Evaluation of vaginal fluid β-human chorionic gonadotrophin for the diagnosis of preterm premature rupture of membranes

Authors


Reprint request to: Dr Maryam Kashanian, Department of Obstetrics and Gynecology, Akbarabadi Teaching Hospital, Tehran University of Medical Sciences, no. 9, Mostaghimi Alley, Khajeh Nasir Toosi Avenue, Tehran 16117, Iran. Email: maryamkashanian@yahoo.com

Abstract

Aim

The purpose of the present study was to evaluate vaginal fluid β-human chorionic gonadotrophin (β-hCG) for the diagnosis of preterm premature rupture of membranes (PPROM).

Material and Methods

An observational cross-sectional study was performed on 123 pregnant women who were in the third trimester of their gestation (28–37 weeks). The patients were divided into three groups: (i) PPROM group (41 cases); (ii) suspected PPROM group (42 cases); and (iii) intact membranes control group (40 cases). Five milliliters of sterile normal saline was inserted into the posterior fornix of the vagina and then vaginal fluid was aspirated and sent to the laboratory for β-hCG measurement. Finally, the amount of β-hCG was compared in the three groups.

Results

β-hCG concentration was 7.71 ± 15.7 mIU/mL in the intact membrane control group, 468.06 ± 366.34 mIU/mL in the PPROM group and 176.43 ± 316.37 mIU/mL in the suspected PPROM group, which showed a significant difference between the three groups (P < 0.001). In order to find an optimal cut-off value for β-hCG, the receiver–operator curve was used and a cut-off value of 79.5 mIU/mL with a sensitivity of 95% and specificity of 84% was determined to be optimal.

Conclusion

β-hCG was higher in the cases of PPROM and patients who were suspected to have PPROM, and may be used as a suitable, fast and reliable test for detecting rupture of membranes.

Introduction

Premature rupture of the fetal membranes (PROM) is defined as rupture of amniotic membranes and amniotic fluid leakage, 1 h or more prior to the onset of labor.[1] PROM is divided into two categories including term PROM (PROM after 37 weeks of gestation) and preterm PROM (PPROM) (before 37 weeks of gestation).[1]

PPROM occurs in 3% of pregnancies and is responsible for one-third of all preterm births;[2] because of this and also because of the higher risk of morbidity and even mortality for the mother, fetus and neonate, it is a focus of special consideration.

PROM is associated with important maternal, fetal and neonatal mortality and morbidity due to infection, cord compression, placental abruption, and fetal prematurity.[3] Its causes are multifactorial, but the most important risk factors are previous preterm labor and previous PPROM.[3]

The diagnosis of rupture of the membranes (ROM) may be made by clinical history and some simple diagnostic tests. Patient history is accurate for about 90% for the diagnosis and should not be ignored.[2, 4]

Regarding the necessity for accurate diagnosis of PPROM, several diagnostic tests have been suggested. Nitrazine test,[4, 5] fern test,[4, 5] and ultrasound[3] have been considered as very useful tests.[3] Biochemical tests have also been suggested, but should only be performed in cases of very suspicious ROM, which other simple tests may not diagnose accurately.

Diagnosis of ROM is simple in the cases of obvious rupture, but when the rupture is scant, it is difficult or even impossible.

Failure of diagnosis may cause a risky delay in performing necessary tasks, such as performing a vaginal examination;[6] also, false diagnosis may cause unnecessary interventions like hospitalization, or labor induction, therefore the diagnosis of suspicious cases is obligatory and vital. Every diagnostic test for this purpose should be simple, reliable and fast.[7]

The diagnosis of ROM is based on the patient's report, and observation of amniotic fluid leakage from the cervix and by pooling of amniotic fluid in the vagina, which is confirmed when a speculum is inserted.

However, there are some cases with a history of fluid leakage, but there is no fluid pooling in the speculum (in the cases with prolonged rupture). Nitrazine test may cause false negative results in 9.4% of cases after 48 h of rupture;[4] also, vaginitis, cervicitis, contamination with urine, semen and blood, and antiseptic use may give false negative results,[4, 7] therefore other methods of diagnosis have been investigated.

Biochemical materials (substances) which have high concentrations in amniotic fluid, including interlukin-6,[8] alpha fetoproteins,[8-10] diamin oxidase,[11, 12] prolactin,[13, 14] urea and creatinine,[15] fetal fibronectin[16] and insulin-like growth factor binding protein I,[17-19] have all been evaluated.

Another substance is β-human chorionic gonadotrophin (β-hCG), which is secreted solely by syncytiotrophoblasts and can be found in amniotic fluid in addition to mother's blood or urine[13, 20] and has been studied for the evaluation of ROM.[6, 20]

The difference between the above-mentioned materials and β-hCG is the latter has simplicity and ease of use as well as being cheaper.

The purpose of the present study is to evaluate β-hCG in vaginal washing fluid for the diagnosis of PPROM.

Method

An observational cross-sectional study was performed in Akbarabadi Teaching Hospital, Tehran, Iran, on pregnant women who were referred to the hospital. A sample size of 120 patients (40 in each group) was sufficient to obtain a power of 90% (α = 0.05, 1-β = 0.085) with a significance level of 5%. These women were considered in three groups: (i) 41 cases were pregnant women who had definite leakage and pooling was positive (PPROM group); (ii) 42 cases were pregnant women who were suspected of having PPROM, with a history of probable fluid leakage and negative pooling (suspected PPROM group); and (iii) 40 cases were pregnant women with intact membranes who had been referred to the prenatal clinic for prenatal care with no positive history and vaginal pooling (control group).

All three groups had a gestational age of 28–37 weeks (according to reliable last menstrual period and ultrasound confirmation of first trimester). Exclusion criteria included: any vaginal bleeding, history of infection or vaginal washing with antiseptics, and occurrence of vaginal intercourse during previous 48 h.

First, a sterile speculum was inserted for all three groups, and the women were evaluated for fluid leakage and vaginal pooling. Then, a fern test and a nitrazine paper test were performed on the posterior fornix discharge for all women. Also, for all women, ultrasound was performed in order to determine the amniotic fluid index (AFI) and reported as normal or decreased AFI.

Positive vaginal pooling was considered as definitive ROM (41 cases). Women who were pooling negative, but fern test was (±) and/or nitrazine paper test was (±) and/or low AFI were considered as suspected ROM.

Finally, 5 cc (mL) sterile normal saline was injected into the vagina with a syringe and after 3–4 min, the fluid of vaginal washing was removed from posterior fornix with the same syringe. The fluid was poured into a tube and a β-hCG test was performed using a commercial kit (baby check with a threshold of 35 mIU/mL for positive results), the results were noted in the form of positive and negative.

The collected fluid was transferred to the laboratory for measurement of β-hCG and was preserved in 0–4°C temperature and then β-hCG titer was measured on all samples. The samples were centrifuged for 3 min with 1500 r.p.m., and β-hCG titer was determined by electrochemiluminescence immunoassay. Total duration of the test was about 18–20 min.

A written informed consent was obtained from all participants and they were fully informed about the study, and institutional review board approval and also institutional ethics committee approval was given to the study.

Statistical analysis was performed using spss 11.5. T-student test, one-way anova and Kruskal–Wallis tests were used for analysis.

Results

In total, 123 pregnant women were analyzed: 41 cases in the PPROM group and 42 cases in the suspected PPROM group together formed the case group of 83 women; and there were 40 women in the intact membrane control group.

Mean age of the women was 25.79 ± 5.59 years, 56 (49.1%) women were primigravida, 35 cases (30.7%) were gravida 2, 18 cases (15.8%) were gravida 3, four cases (3.5%) were gravida 4, and one case (0.9%) was gravida 5.

Mean parity of women was 1.76 ± 0.905. Mean β-hCG of vaginal washing fluid was 199.09 ± 32.689 mIU/mL. Fluid pooling was positive in 41 (33.3%) cases. Nitrazine tests, fern test, and baby check were positive in 39 (31.7%) cases, 48 cases (39%) and 79 (64.2%) cases, respectively. AFI was reported as ‘decreased’ in 44 cases (35.8%).

There was no statistically significant difference between the three groups according to age, parity and gestational age at the time of sampling and history of abortion (Table 1).

Table 1. Characteristics of the patients of three groups
CharacteristicsPPROM group, n = 41Suspected PPROM group, n = 42Control group, n = 40P-value
  1. NS, not significant; PPROM, preterm premature rupture of membranes; SD, standard deviation.
Age (years), mean ± SD24.78 ± 5.2225.64 ± 4.4227 ± 6.23NS
Gestational age, mean ± SD31.41 ± 2.0931.85 ± 2.0432 ± 2.15NS
Parity, mean ± SD1.79 ± 0.871.73 ± 0.821.87 ± 1.04NS
Previous abortion, n (%) n = 195(12.79%)6(14.28%)8(20%)NS

Positive fern test showed a significant difference between the three groups (P < 0.001) and had a sensitivity of 51.8% and specificity of 87.5% for detecting ROM (five women in the control group, 25 women in PPROM group and 18 women in the suspected PPROM group were fern positive) (Table 2).

Table 2. Fern test, nitrazine test, and β-hCG of the patients of three groups
CharacteristicsPPROM group, n = 41Suspected PPROM group, n = 42Control group, n = 40P-value
  1. β-hCG, β-human chorionic gonadotrophin; AFI, amniotic fluid index; PPROM, preterm premature rupture of membranes; SD, standard deviation.
Positive fern test, n (%) n = 4825 (60.97%)18 (42.85%)5 (12.5%)P < 0.001
Positive nitrazine test, n (%) n = 3924 (58.53%)15 (35. 71%)7 (17.5%)P < 0.001
Low AFI, n (%) n = 4429 (70.73%)15 (35. 71%)0 (0%)P < 0.001
β-hCG mIU/mL, mean ± SD391.75 ± 35.94176.43 ± 31.637.71 ± 1.5P < 0.001
Positive baby check, n (%) n = 7939 (95.1%)32 (76.2%)8 (20%)P < 0.001

Positive nitrazine test showed a significant difference between the three groups (P < 0.001) and had a sensitivity of 47% and specificity of 82.5% (seven women in the control group, 24 women in PPROM group and 15 women in the suspected PPROM group were positive for nitrazine test) (Table 2).

The mean amount of β-hCG in the positive baby check group was 298.82 ± 35.49 mIU/mL and in the negative baby check group it was 3.95 ± 1.06 mIU/mL. The amount of β-hCG in the amniotic fluid was significantly higher in the group with positive baby test (P < 0.001) than in the group with negative baby test.

Receiver–operator curve analysis was used to establish the optimal cut-off point for vaginal washing fluid β-hCG and is shown in Figure 1. Optimal cut-off point of β-hCG was calculated as 79.5 mIU/mL with a sensitivity of 93% and specificity of 84%.

Figure 1.

Receiver–operator curve (ROC) analysis for the optimal cut-off point for vaginal washing fluid β-human chorionic gonadotrophin.

Discussion

In the present study, β-hCG of vaginal washing fluid had an optimal cut-off point of 79.5 mIU/mL with a sensitivity of 93% and specificity of 84%. Regarding the importance of the accurate diagnosis of PPROM,[2, 3, 21, 22] it seems reasonable to study further to find the most reliable, fastest and cheapest method in unusual cases of PPROM.

For this reason, different substances have been evaluated, such as interlukin,[6, 8] alpha fetoprotein (AFP),[8-10] diamin oxidas,[11, 12] prolactin,[13, 14] urea and creatinine,[15] fetal fibronectin,[16] and insulin-like growth factor binding protein I.[17]

Also, some significant researches have been performed that may even predict PROM occurrence by evaluating some materials earlier in pregnancy and might be able to consider some therapeutic interventions to prevent it, such as fetal hemoglobin measurement in amniotic fluid during the second trimester,[23] and endothelin 1 in the second trimester,[24] both of which are representative of a great effort to make the diagnosis more accurate.

Some authors are investigating the fact that some of these agents may show false positive results because of contamination with blood and measurement of some materials that are present exclusively in amniotic fluid and are not present in blood, such as proteomics.[25, 26]

Also, vaginal β-hCG has been investigated for the diagnosis of PPROM. In a study that measured the β-hCG level of vaginal fluid,[20] the researchers reported it as a useful marker for ROM in the second and third trimester.

In another study that was performed on 52 women[27] (20 women with intact membranes, 21 women with definitive ROM and 11 women with suspicious ROM), the researchers concluded that vaginal washing fluid hCG is a suitable, cheap and non-invasive method for the diagnosis of PPROM. In this study, an optimal cut-off value of 100 mIU/mL was obtained.

The results of the study correspond with the present study except that the cut-off value of β-hCG was 79.5 mIU/mL and the number of women in the present study was higher. Kim et al.[28] measured the β-hCG of vaginal fluid for the diagnosis of ROM and found it as an appropriate method with a cut-off value of 39.8 mIU/mL. The cut-off value of 65 mIU/mL was obtained for vaginal washing fluid for β-hCG in another study,[6] which was performed on 73 pregnant women who were complaining of fluid leakage. They described this method as a useful and reliable method for the diagnosis of ROM, which is compatible with the present study.

Another study[8] compared hCG, AFP and interlukin-6 in both groups with definitive ROM and intact membranes. In this study, all three agents were higher in the PPROM group. The diagnostic value of AFP was more than hCG, the value of which was more than interleukin-6. However, although the diagnostic value of AFP was more than hCG in this study, it should be considered that the measurement of AFP is more expensive and probably is not applicable in most of the emergency settings.

Prolactin, AFP and β-hCG of vaginal washing fluid were compared in a study[13] for the diagnosis of ROM. This study was performed on 100 women with a gestational age of 28–37 weeks. All three agents were higher in the PPROM group and again, in this study the diagnostic value of AFP was greater. In the present study, this comparison has not yet been performed and it seems more studies with higher sample size are necessary to reach conclusive results. It should also be considered that β-hCG measurement is simpler and can be performed in emergency settings in most centers, whilst measurement of other materials like AFP or prolactin are more expensive, and not always practical to perform in all centers at all times.

In conclusion, β-hCG is higher in the case of ROM and patients who are suspected of having ROM, and may be used as a suitable, fast and reliable test for detecting ROM.

Because of the feasibility and simplicity of β-hCG measurement in most laboratories and hospitals, vaginal fluid β-hCG might be an alternative for diagnosing ROM.

With regard to the low number of studies, it seems that more studies should be performed over a higher number of women to evaluate the diagnostic value of β-hCG and to find a definitive optimal cut-off point for β-hCG in different gestational ages.

Disclosure

None declared.

Ancillary