Lack of evidence for Borrelia burgdorferi seropositivity in Alzheimer disease.

To the Editor: Borrelia burgdorferi sensu lato (BB) is the etiologic agent of Lyme disease. First identified in 1982, this spirochete is strongly neurotropic, resulting in chronic involvement of the central nervous system. Although defined serologic techniques are available, the diagnosis of neuroborreliosis remains difficult, especially for individuals outside an endemic area or when the typical skin lesions are absent. Interestingly, personality changes and cognitive manifestations may be the first, and occasionally, the only symptoms displayed by the patient. Considering this possibility, several authors examined possible BB involvement in disorders characterized by cognitive impairment, such as Alzheimer disease (AD). Available evidence on the relationship between BB chronic infection and AD is controversial with both positive, and negative results, all deriving from small number of subjects. We, therefore, decided to perform a case-control study, assessing the presence for serum antibodies against BB in 50 mild-to-moderate AD patients [on the basis of the National Institute of Neurological and Communicative Disorder and StrokeAlzheimer’s Disease and Related Disorder Association criteria, with computed tomography/magnetic resonance brain scans and complete neuropsychologic test batteries; age: 75± 7.4 y; disease onset: 71±8y; 22/28 M/F; mini-mental state examination (MMSE) 17.4±5.2; mean±SD]. The patients were recruited from our Dementia Outpatient Unit and compared with 2 groups of age-matched, cognitively-spared controls (MMSE> 26): 25 healthy caregivers of the enrolled patients (age: 72±2.4 y; 15/ 10 M/F), and 23 ‘‘neurologic controls’’ from our Department (age: 71±9y; 0/ 23 M/F; 12 affected by polyneuropathy, 4 by amyotrophic lateral sclerosis, 4 by multiple sclerosis, 1 by myopathy, 1 by myasthenia, and 1 by high-grade glioma). Sera were assayed for BB IgG/IgM antibodies by ELFA (Enzyme Linked Fluorescent Assay; VIDAS Lyme Screen, Biomerieux Italia S.p.A.). The BB antibody index was considered positive when >1.00; borderline when between 0.75 and 1; negative when <0.75. All the assayed sera (both patients and controls) presented an index <0.75 (negative result), with the exception of 1 AD patient that displayed a borderline value (1.00). Thus, our results do not support a role for BB in AD, based on the lack of immune response in an appreciable number of AD subjects. It is, however, known that in both the early and late phases of BB infection the antibody index can be within normal limits, suggesting that the direct measurement of BB antigens in the central nervous system would be more definitive. Moreover, our exploratory case-control study might be underpowered to detect a difference between patients and controls. Although specific data are not available, BB seropositivity in our urban/ suburban region is plausibly very low, so that an extremely high number of subjects would need to be recruited to acquire sufficient power to confirm or refute these preliminary results. Nevertheless, the prevalence estimation calculated based on our results in the AD population is extremely low (2% borderline index), allowing us to suggest that, although a specific role for BB in AD pathogenesis cannot yet be rejected, its putative global impact seems to be almost negligible, arguing strongly against population-based strategies aimed at diagnosis or treatment of this illness for the purpose of ameliorating AD.