We thank Drs. Ramchandren and Chervin for their important comments on our study of polyneuropathy in restless legs syndrome (RLS). 1 We agree that RLS may have a relevant impact on the quality of life of patients with peripheral neuropathy, although this point has not been formally investigated. 2 Thus, neuropathic patients should be questioned for symptoms of RLS, which may represent in some cases the main target of symptomatic treatment.
We are confident that our data are representative of the prevalence of RLS in polyneuropathy, as we evaluated a large series of nonselected consecutive outpatients. A reason for lower prevalence in other series 3 may be that severe cases were overrepresented, as RLS, similarly to positive sensory symptoms, is mainly related to the early phases of the disease, disappearing with worsening of neuropathy. We used quite restrictive exclusion criteria in the aim of avoiding a misdiagnosis of polyneuropathy in primary RLS, and of excluding a casual association of polyneuropathy with primary RLS. However, patients with idiopathic neuropathy were not excluded (see our study on page 1255, last but one line 1 ). Family history of RLS suggests a diagnosis of primary/genetic RLS, although the possibility of a genetic neuropathy manifesting with RLS should also be considered 4 in the presence of other signs of polyneuropathy.
It is unclear which share of RLS in the general population is provided by neuropathic patients, 5 but this component could be greater than expected, as RLS is frequent in highly prevalent forms, such as diabetic neuropathy 6,7 and HCV-related cryoglobulinemic neuropathy. 8 We found a 33% prevalence of RLS in neuropathy associated with diabetes or impaired glucose tolerance, often as onset manifestation. 7 The hypothesis of a role of ferritin in axon maintenance is stimulating, but our preliminary data 9 are not in this direction, so we are prone to think that RLS associated with neuropathy belongs to an RLS phenotype not related to ferritin deficiency. 10