Background

Up to 70% of patients with multiple sclerosis (pwMS) suffer from cognitive impairment. Both patients with remittent-recurrent (RR-MS) and progressive (P-MS) MS courses are concerned, especially regarding information processing speed (IPS) alteration. Among IPS tests, Symbol Digit Modalities Test (SDMT), Computerized Speed Cognitive Test (CSCT) and Weschler Adult Intelligence Scale Symbol Subtest (WAIS-IV-S) are used despite several inconveniences.

Objectives

This study aimed to approve the XO (test administered in 30 seconds only) as a new screening test of cognitive impairment in MS patients.

Methods

During a unique appointment, pwMS and healthy volunteers (HV) were assessed. The set of tests consisted in oral and written XO test, oral SDMT, CSCT, WAIS-IV-S, Go/No-Go and Trail Making Test (TMT) in a random order. The questionnaires included Hospital Anxiety and Depression Scale (HADS) and Fatigue Severity Scale (FSS). The primary endpoint was a comparison between two groups of pwMS, with or without altered IPS defined as at least one test with a result below -1.5 standard deviation.

Results

From November 2^nd 2019 to March 13^th 2020, 140 pwMS (89 RR-MS and 51 P-MS), and 117 HV were recruited. Thirty-six percent (50/140) of pwMS and 6% (7/117) of HV showed an IPS alteration (p<0.001). Twenty-six percent (37/140) of pwMS and 13% (15/117) of HV have an executive dysfunction (ED) (p=0.007). The Spearman’s rho between written XO test and IPS alteration was -0.54. Regarding the test-retest experiment (n=59), the Lin’s concordance correlation coefficient was 0.88 [0.83-0.94] for written XO. ROC area of written XO was 0.87 [0.80-0.93]. The threshold of 25.5 is the value beyond which there is an IPS alteration. Assuming 36% of pwMS have an IPS alteration, written XO sensibility is 84%, specificity is 65.6%, likelihood ratio (+) is 2.44, likelihood ratio (-) is 0.244, positive predictive value is 57.5%, and negative predictive value is 88.1%.

Conclusions

A correlation between written XO and IPS alteration was observed. The oral XO test was less relevant. There was neither correlation between XO score and ED, nor correlation between XO mistakes and ED. In 59 MS patients, XO did not show any test-retest effect which is very interesting and distinguishes it from other IPS tests used until now.

We have demonstrated that written XO can be a new useful IPS test for pwMS, without test-retest effect and independent of executive dysfunction. The written XO test is now being standardized using a healthy population of 400 individuals.