Incidence and Factors Influencing Postpartum Bacterial Vaginosis: a Controlled Study
Zhang Dai,Mi Lan,Yang Huixia
DOI: https://doi.org/10.3760/cma.j.issn.0366-6999.20131062
IF: 6.133
2014-01-01
Chinese Medical Journal
Abstract:Vaginitis is a common infectious disease often associated with abnormal vaginal flora. Previous studies have shown that abnormal vaginal flora often cause bacterial vaginosis (BV), a risk factor of preterm delivery. Therefore, BV is a concern in pregnant women. Until now, few studies have focused on changes in vaginal flora in postpartum women. Watts et al1 identified BV as a risk factor for post-cesarean endometritis. However, Yang et al2 found that the incidence of BV is higher in postpartum women than in nonpregnant women. Our study measured the incidence of BV in postpartum patients and the factors influencing the disparity in incidence. Our study was designed to identify whether BV drug therapy is necessary in postpartum women. METHODS Study participants The trial was conducted from April to May 2010 in Peking University First Hospital. A total of 209 women 6 to 8 weeks postpartum underwent routine examination of their vaginal flora in our obstetrics and gynecology clinic. The control group included 200 healthy nonpregnant women being seen for a routine physical examination. This study was approved by the Ethics Committee of Peking University First Hospital. Sampling of vaginal discharge First, we sampled vaginal secretions from the upper third of the vaginal wall using a sterile scraper or cotton swab. We stained the smears using Gram's method. Finally, we examined the smears under the microscope and assigned each a Nugent score.3 Smears with a Nugent score ≥7 are diagnostic of BV, while those with a Nugent score of 4 to 6 could be diagnostic of intermediate BV. Questionnaires We gave participants a questionnaire consisting of objective indicators of postpartum maternal factors, including age, living environment, and breastfeeding status. It also included sexual behavior-related factors, including postpartum sexual activity status, use of contraception, and attitude about sexual activity during pregnancy, and infection-related factors such as vulvar discomfort level and duration. Statistical analysis SPSS14.0 software (SPSS Inc., USA) was used for statistical analysis. The main method used was a χ2 test. P values of 0.05 or less were considered statistically significant. RESULTS General situation The average age of the postpartum women was (30.5±3.9) years, while that of the control group was (30.9±2.4) years. Seventeen (8.1%) of the women reported having sex since giving birth, and 149 women (71.3%) were breastfeeding. Reasons women gave for not having intercourse included a belief that intercourse should not begin until at least 6 weeks after delivery (83.7%); lack of energy (98/209, 46.9%); and no interest (33/209, 15.8%). Incidence of BV The rate of abnormal Nugent score was higher in postpartum women, with most of them having intermediate BV. However, 23.4% (49/209) of all subjects had vulvar discomfort. Of the women with a normal Nugent score, the percentage of women with symptoms was 25.4% (16/63), which was not lower than in the BV group (25%, 2/8) or intermediate BV group (22.5%, 31/138) (Tables 1 and 2).Table 1: Symptoms of postpartum patientsTable 2: Analysis on influencing factors and postpartum bacterial vaginosisFactors influencing postpartum vaginal flora Sexual behavior had no significant effect on postpartum flora. The proportion of intermediate BV in breastfeeding women was significantly higher than that of women who bottle-fed their infants, while there was no significant difference in incidence of BV between the two. Mode of delivery had a greater impact on postpartum flora. Women who had a Cesarean section were more likely to have intermediate BV than women who had delivered vaginally, but the incidence of BV was not significantly different between the two (Table 2). DISCUSSION Postpartum changes in vaginal flora The vaginal flora of healthy women is dominated by Lactobacillus acidophilus, an acid-producing and acid-resistant flora. There is an ecological succession of different microbial floras when age and pregnancy status change. Previous studies have shown that BV can lead to obstetric complications such as premature labor and premature rupture of membranes. Therefore, prenatal BV has been studied more than postnatal BV. About 66% of postpartum women have a Nugent score of 4-6, which is quite different from that of nonpregnant healthy women. This suggests that postpartum women often have obvious abnormal vaginal flora. Women participating in our study mainly chose breastfeeding, which leads to a high in vivo prolactin level and a low estrogen level.4 A lower level of estrogen causes thinner vaginal mucosa and decreased glycogen content, which could explain abnormal postpartum vaginal flora. Such changes could be a physiological phenomenon. Factors influencing postpartum vaginal flora Factors influencing the vaginal flora of healthy nonpregnant women include sexual intercourse, vaginal douching, and menstrual cycle status. Most postpartum women enrolled in our study had not yet returned to a regular menstrual cycle and only 8.1% of them had begun to have sexual intercourse again. Because of these findings, we believe that vaginal flora changes may be more closely related to physiological conditions than to lifestyle factors. Our study showed no significant correlation between postpartum sexual behavior and BV. However, because of the limited number of women who began sexual intercourse in the immediate postpartum period, the influence of sexual intercourse on postpartum BV needs further study. It is puzzling that our study found that breastfeeding population had a higher incidence of BV than women who bottle-fed their infants. We speculate that this phenomenon may be due to the high in vivo prolactin level and low estrogen level in breastfeeding women, which could lead to the slow recovery of vaginal flora. Mode of delivery has a more visible impact on postpartum BV. This study showed a high incidence of BV in Cesarean section patients, which may be due to the relatively slower recovery or lengthier vaginal bleeding after Cesarean section. The effect of the various factors influencing the incidence of BV is mild. The main effect is the rise of the proportion of intermediate BV, whereas the occurrence rate of BV changes little. Intervention in postpartum BV This study shows that postpartum women have a higher probability of having intermediate BV, which is a flora disturbance. Because a vaginal flora disturbance may increase the pathogenicity of other diseases, we should pay close attention to this population. This study also discovered that the rate of symptoms in the population with disturbed flora was not higher than the rate of the normal flora group. We infer that the symptoms may not be due to a vaginal flora disturbance; rather, the vaginal flora disturbance may be related to a temporary low physiological estrogen level. With the end of breastfeeding and recovery of the menstrual cycle, normal vaginal floras are restored. According to the recommendation of the sexually transmitted diseases treatment guidelines,5 patients with BV have symptoms that require treatment. In this way, patients with intermediate BV but no symptoms do not require drug intervention.