Dynamic Magnetic Resonance Imaging (MRI) in Inguinal-Related Chronic Groin Pain (CGP): Comparison With Systematic Surgical Assessment

Objective This study aimed to assess the performance of dynamic MRI in Chronic Groin Pain (CGP) related to the inguinal region, comparing it with surgery as the gold standard. Materials and methods A cohort of 25 consecutive patients exhibiting persistent clinical inguinal-related CGP underwent a pre-surgical pelvis MRI. Imaging encompassed strictly axial Fast Spin Echo (FSE) T1 sequences, both without (static sequence) and with Valsalva Maneuver (VM, dynamic sequence), alongside axial-oblique Proton Density weighted with Fat Saturation (PDFS). Evaluation of these sequences focused on identifying Abdominal Wall (AW) injuries. A consistent surgical approach was employed by the same surgeon across all patients (34 AW injuries in 25 patients). Specificity (Sp), Sensitivity (Se), Negative Predictive Value (NPV), Positive Predictive Value (PPV), and overall accuracy of MRI sequences and their combinations for detecting AW injuries were computed by comparing them to surgical findings. Results Ninety sequences were obtained, revealing that the axial PDFS oblique sequence emerged as the most singularly reliable (Accuracy: 58.82%). The optimal sequence combination was found to be axial T1 combined with axial T1 VM, exhibiting an accuracy of 75.00% (Se: 85.71%, Sp: 70.59%, PPV: 54.55%, NPV: 92.31%, with an average duration of 4 minutes and 31 seconds). Conclusion Based on our findings, we advocate for the adoption of the axial FSE T1 combined with Valsalva Maneuver as a dependable protocol for inguinal-related CGP, characterized by a highly reasonable examination duration.

Recent investigations and consensus conferences have underscored the crucial role of imaging in CGP management [1].Ultrasound (US) is widely used for AW assessment due to its real-time dynamic examination capability, while Magnetic Resonance Imaging (MRI) is preferred for PS and AL assessment [1,2].However, existing literature suffers from limitations, including small sample sizes, inconsistency in imaging protocols, variations in MR scanner types, and a lack of uniformity in surgical approaches [3,4,5].
Recent studies have demonstrated a noteworthy correlation between MRI findings and surgical outcomes, particularly in cases involving AL tendinopathy and deep inguinal canal dehiscence [4,6].Ducouret et al. recently assessed the value of MRI sequences in CGP diagnosis, focusing on AW and AL injuries and systematically comparing them to surgery as the gold standard.They propose a simplified pelvis MRI protocol comprising a 4-sequence cluster for CGP: axial T1 -axial Proton Density weighted with Fat Saturation (PDFS) -sagittal PDFS -coronal T1, offering a reasonable examination time [7].
Regrettably, no prior study has specifically examined the reliability of static and dynamic MRI sequences compared to AW surgical and anatomopathological assessment.Hence, our objective was to evaluate the diagnostic reliability of dynamic MRI sequences in inguinal-related CGP, considering AW injury, and systematically comparing the results with surgery as the gold standard

Patient cohort
A retrospective study was conducted at a single center, involving 25 consecutive athletes (21 males and four females) from January 2016 to October 2020.All participants were referred to our radiology department by a specialized surgeon (GR) in CGP surgery for a pre-abdominal wall (AW) surgical MRI.Among them, five patients presented with AW inguinal-related right-side pain, 11 with AW inguinal-related left-side pain, and complained of AW inguinal-related bilateral pain, resulting in a total of 34 AW injuries in 25 patients.The mean age of the study participants was 40.12 years (from 27.12 to 53.12 years old).Informed consent was obtained from all participants.The study received approval from the local ethic board ("Comité de Protection des Personnes": CPP SOOM III, registration number DC 2015/109).

Inclusion and Exclusion Items
Inclusion criteria encompassed clinical CGP resistant to medical interventions, with pain defined as lasting more than 6 weeks and assessed by a General Practitioner (GP) with ineffective medical treatment and eccentric exercises in physiotherapy.The mean pain time before MRI was 8.2 months (± 0.7 months).Exclusion criteria comprised contraindications to MRI, age below 18 years, and prior pubic surgery.

MRI protocol
All participants underwent pelvis MRI using the same 1.5-Tesla MR scanner All participants underwent pelvis MRI using the same 1.5-Tesla MR scanner (General Electrics (GE) ® Healthcare) with a cardiac coil (16 channels, 32 elements).The following sequences were performed: axial-oblique (in symphysis plane) Proton Density weighted with Fat Saturation (PDFS), axial Fast Spin-Echo T1-weighted (FSE T1

MRI analysis
Each MRI sequence was evaluated by a musculoskeletal (MSK) radiologist (HS: 1-year MSK experience, after CGP MRI training on not included patients with a senior MSK radiologist, and BD: 10 years post-residency experience), who was blinded to patient information, except for the pain side.A minimum delay of 3 weeks was observed between each sequence interpretation.
All images were analyzed using GE® Advantage Workstation 4.5 (GE Healthcare, Chicago, USA).
The MRI images were scrutinized for AW assessment, considering: the presence or absence of inguinal or femoral hernia, fat involvement and trophicity in Rectus Abdominis (RA), presence or absence of a Transversalis Fascia (TF) bulging, defined by an anterior convexity, inguinal orifice diameter (measured at level of the emergence of the inferior epigastric vessels).
An AW was deemed injured if the inguinal orifice diameter exceeded 20 mm or if TF bulging was observed on static and/or dynamic sequences.AW disease severity was graded as follows: Grade 0: Normal abdominal wall, Grade 1: Inguinal orifice < 2 cm, Grade 2: Inguinal orifice between 2 and 3 cm, Grade 3: Inguinal orifice between 3 and 4 cm, Grade 4: Inguinal orifice > 4 cm.

Surgical intervention
All surgical procedures were conducted by the same specialized surgeon (GR) with expertise in CGP.The mainstream suture-based repair employed was the Shouldice technique, characterized by a four-layer reconstruction of the TF using Prolene sutures (Ethicon, Cincinnati, USA), coupled with a conjoint tendon lowering to the iliopubic tract (1,3).Throughout the surgical intervention, the surgeon consistently evaluated and graded the Abdominal Wall (AW).Similar to the criteria used in MRI assessment, the AW was considered injured if the surgeon observed an inguinal orifice diameter exceeding 2 cm or if a Transversalis Fascia (TF) bulging was identified (Figure 3).Histological analysis was performed for one case by the surgeon.The tissue specimen was fixed in a 10% formalin solution for 24 hours and subsequently embedded in paraffin.Three-micron-thick sections were cut and stained with hematoxylin and eosin.

Data and statistical analysis
All statistical analyses were performed with R software (R Core Team, R Foundation for Statistical Computing, Vienna, Austria).We evaluated the performance of the MRI sequence alone and in combination with AW assessment with a t-test.Non-operated AWs were asymptomatic and then considered clinically normal.The significance threshold selected for all the statistical analyses was 0.05.The results from the quantitative variables are presented as mean ± Standard Deviation (SD), minimum, maximum, and median values.The analysis was made on the basis of 34 AWs assessed, considering right and left sides in 25 patients.Specificity, Sensitivity, Positive Predictive Value, Negative Predictive Value, and Accuracy were calculated.Accuracy is defined as the ability to differentiate affected or unaffected patients correctly (Formula: (True Positives + True Negatives)/All).All calculations were made using data from surgery (AW assessment) as the gold standard and the diagnosis was made on the basements of MRI sequences which were studied as the other variable.

Surgical data
The surgeon performed 34 abdominal wall (AW) surgeries on 25 patients: five on the right side, 11 on the left side, and nine bilaterally.

Magnetic Resonance Imaging (MRI) analysis with surgical comparison
A total of 90 sequences were acquired from 25 patients, including 32 axial T1, 24 axial T1 volumetric magnetization (VM), and 34 axial oblique proton density fat saturation (PDFS) sequences.The

Histological evaluation
The histological assessment of one case revealed signs of vascular congestion and neovascularization, accompanied by disorganization of collagen fibers.Notably, there were no observed macrophages or multinucleated cells.Additionally, the examination did not identify hematic deposits, calcification, or a fibroblast repair reaction (Figure 4).

Discussion
The most effective sequence combination identified was axial T1 combined with axial T1 VM, exhibiting an accuracy of 75.00%.Based on these findings, we propose the adoption of axial Fast Spin Echo (FSE) T1 combined with Valsalva Maneuver as a reliable protocol for the assessment of CGP, offering a very reasonable imaging duration.
Axial Proton PDFS in the symphysis plane emerged as the most reliable single sequence for AW assessment.The use of this oblique plane, with a smaller Field of View (FOV), facilitated the discrimination of AW lesions in alignment with the findings of Omar et al [4].
Axial T1 VM demonstrated promise as an efficient sequence for assessing inguinal orifices bilaterally.The utilization of a larger FOV enabled the comparison of bilateral parietal components, while dynamic abdominal pressure sensitized Abdominal Wall Transversalis Fascia (AW TF) bulging, akin to ultrasound (US) practices [8][9][10][11][12][13][14][15].Ultimately, this shortened 4-minute 31-second protocol allowed for a precise assessment of AW by minimizing patient movements, which are a significant source of artifacts that can impede interpretation.
In Nevertheless, our study has several limitations.Firstly, the patient cohort was relatively small (n = 25).However, to our knowledge, it constitutes the largest homogeneous series focusing on AW assessment with a dynamic MRI, evaluating diagnostic value, and systematically comparing results to surgery.Secondly, the retrospective study design reduces the level of evidence.Nonetheless, all patient assessments were consistently performed on the same MR scanner by the same radiologist and surgeon.Lastly, the abdominal wall, being a collagen-rich structure, lacked a validated histological score for grading AW injury.

Conclusions
The study delved into the diagnostic value of static and dynamic MRI sequences, either standalone or combined, in CGP diagnosis, specifically considering AW injuries with compared outcomes to surgical assessments.Based on these findings, we recommend adopting axial Fast Spin Echo (FSE) T1 with Valsalva Maneuver as a reliable sequence for inguinal-related CGP, characterized by a very reasonable examination duration.

Figure 1
Figure1and Figure2show examples of AW MRI and surgery assessment.

FIGURE 1 :
FIGURE 1: A 33-year-old man, professional soccer player (Second league) with bilateral pain.Axial T1 in Abdominal Wall (AW)-related Chronic Groin Pain (CGP) (a) and Axial oblique Proton Density weighted with Fat Saturation (PDFS) in AW (c), with bilateral bulging (bilateral fatty overload): grade 3 in right inguinal orifice (black arrow) and grade 2 in left inguinal orifice (white arrow) in these static MRI sequences.This AW bilateral lesion becomes grade 4 in the right inguinal orifice and grade 3 in the left inguinal orifice in axial T1 VM (b) (same patient, same level).Left AW exposed in AW-related CGP with black arrow = spermatic cord (covered by external oblique aponeurosis).

FIGURE 2 :
FIGURE 2: A 28-year-old man, professional tennis player with left pain.Axial T1 in AW-related Chronic Groin Pain (CGP) (a) and Axial oblique Proton Density weighted with Fat Saturation (PDFS) in Abdominal Wall (AW) (c) with bulging (fatty overload): grade 3 in left inguinal orifice (white arrow) in these static MRI sequences.This AW lesion becomes grade 4 in the left inguinal orifice in axial T1 VM (b) (same patient, same level).Left AW exposed in AW-related CGP (before the Shouldice technique) with black arrow = weakness at conjoint tendon insertion on left Raectus Abdominis.

FIGURE 3 :
FIGURE 3: A 25-year-old man, soccer player (First League) with left pain.Axial T1 in Abdominal Wall (AW)-related Chronic Groin Pain (CGP) (a) and Axial oblique Proton Density weighted with Fat Saturation (PDFS) in AW (c) with left inguinal hernia in left inguinal orifice (white arrow) in these static MRI sequences.The important TF bulging increase in Valsalva Maneuver (VM) in axial T1 VM (b) (same patient, same level).Left AW exposed in AW-related CGP with white arrow = Prolene sutures (lowering to the inguinal ligament).The surgical clamp shows the inferior margin of an external oblique aponeurosis.
MRI findings are summarized in Table1.

TABLE 2 : Summarizing the comparison of MRI sequences with surgery in AW assessment.
[7] Omar et al article., the MRI specificity and sensitivity for CGP injury detection were reported with excellent results for evaluation concerning both wall and musculo-tendinous structures.However, Omar et al.'s study had limitations, including MR imaging from different centers with varied techniques and protocols.Their control group was smaller than their patient group and didn't match for athletic activity, age and sex patients.Additionally, this series did not describe AW patterns in surgical and MRI protocols[4].In the Larbi et al. series, specificity, sensitivity, Negative Predictive Value (NPV), and Positive Predictive Value (PPV) of MRI for AW injuries were not reported.While this pilot study demonstrated a good correlation between MRI and surgery for the inguinal canal, there was a low correlation between MRI and surgery when attempting to grade these lesions.Nevertheless, there was an important correlation when considering only the item "affected versus unaffected" (73% for AW and 100% for AL tendon).Despite systematic histology comparison, Larbi et al.'s study faced limitations, primarily stemming from the small size matched for sports activity, age, and sex[6].In the Ducouret et al. study[7], the overall accuracy in the Abdominal Wall (AW) assessment (63.42%) was lower than the findings reported by Larbi et al. (73%)[6].It's essential to note that this discrepancy takes into account a different analysis method and a larger population, which could potentially be more reflective of real Magnetic Resonance Imaging (MRI) reliability.This study specifically examined the performance of MRI sequences, both independently and in combination, for Chronic Groin Pain (CGP) diagnosis, with a systematic comparison to surgery.The accuracy reached 80.20% for Adductor Longus (AL) and 63.42% for AW.Coronal T1 Fast Spin Echo (FSGE) and axial T1 with Valsalva Maneuver (VM) were identified as the most individually reliable sequences (Accuracy: 91.67% in AL and 83.33% in AW).Comparisons with published literature led the authors to propose an MRI protocol for AL and AW assessment using a four-sequence cluster: coronal T1 combined with axial Proton Density weighted with Fat Saturation (PDFS) (in Pubic Symphysis plane), sagittal PDFS, and axial T1 VM.This four-sequence protocol, excluding Gradient Echo (GE), demonstrated broad acceptability in daily practice, with an accuracy of 77.78%, Sensitivity of 100%, Specificity of 69.23%, Positive Predictive Value (PPV) of 55.56%, and Negative Predictive Value (NPV) of 100%, all within a very reasonable average duration of 10 minutes and 36 seconds[7].