Anda di halaman 1dari 7


Available online at

Journal of the Chinese Medical Association xx (2017) 1e7

Original Article

Association between preterm labor and genitourinary tract infections caused

by Trichomonas vaginalis, Mycoplasma hominis, Gram-negative bacilli, and
Alaa El-Dien M.S. Hosny a, Waleed El-khayat b, Mona T. Kashef a,*, Mohsen N. Fakhry c
Department of Microbiology and Immunology, Faculty of Pharmacy, Cairo University, Cairo, Egypt
Obstetrics & Gynecology Department, Faculty of Medicine, Cairo University, Cairo, Egypt
Microbiologist at Chemipharm Pharmaceutical Industries, Cairo, Egypt
Received July 19, 2016; accepted October 7, 2016


Background: Preterm labor (PTL) is responsible for most cases of neonatal death. In most of these cases, the causes of PTL have not been
established although several risk factors have been described. Therefore, the aim of this study was to investigate risk factors for PTL before 37
gestational weeks among Egyptian women.
Methods: In this case-control study, 117 pregnant women without risk factors for PTL were chosen. The control group (n ¼ 45) had term labor
(gestational weeks  37 weeks), and the case group (n ¼ 72) had PTL (gestational weeks < 37 weeks). The two groups were screened for urinary
and vaginal infections. The role of different demographic characteristics, patient history, and clinical signs were also investigated.
Results: Several risk factors were identified in this study, including age < 20 years, nulliparity, previous abortion and previous preterm birth, menses
vaginal bleeding, a vaginal pH > 5, a positive whiff test, Trichomonas vaginalis infection, Mycoplasma hominis infection, coryneforms heavy
vaginal growth, and any vaginal growth of Gram-negative bacilli. Urinary tract infection with any colony count was not associated with PTL.
Conclusion: Our study demonstrated that the main risk factors for PTL were vaginal infection with T. vaginalis, M. hominis, coryneforms, and
Gram-negative bacilli, and their determinants (vaginal pH > 5, positive whiff test, heavy vaginal bleeding). Both young age (< 20 years) and
poor obstetric history were also the risk factors. Therefore, screening for genitourinary tract infections is strongly recommended to be included in
prenatal care.
Copyright © 2017, the Chinese Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND
license (

Keywords: infection; preterm labor; risk factors; urinary tract infection; vaginal

1. Introduction (PTB). PTB causes most of neonatal deaths and different

forms of neonatal morbidities.1
Preterm labor (PTL) is labor which occurs before 37 The causes of PTB in most cases have not been established
completed weeks of gestation and can lead to preterm birth although several risk factors have been identified.2 These
factors include: (1) poor socioeconomic status and low edu-
cation level3; (2) maternal age of < 20 years and > 35 years4;
(3) heavy manual work2; (4) cervical incompetence, multiple
Conflicts of interest: The authors declare that they have no conflict of interest gestations, previous abortion, previous PTB2,4,5; and (5)
related to the subject matter or materials discussed in this article. genitourinary tract infections.1
* Corresponding author. Dr. Mona T. Kashef, Department of Microbiology
and Immunology, Faculty of Pharmacy, Cairo University, El Qasr El Einy
Genitourinary infections usually lead to PTB. Because many
Street, Cairo 11562, Egypt. of these infections are asymptomatic, underestimation of their
E-mail address: (M.T. Kashef). importance may have been occurred.1 Furthermore, few studies
1726-4901/Copyright © 2017, the Chinese Medical Association. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND
license (
Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),

2 A.E.-D.M.S. Hosny et al. / Journal of the Chinese Medical Association xx (2017) 1e7

focusing on these infections were conducted in Egypt, and they (gestational age, race, age, weight, height, nonprescription drug
investigated only one infection in relation to PTB, such as use, prenatal care received, working during pregnancy, smoking,
chlamydia,6 bacterial vaginosis, or urinary tract infection.7,8 coffee consumption, education, yoghurt consumption, contra-
In the present study, we screened women in term labor (TL) ception use, and use of vaginal douches during pregnancy),
and PTL for the presence of urinary and vaginal infections to pregnancy history (parity, gravidity, previous abortions, and
examine their association with PTL. In addition, the role of previous PTB), medical complications (anemia, vaginal bleeding
different demographic characteristics, patient history, and during pregnancy, urinary tract infection, and other infections),
clinical signs were also investigated. and results of speculum examination, done at the time of sam-
pling (presence of vaginal discharge, other signs of infection as
2. Methods erythema, bleeding, vaginal pH, and the whiff test result).

2.1. Study population 2.4. Sampling

This case-control study was performed involving Egyptian Vaginal swabs were obtained by the attending physician prior
women who were in labor and admitted to the Department of to vaginal examination, and urine samples were collected by the
Obstetrics and Gynaecology, Kasr Al Aini hospital, Cairo, participants themselves. All specimens were transported to the
Egypt. The women were enrolled during a 7-month period laboratory and processed within 18 hours of collection by a
from December 2009 to June 2010. For purposes of this study, microbiologist who was blinded to the group of participants.10
labor was defined as at least three uterine contractions in 10
minutes or 2e3-cm cervical dilatation.9 2.4.1. Vaginal swab samples
The study population (PTL group) consisted of 72 women A total of five swabs were obtained from each woman. The
admitted for PTL (< 37 gestational weeks) with or without first one was used for group B streptococci screening, as previ-
preterm premature rupture of membranes (PPROM). The control ously described.11 The other four swabs were used for screening
group (TL group) consisted of 45 control women admitted for other genital infections using an unmoistened sterile speculum.
TL without any complications of pregnancy such as PPROM, Two of these swabs were placed separately into Amies transport
preterm contractions, or vaginal bleeding ( 37 gestational medium (Oxoid Ltd., UK) for microbial biochemical identifica-
weeks). The gestational age was based on the last menstrual tion. The third swab was returned to its plastic tube containing no
period combined with ultrasonographic data, if present. transport medium for molecular identification tests. Finally, the
The study was approved by Kasr Al Aini hospital ethical final fourth swab was used for Gram stain scoring of vaginal
committee (992009). sample, pH determination, and whiff test.

2.2. Exclusion criteria Isolation and identification of vaginal microorganisms Group B streptococci. The swab was inoculated
Women with any of the following risk factors for PTL were into selective Todd Hewitt broth supplemented with 15 mg/mL
excluded from the study: age < 15 years, placenta previa and nalidixic acid, 10 mg/mL colistin, CNA supplement (as
abruptio placentae, multiple gestations, hydramnios, congenital directed), and 10 mg/mL yeast extract (LIM broth); all the
malformation, intrauterine fetal death, intrauterine growth media were purchased from Oxoid Ltd., UK. Next, the broth
retardation, medical complications requiring long-term or was subcultured to a 5% sheep blood agar (SBA) plate. All
intermittent medications as insulin-requiring diabetes mellitus, cultures were incubated following appropriate atmospheric
chronic hypertension and maternal cardiac disease, pre- conditions.11 Suspected colonies were tested using Strepto-
eclampsia, pregnancy-induced hypertension, erythroblastosis coccal grouping latex test, according to manufacturer's in-
fetalis, Rh isoimmunization, renal disease with a baseline structions (Oxoid Ltd., UK).
creatinine level of > 2.0, autoimmune disease requiring ste- T. vaginalis detection. One of the swabs, placed in
roids, cervical cerclage, uterine fibroids, abdominal trauma, Amies medium, was inoculated into a modified thioglycolate
in vitro fertilization, clinically evident herpes simplex virus, and medium (Oxoid Ltd., UK), incubated aerobically at 35 C for 7
treatment with antibiotics or vaginal douches/pessaries within 4 days, and examined daily for characteristic motility.12
weeks of enrollment. These factors were chosen based on their The other swab, placed in Amies medium, was used to
association with medically-induced preterm delivery. There- inoculate SBA, MacConkey agar, Sabouraud dextrose agar,
fore, this exclusionary regimen facilitates the ability to study the and Rogosa agar (All were from Oxoid Ltd., UK) to isolate
presence of infection as a sole risk factor in PTL. A gynecologist aerobic bacteria, facultative Gram-negative bacilli, yeast spe-
was responsible for the inclusion/exclusion of study participants cies, and lactobacilli, respectively. All inoculated media were
by applying the suitable medical examination or questionnaire. incubated according to the recommended conditions. Only
colonies growing in the third or fourth streak were identified as
2.3. Data collection they indicate a change of vaginal flora to heavy growth of this
A written questionnaire was filled out by a staff nurse prior to Molecular identification testing. Polymerase
sampling, focusing on patient demographic characteristics chain reaction (PCR) was used for the detection of Chlamydia

Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),

A.E.-D.M.S. Hosny et al. / Journal of the Chinese Medical Association xx (2017) 1e7 3

trachomatis, Neisseria gonorrhoeae, Mycoplasma hominis, M. isolates were identified. Cultures growing  3 isolates or any
genitalium, and Ureaplasma. The genomic DNA was extracted urethral nonuropathogen flora (diphtheroids, viridans strepto-
from the swab using the QIAamp DNA Mini kit (Qiagen, cocci, lactobacilli, coagulase-negative Staphylococci other
Germany), according to the manufacturer's instructions. than Staphylococcus saprophyticus, and Bacillus spp.) were
PCR was subsequently performed, as described pre- not identified.24
viously.14e19 Table 1 summarizes the sequence of primers Different biochemical tests were used for microorganism
used for detection of each microorganism and the expected identification including colony characteristics, Gram stain
amplicon size. The positive samples in genus-specific PCR for and wet mount, catalase test (3%), growth on Mannitol salt
mycoplasmas and ureaplasmas were further tested for M. agar (Oxoid, UK), Dry Spot Staphytect Plus kit (Oxoid, UK),
genitalium, M. hominis, and Ureaplasma. disk susceptibility with Polymyxin B (300 units, Oxoid, UK)
and Novobiocin (5 mg, Oxoid, UK), oxidase sticks (Oxoid, Gram stain scoring of vaginal sample, pH determi- UK), RapID STR system (Remel, USA), RapID CB plus
nation, and whiff test. The final fourth swab was rolled over a system (Remel, USA), RapID ONE system (Remel, USA),
glass slide which was Gram stained and scored as described and RapID SS/u system (Remel, USA). All tests were done
previously.20 Clue cells were also recorded. The same swab according to the manufacturer's instructions. Thereafter, the
was then touched to a pH paper (Whatman narrow range pH germ tube test was performed for identification of Candida
paper 4e6, with discrimination of 0.5, UK) for the determi- albicans.22
nation of the vaginal pH, and a drop of 10% potassium hy-
droxide was placed on it (whiff test). A fishy amine odor was 2.6. Statistical analysis
recorded as a positive whiff test.
Bacterial vaginosis (BV) was also diagnosed using the The tests of two or more proportions were done using
Amsel's composite criteria, which suggests that BV should be Fisher's exact test. The p values were from two sample-tailed
suspected if three or more of the following criteria were pre- tests ( Anal-
sent: homogeneous vaginal discharge, elevated vaginal ysis of variance test was used to compare continuous data,
pH > 4.5, production of amine odor in whiff test, and presence such as age and weight between different groups, by the Excel
of clue cells. Women with less than three of the criteria were program after checking data normality using a studentized
considered as normal.21 range test. Odds ratio (OR) and 95% confidence intervals (CI)
were also calculated for different bacterial forms that were
2.5. Urine analysis significantly associated with PTL ( A p
value < 0.05 was considered significant.
Each woman was instructed by a staff nurse regarding the
proper collection of the urine sample. The urine was cultured 3. Results
by the surface streak method on two SBA plates and two
MacConkey agar plates using 1-mL and 10-mL plastic cali- There was no significant difference in the pregnancy
brated loops (Pbi, Italy), incubated aerobically for 48 hours at outcome based on the most tested demographic characteristics
35e37 C.22 Only plates with  10 colonies were consid- (data not shown). In addition, there was no significant differ-
ered.23 The counts were calculated as colony forming units ence in the mean age, weight, or height between the PTL
(CFU)/mL and were placed in one of the following groups: no group and control (TL) group ( p > 0.05) as shown in Table 2;
growth to < 103,  103 to < 104,  104 to < 105, and  105. however, 28.36% of the total PTL women were
Only growth of one or two isolates of possible uropathogens at aged < 20 years compared with only 9.3% of the TL group
a concentration of  103 CFU/ml was considered, and these ( p < 0.05).

Table 1
Primer sequences used in PCR identification of pathogenic microorganisms and their corresponding amplicon size.
Name of microorganism Primer sequence Amplicon size (bp) Reference
Chlamydia trachomatis F: 50 -TCCGGAGCGAGTTACGAAGA-30
Neisseria gonorrhoeae F: 50 -GCTACGCATACCCGCGTTGC-30
Mycoplasmas and ureaplasmas (genus-specific) F: 50 -GTAATACATAGGTCGCAAGCGTTATC-30
Mycoplasma genitalium F: 50 -AGTTGATGAAACCTTAACCCCTTGG-30
281 17
Mycoplasma hominis F: 50 -CAATGGCTAATGCCGGATACGC-30
Ureaplasma F: 50 -GTATTTGCAATCTTTATATGTTTTCG-300 403, 404 (Ureaplasma parvum)
R: 50 - CAGCTGATGTAAGTGCAGCATTAAATT C-3 448 (Ureaplasma urealyticum)
bp ¼ base pair; F ¼ forward primer; R ¼ reverse primer.

Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),

4 A.E.-D.M.S. Hosny et al. / Journal of the Chinese Medical Association xx (2017) 1e7

Table 2 group ( p < 0.05; Table 2). A positive whiff test was signifi-
Comparison of demographic characteristics, pregnancy history, vaginal cantly more common in the PTL group than in the control
bleeding, and results of speculum examination among preterm and term labor
group ( p < 0.01; Table 2).
Characteristics Total PTL cases TL (control) p
3.4. Microbiology
Demographic characteristics
Age (y) 23.13 ± 5.29 (N ¼ 70) 24.45 ± 4.81 (N ¼ 44) 0.1822
Weight (kg) 68.31 ± 13.52 (N ¼ 61) 71.1 ± 11.92 (N ¼ 41) 0.2869
3.4.1. Identification of vaginal microorganisms
Height (cm) 159.69 ± 4.52 (N ¼ 61) 160.05 ± 4.84 (N ¼ 41) 0.7023 N. gonorrhoeae and M. genitalium were not detectable;
Gestational age at 32.38 ± 3.56 (N ¼ 52) 38.91 ± 1.08 (N ¼ 45) < 0.001 thus, they were not considered further. The isolation of group
labor (wk) B streptococci, C. trachomatis and Ureaplasma was compa-
Age < 20 y (No. 19/67 (28.36) 4/43 (9.30) 0.017 rable between the two groups ( p > 0.05; Table 3). Vaginal
infection with T. vaginalis was significantly more common in
total no.) (%) the PTL group than in the TL group (49.21% vs. 28.89,
Pregnancy history (No. with characteristic/total no.) (%) p < 0.05; Table 3). There was a two-fold increase in the risk of
1 parous or more 36/72 (50) 31/45 (68.89) PTL in the group infected with T. vaginalis compared with the
Nulliparous 7/72 (9.72) 0/45 (0) 0.032 noninfected group (OR ¼ 2.38, 95% CI ¼ 1.06e5.37,
Primigravida 29/72 (40.28) 14/45 (31.11)
History of PTB 7/43 (16.28) 0/31 (0) 0.037
p < 0.05; Table 4). M. hominis was significantly more detect-
excluding able in the PTL group ( p < 0.05; Table 3).
primigravida Except for coryneforms, the heavy vaginal growth of aer-
History of 16/43 (37.21) 2/31 (6.45) 0.002 obic microorganisms was similar between the two groups
abortion ( p > 0.05; Table 3). Only coryneform heavy vaginal growth
was significantly more common in the PTL group than in the
Menses vaginal 10/66 (15.15) 1/44 (2.27) 0.047 control group ( p < 0.01; Table 3). There was a 20-fold in-
bleeding crease in the risk of PTL in the group with heavy vaginal
Speculum examination (No. with characteristic/total no.) (%) growth of coryneform compared with the group that did not
Vaginal pH  4.5 27/63 (42.8) 24/43 (55.81) show such growth (OR ¼ 20.64, 95% CI ¼ 1.19e356.58,
Vaginal pH ¼ 5 23/63 (36.5) 18/43 (41.86) 0.017
Vaginal pH > 5 13/63 (20.6) 1/43 (2.33)
p < 0.05; Table 4).
Whiff test positive 37/66 (56.06) 11/44 (25) 0.002
PTB ¼ preterm birth; PTL ¼ preterm labor; TL ¼ term labor.
Data are presented as mean ± SD. Table 3
Comparison of microbiological characteristics among preterm and term labor
3.1. Pregnancy history Characteristic (No. with Total PTL cases TL (control) p
characteristic/total no.) (%)
The PTL group showed a significantly higher number of Vaginal infection
nulliparous women than the TL group (9.72% vs. 0%, GBS positive 12/71 (16.9) 4/45 (8.89) 0.277
p < 0.05). This group also recorded significantly higher fre- Chlamydia trachomatis positive 1/72 (1.39) 1/45 (2.22) > 0.99
Ureaplasma positive 26/72 (36.11) 22/45 (48.89) 0.183
quencies of both previous PTB and abortions, excluding pri- Trichomonas vaginalis positive 31/63 (49.21) 13/45 (28.89) 0.047
migravidas, than the TL group ( p > 0.05 vs. p < 0.01), as Mycoplasma hominis positive 11/72 (15.28) 1/45 (2.22) 0.028
indicated in Table 2. Gram-positive cocci a 10/72 (13.89) 7/45 (15.56) > 0.99
Coryneforms a,b 13/72 (18.06) 0/45 (0) 0.002
3.2. Medical complications Gram-negative bacilli a 4/72 (5.56) 1/45 (2.22) 0.648
Candida a 5/72(6.94) 5/45 (11.11) 0.5050
Lactobacilli a 35/72 (48.61) 19/39 (48.72) > 0.99
Only menses vaginal bleeding during pregnancy was Growth of Gram-negative bacilli 43/72 (59.72) 17/45 (37.78) 0.024
significantly more common in the PTL group than in the TL at any streak
group (15.15% vs. 2.27%, p < 0.05; Table 2). However, other Growth of lactobacilli at any 58/72 (80.56) 33/39 (84.62) 0.797
tested medical complications were comparable between the streak
Urine bacterial count
two groups (data not shown). > 103 CFU/mL 58/72 (80.56) 42/45 (93.34) 0.129
 103 to < 104 CFU/mL 3/72 (4.17) 2/45 (4.44)
3.3. Speculum examination  104 to < 105 CFU/mL 5/72 (6.94) 1/45 (2.22)
 105 CFU/mL 6/72 (8.33) 0/45 (0)
The presence of vaginal discharge or other signs of infec- CFU ¼ colony forming unit; GBS ¼ group B Streptococci; PTL ¼ preterm
tion, such as bleeding, erythema, eruptions, warts, ulcerations, labor; TL ¼ term labor.
Presence of characteristic growth at the third or fourth streak.
or inguinal adenopathy, on speculum examination did not b
Coryneforms: Brevibacterium spp. (group B) in nine cases, Arcanobacte-
differ significantly between the two groups ( p > 0.05) (data rium pyogenes, Acinetobacter calcoaceticus, and Corynebacterium minu-
not shown). A vaginal pH of > 5 (very abnormal pH) was tissimum each in one case and a mixed Brevibacterium spp. (group B) and
significantly more common in the PTL group than in the TL Corynebacterium minutissimum in one case.

Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),

A.E.-D.M.S. Hosny et al. / Journal of the Chinese Medical Association xx (2017) 1e7 5

Table 4
Calculated odds ratio for factors associated with preterm labor.
Infection N (total cases in TL and PTL groups) N with PTL Rate of PTL (%) OR (95% CI) p
Trichomonas vaginalis infection (þve) 44 31 70.45 2.38 (1.06e5.37)
Trichomonas vaginalis infection (ve) 64 32 50.00 Reference 0.0468
Gram negative bacilli infection (þve) 60 43 71.67 2.4 (1.1e5.2)
Gram negative bacilli infection (ve) 57 29 50.88 Reference 0.02
Coryneforms heavy vaginal growth (þve) 13 13 100 20.64 (1.19e356.58)
Coryneforms heavy vaginal growth (ve) 104 59 56.7 Reference 0.037

OR ¼ odds ratio; CI ¼ confidence interval; PTL ¼ preterm labor; TL ¼ term labor.

When considering any vaginal growth of Gram-negative may be due to the consequent thrombin production, which
bacilli and not only heavy growth, this type of growth was stimulates uterine contractions as well as proteolytic activity
significantly more common in the PTL group than in the TL that can lead to PPROM. In addition, this bleeding may be an
group ( p < 0.05; Table 3). There was a two-fold increase in indicator of infection or inflammation.27 As demonstrated in
the risk of PTL when taking into consideration any vaginal other studies, vaginal pH > 5 was associated with PTL. This
Gram-negative bacilli infection compared with the group may be due to the fact that elevated pH is a sign of inflammation
showing no such infection (OR ¼ 2.4, 95% CI ¼ 1.1e5.2, or infection of the endometrium or amniotic fluid with subse-
p < 0.05; Table 4). quent PTL.28
Some studies have suggested an association between BV
3.4.2. Gram staining of vaginal smears and PTL due to released proteolytic enzymes and elevated pH,
There was no significant difference in the vaginal grade, the which can increase the risk of BV by 10-fold.29 The detection
presence of clue cells, or BV diagnosed by the Amsel method of BV did not differ significantly between the two groups.
between the PTL and TL groups (data not shown). Similar results were reported by Discacciati et al30 and Ver-
straelen et al.31 However, several tested bacterial species were
3.5. Urinary tract infection found to be associated with PTL namely T. vaginalis, M.
hominis, and coryneform bacteria. Furthermore, similar results
The urine samples colony count did not differ significantly were previously reported.31e33 Group B streptococci were
between the two groups ( p ¼ 0.129; Table 3). All detected suspected of being associated with PTL; however, this was not
uropathogens were comparable among the two groups (data the case in our study as well as in some other studies.30e34 As
not shown). only one chlamydial infection was detected in the PTL and TL
groups, this study cannot be relied upon to help prove the
4. Discussion association between PTL and chlamydial infection.
This study results also do not support the claimed associ-
Several factors have been documented to be associated with ation between Ureaplasma urealyticum and PTL.35 On the
PTL. However, only a few of these factors were found to be contrary, this study recorded a higher prevalence of these
associated with PTL in this study, which further highlighted species in the TL than in the PTL group, which may be due to
their importance. These factors included women aged < 20 the fact that 60% of healthy women carry U. urealyticum in
years, as noted in other studies.4 A plausible biological their urogenital tract.36 All heavy growing aerobic bacteria,
explanation may be incomplete maternal physical growth and except coryneforms, were not associated with PTL. Gram-
relative malnutrition.25 On the contrary, women aged > 35 negative bacilli were associated with PTL when any growth
years, associated with PTL in the same study,4 were not (whether heavy or not) was used in statistical analysis. This
associated with PTL in this study. This may be due to the supports the claim that Gram-negative bacilli are important
small number of women in this age group who were included placental pathogens responsible for subclinical chorioamnio-
in the study (3 women in each of the TL and PTL group) or nitis and PTB.33
that the risk in this age group arose from the presence of Consistent with other studies, we found no significant as-
chronic diseases where women with chronic diseases were sociation between Candida infection and PTL. Also, lactoba-
excluded in our study.25 cilli, which are the predominating genus in vaginal microbiota
As recorded previously, there was an association between and play an important role in maintaining the natural healthy
nulliparous women as well as previous preterm delivery or balance of these organisms, were found to play a role in PTL
previous abortion and PTL.5,25 This may be due to the presence risk. Their absence was associated with PTL in some studies
of complications during childbirth in nulliparous women, such but not in this study and other studies.12,24 This may be
as obstructed labor, or due to an increased risk of uterine attributed to other factors that have not been addressed in this
infection during a prior abortion that can lead to PTB.25,26 study, such as the diversity of lactobacilli which affects
Furthermore, in accordance with the results of earlier studies, pregnancy duration.37
intense vaginal bleeding (not spotting) was highly associated Most studies recommend considering a urine culture with
with PTL. The association between vaginal bleeding and PTL more than 105 CFU/mL as indicative of the presence of

Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),

6 A.E.-D.M.S. Hosny et al. / Journal of the Chinese Medical Association xx (2017) 1e7

urinary tract infection in PTL women.38 In our study popu- 11. Schrag S, Gorwitz R, Fultz-Butts K, Schuchat A. Prevention of perinatal
lation, none of the tested urine culture count criteria was group B streptococcal disease. Revised guidelines from CDC. MMWR
Recomm Rep 2002;51:1e22.
significantly associated with PTL, although a urine culture 12. Poch F, Levin D, Levin S, Dan M. Modified thioglycolate medium: a
colony count of more than 105 CFU/mL was detectable only in simple and reliable means for detection of Trichomonas vaginalis. J Clin
the PTL group. This may be due to the fact that low-count Microbiol 1996;34:2630e1.
bacteriuria might be an early phase of urinary tract infection 13. Carey JC, Klebanoff MA. Is a change in the vaginal flora associated with
but not a true infection and due to the lower number detected an increased risk of preterm birth? Am J Obstet Gynecol 2005;192:
of  105 CFU/mL in PTL group (8.33%).39 14. Mahony JB, Luinstra KE, Sellors JW, Jang D, Chernesky MA. Confir-
There were several limitations to this study. First, it is a matory polymerase chain reaction testing for Chlamydia trachomatis in
one-hospital study and therefore, may not reflect the true first-void urine from asymptomatic and symptomatic men. J Clin
Egyptian population. Second, screening for other infections Microbiol 1992;30:2241e5.
associated with PTL, such as hepatitis B, HIV, and syphilis, 15. Ho BS, Feng WG, Wong BK, Egglestone SI. Polymerase chain reaction
for the detection of Neisseria gonorrhoeae in clinical samples. J Clin
were not conducted in our study.1 Pathol 1992;45:439e42.
In conclusion, although several factors were considered as 16. Yoshida T, Maeda S, Deguchi T, Miyazawa T, Ishiko H. Rapid detection
risk factors for PTL, only a few of them were significantly of Mycoplasma genitalium, Mycoplasma hominis, Ureaplasma parvum,
associated with PTL in this study. Most of these factors could and Ureaplasma urealyticum organisms in genitourinary samples by PCR-
be attributed to suspected infection. However, special attention microtiter plate hybridization assay. J Clin Microbiol 2003;41:1850e5.
17. Jensen JS, Uldum SA, Søndergård-Andersen J, Vuust J, Lind K. Poly-
should be given to nulliparous women, women showing signs merase chain reaction for detection of Mycoplasma genitalium in clinical
of vaginal infection such as bleeding, and high vaginal pH. samples. J Clin Microbiol 1991;29:46e50.
Infection control during abortion or labor is required to avoid 18. Blanchard A, Ya~nez A, Dybvig K, Watson HL, Griffiths G, Cassell GH.
its adverse effect on subsequent pregnancies. Therefore, it is Evaluation of intraspecies genetic variation within the 16S rRNA gene of
recommended to detect and treat infections with T. vaginalis, Mycoplasma hominis and detection by polymerase chain reaction. J Clin
Microbiol 1993;31:1358e61.
M. hominis, Gram-negative bacilli, and coryneform bacteria to 19. Fanrong K, James G, Zhenfang M, Gordon S, Wang B, Gilbert GL.
avoid their effect on pregnancy outcome in this population. Phylogenetic analysis of Ureaplasma urealyticumesupport for the estab-
Larger studies on Egyptian women are needed to support our lishment of a new species, Ureaplasma parvum. Int J Syst Bacteriol 1999;
findings. This study raises a question about other documented 49:1879e89.
risk factors which were proved not to be associated with PTL 20. Hay PE, Lamont RF, Taylor-Robinson D, Morgan DJ, Ison C, Pearson J.
Abnormal bacterial colonisation of the genital tract and subsequent pre-
in this study, such as infection with group B streptococci and term delivery and late miscarriage. BMJ 1994;308:295e8.
U. urealyticum. These factors need to be reevaluated to 21. Amsel R, Totten PA, Spiegel CA, Chen KC, Eschenbach D, Holmes KK.
determine if any other associated conditions and mechanisms Nonspecific vaginitis. Diagnostic criteria and microbial and epidemiologic
are responsible for their previously recorded association with associations. Am J Med 1983;74:14e22.
PTL rather than the infection itself. 22. Winn Jr WC, Allen SD, Janda WM, Koneman EW, Schreckenberger PC,
Procop GW, et al. Konemann's Color Atlas and textbook of diagnostic
microbiology. 6th ed. Philadelphia: Lippincott Williams & Wilkins;
References 2006.
23. Frimodt-Møller N, Espersen F. Evaluation of calibrated 1 and 10 microl
1. Cram LF, Zapata MI, Toy EC, Baker 3rd B. Genitourinary infections and loops and dipslide as compared to pipettes for detection of low count
their association with preterm labor. Am Fam Physician 2002;65:241e8. bacteriuria in vitro. APMIS 2000;108:525e30.
2. Al-Dabbagh SA, Al-Taee WY. Risk factors for pre-term birth in Iraq: a 24. Semeniuk H, Church D. Evaluation of the leukocyte esterase and nitrite
case-control study. BMC Pregnancy Childbirth 2006;6:13. urine dipstick screening tests for detection of bacteriuria in women with
3. Savitz DA, Kaufman JS, Dole N, Siega-Riz AM, Thorp Jr JM, Kaczor DT. suspected uncomplicated urinary tract infections. J Clin Microbiol 1999;
Poverty, education, race and pregnancy outcome. Ethn Dis 2004;14:322e9. 37:3051e2.
4. Ko YL, Wu YC, Chang PC. Physical and social predictors for pre-term 25. Kozuki N, Lee AC, Silveira MF, Sania A, Vogel JP, Adair L, et al. Child
births and low birth weight infants in Taiwan. J Nurs Res 2002;10:83e9. Health Epidemiology Reference Group (CHERG) Small-for-Gestational-
5. Ezechi OC, Makinde ON, Kalu BE, Nnatu SN. Risk factors for preterm Age-Preterm Birth Working Group. The associations of parity and
delivery in south western Nigeria. J Obstet Gynaecol 2003;23:387e91. maternal age with small-for-gestational-age, preterm, and neonatal and
6. El-Shourbagy M, Abd-el-Maeboud K, Diab KM, el-Ghannam A, infant mortality: a meta-analysis. BMC Public Health 2013;13:S2.
Nabegh L, Ammar S. Genital Chlamydia trachomatis infection in Egyp- 26. Dragoman M, Sheldon WR, Qureshi Z, Blum J, Winikoff B, Ganatra B,
tian women: incidence among different clinical risk groups. J Obstet WHO Multicountry Survey on Maternal Newborn Health Research
Gynaecol Res 1996;22:467e72. Network. Overview of abortion cases with severe maternal outcomes in
7. Darwish A, Elnshar EM, Hamadeh SM, Makarem MH. Treatment options the WHO Multicountry Survey on Maternal and Newborn Health: a
for bacterial vaginosis in patients at high risk of preterm labor and pre- descriptive analysis. BJOG 2014;121:25e31.
mature rupture of membranes. J Obstet Gynaecol Res 2007;33:781e7. 27. Yang J, Hartmann KE, Savitz DA, Herring AH, Dole N, Olshan AF, et al.
8. Dimetry SR, El-Tokhy HM, Abdo NM, Ebrahim MA, Eissa M. Urinary Vaginal bleeding during pregnancy and preterm birth. Am J Epidemiol
tract infection and adverse outcome of pregnancy. J Egypt Public Health 2004;160:118e25.
Assoc 2007;82:203e18. 28. Simhan HN, Caritis SN, Krohn MA, Hillier SL. Elevated vaginal pH and
9. Buhimschi IA, Christner R, Buhimschi CS. Proteomic biomarker analysis neutrophils are associated strongly with early spontaneous preterm birth.
of amniotic fluid for identification of intra-amniotic inflammation. BJOG Am J Obstet Gynecol 2003;189:1150e4.
2005;112:173e81. 29. Jakovljevic A, Bogavac M, Nikolic A, Tosic MM, Novakovic Z, Stajic Z.
10. Holst E, Goffeng AR, Andersch B. Bacterial vaginosis and vaginal mi- The influence of bacterial vaginosis on gestational week of the completion
croorganisms in idiopathic premature labor and association with preg- of delivery and biochemical markers of inflammation in the serum. Voj-
nancy outcome. J Clin Microbiol 1994;32:176e86. nosanit Pregl 2014;71:931e5.

Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),

A.E.-D.M.S. Hosny et al. / Journal of the Chinese Medical Association xx (2017) 1e7 7

30. Discacciati MG, Simoes JA, Silva MG, Marconi C, Brolazo E, Costa ML, 35. Vouga M, Greub G, Prod'hom G, Durussel C, Roth-Kleiner M,
et al. Microbiological characteristics and inflammatory cytokines associ- Vasilevsky S, et al. Treatment of genital mycoplasma in colonized preg-
ated with preterm labor. Arch Gynecol Obstet 2011;283:501e8. nant women in late pregnancy is associated with a lower rate of premature
31. Verstraelen H, Verhelst R, Roelens K, Claeys G, Weyers S, De Backer E, labor and neonatal complications. Clin Microbiol Infect 2014;20:1074e9.
et al. Modified classification of Gram-stained vaginal smears to predict 36. Nassar FA, Abu-Elamreen FH, Shubair ME, Sharif FA. Detection of
spontaneous preterm birth: a prospective cohort study. Am J Obstet Chlamydia trachomatis and Mycoplasma hominis, genitalium and Ure-
Gynecol 2007;196:528.e1e6. aplasma urealyticum by polymerase chain reaction in patients with sterile
32. Mann JR, McDermott S, Gill T. Sexually transmitted infection is pyuria. Adv Med Sci 2008;53:80e6.
associated with increased risk of preterm birth in South Carolina 37. Petricevic L, Domig KJ, Nierscher FJ, Sandhofer MJ, Fidesser M,
women insured by Medicaid. J Matern Fetal Neonatal Med 2010;23: Krondorfer I, et al. Characterisation of the vaginal Lactobacillus micro-
563e8. biota associated with preterm delivery. Sci Rep 2014;4:5136.
33. Donders GG, Van Calsteren K, Bellen G, Reybrouck R, Van den Bosch T, 38. Jain V, Das V, Agarwal A, Pandey A. Asymptomatic bacteriuria & ob-
Riphagen I, et al. Predictive value for preterm birth of abnormal vaginal stetric outcome following treatment in early versus late pregnancy in north
flora, bacterial vaginosis and aerobic vaginitis during the first trimester of Indian women. Indian J Med Res 2013;137:753e8.
pregnancy. BJOG 2009;116:1315e24. 39. Kunin CM, White LV, Hua TH. A reassessment of the importance of “low-
34. Seyyed EZ, Toossi E, Jalalvand A, Sajadi M. Group B streptococci count” bacteriuria in young women with acute urinary symptoms. Ann
investigation in pre-term labors. Med Arh 2013;67:124e5. Intern Med 1993;119:454e60.

Please cite this article in press as: Hosny AE-DMS, et al., Association between preterm labor and genitourinary tract infections caused by Trichomonas
vaginalis, Mycoplasma hominis, Gram-negative bacilli, and coryneforms, Journal of the Chinese Medical Association (2017),