Effect of intra-pregnancy nonsurgical periodontal therapy on inflammatory biomarkers and adverse pregnancy outcomes: a systematic review with meta-analysis

Periodontitis is a chronic destructive inflammatory disease affecting the tooth-supporting tissues, initiated by dental plaque, a biofilm with gram-negative anaerobic microorganism’s predominance [1, 2] and mediated by the inflammatory response of the host [2‐4].

Frequent episodes of bacteremia or dissemination of bacterial endotoxins from periodontal focus might induce systemic activation of the inflammatory response [5, 6] and intense pro-inflammatory cytokine production [7]. Therefore, anti-infective periodontal therapies, once reducing exposure to subgingival pathogenic microorganisms and products, might have an anti-inflammatory potential and therefore be considered an anti-inflammatory intervention [8].

The transit of periodontal pathogens, pro-inflammatory cytokines, and prostaglandins from periodontal pockets to the fetal-placental unit suggests a plausible hypothesis for the association between periodontal disease and complications of pregnancy, as premature births and low birth weight. This possibility resulted in a series of epidemiological and interventional studies, conducted in the last 20 years, to investigate the association between periodontal disease and adverse pregnancy outcomes (APO), with conflicting results [9‐14]. Systematic reviews and meta-analyses suggested this diversity might be, in part, due to methodological inconsistencies which also difficult comparisons between studies. Alternative explanations include variations in the populations assessed, the presence of a range of potential confounding factors, variations in the case-definition of periodontitis across studies, relative obstetric risk, and other factors which are known to influence the prevalence of APO, irrespective of oral status [4, 15].

Measurement of inflammatory biomarkers in gingival crevicular fluid (GCF) or saliva may be used to determine the state of periodontal inflammation, which may be enhanced by periodontal therapy, reducing inflammatory biomarker level of cytokines such as IL-1β, IL-6, IL-8, IL-10, IL-12, IL-17, TNF-α, chemokines (MCP-1, RANTES), and prostaglandins in the affected sites [16]. This reduction of inflammatory biomarker levels is also associated with significant improvements in the clinical parameters of the disease, such as probing depths, percentage of bleeding on probing, and dental plaque [17].

As pregnancy progresses, amniotic fluid levels of inflammatory cytokines, such as TNF-α, IL-1β, and prostaglandin E2 (PGE2), naturally rise until a critical level, thus inducing rupture of the amniotic sac membranes, uterine contraction, cervical dilation, and delivery. Besides normal delivery is controlled by inflammatory signaling, this triggering mechanism can be modified by external stimuli, as infection and inflammatory stressors. In fact, obstetric studies associated increased levels of local and systemic markers of inflammation with APO [15].

A significantly positive linear trend was observed between the degree of periodontal disease and inflammatory mediators among preconception women. The levels of inflammatory mediators (IL-1β, IL-6, β-glucuronidase, and TNF-α) in saliva and in serum (except IL-6) significantly increased with severity of periodontal disease. The elevated concentrations of serum IL-1β, β-glucuronidase, and TNF-α-accompanying saliva mediators suggested periodontal disease might be one of the factors triggering a systemic inflammatory response [18].

In two recent studies conducted with women immediately after delivery, GCF levels of IL-1β, IL-6, TNF-α, and PGE2 and serum levels of TNF-α and PGE2 were significantly increased in women with periodontitis compared with periodontally healthy women. A significantly positive correlation was observed between periodontal disease (PD) and clinical attachment level (CAL) and serum levels of PGE2, considered the main labor trigger. A significantly positive correlation was observed between serum PGE2 and GCF levels of TNF-α. These findings might indicate that periodontal inflammation affects circulatory levels of labor triggers [19]. Severe clinical attachment loss was also associated with higher serum levels of TNF-α and IL-1β [20].

A systematic review was conducted to evaluate association between periodontal inflammatory biomarkers from GCF, as IL-1β, PGE2, and TNF-α, and APO, suggesting that a positive association between GCF inflammatory biomarker level and APO might exist [21]. Therefore, intra-pregnancy nonsurgical periodontal therapy (IPPT), once reducing periodontal inflammatory biomarker level, should interfere with APO. Some interventional studies on the effects of IPPT on APO also tested periodontal inflammatory biomarkers as inflammation resolution signs or prognostic markers to predict pregnancy outcomes [10, 13, 16, 22, 23]. However, there is no systematic review available on the potential association between biomarker changes and APO occurrence after IPPT.

The primary aim of this systematic review with meta-analysis was to analyze the effects of IPPT on periodontal inflammatory biomarkers and gestational adverse outcomes. The focused question was “In pregnant women with chronic periodontitis, does nonsurgical periodontal therapy change inflammatory biomarker level and consequently the occurrence of premature births, low birth weight or maternal preeclampsia?”

The protocol of this study was registered on the International Prospective Register of Systematic Reviews-PROSPERO (https://​www.​crd.​york.​ac.​uk/​prospero/​display_​record.​asp?​ID=​CRD42015027750) under number CRD42015027750. This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses, following the PRISMA checklist (http://​www.​prisma-statement.​org/​) Additional file 1.

This systematic review and meta-analysis evaluated the effect of IPPT on periodontal inflammatory biomarkers and APO. Included studies showed IPPT reduced inflammatory biomarker level from GCF and some from blood serum and improved periodontal clinical parameters (PPD, CAL, and BOP). Two included studies [22, 23] showed a positive relation between reduced serum levels of inflammatory markers and lower premature birth (PB)/low birth weight (LBW) rates.

Intra-pregnancy nonsurgical periodontal therapy, once reducing periodontal pathogens and consequently the inflammatory response, could theoretically decrease APO risk. However, besides IPPT reduced inflammatory biomarkers from GCF [16, 22], it did not remarkably reduce inflammatory biomarker level from blood serum [22], except CRP [23], and did not influence biomarker level from cord blood [13]. Therefore, apparently it was ineffective in stopping inflammatory cascade triggered by periodontitis and intensified by pregnancy.

At the meta-analysis performed, one study [23] showed strong statistical significance for decrease of premature birth rate after nonsurgical periodontal therapy. Two studies [16, 22] showed no difference, and the last one [13] showed 4.08-fold increased risk for premature birth occurrence after intra-pregnancy periodontal therapy (RR = 4.08; 95% CI [0.47–35.24]).

During overall analysis of included studies, however, it was observed tendency for protection of premature births (PB) occurrence after intra-pregnancy nonsurgical periodontal therapy (RR = 0.54; 95% CI [0.38 to 0.77]; p = 0.0007). Besides, two studies [13, 23] showed highly antagonistic results for effect of nonsurgical periodontal therapy on PB. Khairnar et al. [23] had higher weight on analysis (65.9%), compared to Pirie et al. [13] (1.8%). However, methodological issues, as reduced sample size, confounding factors, and high risk of bias [23] affected study comparison and this relationship, so results must be considered cautiously.

Regarding low birth weight (LBW), two studies [13, 23] showed no statistical difference at meta-analysis (RR = 0.78; 95% CI [0.50–1.21]; p = 0.27). Khairnar et al. [23] had higher weight (96.3%) for protective effect of periodontal therapy on LBW occurrence, and Pirie et al. [13] showed 3.06-fold increased risk for LBW after intra-pregnancy nonsurgical periodontal therapy (RR = 3.06, 95% CI [0.33–28.43]). Once again, due to methodological issues and high risk of bias observed [23], results must be considered cautiously.

In this systematic review, two studies [22, 23] showed statistical significance for decrease of premature birth rate after IPPT. These papers also showed reduction for CRP [23] and IL-6 sr [22] from serum blood after IPPT. Pregnancy alters hormonal and immunological responses, leading to increased local periodontal inflammatory response for the same microbiological challenge, once the dental plaque increases in volume during pregnancy, without significant composition changes. Higher level of IL-6 at GCF during pregnancy is accompanied by systemic activation of IL-6. Even if serum increase of IL-6 is not significant, effects on pregnancy might be expressive, once the whole IL-6 cascade activation is observed (sICAM1, sgp130 e IL-6 sr) [22].

A cohort prospective study [25] showed periodontitis increases the level of CRP during pregnancy. Serum blood CRP could be one of the plausible inflammatory biomarkers in the association between periodontitis and APO. Periodontitis acts as a reservoir of bacterial lipopolysaccharides (LPS), which induces the production of CRP [26]. CRP is synthesized in the liver from hepatic action of inducers such as IL-6 and IL-8iso [26], and the maternal systemic stress during pregnancy leads to activation of hepatic route. Elevated CRP levels could increase the inflammatory response by activation of the complement system, release of inflammatory cytokines, resulting in premature rupture of the fetal-placental membrane [27, 28]. Thus, the hypothesis that nonsurgical periodontal therapy might reduce the bacterial load and associated bacteremia, therefore reducing the inflammatory stress during pregnancy and IL-6 cascade, contributes to reduction in CRP levels and, consequently, the incidence of APO [22, 23]. In this review, two studies tested CRP from serum, with opposite results. Khairnar et al. [23] observed reduction of CRP level after IPPT, and Offenbacher et al. [22] found no difference for CRP, but positive for IL-6rs. Both studies observed less PTB after IPPT.

Labor natural process involves liver production of inflammatory mediators during pregnancy that gradually rises till leads to rupture of the membranes, culminating with labor. However, host immune response to chronic periodontitis produces inflammatory mediators that theoretically could stimulate the hepatic route earlier, leading to increased levels of inflammatory mediators, accelerating labor, and causing premature delivery or low birth weight [15]. So, the control of inflammatory biomarker level after IPPT would be an easy way to attest success of therapy and monitor labor evolution.

However, there are conflicting results about correlation between local (GCF/saliva) and systemic levels of inflammatory periodontal biomarkers in women with chronic periodontitis. High cytokine levels were found in GCF of pregnant women with widespread periodontal inflammation but limited periodontal destruction, still, local cytokine levels were not strongly associated with serum levels [29]. Nevertheless, a recent study with a bigger sample of preconception women with periodontitis showed a positive association between elevated concentrations of IL-1β, β-glucuronidase, and TNF-α in saliva and serum [18]. And a study with post-delivery women with periodontitis found an association between serum and GCF levels of TNF-α and PGE2 [19].

Conflicting results were also observed regarding effects of nonsurgical periodontal therapy on systemic inflammatory biomarkers, besides the reduction of local levels on GCF [16, 17, 30]. Some studies found no significant effect on serum inflammatory biomarkers [10, 17], while others found significant reduction [22, 23]. In opposite, increased serum levels of MCP-1, TNF-α, and IL-8 after IPPT were found in one study [16]. This increase may be attributed to differences in individual responses to inflammatory periodontal disease and its treatment [31], or may reflect an increase related to gestational age [16].

The lack of consistent results for IPPT on systemic inflammatory biomarker reduction may be a consequence of the moment of periodontal treatment. It is possible that to diagnose and subsequently treat maternal periodontitis during pregnancy is already too late to halt inflammatory cascade and improve pregnancy outcomes [32]. In fact, all included studies treated pregnant women between 11 and 24 gestational weeks, because, during pregnancy, second trimester is the safest period to periodontal treatment. There is no publicized study about effects of nonsurgical periodontal therapy prior to conception on APO occurrence, although a research on this subject is currently on course [18].

Another possible explanation for conflictive results of IPPT on periodontal inflammatory biomarkers and APO among studies is sample characteristics. Studies in which treatment during pregnancy were effective were conducted among Indian [23] or African-American [22] populations at high risk of preterm delivery, as described before in previous studies [33‐35], and therefore not generalizable to Caucasian populations, as pointed by Pirie et al. [13]. Samples studied by Khairnar et al. [23] and Offenbacher et al. [22] were nonwhite, from mixed ethnicity, with low socioeconomic status, and both studies found reduction of PTB after IPPT. The other two included studies, one [16] used population with low socioeconomic status from mixed ethnicity, the other [13] was conducted with a less mixed population although unbalanced for socioeconomic status, and both found no positive effect of IPPT on APO occurrence. Apparently, some populations/ethnicities should be more prone to benefit from IPPT to reduce occurrence of APO.

An important limitation among the selected papers that should be analyzed is the definition of chronic periodontitis. Chronic periodontitis is an infectious disease resulting in inflammation within the supporting tissues of the teeth, progressive attachment, and bone loss and is characterized by pocket formation and/or gingival recession. It is recognized as the most frequently occurring form of periodontitis. It is prevalent in adults but can occur at any age. The disease is usually associated with the presence of plaque and calculus. Progression of attachment loss usually occurs slowly, but periods of rapid progression can occur [36]. Although this is the American Academy of Periodontology current definition, there is no certainty about included study concepts about chronic periodontitis. For inclusion criteria, author statement at methodology was accepted, but it is not impossible that other forms of periodontitis could be included at studies samples.

Another point of concern is periodontitis case-definition. Currently, there is no single and universally accepted case-definition for chronic periodontitis. Two current case definitions are widely used, one European and another American. The case-definition proposed by Tonetti & Claffey [37] in Group C of the European Workshop in Periodontology mentions a sensitive definition as “presence of proximal attachment loss ≥3 mm in two or more nonadjacent teeth” and an extensive definition with “presence of proximal attachment loss ≥5 mm in ≥30% teeth”. The American definition proposed by Page & Eke [38] from the criteria of the Center for Disease Control/American Academy of Periodontology defines mild periodontitis as “at least two interproximal sites with clinical attachment level (CAL) ≥3mm and at least two interproximal sites with probing depth (PD) ≥4mm (not on the same tooth) OR one site with 5mm PD”; moderate periodontitis as “at least two interproximal sites with CAL ≥4mm (not on the same tooth) OR at least two interproximal sites with PD ≥5mm (not on the same tooth)”; and severe periodontitis as “at least two interproximal sites with CAL ≥6mm (not on the same tooth) OR at least one interproximal site with PD ≥5mm”. None of included studies used any of these definitions. Some used more conservative definitions [16, 23], compatible with mild periodontitis, while other used more extensive definitions [13, 22], compatible with moderate periodontitis. This finding may contribute to confound results.

Beyond the case definition of chronic periodontitis, some limitations of this study include scarce data for LBW as primary outcome, absence of large RCT testing inflammatory biomarkers before and after IPPT and correlating then with APOs occurrence, and great heterogeneity of inflammatory biomarkers tested in available studies, as well as diversity of source (serum blood, cord blood, and GCF). Due to these limitations, further studies are necessary to confirm association between biomarker level after IPPT and APO occurrence rate.

Intra-pregnancy nonsurgical periodontal therapy improved periodontal clinical parameters, decreased periodontal inflammatory biomarker levels from GCF and some from serum blood; however, it did not influence inflammatory biomarker level from cord blood nor consistently reduced preterm birth or low birth weight occurrence.