Development of gestational diabetes mellitus in women with periodontitis in early pregnancy: A population-based clinical study.

AIM: This study aimed to determine whether periodontitis in early pregnancy and periodontal therapy during gestation affect the incidence of gestational diabetes mellitus (GDM) through a population-based clinical study. MATERIALS AND METHODS: Subjects without periodontitis at 1-4 weeks of gestation who met our inclusion criteria were enrolled in the non-periodontitis group. Periodontitis patients who agreed or refused to receive periodontal therapy during pregnancy were separately enrolled in the periodontitis treated or untreated group. At 12-16 weeks of gestation, gingival crevicular fluid (GCF) and venous blood were collected for analyses of bacterial species and serum inflammatory mediators, respectively. At 24-28 weeks of gestation, GDM patients were identified by oral glucose tolerance tests. The association tests were performed using Chi-squared statistics and regression analyses. RESULTS: The complete data of 3523 pregnant women were recorded during the study period. GDM incidence among the untreated periodontitis participants (84/749, 11.21%) was significantly higher than that among the non-periodontitis participants (108/2255, 4.79%) (p < .05), and periodontal treatment during gestation reduced the incidence from 11.21% (untreated group) to 7.32% (38/519, treated group) (p < .05). Based on multiple logistic regression analyses, it was found that periodontitis in early pregnancy was associated with GDM, and three-step regression analyses showed that Porphyromonas gingivalis (P. gingivalis) and the serum TNF-α and IL-8 levels played a role in the association between untreated periodontitis and GDM. Furthermore, Pearson's correlation test indicated that the existence of P. gingivalis in GCF was positively correlated with high serum levels of these two inflammatory mediators. CONCLUSIONS: This study establishes a connection between periodontitis in early pregnancy and GDM and demonstrates that the presence of P. gingivalis is associated with high levels of inflammatory mediators in serum, and thereby may contribute to the development of GDM. In-depth mechanistic studies are needed to further support these findings.

[Relationship between the expression of human leukocyte antigen G and preeclampsia].

OBJECTIVES: To detect the expression of human leukocyte antigen-G (HLA-G) in tissues from pregnant women with preeclampsia and discuss the relationship between HLA-G and preeclampsia. METHODS: Pregnant women with preeclampsia in Maternal and Child Health Hospital of Shaanxi Province from March 2009 to December 2009 were included. Eight were included into mild preeclampsia groups and 22 were included into severe preeclampsia group. And 30 age-matched normal pregnancies were referred as the control group. All women in the three groups received cesarean section. The soluble HLA-G (sHLA-G) levels in peripheral blood, umbilical blood and amniotic fluid were examined by ELISA; the expressions of HLA-G protein in placenta, fetal membrane and umbilical cord were examined by western blot. RESULTS: (1) The sHLA-G levels in peripheral blood, umbilical blood and amniotic fluid in each group. The sHLA-G levels in peripheral blood in mild and severe preeclampsia group were (50 ± 14) and (30 ± 6) µg/L respectively, and the sHLA-G levels in umbilical blood were (34 ± 10) and (26 ± 8) µg/L respectively. All were significantly lower than those in the control group (P < 0.01), which were (100 ± 16) and (70 ± 9) µg/L respectively. There was also statistical difference between mild and severe preeclampsia group (P < 0.01). Although the sHLA-G level in umbilical blood of severe preeclampsia group was lower than that in mild preeclampsia group, there was no statistical difference (P > 0.05). The sHLA-G levels in amniotic fluid in mild and severe preeclampsia groups were (26 ± 7) and (25 ± 5) µg/L respectively, which were lower than that in the control group (27 ± 6) µg/L, but the differences were not significant (P > 0.05). There was no statistical difference between mild and severe preeclampsia groups (P > 0.05). (2) The expression levels of HLA-G protein in placenta, fetal membrane and umbilical cord in each group. The expression levels of HLA-G in placenta and fetal membrane in the control group were 1.59 ± 0.36 and 0.42 ± 0.09 respectively. The expression of HLA-G in placenta was significantly higher than that in fetal membrane (P < 0.05). The expression level of HLA-G in umbilical cord in the control group was 0.24 ± 0.17, statistically different from those in placenta and fetal membrane, respectively (P < 0.01). The expression levels of HLA-G in placenta in mild and severe preeclampsia groups were 0.78 ± 0.21 and 0.29 ± 0.17 respectively, significantly different from the control group (P < 0.01). There was no expression of HLA-G in fetal membrane and umbilical cord in mild and severe preeclampsia groups. CONCLUSIONS: The expressions of HLA-G in the peripheral blood, umbilical blood and placenta in women with preeclampsia are significantly lower than those in normal pregnant women. The abnormal expression of HLA-G might be associated with the pathogenesis of preeclampsia.