ObjectiveTo evaluate the predictive value of disturbed sleep on the ability of pregabalin to reduce pain associated with diabetic peripheral neuropathy (DPN) and post-herpetic neuralgia (PHN). DesignA post-hoc analysis of data pooled from 16 placebo-controlled trials of pregabalin in patients with DPN or PHN. MethodsPain relief at endpoint was compared in patients with mild, moderate, or severe levels of baseline sleep disturbance. Sleep disturbance was based on a scale from 0-10 and scores <4, 4 to 7, and 7 were classified as mild, moderate, and severe, respectively. ResultsPregabalin significantly reduced mean pain scores in the DPN (N=3,056) and PHN (N=1,471) cohorts (mean placebo-adjusted reductions were -0.73 and -1.08 for patients with DPN/PHN, respectively; both P<0.05). The greatest extent of pain relief occurred in patients with severe sleep interference scores at baseline. Data analyses using the pooled DPN/PHN population identified a subset of patients (N=707) exhibiting marked levels of pain relief at endpoint (mean placebo-adjusted reduction of -2.88), all of whom had severe sleep interference scores at baseline. Baseline sleep interference scores were a moderately good predictor of global patient improvement in response to pregabalin treatment in both patient cohorts. Finally, path analysis showed a high degree of association between improvements in sleep and pain relief in patients with DPN/PHN. ConclusionOverall, these data suggest that the presence of comorbid sleep disturbance in patients with DPN/PHN might, in part, predict substantial pain relief in response to pregabalin treatment.