Commentary

This systematic review deals with the question of whether maternal infection - and periodontal disease in particular - is associated with pre-eclampsia, a multifactorial disorder that complicates about 3% of all pregnancies. Pre-eclampsia remains an important cause of maternal and perinatal morbidity and mortality in both developed and developing countries.

It is now well documented that periodontal diseases could have an impact on systemic illnesses, including atherosclerotic cardiovascular disease1. Given the link between cardiovascular disease and pre-eclampsia on the one hand2, and the association between periodontal disease and adverse pregnancy outcomes on the other3, it has recently been hypothesised that pre-eclampsia could be associated with periodontal disease.

Conde-Agudelo and colleagues have produced a good-quality systematic review. In spite of space limitations due to the investigation of multiple infections, results of the systematic review are highly reproducible, both in terms of search strategy and calculation of pooled OR. Results of the meta-analysis suggest that pre-eclampsia is associated with periodontal disease, with some evidence of a dose-response gradient. In addition, a recent Brazilian case-control study4, published after the present systematic review, provided quite a similar OR (OR=1.52; 95% CI: 1.01 to 2.29), adding strength to the hypothesis.

There are nevertheless some limitations to this systematic review that warrant consideration. As stated by the authors, even if adjustments for some confounding factors are made, residual confoundings remain a serious problem in observational studies. Second, the quality assessment method used in this systematic review seems not to be sufficiently discriminative, since the nine studies included are rated as “being of high-quality”, which is obviously not the case. It would have been more appropriate to use the STROBE statement to assess the quality of reporting of observational studies5. Thus the result of the meta-analysis should be interpreted with caution, given that confounding factors and bias may account for a considerable proportion of the association.

Moreover, a recent randomised controlled trial found a non-significant increased risk of pre-eclampsia (OR=1.59, 95% CI: 0.89 to 2.83) among pregnant women who received periodontal scaling and root planning before 21 weeks as compared to women who underwent scaling and root planning after delivery6. This could be explained by the fact that periodontal treatment has been shown to result in acute, short-term systemic inflammation and endothelial dysfunction7.

Finally, this systematic review provides another occasion to point out an important weakness in the current literature. Contrary to what Conde-Aguledo imply, using “the most stringent criteria for diagnosing periodontal disease (pocket depth, clinical attachment level, bleeding on probing and gingival inflammation)” is not, in itself, sufficient to guarantee that studies are comparing similar degrees of exposure to periodontal disease. As stated by D. Matthews in a recent commentary for EBD8, “there is a lack of a standardised protocol to determine periodontal diseases. The research community needs to address this issue and develop consensus”.

Practice point:

To date, observational studies have suggested an association between maternal periodontal disease and pre-eclampsia. It has also been shown that treatment of periodontitis in pregnant women alleviates periodontal disease but, according to current evidence, it would be presumptuous to treat pregnant women in the aim of reducing the risk of pre-eclampsia.