As clinicians, we are encouraged to use symptom-based diagnostic criteria to diagnose irritable bowel syndrome (IBS). The Rome III criteria have been validated in one large study and were found to have a sensitivity of 68.8% and specificity of 79.5%. The positive likelihood ratio (PLR) and negative likelihood ratio (NLR) were found to be 3.35 and 0.39, respectively. Although the Rome IV criteria have recently been published, they have yet to be externally validated. However, as only minor changes have been made, they are likely to perform similarly to the Rome III criteria.
The aim of our study was to ascertain if the accuracy of the Rome III criteria can be improved if combined with simple laboratory tests (haemoglobin and C-reactive protein [CRP]), absence of nocturnal passage of stool, markers of mood (anxiety and depression), somatisation, or combinations thereof.
The study included 318 patients referred from primary to secondary care for investigation of lower gastrointestinal tract symptoms, who underwent complete colonoscopy and provided complete Rome III symptom data. The accuracy of the Rome III modifications was compared to a reference standard of lower abdominal pain or discomfort occurring at least once in a week, in association with a change in bowel habit, the absence of organic lower gastrointestinal disease after colonoscopy, and where appropriate, the exclusion of coeliac disease.
All modifications showed specificity and PLRs greater than that of the Rome III criteria alone. The best performing modifications were: Rome III, with a high level of somatisation (specificity 94.8%, 95% confidence interval [CI]: 90.6–97.5%; PLR 7.27, 95% CI: 3.74–14.2); Rome III, with normal haemoglobin, CRP, and a high level of somatisation (specificity 96.8%, 95% CI: 92.0–99.1%; PLR 7.56, 95% CI: 2.63–21.7); and Rome III, with no nocturnal passage of stool and a high level of somatisation (specificity 99.0%, 95% CI: 96.3–99.9%; PLR 17.3, 95% CI: 4.45–67.6).
Our study has shown that the Rome III criteria can be improved if combined with other relevant markers from the clinical history of the patient and simple laboratory tests. We believe that our findings may aid in the development of an accurate and inexpensive clinical test that is easily administrable, and which will allow the consulting physician to confidently make a positive diagnosis of IBS.