PWE-006 Diagnostic yield and clinical outcome for defaecating proctography and anorectal manometry in patients with chronic constipation

IntroductionDefaecating proctography (DFP) and anorectal manometry (ARM) are both used to investigate chronic constipation but their relative clinical performance is unclear. Our aim was to investigate the diagnostic yield and clinical outcomes of DFP and ARM in chronic constipation.MethodsPatients who had undergone both DFP and ARM over a 3-year period were studied retrospectively. Demographics, treatment and clinical outcomes were recorded. The diagnosis was recorded as “mixed” if investigation showed evidence of both anismus and anatomical problems such as rectocoele, intussusception or prolapse. The clinical outcome was defined as positive if the test resulted in treatment with symptomatic improvement, or resolution at follow-up. To determine whether there was a selection bias in those undergoing both DFP and ARM we additionally looked at the two groups having solely DFP or ARM from the same period.ResultsDFP and ARM group: 43 patients (40 female, 58% surgical referrals; age range 17–85 years; median 46) underwent both DFP and ARM. The diagnostic yield for DFP was higher at 98% (anismus 44%, anatomical 40%, mixed 14%; normal 2%) vs 47% for ARM (anismus 26%, mixed 21%; normal 53%). There was diagnostic concordance in only 11 (26%), partial concordance in 9 (21%) and discordance in 23 (53%) patients. Although the diagnostic yield of DFP was much greater than ARM in this combined group, both tests led to similar positive outcomes regardless (47% in DFP vs 45% in ARM) when tests revealed a pathology. Single investigation groups: 10 patients had DFP alone (8 female, 60% surgical referrals; age range 22–73 years, median 55) with a diagnostic yield of 90% (anismus 30%, anatomical 50%, mixed 10%; normal 10%). The positive outcome in those with a detectable pathology was 33%. 15 patients had ARM alone (14 female, 27% surgical referrals; age range 19–75 years, median 50) with a diagnostic yield of 67% (anismus; 33% normal). The positive outcome in those with a detectable pathology was 70%.ConclusionDFP had a higher diagnostic yield than ARM, but concordance was poor. Greater diagnostic yield did not translate into more positive clinical outcomes either. The clinical impact of additional DFP-based diagnoses is therefore questionable. The single test cohort data suggest that patients having DFP alone are a different clinical population from those who accessed both tests, since diagnostic yields and clinical outcomes were higher for ARM alone. The latter group were