Nocturnal wrist movements are correlated with objective clinical scores and plasma chemokine levels in children with atopic dermatitis

Background Atopic dermatitis (AD) is a distressing disease associated with pruritus and sleep disturbance. Scratching due to pruritus is an important mechanism in the exacerbation of AD but is difficult to document in the home environment. Objectives To evaluate whether nocturnal wrist activities, defined as average acceleration in the early hours of sleep, were correlated with components of the SCORing Atopic Dermatitis (SCORAD) index and various AD-associated chemokine markers. Methods Patients with AD aged under 18 years were recruited and the severity of eczema was assessed with the SCORAD index. Concentrations of plasma AD-associated chemokines [cutaneous T-cell attracting cytokine (CTACK); macrophage-derived chemokine (MDC); thymus and activation regulated chemokine (TARC)], interleukin (IL)-18, serum total IgE, and eosinophil counts were measured in these patients. Healthy children with noninflammatory and nonitchy skin conditions as well as healthy children of staff volunteers were recruited as controls. All children were instructed to wear the DigiTrac monitor on their dominant wrist before sleeping. The monitor was programmed to record limb motion between 22.00 and 08.00 h the following morning. Results Twenty-four Chinese children with AD (mean +/- SD age 12.6 +/- 3.7 years) and 15 normal children (mean +/- SD age 11.9 +/- 3.4 years) were recruited. The median (interquartile range) SCORAD was 54.8 (32.8-70.2). Plasma concentrations in pg mL(-1) of CTACK, MDC, TARC and IL-18 in the patients were 105 (92-172), 1648 (973-4214), 258 (100-850) and 415 (304-539), respectively. When compared with controls, most wrist activities occurred at frequencies between 1 and 3 Hz. These activities were most consistent over the first 3 h of sleeping and correlated significantly with disease severity, extent, intensity, and AD-associated chemokine markers CTACK, MDC and TARC. However, there was no significant correlation between wrist activities and the subjective symptom of pruritus or sleep loss. Conclusions This is the first study to demonstrate that wrist activities, nonintrusively measured by the DigiTrac monitor at home, are closely correlated with the objective clinical scores and levels of peripheral blood chemokine markers for AD but not with the reported symptoms of pruritus or sleep loss. We propose that wrist activities between 1 and 3 Hz for the first 3 h are a good indicator of AD severity in children and should substitute for the pruritus and sleep-loss components of the SCORAD.