Advancing urological practice: the value of combined antegrade and retrograde ureterography for the management of complex ureteral strictures Authors: Kangxiang Xu, Manshun Dong, Chaoqi Liang, Ruoyu Li, Qiuxuan Yu, Jin Liu, Hang Yin, Kuibiao Wang, Xincheng Gao, Yuancheng Zhou, Shuaishuai Chai, Xingyuan Xiao, Bing Li Published: 2026 ABSTRACT AbstractBackgroundComplex proximal and mid-ureteral strictures, caused by calculi, inflammation, fibrosis, or iatrogenic injury, can result in hydronephrosis and renal function loss. Traditional ureterography, when used alone, has limitations in accurately assessing stricture length and detecting multiple segments. We evaluated the diagnostic value of the combined antegrade and retrograde ureterography (CARU) technique in complex strictures, focusing on stricture length, location, and severity.MethodsThis retrospective study collected data from 110 patients diagnosed with proximal and mid-ureteral strictures at our center between January 2023 and November 2025. All patients were evaluated with CARU 4 weeks prior to surgery and subsequently underwent lingual mucosal graft (LMG) ureteroplasty. A comparative analysis was performed to assess the correlation between the preoperative CARU assessment of stricture data and the intraoperative findings. Subgroup analyses were conducted according to stricture etiology and severity.ResultsCARU demonstrated excellent agreement with intraoperative findings for stricture location (κ=0.93, P<0.001) and substantial agreement for stricture severity (κ=0.69, P<0.001). CARU tended to underestimate the true length of ureteral strictures and, consequently, the graft length required for reconstruction. The stricture length measured by CARU (2.89±2.29 cm) was, on average, 0.74 cm shorter than the intraoperative stricture length (3.63±1.88 cm; P<0.001) and 0.93 cm shorter than the intraoperative LMG length (3.82±1.80 cm; P<0.001). In subgroup analyses, CARU-measured stricture length was broadly comparable with both intraoperative stricture length and graft length in congenital or idiopathic cases, but underestimated these lengths in patients with acquired etiologies and severe strictures. Nevertheless, ICC analysis demonstrated good agreement between CARU-based and intraoperative measurements.ConclusionsCARU is a valuable tool for evaluating complex proximal and mid-ureteral strictures, with high accuracy in assessing stricture location and severity. It provides valuable reference for estimating stricture and graft lengths. It could serve as a useful preoperative imaging modality for patients with proximal and mid-ureteral strictures.Keywords: Urography, ureteral strictures, X-ray, diagnostic imaging, radiography Highlight box Key findings • Combined antegrade and retrograde ureterography (CARU) showed excellent agreement with intraoperative findings for stricture location and substantial agreement for stricture severity in complex proximal and mid-ureteral strictures. It also provided a useful preoperative estimate of stricture length, although it tended to underestimate both true stricture length and graft length required for reconstruction. What is known and what is new? • Traditional antegrade or retrograde ureterography alone can identify ureteral strictures, but their ability to define stricture length, severity, and multisegmental disease is limited. • This study shows that CARU offers more comprehensive preoperative assessment in this setting and may better support surgical planning. What is the implication, and what should change now? • CARU can be considered a useful adjunct for preoperative evaluation of complex proximal and mid-ureteral strictures, especially when conventional imaging is insufficient. • Surgeons should allow for a margin of underestimation when planning resection extent and graft harvest. Introduction Complex proximal and mid-ureteral strictures, often caused by calculi, inflammation, fibrosis, or iatrogenic injury, can lead to hydronephrosis, renal function loss, and even renal failure, posing a significant challenge in modern urological practice (1). Recent advancements in imaging and surgical techniques have greatly improved the surgical management of complex strictures (2). Ureteroplasty using onlay grafts or flaps, such as lingual mucosa, colorectal mucosa, buccal mucosa, bladder mucosa, and penile skin, has emerged as an innovative procedure for the management of proximal and midureteral strictures, becoming a research focus in upper urinary tract reconstruction (3-7). Recent advances in robotic surgery have enabled precise and minimally invasive reconstructive procedures, and the robotic platform is increasingly used by urologic reconstructive surgeons (8-11). Treatment selection for ureteral strictures is determined by several key parameters, particularly the etiology, length, location, and severity of the lesion (12). Comprehensive preoperative assessment of stricture complexity is crucial for guiding surgical decision-making (13). However, traditional methods remain limited, and intraoperative assessment of tissue tension and accurate stricture length is still required (14,15). Traditional ureterography, typically performed via retrograde or antegrade approaches, can identify the approximate location of ureteral strictures but is limited in accurately assessing stricture length, severity, and detecting multiple stricture segments when used alone (16). In recent years, our team has innovatively applied combined antegrade and retrograde ureterography (CARU) for diagnosing ureteral strictures, rather than relying solely on either the antegrade or retrograde approach (3,17,18). This innovative imaging technique has not been explored for its diagnostic value in complex proximal and mid-ureteral strictures, particularly regarding length, location, and severity. The aim of this study is to evaluate the diagnostic value of CARU in patients with complex proximal and mid-ureteral strictures, and to further explore its clinical applicability. We present this article in accordance with the STARD reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0129/rc). Methods The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Ethics Committee of Zhongnan Hospital of Wuhan University (No. 2025292K), and individual consent for this retrospective analysis was waived. We retrospectively included 110 patients with proximal or mid-ureteral strictures treated between January 2023, and November 2025. Demographic characteristics, surgical details, and perioperative data were collected and analysed. The aetiology and location of strictures were classified according to the Ureteral Stricture Classification System proposed by Cho and colleagues (19). All patients underwent percutaneous nephrostomy before ureteral reconstruction and were evaluated with CARU approximately 4 weeks before surgery. Stricture length, location, and severity were assessed with CARU. Obliterative or atretic lesions were classified as severe strictures, whereas non-obliterative lesions were classified as stenotic strictures. CARU No patients had fever or urinary tract infection within 3 days before CARU. After cystoscopic identification of the ureteric orifice, a 5 Fr ureteric catheter was inserted retrogradely. When direct insertion was difficult, placement was assisted with a guidewire, and the catheter was advanced 20–30 cm. After catheter placement, the cystoscope was removed and a urethral catheter was inserted for continuous drainage and fixation of the ureteric catheter. All patients underwent CARU using the same fluoroscopic system (Luminos Fusion; Siemens Healthcare GmbH, Erlangen, Germany). The iodinated contrast agent, iopamidol, was diluted with saline to approximately 125 mg iodine/mL and administered under sterile conditions by low-pressure, slow bolus injection, with continuous monitoring of patient symptoms. Patients were observed for 24 h after CARU, and procedure-related adverse events, including fever, urinary tract infection, contrast extravasation, and catheter-related trauma, were recorded from the medical records. Antegrade ureterography The patient was placed supine, and an initial radiograph was obtained to confirm the positions of the nephrostomy tube and ureteric catheter (Figure 1A). After clamping the nephrostomy tube, 5–10 mL of contrast was injected under fluoroscopic guidance to confirm that the catheter tip was within the renal pelvis and to assess pelvicalyceal morphology. Contrast was then infused slowly while the patient’s position was changed from supine to standing and the fluoroscopic angle was adjusted to visualise antegrade passage through the ureter. The flow of contrast was assessed for stenosis, obliteration, extravasation, reflux, and dilatation or peristalsis of the renal pelvis and ureter. The location and severity of any abnormalities were recorded, followed by retrograde ureterography (Figure 1B). CARU imaging technique and methodology. (A) Initial assessment: confirm the position of the nephrostomy tube and ureteral catheter. (B) Antegrade ureterography: inject contrast agent through the nephrostomy tube, adjust the patient’s position from supine to standing, change the fluoroscopic projection angle (blue double-headed arrow), and record the location and severity of any abnormalities (red arrow). (C) Retrograde ureterography: inject contrast agent through the ureteral catheter in the supine position and combine this with antegrade ureterography to assess the length and severity of any ureteral stricture (yellow arrow). (D) Delayed assessment: remove the ureteral catheter and assess antegrade contrast flow through the stricture area in the standing position (purple arrow). Green arrows indicate procedural flow. CARU, combined antegrade and retrograde ureterography. Retrograde ureterography With the nephrostomy tube kept clamped, the ureteric catheter was connected to a syringe. Under fluoroscopic guidance, 5–10 mL of contrast was injected slowly to assess ureteric opacification. The patient’s position or fluoroscopic angle was adjusted to visualise retrograde filling of the ureter, renal pelvis, and calyces. The catheter was then gradually withdrawn to 2–3 cm distal to the suspected stricture, and contrast injection was continued to assess passage across the narrowed segment (Figure 1C). The catheter was subsequently withdrawn slowly with continued contrast injection to visualise the full length of the ureter. The patient was then repositioned upright to assess antegrade passage of contrast through the stricture (Figure 1D). A severe stricture was suspected when there was no prompt downward drainage and the ureteric catheter could not be advanced across the lesion. If contrast cleared promptly, the stricture was classified as stenotic. In borderline cases, strictures with a visible ureteric lumen were classified as stenotic; otherwise, they were classified as severe. After the procedure, the catheter was removed and the nephrostomy valve was reopened. Stricture length was then measured against a reference scale. Intraoperative evaluation of ureteral strictures Intraoperative assessment included stricture location, length, severity, the presence of multisegmental strictures, and lingual mucosal graft (LMG) length. All procedures were performed by the same surgical team using the da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA), led by two reconstructive urologists with more than 10 years of experience. For intraoperative assessment of stricture severity, 10 mL of indocyanine green solution (25 mg indocyanine green diluted in 10 mL normal saline) was injected into the ureter through the nephrostomy tube. Under near-infrared fluorescence imaging, healthy ureter fluoresced green, stenotic segments showed reduced fluorescence, and severely strictured segments showed no fluorescence (Figure 2A,2B). Ureterotomy was extended until a 10 Fr catheter could be passed without resistance into both the spatulated proximal and distal ureteric ends; this criterion was also used to assess the presence of multisegmental strictures (Figure 2C,2D). All patients underwent ureteroplasty with autologous LMG reconstruction (3,17,18). Stricture length was assessed by the operating surgeon on direct visual inspection, with the ureter opened proximally and distally into healthy tissue and fluorescence imaging used as an adjunct. Stricture length and the required LMG length were measured with a graduated catheter or ruler (Figure 2E,2F). Intraoperative evaluation of ureteral strictures. (A,B) Severity assessment: the healthy ureter fluoresced green, the stenotic segment showed minimal fluorescence, and the severe stenotic segment ureter displayed no fluorescence. (C,D) Assessment of multiple strictures: a 10 Fr catheter was inserted into the spatulated proximal and distal ureteral stumps to assess the presence of stricture based on resistance. (E,F) Length assessment: the length of the stricture and the LMG were measured using a catheter with a scale or a graduated ruler. LMG, lingual mucosal graft. Statistical analysis The index test was CARU and the reference standard was intraoperative findings. Categorical outcomes included assessment of stricture location and severity, whereas continuous outcomes compared stricture length measured preoperatively by CARU with that measured intraoperatively to assess agreement. CARU images were interpreted independently by a urologist and a radiologist, each with more than 10 years of experience, who were masked to clinical information and intraoperative findings at the time of image review. Concordant interpretations were accepted as final. In cases of disagreement, a senior radiologist participated in a joint review to reach consensus. Disagreements in intraoperative assessment were resolved through discussion among the surgeons, review of operative videos, and consultation with external experts. If consensus could not be reached, the lead surgeon made the final decision. A stricture length of 2 cm or greater was defined as a long-segment ureteric stricture (2,19). Agreement between CARU and intraoperative findings was assessed with Cohen’s κ. Three lengths were recorded: d1, stricture length measured by CARU; d2, intraoperative stricture length; and d3, intraoperative LMG length. Differences were defined as Δd1 = d2 − d1, Δd2 = d3 − d1, and Δd3 = d3 − d2. For paired data, normally distributed differences are presented as mean ± standard deviation and were analysed with the paired t-test; non-normally distributed differences are presented as median [interquartile range (IQR)] and were analysed with the Wilcoxon signed-rank test. Normality was assessed with the Kolmogorov-Smirnov and Shapiro-Wilk tests. Intraclass correlation coefficients (ICCs) with 95% confidence intervals (CIs) were calculated with a two-way mixed-effects model for absolute agreement to assess concordance between paired measurements. Subgroup analyses by aetiology and stricture severity were prespecified exploratory analyses based on clinical relevance. Statistical analyses were done with SPSS version 31.0, and P values less than 0.05 were considered significant. Results Baseline characteristics of the patients This study included 110 patients with ureteric stricture, and baseline characteristics are shown in Table 1. Of these patients, 77 (70.0%) were male and 33 (30.0%) were female. Age ranged from 8 to 71 years, with a median of 46.5 years. Of 110 strictures, 66 (60.0%) were attributed to internal trauma related to endoscopic procedures (ureteroscopy, lithotripsy, or percutaneous nephrolithotomy), 15 (13.6%) to external trauma (ligation, transection, or thermal injury), 18 (16.4%) to previous ureteric reconstruction, and 11 (10.0%) were idiopathic or congenital; 99 (90.0%) were acquired. No Clavien-Dindo grade 2 or higher complications occurred. Five (4.5%) patients developed transient fever within 24 h after CARU, which resolved with physical cooling alone (Table 1). Table 1Patients’ characteristics, CARU, and intraoperative findingsVariablesResultsAge (years), median (range)46.5 (8 to 71)Gender, n (%) Male77 (70.0) Female33 (30.0)BMI (kg/m2), median (range)23.84 (14.06 to 32.74)Surgical side, n (%) Left56 (50.9) Right54 (49.1)Etiology, n (%) Congenital3 (2.7) Idiopathic8 (7.3) Internal trauma (endoscopic procedure: ureteroscopy, lithotripsy, or PCNL)66 (60.0) External trauma (ligation, transection, or thermal injury) 15 (13.6) Prior ureteral reconstruction18 (16.4)The stricture locations predicted by CARU, n (%) Proximal94 (85.5) Middle9 (8.2) Proximal and middle7 (6.4)The stricture locations identified intraoperatively, n (%) Proximal93 (84.5) Middle8 (7.3) Proximal and middle9 (8.2)The CARU predicted the severity of the stricture, n (%) Obliteration or atresia62 (56.4) Stenotic48 (43.6)The severity of the stricture was identified intraoperatively, n (%) Obliteration or atresia53 (48.2) Stenotic57 (51.8)Lengths (cm), mean ± SD (range; median, IQR) d12.89±2.39 (0.4 to 12.4; 2.40, 2.55) d23.63±1.88 (0.5 to 11.5; 3.15, 2.40) d33.82±1.88 (1.2 to 12.0; 3.30, 2.50) Δd10.73±1.20 (−2.80 to 4.60; 0.75, 1.50) Δd20.93±1.33 (−2.80 to 4.60; 0.90, 1.43) Δd30.19±0.68 (−3.00 to 2.00; 0, 0.50)Complications, n (%) No105 (95.5) I5 (4.5) II–V0Δd1, d2 − d1; Δd2, d3 − d1; Δd3, d3 − d2. BMI, body mass index; CARU, combined antegrade and retrograde ureterography; d1, stricture length measured by CARU; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; IQR, interquartile range; PCNL, percutaneous nephrolithotomy; SD, standard deviation. Patients’ characteristics, CARU, and intraoperative findings Δd1, d2 − d1; Δd2, d3 − d1; Δd3, d3 − d2. BMI, body mass index; CARU, combined antegrade and retrograde ureterography; d1, stricture length measured by CARU; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; IQR, interquartile range; PCNL, percutaneous nephrolithotomy; SD, standard deviation. Location and severity CARU classified 94 (85.5%) strictures as proximal, 9 (8.2%) as mid-ureteric, and 7 (6.4%) as involving both the proximal and mid ureter, whereas intraoperative findings identified 93 (84.5%), 8 (8.3%), and 9 (8.2%) strictures in these categories, respectively. Seven (6.4%) patients had multisegmental strictures. CARU showed excellent agreement with intraoperative assessment for stricture location (κ=0.93; P<0.001). For proximal strictures, CARU had a sensitivity of 100% (95% CI: 96.1–100.0%) and a specificity of 94.1% (95% CI: 73.0–99.0%). For mid-ureteric strictures, sensitivity was 100% (95% CI: 67.6–100.0%) and specificity was 99.0% (95% CI: 94.7–99.8%). For strictures involving both the proximal and mid ureter, sensitivity was 77.8% (95% CI: 45.3–93.7%) and specificity was 100% (95% CI: 96.4–100.0%; Tables 1,2). Table 2Diagnostic performance of CARU for identifying stricture location compared with intraoperative findingsCARUIntraoperative findingsProximalMiddleProximal and middleTotalProximal930194Middle0819Proximal and middle0077Total9389110CARU demonstrated excellent agreement with intraoperative findings for stricture location (Cohen’s κ=0.93). For identifying proximal strictures, sensitivity was 100% (95% CI: 96.1–100%) and specificity was 94.1% (95% CI: 73.0–99.0%). For middle strictures, sensitivity was 100% (95% CI: 67.6–100%) and specificity was 99.0% (95% CI: 94.7–99.8%). For combined proximal and middle strictures, sensitivity was 77.8% (95% CI: 45.3–93.7%) and specificity was 100% (95% CI: 96.4–100%). CARU, combined antegrade and retrograde ureterography; CI, confidence interval. Diagnostic performance of CARU for identifying stricture location compared with intraoperative findings CARU demonstrated excellent agreement with intraoperative findings for stricture location (Cohen’s κ=0.93). For identifying proximal strictures, sensitivity was 100% (95% CI: 96.1–100%) and specificity was 94.1% (95% CI: 73.0–99.0%). For middle strictures, sensitivity was 100% (95% CI: 67.6–100%) and specificity was 99.0% (95% CI: 94.7–99.8%). For combined proximal and middle strictures, sensitivity was 77.8% (95% CI: 45.3–93.7%) and specificity was 100% (95% CI: 96.4–100%). CARU, combined antegrade and retrograde ureterography; CI, confidence interval. CARU classified 62 (56.4%) strictures as severe, whereas intraoperative findings confirmed 53 (48.2%) as severe. CARU showed substantial agreement with intraoperative findings for stricture severity (κ=0.69; P<0.001). For severe strictures, CARU had a sensitivity of 92.5% (95% CI: 82.1–97.0%) and a specificity of 77.2% (95% CI: 64.8–86.2%). The positive predictive value was 79.0% (95% CI: 67.4–87.3%) and the negative predictive value was 91.7% (80.4–96.7%; Tables 1,3). Table 3Diagnostic performance of CARU for identifying stricture severity compared with intraoperative findingsCARUIntraoperative findingsObliteration or atresiaStenoticObliteration or atresia4913Stenotic444Total5357CARU demonstrated substantial agreement with intraoperative findings (Cohen’s κ=0.69); sensitivity 92.5% (95% CI: 82.1–97.0%); specificity 77.2% (95% CI: 64.8–86.2%); positive predictive value 79.0% (95% CI: 67.4–87.3%); negative predictive value 91.7% (95% CI: 80.4–96.7%). CARU, combined antegrade and retrograde ureterography; CI, confidence interval. Diagnostic performance of CARU for identifying stricture severity compared with intraoperative findings CARU demonstrated substantial agreement with intraoperative findings (Cohen’s κ=0.69); sensitivity 92.5% (95% CI: 82.1–97.0%); specificity 77.2% (95% CI: 64.8–86.2%); positive predictive value 79.0% (95% CI: 67.4–87.3%); negative predictive value 91.7% (95% CI: 80.4–96.7%). CARU, combined antegrade and retrograde ureterography; CI, confidence interval. Length Median stricture length measured by CARU was 2.40 cm (range, 0.40–12.40 cm). Median intraoperative stricture length and median intraoperative LMG length were both 3.15 cm (ranges 0.50–11.50 and 1.20–12.00 cm, respectively; Table 1). Pairwise comparisons among CARU-measured stricture length, intraoperative stricture length, and intraoperative LMG length showed significant differences. Mean d1 (2.89±2.29 cm) was 0.74 cm shorter than mean d2 (3.63±1.88 cm; t=6.438, P<0.001), and 0.93 cm shorter than mean d3 (3.82±1.80 cm; t=7.3, P<0.001). For d2 versus d3, the medians were both 3.15 cm, but the IQRs differed (2.30–4.73 vs. 2.50–5.00; Z=−3.285, P=0.001). Despite these differences, agreement between measurements was high, with ICCs of 0.84 (95% CI: 0.77–0.89) for d1 versus d2, 0.80 (95% CI: 0.72–0.86) for d1 versus d3, and 0.93 (95% CI: 0.91–0.96) for d2 versus d3 (Table 4). Table 4Comparison of CARU-measured stricture length, intraoperative stricture length, and lingual mucosal graft length (cm)Comparisond1d2d3Test statisticP valueICC (95% CI)d1 vs. d22.89±2.293.63±1.88NAt=6.438<0.0010.84 (0.77–0.89)d1 vs. d32.89±2.29NA3.82±1.80t=7.3<0.0010.80 (0.72–0.86)d2 vs. d3NA3.15 (2.30–4.73)3.15 (2.50–5.00)Z=−3.2850.0010.93 (0.91–0.96)Normally distributed paired data are presented as mean ± standard deviation and compared using paired t-tests; non-normally distributed data are presented as median (interquartile range) and compared using the Wilcoxon signed-rank test. CARU, combined antegrade and retrograde ureterography; CI, confidence interval; d1, stricture length measured by CARU; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; ICC, intraclass correlation coefficient; NA, not applicable. Comparison of CARU-measured stricture length, intraoperative stricture length, and lingual mucosal graft length (cm) Normally distributed paired data are presented as mean ± standard deviation and compared using paired t-tests; non-normally distributed data are presented as median (interquartile range) and compared using the Wilcoxon signed-rank test. CARU, combined antegrade and retrograde ureterography; CI, confidence interval; d1, stricture length measured by CARU; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; ICC, intraclass correlation coefficient; NA, not applicable. Subgroup analysis Cases were stratified by aetiology into congenital or idiopathic and acquired groups. In the congenital or idiopathic group, no significant differences were seen among d1, d2, and d3. For d1 versus d2, the median values were 3.30 cm (IQR, 1.60–7.30 cm) and 3.50 cm (IQR, 2.50–7.50 cm), respectively (Z=−1.962; P=0.050). For d1 versus d3, the mean values were 4.47±3.49 and 4.56±2.51 cm, respectively (t=0.202; P=0.84). For d2 versus d3, the median values were both 3.50 cm, although the IQRs differed (2.50–7.50 vs. 3.00–5.00 cm; Z=−0.406; P=0.68). Agreement remained high, with ICCs of 0.95 (95% CI: 0.84–0.99) for d1 versus d2, 0.90 (95% CI: 0.68–0.97) for d1 versus d3, and 0.93 (95% CI: 0.76–0.98) for d2 versus d3. By contrast, in the acquired group, significant differences were seen in all pairwise comparisons. For d1 versus d2, the mean values were 2.71±2.07 and 3.50±1.75 cm, respectively (t=6.483; P<0.001). For d1 versus d3, the mean values were 2.71±2.07 and 3.73±1.79 cm, respectively (t=7.791; P<0.001). For d2 versus d3, the median values were 3.00 cm (IQR, 2.00–4.50 cm) and 3.30 cm (IQR, 2.30–5.00 cm), respectively (Z=−3.456; P<0.001). Despite these differences, agreement remained good to excellent, with ICCs of 0.80 (95% CI: 0.72–0.86) for d1 versus d2, 0.77 (95% CI: 0.68–0.84) for d1 versus d3, and 0.94 (95% CI: 0.91–0.96) for d2 versus d3. These findings suggest that, in congenital or idiopathic strictures, paired measurements were broadly comparable and showed high agreement, whereas in acquired strictures, significant pairwise differences were present despite persistently high agreement, consistent with the overall analysis (Table 5). Table 5Stricture length measurements stratified by etiology (congenital or idiopathic vs. acquired) (cm)GroupComparisond1d2d3Test statisticP valueICC (95% CI)Congenital or idiopathic 1d1 vs. d23.30 (1.60–7.30)3.50 (2.50–7.50)NAZ=−1.9620.0500.95 (0.84–0.99)Congenital or idiopathic 2d1 vs. d34.47±3.49NA4.56±2.51t=0.2020.840.90 (0.68–0.97)Congenital or idiopathic 3d2 vs. d3NA3.50 (2.50–7.50)3.50 (3.00–5.00)Z=−0.4060.680.93 (0.76–0.98)Acquired 1d1 vs. d22.71±2.073.50±1.75NAt=6.483<0.0010.80 (0.72–0.86)Acquired 2d1 vs. d32.71±2.07NA3.73±1.79t=7.791<0.0010.77 (0.68–0.84)Acquired 3d2 vs. d3NA3.00 (2.00–4.50)3.30 (2.30–5.00)Z=−3.456<0.0010.94 (0.91–0.96)Normally distributed paired data are presented as mean ± standard deviation and compared using paired t-tests; non-normally distributed data are presented as median (interquartile range) and compared using the Wilcoxon signed-rank test. CI, confidence interval; d1, stricture length measured by combined antegrade and retrograde ureterography; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; ICC, intraclass correlation coefficient; NA, not applicable. Stricture length measurements stratified by etiology (congenital or idiopathic vs. acquired) (cm) Normally distributed paired data are presented as mean ± standard deviation and compared using paired t-tests; non-normally distributed data are presented as median (interquartile range) and compared using the Wilcoxon signed-rank test. CI, confidence interval; d1, stricture length measured by combined antegrade and retrograde ureterography; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; ICC, intraclass correlation coefficient; NA, not applicable. Cases were stratified by stricture severity into severe and stenotic groups. In the severe group, significant differences were seen in comparisons involving d1. For d1 versus d2, the mean values were 2.75±2.15 and 3.33±1.73 cm, respectively (t=3.451; P=0.001). For d1 versus d3, the mean values were 2.75±2.15 and 3.83±1.54 cm, respectively (t=3.375; P=0.001). By contrast, for d2 versus d3, the median values were both 3.00 cm (IQR, 2.00–4.25 cm for both; Z=−0.909; P=0.363), indicating no significant difference. Agreement remained high, with ICCs of 0.81 (95% CI: 0.69–0.88) for d1 versus d2, 0.73 (95% CI: 0.69–0.84) for d1 versus d3, and 0.90 (95% CI: 0.83–0.94) for d2 versus d3. In the stenotic group, significant differences were seen in all pairwise comparisons. For d1 versus d2, the mean values were 3.02±2.42 and 3.90±1.99 cm, respectively (t=5.692; P<0.001). For d1 versus d3, the mean values were 3.02±2.43 and 4.22±2.08 cm, respectively (t=7.225; P<0.001). For d2 versus d3, the median values were 3.50 cm (IQR, 2.50–4.90 cm) and 4.00 cm (IQR, 2.75–5.00 cm), respectively (Z=−3.637; P<0.001). Agreement remained high, with ICCs of 0.86 (95% CI: 0.78–0.92) for d1 versus d2, 0.85 (95% CI: 0.76–0.91) for d1 versus d3, and 0.96 (95% CI: 0.93–0.97) for d2 versus d3. These findings suggest that, in severe strictures, intraoperative stricture length and LMG length were broadly comparable and showed high agreement, whereas the other pairwise comparisons were consistent with the overall analysis (Table 6). Table 6Stricture length measurements stratified by severity (severe vs. stenotic) (cm)GroupComparisond1d2d3Test statisticP valueICC (95% CI)Severe 1d1 vs. d22.75±2.153.33±1.73NAt=3.4510.0010.81 (0.69–0.88)Severe 2d1 vs. d32.75±2.15NA3.83±1.54t=3.3750.0010.73 (0.69–0.84)Severe 3d2 vs. d3NA3.00 (2.00–4.25)3.00 (2.00–4.25)Z=−0.9090.360.90 (0.83–0.94)Stenotic 1d1 vs. d23.02±2.423.90±1.99NAt=5.692<0.0010.86 (0.78–0.92)Stenotic 2d1 vs. d33.02±2.43NA4.22±2.08t=7.225<0.0010.85 (0.76–0.91)Stenotic 3d2 vs. d3NA3.50 (2.50–4.90)4.00 (2.75–5.00)Z=−3.637<0.0010.96 (0.93–0.97)Normally distributed paired data are presented as mean ± standard deviation and compared using paired t-tests; non-normally distributed data are presented as median (interquartile range) and compared using the Wilcoxon signed-rank test. CI, confidence interval; d1, stricture length measured by combined antegrade and retrograde ureterography; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; ICC, intraclass correlation coefficient; NA, not applicable. Stricture length measurements stratified by severity (severe vs. stenotic) (cm) Normally distributed paired data are presented as mean ± standard deviation and compared using paired t-tests; non-normally distributed data are presented as median (interquartile range) and compared using the Wilcoxon signed-rank test. CI, confidence interval; d1, stricture length measured by combined antegrade and retrograde ureterography; d2, intraoperative stricture length measurement; d3, intraoperative lingual mucosal graft length measurement; ICC, intraclass correlation coefficient; NA, not applicable. Discussion Complex proximal and mid-ureteral strictures pose a significant challenge in modern urological practice (1). Comprehensive preoperative assessment of stricture complexity, including evaluation of key parameters such as etiology, length, location, and severity, is essential for guiding surgical decision-making in onlay graft or flap ureteroplasty (12,13). Antegrade pyelography and ureterography via needle nephrostomy were first described by Vela Navarrete in 1971 (20). Tscholl et al. were the first, in 1973, to use a balloon-tipped double-lumen catheter to perform retrograde ureterography and pyelography (21). The 2025 European Association of Urological Trauma Guidelines identify retrograde or antegrade ureterography as the best method for confirming ureteral injury (22). Traditional ureterography, typically performed via retrograde or antegrade approaches, is limited in accurately assessing stricture length, severity, and detecting multiple stricture segments when used alone (16). We performed ureterography combined with retrograde urethrography and antegrade cystourethrography to diagnose complex urethral strictures, as these methods effectively assess pathological conditions, anatomical changes, and landmarks in the lower urinary tract, demonstrating high preoperative diagnostic value (23). Notably, with the widespread adoption of the concept of ureteral rest, particularly for patients with complex ureteral strictures complicated by upper urinary tract infections, percutaneous nephrostomy to alleviate edema and control infection before reconstructive surgery has become a routine approach (24,25). CARU demonstrated excellent agreement with intraoperative assessment for stricture location (κ=0.93, P<0.001), indicating that it can reliably determine the anatomical level of ureteral strictures and may aid preoperative planning. Two misclassification cases were observed. In one case, a proximal stricture was missed because the dilated renal pelvis from hydronephrosis overlapped the stricture segment. In another case, a mid-ureteral stricture was overlooked due to early catheter withdrawal, which limited adequate observation of contrast flow through the mid-ureter. Similarly, CARU showed substantial agreement with intraoperative findings for stricture severity (κ=0.69, P<0.001). The accuracy of CARU is influenced by the resolution and image quality of imaging, as subtle anatomical variations, particularly in complex or multi-segmental strictures, may be challenging to detect (26). Furthermore, in cases of small or progressively worsening strictures, the interval between imaging and surgery may allow for further progression, potentially leading to misdiagnosis (27). The precise localization and assessment of stricture severity by CARU are crucial for guiding preoperative mapping and selecting the optimal surgical approach, underscoring its clinical significance (28,29). CARU tended to underestimate both true stricture length and, consequently, the graft length required for reconstruction, particularly in complex long-segment strictures. On average, CARU measurements were 0.74 cm shorter than intraoperative stricture length and 0.93 cm shorter than intraoperative LMG length. Nonetheless, ICC analysis showed good agreement between CARU-based and intraoperative measurements, supporting the feasibility of CARU for preoperative assessment of ureteric stricture length. This finding is consistent with that of Moncrief and colleagues, who reported that combined retrograde and antegrade urethrography underestimated urethral stricture length by about 1 cm, particularly in longer strictures (30). In patients with ureteric stricture undergoing CARU, contrast may still traverse the narrowed segment and opacify both ends of the stricture. Intraoperatively, however, fibrotic and scarred ureter must be excised, such that the true stricture length is often greater than that estimated by CARU. Furthermore, to ensure complete removal of poorly vascularised tissue and optimise graft survival, about 0.5 cm of healthy ureter is usually resected at both ends, which may further increase the graft length required (3,17,18). For long-segment ureteric strictures, ureteroplasty with onlay grafts or flaps has emerged as an important reconstructive option for proximal and mid-ureteric disease, with favourable clinical outcomes reported in recent years (31). Because graft options differ in obtainable length, harvesting complexity, and surgical indications, more accurate preoperative assessment of stricture length could help guide selection of the most appropriate reconstructive strategy and improve preparation for graft harvest (31). Accordingly, treatment planning for complex proximal and mid-ureteric strictures should take into account the tendency of CARU to underestimate stricture length, incorporating this discrepancy as a margin-of-safety consideration for graft-site assessment, anticipated operative complexity, and intraoperative planning. In our cohort, complex proximal and mid-ureteric strictures were mainly caused by internal trauma related to endoscopic procedures, external trauma, previous ureteric reconstruction, and idiopathic or congenital factors, consistent with most published series (31,32). Subgroup analyses by aetiology showed that, in congenital or idiopathic strictures, CARU-based measurements provided a reasonable estimate of both intraoperative stricture length and the graft length required for reconstruction, suggesting good performance in more localised benign disease. By contrast, underestimation in acquired strictures might reflect more extensive fibrosis and scarring, which can be visualised by CARU but not fully delineated (33). In addition, acquired causes often involve fibrotic extension at both ends of the stricture, necessitating resection of a longer ureteric segment at surgery and thereby contributing to discrepancies between imaging-based and intraoperative measurements (3,17,18). Subgroup analyses showed that, in severe strictures, intraoperative stricture length and graft length were broadly comparable, probably reflecting use of posteriorly augmented anastomotic ureteroplasty (34,35). In complex severe strictures, however, CARU may underestimate disease extent and, consequently, surgical requirements. CARU should therefore not be used in isolation in such cases, and intraoperative assessment of tissue tension and true stricture length remains essential. Rather than replacing other imaging modalities, CARU might best be regarded as a complementary investigation when conventional ureterography is insufficient, particularly when more detailed assessment of stricture location, severity, and length is required. This study has several limitations. First, this was a retrospective single-centre study restricted to patients undergoing LMG ureteroplasty for proximal or mid-ureteric strictures in a specialised reconstructive setting, which might have introduced selection bias and limits generalisability. Nonetheless, CARU-based assessment of stricture location, severity, and length could also have relevance for other reconstructive approaches, particularly patch-based techniques, although this will need validation in larger studies. Second, intraoperative assessors were not masked to CARU findings, and both image acquisition and intraoperative evaluation were potentially operator dependent. Third, the interval between CARU and surgery might have allowed interval changes in stricture characteristics. Fourth, the subgroup analyses were exploratory and should be interpreted cautiously. Finally, interobserver agreement was not assessed, and no direct comparator was included. Conclusions CARU demonstrates high diagnostic value in evaluating complex proximal and mid-ureteral strictures, with excellent accuracy in assessing both location and severity. It provides valuable reference for estimating stricture and graft lengths. Given its advantages in imaging these strictures, it may be a useful preoperative imaging modality for patients with proximal and mid-ureteral strictures. Supplementary The article’s supplementary files as Acknowledgments We would like to thank Professors Jinxiang Hu, Guobin Xu, and Liejun Mei for their professional guidance on this project. We also wish to acknowledge the medical team at the Department of Medical Imaging, Zhongnan Hospital of Wuhan University, for their generous support. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Ethics Committee of Zhongnan Hospital of Wuhan University (No. 2025292K), and individual consent for this retrospective analysis was waived. Footnotes Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0129/rc Funding: This work was supported by the National Natural Science Foundation of China (No. 82470724); Zhongnan Hospital of Wuhan University (No. rcyj20230102 to B.L.); and the National Natural Science Foundation of China (No. 82400810 to C.L.). Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0129/coif). The authors have no conflicts of interest to declare. Data Sharing Statement Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0129/dss References Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0129/prf Supplementary Materials The article’s supplementary files as Data Availability Statement Available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0129/dss