Cervical tears in postpartum hemorrhage patients undergoing examination under anesthesia at a Kenyan Tertiary Hospital: a descriptive cohort study

¹Department of Obstetrics and Gynecology, School of Medicine, Faculty of Health Sciences, University of Nairobi, Nairobi, Kenya

^&Corresponding author
Duncan Mwangangi Matheka, Department of Obstetrics and Gynecology, School of Medicine, Faculty of Health Sciences, University of Nairobi, Nairobi, Kenya

Introduction    

Annually, 295000 women succumb to pregnancy and childbirth-related complications globally, of which obstetric hemorrhage contributes to 27% of fatalities [1,2]. Low-and-middle middle-income countries (LMICs), including Kenya, experience a disproportionate PPH burden despite clinical and policy interventions in place [3,4]. Postpartum hemorrhage is caused by uterine atony (70%), genital tract trauma (19%), retained tissues (10%), and thrombin (1%) [5]. Genital tract trauma is caused by cervical, vaginal and perineal tears, with cervical tears being the commonest form of traumatic PPH [6].

A lot has been done to address PPH morbidity and mortality due to uterine atony [5,7,8]; and a number of studies have shown that cervical tears could now be one of the leading causes of refractory PPH and PPH mortality [8-10]. The incidence of cervical tears ranges from 0.2% to 13%, with the rate being disproportionately higher in LMICs [5,11,12]. A secondary analysis of the WHO CHAMPION trial showed that tears caused 12.8% of responsive PPH and 28% of refractory PPH [8]. Cervical tears could occur as a consequence of obstructed labor, precipitate labor, cervical cerclage, induction and augmentation of labor, instrument delivery, nulliparity, and young maternal age, among others [11-15]. Timely interventions in PPH management are key in reducing morbidity and mortality; this includes antepartum and intrapartum risk screening, quality intrapartum care, early detection and evaluation of refractory PPH to rule out and or manage genital tract trauma [10,16,17]. In Kenya, the incidence, risk factors, management practices, and outcomes of cervical tears are not documented.

The current study was conducted to determine the incidence, patient characteristics, management, and outcomes of cervical tears among patients undergoing EUA due to PPH at KNH between January 2020 and December 2021. A better understanding of characteristics and outcomes of patients with cervical tears in LMICs is relevant to identify strengths, weaknesses, gaps, opportunities for improvement, and guide further approaches to management.

Methods     

Study design: retrospective cohort study.

Setting: this was conducted at Kenyatta National Hospital, a teaching and referral facility in Nairobi, between January 2020 and December 2021.

Participants and procedures: participants were women who had vaginal deliveries after 28 weeks´ gestation and underwent EUA due to PPH within 24 hours of delivery.

Data sources: we identified records of all patients who underwent EUA for PPH (using ICD 10 coding for PPH and EUA).

Variables: we extracted data on patient sociodemographic (age, marital status, residence, level of education and socio-economic status), clinical and reproductive characteristics (referral status, pre-existing medical conditions, parity, gravidity, gestational age, ANC attendance, previous cervical tear, cervical cerclage and fetal birth weight), labor management practices (induction of labor, amniotomy, augmentation of labor, episiotomy, mode of delivery and partograph use), EUA findings, management and outcomes (ICU admission, hysterectomy, length of hospital stay, acute kidney injury (AKI), anemia, sepsis, hypovolemic shock, disseminated intravascular coagulation (DIC), maternal death or absence of complications) using a data abstraction tool.

Statistical methods: counts, mean, and standard deviation were used to describe continuous variables. Counts, frequencies and proportions were used to describe the categorical variables; which were summarized and tabulated. SPSS version 26 was used for data analysis.

Ethical considerations: the research protocol was approved by the Kenyatta National Hospital/University of Nairobi ethics and review committee (P260/03/2022).

Results     

Participants: of the 23610 women who delivered at KNH between January 2020 and December 2021, 11120 (47.1%) had vaginal delivery; of whom 474 (4%) developed PPH, and 206 (1.9%) of those underwent EUA (Figure 1).

Descriptive data: of the 206 women undergoing EUA, most (n=151, 73%) were married and almost half (n=100, 48.5%) had attained secondary school level of education (Table 1). Most (n=195, 95%) women were urban dwellers, unemployed (n=149, 72%), while 23 (11.16%) were under the age of 20. Half (51%) were primiparous, and another half (n=107, 52%) were referrals to KNH.

Main results

Incidence of cervical tears and EUA findings: one hundred and twenty-nine women sustained cervical tears, giving an overall cervical tear incidence of 1.16% and 62.6% (95% CI 55.8% - 68.9%) among those undergoing EUA. Cervical tears occurred alone in 60 patients, with vaginal tears in 50, with vaginal-perineal tears in 12, and with uterine atony in 7 women (Table 2).

Characteristics of women with cervical tears: the mean age of women who sustained a cervical tear was 26.8 (± 6.2) years, with 10 (7.8%) under the age of 20 years. Of the 129 women who sustained cervical tears, the majority (n=102, 79.1%) were married, almost half (n=61, 47.3%) had attained secondary school level of education, and another half (n=63, 49%) were referrals to KNH (Table 1). Most (n=123, 95%) women were urban dwellers, unemployed (n=98, 76%), and multiparous (n=69, 53.5%). At least 3 antenatal visits were attended by 90% of the women, and there was paper evidence of partograph use in 90.8% of the delivery records. Of the 129 women with cervical tears, fifty-two (44.1%) delivered between 39 and 41 weeks´ gestation, hypertension was a comorbidity in 13 (10.1%), while delivery was through SVD in 127 (98.4%) and AVD in 2 (1.6%) (Table 3). In the documented cases, induction of labor was done in 21 (18.1%), amniotomy in 20 (18.7%), labor augmentation in 48 (44.9%), and episiotomy in 14 (12.2%) of the women. The majority (n=65, 53.7%) delivered babies weighing 2500-3499g, followed by 35(28.9%) women who delivered babies weighing 3500-4000g, and 10 (8.3%) women delivered infants weighing >4000g.

Management of cervical tears: of 129 women with cervical tears, repair was done in 126 (97.7%) women, tranexamic acid was administered in 52 (40.3%) women, while blood transfusion was done in 50 (38.8%) women (Table 4). Additional uterotonics after AMTSL were administered to the majority (n= 96, 74.4%) of the women. The commonest uterotonic administered either singly or in combination with other uterotonics was oxytocin in 76 (57.4%), then misoprostol 47 (36.4%), carbetocin 6 (4.7%) and ergometrine 4 (3.1%). Laparotomy was done on 3 (2.3%) women due to attendant uterine atony and secondary PPH.

Outcomes of cervical tears: of 129 women with cervical tears, an adverse maternal outcome was documented in 68 (52.7%) of them. Most common was anemia (n=68, 52.7%), followed by hypovolemic shock (n=9, 7%), ICU admission (n=6, 4.7%), AKI (n=4, 3.1%), DIC (n=3, 2.3%), and infections (n=3, 2.3%) (Table 5). The mean length of hospital stay post EUA was 2.8 days. Hysterectomy was done in 2 (1.6%) of the patients due to attendant uterine atony. There was no maternal mortality due to cervical tears. However, there were 3 maternal mortalities in the non-cervical tear group due to uterine atony.

Discussion     

Our study reported cervical tears as the most common finding (63%) among women undergoing EUA due to PPH; affecting relatively young women. We observed evidence of partograph utilization in most patient records. Anemia was the most commonly reported complication, but fortunately, we did not encounter any fatal outcomes, likely attributed to the EUA and management. In a study on the use of routine colposcopy 6-48 hours after vaginal delivery, cervical injuries were found in 66% of cases, although most of them were small, superficial, and asymptomatic [18]. This compares with the incidence of cervical tears in our study (63%). Moreover, comparing our study to prior research, the rate of cervical tears in our study (1.16%) surpassed rates reported by Parikh et al. (0.2%) and Melamed et al. (0.16%) in North America [11,12]. This discrepancy can be attributed to the distinct study settings and study designs. Notably, our findings align with a study conducted in Pakistan, where cervical tears accounted for a substantial proportion of obstetrical trauma, especially among multiparous (26.53%) and grand multiparous (44.4%) women [6]. Similarly, Mousa et al. noted that trauma was the primary cause of PPH in cases unresponsive to first-line treatment [19].

Efforts to reduce genital trauma during vaginal births have demonstrated the importance of skilled birth attendants and vigilant labor monitoring [6]. The use of partograph in labor management has proven beneficial in diminishing the risk of various complications, including prolonged labor, obstructed labor, caesarean sections, and intrapartum stillbirths [6]. The current study reaffirms the positive impact of partograph use. The World Health Organization recommends specific interventions as first-line treatment for women with PPH, including uterotonic drugs, tranexamic acid, uterine massage, and intrauterine balloon tamponade (UBT) [8,20,21]. Our study found that medical treatment predominantly consisted of uterotonics, with oxytocin being the most commonly used. Tranexamic acid and blood transfusions have also been effectively employed in PPH management [20,22-24].

For cases persistently bleeding despite first-line treatment, early EUA and possible surgical intervention become necessary [8]. Encouragingly, we observed no fatal outcomes in our study, likely attributed to the timely EUA and early management provided. This underscores the importance of ensuring prompt EUAs and early referrals for facilities unable to perform EUAs, as these factors significantly impact patient outcomes [7,25]. Surgical treatment in our study encompassed cervical tear repair, laparotomy, and, in rare cases, postpartum hysterectomy, which was necessary due to uterine atony leading to refractory PPH following EUA.

Our study serves as the inaugural effort to quantify the burden of clinically relevant cervical tears as a cause of PPH in Kenya. It provides essential baseline data for future analytical research. Nevertheless, we acknowledge several limitations, including the retrospective design, which resulted in some missing or incomplete information such as how soon interventions happened, and details on referral notes. Moreover, this was a single-facility study and may not be fully representative of the entire population or other health facilities. There is also a potential for selection bias, as the study was conducted at a teaching and referral hospital where cases may be more severe. Additionally, the descriptive study design prevented us from reporting associations. It is also worth noting that minor non-bleeding cervical lacerations that do not require suturing or EUA may not have been captured in our study, as our focus was on clinically relevant tears causing PPH.

Conclusion     

Cervical tears were the commonest causes of traumatic PPH requiring EUA. Anemia was the most common reported complication, but with no mortalities reported. This study reaffirms the role of vigilant intrapartum and postpartum monitoring for all patients, PPH prevention for those at risk, a high index of suspicion for cervical tears, tranexamic acid use, and EUAs for patients with genital trauma or refractory PPH. Measures to ensure the availability of blood products will help management and reduce the length of hospital stay post EUA.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Duncan Mwangangi Matheka: conceptualisation, initial draft, data extraction, writing, review. George Gwako: conceptualisation, review, supervision, final approval. Eunice Cheserem: writing, review, final approval. Alfred Osoti: initial draft, review, supervision, final approval. All the authors have read and agreed to the final version of this manuscript.

Tables and figure     

Table 1: maternal socio-demographic characteristics of postpartum women who underwent EUA due to PPH at KNH from 2020-2021 (n=206)

Table 2: findings among postpartum women who underwent EUA due to PPH at KNH from 2020-2021 (n=129)

Table 3: clinical characteristics and labor management practices of postpartum women who underwent EUA due to PPH at KNH from 2020-2021

Table 4: surgical and medical management of cervical tears among postpartum women who underwent EUA due to PPH at KNH from 2020-2021

Table 5: maternal outcomes of cervical tears among postpartum women who underwent EUA due to PPH at KNH from 2020-2021

Figure 1: postpartum women who underwent EUA due to PPH at KNH from 2020-2021

References