The normal gestation period lasts forty weeks. According to the World Health Organization (WHO), the birth given before thirty-seventh week of gestation is defined as a preterm birth (PB) and low birthweight (LBW) represents a newborn weighing less than 2500 grams (1). Preterm low birth weight (PLBW) is the most frequent cause of infant mortality and nearly half of serious long-term neurological diseases (2, 3). These deliveries and medical expenses for their treatment have become an economic burden, not only for the affected family but also for the society as a whole (4). Despite the progress in the understanding of reproductive physiology, the prevalence of PLBW has not changed. In some population groups, the prevalence of PLBW has increased (5, 6).
Numerous studies have identified the following risk factors for PLBW: mother's age (under 17 and over 35 years) (7), low socioeconomic level, alcohol, drugs, African-American race, low prenatal care (8), smoking, multiple pregnancy, genitourinary tract infections, mother’s systemic diseases (hypertension, eclampsia, preeclampsia, gestational diabetes) (9). However, those risk factors were not present in 25% of the cases with PLBW (10).
The basic clinical parameters for measuring the severity of periodontal disease (PD) are clinical attachment level (CAL) and pocket probing depth (PPD) together with gingival bleeding and radiological evaluation of osseous resorption. The standardized protocol which is used today for measuring both the clinical attachment level and pocket probing depth with the manual probe was described for the first time more than 50 years ago (11), and has not changed much since then.
Problems for researchers that have emerged from a low reliability of the measurements with the periodontal probe have prompted them to seek more reliable markers that, which would result in a more precise periodontal diagnosis and subsequent treatment planning. Chronic and cyclic nature of the PD increases the likelihood of repeated hematogenous dissemination of periopathogens and of direct microbial exposure of the blood vessels, liver, including a fetal-placental unit of pregnant women. An increased C-reactive protein is a marker of liver activation as the acute phase response, and this increase in pregnant women is associated with premature delivery (12).
Investigations in pregnant rodent models have demonstrated that even low-level exposure to oral microorganisms may result in fetal growth disorders. This is demonstrated in an animal model that used a subcutaneous injection of Porphyromonas gingivalis (13, 14), but also in a human model of experimental periodontal disease (15).
In 1996, Offenbacher et al. (16) were the first researchers to clinically demonstrate that PD is a statistically significant risk factor for PLBW, with the odds ratio (OR) of 7.9 for all PLBW cases, and 7.5 for primiparous PLBW cases, after adjustment for known risk factors.
Davenport et al. failed to find sufficient evidence to link PD with PLBW. After analyzing 236 cases with PLBW and 507 controls with normal delivery, the results showed that with increasing depth of periodontal pockets, the risk for PLBW reduced (OR = 0.83). After the age adjustments had been made, and also, adjustments related to ethnicity, smoking, and infections, this risk was even more reduced (OR = 0.78) (17, 18).
In a meta-analysis of over forty articles (of which eight are highly relevant) on the impact of the PD on PLBW, it has been concluded that the disease significantly increases the risk of PB and/or LBW, and that despite the importance of promoting the oral health in pregnant women, there is still no strong evidence that treatment of diseased periodontium would reduce the risk of PB (19).
The objective of this study was to compare periodontal parameters of mothers who had a preterm delivery and/or a low-birth-weight infant with those who had a normal delivery and a normal birth weight infant and gestation and, subsequently, to investigate the relationship between PD and PLBW.
Material and Methods
Data from periodontal examination and obstetric records of 200 nursing mothers, regardless of their age, who gave birth at the University Clinical Center of Kosovo’s Obstetrics and Gynecology Clinic, were used. The approval of the University of Pristina Medical School Ethics Committee was obtained prior to the commencement of the study. Subjects were randomly selected, and after they had given an oral informed consent, they initially underwent a periodontal examination. Subsequently, their obstetric records were taken.
In addition, the periodontal examination was performed under natural light, using dental mirror and periodontal probe with 1-2-3-5-7-8-9-10 increments (3N Nabers P3N probe; Hu-Friedy Manufacturing, Inc., Chicago, Illinois). Periodontal data included: dental plaque (by Sillness-Löe) (20) and gingival index (by Löe-Sillness) (21) was assessed in the following four tooth surfaces: buccal, lingual, mesial and distal; bleeding on probing (in percentage) was assessed in similar sites as dental plaque and gingival index, which was positive if hemorrhage occurred within 15 seconds after probing; PPD, CAL and gingival recession (GR) (all in millimeters) were assessed in the following six tooth surfaces: mesiolingual, mid-lingual, distolingual, mesiobuccal, mid-buccal and distobuccal. PPD represented the distance from the gingival margin to the base of the pocket; GR represented the distance from the cementoenamel junction to the gingival margin; CAL was calculated from probing depth and gingival recession, representing the distance from the cement-enamel junction to the base of the pocket. The third molars and teeth indicated for extraction were excluded from the evaluation.
Obstetric data were composed of two parts: demographic and actual birth - date of delivery, weight at birth and gestation age. Data from the first part were obtained from the subjects’ obstetric records after the periodontal examination, and from the second part they were obtained from the newborn’s history. Obstetric exclusion criteria were: multiple births (twins, triplets), maternal systemic disease (hypertension, diabetes, etc.), and genitourinary tract infections.
The modified Machtei’s criteria were used for determining whether the subjects had periodontitis or not (22). The periodontitis group included subjects with periodontal pocket depth 5 mm or more in one or more bleeding-positive sites and with clinical attachment level amounting to 6 mm or more in two or more sites. The subjects who did not meet the abovementioned criteria formed the group without periodontitis.
Periodontal parameters were recorded by one blinded periodontist (KM) to case-control status of the subjects. The intra-examiner reliability weighted with kappa coefficients for measurements within ±1 mm ranged from 0.80 to 0.90.
Based on the WHO criteria (1), nursing mothers who gave birth before the 37th week of gestation and infants weighing less than 2500 grams formed case groups with PB and LBW, while others who gave birth after the 37th week of gestation and infants weighing more than 2500 grams formed the control group with normal deliveries.
A statistical analysis was performed using the index of structure, arithmetic mean, and standard deviation. The Student's t-test was used to determine the difference between the arithmetic averages, and the chi square test (χ2) was used to determine the differences between non-parametric data.
The data of this study were analyzed in the sense of differentiating the subjects by their age. Out of two-hundred parturients, thirteen (6.5%) were aged 35 years or older. The mean age of all subjects was 26.9. After the women older than 35 had been excluded, the average dropped to 26.
The results are presented for the duration of gestation and birth weight separately and en bloc.
When comparing periodontal parameters of all subjects with normal gestation vs. PB, PPD and CAL were significantly higher in preterm birth subjects (p=0.009 and p=0.037, respectively). After the subjects who were aged 35 years or older had been excluded, GR was additionally higher in PB subjects (Table 1).
The subjects who gave birth to low-weighted babies had significantly higher dental plaque index (p=0.03) as well as deeper periodontal pockets (p=0.028). After adjustment for age older than 35, apart from the dental plaque index and PPD (p=0.027 and p=0.032, respectively), the CAL was also significantly higher (p=0.025) (Table 2).
The comparison of the birth-weight of women with PD and those without PD has shown that mean weight of the newborn is significantly lower in PD group than in healthy subjects (p=0.0003). After excluding thirteen subjects older than 35 years, this difference became more significant (p=0.0002), which indicates that, in this study, the mean birth-weight was not influenced by mothers’ age (Table 3).
|Mean birth weight (gr ± SD)||Healthy||Periodontitis||p|
|All subjects (n=200)||3208 ± 682.3||2751.4 ± 1008.5||0.0003|
|Subjects older than 35 (n=13)||2661.7 ± 962.0||2728.6 ± 1380.2||0.92|
|Subjects 17-35 years old (n=187)||3232.2 ± 662.0||2754.6 ± 963.9||0.0002|
Also, the duration of gestation was compared for healthy vs. periodontitis groups. The gestation period of the periodontitis group was significantly shorter compared to the non-periodontitis group (p<0.0001). The significance of shorter gestation in periodontitis group did not change even after the age adjustment for parturients older than 35 years old (p=0.0001) had been made (Table 4).
|Mean gestation (weeks ± SD)||Healthy||Periodontitis||p|
|All subjects (n=200)||38.6 ± 2.7||36.4 ± 4.8||<0.0001|
|Subjects older than 35 (n=13)||37.3 ± 3.01||36.0 ± 5.20||0.59|
|Subjects 17-35 years old (n=187)||38.6 ± 2.68||36.5 ± 4.77||0.0001|
After the adjustment for the mothers’ age, the Chi-square test for birth-weight and PD resulted in significant correlation (Chi-square 9.55, p<0.01), while the age-adjusted odds for subjects with PD to give birth to low-weighted babies were 3.2 fold higher than for non-PD subjects (Table 5).
|Birth weight (g)||Healthy||Periodontitis||χ2||OR (95% CI)|
|3.2 (1.5 – 6.8)|
Also, PB and periodontitis showed highly significant correlation when assessed with the Chi-square test (10.48, p<0.01). The odds for women with PD to have a child born before the 37th week of gestation were more than 3 times higher than for women without PD (Table 6).
Some recent studies indicate that infections of periodontal tissues may contribute to morbidity and mortality of individuals with some systemic diseases such as atherosclerosis, myocardial infarction, stroke and premature delivery. Jeffcoat et al. (23) found that women with generalized periodontitis are at risk of preterm delivery 4 to 7 times more frequently than healthy subjects. Our study, although with fewer subjects, showed that women with PD were at risk 3.2 times more frequently for delivering a low birth-weighted offspring and were at risk 3.4 times more frequently for delivering prematurely than women without PD. Our population consisted of subjects of Caucasian race, and due to demographic limitations, we could not find any differences based on the race of the subjects.
In a cohort study of 327 Brazilian low-risk pregnant women with or without PD, it was concluded that it was a risk factor for PLBW and premature rupture of membranes (24).
In the studies of Dasanayake (25) and Dempsey et al. (26), mothers with PLBW had significantly more bleeding sextants than controls. In contrast, finding from previous research by Offenbacher et al. (16) and Mitchell- Lewis et al. (27) did not show an association between the bleeding index and PLBW, which is a finding similar to that of the present study.
It has been pointed out that PD may not be the cause of PB, but may share a combined etiologic mechanism, such as a genetic predisposition for a hyper-inflammatory response (28).
Different associations between periodontal status and preterm birth alone (PB), low birth weight (LBW), or preterm birth associated to low birth weight (PLBW) have been shown in cross-sectional studies (29-36), as well as in case-control studies (37, 38). However, other cohort studies (36, 39-41), as well as case-control studies (17, 42-45), did not find a significant association.
Our investigation has found a significant correlation between periodontitis and low birth-weight (χ2=9.55), as well as between periodontitis and preterm birth (χ2=10.48). The adjusted odds for women with periodontitis to have low-weighted offspring were 3.2 times higher than for women without periodontitis. Also, the same adjusted odds for women with periodontitis to give birth prematurely were 3.4 times higher than for women without periodontitis.
Although this study included parturients regardless of their age, it was found that women with periodontitis had mean birth weight and mean gestation significantly lower than women without periodontitis, even after excluding women older than 35 years.
In the pool of data that try to detect the relationship between periodontal disease and preterm low birth weight, our study suggests that women are likely to give birth prematurely if they have periodontal disease. Within the limitations of this investigation, a definition of periodontal disease might have affected the quality of evidence. The timing of possible direct influence of periodontal disease on the course of pregnancy may improve the methodology of future investigations.