Predicting bladder-to-uterine adhesions in patients undergoing hysterectomy using sliding-bladder sign technique with preoperative ultrasound: a diagnostic accuracy study.

We conducted a prospective observational diagnostic test‐accuracy study at two sites (McMaster University Medical Centre (MUMC) and West Lincoln Memorial Hospital (WLMH)) within a single hospital corporation (Hamilton Health Sciences, Hamilton, ON, Canada). The study was reviewed and approved by the Hamilton Integrated Research Ethics Board (HiREB; ref.: 13001) and was conducted and reported in accordance with the Standards for Reporting of Diagnostic Accuracy Studies (STARD) guidelines 15 . Consecutive patients undergoing a planned laparoscopic hysterectomy at MUMC or WLMH between 1 October 2021 and 31 December 2024 were included in this study. All patients were ≥ 18 years of age, with a benign gynecological condition as the indication for scheduled hysterectomy. All patients underwent preoperative TVS examination to assess the sliding‐bladder sign as an indicator of the presence or absence of BTU adhesions. Patients at MUMC underwent TVS examination immediately prior to surgery, whereas patients at WLMH underwent TVS examination within 6 months before surgery at the Specialized Ultrasound Gynecology & Obstetrics ultrasound clinic (Hamilton, ON, Canada). Patients were excluded if they could not undergo TVS examination, had a current reproductive tract infection (e.g. acute pelvic inflammatory disease, tubo‐ovarian abscess) or were being evaluated for malignancy, as these conditions could significantly alter the pelvic anatomy, tissue planes and organ mobility. All patients provided written informed consent. Preoperative TVS examination (the index test) was performed by an expert gynecological sonologist (M.L.). Intraoperative surgical classification of BTU adhesions (the reference standard) was performed by one of four independent, attending‐level obstetrician gynecologists (OBGYNs) (I.M., A.S., M.M., A.A.M.). Both procedures were conducted independently, and each specialist completed a pre‐specified case report form and reported their findings separately to a third party. The reference standard was intentionally recorded as a binary outcome to address a focused diagnostic accuracy question (presence vs absence of clinically relevant BTU adhesions), rather than to quantify adhesion severity or surgical complexity. The index test was performed by systematically evaluating the anterior compartment for the sliding‐bladder sign, according to the International Deep Endometriosis Analysis (IDEA) consensus 16 , with a slightly filled bladder and the transvaginal probe placed in the anterior fornix. The sliding‐bladder sign was assessed by a combination of pressure applied on the uterus using the probe and pressure of the free hand of the sonologist applied over the suprapubic region, palpating the uterus between the probe and the hand. A variety of ultrasound systems were used for assessment: Voluson E10, Voluson S8, Voluson Swift (GE Healthcare, Zipf, Austria), Samsung V8 and Samsung HS70 (Samsung Medison Co. Ltd., Seoul, South Korea). The sonologist who assessed the sliding‐bladder sign on TVS was not blinded to the clinical and surgical history of the patient. To obtain the reference standard, BTU adhesions were classified by the operating surgeon at the start of anterior‐compartment dissection, with the option to confirm the classification during development of the vesicouterine plane to avoid misclassification based on the superficial peritoneal appearance alone. The study was designed as such because surgical visualization alone may underestimate the density of BTU adhesions if the overlying peritoneum or uterine serosa interface was relatively clear/regular. All procedures were performed by fully trained generalist OBGYNs practicing within the same hospital corporation (Hamilton Health Sciences), providing a consistent generalist attending‐level surgical context, and all cases utilized the same camera and laparoscopic set up. The OBGYN surgeon who classified the state of the BTU adhesions during laparoscopy was blinded to the results of the preoperative TVS findings regarding BTU adhesions, but was not blinded to the clinical and surgical history of the patient or other pertinent ultrasound features. The index test was reported in alignment with the posterior compartment sliding‐sign nomenclature, as per the IDEA consensus 16 . The sliding‐bladder sign was recorded as positive when bladder sliding was observed, indicating the absence of adhesions, or as negative when bladder sliding was absent, indicating the presence of adhesions. The reference standard was reported as no adhesions or adhesions present. BTU adhesions were defined as dense, fibrotic adhesions between the bladder and the uterus that required sharp or blunt dissection to open the vesicouterine plane (i.e. adhesions judged intraoperatively to require deliberate adhesiolysis to establish the vesicouterine plane). Filmy or easily lysed adhesions that did not obstruct dissection were not classified as BTU adhesions. To standardize data capture, a structured case‐report form was used for all participants, including pre‐specified fields for the index test (sliding‐bladder sign recorded as positive or negative), the reference standard (BTU adhesions recorded as present or absent) and relevant clinical variables. Results of the index test and of the reference standard were recorded independently and submitted separately to a third party for entry, to maintain separation of assessments and prevent observer bias. Demographic characteristics, such as age, body mass index, previous pelvic surgery (including excision of endometriosis, myomectomy and Cesarean section) and medical and obstetric history were recorded at the time of surgery. The presence of fibroids or adenomyosis was assessed during TVS examination, in accordance with the International Federation of Gynaecology and Obstetrics (FIGO) 17 and Morphological Uterus Sonographic Assessment (MUSA) 18 classifications. Endometriosis was evaluated and described sonographically in accordance with the IDEA consensus 16 , and was surgically and histologically validated. The data were analyzed using SPSS Statistics version 28 (IBM Corp., Armonk, NY, USA). The diagnostic accuracy of the sliding‐bladder sign for the visualization of BTU adhesions was evaluated using sensitivity, specificity, accuracy, positive predicitive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+) and negative likelihood ratio (LR−), with 95% CI. A P ‐value of < 0.05 was considered statistically significant. A negative sliding‐bladder sign was considered a positive result of the diagnostic test. Logistic regression analysis was performed to predict the presence of BTU adhesions on TVS based on the patient's gynecological surgical history or uterine pathology seen on TVS. The prevalence of BTU adhesions is estimated to be approximately 25% among patients with a history of pelvic surgery 8 , 19 . Based on the current literature, we anticipated a sensitivity of 75% and a specificity of 95% for the sliding‐bladder sign for predicting BTU adhesions 8 . Sample‐size calculations were performed using Buderer's formula for diagnostic test evaluation, which estimates the required number of participants based on the expected sensitivity and specificity and desired precision and confidence level, without requiring a standard deviation given that the outcome is binary (presence or absence of BTU adhesions) 20 . Using a confidence level of 80% and a precision of 10%, the required sample size was 124 patients.

Out of 150 eligible participants who consented to undergo a laparoscopic hysterectomy between October 2021 and December 2024, 128 individuals proceeded with hysterectomy within the designated timeframe and were subsequently included in the analysis. The mean age of the participants was 40.9 ± 7.0 years. A total of 68 (53.1%) participants had a history of pelvic surgery: 37 (28.9%) had undergone previous Cesarean section, 39 (30.5%) had undergone previous surgery for endometriosis and 10 (7.8%) had undergone previous myomectomy. In terms of gynecological pathologies, 44 (34.4%) patients had adenomyosis and 33 (25.8%) had fibroids observed on TVS. Among participants, 79 (61.7%) had sonographically confirmed endometriosis, with 76 (59.4%) having superficial/peritoneal lesions at the time of surgery. Within the anterior compartment, 14 (10.9%) patients had superficial disease along the bladder, vesicouterine peritoneum, broad ligament and abdominal wall. No participant exhibited deep endometriosis of the bladder; all deep endometriotic disease was identified in the posterior compartment. A summary of participant characteristics is given in Table  1 . Demographic and transvaginal ultrasound (TVS) characteristics of 128 patients with indication for hysterectomy who underwent TVS assessment of bladder‐to‐uterine adhesions Data are presented as mean ± SD or n (%). Fifteen patients had multiple previous pelvic surgeries. Fifty patients had multiple gynecological pathologies on TVS. Forty patients had multiple types of endometriosis on TVS; type of endometriosis was surgically confirmed and all cases of deep endometriosis were identified in the posterior compartment. BMI, body mass index. Among the 128 participants, the prevalence of BTU adhesions detected on TVS was 21.9% ( n  = 28) and the prevalence of BTU adhesions confirmed during hysterectomy was 28.1% ( n  = 36). The sliding‐bladder sign on preoperative TVS demonstrated an accuracy of 89.1% (95% CI, 82.3–93.9%), with a sensitivity of 69.4% (95% CI, 51.9–83.6%) and a specificity of 96.7% (95% CI, 90.1–99.3%). The PPV was 89.3% (95% CI, 72.8–96.3%) while the NPV was 89.0% (95% CI, 83.2–93.0%). The LR+ was 21.3 (95% CI, 6.8–66.2) and the LR– was 0.3 (95% CI, 0.2–0.5) (Table  2 ). Univariable logistic regression analysis indicated that several factors were associated with the presence of BTU adhesions on TVS, including previous Cesarean section ( P  < 0.001), parity ( P  = 0.002) and gravidity ( P  = 0.011) (Table  3 ). Significantly higher sensitivity (91.7%) and NPV (98.7%) were seen when participants with previous Cesarean section were excluded (Table  2 ). Predictive value of preoperative transvaginal ultrasound (TVS) sliding‐bladder sign for bladder‐to‐uterine adhesions that were subsequently confirmed during hysterectomy Data in parentheses are 95% CI. CS, Cesarean section; LR+, positive likelihood ratio; LR−, negative likelihood ratio; NPV, negative predictive value; PPV, positive predictive value. Association between patient characteristics and the presence of bladder‐to‐uterine adhesions on transvaginal ultrasound, estimated by univariable logistic regression analysis BMI, body mass index; CS, Cesarean section.

This study evaluated the diagnostic performance of the sliding‐bladder sign on preoperative TVS for predicting BTU adhesions in patients undergoing hysterectomy. This technique achieved 89.1% accuracy, 96.7% specificity and 69.4% sensitivity in identifying BTU adhesions that were subsequently confirmed during hysterectomy. The high specificity supports the use of TVS as an effective tool for detecting BTU adhesions, indicated when bladder sliding is absent, while the lower sensitivity suggests that it is less reliable for ruling out BTU adhesions. Nonetheless, the PPV of 89.3% supports its clinical utility. Sensitivity and NPV improved significantly when patients with a history of Cesarean section were excluded. This may be attributed to the Bayesian diagnostic principle 21 , whereby a history of Cesarean section, being the most predictive risk factor for BTU adhesions, increases the pretest probability and therefore influences the observed diagnostic accuracy. The posterior sliding sign, as assessed using TVS, is a well‐established predictor of pelvic adhesions, particularly in cases of deep endometriosis 7 , 8 , 22 . Our study extends its application to BTU adhesions in the anterior compartment. We found the sliding‐bladder sign to be a highly specific indicator (96.7%) of BTU adhesions, consistent with the studies of Min et al . 10 and Ayachi et al . 8 , which reported similarly high specificity in patients who had undergone prior pelvic surgery. Unlike the study of Min et al . 10 , in which the cohort mainly included women with a history of Cesarean section, our study encompassed a broader population, enhancing the generalizability of the findings. While consistent with prior research, the lower sensitivity (69.4%) of the sliding sign in identifying BTU adhesions suggests that some adhesions may be missed using this technique alone. Min et al . 10 reported higher diagnostic performance for vesicouterine adhesions classified as severe in their cohort. As adhesion severity was not quantified in the present study, our findings should be interpreted as reflecting the diagnostic accuracy of TVS for the presence vs absence of clinically relevant BTU adhesions, rather than for stratification of severity. Preoperative identification of BTU adhesions may be clinically useful by enabling anticipation of altered vesicouterine anatomy and the potential need for adhesiolysis during hysterectomy. Evidence from studies evaluating the sliding sign in the posterior compartment suggests that the absence of organ mobility may be associated with altered operative anatomy in certain clinical contexts 23 , 24 . However, the present study was designed as a diagnostic‐accuracy study focused on detecting the presence of BTU adhesions, and did not assess adhesion severity, operative complexity, intraoperative complications or postoperative outcomes. Accordingly, the findings should be interpreted as providing anatomical information to support preoperative awareness of BTU adhesions, rather than as predictors of surgical difficulty or clinical outcomes. Incorporating assessment of the sliding‐bladder sign into routine preoperative TVS examinations may support identification of anterior‐compartment adhesions and anatomical anticipation prior to hysterectomy. Knowledge of likely BTU adhesions may assist with perioperative preparedness, including planning the operative sequence for vesicouterine dissection and facilitating patient counseling regarding the possibility of adhesiolysis 10 . While BTU adhesions have been identified as a perceived contributor to hysterectomy difficulty in consensus work 14 , quantifying operative complexity, operative time, complications or downstream surgical outcomes was outside the scope of this diagnostic‐accuracy study. Further studies should focus on improving the sensitivity of TVS for anterior‐compartment adhesions without compromising specificity. Large multicenter cohorts are needed to validate these results and assess the impact of factors such as previous pelvic surgery, endometriosis, adenomyosis or pelvic inflammatory disease. Evaluating interobserver variability is crucial to confirm reproducibility across operators. Min et al . 10 previously addressed this issue by demonstrating that the sliding‐bladder sign can be reliably evaluated even by relatively inexperienced clinicians, supporting its potential as a reproducible marker of adhesions across diverse settings. Longitudinal studies should also be conducted to determine whether the preoperative detection of adhesions improves surgical outcomes, recovery and reoperation rates. Important endpoints should focus on the severity of the adhesions and correlation with the complexity of surgical adhesiolysis, as well as the association with clinically relevant pain and quality of life. Ultimately, future research should aim to integrate the consensus‐derived core feature set established by Leonardi et al . 14 with predictive preoperative imaging markers, such as the sliding‐bladder sign, into a standardized preoperative hysterectomy complexity index. This integrated approach would not only enhance the standardization of surgical data collection and nomenclature across studies, but also allow for robust analysis of how specific patient and surgical features correlate with key intraoperative events and postoperative outcomes. The prospective study design and blinding between TVS and surgical findings strengthen the validity of the results and address a gap in the current literature. However, limitations include the single‐center design with a single TVS operator, potentially limiting the reproducibility and generalizability of the results. Multicenter studies would be valuable in confirming the diagnostic accuracy of the sliding‐bladder sign on TVS across diverse healthcare environments. Another limitation is the relatively small sample size. While adequate for the primary analysis, the study may have been underpowered for subgroup analyses. The diagnostic performance of TVS in patients with specific gynecological conditions, such as adenomyosis, endometriosis or fibroids, was not thoroughly explored owing to the limited sample size. Cesarean niches should be evaluated using ultrasound, as they have been reported to be associated with BTU adhesions 25 . Lastly, BTU adhesions were recorded as a binary outcome using the reference standard to address a focused diagnostic‐accuracy question (presence vs absence of BTU adhesions requiring meaningful adhesiolysis). This design does not permit severity grading or direct inference regarding surgical complexity; future work should incorporate standardized severity metrics and operative endpoints to evaluate clinical impact. The sliding‐bladder sign assessed using TVS is a non‐invasive tool for predicting the presence of BTU adhesions that is reliable when bladder sliding is absent. Incorporating this assessment into preoperative ultrasound may support the identification of anterior‐compartment adhesions and anatomical anticipation prior to hysterectomy and can be considered as part of a comprehensive preoperative evaluation.

Bladder‐to‐uterine (BTU) adhesions can alter vesicouterine anatomy, and are commonly encountered during hysterectomy following prior pelvic surgery 1 , 2 , 3 . Cesarean section, the most common surgical procedure in women, frequently leads to substantial adhesion formation soon after surgery 4 . Other causes include endometriosis, pelvic inflammatory disease or other pelvic inflammatory processes (e.g. appendicitis, ovarian torsion), prior bladder surgery and other pelvic surgical procedures (e.g. tubal surgery, myomectomy, (abdominal) cerclage). Hudelist et al . 5 and Reid et al . 6 first described an ultrasound technique to assess organ mobility, termed the ‘sliding sign’. The use of transvaginal ultrasound (TVS) to determine the presence or absence of rectouterine pouch obliteration due to deep endometriosis is already well documented 7 . However, literature regarding the prediction of BTU adhesions using ultrasonography is scarce. Suggested sonographic indicators include a negative sliding‐bladder sign (lack of motion between the abdominal and uterine walls) and uterine peaking (tissue strands with myometrial density between the uterus and the bladder reflection) 4 , 8 , 9 . Min et al . 10 recently evaluated the diagnostic performance of the sliding‐bladder sign on TVS for identifying BTU adhesions and reported good predictive value for the presence of severe BTU adhesions. However, this study was performed in a center with a high volume of cases with a uterine Cesarean niche, which limited its generalizability. Preoperative uncertainty regarding anterior‐compartment adhesions can limit anatomical anticipation and counseling about potential intraoperative findings 1 , 3 . While BTU adhesions have been identified through expert consensus as features perceived to contribute to the difficulty of performing hysterectomy, their impact on surgical course may vary depending on their extent, location and the experience of the surgeon 1 , 11 , 12 , 13 . A Delphi consensus study of Leonardi et al . 14 highlighted anterior‐compartment adhesions as a core feature relevant to hysterectomy planning, underscoring the potential value of identifying them preoperatively. Accordingly, the reliable preoperative detection of BTU adhesions may support informed surgical preparedness and patient counseling by improving awareness of altered vesicouterine anatomy prior to hysterectomy 10 . The aim of this study was to confirm the diagnostic accuracy of preoperative TVS in assessing the sliding‐bladder sign for the presence of BTU adhesions, and to improve the generalizability of the findings in comparison with previous studies.