Second pregnancy vaginal birth after cesarean- impact of maternal age on outcomes from a retrospective cohort study

Study design

A population‐based retrospective cohort study was conducted among women delivering in the secondary and tertiary care maternity centres in Qatar between January 2017 and April 2018. The study utilised data from a maternity registry, including all deliveries over the study period, to investigate outcomes among women in their second pregnancies who had previously undergone cesarean delivery (CD). The study was approved by the Medical Research Centre, Hamad Medical Corporation (MRC‐01‐24‐012) and was exempt from informed consent, since only existing data extracted from patient electronic health records was used.

Study participants

All women in their second viable pregnancies were eligible for inclusion in the study. Those with a previous CD who had antenatal care in government settings, and had a singleton, viable birth of ≥24 weeks gestational age (GA) were included in the study. The participants were divided into exposure groups based on their age at the time of delivery: young age group, YG (<25 years), advanced age group, AG (>35 years), and those with an average age of 25–35 years being the control group (CG), based on definitions of young and advanced maternal age used commonly in obstetric research [11], 14]. The main outcome was the incidence of VBAC, defined as the proportion of women having vaginal births (entire cohort as denominator). Women with missing data in the main exposure and outcome variables were excluded from the analysis.

Data source and variables

The data used for analysis were extracted from the PEARL Peristat maternity registry, excluding all patient-identifying variables. The primary exposure variable was the maternal age at the time of delivery, recorded as a continuous variable in years and then categorised as mentioned. The outcome for adjusted analysis was the mode of delivery (MOD): VBAC vs. repeat CD (including elective and emergency CD for any indication). Those who did not have a planned CD were considered to have opted for TOLAC (VBAC success being number of women having VBAC in those opting for TOLAC). Additionally, women who initially chose for CD but presented in advanced labour and subsequently delivered vaginally were included in this group. Other demographic variables included maternal nationality, body mass index (BMI) at delivery in kg/m²-divided into normal, overweight and obesity categories according to the WHO classification of BMI [15], assisted reproductive technology (ART) for conception, maternal preexisting comorbidities, and location of antenatal care and delivery (central tertiary or peripheral secondary care). These variables were considered important confounders in the association between age and mode of delivery.

Other pregnancy outcomes of interest included gestational diabetes mellitus (GDM)- defined as an abnormal 75 gm glucose tolerance test, anytime between 16 and 32 weeks of gestation depending on the patient’s risk factors [16], gestational hypertension (GHT) defined as the occurrence of high blood pressure after 20 weeks of gestation, including preeclampsia (including severe forms that can involve maternal hepatic, renal, pulmonary and cerebral systems) or eclampsia [17], gestational age- GA at delivery-preterm birth (PTB) if GA<37 completed weeks and post-dated birth if GA>40 weeks, use of oxytocin in women having a successful VBAC [18], birthweight of the baby measured immediately after birth in grams-low birthweight- LBW if<2500 g, and macrosomia if≥4,000 g [19], small for gestation (SGA) based on birthweight centiles<10 % customised for maternal characteristics such as nationality, height, weight, biological sex and GA at delivery, and large for gestation (LGA) based on birthweight centiles≥90 % [20]. The impact of these pregnancy outcomes on MOD was explored to provide insight into the mediating factors between maternal age and MOD. A complicated pregnancy was defined as a pregnancy having any maternal preexisting or new-onset comorbidities or any pregnancy complications, including fetal growth concerns.

Statistical analysis

For the descriptive analysis, all continuous variables were reported as means with standard deviations (SD) or medians with interquartile ranges (IQR), depending on the distribution of the variable. The distributions were analysed using histograms and Shapiro-Wilk test for normality if appropriate. They were compared using One-way ANOVA for means or Kruskal-Wallis test for medians. Categorical variables were described as frequencies and percentages, with comparisons done using the Chi-Square test or Fisher’s exact test as appropriate. The maternal demographics and pregnancy outcomes variables, including MOD, were compared between the three age groups. Additionally, the MOD in women developing other pregnancy outcomes were explored using bar graphs. For descriptive analysis, MOD was divided into three groups- VBAC, emergency CD and elective CD.

The association between age categories and MOD was explored using logistic regression models, yielding crude odds ratios (OR) with 95 % confidence intervals (CI). The logistic regressions were run with no VBAC (or repeat CD including both elective and emergency) vs. VBAC, as the reference category. Variables such as maternal nationality, BMI, assisted conception, preexisting comorbidities and location of antenatal care and delivery were considered as potential confounders and adjusted for, in multivariable logistic regression models to obtain adjusted ORs, after testing for collinearity. The model parameters were represented as a nomogram, which provided a visual representation of the model and provided the probability of having a repeat CD based on the age categories and other covariates in the model. The regression models were repeated in those with a complicated pregnancy vs. uncomplicated pregnancy. A p-value of<0.05 was considered strong evidence against a null hypothesis of no difference between the groups. All analyses were performed using STATA statistical software, version 18 (StataCorp, College Station, TX: StataCorp LLC) [21].

Descriptive analyses

Among the 2,729 women included, 321 belonged to YG, 273 in AG, and the remaining in CG, as shown in Figure 1. The mean age in YG and AG was 22.5 (±1.6) and 38.1 (±2.2) years, respectively, with the youngest mother being 16 years old and the oldest 50. A higher proportion of YG were Qatari, while more than half (55.7 %) in AG belonged to other non-Arab nationalities (Table 1). The mothers in AG had higher BMI, with nearly 62 % belonging to the obesity category. More than three times higher proportion had ART in AG. As expected with advancing age, all preexisting comorbidities, except asthma, were statistically significantly higher in the AG group. Nearly 60 % of all antenatal care was in the tertiary care centre, with lower proportions in YG choosing to attend secondary care facilities.

Figure 1:

Study population and comparison groups; CD- Cesarean delivery, VBAC- vaginal birth after cesarean; TOLAC- trial of labour after cesarean.

Among YG, 27.4 % underwent a VBAC, compared to 17.9 % in CG and only 9.5 % in AG (p<0.001), the VBAC rate in the whole cohort being 18.2 % (Table 2). Nearly 48 % of the cohort agreed to a TOLAC (57.9 % in YG and 41.8 % in AG), 38 % of whom had a successful VBAC. Among those who opted for TOLAC, 47.3 % in YG had a successful VBAC, compared to only 22.8 % in AG (p<0.001). Nearly 58 % of AG chose to have an elective repeat CD, compared to 42.1 % in YG and 52.8 % in the controls.

While 42.1 % in AG had GDM compared to only 20 % in YG (p<0.001), nearly 7 % in AG had PIH compared to only 1.3 % in YG (p=0.003). There were no clinically relevant differences in the incidence of preterm, post-dated births or SFD. Similarly, there were no statistically significant differences in the incidence of LFD, LBW or macrosomia. Nearly 42 % of YG had an uncomplicated pregnancy compared to only 27 % in AG (p=0.001).

The chances of having a successful VBAC were reduced in those having PIH (only 4 %), with a macrosomic or LFD baby (10–12 %), having GDD (13.7 %) or preterm birth (15.5 %), compared to 22.9 % in those who had an uncomplicated pregnancy (Figure 2). In contrast, having a baby with SFD and post-dated pregnancy increased the chances compared to the VBAC rate in the overall cohort. The proportion of women having emergency CD was highest (68 %) in those having a preterm birth, followed by PIH (52 %). Interestingly, 51 % of the whole cohort with an uncomplicated pregnancy chose to have an elective CD. When evaluating the impact of age groups in uncomplicated pregnancies, the VBAC success rates reached 60 % in YG, compared to 29 % in AG (Figure 3).

Figure 2:

Proportion of women delivering vaginally or by cesarean delivery in each of the pregnancy complication groups compared to the overall cohort and women with uncomplicated pregnancies.

Figure 3:

Comparison of TOLAC, VBAC success and VBAC rates in the age groups in complicated vs. uncomplicated pregnancies.

Measures of association between age categories and MOD

Compared to the control group, YG had 43 % lower odds of having a repeat CD (crude OR=0.57, 95 % CI=0.44–0.76, p<0.001), while AG had nearly 2.1 times higher odds (crude OR=2.08, CI=1.37–3.16, p=0.001). After adjustment for confounders (Table 3), the magnitude and direction of the association remained roughly the same (aOR=0.65, CI=0.49–0.85, p=0.002; aOR=1.93, CI=1.26–2.95, p=0.003, respectively).

In the multivariable model, among the medical illnesses, preexisting diabetes and maternal anaemia increased the odds of CD (aOR=2.58, CI=1.03–6.51, p=0.044; aOR=1.43, CI=1.00–2.05, p=0.050, respectively). Maternal essential hypertension also doubled the odds, although was non-significant. Obesity nearly doubled the odds of CD (aOR=1.88, CI=1.33–2.64, p<0.001). Interestingly, antenatal care in the secondary care hospitals increased the odds more than three times compared to the central tertiary centre (aOR=3.23, CI=1.46–7.11, p=0.004).

The multivariable model was visually represented in the nomogram shown in Figure 4. Apart from the care location, maternal age categories had the most impact on the probability of having a repeat CD, specifically the AG group, followed by preexisting diabetes, hypertension, and maternal BMI (presence of any of the above giving a score of atleast 5/10). The visual representation of the model provided an easy assessment of the risk of CD based on the included factors. For example, a woman in the AG group, with preexisting diabetes, in the obese category will have at least 95 % probability of having a CD (total score of 23.2).

As shown in Table 4, in complicated pregnancies, YG had a non-significant 29 % reduced odds of repeat CD, whereas AG had more than double the odds (aOR=2.07, CI=1.21–3.54; p=0.008). Interestingly, in uncomplicated pregnancies, the VBAC rate was 35.1 % compared to only 13.5 % in AG, and the reduced odds of CD in YG remained statistically significant (aOR=0.59, CI=0.39–0.88; p=0.010).

Figure 4:

Nomogram showing the probability of having a repeat cesarean delivery based on the covariates included in the adjusted logistic regression model shown in Table 2; ART- assisted reproduction techniques; CD- cesarean delivery.

Summary of main findings

The results of this study demonstrate an association between extremes of maternal age and mode of delivery after cesarean. Nearly three times higher proportion of younger women had a successful VBAC compared to those in the advanced age group. After adjusting for potential confounders, older women nearly doubled their odds of having a repeat CD. Complications occurring during pregnancy, such as GDM, PIH and larger birthweight, were detrimental towards VBAC rates and were also observed to be higher in older mothers. Despite having fewer complications during pregnancy and a VBAC success rate reaching 60 % in uncomplicated pregnancies, more than two-fifths of younger mothers opted for an elective CD, exposing them to the morbidity associated with future higher-order repeat CDs.

Interpretation of results and comparison with previous literature

Recommendations regarding offering VBAC emerged in the 1980s, as a means of controlling the rates of repeat CDs. Although initially popular, increased incidence of complications and the associated legal implications have reduced the rates [9]. Clinician and patient perspectives regarding TOLAC vary worldwide, resulting in a wide variation in VBAC rates. The definitions also play a part in the disparity, as VBAC rates consider the proportion of vaginal births in the entire cohort of women with previous cesarean (as endorsed by the American College of Obstetricians and Gynecologists, ACOG), whereas VBAC success considers the same, but among those who opted for a TOLAC [22]. Additionally, VBAC attempt rates have also been reported, referring to the proportion of women opting for TOLAC among the total cohort [22], 23]. For example, a large cohort study from UK in 2013 reported a VBAC incidence of 33 % in women in their second pregnancies similar to our study; however, 52.2 % opted for TOLAC, among whom 63.4 % had a successful VBAC [24].

Other European countries, such as Ireland, Italy and Germany, report high VBAC success rates reaching 75–80 %; however, these numbers included women with previous vaginal births [25]. Finland, Sweden, and the Netherlands similarly report TOLAC rates between 45 and 55 %, contrasting with 10–12 % in the USA and Australia [23]. A study comparing a 10-group CD classification system between Norway, Ireland and Slovenia report VBAC rates of 54 %, 39%, 25 % respectively, including all women with previous cesarean and singleton term births [26]. Countries of the Middle East, such as Saudi Arabia report VBAC success rate above 70 % in women without prior vaginal births [27], 28]. In our study consisting of women with no prior vaginal births, nearly 48 % opted for TOLAC, with 38 % being successful (overall VBAC incidence- 18.2 %). The concern is that more than half of the women with uncomplicated pregnancies were still choosing to have elective CDs-representing the challenging path ahead towards addressing patient concerns and improving physician confidence and counselling regarding TOLAC.

The physiological impact of advanced age on vaginal birth has been explored in several studies, attributing the higher risk of labour dystocia and CD to intrapartum decreased myometrial function, uterine blood flow and function of oxytocin receptors [29] in older mothers. Additionally, higher comorbidities such as higher BMI, preexisting diabetes and pregnancy complications such as hypertensive disorders, growth restriction and requirement for early induction of labour in them impact delivery choices, often favouring CD, perceived to be the easy and safe option [30]. These factors obviously impact TOLAC and VBAC rates, as evidenced in our study. However, what is striking is that 54 % of AG mothers with uncomplicated pregnancies opted for elective CD. This could be because majority of AG were expatriates of other nationalities, not hoping for larger families [2], biased towards CD, and who are likely to have an undocumented prior CD of unclear indication that can influence physician acceptance of VBAC.

Although there is scope for improving acceptance of VBAC in older mothers, our study results show that the other extreme age group (YG) is the key to improving the VBAC rates, especially since the VBAC success in this group goes up to 60 % in uncomplicated pregnancies, similar to the success rates in several European countries [24], 25]. Plenty of existing literature emphasises the role of age in TOLAC uptake and VBAC, with prediction models (including the validated Grobman calculator) suggesting nearly 15–20 % reduction in VBAC success for every 5-year age increment [29], 31], in those opting for TOLAC. In our study, we demonstrate a nearly 60 % reduction in VBAC success in AG when directly comparing to YG. Our nomogram also demonstrates that among the predictors included in the model, maternal age contributes most towards increasing the probability of repeat CD in this cohort. The difference noted reflects the fact that we have analysed a specific group who have not delivered vaginally prior.

The concern is that young mothers, especially those starting their reproductive journey just after adolescence, have a longer journey ahead. A larger proportion of nationals belong to this age group, and they need to satisfy the societal expectations of having larger families. Opting for repeat CD simply increases the likelihood of undergoing multiple repeat cesarean deliveries, without any striking benefit over a successful VBAC in those with no clinical contraindications. Fetal morbidity is very low in VBAC, comparable to that of nulliparous delivery, and the incidence of uterine rupture is rare [32]. Repeat CDs carry risks such as morbidly adherent placenta, adhesions, and surgical complications, which increase with each additional cesarean [4], 9]. Promoting VBAC in younger women helps reduce these risks, preserves future fertility and reduces the likelihood of severe complications in future pregnancies. Successful VBACs also result in shorter recovery times, lower infection rates, and fewer long-term health issues such as chronic pain or surgical complications from adhesions [33]. It is also the single most important predictor influencing future VBAC chances [31]. Effective counselling in this age group must include an unbiased discussion of current and future consequences of both MODs, while exploring the underlying reasons for not opting for TOLAC. The results from our study will provide valuable improvements to patient counselling by individualising the risks according to patient characteristics.

A systematic review exploring maternal perspectives and barriers against VBAC choice reported that apart from fear of risk for the baby, pregnant women felt that inconsistent and incomplete information about the choices from healthcare professionals along with coercive counselling favouring CD contributed to their confidence in TOLAC [34]. A similar evaluation of patient perspectives in Saudi Arabia revealed fear for the baby and recommendation of a CD being the top choices for mothers refusing VBAC [13]. This highlights the importance of our study as results from the local population will equip healthcare professionals with adequate and accurate information about risks and benefits, and provide better support and confidence to these mothers. Awareness about the influence of pregnancy factors on VBAC rate can help mothers and physicians proactively manage complications such as diabetes and macrosomia, and delay delivery up to 40 weeks or beyond in uncomplicated pregnancies to enhance success rates.

Exploration of physicians’ and midwives’ perspectives revealed differences in attitudes towards VBAC, ranging from positive in terms of lower risk and cost benefit to negative in terms of adverse outcomes such as uterine rupture [35]. The review suggests that in countries with higher success rates, main decision-making is from the physicians’, whereas in countries with lower rates, maternal decision-making is more assertive. A thorough dynamic risk assessment is required in YG, motivating them to take up VBAC from early pregnancy. Along with this, a positive mindset and confidence from a strong organisational support need to be demonstrated by counselling physicians or midwives, preferably trained and well-educated about factors influencing VBAC in the local population. Our study shows higher odds of repeat CD in peripheral secondary care hospitals compared to tertiary care, which highlights the confidence imparted by the availability of expertise. Younger mothers might advocate for a planned controlled CD that might allow them to manage work or study and family commitments. However, it is the responsibility of counselling professionals to encourage them to think long-term, beyond immediate social considerations and provide reassurance about expertise and healthcare resources available in the country.

Strengths and limitations

This is the first study from the region that explores the impact of age on choice and success of VBAC. It includes a substantial sample of 2,729 women, taken from a comprehensive maternity registry. The sample is therefore representative of this specific population in the country, and the results are generalizable. The categorisation of participants into three distinct age groups is based on widely accepted cutoffs and allows for valid comparisons with previous studies. Another unique advantage is that the multiethnic residential population in Qatar allows for comparison between women from different nationalities and Qatari citizens.

Since the maternal age is calculated from the date of birth recorded in participants’ national information, there is very little possibility of misclassification of exposure. Similarly, the mode of delivery is an absolute objective outcome variable that is very unlikely to be misclassified, thereby reducing any measurement bias. Adjusting for confounding variables such as maternal BMI, nationality, delivery hospital, and preexisting comorbidities strengthens the validity of the results by isolating the effect of maternal age on VBAC success. The study’s outcome variables are directly relevant to clinical decision-making and maternal health, providing actionable insights for healthcare providers.

Being a retrospective study, it is limited by potential biases in data collection and the accuracy of the medical records. Prospective studies could provide more controlled and precise data. The study is conducted in Qatar, and while it provides valuable insights, the findings may not be entirely generalizable to other countries with different healthcare systems and demographic profiles. Despite adjusting for several important confounders, there may still be unmeasured confounders that could affect the results, such as socio-economic status or indications for the initial CD. Exploration of reasons for the choice of repeat CD is lacking, and further qualitative studies will provide more insight. Future prospective studies could provide more definitive evidence and help refine VBAC guidelines and recommendations for different age groups.