MRI features of perianal fistulas: is there a difference between Crohn's and non-Crohn's patients?

Early diagnosis of Crohn's disease in patients presenting with a perianal fistula: systematic review and development of a perianal red flags index

BACKGROUND

Delay in diagnosing Crohn's disease (CD) in patients presenting with perianal abscess (PAA) and/or fistula (PAF) is common. The aim of this study was to identify red flags suggestive of CD.

METHODS

A systematic literature review was conducted to identify symptoms associated with CD in patients presenting with PAA/PAF. A questionnaire including those symptoms, supplemented with items from the International Organization for the Study of Inflammatory Bowel Diseases (IO-IBD) red flags index for luminal CD, was administered to all adult patients presenting with a PAF and eventually diagnosed with CD and matched patients (1:3) from the same study period with a cryptoglandular PAF (2012-2023) at a single non-academic teaching hospital. All patients were asked to recall symptoms/signs experienced during their first PAF.

RESULTS

The systematic review identified 8 articles reporting on 15 clinical characteristics in patients presenting with PAA (n = 2)/PAF (n = 6), supplemented with 13 items from the IO-IBD red flags index (28 items in total). A total of 25 patients with CD and 75 patients with PAF without CD answered the questionnaire. Univariate analysis identified seven items associated with CD (age, family history, > 2 perianal interventions, weight loss, abdominal pain, diarrhoea and fatigue), and four items remained significant in multivariate analysis: age (OR 3.4 [1.0-11.5]), > 2 previous perianal interventions (OR 3.4 [1.0-10.1]), weight loss (OR 14.4 [3.7-55.6]) and abdominal pain (OR 9.8 [1.9-49.8]). Receiver-operating characteristic curve (ROC) analysis showed that a combination of these red flags was associated with good discrimination of CD versus non-CD (AUC 0.83 [0.72-0.94]).

CONCLUSIONS

The perianal red flags index has a good predictive value for early identification of patients with PAF at risk for underlying CD.

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.

Clinical Course of Isolated Recurrent, Persistent Complex Perianal Fistulas Without Luminal Crohn's Disease: A Multicenter Case Series of 24 Patients

Background

Isolated complex perianal fistulas, without luminal evidence of inflammatory bowel disease in the gastrointestinal tract, pose diagnostic and treatment dilemmas for gastroenterologists and colorectal surgeons. For patients who develop recurrent complex fistulas, a presumptive diagnosis of Crohn's disease may be made. It is unclear whether these cases of isolated perianal disease in the absence of luminal inflammation truly represent isolated severe cryptoglandular fistulas or rather an early presentation of Crohn's disease. We aimed to investigate the clinical course and outcomes of patients with isolated complex perianal fistulas.

Methods

In this retrospective multicenter case series across 6 institutions in the United States, we report the clinical course of patients with isolated recurrent complex perianal fistulas, including their diagnostic evaluation, medical and surgical therapies, and clinical outcomes.

Results

All patients (n = 24) required incision and drainage of perirectal abscesses. The majority received setons (n = 19, 79%), more intensive surgical interventions (n = 15, 62.5%, including fistulotomy/sphincterotomy, advancement flap, and ligation of the intersphincteric fistula tract), antibiotics (n = 17, 71%), and biologic therapy (n = 16, 67%). Nine patients (37.5%) underwent a combined medical-surgical approach with biologics and intensive surgical intervention. Despite surgical and/or medical management, active symptomatic complex perianal fistulas persisted in 58% (n = 14) of patients at follow-up (median 5.5 years, interquartile range 2.5-10 years); symptom remission was achieved in 21% (n = 5), and fistula closure in 21% (n = 5).

Conclusions

These cases highlight a multidisciplinary and multimodal approach when treating isolated complex perianal fistulas and their propensity to persist despite the incorporation of advanced therapies.

The relationship between perianal fistula activity and abdominal adipose tissue in Crohn’s disease: an observational study

Objective

The aim of this study was to analyze the relationship between abdominal adipose tissue and perianal fistula activity in patients with Crohn's disease (CD) using cross-sectional imaging.

Methods

Patients with perianal fistulizing CD who underwent pelvic magnetic resonance imaging (MRI) and abdominal computed tomography (CT) were retrospectively enrolled. We scored the fistulas in each patient's MRI images based on Van Assche's classification. The area and density of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) (at the third, fourth, and fifth lumbar (L3, L4, and L5) levels were measured by averaging five slices of measurements at each vertebral level in CT images, and areas were further standardized by the lumbar height2 (heightL1-5). The VAT/SAT ratio (VSR) and VAT/Total adipose tissue (VA/TA) index were calculated. Based on MRI scores, patients were divided into two groups with low and high activity, and their clinical, MRI features, and CT parameters were compared. We evaluated patients with follow-up MRI and compared the differences in clinical and radiological indicators among patients with different outcomes.

Results

Overall, 136 patients were included, 77 in the low-activity group and 59 in the high-activity group. Patients in the high activity group had lower subcutaneous adipose index (all levels, p < 0.05) and visceral adipose index (L3 level, p < 0.01), but higher SAT and VAT density (all levels, p < 0.01), VSR (L5 level, p = 0.07) and VA/TA index (L5 level, p < 0.05).

Conclusion

There were differences in adipose tissue composition among CD patients with different active perianal fistulas.

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.

Deep Learning-Based Magnetic Resonance Imaging Features in Diagnosis of Perianal Abscess and Fistula Formation

There was an investigation of the diagnostic and prognostic effect of magnetic resonance imaging (MRI) based on multimodal feature fusion algorithm for impotence of perianal abscess. In this study, the second to fifth convolution blocks of the visual geometric group network were applied to extract the depth features in the way of transfer learning, and a multimode feature fusion algorithm was constructed. The whole network was trained by maximizing the energy proportion of the feature layers, which was compared with the fully convolutional neural network (FCN) algorithm. Then, this algorithm was adopted to the imaging diagnosis of 50 patients with anorectal diseases admitted to our hospital, and it was found that the similarity coefficient (85.37%), accuracy (80.02%), and recall rate (79.38%) of the improved deep learning algorithm were higher markedly than those of the FCN algorithm (70.18%, 67.82%, and 66.92%) (P < 0.05). As the number of convolutional layers increased, the segmentation accuracy of the convolutional neural network (CNN) algorithm was also improved. The detection rate of the observation group (84%) rose hugely compared with the control group (64%), and the difference was statistically obvious (P < 0.05). Besides, the detection accuracy of abscess location (84%), impotent tract location (80%), and internal orifice location (92%) in patients from the observation group was higher substantially than the accuracy of abscess location (60%), impotent tract location (68%), and internal orifice location (72%) from the control group (P < 0.05). In conclusion, the performance of the multimodal feature fusion algorithm was better, and the MRI image feature analysis based on this algorithm had a higher diagnostic accuracy, which had a positive effect on improving the detection rate, detection accuracy, and disease classification.

MRI of perianal fistulas in Crohn's disease

Perianal fistulas represent one of the most critical complications of Crohn's disease (CD). Management and treatment need a multidisciplinary approach with an accurate description of imaging findings.

AIM

This study aspires to assess the significative role of Magnetic Resonance Imaging (MRI) in the study of perianal fistulas, secondary extensions, and abscess in patients with CD. Therefore it is essential to standardize an appropriate protocol of sequences that allow the correct evaluation of disease activity and complications.

METHODS

We selected and reviewed ten recent studies among the most recent ones present in literature exclusively about pelvic MRI imaging and features in CD. We excluded studies that weren't in the English language.

CONCLUSIONS

MRI has a crucial role in the evaluation and detection of CD perianal fistulas because, thanks to its panoramic and multiplanar view, it gives excellent anatomic detail of the anal sphincters. Today MRI is the gold standard imaging technique for the evaluation of perianal fistulas, mainly because this technique shows higher concordance with surgical findings than does any other imaging evaluation. Surgical treatment is often required in the management of perianal fistula in patients with CD, which often have complex perineal findings.

Tumor Necrosis Factor Inhibitors May Have Limited Efficacy for Complex Perianal Fistulas Without Luminal Crohn's Disease

BACKGROUND

Complex perianal fistulas occurring in the absence of luminal inflammation (isolated perianal disease, IPD) may represent a specific phenotype of Crohn's disease (CD).

AIM

We assessed the effectiveness of tumor necrosis factor (TNF)-antagonists in patients with IPD compared to those with perianal CD (PCD) with luminal inflammation.

METHODS

Patients were identified through our institutional radiology database and were classified as PCD or IPD based on the presence or absence of luminal inflammation by ileocolonoscopy and abdominal enterography. Consecutive adults (> 17 years) with recurrent IPD who were treated with TNF antagonists were matched by age and gender to patients with complex PCD (1:2 ratio). Fistula remission was defined as an absence of fistula drainage. Surgery-free survival was assessed by Cox proportional hazard models.

RESULTS

Twenty-two patients with IPD treated with a TNF antagonist were compared with 44 matched patients with PCD. A similar proportion of patients with IPD and PCD were treated with concomitant immunomodulators (55% vs. 66%) and underwent examinations under anesthesia prior to therapy (36% vs. 46%). Fistula remission at 3, 6, and 12 months was lower for the IPD cohort: 9.5% versus 34%; 19% versus 39%; and 19% versus 43%. Surgical intervention after initiating anti-TNF therapy was more common for patients with IPD (HR 3.99: 95% CI, 1.62-9.83; p = 0.0026).

CONCLUSIONS

Fewer patients with IPD achieved fistula remission, and more required surgical intervention after anti-TNF therapy, suggesting that TNF antagonists may not be as effective in these patients.

Preclinical Model of Perianal Fistulizing Crohn's Disease

BACKGROUND

Fistulizing anoperineal lesions (FAPLs) are common and severe complications of Crohn's disease (CD), exposing patients to the risk of anal sphincter alteration and permanent stoma. Due to the limited efficacy of current treatments, identifying new local therapies is mandatory. However, testing new treatments is currently limited because no relevant preclinical model of Crohn's-like FAPL is available. Thus, a reliable and reproducible experimental model of FAPLs is needed to assess new therapeutic strategies.

METHODS

Twenty-one rats received a rectal enema of 2,4,6-trinitrobenzensulfonic acid (TNBS) to induce proctitis. Seven days later, a transsphincteric fistula tract was created with a surgical thread, instilled with TNBS twice a week until its removal at day 7 (group 1), day 14 (group 2), or day 28 (group 3). In each rat, pelvic MRI was performed just before and 7 days after thread removal. Rats were sacrificed 7 days after thread removal for pathological assessment of the fistula tract.

RESULTS

The optimal preclinical model was obtained in group 3. In this group, 7 days after thread removal, all animals (9 of 9) had a persistent fistula tract visible on MRI with T2-hypersignal (normalized T2 signal intensity: 2.36 ± 0.39 arbitrary units [a.u.] [2.08-2.81]) and elevation of the apparent diffusion coefficient (1.33 ± 0.16 10-3 millimeter squared per seconds [1.18-1.49]). The pathological examination of the fistula tract revealed acute and chronic inflammation, granulations, fibrosis, epithelialization, and proctitis in the adjacent rectum.

CONCLUSIONS

This reproducible preclinical model could be used to assess the effectiveness of innovative treatments in perianal fistulizing CD.

Contrast- vs. non-contrast enhanced MR data sets for characterization of perianal fistulas

PURPOSE

To evaluate the diagnostic efficacy between pre- and post-contrast MRI sequences in perianal fistulas using intra-operative findings as the gold standard.

MATERIALS AND METHODS

Retrospective analysis of 50 patients with a history of perianal fistula and MRI performed between January 2006 and January 2018 was performed. The inclusion criteria were patients who underwent MRI prior to surgery and had a detailed surgical report available. Pre- and post-contrast MR data sets were evaluated by two radiologists at two-week-intervals, assessed fistula type, internal/external opening, presence of abscess/secondary tracts, and confidence scores. The area under the curve (AUC) was used for comparison the diagnostic ability. The sensitivity and specificity were compared using the McNemar's test.

RESULTS

The confidence scores in detecting perianal fistulas were significantly higher in the post-contrast MR data set (p < 0.003). The post-contrast MR data set had similar ability to classify perianal fistulas as combined T2-DWI and isolated T2 data sets in 49/50 cases. For internal/external opening, the post-contrast MR, combined T2-DWI, and isolated T2 data sets had 100% concordance with intra-operative reports. For perianal abscess, there was no significant difference in sensitivity or AUC value between the isolated T2 or combined T2-DWI data sets and post-contrast MR data set (p > 0.05). All MR data sets correctly identified secondary tracts in all 50 cases.

CONCLUSIONS

Although contrast-enhanced MR studies can improve a radiologist's confidence, non-contrast MR studies had similar diagnostic efficacy in identifying perianal fistulas and their complications. Therefore, a non-contrast study may suffice in selected patients such as those with renal impairment.