Acurácia diagnóstica do CompCog: tempo de reação como uma medida de rastreio para comprometimento cognitivo leve

Hartle, Larissa; Martorelli, Marina; Balboni, Giulia; Souza, Raquel; Charchat-Fichman, Helenice

Acessibilidade / Reportar erro

Brasil

Arquivos de Neuro-Psiquiatria

Español English

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Article • Arq. Neuro-Psiquiatr. 80 (6) • Jun 2022 • https://doi.org/10.1590/0004-282X-ANP-2021-0099 copiar

Acurácia diagnóstica do CompCog: tempo de reação como uma medida de rastreio para comprometimento cognitivo leve

Autoria SCIMAGO INSTITUTIONS RANKINGS

RESUMO

Antecedentes

O tempo de reação é afetado em diferentes condições neurológicas, mas não foi muito investigado considerando todos os tipos de comprometimento cognitivo leve (CCL).

Objetivo

Este estudo investigou a acurácia diagnóstica do CompCog, uma bateria computadorizada de rastreio cognitivo focada em medidas de tempo de reação.

Métodos

Uma amostra de 52 idosos passou por uma avaliação neuropsicológica, incluindo o CompCog, e uma consulta médica para serem classificados como grupo controle ou serem diagnósticos com CCL. A acurácia do teste para distinguir entre os dois grupos foi calculada.

Resultados

Os resultados das análises de acurácia diagnóstica mostraram AUC das curvas ROC tão altas quanto 0,915 (CI 0,837-0,993). O subteste com maior sensibilidade e especificidade - subteste de tempo de reação de escolha - apresentou sensibilidade de 91,7% e especificidade de 89,3%. O modelo final de regressão logística classificou corretamente 92,3% dos indivíduos, com especificidade de 92,9% e sensibilidade de 91,7%, e incluiu apenas 4 variáveis de diferentes subtestes.

Conclusões

Em resumo, o estudo mostrou que o tempo de reação avaliado pelo CompCog é uma boa medida de rastreio para diferenciar entre envelhecimento normal e CCL. Medidas de tempo de reação em milissegundos se mostraram melhores que o número de respostas corretas. O teste pode fazer parte de testes clínicos de rotina para atingir o objetivo de rastrear o CCL, indicar outros procedimentos para investigação e diagnóstico e planejar intervenções.

Palavras-chave:
Disfunção Cognitiva; Tempo de Reação; Diagnóstico; Demência; Envelhecimento Cognitivo

Test	Cognitive functions involved and how they are evaluated	Variables
Simple Reaction Time (SRT)	Processing speed. As soon as a white square appears in the middle of the screen, the person should touch the rectangle at the bottom of the screen.	Median reaction time
Choice Reaction Time (CRT)	Processing speed. As a white or orange square appears in the middle of the screen, the person should touch the rectangle of the same color at the bottom of the screen.	Median reaction time; Correct responses; Revised median reaction time (choice reaction time - simple reaction time).
Implicit Learning Test (ILT)	Implicit learning. As one of ten gray squares distributed in the screen turns white, the person should press it. There is a fixed sequence of 25 squares that is repeated four times and one last random sequence.	Median reaction time in each of five tasks; Implicit learning (median reaction time in sequence 4/median reaction time in sequence 1).
Visual and Spatial Short-Term Memory (STM)	Working memory. There are ten gray squares distributed on the screen. One will become white at a time, making a sequence that should be reproduced.	Correct responses; Direct order SPAN; Median reaction time in direct order; Inverse order SPAN; Median reaction time in inverse order.
Face Recognition and Memory (FRM)	Episodic memory. Ten drawings of unknown faces are presented for 30 seconds. The participant should then choose from among ten pairs of faces, the one that was among those initially shown for memorization, in four attempts.	Correct responses and median reaction time for each of the four tasks and for all tasks together.
Inhibitory Control Test (ICT)	Attention and inhibitory control. Squares of different colors will appear in the middle of the screen for one second each: the white ones should be avoided.	Median reaction time; Correct responses; Median reaction time for correct responses; Median reaction time for errors; Errors.
Stroop Test (StT)	Attention and inhibitory control. All tasks have four colored rectangles located at the bottom of the screen. The person should touch the one matching the stimulus that appears in the middle of the screen considering its color without distracters (task 1) and with distracters (tasks 2 and 3).	Interference; Median reaction time and errors for each of the three tasks.
Survey Test (ST)	Attention. Squares of different colors will appear in the middle of the screen for one second each. Participants should press the white ones in the first task, whites and blues in the second and also yellow ones in the third.	Median reaction time, correct responses, reaction time for correct responses, errors and reaction time for errors, for each of the three tasks.

Variable	CG (n = 28) Mean (SD), min-max	MCI (n = 24) Mean (SD), min-max	p-value
Sex*	23/5	17/7	0.335
Age	71.4 (5.7), 62-83	73.9 (6.9), 61-85	0.164
Number of years of education	14.1 (3.3), 6-17	11.6 (5.3), 3-18	0.052
Health problems	1.5 (1.0), 0-3	1.7 (1.4), 0-4	0.403
Medications in use	2.0 (1.9), 0-6	1.9 (2.0), 0-6	0.840
Depressive symptoms	2.8 (1.9), 0-6	3.6 (2.6), 0-9	0.224
Naming	10.0 (0.0), 10-10	9.8 (4.4), 8-10	0.123
Incidental memory	5.9 (1.2), 4-8	5.2 (1.5), 3-8	0.084
Immediate memory**	8.5 (1.1), 6-10	7.2 (1.4), 5-10	0.001
Learning**	9.3 (0.8), 8-10	8.3 (1.1), 6-10	0.001
Delayed recall**	8.6 (1.5), 4-10	7.2 (1.6), 4-10	0.005
Recognition	9.7 (0.4), 9-10	9.5 (0.9), 7-10	0.930
Clock drawing test	6.2 (2.2), 4-10	5.5 (2.5), 1-10	0.259
Verbal fluency**	20.7 (4.3), 13-30	16.6 (4.6), 8-26	0.002
MMSE30	27.0 (2.1), 22-30	25.7 (2.7), 21-29	0.067
Functionality**	20.9 (0.2), 20-21	19.6 (1.6), 15-21	< 0.001

Test	AUC, 95% confidence interval range	Cutoff point (milliseconds)	Sensitivity/specificity
Simple Reaction Time Test
MRT	0.506, 0.339-0.673		Sensitivity/specificity
Choice Reaction Time Test
MRT†	0.915, 0.837-0.993*	689.813	91.7%/89.3%
Correct responses	0.650, 0.498-0.801
Revised MRT	0.705, 0.559-0.852
Implicit Learning Test
MRT 1	0.839, 0.721-0.957*	688.125	83.3%/75%
MRT 2	0.836, 0.720-0.953*	651.531	75%/75%
MRT 3	0.823, 0.703-0.943*	616.159	79.2%/67.9%
MRT 4	0.829, 0.710-0.948*	618.139	75%/71.4%
MRT 5	0.804, 0.674-0.933*	664.784	75%/75%
Implicit learning	0.521, 0.361-0.681
Visual and Spatial Short-Term Memory Test
Correct responses	0.718, 0.577-0.859
Direct order SPAN	0.725, 0.588-0.861
MRT in direct order	0.774, 0.642-0.906*	643.635	75%/71.4%
Inverse order SPAN	0.868, 0.774-0.962*	3.5	95.8%/64.3%
MRT in inverse order	0.781, 0.648-0.915*	668.615	75%/67.9%
Face Recognition and Memory
MRT†	0.896, 0.799-0.993*	1580.791	83.3%/85.7%
MRT 1	0.823, 0.703-0.943*	1905.385	75%/71.4%
MRT 2	0.881, 0.781-0.981*	1506.316	91.7%/75%
MRT 3	0.872, 0.767-0.977*	1430.946	87.5%/75%
MRT 4	0.813, 0.693-0.932*	1486.333	79.2%/71.4%
Correct responses	0.811, 0.685-0.937*	98.750	79.2%/57.1%
Correct responses 1	0.763, 0.628-0.899
Correct responses 2	0.673, 0.537-0.838
Correct responses 3	0.688, 0.537-0.838
Correct responses 4	0.757, 0.619-0.894
Inhibitory Control Test
MRT	0.884, 0.782-0.976*	664.447	87.5%/78.6%
CAMRT	0.871, 0.774-967*	663.279	83.3%/78.6%
EMRT	0.753, 0.615-0.891
Correct responses	0.794, 0.664-0.924*	96.5	75%/75%
Errors	0.794, 0.664-0.924*	3.5	75%/75%
Stroop Test
MRT 1	0.847, 0.732-0.962*	814.839	87.5%/75%
MRT 2	0.799, 0.672-0.926*	911.100	83.3%/67.9%
MRT 3	0.743, 0.604-0.881
Errors 1	0.587, 0.430-0.744
Errors 2	0.541, 0.378-0.704
Errors 3	0.525, 0.336-0.684
Interference	0.506, 0.343-0.669
Survey Test
MRT 1	0.818, 0.703-0.934*	637.844	79.2%/71.4%
Correct responses 1	0.648, 0.497-0.799
CAMRT 1	0.818, 0.703-0.934*	637.844	79.2%/71.4%
Errors 1	0.648, 0.497-0.799
EMRT 1	0.465, 0.307-0.623
MRT 2	0.829, 0.714-0.944*	663.004	79.2%/75%
Correct responses 2	0.798, 0.676-0.919*	97	79.2%/71.4%
CAMRT 2	0.835, 0.721-0.949*	663.004	79.2%/75%
Errors 2	0.798, 0.676-0.919*	3	79.2%/71.4%
EMRT 2	0.573, 0.415-0.731
MRT 3	0.823, 0.707-0.939*	653.629	83.3%/71.4%
Correct responses 3	0.802, 0.678-0.926*	93	75%/75%
CAMRT 3	0.826, 0.711-0.941*	664.223	83.3%/75%
Errors 3	0.802, 0.678-0.926*	7	75%/75%
EMRT 3	0.507, 0.346-0.669

Academia Brasileira de Neurologia - ABNEURO R. Vergueiro, 1353 sl.1404 - Ed. Top Towers Offices Torre Norte, 04101-000 São Paulo SP Brazil, Tel.: +55 11 5084-9463 | +55 11 5083-3876 - São Paulo - SP - Brazil
E-mail: revista.arquivos@abneuro.org

Acompanhe os números deste periódico no seu leitor de RSS

[1] Correspondence: Larissa Hartle; Email: larissahartle@gmail.com.