The honest answer
A pelvic MRI does not diagnose endometriosis. Only a laparoscopy, with tissue sent to a pathologist, does that. Anyone who tells you an MRI can rule endometriosis in or out is oversimplifying, and it is one of the reasons the average UK diagnostic delay is still eight years.
What a good pelvic MRI can do is genuinely useful, and it is under-used in the UK. Done properly, on the right protocol, and read by a radiologist who reports gynaecological MRI every week, it maps out the deep disease that surgery cares about. It finds endometriomas on the ovaries, deep infiltrating nodules on the uterosacral ligaments, disease burrowing into the bowel or bladder wall, ureters being pulled or narrowed, and adenomyosis inside the uterus itself. All of that changes what the operation looks like, who is in the theatre, and how urgently it needs to happen.
So the fair way to think about pelvic MRI for endometriosis is not as a diagnostic replacement for laparoscopy, but as the map the surgeon needs before opening the door. And in a system where the wait for an NHS pelvic MRI is routinely four to six months, getting that map early can matter a great deal.
What a dedicated endometriosis MRI protocol actually is
A pelvic MRI for endometriosis is not the same scan as a general pelvic MRI. It is a protocol, and the difference is what the sequences are optimised to see.
The core of an endometriosis MRI is high-resolution T2-weighted imaging in three planes across the pelvis, with thin slices, a small field of view, and enough resolution to spot 3mm to 5mm nodules on the uterosacral ligaments and rectovaginal septum. Most UK centres add T1 with and without fat saturation to pick up haemorrhagic content typical of endometriomas, and many add a diffusion-weighted sequence. Gadolinium contrast is not routine but is sometimes added if there is a suspicion of bladder wall infiltration or complex ovarian disease.
Preparation is lighter than most patients expect. There is no full bowel prep. You will usually be asked to fast for 4 to 6 hours before the scan, to arrive with a partially full bladder, and in many centres to be given a mild anti-spasmodic on arrival to quieten bowel motion. Some units add a small amount of vaginal or rectal ultrasound gel to distend the fornix and improve visualisation of rectovaginal disease. Scan time is typically 30 to 45 minutes.
Two things matter far more than the machine: the protocol and the radiologist. A 3T scanner does not save a scan reported by a generalist who reads two pelvic MRIs a month. A 1.5T scanner reported by a subspecialist gynae radiologist who reads several a day will find nodules the first one misses.
What MRI reliably finds
Reported by a subspecialist on a proper endometriosis protocol, MRI performs well for the disease that most changes surgical planning. The reliable finds are:
- Deep infiltrating endometriosis (DIE) greater than 5mm. Nodules on the uterosacral ligaments, rectovaginal septum, torus uterinus and pouch of Douglas are usually visible on high-resolution T2 imaging.
- Endometriomas ("chocolate cysts"). These have a characteristic T1 bright, T2 shading appearance and are among the most reliably identified findings on pelvic MRI.
- Adenomyosis. MRI is currently the best non-invasive test for adenomyosis, showing a thickened junctional zone (over 12mm), small myometrial cysts and a globally enlarged uterus. This is frequently missed on ultrasound.
- Hydrosalpinx and tubal involvement. Dilated, fluid-filled tubes are usually clearly visible and matter for fertility planning.
- Ureteric involvement. MRI can show a ureter being tethered, deviated or actively obstructed by disease, which is a finding that changes the urgency of surgery.
- Bowel wall infiltration. Nodules involving the rectum or sigmoid, and the depth of that involvement, guide whether a colorectal surgeon is needed in theatre.
- Bladder wall involvement. Detrusor muscle infiltration is usually clearly seen and, again, changes the operating team.
These are the findings that most reliably alter the surgical plan, and this is where a pelvic MRI pays for itself.
What MRI often misses
The honest limits of MRI matter as much as the strengths. There are three categories the scan is not reliable for:
- Superficial peritoneal endometriosis. Thin, plaque-like disease on the peritoneal surfaces of the pelvis, without a nodular component, is often invisible on MRI. This is the most common form of endometriosis, and its absence on a scan does not rule the disease out.
- Small nodules under 5mm. Below the resolution threshold of most sequences, small deep nodules can be missed even by a subspecialist.
- Adhesions. Bowel loops tethered to the uterus, ovaries adhered to the pelvic side wall, small dense adhesions from previous disease - MRI hints at these indirectly (fixed positions on cine sequences, distortion of anatomy) but cannot map them the way a laparoscopic camera can.
This is the reason a normal MRI in a woman with classic endometriosis symptoms does not mean the pain is not endometriosis. It means the disease may be a form the scan cannot see, and laparoscopy remains the definitive next step.
Why surgical planning changes when MRI is done first
The single most useful thing an MRI does before endometriosis surgery is tell the surgeon who else needs to be in the room.
A straightforward laparoscopy for suspected superficial disease is a gynaecologist alone, a few hours in a day-case theatre. The moment MRI shows deep bowel wall involvement, that changes to a joint operation with a colorectal surgeon on the same list, often in a specialist centre. Bladder infiltration adds a urologist. Ureteric involvement, especially with any obstruction visible on imaging, moves the operation up the priority list because untreated obstruction can silently damage the kidney.
None of this is theoretical - it is the difference between a patient waking up from surgery with the disease fully treated in one operation, and a patient waking up to be told the operation had to be abandoned because the surgeon found more than expected. That second scenario, well documented in UK endometriosis pathways, is exactly what a pre-operative MRI is designed to prevent.
