Last week we talked about pain and how early intervention is important. In order to achieve this early intervention, early diagnosis is critical. The gold standard for diagnosis involves surgery, which can be delayed while trials of medications and other studies are performed. At times, with expert training, ultrasounds and MRI’s can rule in endometriosis (but do not necessarily rule it out). And while there is promising research into biomarkers for endometriosis, none have demonstrated the reliability for widespread use (https://icarebetter.com/labwork-and-blood-tests/ ). Another approach to increase the suspicion of endometriosis and hopefully lead to quicker diagnosis is symptoms.
There has been much research on the symptoms involved with endometriosis, such as chronic pelvic pain. Some researchers are trying to develop “machine learning algorithms (MLA) to predict the likelihood of endometriosis” (Bendifallah et al., 2022). One study in France looked at developing such technology to help indicate to both clinicians and patients a higher probability of endometriosis. One caveat of the study was that they did not use the gold standard of diagnosis for endometriosis (surgical confirmation)- they identified patients “with diagnosis of endometriosis based on previous treatment for endometriosis or clinical examination confirming deep endometriosis, or sonography/MRI detecting ovarian, peritoneal or deep endometriosis.”
They developed a screening tool that utilized “16 clinical and symptom-based features” to make an algorithm to help aid in the “early prediction of endometriosis.” They used metrics such as age, BMI, dysmenorrhea, defecation pain, urinary pain during menstruation, sexual intercourse pain, absenteeism during last 6 months, right shoulder pain near or during menstruation, abdominal pain outside menstruation, low back pain outside menstruation, leg pain suggesting sciatica, mother/daughter history of endometriosis, number of nonhormonal pain treatments used, history of surgery for endometriosis, blood in urine during menstruation, and blood in stool during menstruation.
The researchers hope that utilization of the tool would help “reduce ‘diagnostic wandering’, and hence diagnostic delay, and result in earlier treatment” (Bendifallah et al., 2022). The researchers developed the tool with patients in mind. They report “since delays in diagnosis may contribute to undertreatment, continued pain, and prolonged symptom impact which impairs women’s quality of life, helping patients to recognize their symptoms is a crucial step toward diagnosis and effective management of endometriosis” (Bendifallah et al., 2022).
This research highlights the importance of patients monitoring their symptoms and working with their healthcare provider to achieve that earlier diagnosis and treatment. Some of these symptoms might include:
- Severe pain during menstruation (see “Pain“)
- Pelvic or abdominal pain not associated with menses
- Low back and/or leg pain
- Pain with sex (see “Sexual Functioning“)
- Painful bowel movements
- Stomach problems including nausea, bloating, diarrhea and/or constipation (see “Bowel/GI“)
- Fatigue (see “Fatigue” and “Inflammation“)
- Infertility (see “Fertility Issues“)
For more information on symptoms of endometriosis, see https://icarebetter.com/endometriosis-symptoms/ .
Here is a link to what the researchers used: https://ziwig.com/
Bendifallah, S., Puchar, A., Suisse, S., Delbos, L., Poilblanc, M., Descamps, P., … & Daraï, E. (2022). Machine learning algorithms as new screening approach for patients with endometriosis. Scientific Reports, 12(1), 1-12. Retrieved from https://www.nature.com/articles/s41598-021-04637-2
comments are closed .