Quality-of-life and late effects after transabdominal cerclage: A nationwide matched cohort study.

Lea Kirstine Hansen: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data Curation, Writing – Original draft, Visualization, Project administration, Funding acquisition. Julie Glavind: Conceptualization, Methodology, Writing – Review & Editing, Supervision, Funding acquisition. Andrew H. Shennan: Writing – Review & Editing, Supervision. Niels Uldbjerg: Conceptualization, Methodology, Writing – Review & Editing, Supervision, Funding acquisition. Stina Lou: Conceptualization, Methodology, Writing – Review & Editing, Supervision. Ulrik Schiøler Kesmodel: Methodology, Writing – Review & Editing, Supervision. Pernille Tine Jensen: Conceptualization, Methodology, Validation, Writing – Review & Editing, Supervision, Project administration.

Approval to identify eligible participants was granted by the Central Denmark Region (1–45–70‐80‐24) on November 6th 2024, and the project was registered with the region's Data Protection Office (1–16–02‐69‐25). Data was processed in accordance with Danish law and the EU General Data Protection Regulation. In Denmark, questionnaire studies are exempt from formal ethical approval. All invited participants were asked to provide consent to participate.

This study was funded by the Novo Nordic Foundation (Grant ref.: NNF21OC0071037).

We identified 280 women who underwent transabdominal cerclage and subsequently gave birth by cesarean section in Denmark between 2004 and 2025 (Figure  1 ). Response rates were 73% (203/277) for transabdominal cerclage, 31% (90/287) for cesarean birth, and 34% (98/286) for the vaginal birth group. Endometriosis, fertility treatment, and a prior midtrimester loss or PTB were higher in the transabdominal cerclage group, while age, parity, and medical comorbidity were similar (Table  1 ). Characteristics of participants. Note : Variables are n (%) unless otherwise indicated. Missing items are less than 5 across all variables for all groups. Data are self‐reported. 2 Respondents may have experienced more than one event. SF‐36 scores are presented in Table  2 . For the primary outcome, the General Health Scale, the transabominal cerclage group did not differ from either comparison group. Adjusted mean differences were − 4.0 points (95% CI ‐9.5–1.4) for the cesarean birth group and − 0.6 (95% CI ‐6.2 – 4.9) for the vaginal birth group compared with the transabominal cerclage group, respectively. For all eight SF‐36 scales there was a tendency towards better quality‐of‐life scores in the transabominal cerclage group compared with the cesarean birth and vaginal birth groups. A sensitivity analysis of SF‐36 imputing missing scale scores with the group's worst observed response did not substantially alter the estimates (see supporting information Table  S1 ). Table  3 shows outcomes by time since transabominal cerclage. Mean scores increased with longer time since transabominal cerclage surgery. The adjusted General Health score was 8.5 points higher (95% CI ‐0.26–17) in women who underwent a transabominal cerclage procedure more than 10 years ago compared with those less than 5 years post transabominal cerclage. SF‐36 health‐related quality‐of‐life scores between transabdominal cerclage, cesarean, and vaginal birth groups. Score ranges 0–100. Higher scores indicate better health‐related quality‐of‐life. Note : Comparison groups Transabdominal cerclage and cesarean birth, n  = 203, No transabdominal cerclage and cesarean section, n  = 90, Vaginal birth = No transabdominal cerclage and vaginal birth, n  = 98. Abbreviations: SD = Standard Deviation, CI = Confidence Interval. Multiple linear model adjusted for fertility treatment, endometriosis, previous midtrimester loss, medical co‐morbidities, age and date of giving birth. According to time since cerclage, SF‐36 Health‐related quality‐of‐life scores in women with transabdominal cerclage. Note : Score ranges 0–100. Higher scores indicate better health‐related quality‐of‐life. Women with transabdominal cerclage ( N  = 203): Transabdominal cerclage procedure 10 years, n  = 83. Abbreviations: SD = Standard Deviation, CI = Confidence Interval. Multiple linear model adjusted for fertility treatment, endometriosis, previous midtrimester loss, and medical co‐morbidities. No groups scored ≥8 on either HADS subscale. Adjusted mean differences showed a higher anxiety score in the vaginal birth group compared to transabominal cerclage by 1.6 points (95% CI 0.5–2.7), while the cesarean birth to transabominal cerclage difference was 0.7 (95% CI ‐0.4–1.8). For Depression, adjusted mean differences were 0.7 points in both cesarean birth (95% CI ‐0.01–1.6) and vaginal birth (95% CI ‐0.2–1.5) compared with transabominal cerclage. The chronic pelvic pain prevalences were 16% for the transabominal cerclage group, 8.4% for the cesarean birth group, and 13% for the vaginal birth group, respectively. Compared with the transabominal cerclage group, the adjusted prevalence difference (adj PD) was 4 percentage points lower in the cesarean birth group (adj PD–0.04, 95% C–0.14–0.05) and 1 percentage point higher for the vaginal birth group (adj PD 0.01, 95% CI–0.09–0.11). In women who met the criteria for chronic pelvic pain, the multisite distribution of pelvic pain is illustrated in Figure  2 . Distribution of pelvic pain location in women who reported ≥6 months chronic pelvic pain. Participants could report more than one pain location. Bladder symptoms were uncommon and adjusted mean differences between the groups were ≤0.4 on the Filling, Incontinence, and Voiding scales (Figure  3 ). For Sexual Health, adjusted mean differences (95% CI) in the FSFI Full Scale Score were − 0.39 (−2.2–1.4) for the cesarean birth group and − 1.1 (−2.9–0.70) for the vaginal birth group compared with the transabominal cerclage group. The mean FSFI Full Scale Score was 27.9 (SD 5.4) for the transabominal cerclage and cesarean birth group and 27.1 (SD 4.4) for the vaginal birth group, all above the threshold of 26.55 indicating risk of sexual dysfunction. Mean scores of Female Lower Urinary Tract Symptoms. Responses to the four items administered only to the transabdominal cerclage group are shown in Figure  4 and  5 . Most women with a transabdominal cerclage reported rarely noticing or thinking about their cerclage. When recalling the time immediately following transabdominal cerclage placement, nearly all expressed trust in the procedure's effectiveness. Foreign body sensation from the permanent transabdominal cerclage. N  = 203. Top plot: “How often do you think about your abdominal cerclage.” Bottom plot: “How often do you notice sensations in your pelvic area that you believe are related to your cerclage?”. Women's trust in their transabdominal cerclage's effectiveness. N = 203. Top plot: “I believed my treatment was the best possible to achieve a viable pregnancy.” Bottom plot: “The cerclage gave me a feeling of reassurance about being able to achieve a viable pregnancy.” Twenty‐one women reported having their transabdominal cerclage removed. Of these, 3/203 (14%) underwent removal during pregnancy due to midtrimester loss, 15/203 (71%) had the transabdominal cerclage removed in conjunction with a later hysterectomy, and 3/203 (14%) underwent removal in a non‐pregnant state due to cerclage‐related complications (e.g. pain or erosion).

In this nationwide cohort, a permanent transabominal cerclage was not associated with poorer self‐perceived health, compared with not having a transabominal cerclage, regardless of mode of birth. Longer time since transabominal cerclage had a positive effect on quality‐of‐life in women with a transabominal cerclage. Mental health, bladder discomfort, pelvic pain, and sexual health were comparable between women with a transabominal cerclage and women with no transabominal cerclage. transabominal cerclage‐related complications leading to removal after completed pregnancies were uncommon. Perception of the transabominal cerclage as a foreign body was rare, and women expressed strong trust in the effectiveness of the cerclage. This study included 203 Danish women who underwent transabdominal cerclage during a 21‐year period, allowing analyses of long‐term outcomes by time since cerclage surgery and mode of birth using matched comparison groups. For this study population, it is difficult to find relevant comparison groups in the literature. However, compared with a 12‐month follow‐up study of women who underwent fertility‐sparing trachelectomy with cerclage for cervical cancer, our cohort demonstrated lower rates of urogenital morbidity and slightly affected sexual function with FSFI Full Scale Score of 27.9 in our cohort compared to scores of 20.5 in the trachelectomy cohort clearly indicating sexual dysfunction, 34 , 35 These differences likely reflect the surgical procedure with great risk of damage to the inferior hypogastric plexus during trachelectomy in the cervical cancer cohort and cancer‐related psychological distress in our cohort. Prior research suggests that cesarean birth and longer time since birth are associated with lower quality‐of‐life. 9 , 36 In contrast, our groups showed similar mean SF‐36 General Health scores regardless of mode of birth. Fertility treatments were more frequent in the transabominal cerclage group, consistent with other studies, although transabominal cerclage has been shown not to reduce fertility. 4 , 37 Endometriosis was also more prevalent in the transabominal cerclage group, and was likely a secondary finding during transabominal cerclage laparoscopy. 38 The prevalence of chronic pelvic pain of 8.4–16% in our study was comparable to an 11% prevalence in the general Danish population, likely reflecting background risk rather than transabominal cerclage‐specific morbidity. 24 Transabominal cerclage‐related complications leading to removal of the transabominal cerclage were rare in our cohort (3/203 women). This is consistent with a large transabdominal cerclage study that reported two transabdominal cerclage‐related complications in a cohort of 250 procedures. 4 Most studies do not report any complications. 12 , 13 , 39 Although synthetic intra‐abdominal materials have been linked to late complications after urogynecologic surgery 11 our findings suggest that a permanent transabdominal cerclage carries minimal long‐term symptom burden or morbidity, even after more than two decades since surgery. The smaller caliber of a cerclage suture, compared with broader urogynecologic meshes, may partly explain this. Our findings suggest that transabominal cerclage does not impose long‐term morbidity. This data provides reassurance that women treated with a transabominal cerclage do not need to be concerned about long‐term health consequences. This is particularly important because transabominal cerclage is often considered during a vulnerable period following a loss or PTB. 40 Most women who receive a transabominal cerclage have experienced pregnancy loss or PTB, often with significant emotional distress. 40 In our cohort, all women had at least one birth after their transabominal cerclage, and the trend of longer time since transabominal cerclage and higher quality‐of‐life scores may reflect psychological recovery and reassurance after successful pregnancies. Similarly, the consistently higher SF‐36 scores across domains in women with transabominal cerclage compared with matched women without transabominal cerclage may indicate improved self‐perceived health after overcoming previous adverse outcomes. Although this study provides a broader understanding of transabominal cerclages' long‐term effects, some questions remain regarding the underlying mechanisms of the quality‐of‐life scores observed in this study. To better isolate the effect of the procedure itself, future studies could include comparison groups of women with comparable obstetric risk profiles managed without transabominal cerclage, that is treated with transvaginal cerclage. Including women who do not conceive after transabominal cerclage in such follow‐up studies is equally important to capture the full range of long‐term health. The conceptual framework for the questionnaire package was informed by a literature review and input from clinical experts and patients. The package integrated validated instruments with pilot tested, study‐specific items, ensuring both psychometric robustness and clinical relevance. Face and content validity of the full questionnaire package was established. Additional strengths of the study include the prespecified MID for the primary outcome, the use of matched comparison cohorts to account for mode of birth and the high response rate in the transabdominal cerclage group. Some limitations should be acknowledged. The differences in response rates between women with transabominal cerclage and no transabominal cerclage may have increased the risk of confounding, although statistical models were adjusted for the matching variables. The difference may also have introduced a risk of selection bias, especially in the groups with lower participation. Without baseline data on non‐responders, the direction and magnitude of this potential bias cannot be determined. This may affect the comparability between groups and should be considered when interpreting the findings. The generalizability of our findings should also be considered. This study evaluates patient‐reported outcomes in women living with a transabominal cerclage, and these experiences may be broadly relevant across settings. However, transabominal cerclage is a specialized procedure generally performed in few centers, and differences in healthcare systems, access to follow‐up care, and socioeconomic context may influence patient‐reported quality of life. In addition, all women in the transabominal cerclage group had achieved at least one birth following cerclage placement, which may limit generalizability to women who did not give birth after a transabominal cerclage. Information on some potentially relevant baseline characteristics, including race/ethnicity and BMI, was not available and could therefore not be adjusted for in the analyses.

Transabdominal cerclage does not appear to be associated with impaired quality of life or persistent late effects. These findings may guide counseling of women in need of a transabdominal cerclage.

Women at risk of recurrent midtrimester pregnancy loss or spontaneous preterm birth due to a weak or short cervix may be offered a cerclage, a suture of the cervix, to reduce their risk. 1 , 2 , 3 A cerclage is performed transvaginally (TVC) or transabdominally, with transabdominal cerclage being reserved for women with prior failed TVC or extensive cervical surgery. However, transabdominal cerclage is also performed for broader indications, including one or more midtrimester loss or PTB and multiple uterine evacuations at advanced gestational ages. 4 , 5 , 6 Obstetric outcomes after transabdominal cerclage are generally favorable. 3 , 6 , 7 However, there are disadvantages of a transabdominal cerclage. It requires at least two surgeries: the transabdominal cerclage procedure and the cesarean birth. Further, transabdominal cerclages generally remain in situ for future pregnancies, leaving a permanent, synthetic suture for years, often beyond the reproductive period. 7 Synthetic intra‐abdominal materials, cesarean birth, and prior adverse pregnancy outcomes have been associated with late complications and poorer long‐term health‐related quality‐of‐life. 8 , 9 , 10 , 11 Most studies on long‐term outcomes with transabdominal cerclages emphasize obstetric endpoints, such as midtrimester loss, PTB, and live birth, while providing little or no data on maternal non‐obstetric outcomes or late surgical complications. 3 , 12 , 13 To date, long‐term effects of transabdominal cerclage have only been described in case reports, and no studies have evaluated quality‐of‐life or patient reported cerclage‐related discomfort. 14 , 15 , 16 We aimed to evaluate maternal health‐related quality‐of‐life and late effects in women who underwent transabdominal cerclage, using women who gave birth by cesarean section or vaginally as comparison groups.

The authors declare no conflicts of interest.

This is a nationwide cohort study with two matched comparison groups investigating quality‐of‐life and late effects in women who underwent transabdominal cerclage in Denmark between 2004 and 2025. We identified all women who underwent a transabdominal cerclage and later had a cesarean birth using NOMESCO procedural codes for abdominal cerclage and cesarean section and ICD‐10 codes for vaginal birth. The transabdominal cerclage group was matched 1:1:1 to two comparison groups(1): women without transabdominal cerclage with a cesarean birth, matched on year of cesarean birth (+/− 6 months) and age (+/− 2 years); and (2)women without transabdominal cerclage with non‐instrumental vaginal birth and no cesarean birth, matched on year of vaginal birth (+/− 6 months) and age (+/− 2 years). The two comparison groups were included to help distinguish potential effects related to the transabdominal cerclage from those associated with cesarean birth, alone. To minimize transient post‐partum effects on outcomes, women were ineligible if they had given birth within 6 months. Study invitation letters were sent via REDCap, Aarhus University. 17 After electronic consent, participants gained access to the questionnaire package. Non‐responders and partial responders received up to three reminders at least 4 days apart. Following a comprehensive literature search to identify relevant domains for women with a transabdominal cerclage, the following issues were conceptualized: pelvic pain, bladder discomfort, sexual health, trust in the cerclage, foreign body sensation, anxiety and depression, and quality‐of‐life. Relevant validated questionnaires were searched for each issue. Any version tested in a Danish population and covering the relevant domains was chosen to ensure cross‐cultural comparability. If relevant and validated instruments were not available, ad hoc items were developed and underwent qualitative pilot testing. To ensure face‐ and content validity, the full questionnaire package was reviewed by specialists in obstetrics and gynecology and by a patient panel of women who had previously undergone a transabdominal cerclage or TVC. The final questionnaire package was pilot tested by the patient panel. SF‐36 is a quality‐of‐life instrument that comprises eight scales: physical functioning, role limitations due to physical health, role limitations due to emotional problems, energy/fatigue, emotional well‐being, social functioning, pain and general health. 18 , 19 The reliability and validity are well‐established across diverse populations, including the Danish population. 20 , 21 All items are scored on a Likert scale ranging from 0 to 100. A high score indicates a better health state. The primary outcome in this study was the General Health scale. It was selected because it captures how respondents assess their health in the context of daily life. The General Health scale includes five items: a global self‐rating of health (“In general, would you say your health is…”), perceived susceptibility to illness (“I seem to get sick a little easier than other people”), comparative health (“I am as healthy as anybody I know”), expected health trajectory (“I expect my health to get worse”), and a positive global appraisal (“My health is excellent”).We hypothesized that living with a permanent cerclage, including any associated symptoms, could be reflected in differences in self‐perceived health in the context of daily life. This instrument was developed for patients with somatic illness and is used to screen for symptoms of anxiety and depression. 22 , 23 HADS consists of 14 items comprising two 7‐item subscales, anxiety and depression, with a 1 week time frame. Each subscale is scored on a Likert scale ranging from 0 to 21, with higher scores indicating more symptoms, and predefined thresholds of 8–10 for borderline abnormal and 11–21 for abnormal. This instrument was developed to estimate prevalence and characterize chronic pelvic pain. 24 It includes 18 items across three sections: Background variables, a chronic pelvic pain specific section assessing pain presence (≥6 months duration; yes/no) which is conditional for an item on pain location (7 pelvic and lower abdominal regions; yes/no), pain frequency (4 point Likert scale), pain intensity (numerical rating scale (NRS) 0–10) and pain‐related impact on daily life (5 point Likert scale) and a final section on prior pelvic trauma/surgery and dyspareunia. 24 A chronic pelvic pain prevalence can be calculated from the item on pain presence. The chronic pelvic pain subscale has good internal consistency (Cronbach's ⍺ = 0.82). 24 Our analyses included the chronic pelvic pain subscale, as the other sections capture correlating variables such as age and socioeconomic status, rather than validated subscales. ICIQ‐FLUTS was developed to assess female lower urinary tract symptoms in the past 4 weeks. 25 The instrument comprises three scales: Filling ( daytime frequency, nocturia, urgency, and bladder pain ), Voiding ( hesitancy, straining, and voiding difficulty ), and Incontinence ( urge and stress urinary incontinence, leakage, and nocturnal enuresis ). The validity and reliability were tested on women aged 24–80 years, with and without clinical symptoms, and showed good psychometric function (Cronbach's ⍺ = 0.78). 25 The scales are Likert scales ranging from 0 to 16 for Filling, 0 to 12 for Voiding, and 0 to 20 for Incontinence. Higher scores in each scale indicate more symptoms. There is no suggested threshold indicating disease or dysfunctions and no defined minimally important difference between groups. FSFI consists of 19 items that assess sexual function over the past 4 weeks across six scales: Sexual desire, Arousal, Lubrication, Orgasm, Satisfaction, and Pain. 26 FSFI has been validated in women with and without sexual dysfunction. 27 The subscale scores range from 0 to 6 and the Full Scale Score ranges from 2 to 36, with higher scores indicating better sexual function. An FSFI Full Scale Score of 26.55 has been defined as a valid cut‐off for differentiating women with and without sexual dysfunction. 27 The literature search identified no validated instruments for these themes, and two study‐specific items were developed for each of which two items addressed the awareness and reminding of having the cerclage and two items addressed trust in the cerclage's effectiveness to prevent PTB. Self‐reported baseline data included parity (number and year of birth), prior midtrimester loss (number) and PTB (number), fertility treatment (any), endometriosis (y/n), and other relevant medical co‐morbidities (y/n). Medical co‐morbidities were defined as the presence of one or more chronic conditions relevant to women of reproductive age, based on the definition by Joelving et al. 28 These included endocrine, autoimmune, cardiovascular, neurological, pulmonary, and psychiatric conditions. Participants were also asked whether their transabdominal cerclage had been removed and, if so, the reason for removal. No formal sample size calculation was performed. The study was designed as a nationwide cohort including all identifiable women who underwent transabdominal cerclage in Denmark during the study period, and the sample was therefore determined by the available population. Transabominal cerclage is a specialized and rare procedure. On average over the study period, approximately 25 transabominal cerclage procedures were performed per year in Denmark (unpublished data). In comparison, the Danish annual number of births is approximately 60 000. As transabominal cerclage is generally permanent and may be followed by more than one birth, procedure counts cannot be directly related to the number of births. For the primary outcome, the SF‐36 General Health scale no minimally important difference (MID) has been established in a population comparable to ours. Ideally, population‐specific MID estimates are derived using anchor‐based methods reflecting patient‐perceived change. However, such data were not available in this study. Therefore, we based our prespecified MID on published estimates from other clinical populations using the SF‐36. Previous studies report MID of 10 and 8 points for patients with endometriosis and Crohn's disease, respectively. 29 , 30 We selected an MID of 10 points, since the endometriosis population was considered the most relevant comparator in the absence of population‐specific estimates, and to reduce the risk of overestimating an effect. Participant flow was visualized according to the STROBE principles, Figure  1 . 31 Continuous variables were summarized by means and standard deviations, and categorical variables by counts and percentages. Flow chart of participants. Questionnaire responses were scored according to their manuals. 19 , 24 , 25 , 26 , 32 For SF‐36 we used the RAND Medical Outcomes Study approach. 19 For the Chronic Pelvic Pain questionnaire, we reported chronic pelvic pain prevalence and single item responses. FSFI domain scores were obtained by multiplying the scale sum score by a domain‐specific weight to a 0–6 range. The Full Scale Score 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 is the sum of the six weighted domain scores. Scale, prevalence and total scores were compared with the two comparison groups using multiple linear regression with transabominal cerclage as reference. Within the transabominal cerclage group, scores were compared to time since the transabominal cerclage procedure (5–10 and >10 years) using <5 years since the procedure as reference. Results were presented as mean scores, crude and adjusted mean differences with 95% confidence intervals. We adjusted the mean differences for prespecified confounders (endometriosis, fertility treatment, prior mid‐trimester loss, medical co‐morbidity, 28 age and date of cesarean or vaginal birth). Missing responses were handled separately for each instrument. For SF‐36, each scale consists of multiple items reflecting the same construct. As the internal consistency for each scale is high (Cronbach's ⍺ ≥ 0.78), omitting one or two items should not change the overall scale score substantially. 19 Accordingly, when item responses were missing, scale scores were computed from the mean of answered items following the scoring manual. 19 For HADS we applied the recommended half rule: if at least half of the items within a subscale were answered, missing items were imputed with the respondent's subscale mean, and the subscale score was then calculated 33 For ICIQ‐FLUTS, FSFI, and the Pelvic Pain instrument, no recommended approach for handling missing items was identified. To preserve construct validity, we used complete‐case scoring for IC1IQ‐FLUTS, FSFI, and Pelvic Pain, computing scale scores only when all items were complete. Baseline data were unavailable for complete non‐responders. For partial responders, we assessed robustness in a sensitivity analysis of SF‐36 by imputing missing items with the lowest observed score within the respondent's group. All analyses were performed in R version 4.5.1 (R Foundation for Statistical Computing, Vienna, Austria, 2025).

Table S1. Sensitivity analysis for SF‐36 Health related quality of life scores between transabdominal cerclage, cesarean and vaginal birth groups.