Abstract
Background and objective: The COVID-19 pandemic severely disrupted global maternity care, prompting concerns that logistical pressures and infection risks lowered the clinical threshold for performing cesarean sections (LSCS). This study aimed to determine whether delivering during the pandemic epoch independently increased a patient's odds of undergoing a surgical delivery, adjusting for pandemic-induced demographic shifts using the WHO Robson Ten Group Classification System.
Methods: A retrospective cohort study of 3,656 deliveries was conducted at a tertiary care center, comparing pre-pandemic (n = 2,819) and pandemic (n = 837) epochs. Baseline demographics, antenatal comorbidities, and delivery outcomes were analyzed. A multivariable binomial logistic regression model was constructed to isolate the independent effect of the pandemic on delivery mode, adjusting for maternal age, comorbidities, and Robson mega-groups.
Results: The pandemic induced a significant "triage effect," concentrating high-risk admissions. The pandemic cohort was significantly older (p < 0.001) and possessed a higher burden of metabolic comorbidities, including gestational diabetes mellitus (38.2% vs. 26.6%, p < 0.001). Despite this increased clinical complexity, the overall crude LSCS rate remained stable (33.5% vs. 34.8%, p = 0.518). Multivariable regression confirmed that the pandemic epoch did not independently increase the odds of a surgical delivery (aOR = 0.942; 95% CI: 0.768-1.156; p = 0.567). Surgical intervention was overwhelmingly driven by established obstetric indications, notably previous uterine scars (Robson 5; aOR = 22.45), malpresentations or multiple gestation (Robson 6-9; aOR = 10.70), and severe preeclampsia (aOR = 3.05). Unadjusted associations between metabolic comorbidities and LSCS lost significance in the adjusted model, proving them to be confounding variables mediated by age and prior uterine scars.
Conclusion: While the COVID-19 lockdown fundamentally shifted the demographic profile of obstetric admissions toward a highly complex, comorbid population, evidence-based surgical thresholds were rigorously maintained. Institutional increases in surgical urgency were a direct mathematical reflection of the triaged patient profile rather than a degradation of clinical practice standards.
Keywords: cesarean section, covid-19 pandemic, high-risk pregnancy, maternal morbidity, multivariable analysis, obstetric triage, robson classification
Introduction
The COVID-19 pandemic precipitated an unprecedented disruption of global healthcare systems, fundamentally altering the landscape of maternity care. To mitigate the spread of the SARS-CoV-2 virus, sweeping lockdowns were implemented, healthcare resources were reallocated, and strict infection control protocols were enforced. Consequently, obstetric care-seeking behaviors shifted dramatically. Fear of nosocomial infection, compounded by severe mobility restrictions, led many low-risk pregnant women to bypass large metropolitan hospitals in favor of peripheral midwifery-led units, smaller local clinics, or home births [1]. In contrast, tertiary care facilities were rapidly transformed into high-acuity triage hubs, absorbing a heavily concentrated influx of complicated and high-risk pregnancies [2,3].
During the initial waves of the pandemic, this sudden reorganization of patient volume coincided with alarming reports in the global literature detailing sharp spikes in institutional cesarean section rates [4]. Existing studies have raised clinical and ethical concerns regarding whether the severe logistical constraints of the pandemic - including acute staffing shortages, resource rationing, and the deliberate strategy to minimize healthcare provider exposure during prolonged vaginal labors - inadvertently lowered the clinical threshold for surgical deliveries [1,5].
The World Health Organization (WHO) had recommended Robson Ten Group Classification System serves as the global gold standard for assessing, comparing, and monitoring cesarean delivery rates [6]. By categorizing women into mutually exclusive groups based on fundamental obstetric characteristics, parity, previous uterine scars, gestational age, onset of labor, and fetal presentation, the Robson classification effectively mitigates the confounding variables that plague crude data comparisons [7].
This study utilized the Robson classification framework to investigate the core clinical drivers of caesarean sections during the lockdown period. Specifically, we sought to determine if delivering during the COVID-19 pandemic independently elevated a patient's odds of undergoing a caesarean delivery.
Materials and methods
Study design and setting
A retrospective, single-center cohort study was conducted at a tertiary care teaching hospital in Puducherry. The study evaluated obstetric admissions over a 48-month period, encompassing a pre-pandemic epoch, January 2018 to February 2020, and a COVID-19 pandemic epoch, March 2020 to December 2021. The study protocol was reviewed and approved by the Institutional Ethics Committee (IEC number - RC/2024/09), and the requirement for informed patient consent was waived due to the retrospective nature of the de-identified data analysis.
Study population
All pregnant women who delivered a viable fetus (≥28 weeks of gestation) at the study institution during the defined 48-month timeframe were eligible for inclusion. Cases with incomplete or missing primary outcome data (mode of delivery or gestational age) were excluded from the final analysis. A total of 3,656 deliveries met the inclusion criteria, yielding a pre-pandemic cohort of 2,819 patients and a pandemic cohort of 837 patients.
Data collection and variable definitions
De-identified clinical data were systematically extracted from the hospital's electronic and physical obstetric medical records. The extracted variables included maternal demographics (age, gravidity, parity, and obstetric history), gestational age at delivery, labor characteristics (onset of labor, fetal presentation), and primary mode of delivery (vaginal delivery, elective lower segment cesarean section (LSCS), or emergency LSCS).
Maternal comorbidities and labor complications were standardized from raw clinical text into binary variables. Key clinical covariates included gestational diabetes mellitus (GDM), maternal obesity, thyroid disorders (clinical and subclinical hypothyroidism), chronic or gestational hypertension, severe preeclampsia/eclampsia, and bad obstetric history (BOH). Acute intrapartum surgical indications, such as fetal distress, cephalopelvic disproportion (CPD), and cord prolapse, were also recorded.
Robson Ten Group Classification System
To objectively compare cesarean rates and adjust for the shifting baseline characteristics of the study cohorts, all patients were categorized according to the WHO Robson Ten Group Classification System [8]. To ensure statistical stability for multivariable modeling and to prevent sample size fragmentation, the 10 standard Robson groups were consolidated into five robust clinical mega-groups based on established obstetric risk profiles:
Robson 1 & 2: Low-risk, nulliparous women with a single cephalic pregnancy at term (the standard reference population).
Robson 3 & 4: Multiparous women (without a previous uterine scar) with a single cephalic pregnancy at term.
Robson 5: All women with at least one previous uterine section, single cephalic pregnancy at term.
Robson 6-9: Absolute or high-risk indications, including all multiple gestations, breech presentation, abnormal fetal lies (transverse, oblique), and abnormal lies with previous scars.
Robson 10: All women with a single cephalic pregnancy resulting in a preterm delivery (<37 weeks of gestation).
Statistical analysis
Statistical analyses were performed using R software (version 4.2.3; R Development Core Team, Vienna, Austria). Continuous variables, such as maternal age, were assessed for normality and presented as means with standard deviations (± SD); comparisons between the pre-pandemic and pandemic cohorts were conducted using independent Student's t-tests. Categorical variables were presented as frequencies and percentages, and group differences were analyzed using Pearson's chi-square test.
To identify independent predictors of the primary outcome (delivery mode: LSCS vs. NVD), both unadjusted and adjusted binomial logistic regression models were constructed. The multivariable regression model incorporated time period (pandemic vs. pre-pandemic), maternal age, antenatal comorbidities, and the categorized Robson mega-groups as independent covariates. The results of the regression analyses were reported as unadjusted odds ratios (OR) and adjusted odds ratios (aOR) with 95% confidence intervals (CI). Overall model fit was evaluated using the omnibus Chi-square test and Nagelkerke's pseudo-R2. For all analyses, a two-tailed p-value of < 0.05 was considered statistically significant.
Results
Baseline demographic and clinical shifts
A total of 3,656 deliveries met the inclusion criteria (2,819 pre-pandemic; 837 pandemic). The onset of the pandemic was associated with a significant demographic shift toward high-risk pregnancies (Table 1). The pandemic cohort was significantly older (p < 0.001) and comprised a higher proportion of multiparous women (p = 0.002) and preterm deliveries (14.7% (n = 123) vs. 11.8% (n = 332), p < 0.001). Furthermore, the burden of antenatal comorbidities intensified markedly; 82.1% (n = 687) of the pandemic cohort presented with at least one complication compared to 74.7% (n = 2,106) pre-pandemic (p < 0.001). Most notably, the prevalence of GDM (38.2% (n = 320) vs. 26.6% (n = 751), p < 0.001), maternal obesity (11.2% (n = 94) vs. 4.4% (n = 125), p < 0.001), and thyroid disorders (25.7 (n = 215) % vs. 21.2% (n = 215), p = 0.007) spiked during the lockdown epoch.
Table 1. Baseline demographic and obstetric characteristics.
Age (years) - An independent t-test was applied. Categorial variables: Chi-square test was applied.
LSCS, lower segment cesarean section; NVD, normal vaginal delivery; IUGR, intrauterine growth restriction; FGR, fetal growth restriction; PPROM, preterm premature rupture of membranes; PROM, premature rupture of membranes; GDM, gestational diabetes mellitus
| Baseline Demographic and Obstetric Characteristics | Pre-pandemic (n=2819) | Pandemic (n=837) | Total (N=3656) | Test Statistic (t/χ²) | p |
| Age (years), Mean (SD) | 25.9 (4.4) | 27.4 (4.4) | 26.3 (4.4) | 8.32 | <0.001 |
| Gravida Status | |||||
| Grand multigravida | 56 (2.0) | 9 (1.1) | 65 (1.8) | 32.04 | <0.001 |
| Multigravida | 1385 (49.1) | 503 (60.1) | 1888 (51.6) | ||
| Primi | 1378 (48.9) | 325 (38.8) | 1703 (46.6) | ||
| Parity Status | |||||
| Multipara | 211 (7.5) | 57 (6.8) | 268 (7.3) | 17.67 | <0.001 |
| Nullipara | 1591 (56.4) | 411 (49.1) | 2002 (54.8) | ||
| Primipara | 1017 (36.1) | 369 (44.1) | 1386 (37.9) | ||
| Live Children Status | |||||
| L0 | 1627 (57.7) | 418 (49.9) | 2045 (55.9) | <0.001 | |
| L1 | 1021 (36.2) | 376 (44.9) | 1397 (38.2) | 20.99 | |
| L2 | 164 (5.8) | 42 (5.0) | 206 (5.6) | ||
| L>/=3 | 7 (0.2) | 1 (0.1) | 8 (0.2) | ||
| Abortion Status | |||||
| Nulliabortus | 2247 (79.7) | 631 (75.4) | 2878 (78.7) | 7.97 | 0.047 |
| Primiabortus | 431 (15.3) | 161 (19.2) | 592 (16.2) | ||
| A2 | 101 (3.6) | 33 (3.9) | 134 (3.7) | ||
| A3 or more | 40 (1.4) | 12 (1.4) | 52 (1.4) | ||
| Single/Multiple-Ton Pregnancy | |||||
| Multiple | 69 (2.4) | 27 (3.2) | 96 (2.6) | 1.528 | 0.216 |
| Single | 2750 (97.6) | 810 (96.8) | 3560 (97.4) | ||
| Pre/Post/Term | |||||
| Post-term | 306 (10.9) | 43 (5.1) | 349 (9.5) | <0.001 | |
| Preterm | 332 (11.8) | 123 (14.7) | 455 (12.4) | 27.17 | |
| Term | 2181 (77.4) | 671 (80.2) | 2852 (78.0) | ||
| Gestation Age at Delivery | |||||
| Early preterm | 45 (1.6) | 14 (1.7) | 59 (1.6) | <0.001 | |
| Late preterm | 239 (8.5) | 92 (11.0) | 331 (9.1) | 27.67 | |
| Post-term | 306 (10.9) | 43 (5.1) | 349 (9.5) | ||
| Preterm | 48 (1.7) | 17 (2.0) | 65 (1.8) | ||
| Term | 2181 (77.4) | 671 (80.2) | 2852 (78.0) | ||
| Antenatal comorbidities | 2106 (74.7) | 687 (82.1) | 2793 (76.4) | 19.45 | <0.001 |
| GDM | 751 (26.6) | 320 (38.2) | 1071 (29.3) | 41.86 | <0.001 |
| Hypertensive disorders (GHTN or chronic HTN) | 157 (5.6) | 54 (6.5) | 211 (5.8) | 0.924 | 0.336 |
| Preeclampsia & eclampsia | 65 (2.3) | 15 (1.8) | 80 (2.2) | 0.796 | 0.372 |
| Hypothyroidism | 598 (21.2) | 215 (25.7) | 813 (22.2) | 7.470 | 0.006 |
| Anemia | 257 (9.1) | 36 (4.3) | 293 (8.0) | 20.302 | <0.001 |
| Fetal growth restriction | 156 (5.5) | 36 (4.3) | 192 (5.3) | 1.971 | 0.160 |
| Oligohydramnios | 162 (5.7) | 19 (2.3) | 181 (5.0) | 16.578 | <0.001 |
| Polyhydramnios | 17 (0.6) | 11 (1.3) | 28 (0.8) | 4.295 | 0.038 |
| Membrane rupture (PROM/PPROM) | 230 (8.2) | 77 (9.2) | 307 (8.4) | 0.909 | 0.341 |
| Obesity | 125 (4.4) | 94 (11.2) | 219 (6.0) | 52.937 | <0.001 |
| Bad obstetric history | 41 (1.5) | 11 (1.3) | 52 (1.4) | 0.090 | 0.764 |
| Rh-negative | 149 (5.3) | 40 (4.8) | 189 (5.2) | 0.338 | 0.561 |
| Previous LSCS | 381 (13.5) | 130 (15.5) | 511 (14.0) | 2.182 | 0.140 |
| Labour Characteristics | |||||
| Onset of Labour | |||||
| Spontaneous | 1284 (45.5) | 374 (44.7) | 1658 (45.4) | 0.315 | 0.854 |
| Induced | 942 (33.4) | 280 (33.5) | 1222 (33.4) | ||
| Prelabour Cs | 593 (21.0) | 183 (21.9) | 776 (21.2) | ||
| Mode of Delivery | |||||
| NVD | 1875 (66.5) | 546 (65.2) | 2421 (66.2) | 15.574 | <0.001 |
| Elective LSCS | 261 (9.3) | 115 (13.7) | 376 (10.3) | ||
| Emergency LSCS | 683 (24.2) | 176 (21.0) | 859 (23.5) | ||
| Previous LSCS | 263 (9.3) | 78 (9.3) | 341 (9.3) | 0 | 0.993 |
| Complications During Delivery/Labour That Indicated LSCS | |||||
| Fetal distress | 206 (7.3) | 36 (4.3) | 242 (6.6) | 9.438 | 0.002 |
| Malpresentation | 92 (3.3) | 28 (3.3) | 120 (3.3) | 0.014 | 0.907 |
| Cephalopelvic disproportion (CPD) & contracted pelvis | 143 (5.1) | 48 (5.7) | 191 (5.2) | 0.571 | 0.450 |
| Failure to progress/arrest of labor | 16 (0.6) | 10 (1.2) | 26 (0.7) | 3.595 | 0.058 |
| Meconium-stained liquor | 3 (0.1) | 0 (0.0) | 3 (0.1) | 0.066 | 0.797 |
| Placental complications | 33 (1.2) | 10 (1.2) | 43 (1.2) | 0.003 | 0.955 |
| Hypertensive disorders (preeclampsia, eclampsia) | 21 (0.7) | 12 (1.4) | 33 (0.9) | 3.423 | 0.064 |
| Multiple gestation | 34 (1.2) | 15 (1.8) | 49 (1.3) | 1.676 | 0.195 |
| IUGR/FGR | 9 (0.3) | 2 (0.2) | 11 (0.3) | 0.139 | 0.710 |
| Membrane issues (PROM, PPROM) | 30 (1.1) | 4 (0.5) | 34 (0.9) | 2.408 | 0.121 |
| Cord complications (prolapse, presentation) | 5 (0.2) | 1 (0.1) | 6 (0.2) | 0.132 | 0.176 |
| Robson Group | |||||
| 1 | 605 (21.5) | 137 (16.4) | 742 (20.3) | 30.26 | <0.001 |
| 2 | 739 (26.2) | 185 (22.1) | 924 (25.3) | ||
| 3 | 466 (16.5) | 162 (19.4) | 628 (17.2) | ||
| 4 | 272 (9.6) | 106 (12.7) | 378 (10.3) | ||
| 5 | 304 (10.8) | 109 (13.0) | 413 (11.3) | ||
| 6 | 49 (1.7) | 9 (1.1) | 58 (1.6) | ||
| 7 | 39 (1.4) | 8 (1.0) | 47 (1.3) | ||
| 8 | 69 (2.4) | 28 (3.3) | 97 (2.7) | ||
| 9 | 14 (0.5) | 3 (0.4) | 17 (0.5) | ||
| 10 | 262 (9.3) | 89 (10.6) | 351 (9.6) | ||
| Robson Group Category | |||||
| Robson 1&2 | 1344 (47.7) | 322 (38.5) | 1666 (45.6) | <0.001 | |
| Robson 3&4 | 738 (26.2) | 268 (32.1) | 1006 (27.5) | ||
| Robson 5 | 304 (10.8) | 109 (13.0) | 413 (11.3) | 24.20 | |
| Robson 6-9 | 171 (6.1) | 48 (5.7) | 219 (6.0) | ||
| Robson 10 | 262 (9.3) | 89 (10.6) | 351 (9.6) | ||
Delivery modes and surgical indications
Despite the significantly higher clinical complexity of the pandemic cohort, the overall binary rate of cesarean delivery remained statistically stable between the two epochs (33.8% overall LSCS; p = 0.518). However, the urgency and indications for surgical intervention shifted significantly. While the onset of labor (spontaneous vs. induced) remained unchanged (p = 0.854), there was a marked transition from emergency to elective procedures (p < 0.001). Elective LSCS procedures increased significantly during the pandemic (13.7% (n = 115) vs. 9.3% (n = 261)), whereas emergency LSCS procedures decreased (21.0% (n = 176) vs. 24.2% (n = 24.2)). Correspondingly, intrapartum fetal distress as a primary surgical indication dropped significantly during the lockdown months (4.3% (n = 36) vs. 7.3% (n = 206); p = 0.003). Obstetric indications for surgery, such as cephalopelvic disproportion (5.2% (n = 191) overall, p = 0.505) and malpresentation (3.3% (n = 120) overall, p = 0.995), showed no significant variation between the two time periods.
Robson Ten Group Classification System dynamics
The changing clinical profile of the hospital was most clearly reflected in the shifting distribution of the Robson classification groups (p < 0.001). The proportion of standard, low-risk nulliparous women with single cephalic term pregnancies (Robson Groups 1 & 2), the primary benchmark population for evaluating a facility's baseline obstetric care, contracted significantly from 47.7% (n = 1,344) in the pre-pandemic period to 38.5% (n = 322) during the pandemic.
In contrast, the facility absorbed a significantly higher proportion of patients with established obstetric histories. Deliveries among low-risk multiparous women (Robson Groups 3 & 4) increased from 26.2% (n = 738) to 32.1% (n = 268). Most notably, women presenting with a previous cesarean scar (Robson Group 5), a primary biological driver of repeat surgical deliveries, rose from 10.8% (n = 304) to 13.0% (n = 109) of the hospital's total volume. Finally, the proportion of preterm deliveries (Robson Group 10) increased (9.3% (n = 1344) vs 10.6% (n = 89)), aligning with the higher maternal comorbidity burden and increased elective surgical triage observed during the lockdown months.
Unadjusted and multivariable predictors of cesarean delivery
To determine the independent predictors of cesarean delivery, both unadjusted and adjusted binomial logistic regression models were analyzed (Table 2). The multivariable model was constructed to account for the significant demographic and clinical shifts observed during the pandemic, incorporating maternal age, relevant antenatal comorbidities, and the grouped Robson classification as covariates.
Table 2. Multivariable logistic regression to determine predictors of cesarean delivery.
LSCS, lower segment cesarean section; NVD, normal vaginal delivery; GDM, gestational diabetes mellitus; GHTN, gestational hypertension; HTN, hypertension
| Parameters | LSCS (N=1235) | NVD (N=2421) | Unadjusted OR (95% CI) | p-value | Adjusted OR (95% CI) | p-value |
| Time Period | ||||||
| Pandemic | 291 (23.6) | 546 (22.6) | 1.06 (0.90-1.25) | 0.492 | 0.942 (0.768-1.156) | 0.567 |
| Pre-pandemic | 944 (76.4) | 1875 (77.4) | ||||
| Age Mean (SD) | 27.2 (4.7) | 25.8 (4.2) | 1.43 (1.13-1.73) | <0.001 | 1.073 (1.052-1.096) | <0.001 |
| GDM | ||||||
| Present | 421 (34.1) | 650 (26.8) | 1.41 (1.22-1.63) | <0.001 | 1.138 (0.945-1.371) | 0.174 |
| Absent | 814 (65.9) | 1771 (73.2) | ||||
| Hypertensive Disorders (GHTN or Chronic HTN) | ||||||
| Present | 91 (7.4) | 120 (5.0) | 1.52 (1.15-2.02) | 0.003 | 1.305 (0.934-1.824) | 0.119 |
| Absent | 1144 (92.6) | 2301 (95.0) | ||||
| Preeclampsia & Eclampsia | ||||||
| Present | 53 (4.3) | 27 (1.1) | 3.97 (2.5-6.42) | <0.001 | 3.053 (1.792-5.203) | <0.001 |
| Absent | 1182 (95.7) | 2394 (98.9) | ||||
| Thyroid Issues (Hypothyroidism and Subclinical Hypothyroidism) | ||||||
| Present | 304 (24.6) | 509 (21.0) | 1.23 (1.04-1.44) | 0.014 | 1.136 (0.932-1.385) | 0.207 |
| Absent | 931 (75.4) | 1912 (79.0) | ||||
| Bad Obstetric History | ||||||
| Present | 27 (2.2) | 25 (1.0) | 2.14 (1.23-3.73) | 0.005 | 2.275 (1.12-4.622) | 0.023 |
| Absent | 1208 (97.8) | 2396 (99.0) | ||||
| Robson Group | ||||||
| Robson 1&2 | 432 (35.0) | 1234 (51.0) | 1 | |||
| Robson 3&4 | 91 (7.4) | 915 (37.8) | 0.28 (0.22-0.36) | <0.001 | 0.23 (0.178-0.296) | <0.001 |
| Robson 5 | 373 (30.2) | 40 (1.7) | 26.59 (19.01-37.92) | <0.001 | 22.45 (15.847-31.803) | <0.001 |
| Robson 6-9 | 178 (14.4) | 41 (1.7) | 12.38 (8.73-17.86) | <0.001 | 10.701 (7.452-15.369) | <0.001 |
| Robson 10 | 160 (13.0) | 191 (7.9) | 2.39 (1.89-3.03) | <0.001 | 1.903 (1.485-2.438) | <0.001 |
The regression analysis confirmed that delivering during the COVID-19 pandemic epoch did not increase a patient's odds of undergoing a cesarean section. This remained true in both the crude, unadjusted analysis (OR = 1.06; 95% CI: 0.90-1.25; p = 0.492) and the final multivariable model (aOR = 0.942; 95% CI: 0.768-1.156; p = 0.567).
The analysis revealed a profound confounding effect regarding maternal metabolic comorbidities. In the unadjusted analysis, GDM (OR = 1.41, p < 0.001), hypertensive disorders (OR = 1.52, p = 0.003), and thyroid issues (OR = 1.23, p = 0.014) were all significantly associated with higher odds of surgical delivery. However, after mathematically adjusting for maternal age and obstetric history, all three of these conditions lost their statistical significance (p > 0.05). This demonstrates that their apparent influence on the delivery mode was entirely mediated by advancing maternal age and previous surgical scars, rather than the metabolic conditions acting as independent triggers for surgery.
The true, independent drivers of surgical delivery were rooted in severe maternal morbidity and established obstetric history. Preeclampsia and eclampsia remained a massive independent risk factor, tripling the odds of an LSCS (aOR = 3.05; p < 0.001), while advancing maternal age increased the odds by 7.3% per year (aOR = 1.07; p < 0.001). A BOH also significantly more than doubled the odds of surgical intervention (aOR = 2.27; p = 0.023).
Finally, the Robson classification served as the strongest predictor of delivery mode. Using low-risk, term nulliparous women (Robson Groups 1 & 2) as the reference baseline, the odds of LSCS were overwhelmingly driven by previous cesarean scars (Robson 5; aOR = 22.45, p < 0.001) and absolute surgical indications, such as malpresentations and multiple gestations (Robson 6-9; aOR = 10.70, p < 0.001). Preterm deliveries (Robson 10) also independently nearly doubled the odds of surgery (aOR = 1.90; p < 0.001), while low-risk multiparous women (Robson 3 & 4) exhibited a highly significant 77% reduction in surgical odds (aOR = 0.23; p < 0.001).
Discussion
The primary objective of this study was to determine whether the logistical and clinical pressures of the COVID-19 pandemic altered the threshold for performing cesarean deliveries. Our findings demonstrate that, while the pandemic caused a profound demographic shift toward older, multiparous, and highly comorbid admissions, delivering during the lockdown epoch did not independently increase a patient's odds of undergoing a surgical delivery (aOR = 0.942). Furthermore, while crude, unadjusted analyses suggested that metabolic conditions, such as GDM and hypertensive disorders, significantly increased surgical risk, multivariable adjustments revealed these to be confounding variables. Their influence was heavily mediated by advancing maternal age and prior uterine scars. Ultimately, clinical decision-making remained resolute; surgical interventions were driven entirely by established, absolute obstetric indications rather than pandemic-induced logistical pressures.
The profound baseline demographic shifts observed within our cohort - specifically the significant contraction of low-risk, term nulliparous admissions (Robson Groups 1 & 2) and the concurrent surge in multiparous women with previous uterine scars (Robson Group 5) - are emblematic of a broader, global "triage effect" precipitated by the COVID-19 pandemic. Throughout the initial waves of the outbreak, stringent mobility restrictions, the reallocation of healthcare resources, and widespread patient apprehension regarding nosocomial SARS-CoV-2 transmission fundamentally altered obstetric care-seeking behaviors [2,9].
A comprehensive systematic review by Chmielewska et al. evaluating the global impact of COVID-19 on maternal outcomes demonstrated a precipitous decline in overall institutional deliveries, particularly in low- and middle-income countries (LMICs) [10]. Healthy women with uncomplicated, low-risk pregnancies frequently bypassed major metropolitan hospitals, opting instead for peripheral health centers, midwifery-led maternity units, or home births to distance themselves from the epicenters of the viral outbreak [1].
Consequently, tertiary care facilities were inadvertently transformed into high-acuity triage hubs. The concentration of clinical complexity observed in our study, characterized by older maternal age, higher parity, and a significantly heavier burden of metabolic comorbidities such as GDM, directly mirrors trends reported in international tertiary cohorts during the lockdown epochs. Patients presenting with a previous uterine section (Robson Group 5) or absolute surgical indications (Robson Groups 6-9) could not safely utilize peripheral centers and were funneled into the tertiary system, effectively saturating the admission profile with high-risk pregnancies [3,11].
This concentration of clinical complexity provides a critical epidemiological context for interpreting pandemic-era surgical rates. Early pandemic literature by Bhatia et al. [4] had reported alarming, unadjusted spikes in institutional cesarean delivery rates. However, our Robson-stratified analysis corroborates the emerging consensus that these initial reports were heavily influenced by a "denominator fallacy". The apparent surge in the crude surgical rate at tertiary centers was not driven by a diminished threshold for surgical intervention among obstetricians facing logistical panic; rather, it was a mathematical artifact caused by the disappearance of the denominator, the low-risk vaginal deliveries that had been diverted elsewhere [12,13]. By utilizing the Robson classification, our study unmasks this triage effect, confirming that, when adjusted for the shifting influx of surgical risk, institutional adherence to evidence-based delivery protocols remained remarkably resilient.
In our unadjusted analyses, the pandemic cohort's heavily concentrated burden of metabolic conditions, specifically GDM, chronic/gestational hypertension, and thyroid disorders, appeared to significantly increase the odds of surgical delivery. However, upon adjusting for maternal age and the Robson classification, these metabolic comorbidities completely lost their statistical significance.
This statistical attenuation highlights a well-documented epidemiological phenomenon: metabolic conditions frequently serve as proxy indicators for underlying demographic risk factors rather than acting as independent mandates for surgery [14,15]. Our adjusted model (aOR = 1.138; p = 0.174) aligns with recent global consensus demonstrating that uncomplicated GDM, when properly managed and in the absence of suspected fetal macrosomia, does not inherently justify a departure from vaginal delivery protocols [16].
The fact that these metabolic variables washed out in the multivariable model provides compelling evidence of clinical restraint. It indicates that the institution's obstetric department did not utilize the mere presence of a COVID-era, GDM, or hypertension diagnosis as a convenient threshold to expedite labor via surgery. Instead, surgical interventions were strictly reserved for cases where these metabolic conditions cascaded into acute, life-threatening complications or when they intersected with established obstetric risk factors.
By neutralizing the confounding effects of the pandemic's demographic shift, the regression model exposes the true, independent drivers of surgical delivery in our cohort. These drivers were rooted entirely in severe maternal morbidity and established obstetric indications, underscoring a rigorous adherence to global obstetric guidelines despite the logistical chaos of the COVID-19 lockdowns.
The most profound predictor of surgical delivery was the presence of a previous uterine scar (Robson Group 5), which increased the odds of a repeat LSCS more than 22-fold (aOR = 22.45). This finding is consistent with global data from the WHO, which consistently identifies Robson Group 5 as the single largest contributor to overall institutional cesarean rates worldwide, driven by the inherent risks of uterine rupture during a trial of labor after cesarean (TOLAC) [6,17]. Similarly, absolute indications, such as malpresentations, abnormal lies, and multiple gestations (Robson Groups 6-9), exhibited a nearly 11-fold increase in surgical odds, reflecting non-negotiable barriers to safe vaginal delivery [7].
Furthermore, while chronic hypertension lost its significance, preeclampsia and eclampsia remained massive independent risk factors, tripling the odds of surgical intervention (aOR = 3.05). This divergence perfectly illustrates evidence-based triage. Unlike chronic hypertension, severe preeclampsia represents an acute, progressive endothelial crisis where immediate, and often surgical, delivery remains the only definitive treatment to prevent maternal stroke, hepatic rupture, or fetal demise [7].
Further, the isolation of these specific, severe clinical determinants, advanced maternal age, previous surgical scars, malpresentations, and preeclampsia, confirms that the fundamental mechanisms of obstetric decision-making remained intact. The crude increase in surgical volume was a mathematical inevitability dictated by the triaged, high-risk physiology of the patients admitted, rather than a degradation of clinical practice standards by the attending physicians.
The primary strength of this study is its robust sample size (N = 3,656) and the integration of the WHO Robson classification with multivariable modeling, which successfully isolated true clinical drivers from pandemic-induced demographic confounders. However, as a single-center retrospective study, the findings may lack broad generalizability. Further, low-risk women who bypassed tertiary hospitals during lockdowns, meaning our dataset inherently reflects a highly triaged population. Additionally, unmeasured logistical confounders, such as acute staffing shortages or provider fatigue, could not be captured.
Conclusions
The COVID-19 pandemic induced a profound "triage effect," fundamentally shifting the demographic profile of obstetric admissions toward older, multiparous, and highly comorbid patients. However, despite these surging clinical complexities and the unprecedented logistical pressures of the lockdown, delivering during the pandemic epoch did not independently increase a woman's odds of undergoing a cesarean delivery. Multivariable analysis confirms that surgical interventions remained strictly dictated by absolute, evidence-based obstetric indications - predominantly previous uterine scars, malpresentations, and severe preeclampsia. These findings highlight the remarkable resilience of the obstetric care system, demonstrating that rigorous clinical thresholds for surgical delivery were successfully maintained during a global public health crisis.
Disclosures
Human subjects: Informed consent for treatment and open access publication was obtained or waived by all participants in this study. PIMS Institute Ethics Committee issued approval IEC: RC/2024/09. Approved from 13.02.2025 for the period of 1 year until 12.02.2026.
Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.
Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following:
Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work.
Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work.
Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.
Author Contributions
Concept and design: Thamizhmaran Sundararajan, Pavithra Devi Babu, Sreelakshme Nilakandan, Kiruba Haridassan
Acquisition, analysis, or interpretation of data: Thamizhmaran Sundararajan, Pavithra Devi Babu, Sreelakshme Nilakandan, Kiruba Haridassan
Drafting of the manuscript: Thamizhmaran Sundararajan, Pavithra Devi Babu, Sreelakshme Nilakandan, Kiruba Haridassan
Critical review of the manuscript for important intellectual content: Thamizhmaran Sundararajan, Pavithra Devi Babu, Sreelakshme Nilakandan, Kiruba Haridassan
Supervision: Pavithra Devi Babu
References
- 1.The impact of the coronavirus (COVID-19) pandemic on maternity care in Europe. Coxon K, Turienzo CF, Kweekel L, Goodarzi B, Brigante L, Simon A, Lanau MM. Midwifery. 2020;88:102779. doi: 10.1016/j.midw.2020.102779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Effect of the COVID-19 pandemic response on intrapartum care, stillbirth, and neonatal mortality outcomes in Nepal: a prospective observational study. Kc A, Gurung R, Kinney MV, et al. Lancet Glob Health. 2020;8:0–81. doi: 10.1016/S2214-109X(20)30345-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Maternal and neonatal morbidity and mortality among pregnant women with and without COVID-19 infection: the INTERCOVID multinational cohort study. Villar J, Ariff S, Gunier RB, et al. JAMA Pediatr. 2021;175:817–826. doi: 10.1001/jamapediatrics.2021.1050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.The effect of COVID-19 on general anaesthesia rates for caesarean section. A cross-sectional analysis of six hospitals in the north-west of England. Bhatia K, Columb M, Bewlay A, et al. Anaesthesia. 2021;76:312–319. doi: 10.1111/anae.15313. [DOI] [PubMed] [Google Scholar]
- 5.Coronavirus disease 2019 (COVID-19) pandemic and pregnancy. Dashraath P, Wong JL, Lim MX, et al. Am J Obstet Gynecol. 2020;222:521–531. doi: 10.1016/j.ajog.2020.03.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Use of the Robson classification to assess caesarean section trends in 21 countries: a secondary analysis of two WHO multicountry surveys. Vogel JP, Betrán AP, Vindevoghel N, et al. Lancet Glob Health. 2015;3:260–270. doi: 10.1016/S2214-109X(15)70094-X. [DOI] [PubMed] [Google Scholar]
- 7.WHO statement on caesarean section rates. Betran AP, Torloni MR, Zhang JJ, et al. https://www.who.int/publications/i/item/WHO-RHR-15.02. BJOG. 2016;123:667–670. doi: 10.1111/1471-0528.13526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Robson classification: implementation manual. [ Jun; 2026 ]. 2017. https://www.who.int/publications/i/item/9789241513197 https://www.who.int/publications/i/item/9789241513197
- 9.Global changes in maternity care provision during the COVID-19 pandemic: a systematic review and meta-analysis. Townsend R, Chmielewska B, Barratt I, et al. EClinicalMedicine. 2021;37:100947. doi: 10.1016/j.eclinm.2021.100947. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Effects of the COVID-19 pandemic on maternal and perinatal outcomes: a systematic review and meta-analysis. Chmielewska B, Barratt I, Townsend R, et al. Lancet Glob Health. 2021;9:0–72. doi: 10.1016/S2214-109X(21)00079-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.The effect of the COVID-19 pandemic on maternal health due to delay in seeking health care: experience from a tertiary center. Goyal M, Singh P, Singh K, Shekhar S, Agrawal N, Misra S. Int J Gynaecol Obstet. 2021;152:231–235. doi: 10.1002/ijgo.13457. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Profile of cesarean section during COVID-19 using Robson ten group classification system in a tertiary teaching hospital, Nepal: a cross-sectional study. Shrestha N, Regmi P, Shrestha S, Nyaupane K. Ann Med Surg (Lond) 2023;85:1699–1704. doi: 10.1097/MS9.0000000000000622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.The ten-group Robson classification: a single centre approach identifying strategies to optimise caesarean section rates. Tanaka K, Mahomed K. Obstet Gynecol Int. 2017;2017:5648938. doi: 10.1155/2017/5648938. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gestational diabetes mellitus and late preterm birth: outcomes with and without antenatal corticosteroid exposure. Nazeer SA, Chen HY, Chauhan SP, Blackwell SC, Sibai B, Fishel Bartal M. Am J Obstet Gynecol MFM. 2024;6:101268. doi: 10.1016/j.ajogmf.2023.101268. [DOI] [PubMed] [Google Scholar]
- 15.Prevalence of gestational diabetes and risk of progression to type 2 diabetes: a global perspective. Zhu Y, Zhang C. Curr Diab Rep. 2016;16:7. doi: 10.1007/s11892-015-0699-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.ACOG practice bulletin no. 190: gestational diabetes mellitus. American College of Obstetricians and Gynecologists. Obstet Gynecol. 2018;131:0–64. doi: 10.1097/AOG.0000000000002501. [DOI] [PubMed] [Google Scholar]
- 17.Global epidemiology of use of and disparities in caesarean sections. Boerma T, Ronsmans C, Melesse DY, et al. Lancet. 2018;392:1341–1348. doi: 10.1016/S0140-6736(18)31928-7. [DOI] [PubMed] [Google Scholar]