Premature infant births are on the increase. In an editorial published in the British Medical Journal (21 April 2006), Professor Andrew Shennan states that this trend could have a considerable impact on society. Research from Denmark shows a 22% increase in premature births between 1995 and 2004. The researchers predict that this pattern is likely to be reflected in countries such as Ireland.
Out of the 60,000 babies born in Ireland each year, around 3,000 are premature. Premature births are increasing in Ireland due to a greater number of couples opting for in vitro fertilisation (IVF) therapy. In addition, innovations in the treatment of premature infants has led to more babies being delivered prematurely in order to treat complications, according to the Department of Neonatology at Holles Street Hospital in Dublin.
Around 600 of the 3,000 premature infants born each year are very immature, born at least six weeks before their due date, and around 10% of all newborn babies will need admission to a neonatal intensive care unit. Dr John Murphy, consultant neonatologist at Holles Street, says there is a growing need for investment, research and support in the area of premature infants (www.irishhealth.com).
Premature delivery – pre-term birth (PTB) and low birth weight (LBW) – is closely associated with infant morbidity and long-term health problems. LBW is defined as an infant weighing less than 5.5 pounds at full term, while PTB remains one of the most serious problems in obstetrics. PTB is defined as a gestational age at birth of less than 37 completed gestational weeks. It is further classified into three main categories: mild (32-36 weeks), very pre-term (28-31 weeks) and extremely pre-term (less than 28 weeks), with average frequencies of 85%, 10% and 5% respectively (Moutquin J, 2003; Lawn JE et al, 2005). Despite major preventive efforts, the incidence of PTB has remained constant at about 5-10% of live births in most countries over the past two decades (Bibby E, Stewart A, 2004).
There is mounting evidence that suggests chronic oral infection as a trigger for premature birth and low birth weight (Offenbacher S et al, 1996). While periodontitis is often considered to be a local oral infection, over the last decade it has become increasingly more evident that both local and systemic inflammation may occur (Genco RJ, 1992).
Can a causal relationship between periodontitis and PTB be established and, if so, can subsequent intervention therapies prevent pre-term birth?
It was in 1996 that researchers first reported an association between maternal periodontal disease and delivery of a pre-term infant (Offenbacher S et al). A case control study of 124 pregnant women found that women who delivered at less than 37 weeks’ gestation or who had an infant that weighed less than six pounds had significantly worse periodontal disease than women in the control group. The researchers suggested a seven-fold increased risk of pre-term delivery and considered periodontal disease to be a previously unrecognised and clinically significant risk factor for delivery of a pre-term LBW infant.
This was followed by another independent study of 55 matched pairs of women, where mothers with a healthy periodontium and low gingival index were found to be at lower risk for LBW babies (Dasanayake AP, 1998).
Davenport et al (2002), however, found no such association. He examined 236 infants born at less than 37 weeks’ gestation and weighing less than six pounds and compared them to a random sample of 507 control infants who were born at 38 weeks or more and weighing six pounds or more. They found that, on average, deeper pockets at delivery were associated with a reduction in the risk of delivery of a pre-term, LBW infant.
In discussing their results, the authors considered that a difference in study population may have had a limited effect on the outcome. The Offenbacher study consisted of 124 women who were slightly younger, 58% of whom were black. The Davenport study, however, was much larger (743 women), and subjects were recruited from a predominantly Bangladeshi (53%) community residing in East London.
The authors concluded that the biological mechanisms through which periodontal disease could cause LBW may be potentiated by the presence of other specific genetic and environmental factors.
Despite disparate findings, these early studies led to the hypothesis that the major periodontal pathogens increase circulating levels of pro-inflammatory mediators (such as PGE2, IL1, IL6 and other cytokines) that, in turn, trigger pre-term delivery.
Continued research into the link between periodontal infection and PTB appears to be based on epidemiological data, with a number of causal mechanisms also being suggested. These include elevated pro-inflammatory mediators, an association between periodontal disease and bacterial vaginosis (BV), maternal and foetal humoral response and elevated C-reactive protein (CRP) levels.
It has long been recognised that maternal infection and subsequent inflammation is a risk factor for an adverse pregnancy outcome.
Inflammation has been implicated in the mechanisms responsible for pre-term and term parturition, as well as foetal injury. Out of all of the suspected causes of pre-term labour and delivery, infection and/or inflammation is the only pathological process for which both a firm causal link with pre-term birth and a molecular pathophysiology have been defined (Romero R et al, 2006).
The common pathway appears to be activation of the inflammatory cascade. Bacterial colonisation and/or inflammation induces production of pro-inflammatory cytokines that lead to neutrophil activation and the synthesis and release of uterotonins such as prostaglandins. These cause uterine contractions and metalloproteinases that weaken foetal membranes and remodel cervical collagen (Park JS, 2005).
Local production of pro-inflammatory cytokines in the gingival crevicular fluid rises with increasing inflammation (Orozco A et al, 2006). One hypothesis to explain the association between periodontal disease and pre-term birth/low birth weight is that PTB and LBW may be indirectly mediated through translocation of bacteria or bacterial products in the systemic circulation. Transient bacteraemias occur in subjects with marginal periodontitis or gingivitis, and it is possible that bacteria and their products may reach the placental membranes haematogenously and provide the inflammatory effect to induce pre-term labour (Lopez NJ, 2005). The fact that intra-amniotic levels of many inflammatory mediators, such as prostaglandin E2, increase during pregnancy and peak at delivery suggests that these mechanisms play an important role in normal physiologic parturition.
There is increasing evidence to suggest that the presence of a sub-clinical endogenous Gram-negative infection of periodontal disease may present a systemic challenge sufficient to initiate the onset of premature labour. It has also been suggested that gingival crevicular levels of PGE2 correlate to the amniotic fluid levels of PGE2 and may be used to provide an indirect estimate of the amniotic fluid levels of PGE2 (Damare SM et al, 1997).
Bacterial vaginosis (BV) is the most common cause of premature birth. Recent studies have confirmed its association with pelvic inflammatory disease and adverse pregnancy outcomes.
BV has been associated with low birth weight (Hillier SL et al, 1995) and pre-term birth, with odds for pre-term birth estimated to be from 1.8 to 2.8 (Meis PJ et al, 1995; Hay PE et al, 1994).
It has been suggested that BV and periodontal disease are interrelated. Both diseases are associated with high bacterial burden and an increase of pro-inflammatory cytokines and other mediators of inflammation (Mattsby-Baltzer I et al, 1998; Orozco A et al, 2006).
Maternal and foetal humoral responses to oral pathogens have also been identified as possible risk factors or markers for pre-term delivery among pregnant women with periodontal disease. A lack of maternal IgG antibody to pathogenic oral organisms was associated with an increased rate of pre-term delivery. There was also a 2.9-fold higher prevalence of neonatal IgM seropositivity for one or more oral pathogens among pre-term babies, compared to full-term babies. These results led the researchers to conclude that maternal periodontal infection without adequate maternal antibody response is associated with periodontal organisms that may be passed on to the foetus and result in pre-term delivery (Madianos PM et al, 2001; Boggesse KA, 2003).
Research has revealed that periodontal infections contribute to elevated systemic C-reactive protein (CRP) levels in humans (Noack B et al, 2001). New research has recently suggested that pregnant women with periodontitis have 65% higher CRP levels compared to periodontally healthy women (Pitiphat W et al, 2006). This study examined serum samples collected during early pregnancy. Smokers and diabetic patients were excluded. The authors concluded that periodontitis may increase CRP levels in pregnancy, which could mediate the association between periodontal disease and adverse pregnancy outcome.
Systemic markers of inflammation may also be regarded as predictive markers for PTB. Therefore, changes in these markers in periodontitis may be part of the explanation as to why periodontitis is associated with PTB and negative pregnancy outcomes in epidemiological studies. It is hypothesised that daily episodes of a bacteraemia originating from periodontal lesions are the cause for the changes in systemic markers in patients with periodontitis (Loos BG, 2005).
All of this research suggests that the treatment of periodontal disease should reduce the risk of PTB. To investigate this, Scannapieco et al (2003) conducted a systematic review of the evidence linking periodontal disease to PTB and LBW, and the effect of periodontal therapy on adverse pregnancy outcomes. With a cut-off date of October 2002, Scannapieco found only moderate evidence to support an association between the two conditions.
Since this time, however, further study has been undertaken. Jeffcoat et al (2003) carried out a randomised study of a group of women with periodontitis of between 21 and 25 weeks’ gestation. They concluded that periodontal treatment may reduce PTB in this group but that adjunctive metronidazole therapy did not improve pregnancy outcome.
Lopez et al (2002) claimed a five-fold reduction in the rate of prematurity following intervention trials. A further randomised, controlled trial was undertaken to determine the effect of routine plaque control and scaling on the pregnancy outcomes of 870 women with gingivitis (Lopez NJ et al, 2005). The research reported that periodontal treatment significantly reduced the PTW/LBW rate in this population of women with pregnancy-associated gingivitis.
New studies are currently recruiting patients for multi-centred randomised clinical trials, including one in Australia, the Obstetrics and Periodontal Therapy Study and the MOTOR study (for further details visit www.clinicaltrials.gov).
Common risk factors
The difficulty in establishing a causal relationship between PTB and periodontitis lies in the fact that both conditions share a number of aspects in common. From an epidemiological standpoint, both conditions are relatively prevalent, confounding a statistical relationship. They also share a number of risk factors, including low socioeconomic status, smoking, alcohol consumption, caffeine intake, diabetes and stress (Obaid-ur-Rahman M et al, 1998). Stress also increases cytokine production, which may independently lead to PTB or perhaps increase susceptibility to infection and, subsequently, PTB.
When designing and conducting further studies, it will be necessary to:
• Identify whether periodontal disease was present prior to pregnancy
•Test the hypothesis linking periodontitis with PTB and LBW
• Assess the gestational age correctly
• Analyse separately the babies delivered on term and pre-term
• Obtain sound information about confounders (e.g. tobacco use before pregnancy)
• Provide a control for them in the analysis (Vettorel MV et al, 2006).
The cause of PTB is often unknown. The mechanisms by which periodontal disease may cause PTB or LBW remain speculative and continued research is required. It is possible that maternal periodontal disease is a surrogate for other factors that may predispose to PTB.
While researchers strive to provide a definitive answer, monitoring the oral health of our pregnant patients remains part of a whole health approach to oral care. If intervention trials are shown to improve pregnancy outcome, then periodontal treatment will become an integral part of obstetric care in the prevention of pre-term birth and low birth weight babies.