Adjunct Professor, Stanford School of Medicine
Former President, ASRM
Endometriosis sufferers often require multiple visits across multiple years before their condition is diagnosed. There are several reasons endometriosis is so poorly diagnosed, including:
Before we look at each form of diagnosis, we want to say a word about “accuracy.” Accuracy of a test can be broken down into two concepts: a test’s “positive predictive value” or PPV and its “negative predictive value” or NPV. We will refer to this constantly and so it’s worth committing to memory.
For each of the major diagnostic modalities, we break down their level of accuracy in diagnosing endometriosis depending on where the endometriosis may reside.
Patients can undergo surgery during which cameras or scopes are used by a gynecologist or reproductive endocrinologist to make a diagnosis of endometriosis. Surgery is technically considered the only way to definitively diagnose endometriosis and is considered “the gold standard,” but in current practice this is more a means for standardizing patients in research studies and many patients are presumably diagnosed and treated without a surgical diagnosis.
During surgery, endometriosis can be diagnosed two ways:
Surgery has a number of crucial benefits but also real drawbacks. For that reason, many doctors and societies recommend against it when used solely for the purpose of diagnosis.
Surgery as a diagnostic has five major benefits: it’s accurate, allows the doctor to treat the disease in that moment, allows the doctor to double check whether the fallopian tubes are truly open or closed, and helps the doctor characterize how widespread the disease has become.
Strong Accuracy: Surgery’s chief benefit is that it’s more accurate in diagnosing endometriosis than any other method we have. When the diagnosis is made by:
Among patients with symptoms, when the doctor does surgery and visually determines there is no endometriosis, in 6% of cases they are wrong and a biopsy reveals the patients actually do have endometriosis. This means visual inspection has a 94% Negative Predictive Value (says I don’t have it and I really don’t).
Ability To Treat At The Same Time: During diagnostic surgery, some clinicians can also treat (remove or destroy) the areas of endometriosis. As we’ll discuss in the next lesson, this has value for reducing pain, but may not necessarily improve pregnancy rates.
Ability To Inspect Fallopian Tubes: During diagnostic surgery, doctors can also inspect the fallopian tubes to determine whether they are open. Open tubes are a requirement for the sperm and egg to meet and for natural conception or intrauterine insemination (IUI), to work.
Surgical inspection is more accurate than other methods to look at the tubes like HSG or FemVue ultrasound. For instance, in 23% of cases an HSG says a woman’s tubes are closed when they are actually open. The implication is that for some patients who are told they must do IVF on account of their closed tubes, the option to conceive naturally, or through IUI, may still exist. For women who have been told they have a tubal blockage, visual inspection can be a valuable benefit of diagnostic surgery for endometriosis.
Ability To “Stage” Endometriosis: During surgery a doctor can get a sense for how widespread the endometriosis is. This often correlates into a “staging system.” The most popular staging system is the ASRM system whereby doctors rank the endometriosis on a 1 - 4 scale (4 being the most widespread).
The ASRM staging system is useful in that it can help your doctor know what research is relevant to your case and maybe the next logical step for treatment. In this regard, being able to “stage” your case is useful.
However, ASRM staging system has poor reproducibility (50% of the time doctors disagree on the stage) and has virtually no predictive value of a patient’s likelihood to conceive.
The one staging system we believe can credibly predict outcomes post-surgery is known as the Endometrial Fertility Index (EFI). We’ll address it in the next lesson where we discuss the likelihood of success by each treatment option (like surgery or IVF).
Diagnosis And Removal Of Other Abnormalities: During surgery other abnormalities can be found and can possibly be treated, namely, in very rare circumstances, ovarian cancer.
Surgery as a diagnostic has many drawbacks and for that reason many recommend against using it purely to make a diagnosis. Here are the reasons why:
Given this, the American Society of Reproductive Medicine recommends against using surgery solely to make a diagnosis of endometriosis in patients who don’t have symptoms. However, it may be necessary if another diagnosis needs to be excluded or if your physician thinks there may be benefit of surgically treating endometriosis.
When deciding whether surgery should be used to diagnose endometriosis, you should be aware of bias from your doctor’s background. A minimally invasive surgeon or gynecologist who performs surgery frequently is more likely to recommend surgery for diagnosis. On the other hand, a reproductive endocrinologist (who belongs to ASRM, the society that recommends against surgery for diagnostic purposes) specializes in assisted reproduction and may be less likely to perform surgery and more likely to recommend other treatments like IVF. Both perspectives can be correct depending on the situation, and while potentially confusing it can often be helpful to get a second opinion before choosing a particular treatment.
Because surgery is invasive, many doctors will make a “presumptive diagnosis” of endometriosis based on a patient’s symptoms. There is a long list of potential symptoms caused by endometriosis — most of which can also be caused by other conditions — making diagnosis tricky. The most common symptoms that should raise your suspicion of endometriosis include:
In conjunction with taking account of symptoms, a doctor will also perform a physical exam. During an exam, a doctor will place a duck-billed instrument called a speculum into the vagina to better visualize the cervix and vagina. Then with one hand placed on the lower abdomen and the other in the vagina, the doctor will try to palpate the uterus, ovaries, the ligaments that support the uterus, and any masses that may be present and appreciate any areas that are tender.
Occasionally, doctors will prescribe continuous birth control pills, progesterone, or lupron upon making a diagnosis based on symptoms. These drugs starve endometriosis (if it exists) of estrogen needed to grow. If the patient’s pain symptoms go away after birth control, the doctor will feel more confident in their initial presumptive endometriosis diagnosis. Of course, this will not be an option for women trying to conceive now.
The accuracy of taking a history and doing a physical exam is mediocre and varies depending upon where the endometriosis may reside. Accuracy for endometriosis on the ovary is high because on a normal ovaries doctors can barely feel it in on physical exam. If a doctor can feel it, or it’s tender, it may well be enlarged by endometriomas (endometriosis-filled sacs on the ovary).
More problematic is that doctors doing a physical have a very low Positive Predictive Value (says I have it & I do) for endometriosis in other parts of the pelvis. In over half of the instances, when a doctors diagnoses endometriosis not on the ovary using a physical exam, the patient does not in fact have it.
During an ultrasound, a small plastic probe is placed into the vagina and repositioned until all the structures in question have been seen. It can be done in a general OBGYN provider’s office, in a radiology office, and is always performed in a reproductive endocrinologists office as part of an infertility workup. Ultrasound is the imaging study of choice in diagnosing endometriosis according to American College of Obstetrics and Gynecology. Its major drawback deals in issues of accuracy when diagnosing non-ovarian endometriosis.
Generally speaking, the accuracy of ultrasound for diagnosing endometriosis is solid (better than a physical exam, worse than surgery). As for detecting endometriosis on the ovary, ultrasound is a good tool, as you can see below. For that reason, ultrasound has a good Positive Predictive Value (says I have it & I do) and Negative Predictive Value (says I don’t have it & I don’t).
However, ultrasound is far less reliable in diagnosing endometriosis elsewhere in the pelvis. Specifically, the Positive Predictive Value (says I have it & I do) is as low as 50%, leading many women to being wrongly told they have endometriosis, when they don’t. You’ll recall physical exams suffer from this exact same issue — in this regard ultrasound is likely better, though still not good.
Like ultrasound, Magnetic Resonance Imaging (MRI) is an imaging technique that takes pictures of internal structures. During an MRI, patients are asked to lie still on a bed that slides slowly through a machine. A doctor who specializes in medical imaging, a radiologist, looks at the images and makes a determination if there is endometriosis.
Generally speaking, MRI is no better than ultrasound, but is far less accessible and more costly, so it’s often a worse option. That said, MRI is invaluable in diagnosing endometriosis on the bowel or bladder and, when there is a credible reason it could reside there, MRI can have a valuable role to play in diagnosing it and planning for surgery.
One final word to say about MRI is that the accuracy data we display here is probably the best one could expect from MRI. Ours comes from studies using expert radiologists to interpret the images. Many radiologists are generalists and likely few are as competent as those participating in studies. MRI’s accuracy, in less experienced hands, may be quite a bit worse.
Biomarkers are an attractive diagnostic method for any medical condition, in theory providing a diagnosis with a simple blood test. For example, hCG is produced by a pregnancy and when it’s present in the blood, accurately reflects a pregnancy is underway. Unfortunately no biomarker is currently accurate enough to be recommended to diagnose endometriosis.
CA-125, a biomarker used to follow certain ovarian cancers, is often elevated in patients with endometriosis. While a recent meta-analysis demonstrated specificity (when the test says you have it, you actually have it) of 93%, its ability to detect all women who have endometriosis (known as sensitivity) is low (52%) and even lower for patients with mild or moderate endometriosis. Research is underway to develop more accurate biomarkers for endometriosis so hopefully a blood test, rather than a surgery, will be all that is required for a definitive diagnosis in the future.