Relationship between cervical swab culture and placental histological evidence of amniotic fluid infection in preterm labor

An appropriate sample size was calculated by using the formula (N) = Z²PQ/D², where N was the desired sample size of the studied population, Z was the standard normal deviate, set at 1.96 which corresponded to the 95% confidence interval level, P was set at 0.5, from the prevalence of HCA in preterm labor obtained from Goldenberg et al. that was 49.8%, D was the degree of accuracy set at 0.1. Q was set at 0.5 (1-0.5). The 10% relative acceptable was added to the calculated sample size. N was 96, and after adding 10% relative acceptable, the total target sample size was increased to 106.

After receiving approval for this retrospective observational study from our Institutional Review Board at the Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand (approval No. 245/55) and permission to collect data from medical records by the Director of King Chulalongkorn Memorial Hospital, the data of the pregnant women who had preterm labor with intact membranes and delivered at the Hospital between January 2012 and May 2013 were retrieved. Patient identities were kept confidential.

The gestational age of the pregnant women diagnosed as preterm labor was between 20 weeks to 36+6 weeks [11]. True labor was diagnosed when patients had regular uterine contractions (4 in 20 minutes or 8 in 60 minutes) and significant cervical changes (cervical dilatation of more than 1 cm and/or cervical effacement of not less than 80%). All participants must have >3 antenatal visits and had certain gestational age, which was either calculated from reliable last menstrual period (LMP) or sonographic assessment of gestational age before 28 weeks of gestation. Any pregnant women with placenta previa or history of vaginal douche or antibiotics used within 7 days were excluded from the study.

A standard procedure for cervical swab culture was used. In brief, a specimen was collected by swabbing around the endocervical area. Then, the specimen was kept in sterile plastic tube containing Amie’s charcoal transport media before sending to the microbiological laboratory within two hours. The culture media used were blood agar, MacConkey agar, and Thayer-Martin agar. A negative result could be reported after 48 hours.

The HCA or pathological evidence of amniotic fluid infection was defined as the presence of acute inflammation in the placenta, which is classified into maternal and fetal inflammatory responses. The maternal inflammatory responses can be further classified into 3 stages as: (1) acute subchorionitis, (2) acute chorioamnionitis, and (3) necrotizing chorioamnionitis. Fetal inflammatory responses can also be further classified into 3 stages as (1) umbilical phlebitis and chorionic plate vasculitis, (2) umbilical arteritis and umbilical vasculitis, and (3) concentric perivasculitis [12].

The patients data were recorded in a “Case Record Form”, which is composed of general history, obstetric history, risk factors of preterm labor, symptoms and signs of chorioamnionitis, treatment, duration or onset of labor to placental delivery, neonatal outcome, complication of delivery, laboratory investigation, result of cervical swab culture, and result of histopathological examination of the placenta. The general and obstetric histories were taken from medical records, while the results of cervical swab culture and histopathological examination of the placenta were retrieved from the hospital computer database.

SPSS for Windows, version 17 (SPSS Inc, Chicago, IL, USA) was used for statistical analysis. Descriptive statistics were used to describe the qualitative and quantitative variables. The quantitative data was shown in mean ± SD and the qualitative data was shown in percentage. A chi square or Fisher exact test was used to test for significance of associations. P < 0.05 was considered significant.

We included data from 104 patients who had preterm labor with intact membranes during January 2012 to May 2013. General maternal characteristics are shown in Table 1. The majority of patients are Primigravidae. Most preterm labors occurred between 34-36 weeks of gestation, followed by 28-33 weeks. Only 2 patients had <28 weeks of gestation. The majority of patients had delivery within 48 h of the onset of labor. The most common risk factor for preterm labor in this study was low body mass index (BMI) before pregnancy, followed by teenage pregnancy, previous preterm birth, and advanced maternal age. Only 18 patients had clinical signs or symptoms of chorioamnionitis. As shown in Table 2, the most common clinical manifestation was maternal tachycardia, followed by maternal fever, maternal leukocytosis, fetal tachycardia, and foul smelling amniotic fluid.

The cervical swab culture (Table 3) revealed bacterial infection in 12 cases.

Table 4 shows the relationship between the results of cervical swab culture and the HCA. There was evidence of amniotic fluid infection detected by placental pathology in 10 cases, while there was a positive cervical culture in 12 cases. There was only one case that showed a positive result for both cervical swab culture and HCA. In cases with HCA, acute chorioamnionitis was most commonly observed and present in up to 7 of 10 cases. Fetal inflammatory response was noted in only one case. There was no significant association between cervical swab culture and evidence of amniotic fluid infection in placental pathology (P = 0.87, odds ratio 0.84, 95% confidence interval 0.1 to 7.27)

Finally, we studied the relationship between the clinical signs or symptoms of chorioamnionitis and cervical swab culture and the HCA, and found that HCA was detected in 3 out of 18 cases (17%) of clinical chorioamnionitis. Similarly, positive cervical swab culture was shown in only 2 out of 18 cases (11%) of clinical chorioamnionitis.

In this study, we found no significant relationship between cervical swab culture and histological evidence of amniotic fluid infection. Surprisingly, the rates of positive cervical swab culture and HCA were very low in this study. The low rate of positive cervical swab culture was probably because of technical pitfalls of specimen collection and transportation procedure before submission for microbiological analysis. Because this study was retrospective, there are possibly many hard-to-control factors. The process of conducting the cervical swab culture is also important to the interpretation of results [13]. Ovalle et al. [10] found that 63.5% of cervical swab cultures in preterm labor were positive, while the culture from amniotic cavity was positive in only 24%. A study of the correlation of HCA and maternal characteristics in preterm labor by Oguntemi et al. [5] found that of 254 cases of preterm labor with HCA, only 127 patients or 50% had positive result of hemoculture, urine culture, or cervical swab culture. This may imply that cervical and intra-amniotic cavity colonization and histological inflammatory responses need time to incubate from initial bacterial invasion to histological and clinical manifestation.

The low rate of HCA despite the calculated sample size and thorough histological evaluation may be explained by the differences in study methodology. Our population was based on the initial admission at the onset of preterm labor, while the placenta was subsequently collected after delivery. With early medical treatment and the difference of time to delivery, these may alter the histological features of inflammation.

The result of cervical swab culture in this study showed 11.5% of bacterial infection and 1.9% of Group Β streptococcus infection, which is low when compared with the result of a prospective study [14]. A prospective study entitled “Antenatal microbiologic and maternal risk factors associated with prematurity” was conducted in 202 women with gestational age between 20-28 weeks. It aimed to evaluate the possible influences of lower genital tract infection or bacterial conditions on obstetric outcomes. The results revealed the presence of bacterial vaginosis in 18.7%, Mycoplasma hominis 10.8%, and Staphylococcus aureus 3.0% [14]. The lower rate of bacterial infection in this study may be a consequence of the difference in gestational ages of the studied population. Nevertheless, a study from Siriraj Hospital, Thailand, [15] found that 9.8% had positive pathological organisms. A positive bacterial culture from the genitourinary tract was not found to be a significant predictor of preterm delivery, which is similar to findings in this study.

We found no significant relationship between evidence of amniotic fluid infection in placental pathology and cervical swab culture or clinical of chorioamnionitis in preterm labor with intact membranes. However, because of some limitations in this study, the role of cervical swab culture in preterm labor remains inconclusive. Further studies with improved specimen collection, transportation, and laboratory procedures are be needed to assess the role of cervical swab culture.