More than one in four women who had a cesarean delivery at a Chinese tertiary referral center developed a defect at their surgical scar within six months of the operation, according to a retrospective cohort study published in BMC Women’s Health. Researchers analyzed records from 156 women who gave birth by cesarean between 2021 and 2023 and underwent routine postoperative transvaginal ultrasound one to six months after surgery. They found that 43 of the women, or 27.56 percent, had a previous cesarean scar defect, a condition known as PCSD in which the uterine wall fails to heal cleanly at the incision site.
The diagnosis relied on specific imaging criteria rather than symptoms. A defect was identified when a wedge-shaped or cystic anechoic area at the cesarean scar communicated with the uterine cavity and disrupted the continuity of the myometrium, the muscular layer of the uterine wall. Because the study was conducted at a tertiary referral center, the cohort was a referred population, which may affect how the prevalence figure generalizes to other settings. The women were divided into two groups for analysis: 43 with PCSD and 113 without the condition.
To understand why some women develop these defects and others do not, the team used multivariate logistic regression to sift through potential risk factors. Four independent predictors emerged. Women with a higher number of cesarean deliveries had roughly three times the odds of a scar defect, with an odds ratio of 3.19 and a 95 percent confidence interval of 1.27 to 8.05. Emergency cesarean section, as opposed to a planned operation, was associated with an odds ratio of 3.25 (95 percent CI: 1.38 to 7.64). A retroverted uterus, meaning the uterus tilts backward rather than forward, carried an odds ratio of 2.28 (95 percent CI: 1.04 to 5.01). Finally, an incision positioned low on the uterus near the internal cervical os, the opening between the uterus and the cervix, was linked to an odds ratio of 2.91 (95 percent CI: 1.16 to 7.70).
These factors point to a combination of surgical and anatomical influences on scar healing. Repeated cesareans may thin the myometrium at the site of prior incisions, while emergency operations can involve different clinical circumstances than scheduled deliveries. A retroverted uterus may alter tension and blood flow at the lower uterine segment, and a low incision sits close to the cervix, where the muscular wall is thinner. The confidence intervals for all four predictors excluded the null value of one, indicating statistical support for each association in this cohort, though the wide intervals reflect the modest sample size.
Beyond identifying risk factors, the researchers built a nomogram, a graphical scoring tool that translates a patient’s individual characteristics into an estimated probability of developing PCSD. The model combines the four predictors into a single risk estimate that clinicians could, in principle, use during postoperative counseling. The tool was tested within the same cohort used to develop it, an approach the authors describe explicitly as internal validation rather than proof of performance in new populations.
The model’s discrimination, meaning its ability to separate women who developed defects from those who did not, was quantified with the area under the receiver operating characteristic curve. The apparent AUC was 0.858 with a 95 percent confidence interval of 0.793 to 0.923. After 500 bootstrap resamples, a statistical technique that estimates how much the model’s performance is inflated by being fit to its own data, the optimism-corrected AUC settled at 0.821. Values above 0.8 are generally considered indicative of good discrimination, though these figures describe the development cohort only.
Calibration, which measures how closely predicted probabilities match observed outcomes, was described as acceptable. The calibration slope was 0.874 and the intercept was −0.032, values close to the ideal of 1 and 0 respectively. The Hosmer–Lemeshow test yielded a P value of 0.609, indicating no statistically significant mismatch between predicted and observed risk groups, and the Brier score, which combines discrimination and calibration into a single error measure, was 0.163. Decision curve analysis suggested positive net benefit across threshold probabilities of 0.14 to 0.86 in the development cohort, meaning the model could theoretically add value across a broad range of clinical decision thresholds.
The authors are careful about the limits of their work. The nomogram, they conclude, showed acceptable apparent and internally validated performance for estimating PCSD risk in this development cohort and may support early postoperative risk stratification and counseling, but external validation is required before clinical implementation. Because the study is retrospective and single-center, and because the population was referred to a tertiary facility, the prevalence estimate and the model’s coefficients may not transfer directly to community hospitals or other health systems. Prospective studies in independent cohorts would be needed to confirm both the risk factors and the predictive tool.
Scar defects after cesarean delivery matter clinically because they can alter the structure of the uterine wall in women who may become pregnant again, and identifying which patients face higher odds could help clinicians tailor follow-up imaging and counseling. This study contributes a quantified prevalence figure for a referred cohort, a short list of modifiable and non-modifiable risk factors, and a risk-estimation tool whose real-world usefulness now awaits testing outside the dataset where it was born.
Subject of Research: Prevalence and risk factors for previous cesarean scar defect after cesarean delivery
Article Title: Prevalence and risk factors for previous cesarean scar defect formation: a retrospective cohort study
Article References: Li, S., Cheng, H., Tao, W., & Li, C. (2026). Prevalence and risk factors for previous cesarean scar defect formation: a retrospective cohort study. BMC Women's Health. https://doi.org/10.1186/s12905-026-05050-5
Image Credits: AI Generated
DOI: 10.1186/s12905-026-05050-5
Keywords: cesarean scar defect, PCSD, cesarean delivery, transvaginal ultrasound, nomogram, retrospective cohort, risk factors, myometrium, retroverted uterus, obstetrics, Prevalence, risk
News Source: Ophelia Keating. (November 6, 2026). Cesarean scar defects found in over a quarter of women in new study. Scienmag.
