The Use of Actigraphy Differentiates Sleep Disturbances in Active and Inactive Crohn's Disease

Sleep disturbances (SDs) are commonly reported in patients with Crohn's disease (CD). Several survey instruments assessing subjective measures of insufficient sleep have identified SDs in subjects with CD. However, there are limited data on objective measures of SDs in these patients as they relate to disease activity. In this prospective cross-sectional study, we compared objective estimates of sleep obtained using multiday wrist actigraphy in individuals with CD with varying disease activity.

Eighty patients with a diagnosis of CD were recruited to take part in the study. Participants were stratified by disease activity into remission, mild disease, and moderate to severe disease groups using the Harvey-Bradshaw Index and C-reactive protein levels. Participants were excluded on the basis of significant comorbidity (Charlson Comorbidity Index ≥3), a known history of a sleep disorder, or the concomitant use of systemic corticosteroids. Participants completed surveys, including the PROMIS-SD Short Form 8a, the Epworth Sleepiness Scale, and the Women's Health Initiative Insomnia Rating scale, and were provided with an accelerometer that estimated sleep-wake patterns over 7 days. Comparisons of actigraphic sleep parameters were performed between disease activity groups. Multivariate logistic regression analyses were performed using covariates determined a priori to have an association with sleep disturbance in CD through a review of the literature.

Of the 80 participants enrolled in the study, 72 completed 5 days of actigraphy data: 28 subjects in remission, 22 subjects with mild disease activity, and 22 subjects with moderate to severe disease activity. Self-reported sleep characteristics assessed by questionnaires were similar between groups. By actigraphy, individuals with moderate to severe CD spent a significantly longer time awake after falling asleep compared with subjects with remissive disease or compared with subjects with mild disease (65.8 minutes vs 44.3 minutes and 49.1 minutes, respectively; each P < 0.05). Individuals with moderate to severe CD had significantly lower sleep efficiency compared with those with remissive CD (86.6% vs 89.9%; P = 0.03). In the multivariate analyses, moderate to severe CD disease activity was significantly associated with an increased amount of fragmented sleep (odds ratio [OR], 3.70; 95% confidence interval [CI], 1.23-11.32; P = 0.02; WASO ≥ 60 minutes). Moreover, the use of controlled substances was associated with poor sleep efficiency (OR, 3.86; 95% CI, 1.01-14.7; P = 0.04; SE ≤ 85.5%).

This is the first study to objectively quantify disturbed sleep using wrist actigraphy in adults with CD with varying disease activity. Wrist actigraphy may serve as a useful modality for discerning SD in subjects with active vs remissive disease that is not evident with questionnaires alone. Although we determined that disease severity is a significant factor that leads to SDs in CD, larger studies using these objective measures may help determine the contribution of other factors.