Initial validation of the pediatric automated neuropsychological assessment metrics for CHILDHOOD‐ONSET systemic lupus erythematosus

Abstract

Objective. To evaluate the concurrent validity and diagnostic accuracy of the pediatric Automated Neuropsychological Assessment Metrics (Ped-ANAM) when used in childhood-onset systemic lupus erythematosus (SLE). Methods. Formal neuropsychological testing and the Ped-ANAM were performed on 27 children with SLE who had not been previously diagnosed with neuropsychiatric SLE. Performance when completing the 10 Ped-ANAM tests was based on accuracy (AC), mean time to correct response, coefficient of variation of the time required for a correct response (CVc), and throughput. Formal neuropsychological testing was used as a criterion standard for diagnosing neurocognitive dysfunction (NCD; yes/no). Results. NCD was common and present in 16 (59%) of 27 participants. Ped-ANAM performance parameters were often moderately correlated with the Z scores on formal neuropsychological testing. The NCD group differed significantly (P < 0.05) from the normal cognition group in 3 Ped-ANAM tests: CVc with mathematical processing (MTH-CVc), AC with continuous performance test (CPT-AC), and CVc with spatial processing (SPD-CVc). Areas under the receiver operating curves (AUCs) ranged between 0.75 and 0.84 when each of these parameters (CPT-AC, MTH-CVc, SPD-CVc) was used to identify NCD independently. The AUC was improved to 0.96 for the combined assessment. Conclusion. The Ped-ANAM has concurrent validity when used in children with SLE. Initial validation suggests that the Ped-ANAM could be a useful screening tool for NCD in children with SLE.

Psychopathology, Neurodiagnostic Testing, and Imaging

“…3 The identified frequency of cognitive impairment using the ANAM was 69% in these adult patients with SLE 3 and 59% in a study of childhood-onset SLE. 38 In another study of adult patients with SLE studied within 9 months of diagnosis, Petri and colleagues 39 found cognitive impairment in 21% to 61% of cases, depending on the stringency of the definition of impairment. Hanly and colleagues 9 also found a range of cognitive impairment ranging from 11% to 50%, compared with locally recruited healthy control subjects, again depending on the stringency of the decision rules.…”

Pathogenesis of neurocognitive and neuropsychiatric manifestations in childhood-onset lupus: an overview

“…This cohort was also significantly more likely to perform poorly more specifically in measures of visuoconstructional processing and short term memory than those childhood-onset SLE patients without cognitive dysfunction. 3 Further studies by Brunner et al, including one presented at the 2011 ACR Conference, have demonstrated the reproducibility and validity of Ped-ANAM, with recommendations for its use in clinical practice. 44,49 However, it still has not been formally compared to other standard battery tests and although potentially cheaper and more efficient than other battery tests, without the requirement of a trained psychologist to administer, it still remains only a research tool at present for most pediatric/ adolescent rheumatology departments.…”

“…Useful information on pediatric SLE has been gained using such tests as was shown in a study by Brunner et al in 2007, where 59% of the childhood-onset SLE patients assessed with standardized neuropsychometric tests had neurocognitive dysfunction. This cohort was also significantly more likely to perform poorly more specifically in measures of visuoconstructional processing and short term memory than those childhood-onset SLE patients without cognitive dysfunction.…”

“…They compared subjects with age-and ethnicity-matched healthy controls. The results were categorized using firstly the Brunner et al 3 Categorization method, of assessing 4 domains (memory, psychomotor speed, visual construction processing and attention/executive functioning) using 11 tests, then a method by Mikdashi et al,5 where NCI is dichotomized into impairment and cognitive decline (also with focal versus multifocal domain impairment). Finally they used the Muscal et al 42 method, which tests the ACR recommended seven domains in total, expanded to include academic achievement as a separate domain, using 19 individual tests.…”

“…However in children, NPSLE is probably more common and more severe, with some studies reporting a prevalence rate of up to 95%. [1][2][3][4][5] However, many studies have reported widely varying figures on the prevalence of NPSLE in both adults and children. The heterogeneity of this clinical subset of SLE complicates elucidation of its pathogenesis.…”