Mesh implants for the treatment of stress urinary incontinence (SUI) and pelvic organ prolapse (POP) have been under scrutiny due to their potential complications. Women are referred to our tertiary centre with complaints that may be related to their implants. We use Magnetic Resonance Imaging (MRI) as an assessment tool in this patient population. However, there is limited published data in the use of MRI to evaluate mesh implant complications [1,2].  

The aim of this study is to present our experience as a tertiary centre of using MRI selectively in assessing women presenting with potential implant complications, to evaluate the concordance of MRI reports to clinical  and surgical findings and to evaluate the role of MRI in this patient group.

A retrospective cohort study of a sub-set of women who presented to our unit with suspected complications, from either vaginally and / or abdominally placed mesh implants for POP and/or SUI and were investigated with MRI , between January 2018 – December 2019 was carried out.

We looked at the primary complaint of the woman, clinical symptoms and signs leading to an MRI request, MRI results and intra-operative findings, if the woman had surgery for this reason, over this time period.

MRI was carried out in these women to potentially identify the extent of any infection and to confirm location and placement of the implants. Additionally, an MRI was done in women presenting with symptoms that were not clinically attributable to implants to see if mesh-related pathology was apparent on imaging. In some cases, it was also used to investigate other causes for patient symptoms (e.g. pain due to musculoskeletal causes).
All of these MRIs were reviewed by our mesh centre radiology team, as part of our Multidisciplinary team (MDT) mesh service, even if the MRI was performed in a different unit.

On MRI: 
A. An implant was deemed to be ‘partially’ seen if the radiologist reported that only part of the implant was visible. If the report did not mention the implant, this data has been recorded as such. 
B. mesh-related complications were identified, which included: 
1.	Infection- defined as evidence of abscess, sinus tract or collections directly associated with the mesh implant, 
2.	Mesh rupture- defined as the detachment of an abdominal mesh from point of attachment
3.	Mesh exposure.

65 women with either vaginal implants, abdominal mesh or both presented with suspected complications and were investigated with an MRI during this time period. 47 patients had only 1 mesh implant in situ, 12 patients had 2 separate mesh implants and 6 had 3 separate mesh implants in place.

26 women had transobturator tapes (TOT) implants in situ, 8 had a Retropubic-tape (RP-Tape) mesh implants in situ, 20 had trans-vaginal mesh (TVM) implants for POP and 23 had abdominal mesh for POP (sacrocolpopexy / rectopexy).

Women had more than one presenting complaint. 45 women presented with pain, 11 with vaginal mesh exposure, 5 with urinary symptoms, 1 with bowel symptoms, 6 with vaginal discharge, 7 with dyspareunia. 3 women also had concurrent POP However, 3 women presented with a  recurrent POP following a sacrocolpopexy as their only symptom. 

Looking at the separate implant types, MRI was able to detect: 
o	Abdominal meshes in 91.3% (21/23) of women,   
o	Retropubic tape in 75% (6/8) of women, 
o	TOT mesh in 42.3% (11/26) of women,
o	TVM in 50 % (10/20) of women. 

Overall, MRI picked up pathology, both mesh-related and not, in 23.1% (15/65) of women. MRI detected no positive radiological findings  in 50 women. 

4/15 patients in whom the initial complaint was of pelvic pain had an MRI which showed no mesh related complications, but which revealed other pathology which could explain their pain. 2 of them were subsequently referred to Orthopaedics for further investigation of degenerative hip changes, 1 to Neurology for investigation of spinal abnormalities and 1 to General Surgery for investigation of proctitis. 

In the 11/15 women in whom a mesh-associated complication was detected, the mesh itself was identified on MRI in 7 of them, was partially seen in 3 and was not identified in 1 of them. The positive findings are described below: 

o	In the 5 women with mesh-associated infection (Table 1), MRI showed more extensive pelvic infection than the initial clinical findings suggested (example Picture 1) and helped inform surgical management and patient counselling. All 5 women underwent surgery and MRI were confirmed at time of surgery. Of these 5 women, MRI also identified the concurrent vaginal mesh exposure in 1 patient,


o	1 woman had mesh exposure in the bladder with concomitant vaginal exposure and 1 woman had urethral exposure on MRI. Cystoscopy and subsequent surgery confirmed the findings. 


o	With abdominal meshes, 2 women had mesh rupture on MRI.  MRI findings were confirmed on surgery in one woman and the other 1 is awaiting surgery. 

o	1 woman, who presented with groin pain, was found to have inflammation linked to the right arm of her TOT implant on MRI. She is awaiting surgery for tape removal. 


o	Overall, MRI identified vaginal mesh exposure in 18.8% (3/16) of patient with clinically confirmed vaginal mesh exposure. 1 woman had vaginal mesh exposure alone on MRI and 2 had concomitant MRI findings, (1 with pelvic infection, other with bladder mesh exposure, as described above).

Looking at the separate implant types, MRI was able to detect abdominal meshes in 91.3% (21/23) of women, retropubic tape in 75% (6/8) of women, TOT mesh in 42.3% (11/26) of women and TVM in 50 % (10/20) of women. Overall, MRI picked up pathology, both mesh-related and not, in 23.1% (15/65) of women.

Non-mesh related pathology that could be a cause for pain as a symptom was found in 7.4% (4/54) of women.  MRI confirmed infection in 100% of cases with clinically suspected implant-related infection and helped identify its extent. 

MRI picked up abdominal mesh rupture in 2 cases- 1 case was confirmed at surgery and the other is awaiting surgery. 

Diagnosis of vaginal exposure is poor. This, however, is something that is clinically apparent. MRI was however good at identifying more extensive infection accompanying exposure, when clinically suspected.

MRIs have a high likelihood of identifying abdominal meshes and RP-tapes but poorer at visualising vaginal meshes and TOTs. 

MRIs are able to detect mesh related infection, in every case in this series, and often shows more extensive disease than is first clinically apparent. This can be useful in patient counselling and planning surgery.

MRI may be helpful in identifying other causes for pelvic pain, but the pick up in our selected population is low. 

MRI can be a useful tool in the assessment of this complex symptomatic patient group, to aid decision-making in their clinical and surgical management. However, this has its limitations.

NICE. (2019). Urinary incontinence and pelvic organ prolapse in women: management (p. 40).
Mahoney, C., Hindle, A., & Rajashenkar, B. (2019). MR scan evaluation of pelvic organ prolapse mesh complications and agreement with intra-operative findings. International Urogynaecology Journal.