OBJECTIVE: The digital rectal examination (DRE) is commonly recommended in the evaluation of individuals with undifferentiated abdominal pain (UAP) despite negligible evidence. We aimed to determine its clinical utility. METHODS: We prospectively enrolled 893 UAP subjects in an observational, convenience sample study at our urban, academic emergency department. Exclusion criteria were age younger than 18 years, vaginal or rectal bleeding, pregnancy, and primarily flank pain or vaginal complaint. Data forms were completed before laboratory testing, imaging, or other diagnostic maneuvers; and structured telephone and chart follow-up was performed. Physicians were asked to document DRE findings contemporaneously as well as their perception of the DRE's diagnostic impact. Final diagnosis in each case was compared with DRE result. RESULTS: Five hundred thirty-eight of 893 (60%) subjects had a DRE performed at the discretion of the examining physician. The DRE was classified as "not useful" in 494 of 538 (92%). In the remaining 44 subjects, physicians reported having their differential diagnosis influenced by the DRE. Seventeen of 538 (3%; 0.95 confidence interval, 1.9%-5.0%) were diagnostically helped, and 12 of 538 (2%; 0.95 confidence interval, 1.3%-3.9%) were diagnostically harmed. Eleven of 538 (2%) were lost to final follow-up. Sensitivity analysis indicates that unless all 11 had diagnostically helpful DREs, the DRE was statistically as likely to be harmful as helpful. CONCLUSION: Differential diagnosis was unaffected in most subjects undergoing DRE, and it appears as likely to be harmful as helpful in predicting final diagnosis. Given the discomfort and minimal predictive value of the DRE in this setting, highly selective use seems reasonable.
OBJECTIVE: The digital rectal examination (DRE) is commonly recommended in the evaluation of individuals with undifferentiated abdominal pain (UAP) despite negligible evidence. We aimed to determine its clinical utility. METHODS: We prospectively enrolled 893 UAP subjects in an observational, convenience sample study at our urban, academic emergency department. Exclusion criteria were age younger than 18 years, vaginal or rectal bleeding, pregnancy, and primarily flank pain or vaginal complaint. Data forms were completed before laboratory testing, imaging, or other diagnostic maneuvers; and structured telephone and chart follow-up was performed. Physicians were asked to document DRE findings contemporaneously as well as their perception of the DRE's diagnostic impact. Final diagnosis in each case was compared with DRE result. RESULTS: Five hundred thirty-eight of 893 (60%) subjects had a DRE performed at the discretion of the examining physician. The DRE was classified as "not useful" in 494 of 538 (92%). In the remaining 44 subjects, physicians reported having their differential diagnosis influenced by the DRE. Seventeen of 538 (3%; 0.95 confidence interval, 1.9%-5.0%) were diagnostically helped, and 12 of 538 (2%; 0.95 confidence interval, 1.3%-3.9%) were diagnostically harmed. Eleven of 538 (2%) were lost to final follow-up. Sensitivity analysis indicates that unless all 11 had diagnostically helpful DREs, the DRE was statistically as likely to be harmful as helpful. CONCLUSION: Differential diagnosis was unaffected in most subjects undergoing DRE, and it appears as likely to be harmful as helpful in predicting final diagnosis. Given the discomfort and minimal predictive value of the DRE in this setting, highly selective use seems reasonable.