Prevalence of isthmocele according to diagnostic hysteroscopy, transvaginal ultrasound, and magnetic resonance imaging in non-pregnant women: cross-sectional studySUMMARY
OBJECTIVE: The aim of this study was to estimate the prevalence of isthmocele using transvaginal ultrasound, magnetic resonance imaging and diagnostic hysteroscopy.
METHODS: A cross-sectional study conducted from 2020 to 2022 in tertiary academic hospital with ninety non-pregnant women with a history of cesarean section. Transvaginal ultrasound (n=90), magnetic resonance imaging (n=65) and office diagnostic hysteroscopy (n=81) were performed within a short interval for each patient.
RESULTS: The women had a mean age of 34.8±6.3 years, mean number of 2.0±1.1 cesarean sections and mean duration of 40 months since their last cesarean. The prevalence of isthmocele with 95%CIs was 11.3% (5.5–19.7%) on transvaginal ultrasound, 66.1% (53.3–77.4%) on magnetic resonance imaging, and 43.2% (32.2–54.6%) on diagnostic hysteroscopy. When transvaginal ultrasound was considered reference, accuracy for diagnostic hysteroscopy was 57.89% (44.1%; 70.6%) and magnetic resonance imaging, 43.86% (30.9; 57.5%); when magnetic resonance imaging was reference, accuracy for diagnostic hysteroscopy was 54.39% (40.75; 67.43%). Agreement analysis (kappa) among the methods showed low scores between diagnostic hysteroscopy and magnetic resonance imaging regarding transvaginal ultrasound (0.126, p<0.001 and 0.103, p<0.001, respectively) and between diagnostic hysteroscopy and magnetic resonance imaging (0.206, p=0.014).
CONCLUSION: The prevalence of isthmocele was lower in transvaginal ultrasound and higher in magnetic resonance imaging. Accuracy varied from 43 to 58% among the three methods and agreement between tests for diagnosing isthmocele was low.
KEYWORDS:
Ultrasonography; Hysteroscopy; Cesarean section; Magnetic resonance imaging
INTRODUCTION
Isthmocele is defined as a dehiscence at the site of a previous cesarean section scar, which acts as a reservoir collecting blood during the menstrual period, leading to abnormal uterine bleeding1,2. The prevalence of isthmocele varies from 19 to 84%, depending on the method used for diagnosis and when we individualize the diagnosis by hysteroscopy, we observe that data in the literature are scarce3,4. This variation is possibly due to differing definitions of isthmocele, the types of studies conducted, and the diagnostic methods employed5,6.
When assessed using transvaginal ultrasound (TVUS), isthmocele appears as an anechoic triangular area in the anterior region of the isthmus, easily detectable in non-pregnant women. Its shape and morphology can vary, presenting as a round, square, or wedge-shaped cavity. On diagnostic hysteroscopy (DxHyst), isthmocele appears as a cleft in the anterior uterine wall7. Magnetic resonance imaging (MRI) allows for clear delineation of the defect borders and measurement of myometrial thickness at the scar level. Despite several diagnostic techniques, there is no consensus on a gold standard for diagnosing isthmocele, although some studies propose standardizing classification using ultrasound8,9.
Despite several diagnostic techniques, there is no consensus on a gold standard for diagnosing isthmocele, although some studies propose standardizing classification.
The Diagnostic performance of 3D-TVUS does not appear to be significantly better than of 2D-TVUS9. However, both MRI and hysteroscopy, despite their higher costs, offer excellent accuracy8,10,11. Therefore, this study aims to estimate the prevalence of isthmocele using 2D-TVUS, MRI, and DxHyst, and to evaluate the accuracy and agreement between these.
METHODS
We conducted a cross-sectional study at tertiary hospital, from November 2020 to December 2022. The study included 90 women with a history of cesarean section between 2015 and 2020, selected from a total of 452 women who had undergone cesarean sections. Of these, 277 were invited via telephone, email or during postpartum consultation. Women aged between 18 and 45 Women aged 18–45 years who had at least one cesarean section at least 6 months prior to study enrollment and within a maximum interval of 5 years were included. Pregnant women and those using hormonal or non-hormonal intrauterine devices were excluded. The study was approved by the Institutional Review Board of University of Campinas (CAAE 02238118900005404). The STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) checklist for cross-sectional studies were followed.
Participants underwent two-dimensional TVUS and DxHyst by independent examiners on the same day to establish the diagnosis of isthmocele. MRI was performed the following day at an ultrasound clinic. Each exam was conducted by experienced physicians, each with at least 10 years of practice. The physicians were blinded to the results of the other examinations.
Data were collected on the following variables: age, body mass index (BMI), age at menarche, parity, presence and number of cesarean sections, time elapsed since the last cesarean section, and the indication for cesarean section (elective or emergency).
Technical details about the instruments have been previously described in a published study with the same cohort of women to assess risk. Two-dimensional TVUS was performed using Toshiba Xario or Voluson E8 devices to evaluate the uterus and adnexa. The isthmocele was assessed for its width (W), depth (D), length (L) and residual myometrial thickness (RMT). MRI was performed using the Achieva 1.5 Tesla equipment (Philips, Eindhoven, Netherlands) with a six-receptor coil dedicated to high-resolution pelvic examinations. The exams were conducted with patients in a supine position, with moderate bladder repletion with no vaginal gel. The diagnosis of other uterine pathologies and the signal intensity of the contents of the endometrial cavity and isthmocele contents were described. Parameters regarding the length (L), depth (D), width (W), and RMT of the isthmocele were measured. Diagnostic hysteroscopy was performed on an outpatient basis without anesthesia, following the technique described by Bettocchi and Selvaggi in 199712, using saline solution as the distension medium. The isthmocele was evaluated for the presence of tent-shaped retraction on the anterior wall of the myometrium at the isthmus and the presence or absence of edge hypertrophy in this area.
Statistical analysis
Sample size calculation was described in a previous publication analyzing factors associated with isthmocele13. Since the primary outcome for this study differs from the previous one, we used a convenience sample. In the previous publication, a composite variable derived from all three exams was used to calculate the global prevalence of isthmocele. However, since this study aims to identify the prevalence of isthmocele from each of the three exams individually, the composite variable was not used here. Descriptive analysis was performed to evaluate the frequency, means and standard deviations of the variables. Prevalence was displaying as proportion plus 95%CIs. Given the absence of a gold standard diagnostic test for isthmocele, 2D-TVUS was used as the to calculate the sensitivity, specificity, positive predictive value, negative predictive value and accuracy plus 95%CIs of the tests. To evaluate the agreement between the methods, the kappa coefficient kappa coefficient was calculated, and the McNemar test was used to compare the prevalence of isthmocele between the methods in related samples. Statistical analysis was performed using SAS version 9.4 (Cary, NC, USA) with a significance level of 5%.
RESULTS
Table 1 shows the sociodemographic characteristics of the patients. The mean age was 34.8±6.3 years and the mean number of cesarean sections was 2.0±1.1, with 55.6% of women having two or more cesarean sections. The time elapsed from the last birth to inclusion was 40.1±14.7 months.
The prevalence of isthmocele varied according to the diagnostic method: 11.3% (5.5–19.7%) with TVUS, 66.1% (53.3–77.4%) with MRI and 43.2% (32.2–54.6%) with hysteroscopy. The mean uterine volume was 111.3±57.7 cm3 for TVUS and 98.3±44.7 cm3 for MRI (p=0.542). The cesarean section scar was identified in 100% of MRI exams and 71% of TVUS exams. Hysteroscopy found that 85% of anteverted uteri were anteverted and 10 participants had their exams abbreviated due to pain or cervical stenosis. RMT was greater than 3 mm in 42 (97.6%) subjects with isthmocele. Regarding isthmocele measurements, TVUS showed a larger width than MRI; all other measures were similar (Table 2).
Assessment of transvaginal ultrasound, magnetic resonance imaging, and diagnostic hysteroscopy for isthmocele.
Table 3 compares the three diagnostic methods for isthmocele. MRI and DxHyst had the highest sensitivity values, and DxHyst had the highest accuracy rates. However, the agreement rate for diagnosing isthmocele among the methods was low. When comparing MRI results to TVUS (used as the reference in this study), there was a significant difference in the prevalence of isthmocenle on MRI (p<0.001) with an accuracy of 43.86% (30.98%; 57.57%). When comparing hysteroscopy to TVUS, a higher prevalence was also found with hysteroscopy (p<0.001) and an accuracy of 57.89% (44.13%; 70.60%). When comparing the two most accurate methods, hysteroscopy and MRI, there was still low agreement between them (p=0.014). MRI had an accuracy of 54.39% (40.75%; 67.43%) when compared to hysteroscopy (Table 3).
DISCUSSION
In our study, the prevalence of isthmocele varied according to the diagnostic method with MRI and hysteroscopy presenting the highest proportions (66 and 43%). MRI had an accuracy of 43.8% and the hysteroscopy of 57.89% considering TVUS as a reference. There was low agreement between diagnostic methods.
We evaluated the accuracy of three methods—TVUS, MRI, and DxHyst., compared in pairs for diagnosing isthmocele. This study is one of the few that assess all three methods for this disorder. However, our results did not show high accuracy rates compared to some other studies in the literature.
Most of the recommendations suggest that TVUS should be the starting point for diagnosing isthmocele, as acquiring the simple sagittal plane and performing contrast sonohysterography are straightforward and require only medium-level diagnostic skills9,14. TVUS can be complemented by contrast-enhanced sonohysterography, using intrauterine instillation of saline solution or gel. Second-generation contrast agents that have been tested during contrast-enhanced US have shown superior accuracy with conventional 2D TVUS, according to a recent systematic review15. In the current literature, as well as in our study, prevalence rates vary according to the diagnostic method, ranging from 24 to 70% with TVUS and from 56 to 84% with sonohysterography14,16. Antila-Långsjö et al, in a study carried out with 371 women undergoing TVUS and sonohysterography, observed that the isolated use of TVUS failed to diagnose about half of the isthmoceles in this population10. Similarly, in our study, compared with MRI, only a quarter of isthmoceles were identified on TVUS.
Variation in the prevalence rates of isthmocele by TVUS may depend on the examiner’s experience, the complexity of the lesions, the type of suture, and the time elapsed since the cesarean section and diagnosis. A recent study found a 42.3% prevalence of isthmocele six weeks after c-section and an association between locking continuous sutures and the presence of more niches16,17.
TVUS isthmocele dimensions were similar to MRI measurements, except for the isthmocele ´width. A recent study analyzed whether these dimensions could predict the ability to confirm isthmocele and found that defect length and width significantly contributed to the diagnosis18.
MRI is an expensive exam and usually has higher sensitivity and accuracy rates than TVUS for diagnosing isthmocele. Our accuracy rates were lower than usual, but the prevalence rates were similar to previous studies19. Generally, MRI accuracy rates are higher than TVUS19,20. Despite its higher cost, MRI, like hysteroscopy depends on more advanced technology but is more accurate than TVUS7,8,19. Gupta et al. did not find a statistically significant difference between MRI and TVUS for diagnosing isthmocele19. We used a 1.5 T MRI to detect isthmocele. Although we did not compare it to other MRI technologies, such as 3.0 T, a recent study using 3.0 T MRI found an accuracy of 92.86% compared to 69.64% for TVUS20. Moreover, MRI derived parameters may be used to differentiate an abnormal post-cesarean uterine scar from a normal one, similar to our study’s findings21.
Most of the studies have addressed hysteroscopy as both a diagnostic and treatment method for isthmocele. It is a well-established method for evaluating the uterine cavity22,23. An advantage of DxHyst is its ability to exclude other endometrial pathologies that could cause similar symptoms. However, a disadvantage is that it cannot measure the thickness of the residual myometrium over the niche24,25. Despite this, its accuracy rate is similar to MRI.
The study has several limitations. The small sample size, resulting from the recruitment phase during the COVID-19 pandemic, might have increased the odds of a type 2 error. Additionally, this was a single-center study, which may limit external generalizability. There is also potential response bias as we do not have the profile of non-responders to the study invitation.
Therefore, we propose the need for new further, prospective, cohort studies to define the most appropriate diagnostic method for identifying isthmocele. TVUS should be performed by two different operators to calculate inter-variability influence, with similar time intervals from the last cesarean section, and from different centers. Larger samples would allow for a more comprehensive study of this subject, using statistical tools to correct for confounding bias, such as propensity score analysis. Comparisons with MRI and hysteroscopy should also be included to develop a more specific algorithm for diagnosing this disorder. As the literature suggests, TVUS has good diagnostic test parameters and should be the primary choice for investigating isthmocele. Additionally, studies focusing on the prevention of isthmocele should be planned, given the rising cesarean section rates worldwide.
DISCLOSURE STATEMENT
ETHICAL APPROVAL
ACKNOWLEDGMENTS
We thank Helymar da Costa Machado for the statistical analysis of this study.
DATA AVAILABILITY STATEMENT
The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.
REFERENCES
-
1. Morris H. Surgical pathology of the lower uterine segment caesarean section scar: is the scar a source of clinical symptoms? Int J Gynecol Pathol. 1995;14(1):16-20. https://doi.org/10.1097/00004347-199501000-00004
» https://doi.org/10.1097/00004347-199501000-00004 - 2. Tulandi T, Cohen A. Emerging manifestations of cesarean scar defect in reproductive-aged women. J Minim Invasive Gynecol. 2016;23(6):893-902.
-
3. Ofili-Yebovi D, Ben-Nagi J, Sawyer E, Yazbek J, Lee C, Gonzalez J, et al. Deficient lower-segment cesarean section scars: prevalence and risk factors. Ultrasound Obstet Gynecol. 2008;31(1):72-7. https://doi.org/10.1002/uog.5200
» https://doi.org/10.1002/uog.5200 -
4. Bij Vaate AJ, Voet LF, Naji O, Witmer M, Veersema S, Brölmann HA, et al. Prevalence, potential risk factors for development and symptoms related to the presence of uterine niches following cesarean section: systematic review. Ultrasound Obstet Gynecol. 2014;43(4):372-82. https://doi.org/10.1002/uog.13199
» https://doi.org/10.1002/uog.13199 -
5. Wang CB, Chiu WW, Lee CY, Sun YL, Lin YH, Tseng CJ. Cesarean scar defect: correlation between cesarean section number, defect size, clinical symptoms and uterine position. Ultrasound Obstet Gynecol. 2009;34(1):85-9. https://doi.org/10.1002/uog.6405
» https://doi.org/10.1002/uog.6405 -
6. Osser OV, Jokubkiene L, Valentin L. Cesarean section scar defects: agreement between transvaginal sonographic findings with and without saline contrast enhancement. Ultrasound Obstet Gynecol. 2010;35(1):75-83. https://doi.org/10.1002/uog.7496
» https://doi.org/10.1002/uog.7496 - 7. Fabres C, Aviles G, Jara C, Escalona J, Muñoz JF, Mackenna A, et al. The cesarean delivery scar pouch: clinical implications and diagnostic correlation between transvaginal sonography and hysteroscopy. J Ultrasound Med. 2003;22(7):695-700.
- 8. Anter ME, Gad MS, Fahmy EM, Aldeen A, Halaby A. Endovaginal ultrasound versus hysterosalpingography in evaluation of cesarean scar. Egyptian J Hospital Med. 2021;82(4):773-7.
- 9. Jordans IPM, Leeuw R, Stegwee SI, Amso NN, Barri-Soldevila PN, Bosch T, et al. A practical guideline for examining a uterine niche using ultrasonography in non-pregnant women: a modified Delphi method amongst European experts. Ultrasound Obs Gynecol. 2019;53(1):107-15.
-
10. Antila-Långsjö R, Mäenpää JU, Huhtala H, Tomás E, Staff S. Comparison of transvaginal ultrasound and saline contrast sonohysterography in evaluation of cesarean scar defect: a prospective cohort study. Acta Obstet Gynecol Scand. 2018;97(9):1130-6. https://doi.org/10.1111/aogs.13367
» https://doi.org/10.1111/aogs.13367 -
11. Savukyne E, Machtejeviene E, Paskauskas S, Ramoniene G, Nadisauskiene RJ. Transvaginal sonographic evaluation of cesarean section scar niche in pregnancy: a prospective longitudinal study. Medicina (Kaunas). 2021;57(10):1091. https://doi.org/10.3390/medicina57101091
» https://doi.org/10.3390/medicina57101091 - 12. Bettocchi S, Selvaggi L. A vaginoscopic approach to reduce the pain of office hysteroscopy. J Am Assoc Gynecol Laparosc. 1997;4(2):255-8.
-
13. Leonardo-Pinto JP, Brito LGO, Belluomini RTP, Benetti-Pinto CL, Yela DA. Factors associated to the presence of isthmocele diagnosed by pelvic ultrasound, magnetic resonance imaging or diagnostic hysteroscopy: a cross-sectional study. Reprod Sci. 2024;31(12):3908-12. https://doi.org/10.1007/s43032-024-01711-8
» https://doi.org/10.1007/s43032-024-01711-8 -
14. Dominguez JA, Pacheco LA, Moratalla E, Carugno JA, Carrera M, Perez-Milan F, et al. Diagnosis and management of isthmocele (cesarean scar defect): a SWOT analysis. Ultrasound Obstet Gynecol. 2023;62(3):336-44. https://doi.org/10.1002/uog.26171
» https://doi.org/10.1002/uog.26171 -
15. Hosseini RS, Ebrahimi PS, Shokrani A. Diagnostic value of enhanced-contrast ultrasound for cesarean scar pregnancy: a systematic review. Eur J Obstet Gynecol Reprod Biol. 2024;293:119-24. https://doi.org/10.1016/j.ejogrb.2023.12.028
» https://doi.org/10.1016/j.ejogrb.2023.12.028 - 16. Zampieri G, Nitescu B, Pacu I, Neacsu A, Ionescu CA. Isthmocele an iatrogenic pathology: a prospective study in a tertiary unit. Biomed Rep. 2024;22(1):2
-
17. Osser OV, Jokubkiene L, Valentin L. High prevalence of defects in Cesarean section scars at transvaginal ultrasound examination. Ultrasound Obstet Gynecol. 2009;34(1):90-7. https://doi.org/10.1002/uog.6395
» https://doi.org/10.1002/uog.6395 - 18. Allameh Z, Rouholamin S, Rasti S, Adibi A, Foroughi Z, Goharian M et al. A transvaginal ultrasound-based diagnostic calculator for uterus post-cesarean scar defect. BMC Womens Health. 2023;23(1):558.
-
19. Gupta T, Singal K, Gupta N, Kohli S, Kanyal M. Comparative study of USG and MRI in evaluation of isthmocele. J Obstet Gynaecol India. 2021;71(3):292-6. https://doi.org/10.1007/s13224-021-01433-w
» https://doi.org/10.1007/s13224-021-01433-w - 20. Guo S, Wang H, Yu X, Yu Y, Wang C. The diagnostic value of 3.0 T MRI in cesarean scar pregnancy. Am J Transl Res. 2021;13(6):6229-335.
-
21. Kumar I, Verma A, Matah M, Satpathy G. Utility of multiparametric MRI in caesarean section scar characterization and preoperative prediction of scar dehiscence: a prospective study. Acta Radiol. 2017;58(7):890-6. https://doi.org/10.1177/0284185116675659
» https://doi.org/10.1177/0284185116675659 - 22. Raimondo G, Grifone G, Raimondo D, Seracchioli R, Scambia G, Masciullo V. Hysteroscopic treatment of symptomatic cesarean-induced isthmocele: a prospective study. J Minim Invasive Gynecol. 2015;22(2):297-301.
-
23. Gubbini G, Centini G, Nascetti D, Marra E, Moncini I, Bruni L, et al. Surgical hysteroscopic treatment of cesarean-induced isthmocele in restoring fertility: prospective study. J Minim Invasive Gynecol. 2011;18(2):234-7. https://doi.org/10.1016/j.jmig.2010.10.011
» https://doi.org/10.1016/j.jmig.2010.10.011 -
24. Fabres C, Arriagada P, Fernández C, Mackenna A, Zegers F, Fernández E. Surgical treatment and follow-up of women with intermenstrual bleeding due to cesarean section scar defect. J Minim Invasive Gynecol. 2005;12(1):25-8. https://doi.org/10.1016/j.jmig.2004.12.023
» https://doi.org/10.1016/j.jmig.2004.12.023 - 25. Kremer TG, Ghiorzi IB, Dibi RP. Isthmocele: an overview of diagnosis and treatment. Rev Assoc Med Bras. 2019;65(5):714-21.
