Abstract
Background:
Preterm premature rupture of membrane (PPROM) can lead to serious consequences such as intrauterine infection, prolapse of the umbilical cord, and neonatal respiratory distress syndrome. Genital infection is a very important risk which closely related with PPROM. The preliminary study only made qualitative research on genital infection, but there was no deep and clear judgment about the effects of pathogenic bacteria. This study was to analyze the association of infections with PPROM in pregnant women in Shaanxi, China, and to establish Bayesian stepwise discriminant analysis to predict the incidence of PPROM.
Methods:
In training group, the 112 pregnant women with PPROM were enrolled in the case subgroup, and 108 normal pregnant women in the control subgroup using an unmatched case-control method. The sociodemographic characteristics of these participants were collected by face-to-face interviews. Vaginal excretions from each participant were sampled at 28–36+6 weeks of pregnancy using a sterile swab. DNA corresponding to Chlamydia trachomatis (CT), Ureaplasma urealyticum (UU), Candida albicans, group B streptococci (GBS), herpes simplex virus-1 (HSV-1), and HSV-2 were detected in each participant by real-time polymerase chain reaction. A model of Bayesian discriminant analysis was established and then verified by a multicenter validation group that included 500 participants in the case subgroup and 500 participants in the control subgroup from five different hospitals in the Shaanxi province, respectively.
Results:
The sociological characteristics were not significantly different between the case and control subgroups in both training and validation groups (all P > 0.05). In training group, the infection rates of UU (11.6% vs. 3.7%), CT (17.0% vs. 5.6%), and GBS (22.3% vs. 6.5%) showed statistically different between the case and control subgroups (all P < 0.05), log-transformed quantification of UU, CT, GBS, and HSV-2 showed statistically different between the case and control subgroups (P < 0.05). All etiological agents were introduced into the Bayesian stepwise discriminant model showed that UU, CT, and GBS infections were the main contributors to PPROM, with coefficients of 0.441, 3.347, and 4.126, respectively. The accuracy rates of the Bayesian stepwise discriminant analysis between the case and control subgroup were 84.1% and 86.8% in the training and validation groups, respectively.
Conclusions:
This study established a Bayesian stepwise discriminant model to predict the incidence of PPROM. The UU, CT, and GBS infections were discriminant factors for PPROM according to a Bayesian stepwise discriminant analysis. This model could provide a new method for the early predicting of PPROM in pregnant women.
Keywords: Bayesian Stepwise Discriminant Analysis, Etiological Factors, Infection, Preterm Premature Rupture of Membranes
INTRODUCTION
Preterm premature rupture of membrane (PPROM) is a common perinatal complication in pregnant women. It is responsible for one-third of all preterm births. The worldwide prevalence of PPROM ranges from 2% to 10%. PPROM might occur among women of the reproductive age group, especially during the period of pregnancy before 37 weeks of gestation. PPROM can cause maternal and fetal infection in pregnant women and their unborn children, a lower Apgar score, pulmonary hypoplasia, preterm delivery, and a low birth weight. However, the etiology of PPROM is unclear. PPROM may be caused by cervical incompetence, genital infections, and uterine abnormality. Some studies have shown that a history of PPROM, race, smoking status, poor nutrition, and genital infection are risk factors for PPROM. This study developed a model to explore genital infections that might activate inflammatory cells and then induce PPROM. The etiologies of genital infection include Chlamydia trachomatis (CT), Ureaplasma urealyticum (UU), Candida albicans, syphilis, Neisseria gonorrhoea (NG), group B streptococci (GBS), herpes simplex virus (HSV), and bacterial vaginosis (BV).[1,2]
Genital infections might cause a release of cytokines and other inflammatory mediators that may weaken the membrane and cause PPROM. Studies by Chow and Blas showed that CT infection was associated with the occurrence of PPROM.[3,4] Pregnant women with BV more readily developed PPROM than women without BV.[4,5,6] Candidiasis infection in pregnant women with PPROM is controversial, and a recent study showed that the treatments for candidiasis might reduce the incidence of PPROM.[7] Pregnant women who were infected with NG had a six-time higher risk of developing PPROM than women without NG infection. GBS might cause the activation of inflammatory cells in fetal membranes, which could lead to PPROM.[5,8]
Although some studies have reported that PPROM was related to genital infections, the proportions of women with confirmed genital infections with or without PPROM in China are unknown.[3,4,5,7,8,9] Studies of the relationship between genital infection and PPROM are still rare. This study aimed to determine the association between etiological infection and PPROM. Discriminant analysis is a multivariate statistical method that can distinguish newly acquired samples according to the quantitative characteristics of the existing observational sample. In this study, a Bayesian stepwise discriminant model was established, and a corresponding linear discriminant function was built. This model could predict and reduce the occurrence of PPROM.
METHODS
Ethical approval
The study was conducted in accordance with the Declaration of Helsinki and was approved by Institutional Review Board of Shaanxi Provincial People's Hospital. Informed written consent was obtained from all the participants before their enrolment in this study.
Study design and participants
Abnormal vaginal discharge was examined in each of the participants. The quantitative levels of CT, UU, NG, C. albicans, GBS, HSV-1, and HSV-2 were detected in each of the participants. Based on the etiological detection, a type of linear discriminant analysis was used to discriminate between normal pregnant women and those with PPROM. The accuracy of the Bayesian stepwise discriminant model was validated by both a training group (including 112 cases in case subgroup and 108 cases in control subgroup) and a multicenter validation group (including 500 cases in case subgroup and 500 cases in control subgroup). An unmatched case-control design was used in this study. Inclusion criteria for normal pregnant women were as follows: women with 28–36+6 weeks of gestation, no use of any antibiotics within 2 months, and no history of any chronic diseases (such as diabetes, cardiovascular disease, and hypertension). Inclusion criteria for the PPROM patients were as follows: pregnant women with 28–36+6 weeks of gestation, membrane rupture within 12 h, no use of any antibiotics within 2 months, and no history of any basic diseases (such as diabetes, cardiovascular disease, and hypertension). The PPROM is defined as the onset of amniotic fluid leakage from the vagina before the onset of uterine contractions at less than 37 weeks’ gestational age.[8] The PPROM includes having a history of drainage of clear fluid that wets the perineum and runs along the thighs and legs as well as a sterile speculum examination showing fluid pooling in the posterior vaginal fornix or fluid freely flowing from the cervix. The laboratory definition of PPROM is positivity for insulin-like growth factor-binding protein 1 in the vaginal discharge.[9]
In the preliminary study, 112 pregnant women with PPROM and 108 normal pregnant women were randomly recruited from Department of Obstetrics and Gynecology, Shaanxi Provincial People's Hospital between June 2011 and May 2012. Bayesian stepwise discriminant analysis was used to analyze the etiological infections of CT, UU, C. albicans, GBS, HSV-1, and HSV-2. A multicenter validation group included 500 pregnant women with PPROM (case subgroup) and 500 normal pregnant women (control subgroup) from five different hospitals in the Shaanxi province between June 2012 and January 2013, respectively. These five hospitals were Northwest Women and Children Hospital, Xi’an Fourth Hospital, Xi’an Gaoxin Hospital, Chang’an Hospital, and Xianyang 215 Hospital in Shaanxi province.
Data and specimen collection
Face-to-face questionnaires were used to collect the sociodemographic characteristics (including age, gravidity, parity, marital status, and occupation) and gynecological histories (including obstetric history, past history of PPROM, and history of trauma to the cervix). A vaginal swab and a cervical swab were collected within 12 h of membrane rupture of PPROM cases, and the control group were collected at 28–36+6 weeks of gestation during routine examination.
Nucleic acid extraction
Each swab was suspended in the 1.5 ml of sterile saline (0.85%). Nucleic acid was extracted from the swab specimens using QIAamp MiniStool kit (QIAGEN, Hilden, Germany) following the manufacturer's instruction, and the DNA was eluted in the 45 μl of elution buffer.
Quantitation of etiological agents by real-time polymerase chain reaction
Real-time polymerase chain reaction (PCR; Triplex International Biosciences Co., LTD., China), following the manufacturer's instructions, was used to detect NG, UU, CT, GBS, and C. albicans in the vaginal swabs and HSV-1 and HSV-2 in the cervical swabs. The threshold of detection of the PCR was equal to or greater than 103 copies/ml.
Antibody against HIV and syphilis detection
All participants’ serum was collected to detect antibody against HIV and syphilis using enzyme linked immunosorbent assay (ELISA) kits (Shanghai Kehua Bioengineering Co., Ltd., China).
Statistical analysis
The data were analyzed using SPSS version 19.0 software (SPSS Inc., Chicago, IL, USA). Data were considered to have a normal distribution if the P value given by the Shapiro-Wilk test was more than 0.05. If the test data set did not show a normal distribution, the data could be normalized by logarithmic transformation. The mean levels of infectious agents were compared using the Wilcoxon two-sample test method. The Chi-square test was used to analyze the differences between categorical data. The value of P < 0.05 was considered to be statistically significant.
Furthermore, the original data for the etiological factors of PPROM were log-transformed and then translated as discriminant functions. The translated data were analyzed by a forward selection method (sle = 0.1, sls = 0.1). Significant variables were identified by Bayesian stepwise discriminant analysis.
The quantitative levels of the etiological agents for each of the pregnant women were skewed. Hence, these data were converted into a log-normal distribution. Linear combinations of data were used to form discriminant functions for the separation of categories by minimization of the within-class and between-class ratios of the sum of squares. Bayesian stepwise discriminant analysis was used to distinguish normal pregnant women from those with PPROM. Forward stepwise analysis was used to select significant variables for the discriminant analysis. An obvious difference in the selected variables was observed when the translated variables were used.
RESULTS
Baseline characteristics of the training group
There were no significant differences in age, gravidity, parity, marital status, and occupation between the normal pregnant women (control subgroup) and those with PPROM (case subgroup) in the training group [all P > 0.05, Table 1].
Table 1.
Baseline characteristics of all participates in training group of this study
| Characteristics | Case subgroup (n = 112) | Control subgroup (n = 108) | χ2 | P |
|---|---|---|---|---|
| Marital status, n (%) | 0.303 | 0.860 | ||
| Single | 2 (1.8) | 1 (0.9) | ||
| Married/cohabiting | 109 (97.3) | 106 (98.2) | ||
| Divorced/separated | 1 (0.9) | 1 (0.9) | ||
| Age, n (%) | 0.897 | 0.085 | ||
| <20 years | 1 (0.9) | 1 (0.9) | ||
| ≥20 years and <35 years | 89 (79.5) | 91 (84.3) | ||
| ≥35 years | 22 (19.6) | 16 (14.8) | ||
| Gravidity, n (%) | 0.598 | 0.439 | ||
| Primegravidae | 85 (75.9) | 77 (71.3) | ||
| Gravida ≥2 | 27 (24.1) | 31 (28.7) | ||
| Parity, n (%) | 0.429 | 0.513 | ||
| Primipara | 94 (83.9) | 87 (80.6) | ||
| Secondary ≥2 | 18 (16.1) | 21 (19.4) | ||
| Occupation, n (%) | 0.001 | 0.984 | ||
| House wife | 52 (46.4) | 50 (46.3) | ||
| Employee/business | 60 (53.6) | 58 (53.7) |
Univariate analysis of the etiologic agents in the training group
All participants were negative NG, HIV, and syphilis. In the training group, there were significant differences in the positive rates of abnormal vaginal discharge, UU, CT, and GBS between the normal pregnant women and those with PPROM [all P < 0.05; Table 2]. To study the effects of different etiological agents on PPROM, the quantitative levels of UU, CT, HSV-2, and GBS were converted into log-normal distribution data. The quantitative levels of UU, CT, GBS, and HSV-2 showed significant differences between the normal pregnant women and those with PPROM [all P < 0.05; Table 3]. However, the C. albicans and HSV-1 distributions were not significantly different between the normal pregnant women and those with PPROM [all P > 0.05, Table 3]. Positive rates of each etiological agent were analyzed using Chi-square test. The translated data were analyzed using Wilcoxon two-sample test method. Then, the translated data were separately analyzed by a forward selection method, and significant variables were selected for the Bayesian stepwise discriminant analysis.
Table 2.
Infectious status of training group in this study
| Variables | Case subgroup (n = 112) | Control subgroup (n = 108) | χ2 | P |
|---|---|---|---|---|
| Abnormal vaginal discharge, n (%) | 17.774 | 0.001 | ||
| Yes | 53 (47.3) | 22 (20.4) | ||
| No | 59 (52.7) | 86 (79.6) | ||
| Candida albicans, n (%) | 2.711 | 0.100 | ||
| Positive | 7 (6.3) | 2 (1.9) | ||
| Negative | 105 (93.7) | 106 (98.1) | ||
| HSV-2, n (%) | 0.376 | 0.617 | ||
| Positive | 1 (0.9) | 2 (1.9) | ||
| Negative | 111 (99.1) | 106 (98.1) | ||
| UU, n (%) | 4.817 | 0.028 | ||
| Positive | 13 (11.6) | 4 (3.7) | ||
| Negative | 99 (88.4) | 104 (96.3) | ||
| CT, n (%) | 7.105 | 0.008 | ||
| Positive | 19 (17.0) | 6 (5.6) | ||
| Negative | 93 (83.0) | 102 (94.4) | ||
| GBS, n (%) | 11.098 | 0.001 | ||
| Positive | 25 (22.3) | 7 (6.5) | ||
| Negative | 87 (77.7) | 101 (93.5) | ||
| HSV-1, n (%) | 1.218 | 0.270 | ||
| Positive | 5 (4.5) | 2 (1.9) | ||
| Negative | 107 (95.5) | 106 (98.1) |
UU: Ureaplasma urealyticum; CT: Chlamydia trachomatis; GBS: Group B Streptococci; HSV-1: Herpes simplex virus type 1; HSV-2: Herpes simplex virus type 2.
Table 3.
Distribution of etiological agents in training group of this study
| Etiological agents | Case subgroup* (n = 112) | Control subgroup* (n = 108) | Z | P |
|---|---|---|---|---|
| UU | 6.16 | 5.03 | 3.534 | 0.002 |
| CT | 5.75 | 3.72 | 11.521 | 0.001 |
| GBS | 7.93 | 6.51 | 3.249 | 0.004 |
| Candida albicans | 4.66 | 4.66 | 0.168 | 0.569 |
| HSV-1 | 3.59 | 3.61 | –0.264 | 0.795 |
| HSV-2 | 4.12 | 3.72 | 3.586 | 0.002 |
*The quantitative levels of UU, CT, HSV-2, and GBS were converted into log-normal distribution data. UU: Ureaplasma urealyticum; CT: Chlamydia trachomatis; GBS: Group B streptococci; HSV-1: Herpes simplex virus type 1; HSV-2: Herpes simplex virus type 2.
Bayesian stepwise discriminant analysis
The Bayesian stepwise discriminant analysis is described in statistical language as follows: Assume g populations follow g multivariate normal distributions. Probability of misclassifying a subject in class i into class j, P(i|j); Loss due to misclassification, a(j|i). The Bayesian criterion: minimize the expected misclassification loss.
Classification function:
f1 = a10 + a11x1 + ... + a1pxp
f2 = a20 + a21x1 + ... + a2pxp
fg = ag0 + ag1x1 + ... + agpxp
aj0, aj1., ajp(j = 1,2.... g): the parameters to be estimated; fji represents positively related to the probability of being in the jth population.
Bayesian stepwise discriminant analysis was used to establish a function using retrospective data, and then the individual observation indices were introduced into the equation, according to the results of the individual, to infer the type of a statistical method.
Two Bayesian function equations were established based on the discriminant coefficients. To investigate the contribution of the etiological factors, the tests of equality of three groups (UU, CT, and GBS) were statistically different (P < 0.05), then Bayesian discriminant method could be carried out. The significance test of the discriminant function are shown in Table 4, Wilks’ λ value was 0.530, Chi-square value was 137.535, so the discriminant result was proved to be effective. The classification function of Bayesian model was established as follows:
Table 4.
Results of Bayesian stepwise discriminant function
| Variables | F | P | Coefficients | Wilks’ λ | χ2 | P | |
|---|---|---|---|---|---|---|---|
| PPROM | Non-PPROM | ||||||
| UU | 52.999 | <0.001 | 0.441 | 0.381 | 0.530 | 137.535 | <0.001 |
| CT | 31.247 | <0.001 | 3.347 | 3.263 | |||
| GBS | 125.065 | <0.001 | 4.126 | 2.642 | |||
UU: Ureaplasma urealyticum; CT: Chlamydia trachomatis; GBS: Group B streptococci; PPROM: Preterm premature rupture of membranes.
f1 = −95.383+0.441x1 + 3.347x2 + 4.126x3
f2 = −71.580+0.381x1 + 3.263x2 + 2.642x3
X1 is the distribution of UU, X2 is the distribution of CT, and X3 is the distribution of GBS. f1 is the function for the PPROM group, and f2 is the function for the non-PPROM group.
The results of the Bayesian stepwise discriminant analysis showed that UU, CT, and GBS infections were key factors that could affect the occurrence of PPROM, with coefficients of 0.441, 3.347, and 4.126, respectively [Table 4]. According to the Bayesian stepwise discriminant model, associations were observed among UU, CT, and GBS infections, and PPROM. No associations were found among HSV-1, HSV-2, C. albicans, and PPROM. The Bayesian stepwise discriminant analysis was used to differentiate normal pregnant women from those with PPROM. The results showed that the accuracy of this method was 84.1%.
Validation of Bayesian stepwise discriminant analysis
All classification rules developed through Bayesian stepwise discriminant analysis should be prospectively validated before their use in clinical practice; therefore, we designed a prospective validation group. There were no significant differences in age, gravidity, parity, marital status, and occupation between the normal pregnant women (control subgroup, 500 cases) and those with PPROM (case subgroup, 500 cases) in the validation group [all P > 0.05, Table 5]. The distributions of abnormal vaginal discharge, UU, CT, GBS, and C. albicans showed significant differences between two subgroups [all P < 0.05, Table 6]. However, the distributions of HSV-2 and HSV-1 were not significantly different [all P > 0.05, Table 6]. The log-transformed quantification of quantitative levels of UU, CT, HSV-2, and GBS showed statistical differences between the case and control groups but C. albicans did not show statistical difference [Table 7]. After the bias discriminant function cross-validation, the accuracy of this method was 86.8% to separate normal pregnant women and PPROM women in validation group [Table 8].
Table 5.
Baseline characteristics of all participants in validation group of this study
| Characteristics | Case subgroup (n = 500) | Control subgroup (n = 500) | χ2 | P |
|---|---|---|---|---|
| Marital status, n (%) | 1.001 | 0.606 | ||
| Single | 4 (0.8) | 2 (0.4) | ||
| Married/cohabiting | 495 (99.0) | 496 (99.2) | ||
| Divorce/separated | 1 (0.2) | 2 (0.4) | ||
| Age, n (%) | 1.450 | 0.484 | ||
| <20 years | 6 (1.2) | 6 (1.2) | ||
| ≥20 years and <35 years | 432 (86.4) | 444 (88.8) | ||
| ≥35 years | 62 (12.4) | 50 (10.0) | ||
| Gravidity, n (%) | 2.067 | 0.150 | ||
| Primegravidae | 324 (64.8) | 302 (60.4) | ||
| Gravida ≥2 | 176 (35.2) | 198 (39.6) | ||
| Parity, n (%) | 2.569 | 0.109 | ||
| Primipara | 417 (83.4) | 435 (87.0) | ||
| Secondary ≥2 | 83 (16.6) | 65 (13.0) | ||
| Occupation, n (%) | 0.065 | 0.799 | ||
| Home maker | 219 (43.8) | 215 (43.0) | ||
| Employed/business | 281 (56.2) | 285 (57.0) |
Table 6.
Infectious status of validation group in this study
| Variables | Case subgroup (n = 500) | Control subgroup (n = 500) | χ2 | P |
|---|---|---|---|---|
| Abnormal vaginal discharge, n (%) | 9.019 | 0.003 | ||
| Yes | 178 (35.6) | 134 (26.8) | ||
| No | 322 (64.4) | 366 (73.2) | ||
| Candida albicans, n (%) | 4.384 | 0.036 | ||
| Positive | 23 (4.6) | 11 (2.2) | ||
| Negative | 477 (95.4) | 489 (97.8) | ||
| HSV-2, n (%) | 2.016 | 0.156 | ||
| Positive | 6 (1.2) | 2 (0.4) | ||
| Negative | 494 (98.8) | 498 (99.6) | ||
| UU, n (%) | 22.353 | 0.001 | ||
| Positive | 67 (13.4) | 24 (4.8) | ||
| Negative | 433 (86.6) | 476 (95.2) | ||
| CT, n (%) | 23.349 | 0.001 | ||
| Positive | 57 (11.4) | 17 (3.4) | ||
| Negative | 443 (88.6) | 483 (96.6) | ||
| GBS, n (%) | 36.424 | 0.001 | ||
| Positive | 78 (15.6) | 21 (4.2) | ||
| Negative | 422 (84.4) | 479 (95.8) | ||
| HSV-1, n (%) | 0.504 | 0.478 | ||
| Positive | 5 (1.0) | 3 (0.6) | ||
| Negative | 495 (99.0) | 497 (99.4) |
UU: Ureaplasma urealyticum; CT: Chlamydia trachomatis; GBS: Group B streptococci; HSV-2: Herpes simplex virus type 2; HSV-1: Herpes simplex virus type 1.
Table 7.
Distribution of etiological agents in validation group in this study
| Etiological agents | Case subgroup* (n = 500) | Control subgroup* (n = 500) | Z | P |
|---|---|---|---|---|
| UU | 7.70 | 5.55 | 7.280 | 0.001 |
| CT | 6.12 | 3.95 | 11.183 | 0.001 |
| GBS | 8.44 | 7.55 | 4.453 | 0.001 |
| Candida albicans | 4.50 | 4.70 | −2.291 | 0.091 |
| HSV-1 | 3.17 | 3.22 | −0.626 | 0.539 |
| HSV-2 | 3.78 | 3.37 | 4.684 | 0.001 |
*The quantitative levels of UU, CT, HSV-2, and GBS were converted into log-normal distribution data. UU: Ureaplasma urealyticum; CT: Chlamydia trachomatis; GBS: Group B streptococci; HSV-1: Herpes simplex virus type 1; HSV-2: Herpes simplex virus type 2.
Table 8.
Cross validation of training and validation groups in this study
| Groups | PPROM | Non-PPROM | Total |
|---|---|---|---|
| Training group*, n (%) | |||
| Case subgroup | 97 (81.5) | 15 (18.5) | 112 (100.0) |
| Control subgroup | 20 (18.4) | 88 (86.6) | 108 (100.0) |
| Validation group†, n (%) | |||
| Case subgroup | 426 (85.2) | 74 (14.8) | 500 (100.0) |
| Control subgroup | 66 (13.2) | 434 (86.8) | 500 (100.0) |
*The accuracy was 84.1% to separate normal pregnant women and PPROM women in training group; † The accuracy was 86.8% to separate normal pregnant women and PPROM women in validation group. PPROM: Preterm premature rupture of membranes.
DISCUSSION
The mechanisms of PPROM are unclear. The presence of infections may cause PPROM through the release of inflammatory cytokines and proteases.[6,9] Indeed, genital infection has been identified as a risk factor for PPROM. Infection may impair the antimicrobial effect of the pregnant cervix, making it more susceptible to other microbes.[10,11] This study investigated the associations between the selected genital infections (abnormal vaginal discharge, UU, CT, GBS, NG, C. albicans, HSV-1, HSV-2, HIV, and syphilis) and PPROM in Shaanxi province, China. The distributions of abnormal vaginal discharge, UU, CT and GBS were significantly different between normal pregnant women and those with PPROM. However, the distributions of C. albicans, HSV-1, HSV-2, were not significantly different between normal pregnant women and those with PPROM. The relationship between abnormal vaginal discharge and PPROM has been reported in other studies.[12,13,14] some studies have shown that C. albicans was protective against PPROM.[11,15] Pregnant women with C. albicans were 73% less likely to develop PPROM than pregnant women without C. albicans. The possible reason for this finding might be that the amniotic fluid washes out the yeast cells, which could lead to negative results.[16,17,18] We found that the positive rate of C. albicans had no statistical difference between two subgroups in the training group, but had significant different between two subgroups in the validation group, and bias might be caused by the sample scale. Different outcomes were reported about the relationship between HSV-1 and HSV-2 with PPROM.[19] In the training group, the positive rate of HSV-2 showed no statistical difference, but the quantitative level of HSV-2 was significantly different between normal pregnant women and those with PPROM. This result might be caused by the use of different detection methods for HSV-1 and HSV-2 in pregnant women.[20,21,22]
Some studies have shown that infections with HIV, syphilis, and NG might be risk factors for PPROM in pregnant women.[8,22] In this study, all participants were HIV, syphilis and NG negative, so these three pathogens were not included in the analysis of this study.
This study showed that CT infection was associated with PPROM. Some studies have shown that CT infection of pregnant women could cause release of inflammatory mediators that could be implicated in membrane rupture.[15,23,24] Some studies have shown that infection with GBS might release cytokines and other inflammatory modulators which could cause membrane rupture.[23,25] This study found an association between PPROM and GBS. In this study, the prevalence rates of GBS in the women with PPROM ranged from 4.2% to 22.3%, similar with the results of other studies.[14,26,27]
The associations between etiological factors and PPROM are still unclear, and no tool is available to evaluate the association between quantitative levels of etiological agents and PPROM.[24,28] In this study, we established a Bayesian stepwise discriminant model to identify normal pregnant women and those with PPROM. We found that CT, UU and GBS infections were associated with PPROM. Using this method, 84.1% and 86.8% of the pregnant women with PPROM could be distinguished from the normal pregnant women in the training and validation groups, respectively. However, the cause of PPROM is complicated, only main etiological agents were involved in this study, but noninfectious factors were not included, so some pregnant women with PPROM could not be distinguished from normal pregnant women using this Bayesian stepwise discriminant analysis.
In this study, a Bayesian stepwise discriminant model was established to predict the incidence of PPROM. The UU, CT, and GBS infections were discriminant factors for PPROM according to a Bayesian stepwise discriminant analysis. This model could provide a new method for the early predicting of PPROM in pregnant women to hopefully reduce the incidence of PPROM.
Financial support and sponsorship
This work was supported by grants from the Natural Science Found of Shaanxi Province (No. 2014JM2-8200 and No. 2010JM4031).
Conflicts of interest
There are no conflicts of interest.
Footnotes
Edited by: Xin Chen
REFERENCES
- 1.Luo X, Pan J, Wang L, Wang P, Zhang M, Liu M, et al. Epigenetic regulation of lncRNA connects ubiquitin-proteasome system with infection-inflammation in preterm births and preterm premature rupture of membranes. BMC Pregnancy Childbirth. 2015;15:35. doi: 10.1186/s12884-015-0460-0. doi: 10.1186/s12884-015-0460-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ferrazzi E, Muggiasca ML, Fabbri E, Fontana P, Castoldi F, Lista G, et al. Assessment of fetal inflammatory syndrome by “classical” markers in the management of preterm labor: A possible lesson from metabolomics and system biology. J Matern Fetal Neonatal Med. 2012;25:54–61. doi: 10.3109/14767058.2012.716984. doi: 10.3109/14767058.2012.716984. [DOI] [PubMed] [Google Scholar]
- 3.Chow JM, Kang MS, Samuel MC, Bolan G. Assessment of the association of Chlamydia trachomatis infection and adverse perinatal outcomes with the use of population-based chlamydia case report registries and birth records. Public Health Rep. 2009;124(Suppl 2):24–30. doi: 10.1177/00333549091240S205. doi: 10.1177/00333549091240S205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Blas MM, Canchihuaman FA, Alva IE, Hawes SE. Pregnancy outcomes in women infected with Chlamydia trachomatis: A population-based cohort study in Washington State. Sex Transm Infect. 2007;83:314–8. doi: 10.1136/sti.2006.022665. doi: 10.1136/sti.2006.022665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Li W, Han L, Yu P, Ma C, Wu X, Moore JE, et al. Molecular characterization of skin microbiota between cancer cachexia patients and healthy volunteers. Microb Ecol. 2014;67:679–89. doi: 10.1007/s00248-013-0345-6. doi: 10.1007/s00248-013-0345-6. [DOI] [PubMed] [Google Scholar]
- 6.Ganor-Paz Y, Kailer D, Shechter-Maor G, Regev R, Fejgin MD, Biron-Shental T, et al. Obstetric and neonatal outcomes after preterm premature rupture of membranes among women carrying group B streptococcus. Int J Gynaecol Obstet. 2015;129:13–6. doi: 10.1016/j.ijgo.2014.10.024. doi: 10.1016/j.ijgo.2014.10.024. [DOI] [PubMed] [Google Scholar]
- 7.Roberts CL, Rickard K, Kotsiou G, Morris JM. Treatment of asymptomatic vaginal candidiasis in pregnancy to prevent preterm birth: An open-label pilot randomized controlled trial. BMC Pregnancy Childbirth. 2011;11:18. doi: 10.1186/1471-2393-11-18. doi: 10.1186/1471-2393-11- [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mercer BM. Preterm premature rupture of the membranes: Diagnosis and management. Clin Perinatol. 2004;31:765–82, vi. doi: 10.1016/j.clp.2004.06.004. doi: 10.1016/j.clp.2004.06.004. [DOI] [PubMed] [Google Scholar]
- 9.Xing X, Wenli G. Obsterics and Gynecology. 8th ed. Beijing: People's Health Press; 2013. pp. 133–4. [Google Scholar]
- 10.Sweeney EL, Kallapur SG, Gisslen T, Lambers DS, Chougnet CA, Stephenson SA, et al. Placental infection with Ureaplasma species is associated with histologic chorioamnionitis and adverse outcomes in moderately preterm and late-preterm infants. J Infect Dis. 2016;213:1340–7. doi: 10.1093/infdis/jiv587. doi: 10.1093/infdis/jiv587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lorthe E, Quere M, Kayem G. Prolonged latency after preterm premature rupture of membranes: An independent risk factor for neonatal sepsis? Am J Obstet Gynecol. 2017;216:84. doi: 10.1016/j.ajog.2016.08.022. doi: 10.1016/j.ajog.2016.08.022. [DOI] [PubMed] [Google Scholar]
- 12.Nakubulwa S, Kaye DK, Bwanga F, Tumwesigye NM, Mirembe FM. Genital infections and risk of premature rupture of membranes in Mulago Hospital, Uganda: A case control study. BMC Res Notes. 2015;8:573. doi: 10.1186/s13104-015-1545-6. doi: 10.1186/s13104-015-1545-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Qin L, Li XX, Zhang LX, Zhang LL, Xi FY, Yan L, et al. Preterm premature rupture of membranes vaginal abnormal effects on maternal and infant outcomes (in Chinese) Shaanxi Med J. 2013;42:1220–1. doi: 10.3969/j.issn.1000-7377.2013.09.051. [Google Scholar]
- 14.French JI, McGregor JA, Draper D, Parker R, McFee J. Gestational bleeding, bacterial vaginosis, and common reproductive tract infections: Risk for preterm birth and benefit of treatment. Obstet Gynecol. 1999;93:715–24. doi: 10.1016/s0029-7844(98)00557-2. doi: 10.1016/S0029-7844(98)00557-2. [DOI] [PubMed] [Google Scholar]
- 15.Dadkhah F, Kashanian M, Eliasi G. A comparison between the pregnancy outcome in women both with or without threatened abortion. Early Hum Dev. 2010;86:193–6. doi: 10.1016/j.earlhumdev.2010.02.005. doi: 10.1016/j.earlhumdev.2010.02.005. [DOI] [PubMed] [Google Scholar]
- 16.Palacio M, Kühnert M, Berger R, Larios CL, Marcellin L. Meta-analysis of studies on biochemical marker tests for the diagnosis of premature rupture of membranes: Comparison of performance indexes. BMC Pregnancy Childbirth. 2014;14:183. doi: 10.1186/1471-2393-14-183. doi: 10.1186/1471-2393-14- [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lorthe E, Ancel PY, Torchin H, Kaminski M, Langer B, Subtil D, et al. Impact of latency duration on the prognosis of preterm infants after preterm premature rupture of membranes at 24 to 32 weeks’ gestation: A National population-based cohort study. J Pediatr. 2017;182:47–52.e2. doi: 10.1016/j.jpeds.2016.11.074. doi: 10.1016/j.jpeds.2016.11.07. [DOI] [PubMed] [Google Scholar]
- 18.Surve MV, Anil A, Kamath KG, Bhutda S, Sthanam LK, Pradhan A, et al. Membrane vesicles of group B streptococcus disrupt feto-maternal barrier leading to preterm birth. PLoS Pathog. 2016;12:e1005816. doi: 10.1371/journal.ppat.1005816. doi: 10.1371/journal.ppat.1005816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cordeiro CN, Althaus J, Burke A, Argani C. Herpes simplex virus cervicitis mimicking preterm premature rupture of membranes. Obstet Gynecol. 2015;126:378–80. doi: 10.1097/AOG.0000000000000700. doi: 10.1097/AOG.0000000000000700. [DOI] [PubMed] [Google Scholar]
- 20.Furman B, Shoham-Vardi I, Bashiri A, Erez O, Mazor M. Clinical significance and outcome of preterm prelabor rupture of membranes: Population-based study. Eur J Obstet Gynecol Reprod Biol. 2000;92:209–16. doi: 10.1016/s0301-2115(99)00257-2. doi: 10.1016/S0301-2115(99)00257-2. [DOI] [PubMed] [Google Scholar]
- 21.van der Ham DP, van Kuijk S, Opmeer BC, Willekes C, van Beek JJ, Mulder AL, et al. Can neonatal sepsis be predicted in late preterm premature rupture of membranes? Development of a prediction model. Eur J Obstet Gynecol Reprod Biol. 2014;176:90–5. doi: 10.1016/j.ejogrb.2014.02.003. doi: 10.1016/j.ejogrb.2014.02.003. [DOI] [PubMed] [Google Scholar]
- 22.Waters TP, Mercer B. Preterm PROM: Prediction, prevention, principles. Clin Obstet Gynecol. 2011;54:307–12. doi: 10.1097/GRF.0b013e318217d4d3. doi: 10.1097/GRF.0b013e318217d4d3. [DOI] [PubMed] [Google Scholar]
- 23.Alger LS, Lovchik JC, Hebel JR, Blackmon LR, Crenshaw MC. The association of Chlamydia trachomatis, Neisseria gonorrhoeae, and group B streptococci with preterm rupture of the membranes and pregnancy outcome. Am J Obstet Gynecol. 1988;159:397–404. doi: 10.1016/s0002-9378(88)80093-0. doi: 10.1016/S0002-9378(88)80093-0. [DOI] [PubMed] [Google Scholar]
- 24.Yan SF, Liu XY, Cheng YF, Li ZY, Ou J, Wang W, et al. Relationship between intrauterine bacterial infection and early embryonic developmental arrest. Chin Med J. 2016;129:1455–8. doi: 10.4103/0366-6999.183411. doi: 10.4103/0366-6999.183411. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Eleje GU, Adinma JI, Ugwuanyi DC, Ikechebelu JI, Okafor CI, Ezeama CO, et al. Genital tract microbial isolate in women with preterm pre-labour rupture of membranes in resource-constrained community setting. J Obstet Gynaecol. 2015;35:465–8. doi: 10.3109/01443615.2014.970145. doi: 10.3109/01443615.2014.970145. [DOI] [PubMed] [Google Scholar]
- 26.Mikamo H, Sato Y, Hayasaki Y, Kawazoe K, Hua YX, Tamaya T, et al. Bacterial isolates from patients with preterm labor with and without preterm rupture of the fetal membranes. Infect Dis Obstet Gynecol. 1999;7:190–4. doi: 10.1155/S1064744999000320. doi: 10.1155/S1064744999000320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Dechen TC, Sumit K, Ranabir P. Correlates of vaginal colonization with group B streptococci among pregnant women. J Glob Infect Dis. 2010;2:236–41. doi: 10.4103/0974-777X.68536. doi: 10.4103/0974-777X.68536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Ye F, Chen ZH, Chen J, Liu F, Zhang Y, Fan QY, et al. Chi-squared automatic interaction detection decision tree analysis of risk factors for infant anemia in Beijing, China. Chin Med J. 2016;129:1193–9. doi: 10.4103/0366-6999.181955. doi: 10.4103/0366-6999.181955. [DOI] [PMC free article] [PubMed] [Google Scholar]