Periodontol 2000. 2015 Feb;67(1):87-130. doi: 10.1111/prd.12073.
Preterm birth is a major cause of neonatal morbidity and mortality in both developed and developing countries. Preterm birth is a highly complex syndrome that includes distinct clinical subtypes in which many different causes may be involved. The results of epidemiological, molecular, microbiological and animal-model studies support a positive association between maternal periodontal disease and preterm birth. However, the results of intervention studies carried out to determine the effect of periodontal treatment on reducing the risk of preterm birth are controversial. This systematic review critically analyzes the methodological issues of meta-analyses of the studies to determine the effect of periodontal treatment to reduce preterm birth. The quality of the individual randomized clinical trials selected is of highest relevance for a systematic review. This article describes the methodological features that should be identified a priori and assessed individually to determine the quality of a randomized controlled trial performed to evaluate the effect of periodontal treatment on pregnancy outcomes. The AMSTAR and the PRISMA checklist tools were used to assess the quality of the six meta-analyses selected, and the bias domain of the Cochrane Collaboration's Tool was applied to evaluate each of the trials included in the meta-analyses. In addition, the methodological characteristics of each clinical trial were assessed. The majority of the trials included in the meta-analyses have significant methodological flaws that threaten their internal validity. The lack of effect of periodontal treatment on preterm birth rate concluded by four meta-analyses, and the positive effect of treatment for reducing preterm birth risk concluded by the remaining two meta-analyses are not based on consistent scientific evidence. Well-conducted randomized controlled trials using rigorous methodology, including appropriate definition of the exposure, adequate control of confounders for preterm birth and application of effective periodontal interventions to eliminate periodontal infection, are needed to confirm the positive association between periodontal disease and preterm birth.
早产是发达国家和发展中国家新生儿发病率和死亡率的主要原因。早产是一种高度复杂的综合征，包括不同的临床亚型，其中许多不同的原因可能涉及。流行病学、分子、微生物学和动物模型研究的结果支持牙周病与早产之间存在正相关关系。然而，为确定牙周治疗对降低早产风险的影响而进行的干预研究的结果存在争议。本系统评价批判性地分析了这些研究的荟萃分析中的方法学问题，以确定牙周治疗对降低早产风险的效果。所选随机临床试验的质量对于系统评价至关重要。本文描述了应事先确定并单独评估的方法学特征，以确定评估牙周治疗对妊娠结局影响的随机对照试验的质量。使用 AMSTAR 和 PRISMA 清单工具评估了所选六项荟萃分析的质量，并应用 Cochrane 协作工具的偏倚域评估了纳入荟萃分析的每项试验。此外，还评估了每个临床试验的方法学特征。荟萃分析中纳入的大多数试验都存在严重的方法学缺陷，威胁到其内部有效性。四项荟萃分析得出牙周治疗对早产率无影响的结论，而另外两项荟萃分析得出治疗可降低早产风险的积极结论，这些结论都不是基于一致的科学证据。需要进行设计严谨、方法学可靠的随机对照试验，包括明确暴露定义、充分控制早产的混杂因素以及应用有效的牙周干预措施消除牙周感染，以证实牙周病与早产之间的正相关关系。
