Diagnostic value of lumbar puncture for the etiological assessment of uveitis: a retrospective cohort of 188 patients.

ABSTRACT

AIM: To assess the relevance of lumbar puncture (LP) for the etiological diagnosis of uveitis and to establish predictive factors associated with its contributory use. METHODS: We performed a retrospective study of patients with de novo uveitis who were referred to our tertiary hospital for etiological diagnosis of uveitis, between January 2003 and July 2018. We included patients who underwent a LP as part of the etiological assessment of uveitis. LP was considered as contributory if it led to the etiological diagnosis or to correct the initially suspected diagnosis. RESULTS: One hundred eighty eight of the 1211 patients referred for evaluation (16%) had an LP, among these patients, 93 (49.4%) had abnormal results including 69 (36.7%) patients with hypercellularity, 69 (36.7%) with hyperproteinorachia, and 28 (14.9%) with oligoclonal bands and/or increased IgG index. LP was considered as contributing to the diagnosis in only 31 (16.4%) cases, among which there were 10 (5.3%) contributions to the etiological diagnosis and 21 (11.2%) modifications in the diagnosis classification. Multivariate analysis established that African ethnicity (p < 0.001), bilateral uveitis (p = 0.01), presence of macular edema or retinal serous detachment (p = 0.048), presence of retinal vasculitis (p < 0.001), presence of neurological signs or symptoms (p = 0.01), and contributing cerebral MRI (p < 0.001) were all significantly associated with a contributory LP. LP did not lead to any therapeutic modification. CONCLUSION: LP direct contribution to the diagnosis was rare and most often detected non-specific abnormalities. LP should be performed only in cases of neurological clinical signs or symptoms, suspicion of multiple sclerosis, Vogt-Koyanagi-Harada, or syphilis.