.01; four trials, 466 women; high-certainty evidence), probably no difference compared to clindamycin (RR 0.96, 95% CI 0.89 to 1.04; one trial, 101 women; moderate-quality evidence), and evidence is very low certainty when compared to azithromycin so the effect is not certain (RR 0.89, 95% CI 0.71 to 1.12; two trials, 144 women; very low-certainty evidence). Azithromycin versus erythromycin (average RR 1.11, 95% CI 1.00 to 1.23; six trials, 374 women; I(2) = 53%; moderate-certainty evidence) probably have similar efficacy though results appear to favour azithromycin. Clindamycin versus erythromycin (RR 1.06, 95% CI 0.97 to 1.15; two trials, 173 women; low-certainty evidence) may have similar numbers of women with a microbiological cure between groups.Evidence was downgraded for design limitations, inconsistency, and imprecision in effect estimates. Side effects of the treatment (maternal) (secondary outcome) Antibiotics versus placebo: side effects including nausea, vomiting, and abdominal pain, were reported in two studies (495 women) but there was no clear evidence whether erythromycin was associated with more side effects than placebo and a high level of heterogeneity (I(2) = 78%) was observed (average RR 2.93, 95% CI 0.36 to 23.76). There was no clear difference in the number of women experiencing side effects when clindamycin was compared to placebo in one small study (5/41 versus 1/44) (RR 6.35, 95% CI 0.38 to 107.45, 62 women). The side effects reported were mostly gastrointestinal and also included resolving skin rashes. One antibiotic versus another antibiotic: There was no clear difference in incidence of side effects (including nausea, vomiting, diarrhoea and abdominal pain) when amoxicillin was compared to azithromycin based on data from one small study (36 women) (RR 0.56, 95% CI 0.24 to 1.31).However, amoxicillin was associated with fewer side effects compared to erythromycin with data from four trials (513 women) (RR 0.31, 95% CI 0.21 to 0.46; I(2) = 27%). Side effects included nausea, vomiting, diarrhoea, abdominal cramping, rash, and allergic reaction.Both azithromycin (RR 0.24, 95% CI 0.17 to 0.34; six trials, 374 women) and clindamycin (RR 0.44, 95% CI 0.22 to 0.87; two trials, 183 women) were associated with a lower incidence of side effects compared to erythromycin. These side effects included nausea, vomiting, diarrhoea and abdominal cramping.One small study (101 women) reported there was no clear difference in the number of women with side effects when amoxicillin was compared with clindamycin (RR 0.57, 95% CI 0.14 to 2.26; 107 women). The side effects reported included rash and gastrointestinal complaints. Other secondary outcomes Single trials reported data on repeated infections, preterm birth, preterm rupture of membranes, perinatal mortality and low birthweight and found no clear differences between treatments.Many of this review's secondary outcomes were not reported in the included studies. Treatment with antibacterial agents achieves microbiological cure from C.trachomatis infection during pregnancy. There was no apparent difference between assessed agents (amoxicillin, erythromycin, clindamycin, azithromycin) in terms of efficacy (microbiological cure and repeat infection) and pregnancy complications (preterm birth, preterm rupture of membranes, low birthweight). Azithromycin and clindamycin appear to result in fewer side effects than erythromycin.All of the studies in this review were conducted in North America, which may limit the generalisability of the results. In addition, study populations may differ in low-resource settings and these results are therefore only applicable to well-resourced settings. Furthermore, the trials in this review mainly took place in the nineties and early 2000's and antibiotic resistance may have changed since then.Further well-designed studies, with appropriate sample sizes and set in a variety of settings, are required to further evaluate interventions for treating C.trachomatis infection in pregnancy and determine which agents achieve the best microbiological cure with the least side effects. Such studies could report on the outcomes listed in this review.