Transperineal ultrasonography in perianal Crohn's disease and recurrent cryptogenic fistula-in-ano

Development and validation of the MRI-based deep learning classifier for distinguishing perianal fistulizing Crohn's disease from cryptoglandular fistula: a multicenter cohort study

Background

A singular reliable modality for early distinguishing perianal fistulizing Crohn's disease (PFCD) from cryptoglandular fistula (CGF) is currently lacking. We aimed to develop and validate an MRI-based deep learning classifier to effectively discriminate between them.

Methods

The present study retrospectively enrolled 1054 patients with PFCD or CGF from three Chinese tertiary referral hospitals between January 1, 2015, and December 31, 2021. The patients were divided into four cohorts: training cohort (n = 800), validation cohort (n = 100), internal test cohort (n = 100) and external test cohort (n = 54). Two deep convolutional neural networks (DCNN), namely MobileNetV2 and ResNet50, were respectively trained using the transfer learning strategy on a dataset consisting of 44871 MR images. The performance of the DCNN models was compared to that of radiologists using various metrics, including receiver operating characteristic curve (ROC) analysis, accuracy, sensitivity, and specificity. Delong testing was employed for comparing the area under curves (AUCs). Univariate and multivariate analyses were conducted to explore potential factors associated with classifier performance.

Findings

A total of 532 PFCD and 522 CGF patients were included. Both pre-trained DCNN classifiers achieved encouraging performances in the internal test cohort (MobileNetV2 AUC: 0.962, 95% CI 0.903-0.990; ResNet50 AUC: 0.963, 95% CI 0.905-0.990), as well as external test cohort (MobileNetV2 AUC: 0.885, 95% CI 0.769-0.956; ResNet50 AUC: 0.874, 95% CI 0.756-0.949). They had greater AUCs than the radiologists (all p ≤ 0.001), while had comparable AUCs to each other (p = 0.83 and p = 0.60 in the two test cohorts). None of the potential characteristics had a significant impact on the performance of pre-trained MobileNetV2 classifier in etiologic diagnosis. Previous fistula surgery influenced the performance of the pre-trained ResNet50 classifier in the internal test cohort (OR 0.157, 95% CI 0.025-0.997, p = 0.05).

Interpretation

The developed DCNN classifiers exhibited superior robustness in distinguishing PFCD from CGF compared to artificial visual assessment, showing their potential for assisting in early detection of PFCD. Our findings highlight the promising generalized performance of MobileNetV2 over ResNet50, rendering it suitable for deployment on mobile terminals.

Funding

National Natural Science Foundation of China.

Role of transperineal ultrasound in evaluation of patients with perianal inflammatory disorders

Background

The aim of this study was to study the degree of accuracy of transperineal ultrasound (TPUS) in pre-operative evaluation and follow-up of cases with perianal inflammatory disorders in relation to MRI findings. Thirty patients with symptoms and signs of perianal inflammatory disorders underwent TPUS and pelvic MRI examinations. The radiologist who performed the TPUS was blinded to MR findings and was asked to comment on the site of the internal opening and extension of the tract, any related side branches or collections.

Results

The overall accuracy of TPUS in detection of the site of internal opening of perianal tracts was 83.33%, with sensitivity of 91.30%. TPUS showed sensitivity, specificity and accuracy of 100% in detection of horseshoe collections. The overall accuracy of TPUS in detection of perianal abscesses was 93.33% with specificity of 100%. TPUS showed sensitivity of 50.00%, specificity of 92.31% and accuracy of 86.67% in visualization of side branches. TPUS showed sensitivity of 33.33%, specificity of 88.89% and overall accuracy of 83.33% in detection of supralevator extension of perianal tracts. TPUS correctly classified (23/30) perianal tracts with accuracy of 76.7%.

Conclusions

TPUS can be used as a first diagnostic tool in patients with perianal inflammatory disorders and related pathologies, particularly horseshoe collections which can change the operative decision made by the surgeon. TPUS can be used in repeated follow-up during the course of management of perianal inflammatory disorders or when other imaging modalities are unavailable or contra-indicated.

Practical approach to linear endoscopic ultrasound examination of the rectum and anal canal

Standard endosonographic examination of the rectal area is usually performed with radial endoscopic ultrasound (EUS). However, in recent years, widespread availability of linear EUS for assessing various anatomical regions in the gastrointestinal tract has facilitated its use in the assessment of anorectal disorders. Currently, many rectal and anal diseases, including perianal abscesses, fistulae, polyps, and neoplastic lesions, can be well-visualized and evaluated with linear EUS. The aim of this review is to shed light on the anatomy and systematic examination of the anorectal region with linear EUS and clinical implications for different anorectal pathologies.

Strategies to Distinguish Perianal Fistulas Related to Crohn's Disease From Cryptoglandular Disease: Systematic Review With Meta-Analysis

BACKGROUND

Management of perianal fistulas differs based on fistula type. We aimed to assess the ability of diagnostic strategies to differentiate between Crohn's disease (CD) and cryptoglandular disease (CGD) in patients with perianal fistulas.

METHODS

We performed a diagnostic accuracy systematic review and meta-analysis. A systematic search of electronic databases was performed from inception through February 2021 for studies assessing a diagnostic test's ability to distinguish fistula types. We calculated weighted summary estimates with 95% confidence intervals for sensitivity and specificity by bivariate analysis, using fixed effects models when data were available from 2 or more studies. The Quality Assessment of Diagnostic Accuracy Studies tool was used to assess study quality.

RESULTS

Twenty-one studies were identified and included clinical symptoms (2 studies; n=154), magnetic resonance imaging (MRI) characteristics (3 studies; n=296), ultrasound characteristics (7 studies; n=1003), video capsule endoscopy (2 studies; n=44), fecal calprotectin (1 study; n=56), and various biomarkers (8 studies; n=440). MRI and ultrasound characteristics had the most robust data. Rectal inflammation, multiple-branched fistula tracts, and abscesses on pelvic MRI and the Crohn's ultrasound fistula sign, fistula debris, and bifurcated fistulas on pelvic ultrasonography had high specificity (range, 80%-95% vs 89%-96%) but poor sensitivity (range, 17%-37% vs 31%-63%), respectively. Fourteen of 21 studies had risk of bias on at least 1 of the Quality Assessment of Diagnostic Accuracy Studies domains.

CONCLUSIONS

Limited high-quality evidence suggest that imaging characteristics may help discriminate CD from CGD in patients with perianal fistulas. Larger, prospective studies are needed to confirm these findings and to evaluate if combining multiple diagnostic tests can improve diagnostic sensitivity.

ECCO-ESGAR Topical Review on Optimizing Reporting for Cross-Sectional Imaging in Inflammatory Bowel Disease

BACKGROUND AND AIMS

The diagnosis and follow up of patients with inflammatory bowel disease [IBD] requires cross-sectional imaging modalities, such as intestinal ultrasound [IUS], magnetic resonance imaging [MRI] and computed tomography [CT]. The quality and homogeneity of medical reporting are crucial to ensure effective communication between specialists and to improve patient care. The current topical review addresses optimized reporting requirements for cross-sectional imaging in IBD.

METHODS

An expert consensus panel consisting of gastroenterologists, radiologists and surgeons convened by the ECCO in collaboration with ESGAR performed a systematic literature review covering the reporting aspects of MRI, CT, IUS, endoanal ultrasonography and transperineal ultrasonography in IBD. Practice position statements were developed utilizing a Delphi methodology incorporating two consecutive rounds. Current practice positions were set when ≥80% of the participants agreed on a recommendation.

RESULTS

Twenty-five practice positions were developed, establishing standard terminology for optimal reporting in cross-sectional imaging. Assessment of inflammation, complications and imaging of perianal CD are outlined. The minimum requirements of a standardized report, including a list of essential reporting items, have been defined.

CONCLUSIONS

This topical review offers practice recommendations to optimize and homogenize reporting in cross-sectional imaging in IBD.

The treatment of anal fistula: second ACPGBI Position Statement - 2018

It is over 10 years since the first ACPGBI Position Statement on the management of anal fistula was published in 2007. This second edition is the result of scrutiny of the literature published during this time; it updates the original Position Statement and reviews the published evidence surrounding treatments for anal fistula that have been developed since the original publication.

Comparison of Transperianal Ultrasound With Colonoscopy and Magnetic Resonance Imaging in Perianal Crohn Disease

OBJECTIVE

Perianal fistulae and/or abscesses are common complications of Crohn disease (CD), especially in children. Magnetic resonance imaging (MRI) and gross examination under anesthesia are accurate diagnostic modalities for evaluating perianal lesions. Both methods, however, are expensive and have some limitations for use in children. This study aims to assess the accuracy of transperianal ultrasound (TPUS) and colonoscopic examination, compared with MRI, in pediatric patients with perianal CD (PACD).

METHODS

Thirty-eight children and adolescents with PACD who underwent MRI, TPUS, and gross colonoscopic examination under sedation were included. Fistulae were classified according to Parks' and St. James's University Hospital classifications. Abscesses were identified by their presence and location on each modality.

RESULTS

Fifty-nine fistulae (26 superficial, 23 intersphincteric, 10 transsphincteric) and 16 abscesses were detected using MRI. Fifty-nine fistulae and 10 abscesses were detected using TPUS. Forty-five fistulae (sensitivity 76.3%, positive predictive value [PPV] 84.2%, and kappa value 0.296] and 9 abscesses (sensitivity 56.3%, PPV 90.0%, and kappa value 0.624) on TPUS corresponded with MRI findings. Forty-six fistulae and fifteen abscesses were detected using colonoscopy. Forty fistulae (sensitivity 67.8%, PPV 89.9%, and kappa value 0.369) and 7 abscesses (sensitivity 43.8%, PPV 48.8%, and kappa value 0.304) on colonoscopy corresponded with MRI findings. With respect to abscess, agreement between TPUS and MRI was superior to that between colonoscopy and MRI (Spearman rho 0.651 vs 0.304, P = 0.007).

CONCLUSION

TPUS is an appropriate, simple, and real-time method for detecting perianal fistula and/or abscess, in children with PACD.

Role of ultrasound in colorectal diseases

Ultrasound is an undervalued non-invasive examination in the diagnosis of colonic diseases. It has been replaced by the considerably more expensive magnetic resonance imaging and computed tomography, despite the fact that, as first examination, it can usefully supplement the diagnostic process. Transabdominal ultrasound can provide quick information about bowel status and help in the choice of adequate further examinations and treatment. Ultrasonography, as a screening imaging modality in asymptomatic patients can identify several colonic diseases such as diverticulosis, inflammatory bowel disease or cancer. In addition, it is widely available, cheap, non-invasive technique without the use of ionizing radiation, therefore it is safe to use in childhood or during pregnancy, and can be repeated at any time. New ultrasound techniques such as elastography, contrast enhanced and Doppler ultrasound, mini-probes rectal and transperineal ultrasonography have broadened the indication. It gives an overview of the methodology of various ultrasound examinations, presents the morphology of normal bowel wall and the typical changes in different colonic diseases. We will pay particular attention to rectal and transperineal ultrasound because of their outstanding significance in the diagnosis of rectal and perineal disorders. This article seeks to overview the diagnostic impact and correct indications of bowel ultrasound.

The use of ultrasound in inflammatory bowel disease

Imaging in inflammatory bowel disease (IBD) plays a pivotal role in the primary diagnosis, as well as during the management of patients with known IBD. The evolution of ultrasound equipment and the growing expertise of examiners have both enhanced the role of intestinal ultrasound in the assessment of the gastrointestinal tract in IBD patients. Intestinal ultrasound has been shown to have high sensitivity and specificity, as well as high positive and negative predictive value, in the detection or exclusion of intestinal inflammatory activity in IBD. The obvious advantages of intestinal ultrasound over other imaging modalities include non-invasiveness, rapid availability and low costs. This review summarizes the current developments in the use of intestinal ultrasound for the detection of IBD and its complications, and discusses its use in the management of patients with IBD. Indications for the use of intestinal ultrasound in daily practice are presented, expanded by new developments such as contrast-enhanced ultrasonography and elastography.

Prospective Comparison of Magnetic Resonance Imaging, Transrectal and Transperineal Sonography, and Surgical Findings in Complicated Perianal Crohn Disease

OBJECTIVES

Magnetic resonance imaging (MRI) and transrectal sonography are the two accepted imaging modalities for evaluation of perianal fistulas and abscesses. Transperineal sonography is a new technique that is easy to learn and can be performed at any time. The purpose of this study was to prospectively compare the diagnostic accuracy of MRI, transrectal sonography, and transperineal sonography with surgical findings in patients with perianal Crohn disease.

METHODS

All patients with perianal Crohn disease underwent MRI, transrectal sonography, and transperineal sonography within a few days before surgery. Fistulas were classified as simple (43.8%) or complex (52.2%) based on surgical findings.

RESULTS

Twenty-three patients with active perianal Crohn disease (12 women and 11 men; mean age, 29.9 years; current therapy: antibiotics, 69.6%; azathioprine, 56.5%; and biologics, 73.9%; previous surgery, 26.1%; and proportion of smokers, 39.1%) were included. Sensitivity values for MRI, transrectal sonography, and transperineal sonography for diagnosis of fistulas were 84.6%, 84.6%, and 100%, respectively. Transperineal sonography was more sensitive for diagnosis of perianal abscesses than MRI and transrectal sonography (100%, 58.8%, and 92.8%).

CONCLUSIONS

Transperineal sonography is a very accurate diagnostic method with outstanding sensitivity compared with MRI and transrectal sonography for evaluation of complicated perianal Crohn disease. Due to its simplicity and low cost, it is recommended that transperineal sonography be the first diagnostic modality in these cases.

Transperineal ultrasonography: First level exam in IBD patients with perianal disease

BACKGROUND

A pelvic magnetic resonance imaging (MRI) represents the front-line method for evaluating perianal disease in patients with inflammatory bowel disease (IBD). Recently, transperineal ultrasonography (TPUS) has been proposed as a simple, safe, time-sparing and useful diagnostic technique to assess different pathological conditions of the pelvic floor.

AIM

The aim of this prospective single centre study was to evaluate the accuracy of TPUS versus MRI for the detection and classification of perineal disease in IBD patients.

METHODS

From November 2013 to November 2014, 28 IBD patients underwent either TPUS or MRI. Fistulae and abscesses were classified according to Parks' and AGA's classification methods. A concordance was assessed by k statistics.

RESULTS

Overall, 33 fistulae and 8 abscesses were recognized by TPUS (30 and 7 by MRI, respectively). The agreement between TPUS and MRI was 75% according to Parks' classification (k=0.67) and 86% according to AGA classification (k=0.83), while it was 36% (k=0.34) for classifying abscesses.

CONCLUSIONS

TPUS proved to be as accurate as MRI for detecting superficial and small abscesses and for classifying perianal disease. Both examinations may be performed at the initial presentation of the patient, but TPUS is a cheaper, time-sparing procedure. The optimal use of TPUS might be in follow-up patients.

Imaging of perianal fistulizing Crohn's disease

Perianal fistula is a complication of Crohn's disease that carries a high morbidity. It is a channel that develops between the lower rectum, anal canal and perianal or perineal skin. The development of perianal fistulas typically connotes a more aggressive disease phenotype and may warrant escalation of treatment to prevent poor outcomes over time. Based on fistula anatomy, debris can form inside these tracts and cause occlusion, which subsequently leads to abscess formation, fever and malaise. The clinical presentation is often with complaints of pain, continuous rectal drainage of fecal matter as well as malodorous discharge. Considering that the presence of fistulas often indicates refractory and aggressive disease, early identification of its presence is important. Some patients may not have the classic symptoms of fistulizing disease at presentation and others may have significant scarring and/or pain from previous fistulizing episodes, which can make an accurate assessment on physical exam alone problematic. As a result, utilizing diagnostic imaging is the best means of identifying the early signs of perianal fistulas or abscess formation in these patients. Several imaging modalities exist which can be used for diagnosis and management. Endoscopic ultrasound and pelvic MRI are considered the most useful in establishing the diagnosis. However, a combination of multiple imaging modalities and/or examination under anesthesia is probably the most ideal. Incomplete characterization of the fistula tract(s) extent or the presence of abscess carries a high morbidity and far-reaching personal expense for the patient - promoting worsening of the disease.

Failure of gastroenterologists to apply intestinal ultrasound in inflammatory bowel disease in the Asia-Pacific: a need for action

Intestinal ultrasound (IUS) is a cheap, noninvasive, risk-free procedure that is significantly underutilized in the diagnosis and management of patients with inflammatory bowel disease (IBD) in the Asia-Pacific region. More cost-effective methods of monitoring disease activity are required in light of the increasing global burden of IBD (especially in Asia), the advent of personalized medicine, and the rising cost of healthcare. IUS is a prime example of a technique that meets these needs. Its common clinical applications include assessing the activity and complications of IBD. In continental Europe, countries such as Germany and Italy use this imaging tool as the standard of care and have integrated it into management protocols. There are formal training programs in these countries to train gastroenterologists in IUS, and it is used in an outpatient setting during patient consultations. Barriers to its use in the Asia-Pacific region include lack of experience and research data, and there are few established centers with active training programs. These concerns can be addressed by investing more in IUS service provision and by increasing allocation of resources toward local research and training. Increased uptake of IUS will ultimately benefit patients with IBD.

Methods for preoperative assessment of anal fistula

Anal fistula is one of the most common perianal diseases, and methods for its preoperative assessment are diverse. Traditional methods include physical examination, probing, Goodsall's law, methylene blue test, and so on. Although traditional methods play a role, their accuracy for complex anal fistulas is low. Therefore, their application must be combined with other methods. Fistulography fails to depict the relationship of the fistula to the sphincter complex, and computed tomography lacks specificity to distinguish between fistula and surrounding muscle tissue. Both methods are rarely used now. Magnetic resonance imaging and three-dimensional anorectal endosonography have become routine preoperative examinations and play an important role for guiding precise surgery. However, they also have their own shortcomings. In this paper, we will summarize a variety of methods for preoperative assessment of anal fistula.

Transperineal perineal ultrasound versus magnetic resonance imaging in the assessment of perianal Crohn's disease

BACKGROUND

To assess the accuracy of transperineal ultrasound (TPUS), in comparison with magnetic resonance imaging (MRI), in classifying perianal Crohn's disease (CD) according to Parks' classification and the American Gastroenterological Association criteria.

METHODS

Fifty-nine consecutive patients with confirmed or suspected perianal CD underwent TPUS and MRI within 10 days. An independent expert surgical evaluation, which took into account proctological and MRI findings, was used as the gold standard. Fistulae and perianal disease were classified according to Parks' classification and American Gastroenterological Association criteria, respectively.

RESULTS

Forty-six patients showed 64 fistulae (9 intersphincteric, 34 transsphincteric, 2 suprasphincteric, 9 extrasphincteric, and 10 anovaginal) and 23 abscesses were also found. Fifty-one of 54 perianal fistulae (per-lesion sensitivity: 94.4%) and 9 of 10 anovaginal fistulae (sensitivity: 90.0%) were detected and 58 were correctly classified by TPUS (sensitivity: 90.6%; positive predictive value: 93.4%). Overall, TPUS correctly detected and classified the fistulae in 89% of patients and the agreement for classifying perianal fistulae between TPUS and MRI was excellent (K value: 0.783). In contrast, 11 of 23 abscesses were correctly diagnosed by TPUS (sensitivity: 47.8%); although 14 abscesses were diagnosed by TPUS, only 11 were confirmed by MRI (positive predictive value: 78.6%). Overall, TPUS correctly detected and classified fistulae and associated abscesses in 67.3% of patients. Agreement between MRI and TPUS in discriminating simple (15) and complex (29) perianal disease was fairly good (K value: 0.57).

CONCLUSIONS

TPUS is a simple and accurate diagnostic method for classifying perianal fistulae in CD and could be used for the preliminary assessment and follow-up of perianal CD.