Memory assessment in illiterate patients: The diagnostic value of the TNI 93

Abstract

AbstractBackgroundAssessment of cognitive functioning in illiterate patients is challenging because only few validated tests are available. The “test des neuf images ‐93” (TNI‐93) is a short test (5 to 10 minutes) based on nine black and white drawings, that was developed to assess episodic memory in this specific population (Maillet, 2016). Subjects are asked to learn the nine pictures using the principle of encoding specificity. It has been shown that the TNI‐93 is a suitable test to detect dementia in the elderly low educated population, but it remains unknown whether it is specifically altered in Alzheimer’s disease (AD). The objective of the present study was to examine the diagnostic value of the TNI 93 and its sub‐scores for the screening and diagnosis of biologically confirmed AD in patients explored for cognitive impairment.MethodWe retrospectively collected data of all patients seen in the memory center of Avicenne Hospital between January 2009 and November 2019 who underwent lumbar puncture, brain imaging, and neuropsychological assessment including the TNI 93. We classified patients into two groups: AD/ Non‐AD (NAD) according to the criteria AT(N) proposed by the working group "National Institute on Aging ‐ Alzheimer's Association”(Jack, 2018). We then compared scores at the TNI‐93 between AD and NAD patients.ResultAmong the 113 patients included, 52 presented with AD. Age (mean: 67.5 ± 10.1), gender (49.6% women), and education (mean: 9.2 ± 5.5 years) did not differ between both groups. On the TNI‐93, AD patients were significantly more impaired than NAD patients in immediate recall [mean AD: 5.5; mean NAD: 7; p=0.003], free recall [mean AD: 3.51; mean NAD: 4.78; p=0.018] and total recall [mean AD: 5.53; mean NAD: 6.93; p = 0.031]. In the AD group, there were more intrusions both during encoding (mean AD: 1.23; mean NAD: 0,67; p= 0.056) and restitution (mean AD: 1.22, mean NAD: 0.34; p=0.007).ConclusionThe TNI 93 seems an interesting tool to discriminate between Alzheimer's disease and other causes of cognitive impairment notably in low educated population.