2 million pregnant women, mostly in low and middle-income countries, have syphilis in pregnancy resulting in adverse outcomes in 69% of cases. Given known low costs of screening and treatment, the figures are appalling. Is it really lack of evidence for the effectiveness of interventions that is holding things back? Yet one can only support the aims of a recent systematic review in the Lancet in seizing any opportunity to attract wider public attention to this issue.
The problem for the reviewers, however, is that randomised controlled trials providing the kind of evidence the authors require would no longer be ethical. Given the low-grade nature of the available evidence, there has been a tendency to conclude that there existed no intervention studies showing the effect of interventions on preterm birth. Not content with this, the authors of this review return again to the evidence in order to glean whatever may be gleaned.
The resulting 10 studies are very heterogeneous. As for the interventions themselves, it is evidently hard to generalize beyond the important common features of decentralized testing and same-day treatment. The outcomes are similarly varied and include, in various combinations: i. proportion of women receiving ante-natal screening (5 studies); ii. proportion of women receiving at least one dose of penicillin (6); iii. infants born with congential syphilis (4); iv. peri-natal deaths (3); v. stillbirths (3). Evidence of increased uptake of testing and treatment (i. and ii.) is disappointingly inconclusive; but evidence of adverse pregnancy outcomes (especially iv. and v. ) is stronger, allowing the inclusion in the headline findings of a figure for reduced peri-natal death (RR) of 0.46, and for still birth of 0.42.
Given the small number of studies and their heterogeneity, would we have learnt more about the components of these studies from a narrative presentation? Is the preference for the systematic review format driven by the need to obtain useful headline figures that would furnish statistical ammunition for the worthy cause of syphilis prevention?
It is interesting that in one study reviewed (Potter et al.) syphilis testing was increased only in an intervention that included improved HIV-PMTCT services.
S. Hawkes, “Effectiveness of interventions to improve screening for syphilis in pregnancy: a systematic review and meta-analysis”, The Lancet, published online 16th June 2011