Mo1833 - the Initial Trough Concenrtation at 36 Hours after Starting Tacrolimus is Important for the Personalized Medicine Strategy in Patients with Ulcerative Colitis

sleeping divided by time in bed.The average of up to seven days of data was then calculated."Clinically active" was defined as Harvey-Bradshaw Index of > 4 for CD or Simple Clinical Colitis Activity Index of $ 1 for UC.Each parameter was compared between disease states, then logistic regression for the outcome of clinically active disease.Multivariate analysis included age, sex, diagnosis, and comorbid sleep disorder with each sleep parameter.Results: 38 patients (26 CD, 12 UC) were followed for an average of 8.87 nights.There were no significant differences between CD and UC for age, male sex, or clinically active disease, though there was a difference in the presence of comorbid sleep disorders (table 1).A change in average sleep fragmentation by 0.1 was not predictive of clinically active disease in the univariate model, but was predictive in the multivariate model (OR 1.78 (95% CI 1.004 -3.11), p = 0.02) (table 2).While restlessness (OR 1.14 (95% CI 0.34 -3.82), p = 0.82) and sleep efficiency (OR 0.019 (95% CI 8x10-5 -4.41), p = 0.14) were not predictive.Conclusions: Worse sleep fragmentation measured by a wearable activity tracking device is associated with increased odds of clinical disease activity in IBD.Changes in sleep fragmentation patterns may have utility in a remote patient monitoring system as an early indicator of clinical disease activity.Table 1.Baseline characteristics and sleep parameters.Table 2. Multivariate logistic regression for clinically active disease.