| Sign In to gain access to subscriptions and/or personal tools. |
Higher Risk of Preterm Birth and Low Birth Weight in Women with Periodontal Disease
1 Department of Conservative Dentistry, Section of Periodontics, School of Dentistry, University of Chile, Casilla Postal 89, Santiago 650363, Chile; and Correspondence: *corresponding author, nlopez{at}interactiva.cl
Pregnant women with periodontal disease (PD) may be at increased risk for having preterm low-birth-weight (PLBW) children. We investigated whether the maintenance of the mothers' periodontal health after 28 weeks' gestation reduces the risk of PLBW. Of the 639 women studied, 406 had gingivitis and received treatment before 28 weeks' gestation, and 233 had PD and were treated after delivery. Data about previous and current pregnancies and known risk factors were obtained from patients' medical records. Primary outcomes were delivery before 37 weeks' gestation or an infant with birth weight below 2500 g. The incidence of PLBW was 2.5% in periodontally healthy women, and 8.6% in women with PD (p = 0.0004, relative risk = 3.5, 95% CI, 1.7 to 7.3). Risk factors significantly associated with PLBW were previous PLBW, PD, fewer than 6 pre-natal visits, and low maternal weight gain. PD was associated with both preterm birth and low birth weight, independent of other risk factors.
Key Words: preterm birth, low birthweight, periodontal disease adverse effects, pregnancy, risk factors
Low birth weight (LBW), a major determinant of neonatal infant morbidity and mortality (Kramer, 1987), has been treated as a single entity in most studies, although it can result from either preterm birth (PTB) or intra-uterine growth restriction, or both. A case-control study found that periodontal infection may be a potential independent risk factor for preterm low birth weight (PLBW) (Offenbacher et al., 1996), and two prospective studies showed an association between preterm birth and periodontal infection (Jeffcoat et al., 2001; Mitchell-Lewis et al., 2001).
Periodontal diseases are a group of infectious diseases caused by predominantly Gram-negative, anaerobic, and microaerophilic bacteria that colonize the subgingival area. Inflamed periodontal tissues produce significant amounts of pro-inflammatory cytokines, mainly interleukin 1 beta (IL-1β), IL-6, prostaglandin E2, and tumor necrosis factor alpha (TNF- We undertook a concurrent cohort study with intervention to determine the association between periodontal disease and preterm low birth weight. Two groups of pregnant women were used: a group of women with gingivitis or mild periodontitis who received periodontal treatment before 28 weeks' gestation, and a group of women with periodontal disease who received no periodontal treatment during pregnancy. For the objective of the current study, to determine the relationship between periodontal disease and adverse delivery outcomes, women with gingivitis or mild periodontitis treated before 28 weeks' gestation were periodontally healthy and without periodontal infection, and were used as a comparison group. The null hypothesis tested was that there are no significant differences in the incidence of preterm low birth weight in women with periodontal disease compared with that of periodontally healthy women.
Patient Population The study population consisted of pregnant women of low socio-economic status who received uniform pre-natal care in a public health clinic in Santiago, Chile, and delivered at El Salvador Hospital between April, 1998, and December, 2000. Routine pre-natal care included screening for pregnancy complications, nutritional advice, stress reduction, education about the symptoms of preterm labor, correction of identified risk factors, and referral to the high-risk obstetric clinic when appropriate.
Selection of Subjects Potential participants were identified by the midwives who attended the pre-natal care clinics. A total of 945 women who gave verbal consent to a dental examination were referred to the investigators. All of these women received a clinical oral examination, the patient records were thoroughly examined, relevant data were extracted, and eligibility for the study was determined. Of the 945 women, 45 were ineligible because they declared intention to deliver at a hospital different from that of the study, and 19 refused to participate. In total, 881 women were selected.
Measurement of Periodontal Status Dental examinations and periodontal treatment were performed in a dental clinic located in the pre-natal care center.
Criteria for Periodontal Diagnosis All of the women who did not fulfill all of the criteria for periodontal disease showed gingival redness and bleeding on probing at more than 25% of sites and were diagnosed as having gingivitis or mild periodontitis. Of the 881 women who were selected to participate in the study, 263 had periodontal disease, and 618 had gingivitis or mild periodontitis. Of these women, 159 refused to receive periodontal treatment during pregnancy, but they accepted a new periodontal examination between 28 and 30 weeks' gestation. The results of the analyses of women with untreated gingivitis will be reported in the near future. An informed written consent was obtained from each volunteer, and the study protocol was approved by the institutional review board.
Periodontal Intervention Two hundred and sixty-three women with periodontal disease received no periodontal treatment during pregnancy, and they were monitored every 4-6 weeks during the gestational period so that any worsening of their periodontal status could be detected. All of the women with periodontal disease were treated after delivery. Caries lesions were treated, and teeth indicated for extraction were extracted from all patients. A new periodontal examination was performed between 28 and 30 weeks' gestation.
Recording of Maternal Characteristics A normal standard of the weight-to-height proportion established for Chilean women was used for the evaluation of nutrition status. At every week of gestation, pregnant women should gain weight within the recommended weight range for her weight-to-height proportion. According to the weight gain, women were assigned to one of the following categories: underweight, normal weight, overweight, or obese. Pre-natal care was categorized as that starting before 12 weeks' gestation, between 13 and 20 weeks' gestation, or after 20 weeks' gestation. The number of pre-natal visits and the onset of pre-natal care were used for assessment of the adequacy of pre-natal care. Cigarette smoking and the number of alcoholic drinks consumed per week were recorded. Smoking more than 5 cigarettes per day was considered tobacco abuse. According to the protocol of the pre-natal care program, women with urinary infections or with asymptomatic bacteriuria were treated with oral nitrofurantoin for 10 days. Women with vaginosis were treated with locally applied antibiotics according to the results of the microbiological culture.
Definition of Pregnancy Outcomes For the analyses of data, women were grouped according to pregnancy outcomes into: a preterm-birth group (PTB) if they delivered before 37 weeks of gestation, a low-birth-weight group (LBW) if they delivered a baby with a birth weight under 2500 g, and a preterm/low-birth-weight group (PLBW) if they delivered either a preterm or a low-birth-weight baby.
Statistical Analysis Unadjusted and adjusted risk ratios were calculated with 95% confidence intervals. Statistical analysis was performed with the SAS system (version 6.12, Cary, NC, USA). Statistical significance was defined as P < 0.05.
Of the 722 women enrolled in the study, 83 (11.5%) were excluded for various reasons. Of these, 53 (11.5%) were in the group of periodontally healthy women and 30 (11.2%) in the group with periodontal disease. Of the excluded women, 60 were lost to follow-up, 14 had a spontaneous abortion, and nine had medically indicated preterm delivery due to pre-eclampsia, gestational diabetes, or polyhydroamnios. There were 639 women who finished. Of these, 406 were in the group of periodontally healthy women and 233 in the group with periodontal disease. The mean age of the women was 25 (SD ± 4.5), 24.4% were single, 35.5% had fewer than 12 years of education, 40.8% were primiparous, and 18.3% were smokers. Alcohol consumption, as a variable, was eliminated from the analyses, since no woman declared drinking more than 2 drinks per week. The total incidence of PLBW was 4.7% (30/639). Of these, 18 were preterm births and 12 were low-birth-weight infants. The incidence of PLBW was 2.5% (10/406) in periodontally healthy women and 8.6% (20/233) (P = 0.001) in women with periodontal disease. The relative risk for a woman with periodontal disease having a PLBW was 3.5 (95% confidence interval 1.7 to 7.3; P = 0.004). The total incidence of preterm birth was 2.8% (18/639). The incidence was 1.5% (6/406) in periodontally healthy women and 5.2% (12/233) (P = 0.014) in women with periodontal disease. The relative risk for a woman with periodontal disease having a preterm birth was 3.5 (95% confidence interval 1.3 to 9.2; P = 0.006). The total incidence of low birth weight was 1.9% (12/639). It was 1.0% (4/406) in periodontally healthy women and 3.4% (8/233) (P = 0.024) in women with periodontal disease. The relative risk for a woman with periodontal disease of having a low-birth-weight infant was 3.5 (95% confidence interval 1.06 to 11.4; P = 0.028).
Table 1
The periodontal characteristics of women with periodontal disease showed that they had moderate to severe periodontal disease, while those of women who were treated were compatible with a healthy gingival and a healthy periodontal status (Table 2
Women with PLBW had significantly more severe and extended gingival inflammation and poorer periodontal status than women with normal parturition.
In the univariate analysis for PLBW, a significant association was found with a previous PLBW, fewer than 6 pre-natal visits, periodontal disease, a previous abortion, and low maternal weight gain (Table 3
In the univariate analysis for preterm birth, significant associations with a previous PLBW (risk ratio 9.06, 95% confidence interval 3.5 to 23.5; P = < 0.0001) and periodontal disease (risk ratio 3.5, 95% confidence interval 1.3 to 9.1; P = 0.006) were found. Periodontal disease was the only risk factor significantly associated with LBW in the univariate analysis (risk ratio 3.5, 95% confidence interval 1.06 to 11.4; P = 0.028).
Table 4
The risk factors for LBW were periodontal disease (adjusted risk ratio 3.6, 95% confidence interval 1.07 to 12.2; P = 0.028), and onset of pre-natal care after 20 weeks of gestation (adjusted risk ratio 3.3, 95% confidence interval 1.03 to 10.3; P = 0.034).
Patients in our study were relatively homogeneous, based on social and demographic factors reported as related to PLBW. The distribution of several known risk factors for PLBW was similar in both groups (Table 2 The definition of preterm birth used in our study includes births that followed spontaneous labor or spontaneous rupture of membranes, because there is considerable evidence that the risk factors for both are similar, and the distinction is artificial (Guinn et al., 1995). Since the determinants of PTB and intra-uterine growth restriction appear to differ (Kramer, 1987), we analyzed our data evaluating the risk factors for the two components of low birth weight together, and for each of the components individually, to determine which risk factors affect PTB and which LBW. Relatively few significant associations were found. The risk factors that showed significant association with PLBW were a previous PLBW, periodontal disease, fewer than 6 pre-natal visits, and low maternal weight gain. The factors exhibiting the largest risk ratios for PLBW and for PTB were a previous PLBW and fewer than 6 pre-natal visits. However, periodontal disease was the only risk factor associated with both PTB and LBW. The relationship between periodontal disease and PTB, and periodontal disease and LBW, was consistently maintained without substantial changes after adjustment for other risk factors and covariates in the logistic regression models constructed. These observations strongly suggest that periodontal disease is an independent risk factor for both PTB and LBW. The risk factors associated with PTB and LBW in our study are in concordance with those reported in several other studies. A previous history of PLBW is one of the most important risk factors for a subsequent PTB (Guick et al., 1984; Hediger et al., 1989; Wen et al., 1990; de Hass et al., 1991; Berkowitz and Papiernik, 1993), and low maternal weight gain has also been shown to increase the risk of PTB in several studies (Kramer, 1987; Abrams et al., 1989; Hediger et al., 1989; Carmichael and Abrams, 1997). Low maternal weight gain and inadequate pre-natal care are risk factors considered weakly associated with PTB in retrospective studies (Abrams et al., 1989; Carmichael and Abrams, 1997). However, in our study, fewer than 6 pre-natal visits showed adjusted risk ratio values that were consistently associated with PLBW and with PTB. Inadequate pre-natal care is often cited as a risk factor for poor pregnancy outcomes in low socio-economic status and poorly educated women (Sokol et al., 1980; United States General Accounting Office, 1987). Several studies have shown that adequate utilization of pre-natal care is associated with improved birth weights and lower risk of PTB (Quick et al., 1981; Greenberg, 1983; Donaldson and Billy, 1984). The women in our study had free access to a well-designed pre-natal care program to control many of the known risk factors for PLBW. Free provision of pre-natal care is advisable as an effective means of reducing preterm births based on the observation that PTB is less likely among women who seek pre-natal care early or have more pre-natal visits. The results of our study agree with those of studies showing that making more pre-natal care available to women does not reduce preterm births (Fink et al., 1992; Fiscella, 1995). Urinary infections and vaginosis are well-known risk factors for PLBW (Romero and Mazor, 1988; Holst et al., 1994; Paige et al., 1998). However, these factors were not associated with PLBW in our study, probably because the antibiotic therapy given to the women could either eliminate these infections or modify their effects on the pregnancy outcome. Cigarette smoking has been related to PTB and LBW (Shiono et al., 1986; Kierse, 1989), but this factor was not associated with any of the pregnancy outcomes in our study, probably due to the low proportion of women who smoked. Periodontal disease has been only recently identified as a potential risk factor for PLBW (Offenbacher et al., 1996), and it might be one of the factors associated with some of the approximately 50% of preterm births that occur in women without established risk factors (Kramer, 1987).
The mechanisms by which periodontal disease may cause preterm LBW or PTB have still not been elucidated, but there is evidence that this association has biologically feasible bases. It has been suggested (Offenbacher et al., 1996) that the effect of periodontal disease on PLBW could result from stimulation of fetal membranes on prostaglandin synthesis by cytokines produced by inflamed gingival tissues, or through the effect of endotoxin derived from periodontal infection. Endotoxin can stimulate prostaglandin production by macrophage amnion (Romero et al., 1988) and decidua in vitro (Romero et al., 1989). In animal models, it has been shown that endotoxin produces fetal growth retardation (Beckman et al., 1993; Offenbacher et al., 1998). On the other hand, peripheral monocytes obtained from some patients with periodontal disease showed enhanced release of inflammatory mediators such as PGE2, IL-β, and TNF- The results of our study, in which all evaluations were conducted before the babies were born, show that periodontal disease is an independent risk factor for PLBW and affords more than a three-fold increase in the risk for PTB and for LBW. These adverse pregnancy outcomes are frequently associated with potentially correctable lifestyles, or with infectious diseases that, like periodontal disease, can be eliminated prior to or during pregnancy.
This study was supported by project grant 1981094 from the Fondo de Investigación Cientifica y Tecnológica (FONDECYT). The collaboration of Drs. Violeta Pavez and Isabel Da Silva in the treatment of patients, and of Ms. Monica Rubilar and Valeria Vargas in the selection of patients, is greatly appreciated. The authors thank Ms. Ana Morales, Director of Consultorio General Carol Urzua, for the clinical facilities given to undertake the present study. Received for publication May 29, 2001. Revision received November 11, 2001. Accepted for publication November 27, 2001.
Journal of Dental Research, Vol. 81, No. 1,
58-63 (2002) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
), which may have systemic effects on the host. 





