RESUMEN
BACKGROUND: Intrathecal immunoglobulin synthesis (ITS) plays a crucial role in the diagnosis of multiple sclerosis (MS). Traditionally, the gold standard method for detecting ITS has been through the analysis of oligoclonal bands (OCB). However, the paradigm has shifted with the introduction of the free kappa light chains (FKLC) method. In fact, a recent consensus recommends evaluating FKLC index (FKLCi) as the primary approach and reserving oligoclonal bands with borderline results. The objective of our study is to investigate the diagnostic efficiency of combining FKLC with other methods to predict ITS while minimizing the reliance on OCB. METHODS: A total of 192 patients were included in the study, consisting of 145 individuals diagnosed with multiple sclerosis (pwMS) and 46 with other neurological diseases (controls). Among the MS cases, 100 patients were assigned to the Training Cohort (TC), while an external Validation Cohort (VC) comprised of 45 MS patients was established. Diagnostic efficiency was assessed for FKLCi, OCB, Link index, and the Reiber formula for IgG and FKLC. Optimal cutoff values for Link index and FKLCi were also determined. The last procedure was developed for diverse algorithms using the parameters mentioned above, which included the optimal cutoffs previously obtained. The calculations were conducted independently for both the TC and the VC, as well as for a composite cohort formed by combining data from all patients (OC) RESULTS: One algorithm, named KRO, was developed based on the determination of FKLCi and Reiber Formula as the primary diagnostic parameters. For cases where the FKLCi result was mildly increased, OCB was utilized as a supplementary test. The KRO algorithm demonstrated superior diagnostic accuracy in the OC (89%), resulting in a reduction of OCB consumption by 91%. DISCUSSION: The KRO algorithm demonstrated superior sensitivity and accuracy although lower specificity and NPV compared to the use of FKLCi and OCB alone. The present research aligns with the new consensus recommendations regarding the diagnostic approach. Our findings indicate that employing a combined marker approach via KRO could prove to be a proficient screening tool for multiple sclerosis. This approach also holds the potential to address inherent limitations associated with each individual marker. However, to further validate and solidify the efficacy of our algorithm, additional studies involving larger cohorts are warranted.