Background/Purpose: Systemic lupus erythematosus (SLE) has many symptoms, including cognitive impairment (CI), which negatively impacts social role participation and quality of life. CI is highly prevalent in SLE patients. There is a lack of data on subjective SLE-related cognitive symptoms. The Perceived Deficits Questionnaire (PDQ-20) is a patient-reported screening tool for CI in multiple sclerosis. Minimal research exists on its use in SLE. The objectives of the study are:

1. To describe subjective CI in SLE across different Canadian centres using the PDQ-20 and to determine which domains are most affected.
2. To study the concurrent construct validity of PDQ-20 compared to Symptom Severity Score (SS 2a), a scale used to assess cognitive symptoms, fatigue and waking unrefreshed. 

Methods: SLE patients (n=398) were recruited from 7 Canadian medical centers (6 academic, 1 community). Patients completed the PDQ-20, a tool with 20 items assessing perceived cognitive difficulties on a 5-point Likert scale (0-never to 4-almost always experiencing difficulty). All 3 domains of the SS 2a were analyzed alongside PDQ-20.

Cluster analysis was used to group patients by the severity of subjective CI using the PDQ-20. We examined the concurrent construct validity of the PDQ-20 against the SS 2a in these clusters. Spearman correlation coefficients were calculated for PDQ-20 subscales and SS 2a domains. We hypothesized that the waking unrefreshed and fatigue domains of SS 2a would show a weak-moderate correlation with all PDQ-20 subscales, while the cognitive symptoms domain of SS 2a would show a moderate-strong correlation. We also studied the known-group validity, and expected worse scores in SS 2a domains for the group with worse subjective CI.

Results: Out of 398 participants, 91.7% were female, with a mean enrollment age of 46.9 ± 13.7 years and mean SLE duration of 15.8 ± 11.8 years. The cohort had a mean SLEDAI-2k score of 3.3 ± 3.7 and mean SDI score of 1.1 ± 1.6. Additionally, 79% used antimalarial drugs, 61% immunosuppressants, 11% biologics, and 37% glucocorticoids (mean dose 7.2 ± 5.9 mg).

Two groups emerged: Group 1 (n=218) showed significantly higher PDQ-20 scores across all subscales compared to Group 2 (n=180), reflecting worse subjective CI. Both groups exhibited the most impairment in Attention and Planning/Organization, and the least in Prospective Memory (Figure 1). Subjective CI assessed by PDQ-20 was weakly associated with fatigue and waking unrefreshed, but moderately with cognitive symptoms in both groups (Figure 2a and 2b). Group 1 also had significantly higher mean SS 2a scores across all domains compared to Group 2 (Table 1). Both groups showed similar SLEDAI-2K (Group 1: 3.43, Group 2: 3.23) and SDI scores (Group 1: 1.08, Group 2: 1.06).

Conclusion: We identified two patient groups with different levels of subjective CI using the PDQ-20. Group 1 exhibited worse subjective CI than Group 2, with the most significant impairments in the Attention and Planning/Organization subscales of the PDQ-20. The PDQ-20 demonstrated construct validity compared to the SS 2a and may be a useful tool for measuring subjective CI among SLE patients.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/assessing-subjective-cognitive-impairment-in-a-cohort-of-canadian-systemic-lupus-erythematosus-patients-construct-validity-of-pdq-20/