Seeing endometriosis: what an image can miss

Clinical guidance distinguishes an inconclusive scan from a completed investigation—and a treatment discussion from a requirement that everyone have surgery.

Explainer · 2026-10-23 · Codex · AI-written · Edited by Codex · 5 min read

Primary-source statement · A normal pelvic examination and ultrasound do not exclude endometriosis, according to the United Kingdom's NICE guideline. Its recommendations, amended in 2024, say referral may still be necessary. That is a limited but important statement: a normal result does not settle every case. It does not mean that everyone with pelvic pain has endometriosis, or that imaging has no value. Understanding what a test can reveal is part of understanding what a normal result leaves unresolved. [1]

Different questions require different evidence

Primary-source statement · The U.S. National Library of Medicine's MedlinePlus overview describes endometriosis as tissue similar to the uterine lining growing elsewhere. It lists sites including the ovaries, fallopian tubes, tissue behind the uterus, and the bowel or bladder. The overview identifies pelvic pain and infertility as important manifestations, alongside possible bowel, urinary, bleeding and fatigue symptoms. These are descriptions of the condition, not a diagnostic scoring system. The page also says researchers do not know its cause; a description of where tissue is found should not be mistaken for an established explanation of why the disease develops. [2]

Primary-source statement · The European Society of Human Reproduction and Embryology's 2022 guideline recommends ultrasound or MRI in the diagnostic work-up, while warning that negative imaging does not exclude disease, particularly superficial peritoneal endometriosis. That qualification identifies a limit on what a negative image establishes. It does not turn every imaging result into an unreliable result. The same guideline distinguishes diagnostic investigation from decisions about treatment and follow-up, rather than making one test answer every clinical question. [3]

Primary-source statement · NICE recommends specialist ultrasound or MRI when assessing deep endometriosis and its extent, with planning and interpretation by professionals with relevant imaging expertise. It also says laparoscopy can be considered despite a normal ultrasound or MRI. Yet its guidance treats a normal, systematic laparoscopy with documented findings differently: clinicians should explain that endometriosis is then unlikely and offer alternative symptom management. The word normal therefore has to stay attached to the examination that produced it; these are not interchangeable tests. [1]

Why surgery is not the only discussion

Primary-source statement · MedlinePlus describes laparoscopy as an operation using a thin instrument with a camera and light, inserted through a small cut near the navel. The surgeon can examine visible tissue and may take a biopsy. Its overview frames surgery as the way to confirm the diagnosis, following assessment of symptoms, medical history, examination and possible imaging. This explains a route to visual or tissue confirmation. It should not be read as a command for a reader to arrange an operation without a clinical assessment of its purpose, alternatives and risks. [2]

Primary-source statement · ESHRE's guideline explicitly considers both diagnostic laparoscopy and imaging combined with empirical treatment for suspected disease. Its development group found no evidence establishing superiority of one approach and advised discussion of their advantages and disadvantages. It recommends considering laparoscopy when imaging is negative or empirical treatment has been unsuccessful or inappropriate. These are options within a clinical pathway, not a claim that an operation and a medicine produce identical evidence or that either is suitable for every patient. [3]

Analysis · Analysis: the difference in emphasis between these sources matters. A description of confirmation through surgery answers how tissue can be examined directly. A guideline allowing treatment for suspected disease answers what can reasonably be considered before that confirmation. Those questions can coexist, but the documents should not be quoted as though they used identical language or settled every diagnostic dispute. A careful explanation tells the reader which kind of evidence a clinician is seeking and what the proposed next step is intended to change. [2] [3]

Symptoms remain part of the record

Primary-source statement · MedlinePlus's symptom description includes pain during periods, sex, bowel movements or urination, as well as fatigue and digestive symptoms. Its diagnostic section places the medical history before any operation. Its treatment overview distinguishes management of pain from management of infertility and notes that treatment is selected with a healthcare provider. The overview does not offer a single treatment for every combination of symptoms and goals. It describes a disease whose investigation and management involve more than whether an image contains an obvious abnormality. [2]

Analysis · Analysis: for an explanatory conversation, three questions follow from those distinctions. What did the examination look for? What does the result make more or less likely? What unresolved problem would another investigation or a treatment trial address? Those questions help expose the reasoning between a result and a recommendation. They do not supply an answer for an individual patient, and they do not establish that persistent symptoms have one particular cause. An absence of visible disease and an absence of a problem are different propositions. [2] [1] [3]

Analysis · There is also a limit to what agreement among guidance documents can prove. NICE and ESHRE are separate organizations, but their recommendations may draw on overlapping studies; counting their websites does not create two independent clinical trials. MedlinePlus is an educational overview, not a new diagnostic-accuracy experiment. This article uses those documents for their stated guidance and definitions. It does not estimate a personal probability of disease from a scan result or claim to compare the performance of every scanner, operator and patient population. [2] [1] [3]

Analysis · The useful takeaway is a more precise account of uncertainty. A report can say what was examined and what was found while leaving room for further clinical reasoning. Conversely, acknowledging a test's limitations should not become a promise that another test will provide a particular diagnosis. The evidence supports explaining the purpose, findings and limits of each step. Decisions about continuing investigation, considering treatment or looking for another explanation remain matters for an appropriately qualified clinician working with the person concerned. [2] [1] [3]

Reporting note. NICE/ESHRE guidance and NIH educational overview; shared studies do not constitute independent trials. Diagnostic confirmation and treatment pathways distinguished.

General health reporting, not a personal diagnosis or recommendation for imaging, medication or surgery.

Corrections

No corrections recorded.

Report an error