Relationship between Intrauterine Bacterial Infection and Early Embryonic Developmental Arrest.

BACKGROUND: Early embryonic developmental arrest is the most commonly understudied adverse outcome of pregnancy. The relevance of intrauterine infection to spontaneous embryonic death is rarely studied and remains unclear. This study aimed to investigate the relationship between intrauterine bacterial infection and early embryonic developmental arrest.
METHODS: Embryonic chorion tissue and uterine swabs for bacterial detection were obtained from 33 patients who underwent artificial abortion (control group) and from 45 patients who displayed early embryonic developmental arrest (trial group).
RESULTS: Intrauterine bacterial infection was discovered in both groups. The infection rate was 24.44% (11/45) in the early embryonic developmental arrest group and 9.09% (3/33) in the artificial abortion group. Classification analysis revealed that the highest detection rate for Micrococcus luteus in the early embryonic developmental arrest group was 13.33% (6/45), and none was detected in the artificial abortion group. M. luteus infection was significantly different between the groups (P < 0.05 as shown by Fisher's exact test). In addition, no correlation was found between intrauterine bacterial infection and history of early embryonic developmental arrest.
CONCLUSIONS: M. luteus infection is related to early embryonic developmental arrest and might be one of its causative factors.

INTRODUCTION

Early embryonic developmental arrest is a state of embryonic growth arrest in early pregnancy in which either no developed embryonic bud can be detected through B-scan or no embryonic heartbeat is detected despite the presence of embryonic bud.[1] This condition is a complication commonly observed in early pregnancy and accounts for a significant proportion of all cases of spontaneous abortion. Its occurrence has gradually increased in recent years, and it has become a subject of interest among clinicians and researchers.

Diverse factors are associated with early embryonic developmental arrest, and some of these factors remain unclear.[2] Of the identified causes, hereditary factors, immune and endocrine systems, nutrition and environment, and infection have been the focus of studies. Of all known factors, genetic abnormality provides the highest contribution to early embryonic developmental arrest.[1] The second most potent factor is immune and endocrine system, followed by nutrition and environment and infection.[345] Despite its relatively low effect, increasing studies have reported a crucial status in infective causes among the remaining unknown causative components for early embryonic developmental arrest, with directly relevant microorganisms, including Gardnerella vaginalis, Mycoplasma urealyticum, and Mycoplasma hominis.[6789]

Studies have shown that Listeria monocytogenes can cause spontaneous abortion.[10] L. monocytogenes might enter the body through ingestion of contaminated food, as well as through contact with the eyes, abraded skin, and mucous membranes. When this pathogen persists in the vagina or cervix, the bacterium may infect a pregnant host and its fetus or newborn through placental or obstetric canal transmission. L. monocytogenes infection through sexual contact may also cause abortion.[11] Although the uterus is commonly regarded as an aseptic environment, intrauterine bacterial proliferation resulting from infection or other factors might adversely affect the embryonic development. This study aimed to investigate the relationship between intrauterine bacterial infection and early embryonic developmental arrest through comparative analyses of the conditions for bacterial infection in the early embryonic developmental arrest group and artificial abortion group.

METHODS

Sample collection and ethical approval

Seventy-eight cases of early pregnancy (8–12 weeks) were identified at Peking Union Medical College Hospital between July 1, 2014 and September 30, 2014. Among these cases, 45 patients requesting for curettage because of embryonic developmental arrest were designated as the trial group and 33 patients with normal embryonic development and requesting for artificial abortion were designated as the control group. This study used the following criteria to constitute the control and trial groups. The artificial abortion group (control group) should meet all the following conditions: without history of spontaneous abortion (or embryonic diapause); regular menstrual cycle; consistency between menopausal days and results of gestational sac ultrasound; detection of fetal heart beat in preoperative ultrasonography; normal results of liver function tests; and negative result for infection index (hepatitis B, HIV, and syphilis). The following were the criteria for spontaneous abortion caused by embryonic developmental arrest (trial group): without history of spontaneous abortion (or embryonic diapause); regular menstrual cycle; and gestational sac observed through ultrasound, although embryo bud and embryo heartbeat were not detected or embryo heartbeat was detected early and eventually disappeared. If the reason for spontaneous abortion in certain cases was identified, such cases were excluded. No significant disparities in terms of age and gravidity were observed between the two groups. Vaginal cleanliness in both groups was examined prior to operation, and operation was not conducted in patients with vaginal infection or in patients recovering from postinfection treatment. This study was approved by the Ethics Committee of Peking Union Medical College Hospital. We obtained informed consent from each participant before we collected and tested the samples. Information obtained from each patient was kept strictly confidential. If the results indicated infection, the participants were treated subsequently.

Sample preparation

A routine vulval, vaginal, and cervical disinfection was performed, followed by aseptic cotton swabbing along the uterine cavity and then daubing for one cycle. Chorionic tissue was directly taken from the uterus, and the cotton swabs and tissue were placed in sterile centrifuge tubes containing 5 ml of preliminary Listeria enrichment broth 1 (LB1) (Beijing Land Bridge Technology CO., Ltd., China) for testing. Chorionic tissue was cut, ground, and cultured at 30°C for 24 h in 10 ml of replenished LB1. Following centrifugation at 200 × g for 5 s, the swab samples in the tubes were cultured at 30°C for 24 h in 10 ml of replenished LB1.

Bacterial isolation

The national standard GB 4789.30-2010 test method for food safety was used to isolate L. monocytogenes. After 24 h of culture at 30°C in 10 ml of LB1, 1 ml of each sample was transferred and sub-cultured in 10 ml of secondary Listeria enrichment broth 2 (LB2) (Beijing Land Bridge Technology CO., Ltd., China) for 24 h at 30°C. LB2 was obtained and streaked onto PALCAM agar (Beijing Land Bridge Technology CO., Ltd., China) and CHROMagar™ Listeria plates (Zhengzhou Biocell Biotechnology Co., Ltd., China) and then cultured for 24 h at 37°C. Five suspected colonies of L. monocytogenes were selected from each plate, streaked onto brain–heart infusion agar (Beijing Land Bridge Technology CO., Ltd., China) plates (in cases where no specific colonies exist, five single colonies with different morphologies were randomly chosen), and cultured for 24 h at 37°C. In accordance with the hospital's general sampling methods for assays in obstetrics and clinical gynecology, the cotton swabs or samples touched by inoculation loops were directly smeared onto China blue agar (Beijing Land Bridge Technology CO., Ltd., China) and blood agar (Beijing Land Bridge Technology CO., Ltd., China) plates for culturing without preliminary enrichment.

Bacterial identification

For observation of colony characteristics, the isolated bacteria were subjected to Gram-staining and microscopy. A Vitek 2 biochemistry identifier (bioMérieux, Marcy l’Etoile, France) was used to identify the isolated bacteria.

Statistical analysis

The Chi-square test and Fisher's exact test were used in this study. SPSS 19.0 (SPSS Inc., Chicago, USA) was used for statistical analysis, and significance was set at P < 0.05.

RESULTS

Bacterial detection in intrauterine

Table 1 shows the eight bacterial species that caused infection in the uterine cavity of the trial or control group; these species included Micrococcus luteus, which was found in six samples and showed the highest detection percentage of 7.69% (6/78). Sphingomonas paucimobilis was the second highest at 2.56% (2/78), and each of the six other species was detected once. In addition, S. paucimobilis was detected both in the trial and control groups, whereas Micrococcus kristinae and Gemella morbillorum were found only in the control group. L. monocytogenes was not detected in either the trial or control group.

Table 1

Bacterial detection in intrauterine of patients with either spontaneous or artificial abortion (N = 78)

Bacterium	Gram stain (+/−)	Cases (spontaneous/artificial), n	Infection ratio (%)
Micrococcus luteus	+	6 (6/0)	7.69
Sphingomonas paucimobilis	−	2 (1/1)	2.56
Acinetobacter baumannii	−	1 (1/0)	1.28
Micrococcus kristinae	+	1 (0/1)	1.28
Staphylococcus epidermidis	+	1 (1/0)	1.28
Pseudomonas aeruginosa	−	1 (1/0)	1.28
Ochrobactrum anthropi	−	1 (1/0)	1.28
Gemella morbillorum	+	1 (0/1)	1.28

Intrauterine infection ratio

Eleven cases of bacterial infection were detected among the 45 cases in the trial group, accounting for a detection rate of 24.44%. Three cases of infection were detected in the control group consisting of 33 cases, accounting for a detection rate of 9.09%. Chi-square test revealed an insignificant difference (P > 0.05) in intrauterine bacterial infection between the early embryonic developmental arrest group and artificial abortion group [Table 2].

Table 2

Comparison of intrauterine bacterial infection in spontaneous abortion induced by embryonic developmental arrest and artificial abortion

Experimental group	Total cases, N	Bacterium positive
		Cases, n	Infection rate (%)
Spontaneous abortion	45	11	24.44
Artificial abortion	33	3	9.09

Comparison of intrauterine Micrococcus luteus infection

Six cases (13.33%) of M. luteus infection were detected in the trial group, whereas no M. luteus infection was detected in the control group. Given that M. luteus infection was not detected in the control group, Fisher's exact test was employed. The result indicated a significant difference (P < 0.05) in intrauterine M. luteus infection between the early embryonic developmental arrest group and artificial abortion group [Table 3].

Table 3

Comparison of the incidence of intrauterine bacterial infection by Micrococcus luteus in spontaneous abortion induced by embryonic developmental arrest and artificial abortion

Experimental group	Total cases, N	Micrococcus luteus-positive
		Cases, n	Infection rate (%)
Spontaneous abortion	45	6	13.33
Artificial abortion	33	0	0

Early embryonic developmental arrest history and intrauterine infection

Table 4 shows no significant differences in the total infection rate or individual M. luteus infection rate among patients who have had early embryonic developmental arrest once, twice, and three times or more (P > 0.05).

Table 4

Frequency of spontaneous abortion induced by embryonic developmental arrest and intrauterine bacterial infection

Frequency of spontaneous abortion	Total cases, N	Bacterium positive		Micrococcus luteus-positive
		Cases, n	Infection rate (%)	Cases, n	Infection rate (%)
1	31	6	19.35	4	12.90
2	8	3	37.50	1	12.50
≥3	6	2	33.33	1	16.67

DISCUSSION

The causes of early embryonic developmental arrest in early pregnancy include genetic factors, immune and endocrine systems, nutrition and environment, and infective factors.[135] Studies have attributed the number of adverse occurrences in pregnancy, including spontaneous premature delivery, premature rupture of membranes, chorioamnionitis, and recurrent abortion, to embryonic tissue infection. However, whether intrauterine infection, including embryonic tissue infection, is a causative factor for early embryonic developmental arrest requires further investigation.

L. monocytogenes is a virulent food-borne pathogen recognized for its capability for contamination through ingestion of foodstuffs and sexual contact in pregnant women, leading to miscarriage in severe cases.[11] Through the method used to detect L. monocytogenes, this study isolated and identified cultivated bacteria from two groups of patients undergoing either early embryonic developmental arrest or artificial abortion; this work attempted to elucidate the relationship between L. monocytogenes infection and early embryonic developmental arrest. A clinical method for bacterial testing was used simultaneously, namely direct smearing onto China blue agar and blood agar plates without preliminary enrichment. No L. monocytogenes was detected, and this finding was possibly caused by the combination of inherently low morbidity of the pathogen and limited sample size. No L. monocytogenes was also detected through the clinical test method. However, given that the culture method for L. monocytogenes involved a two-time pre-enrichment step for 24 h each prior to plate cultivation, other bacterial species were detected. This finding might imply that: (1) intrauterine bacterial infections are present in certain patients; (2) low bacterial infection levels result in misdetection in samples directly smeared onto China blue and blood agar plates; and (3) certain bacteria require selective enrichment and suitable media for successful isolation. Thus, assay culture media and preliminary enrichment procedures are suggested for inclusion in clinical testing.

Seven species of intrauterine infective bacteria were detected in this study, all being conditioned pathogens. M. luteus is a Gram-positive coccus that belongs to the family Micrococcaceae and found in soil, dust, water, and air; M. luteus is also one of the normal floras found in mammalian skin.[12] However, M. luteus can cause illness when it colonizes other areas of the body. M. luteus infection induces insidious manifestation of pyogenic liver abscess,[13] meningitis,[14] and myocarditis.[15] Our results demonstrated that the history of early embryonic developmental arrest carried no significant bearing on infection, supporting the previously reported noncorrelation between mycoplasma infection and number of early embryonic developmental arrests.[8] No distinct disparity in intrauterine bacterial infection was observed between the artificial abortion group and early embryonic developmental arrest group. By contrast, six cases of M. luteus infection were detected in the early embryonic developmental arrest group, whereas none was found in the control group; this disparity was statistically significant. This finding hinted at the correlation between the presence of M. luteus in the intrauterine cavity and early embryonic developmental arrest. Based on all of the findings presented, the role of M. luteus in early embryonic developmental arrest must be continuously investigated to reduce the incidence of spontaneous abortion.

We used the same method to isolate bacterium from vaginal and cervical secretions. The bacteria found in the vaginal and cervical secretions were quite different from those isolated from the intrauterine. This difference was possibly caused by the environment. For instance, M. luteus was isolated from a relatively closed environment. Although we did not find a relationship between vaginal infection and intrauterine infection, the results showed that unclean sexual activity might lead to intrauterine infection, even though examination of vaginal secretion yielded normal findings.

Studies have shown that bacteria can migrate to and thrive in the uterus where intrauterine infection is engendered by vaginal and cervical inflammation.[16] As miscarriage induced by early embryonic developmental arrest typically occurs during early pregnancy, preconception health education and health examinations should be vigorously addressed to facilitate advance treatment of any inflammation and reduce the effects of infection on embryonic development. Concurrently, further research on the functional mechanisms of infection in embryonic development should be undertaken to bolster the theoretical foundation for the prevention of early embryonic developmental arrest.

Financial support and sponsorship

This work was funded by National “Twelfth Five-Year” Plan for Science and Technology Support (No. 2013BAI12B01 and No. 2012BAI31B06).

Conflicts of interest

There are no conflicts of interest.