Endometriosis early pointers

Earlier diagnosis of endometriosis

Key Points

Highlight:

• Composite pointers in GP records may be valuable for early diagnosis of endometriosis.

Importance:

• Endometriosis symptoms are non-specific and can occur a long time from diagnosis. Hence, new useful pointers in patients’ primary care electronic records would significantly enhance endometriosis diagnosis.

What's done here:

• A case-control study (366 cases) using anonymised primary care electronic health records from Scotland.
• The 366 cases of endometriosis were between 1994 and 2010, and 1453 matched population controls were obtained.
• Composite pointers include patterns of symptoms, prescribing, or investigations, in combination or over time.
• Conditional logistic regression was used to analyze the presence of pointers during the three years before diagnosis of endometriosis.

Data:

Some composite pointers that predicted endometriosis include:

• Pain and menstrual symptoms occurring within the same year (odds ratio [OR] 6.5, 95% confidence interval [CI] = 3.9 to 10.6)
• Lower gastrointestinal symptoms occurring within 90 days of gynecological pain (OR 6.1, 95% CI = 3.6 to 10.6).

Limitations:

• No secondary care records to validate the reliability of GPs’ endometriosis diagnoses.
• Only approximately half of the cases had cardinal clinical features of endometriosis recorded before diagnosis.
• The duration of the database was short and cannot capture a female’s full reproductive period.

Lay Summary

Endometriosis can take a long time to diagnose, thus associated with prolonged symptoms of pain, subfertility and patient frustration/demoralization. It can be difficult to diagnose clinically because the symptoms are common and non-specific so that GP can misconceive them as part of the regular menstruation experience or other conditions. Currently, there is limited benefit from biomarkers and imaging technology for diagnosis.

Also, composite patterns of symptoms, such as repeated episodes of dysmenorrhoea can have diagnostic value. However, these composite pointers in patients’ electronic records have not been assessed as a possible avenue to early diagnose endometriosis. Therefore, this study aimed to construct datasets from electronic health records to examine the association between conventional and composite features to predict endometriosis. Furthermore, the authors examine whether there was a period of these features to the date of diagnosis.

This case-control study contained 366 cases and 1453 randomly selected females. Endometriosis cases were between 1994 and 2010. The electronic health records were obtained from the Practice Team Information database in Scotland. The composite pointers included patterns of symptoms, prescribing, or investigations, in combination or over time.

Two important composite pointers that predicted endometriosis were (i) pain and menstrual symptoms occurring within the same year (odds ratio [OR] 6.5, 95% confidence interval [CI] = 3.9 to 10.6) and (ii) lower gastrointestinal symptoms occurring within 90 days of gynaecological pain (OR 6.1, 95% CI = 3.6 to 10.6). Hence, there was predictive value to examine several composite features for use in subsequent endometriosis diagnosis in routine records. Thus, this study showed that new useful pointers in patients’ primary care electronic records may facilitate endometriosis diagnosis.

Research Source: https://www.ncbi.nlm.nih.gov/pubmed/29109114