Periodontal infection and preterm birth: successful periodontal therapy reduces the risk of preterm birth
Article first published online: 11 MAR 2011
© 2011 The Authors BJOG An International Journal of Obstetrics and Gynaecology © 2011 RCOG
BJOG: An International Journal of Obstetrics & Gynaecology
Volume 118, Issue 5, page 635, April 2011
How to Cite
Di Mario, S., Spettoli, D., Alessandrini, C., Erenbourg, A., Ronfani, L. and Basevi, V. (2011), Periodontal infection and preterm birth: successful periodontal therapy reduces the risk of preterm birth. BJOG: An International Journal of Obstetrics & Gynaecology, 118: 635. doi: 10.1111/j.1471-0528.2011.02913.x
- Issue published online: 11 MAR 2011
- Article first published online: 11 MAR 2011
- Accepted 17 December 2010.
Despite the great interest in reading the paper from Jeffcoat et al.1 on the treatment of periodontal infection and preterm birth, a few methodological issues in the study design and interpretation raise doubts on the validity of its conclusion. The aim of the paper was to assess the effect of periodontal therapy in reducing the risk of preterm birth and it concluded that the treatment successfully decreased the incidence of preterm birth.
The study was originally designed as a randomised controlled trial. Nevertheless the effectiveness of periodontal therapy is only demonstrated in a nonrandomised comparison between subgroups within the study population: women treated successfully compared with women treated without success. Therefore, there is no certainty that the two populations were comparable with respect to known and unknown confounding factors.
Although the authors claimed that they would have conducted a stratified analysis based on severity of periodontal disease (mild versus moderate or severe) and previous preterm birth, such an analysis has neither been published nor commented on. Previous preterm delivery is a well-recognised risk factor for preterm birth and not considering it in such an analysis creates a major methodological issue, particularly as the comparison was carried out between two nonrandomised groups.
Furthermore, the study indicated a preterm rate of 49% (52.4% among treated women and 45.6% among untreated women), well above the 4.9% rate of the Periodontal Infections and Prematurity Study (PIPS),2 of which the present study is part. As 55% of the PIPS population is represented by the study population in Jeffcoat et al., such a huge discrepancy is not understandable and the authors have not commented on it to justify such a difference.
Finally, the conclusions of the present study are not consistent with evidence from previous studies:2–4 the authors explained the difference by the diverse inclusion criteria (women were selected only if they showed at least three sites with 4 mm or more of attachment loss) and the fact that only successful treatment of periodontal disease should be considered to assess treatment efficacy in reducing preterm birth.
In our opinion there is an alternative explanation: the applied study design does not permit the control of confounders and bias and therefore the lack of randomisation and of control for previous preterm birth hinders the validity of this study.