Association of Cerebrospinal Fluid Parameters and Neurofilament Light Chain With Retinal Nerve Fiber Layer Thickness in Multiple Sclerosis

Nik Krajnc; Patrick Altmann; Katharina Riedl; Christoph Mitsch; Thomas Berger; Fritz Leutmezer; Paulus Rommer; Berthold Pemp; Gabriel Bsteh

doi:10.3389/fneur.2022.814734

Association of Cerebrospinal Fluid Parameters and Neurofilament Light Chain With Retinal Nerve Fiber Layer Thickness in Multiple Sclerosis

Front Neurol. 2022 Mar 7:13:814734. doi: 10.3389/fneur.2022.814734. eCollection 2022.

Authors

Nik Krajnc¹, Patrick Altmann¹, Katharina Riedl¹, Christoph Mitsch², Thomas Berger¹, Fritz Leutmezer¹, Paulus Rommer¹, Berthold Pemp², Gabriel Bsteh¹

Affiliations

¹ Department of Neurology, Medical University of Vienna, Vienna, Austria.
² Department of Ophthalmology, Medical University of Vienna, Vienna, Austria.

Abstract

Introduction: Multiple sclerosis (MS) pathophysiology comprises both inflammatory and neurodegenerative characteristics. Cerebrospinal fluid (CSF) analysis allows for assessment of inflammation while neurofilament light chain can indicate neuroaxonal damage. Retinal thinning is a robust prognostic biomarker for neurodegeneration in MS. To date, an association between CSF parameters upon MS diagnosis and retinal thinning has not been investigated.

Aims and objectives: We aimed to determine whether CSF parameters are associated with the evolution of retinal layer thinning in people with MS (pwMS).

Methods: For this longitudinal observational study, we investigated pwMS from the Vienna MS database (VMSD), who had undergone (1) a diagnostic lumbar puncture (LP) between 2015 and 2020, and (2) simultaneous optical coherence tomography (OCT) and/or (3) a follow-up OCT scan. Linear stepwise regression models were calculated with OCT parameters (peripapillary retinal nerve fiber layer [pRNFL] thickness at LP and at follow-up, annualized loss of pRNFL thickness [aLpRNFL]) as a dependent variable, and CSF parameters (white blood cell [WBC] count, total protein [_CSFTP], CSF/serum albumin ratio [Q_alb], intrathecal synthesis of immunoglobulins, neurofilament light chain [NfL] in both CSF and serum [_CSFNfL/sNfL]) as independent variables adjusted for age, sex, and disease duration.

Results: We analyzed 61 pwMS (median age 30.0 years [interquartile range 25.5-35.0], 57.4% female, median disease duration 1.0 month [IQR 0-2.0] before LP, median follow-up 1.9 years [IQR 1.1-3.5]). _CSFNfL and sNfL measurements were available in 26 and 31 pwMS, respectively. pRNFL thickness at LP was inversely associated with the CSF WBC count (β = -0.36; 95% CI -0.51, -0.08; p = 0.008). We did not find any association between other CSF parameters, including _CSFNfL, sNfL, and aLpRNFL.

Conclusions: Increased WBC count as an indicator of acute inflammation and blood-brain-barrier breakdown seems to be associated with the amount of retinal thickness already lost at the time of LP. However, neither routine CSF parameters nor a singular NfL measurement allows the prediction of future retinal thinning.

Keywords: RNFL; biomarker; cerebrospinal fluid; multiple sclerosis; neurofilament; optical coherence tomography.