NERVOUS SYSTEM INVOLVEMENT IN A COHORT OF ITALIAN PATIENTS WITH PRIMARY SJÖGREN'S SYNDROME. CLINICAL AND SEROIMMUNOLOGICAL CORRELATIONS

Background: Neurological involvement in primary Sjogren's syndrome (pSS) is a still debated issue with a prevalence ranging from 1.8-20% and 8.2 - 25% for CNS and PNS respectively.Objectives: To evaluate CNS and/or PNS involvement in a cohort of Italian patients with pSS and correlate it with other clinical and immunological manifestations of the disease.Methods: A large cohort of 380 patients with pSS, consecutively observed at our unit from January 1990 to December 2002, were retrospectively evaluated for CNS and PNS involvement, through a case-control study with clinical and instrumental investigations. All the patients fulfilled the revised 2002 ECSG criteria for pSS. Demographic, clinic and sero-immunological parameters of patients with neurologic involvement were compared with data from 107 pSS patients without neurological complications randomly selected from database, matched for sex, age and duration of disease.Results: 63 pts (16.5%, 58 female e 5 male) with pSS had neurological complications: 40 (10.5%, 36 female e 4 male) PNS complications and 23 (6%, 22 female e 1 male) CNS complications. 2 pts had involvement of both CNS and PNS. The average age of PNS patients (65 ± 12) was higher either than controls (55.2 ± 14.3, p = < 0.001*) and pts with CNS involvement (58 ± 12.6, p = 0.03*). A neurological manifestation was present at onset in 52.17% of pts with CNS disease (12/23) and in 45% of pts with PNS involvement (18/40). Among the CNS manifestations we observed a substantial equivalence of focal (10/23) and diffuse symptoms (9/23). Further 4 cases had a "miscellaneous" pattern of neurological involvement: narcolepsy (1pt), SM-like syndrome (2 pts), isolated bilateral optic neuropathy (1 pt). 2 pts had a concomitant PNS involvement. Mild, isolated headache or minor vague psychiatric complaints such as anxiety or depression were not taken into account when diagnosing neurological involvement related to pSS because of their too low specificity. Brain MRI demonstrated, in 17/21 pts (80%) with CNS involvement, the presence of multiple scattered T2-weighted hyperlucent small areas in the white cerebral matter, prevalent at the cortico-medullar junction and in the periventricular areas mainly documented in focal disease (87.5%) rather than diffuse non-focal disease (77.7%). Brain SPECT using 99Tcm-HMPAO showed perfusional defects in the same percentage between pts with focal (85.7%) and diffuse manifestations (87.5%).The most frequently observed PNS disorders were lower limbs sensory polyneuropathy (14/40, 35%) and sensory-motor polyneuropathy (8/40, 20%), followed by cranial nerve mononeuropathy and mononeuritis multiplex; we also observed 2 cases of myasthenia gravis.Sero-immunological parameters, including cryoglobulins and monoclonal gammopathy, did not significantly differ between pts with and without neurological involvement. No differences were observed either in the frequency of extraglandular features nor about cardiovascular risk factors between the pts with and without nervous system disease.Conclusion: On the basis of our results, it is impossible to characterize a subset of pSS patient with clinical or serological features related to, or predicting, neurological involvement. So, when clinically suggested, in pSS pts a careful neurological investigations is recommended.