MRI before CSF: prospective evidence that the 2024 McDonald criteria reduce lumbar punctures

Gilmartin, Christopher GS, Clarke, Margareta A, Allen, Christopher M, Vagharia, Ram, Abdel-Fahim, Rasha, Tallantyre, Emma ORCID: https://orcid.org/0000-0002-3760-6634, Schmierer, Klaus, Morgan, Paul S, Dineen, Rob A and Evangelou, Nikos 2026. MRI before CSF: prospective evidence that the 2024 McDonald criteria reduce lumbar punctures. Multiple Sclerosis and Related Disorders , 107984. 10.1016/j.msard.2026.107984

PDF - Accepted Post-Print Version Available under License Creative Commons Attribution. Download (1MB)

Abstract

Background The 2024 McDonald criteria for the diagnosis of multiple sclerosis (MS) introduced MRI biomarkers including the central vein sign (CVS), which may substitute for cerebrospinal fluid (CSF) testing in certain circumstances. We evaluated how the sequential application of MRI biomarkers incorporated into the 2024 criteria changes the need for CSF examination. Methods Our primary analysis used the DECISIve cohort, a prospective, multicentre study of patients with typical clinically isolated syndrome who all required lumbar puncture (LP) for diagnosis under 2017 MS diagnostic criteria. Participants received both LP and MRI with T2* susceptibility-based imaging. We applied the 2024 criteria sequentially, evaluating MRI topographies and CVS status prior to determining whether LP was still required. To assess generalisability, we evaluated a separate consecutive single-centre real-world cohort. Results In DECISIve (n=98), 70% (69/98) could be diagnosed without LP under 2024 criteria when MRI with CVS analysis was performed first: 21% (21/98) had lesions in ≥4 topographies, 72% (48/67) with 2–3 topographies were CVS-positive. In the real-world cohort (n=87), 69% (60/87) met the 2024 criteria without LP, consistent with the trial finding. There was significant concordance between paraclinical tests in DECISIve: 90% (19/21) of participants with ≥4 topographies were also CVS-positive, and 81% (56/69) of OCB-positive participants were CVS-positive. Only 11% (10/88) of participants with at least two topographies were solely OCB-positive. Conclusions Assessing MRI with CVS first removes the need for LP in most patients, a finding corroborated in real-world practice. Collaborative effort is needed to develop CVS/PRL expertise across centres.

Item Type:	Article
Date Type:	Published Online
Status:	In Press
Schools:	Schools > Medicine
Publisher:	Elsevier BV
ISSN:	2211-0348
Date of First Compliant Deposit:	26 October 2026
Date of Acceptance:	11 October 2026
Last Modified:	26 Oct 2026 12:15
URI:	https://orca.cardiff.ac.uk/id/eprint/190416

Actions (repository staff only)

Edit Item