A retrospective cohort study was performed at King Saud University Medical City from January 2016 to December 2019. All deliveries occurring at or beyond 20 weeks of gestation were screened for eligibility using the institution’s electronic medical record system. This database provided comprehensive documentation of maternal demographics, antenatal history, operative details, and neonatal outcomes. Women were categorized into four groups according to the number of previous CDs: 2 previous CDs, 3 previous CDs, 4 previous CDs, and 5 or more previous CDs. Parity, gravidity, and the total number of CDs during the reproductive period were compared across groups. Parity was defined as the number of births at $\ge$24 weeks’ gestation, regardless of live or stillbirth status; gravidity was defined as the total number of pregnancies; and CD was defined as delivery via abdominal and uterine incisions. Delivery type was classified as elective or emergency. Maternal variables included age, gestational age at delivery, and parity. Gestational age at delivery referred to the current pregnancy and was determined using the final obstetric ultrasound or a confirmed last menstrual period. Intraoperative variables included abnormal placentation, bladder or bowel injury, uterine rupture, major hemorrhage, blood transfusion, cesarean hysterectomy, and intensive care unitadmission. Neonatal outcomes included Apgar scores at 1 and 5 minutes, birth weight, and NICU admission.

Eligible participants were women with a documented history of 2 or more previous CDs who delivered at $\ge$20 weeks’ gestation during the study period. Women with incomplete medical records, unknown obstetric history, or those who underwent a TOLAC or a successful VBAC were excluded to ensure a uniform cohort of women undergoing repeated CDs. A total of 2101 women met the eligibility criteria and were included in the analysis.

Data were obtained from the hospital’s electronic medical record system using a standardized abstraction form completed by trained research personnel. Maternal demographic characteristics, medical and obstetric histories, operative details, and neonatal outcomes were recorded. Gestational age referred to the current pregnancy and was determined using the final obstetric ultrasound or a confirmed last menstrual period. Operative notes were reviewed to identify intraoperative complications, including bladder injury, bowel injury, uterine scar rupture, abnormal placentation, and hysterectomy. Neonatal records were examined to determine Apgar scores, birth weight, and NICU admission.

Maternal and neonatal variables were defined a priori and applied uniformly across all groups. Maternal variables included age, gravidity, parity, gestational age at delivery, diabetes mellitus, hypertension, and preeclampsia. Intraoperative outcomes included abnormal placentation, adhesions, bladder or bowel injury, uterine rupture, PPH, blood transfusion, operative time, and ICU admission. Neonatal variables included Apgar scores, prematurity status, birth weight, and NICU admission. Parity was defined as the number of births at $\ge$24 weeks’ gestation, regardless of live or stillbirth status, and gravidity as the total number of pregnancies. Grand multiparity was defined as parity $\ge$5, and high gravidity as gravidity $\ge$7. These variables are reported as separate rows in Table 1 to avoid duplication and ensure clarity. The distribution of these demographic and obstetric characteristics across the 4 CD groups is presented in Table 1, which summarizes the baseline features of the study population.

This retrospective cohort analysis used all available data from the study period; therefore, a formal priori power calculation was not conducted. Post-hoc, our cohort of 2101 women provided $>$80% power ($\alpha$ = 0.05) to detect an odds ratio (OR) of $\ge$2.0 for primary maternal outcomes (e.g., hysterectomy) between the reference group and the group with 5 or more prior CDs, given the observed outcome frequencies. The significant dose-response relationships (p-trend 0.05) further support the adequacy of the sample size to detect clinically important gradients in risk. Descriptive statistics were presented as frequencies and percentages for categorical variables. Continuous variables were compared across the 4 CD groups using the Kruskal-Wallis test, as appropriate for non-normally distributed data. Associations between categorical variables were assessed using the Chi-square test or Fisher’s exact test, as applicable, and p-values are reported in the Results section [18, 19]. The primary exposure, the number of prior CDs, was treated as an ordered categorical variable with four levels (2, 3, 4, and $\ge$5 prior CDs). Multivariable logistic regression was used to evaluate the independent association between this exposure and binary maternal and neonatal outcomes. Women with 2 prior CDs were designated as the reference group, as this group represented the lowest-risk repeat cesarean cohort, allowing for the evaluation of incremental risk with higher-order procedures.

For each outcome, we report both crude and adjusted odds ratios (AORs) with 95% confidence intervals (CIs). The final adjusted models included prespecified potential confounders identified a priori: maternal age, parity, preexisting diabetes, chronic hypertension, and emergency delivery status. The assumption of proportional odds was assessed for the ordinal exposure using a test of parallel lines and was not violated in the primary models. To evaluate a dose-response relationship, we calculated a p-value for linear trend across the ordered exposure categories by modeling the CD number as a continuous variable in the logistic regression [20]. A p-value of 0.05 was considered statistically significant. All analyses were performed using IBM SPSS Statistics version 26 (IBM Corp., Armonk, NY, USA).

Maternal characteristics varied significantly according to the number of prior CDs, highlighting a clear dose-response relationship between repeated cesareans and maternal risk. Women with higher-order CDs (4 or more prior CDs) exhibited elevated parity and a higher prevalence of chronic maternal conditions, including preexisting diabetes mellitus and hypertension. Additionally, the incidence of placenta previa and placenta accreta spectrum disorders increased progressively with the number of prior CDs, reaching the highest rates among women undergoing 4 or more procedures. Conversely, maternal age distribution, gestational diabetes, and preeclampsia showed no marked differences across the groups, suggesting that certain risks are more strongly related to surgical history than to maternal age or pregnancy-specific complications. Inadequate antenatal care was more commonly reported among higher-order CD groups, potentially reflecting both cumulative obstetric complexity and socio-demographic factors.

To examine the impact of delivery timing on these maternal risks, the cohort was stratified into preterm (28–36 weeks) and term ($\ge$37 weeks) births. Preterm deliveries often result from maternal or fetal compromise and are influenced by acute clinical indications, whereas term deliveries are more likely to be elective or scheduled. Maternal baseline characteristics stratified by the number of prior cesareans among all deliveries are summarized in Table 2A, allowing a clear comparison across the 2 CD, 3 CD, 4 CD, and $\ge$5 CD groups. Percentages reflect the proportion of women within each CD group exhibiting the specified characteristic. Chi-square or Fisher’s exact tests were performed to compare categorical variables across groups (p-values included in table footnotes).

In addition to baseline maternal characteristics, intraoperative and postoperative complications also increased with the number of prior CDs, reflecting cumulative surgical risk. Emergency cesarean procedures, PPH, and unplanned hysterectomy were more frequent among women with 4 or more prior CDs, whereas bladder injury, uterine rupture, and thromboembolism remained rare across all groups. These outcomes are summarized in Table 2B, illustrating a clear trend of increasing morbidity with higher-order cesareans.

Overall, these results demonstrate that higher-order CDs are associated with progressively increasing maternal risk, including both baseline comorbidities and intraoperative and postoperative complications. The data also highlight the importance of stratification by gestational age, as maternal and obstetric indications may differ between preterm and term births. In subsequent sections, these maternal characteristics will be evaluated alongside neonatal outcomes and adjusted analyses to quantify the independent risks associated with repeated CDs.

Neonatal outcomes worsened progressively with increasing numbers of prior CDs. After adjusting for key maternal confounders, including age, parity, diabetes, hypertension, gestational age, and adequacy of antenatal care, infants born to mothers with higher-order CDs demonstrated significantly increased risks of adverse outcomes compared with infants born after 2 CDs. Both preterm (28–36 weeks) and term ($\ge$37 weeks) infants were affected, although the magnitude of risk varied across gestational strata. Table 3 summarizes neonatal outcomes by the number of previous CDs, and presents both raw frequencies and percentages alongside AORs relative to the 2 CD reference group. Neonatal morbidity demonstrated a significant graded association with the number of prior CDs. After adjustment for key maternal confounders, the odds of adverse outcomes increased progressively across the CD groups, with the highest risks observed among infants born to mothers with 5 or more prior procedures (Table 3).

This pattern of cumulative risk was evident in both preterm and term infants. For example, among preterm infants, the AORs of NICU admission were 3.1-fold higher (95% CI: 1.8–5.4) for those in the $\ge$5 CD group compared to the 2 CD reference group. A similar incremental increase was observed for low Apgar scores. These findings indicate that the number of prior CDs is an independent predictor of neonatal morbidity, with a clear escalation in risk associated with each additional prior surgery.

In summary, both preterm and term neonates born after higher-order CDs demonstrated progressively higher rates of low Apgar scores and NICU admission. These associations remained statistically significant after adjustment for relevant maternal and obstetric factors, which indicates that the number of prior CDs is an independent predictor of neonatal morbidity.

Neonatal morbidity increased progressively with the number of prior CDs. After adjustment for maternal age, parity, preexisting diabetes, gestational diabetes, hypertension, gestational age at delivery, and adequacy of antenatal care, higher-order CDs were associated with greater odds of low Apgar scores at 5 minutes and NICU admission. Among preterm infants (28–36 weeks), the AORs of a low Apgar score were 1.8 for 3 versus 2 CDs (95% CI: 0.9–3.4; p = 0.11), 2.1 for 4 versus 2 CDs (95% CI: 1.0–4.3; p = 0.045), and 2.8 for $\ge$5 versus 2 CDs (95% CI: 1.2–6.5; p = 0.013), with a significant linear trend across the number of CDs (p-trend = 0.01). The AORs of NICU admission in preterm infants similarly increased with the number of CDs: 1.7, 2.2, and 3.1 for 3, 4, and 5 or more versus 2 CDs, respectively (p-values 0.015, 0.001, 0.001; p-trend 0.001). In term infants ($\ge$37 weeks), AORs of low Apgar scores were 1.4, 1.7, and 2.0 for 3, 4, and 5 or more cesareans versus 2 (p = 0.12, 0.038, 0.022; p-trend = 0.002). For NICU admission, the AORs were 1.5, 2.0, and 2.2 for 3, 4, and 5 or more CDs versus 2 (p = 0.012, 0.002, 0.001; p-trend = 0.001). Table 4 presents the adjusted associations between the number of prior CDs and maternal and neonatal morbidity.

These adjusted results demonstrate that the number of prior CDs is an independent predictor of morbidity, with a pronounced escalation in risk evident after the third procedure.

Several limitations should be acknowledged. The retrospective design may not fully reflect current management practices. Additionally, as this was a retrospective analysis including all available cases, no formal a priori power calculation or post-hoc power testing was performed, which may limit statistical precision for rare outcomes despite the large overall sample size. Data on variables such as body mass index, adhesion severity, and detailed placenta accreta spectrum classification were unavailable, potentially influencing outcomes. Finally, the lack of a comparative cohort of women with repeated vaginal births prevents direct assessment of cesarean-specific versus parity-related risks. Despite these limitations, our study provides robust, adjusted evidence of the cumulative operative risk associated with multiple repeat CDs.

In summary, higher-order CDs are independently associated with significantly increased maternal and neonatal morbidity, with a pronounced escalation after the third procedure. These findings support the need for careful preoperative counseling, risk stratification, and multidisciplinary planning to optimize outcomes for women with multiple prior CDs.