Association between opioid usage and rectal dysfunction in constipation: A cross-sectional study of 2754 patients.

Abstract

Opioid use has reached epidemic proportions. In contrast to the known effect of opioids on gut transit, the effect on rectal sensorimotor function has not been comprehensively investigated. Cross-sectional (hypothesis-generating) study of anorectal physiology studies in 2754 adult patients referred to a tertiary unit (2004-2016) for investigation of functional constipation (defined by "derived" Rome IV core criteria). Statistical associations between opioid usage, symptoms, and anorectal physiological variables were investigated. Opioids were sub-classified as prescriptions for mild-moderate or moderate-severe pain. A total of 2354 patients (85.5%) were classified as non-opioid users, 162 (5.9%) as opioid users for mild-moderate pain, and 238 (8.6%) for moderate-severe pain. Opioids for moderate-severe pain were associated with increased symptomatic severity (Cleveland Clinic constipation score 18.5 vs 15.1; mean difference 2.9 [95%-CI 2.3-3.6]; P<.001), rectal hyposensitivity (odds ratio 1.74 [95%-CI 1.23-2.46]; P=.002), functional evacuation disorders (odds ratio 1.73 [95%-CI 1.28-2.34]; P<.001), and delayed whole-gut transit (odds ratio 1.68 [95%-CI 1.19-2.37]; P=.003). Differences in anorectal variables between opioid users for mild-moderate pain and non-opioid users were not statistically significant. Hierarchical opioid use (non vs mild-moderate vs moderate-severe) was associated with decreasing proportions of patients with no physiological abnormality on testing (40.2% vs 38.1% vs 29.2%) and increasing proportions with both delayed whole-gut transit and rectal sensorimotor dysfunction (16.6% vs 17.5% vs 28.5%). Opioid use is over-represented in patients referred for investigation of constipation. Opioids for moderate-severe pain are associated with rectal sensorimotor abnormalities. Further studies are required to determine whether this association indicates causation.