Association between clinical and ultrasound diagnoses of aetiologies of vaginal bleeding in the first trimester

Nkengfua Samuel; Yauba Saidu; Ngalame Abigail; Kwasseu Gaetan; Kemah Ben-Lawrence; Halle-Ekane Gregory Edie

doi:10.4103/jihs.jihs_7_22

ORIGINAL ARTICLE

Year : 2022 | Volume : 10 | Issue : 2 | Page : 71-75

Association between clinical and ultrasound diagnoses of aetiologies of vaginal bleeding in the first trimester

Nkengfua Samuel¹, Yauba Saidu², Ngalame Abigail¹, Kwasseu Gaetan¹, Kemah Ben-Lawrence³, Halle-Ekane Gregory Edie⁴
¹ Department of Public Health and Hygiene, Faculty of Health Sciences, University of Buea, Buea, Cameroon
² Clinton Health Access Initiative, Yaounde; Institute for Global Health, Siena University, Siena, Italy
³ Department of Obstetrics and Gynaecology, University Hospitals of North Midlands, United Kingdom; Health Education and Research Organizations (HERO), Douala
⁴ Department of Medicine, Faculty of Health Sciences, University of Buea, Buea; Department of Obstetrics and Gynaecology, Douala General Hospital, Douala, Cameroon

Date of Submission	07-Sep-2022
Date of Decision	10-Dec-2022
Date of Acceptance	13-Dec-2022
Date of Web Publication	16-May-2023

Correspondence Address:
Dr. Nkengfua Samuel
Department of Public Health and Hygiene, Faculty of Health Sciences, University of Buea, Buea
Cameroon

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jihs.jihs_7_22

Abstract

Introduction: The prevalence of first-trimester vaginal bleeding (FTVB) ranges worldwide from 4% to 24%. In resource-limited settings, there is a heavy reliance on clinical presentation to establish the cause of bleeding due to the limited availability of ultrasounds (US) imaging modalities. Objectives: The objective is to determine the association between clinical and US diagnoses in FTVB at the Buea and Limbe regional hospitals (BRH and LRH). Subjects and Methods: We recruited 144 participants, with <14 weeks of gestational age at the BRH and LRH. Data on sociodemographic, clinical, and US diagnoses were obtained from patients. Statistical significance was considered at P < 0.05. Results: From 144 participants enrolled, 55 (38.2%) were from BRH and 89 (61.8%) were from LRH. The main causes of FTVB were threatened abortions (42.36%), incomplete abortion (27.8%), complete abortion (3.5%), missed abortion (1.4%), ectopic pregnancy (22.91%), molar pregnancy (0.91%), and pregnancy with fibroids (1.4%). Cohen's kappa analysis revealed a good degree of agreement between clinical and US diagnoses for elective consultations (κ = 0.634, P < 0.001) and an average degree of agreement for those consulting as referrals (κ = 0.520, P < 0.001). Our study revealed a significant statistical association between clinical and US diagnoses for the following pathologies: threatened abortion (P < 0.001), incomplete abortion (P = 0.004), complete abortion (P < 0.019), and ectopic pregnancy (P < 0.001). Conclusion: This study reveals an overall average consistency between clinical and US diagnoses, especially for common etiologies of bleeding in early pregnancy. Hence, the need for US to confirm clinical diagnoses.

Keywords: Association, clinical and ultrasound diagnoses, first-trimester bleeding

How to cite this article:
Samuel N, Saidu Y, Abigail N, Gaetan K, Ben-Lawrence K, Edie HEG. Association between clinical and ultrasound diagnoses of aetiologies of vaginal bleeding in the first trimester. J Integr Health Sci 2022;10:71-5

How to cite this URL:
Samuel N, Saidu Y, Abigail N, Gaetan K, Ben-Lawrence K, Edie HEG. Association between clinical and ultrasound diagnoses of aetiologies of vaginal bleeding in the first trimester. J Integr Health Sci [serial online] 2022 [cited 2023 Jun 2];10:71-5. Available from: https://www.jihs.in/text.asp?2022/10/2/71/377154

Introduction

Bleeding in early pregnancy is often an obstetric emergency more common in the first trimester.^[1] Obstetric emergencies including miscarriages and ectopic pregnancies require further imaging to accurately diagnose and treat them.^[2] Globally, the prevalence of first-trimester vaginal bleeding (FTVB) ranges from 7% to 24%, and in sub-Saharan Africa from 3% to 17%.^[2] It is associated with spotting in 25% of all pregnancies and about 50% of all those who bleed may lose the pregnancy.^[3] The etiological diagnoses of FTVB are made where ultrasonographic modalities are readily available.^[4] However, in low-income countries, these imaging modalities are not made readily available hence there is heavy reliance on clinical assessment in making diagnoses of FTVB.^[5] The implications of missed diagnoses are a threat to the survival of the mother and pregnancy.^[6] Therefore, there is an urgent need of increasing the availability of ultrasonographic imaging modalities. In sub-Saharan Africa, the concordance varies from 25% to 84% but very little was done as far as associating the accuracy of those diagnoses made clinically with those obtained from the ultrasound (US) in Cameroon.^{[7],[8],[9],[10],[11]} Hence, the purpose of this study was to identify the prevalent causes of FTVB and determine the degree of agreement coupled with the association between clinical and ultrasonographic diagnoses of FTVB in our study setting.

Subjects and Methods

Study design

A hospital-based cross-sectional study was conducted involving 144 participants using the Lorentz formula who gave written informed consent.

Study setting and period

The participants were enrolled at the radiologic units, obstetric and gynecologic units of the Buea and Limbe Regional hospitals (BRH and LRH) from February 5, 2020, to April 30, 2020.

Study population and sampling method

The study population comprised all pregnant women who came for consultation and were 14 completed weeks or less of gestational age at the BRH and LRH. The sampling method was consecutive and nonprobabilistic.

Inclusion criteria

This study included all pregnant women that were bleeding and had completed at most 14 weeks of gestational age, who came for consultation and gave written consent.

Noninclusion criteria

All participants with nonobstetrical cause of vaginal bleeding that had completed at most 14 weeks of gestational age and those who did not give informed consent were not included.

Ethical consideration

Ethical approval was obtained from the Institutional Review Board of the Faculty of Health Sciences, University of Buea (2020/1051-01/UB/SG/IRB/FHS).

Data collection method

Structured questionnaires were administered to collect participants' sociodemographic data. The physical examination and clinical diagnoses were obtained from participants' medical booklets. The US diagnoses were gotten from the radiologic result reported by the sonographer.

Statistical analysis

Data were analyzed using Statistical Package for Social Sciences (SPSS) version 25.0 (SPSS, IBM, Chicago, Illinois, U.S.A). Numerical variables were summarized using mean and standard deviation and categorical variables using frequencies and percentages. The degree of agreement between clinical and US diagnoses of the participant was scaled following Cohen's Kappa analysis; 0.0–0.2 was considered poor, 0.2–0.4 was considered fair, 0.4–0.6 was considered average, 0.4–0.8 was considered a good, and 0.8–1.0 was considered excellent. The univariate logistic regression technique was used to determine any association between clinical and US diagnoses in FTVB.

Statistical significance was considered at P < 0.05.

Results

One hundred and forty-four participants aged between 15- and 40-year-old with a mean of 28 (±5 years) were enrolled in this study. The modal age group was 25–29-year-old representing 35.4% [Table 1] and the mean gestational age was 8 weeks [Table 2]. Majority of the participants were students 74 (51.4%) and for the marital status, a greater proportion of the participants were single 121 (84.0%). Furthermore, most of the participants 80 (55.56%) had had more than one pregnancy and 78 (54.17%) participants of the total sample size were nulliparous. Spontaneous bleeding was reported in 131 (90.9%) participants while in 13 (9.0%), bleeding was caused by uterotonics [Table 2]. The clinical diagnoses included threatened abortions which represented 61 (42.4%), incomplete abortion 40 (27.8%), complete abortion 5 (3.5%), missed abortion 2 (1.4%), ectopic pregnancy 33 (22.9%), molar pregnancy 1 (0.91%), and pregnancy with fibroids 2 (1.4%) [Table 3]. The US diagnoses included threatened abortions which represented 51 (35.4%), incomplete abortion 33 (13.9%), complete abortion 7 (4.9%), missed abortion 4 (2.8%), ectopic pregnancy 40 (27.8%), molar pregnancy 2 (1.4%), pregnancy with fibroids 3 (2.8%), blighted ovum 2 (1.4), and heterotopic pregnancy 1 (0.69) [Table 4]. The highest percentage concordance of 81.5% was identified with the obstetrician–gynecologist and the least of 28.6% was identified with the nurses through the general practitioners 70.3% and midwives 42.9% [Table 5]. Cohen's kappa analysis revealed a good degree of agreement between clinical and US diagnoses for participants who came for elective consultations (κ = 0.634, P < 0.001) and an average degree of agreement for those who came in as referrals (κ = 0.520, P < 0.001) [Table 6]. Our study revealed a statistically significant association between clinical and US diagnoses for the following pathologies: Threatened abortion (P < 0.001), incomplete abortion (P = 0.004), complete abortion (P < 0.019), and ectopic pregnancy (P < 0.001) [Table 7].

Table 1: Demographics of participants (n=144)

Click here to view

Table 2: Obstetrics characteristics of participants (n=144)

Click here to view

Table 3: Stratification of clinical diagnosis based on the type of health-care provider (n=144)

Click here to view

Table 4: Ultrasound diagnosis made by type of health-care provider (n=144)

Click here to view

Table 5: The concordance between clinical and ultrasound diagnosis (n=144)

Click here to view

Table 6: Degree of agreement between clinical and ultrasound diagnosis (n=144)

Click here to view

Table 7: Association between clinical and ultrasound diagnosis (n=144)

Click here to view

Discussion

From our study, we clinically identified seven pregnancy-related etiologies of bleeding in early pregnancy including threatened abortion, incomplete abortion, complete abortion, missed abortion, ectopic gestation, pregnancy with fibroids and molar pregnancy. This is similar to a study carried out by Vidya et al. in India.^[12] This similar trend can be explained by the similarity in the prevalence of FTVB in India and Cameroon. However, Neossi Guena et al. failed to identify uterine fibroids due to their large low-risk population of multiparous women who traditionally are of lower risk for fibroids.^[11] The failure to identify molar pregnancies by Neossi et al. can be explained by the small sample size of their study as well as the low prevalence of the disease.^[11]

However, the findings in our study are different from Tiparse et al. who made nine clinical diagnoses of FTVB including ovarian cyst and blighted ovum.^[4] The failure from our study to identify ovarian cyst and blighted ovum can be explained by the disparity in the level of diagnosticians in our setting, which varied from the nurses to obstetric/gynecologist through midwives and general practitioners compared to their setting, which only involved the gynecologist. The common etiologies of bleeding in early pregnancy from our study include sexual intercourse, physical effort, trauma from road traffic crash, and uterotonic use. Neossi Guena et al. identified similar circumstances of bleeding in early pregnancy.^[11] More so, from our study, the most common diagnosis of FTVB made was threatened abortion and represented 42.4%. It was spontaneous and the nature of bleeding varied from light to moderate in most participants. Singh found out that 48% of their diagnoses were threatened abortion.^[13] Vidya et al. found out that 40% of their diagnoses were threatened abortion.^[12] Khatod et al. found out that 47.6% of their diagnoses were threatened abortion.^[14] In these previous studies, bleeding per vagina was spontaneous more than 50% of the time and the nature of bleeding varied from light to moderate in most participants in their respective study setting. This is explained by the fact that in the 1^st week of pregnancy, spotting is the most common symptom of bleeding, caused by implantation of the conceptus into the endometrium.^[15] Moreover, this bleeding generally is diagnosed as threatened abortion, to rule out any external factor, which could have been at the origin of this bleeding. Moreover, we found out that 4 in 7 diagnoses of FTVB made clinically were done accurately. This implies that 43% of various diagnoses made clinically were missed. This is similar to Tiparse et al. who found 41% of various diagnoses made clinically to be wrong.^[4] The implication of the low accuracy is that relying on clinical diagnosis alone, more than 40% of the diagnoses would be missed. However, this is inferior to Gupta et al. who obtained a clinical accuracy of 96% done by the obstetric gynecologist and medical practitioners.^[7] This difference is explained by the disparity in the level of diagnosticians in our setting, which varied from the nurses to obstetric gynecologist through midwives and general practitioners. From our study, the highest concordance between clinical and ultrasonographic diagnoses was observed amongst the obstetrics and gynecology specialist group followed by the general practitioner, nurses, and midwives. This is similar to Aronu et al., who found that the highest concordance between clinical and ultrasonographic diagnoses was observed in obstetrics and gynecology specialist group and the least among nurses and midwives.^[16] This disparity is explained by the fact that the obstetrician–gynecologist has undergone the medical curriculum, permitting him to acquire the clinical acumen required in making a clinical diagnosis of bleeding in early pregnancy. Moreover, our study revealed an average degree of agreement between clinical and US diagnoses, for those who consulted as referrals, compared to a good agreement for those who consulted electively. This is similar to Aronu et al. who found an average degree of agreement between clinical and US diagnoses for those who consulted as referrals, compared to a good agreement for those who consulted electively.^[16] This similarity can be explained by the fact that those involved in referring included nurses and midwives who did not undergo the medical curriculum, permitting them to have the ability to make good clinical judgment and take quick decisions on pregnant women with FTVB. From our study, the association between clinical and US diagnoses of FTVB is diagnostician dependent and was found to be low with nurses and midwives. The implication of this low association is that relying on nurses and midwives to make a diagnosis of FTVB will lead to more than half of the diagnoses missed.

Conclusion

Bleeding in the first trimester of pregnancy is a common obstetrical problem needing a US assessment at the BRH and LRH. It is a source of anxiety to both the obstetrician–gynecologist and the patients. The prevalent causes of FTVB included threatened abortion, incomplete abortion, complete abortion, missed abortion, ectopic pregnancy, molar pregnancy, and pregnancy with fibroids. Our study revealed a low concordance between clinical and US diagnosis compared to other studies. There was an average and good degree of agreement between clinical and US diagnoses for FTVB depending on whether the cases were referred or elective. There was also a significant association between clinical and US diagnoses for the following pathologies; threatened abortion, incomplete abortion, complete abortion, and ectopic gestation. This highlights the need for the availability of US devices and trained sonographers to guide the clinician for better management of patients.

Acknowledgment

I would like to thank Dr. Ekongefeyin Sintieh Nchinda Ngek for training me in data analysis through his forum Sintieh Research Academy (SRA). Also, I would like to appreciate my colleagues for their moral support.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1.	Prager S, Dalton VK, Allen RH. American College of Obstetricians and Gynaecologists' Committee on Practice Bulletins – Gynaecology. ACOG practice bulletin no. 200: Early pregnancy loss. Obstet Gynecol 2018;132:e197-207.
2.	Deutchman M, Tubay AT, Turok D. First trimester bleeding. Am Fam Physician 2009;79:985-94.
3.	Baman K. First trimester vaginal bleeding at the reference health centre of Bougouni. End Study Med Thesis Univ Bamako, Mali 2010;9;38-52.
4.	Tiparse A, Gandhi B, Patel A. Ultrasonographic evaluation of first trimester bleeding. Int J Reprod Contracep Obstet Gynaecol 2017;6:3614-7. [doi: 10.18203/2320-1770ijrcog 20173495].
5.	Hoey R, Allan K. Does speculum examination have a role in assessing bleeding in early pregnancy? Emerg Med J 2004;21:461-3.
6.	Sumathy K. A comparative study of ultrasonographic evaluation of vaginal bleeding in pregnancy with clinical examination. Int J Reprod Contracep Obstet Gynaecol 2019;8:1250-3.
7.	Gupta N, Samariya M, Choudhary D, Yadav K, Kannoujiya P. Ultrasonographic evaluation of first trimester bleeding per vaginum. Int J Reprod Contracep Obstet Gynaecol 2016;5:3085-7.
8.	Coulibaly N. Study on first trimester bleeding at the obstetric and gynaecologic department of the reference health Center of commune V of Bamako District. Study Med Thesis, Univ Bamako 2010;9:22-81.
9.	Sun L, Tao F, Hao J, Su P, Liu F, Xu R. First trimester vaginal bleeding and adverse pregnancy outcomes among Chinese women: From a large cohort study in China. J Matern Fetal Neonatal Med 2012;25:1297-301.
10.	Uduma FU, Abaslattai A, Eduwem DU, Ekanem M, Okere PC. Utility of first trimester obstetric ultrasonography before 13 weeks of gestation: A retrospective study. Pan Afr Med J 2017;26:121.
11.	Neossi Guena M, Guena MN, Alapha FZ, Kemegne DC, Nana AN, Zeh OF, et al. Ultrasound study of first trimester bleeding. Open J Radiol 2019;09:58-68. [doi: 10.4236/ojrad. 2019; 910-06].
12.	Vidya A, Gururaj D, Uma AS, Ruma R. Ultrasonographic evaluation of first trimester vaginal bleeding. US Natl Lib Med Enlisted J Dep Obstet Gynaecol Al-Med Coll Hosp Athani Road Karnataka India 2016;9:107-11.
13.	Singh DK. Assessment of first trimester vaginal bleeding using ultrasound sonography. Asian J Biomed Pharm Sci 2016;6:54-6.
14.	Khatod LV, Legha D. Clinical evaluation of first trimester bleeding reported in tertiary care institute. MedPulse Int J Gynaecol 2017;3:118-22.
15.	Rose K, Joseph A, Hillary M. Sonographic findings in first trimester vaginal bleeding at Moi teaching and referral hospital. J Sci Innov Res 2019;8:46-51.
16.	Aronu ME, Okafor CO, Mbachu II, Iloraah US, Ikeako L, Okafor CI. A review of the correlation between clinical diagnosis and ultrasound diagnosis in first trimester vaginal bleeding. Ann Med Health Sci Res 2018;8:120-4.