Supraspan memory performance is impaired in subjective cognitive impairment compared to cognitively unimpaired individuals

Abstract

The objective cognitive status is incompletely known in Subjective Cognitive Impairment (SCI) and similar constructs. To characterize subspan and supraspan memory performance in groups with SCI, Mild Cognitive Impairment (MCI), and Cognitively Unimpaired individuals (CU) in relation to brain atrophy and CSF biomarkers. Performance on subspan (Digit Span Forward) and supraspan memory (trial 1 in the Rey Auditory Verbal Learning test) was investigated in patients diagnosed with SCI (n = 237) and MCI (n = 1038) from memory clinics in the Stockholm region and CU individuals from Stockholm university, Department of Psychology (n = 124). All participants had an extensive cognitive assessment. In addition, MCI and SCI patients had a comprehensive clinical examination, brain imaging scan (medial temporal lobe, cortical global atrophy, and white-matter-hyperintensities), and CSF biomarker analyses (Abeta, p-tau, and total-tau). Groups differed significantly in demographic characteristics, which were adjusted for in all analyses. The three groups differed significantly on supraspan, but not subspan, memory performance in line with the severity of cognitive impairment. The test-by-group interaction was also significant showing that SCI was characterized by selective supraspan impairment in conjunction with memory overload and in contrast to the previous conception of no objective impairment in SCI. CSF biomarkers and brain abnormality in MTA, GCA, and WMH did not provide additional predictive power over the diagnostic group on supraspan in SCI. Supraspan memory performance differed by diagnostic group in relation to the degree of cognitive impairment in memory clinic participants (CU > SCI > MCI) indicating objective memory impairment in SCI interpreted as due to supraspan overload.

Keywords: Memory; Subjective cognitive impairment; Subspan; Supraspan.

Fig. 1

Score on tests of subspan (Digit Span Forward) and supraspan (RAVL trial 1) memory with 95% confidence interval for cognitively unimpaired individuals, subjective cognitive impairment and Mild Cognitive Impairment (CU, SCI, and MCI).

Fig. 2

(A) A line graph with 95% CI showing the significant diagnosis-by-CSF Abeta interaction across tests. (B) A line graph with 95% CI showing the significant diagnosis-by-test interaction across CSF Abeta abnormality. (C) A line graph with 95% CI showing the non-significant CSF Abeta-by-test interaction across diagnostic groups.