Keywords Tuberous Sclerosis - Neurocutaneous Syndromes - Neuropsychology - Cognition
Palavras-chave Esclerose Tuberosa - Síndromes Neurocutâneas - Neuropsicologia - Cognição
INTRODUCTION
Tuberous sclerosis complex (TSC) is an autosomal dominant genetic disorder with variable expressivity,[1 ]
[2 ] and an estimated prevalence of nearly 1:6 thousand live births.[2 ]
[3 ] The pathophysiology of the disease relates to pathogenic DNA variants in either the TSC1 (OMIM 605284; 9q34) or TSC2 (OMIM 191092; 16p13.3) genes, commonly identified in 80 to 90% of patients.[1 ]
[4 ]
[5 ]
Loss of TSC1 or TSC2 function disrupts the TSC1/2-TBC1D7 complex, often causing abnormal cell growth[6 ] that results in benign tumors in multiple organs, known as hamartomas,[7 ] which are most frequently found on the skin (hypomelanotic macules, facial angiofibromas, shagreen patch), the brain (subependymal nodules and subependymal giant-cell astrocytomas, SEGAs), the heart (cardiac rhabdomyomas), the kidney (angiomyolipomas), and the retina (hamartomas and achromia).[8 ]
Recently, the definite genetic diagnosis of TSC has been reaffirmed, and major and minor features were further reviewed for the clinical diagnosis. Among eleven major and seven minor features, a clinically-definite diagnosis can be established with the observation of the presence of two major features or one major feature plus at least two minor features. If either one major feature or two or more minor features are identified, a possible diagnosis is established.[2 ]
The neuropsychiatric manifestations commonly associated with TSC include learning disabilities and behavioral problems,[9 ] and they have led to the introduction of the term TSC-associated neuropsychiatric disorders – TAND.[10 ] The diversity of cognitive phenotypes is an important characteristic of TSC.[11 ]
[12 ] While some patients present intellectual disability (ID),[9 ] others present intellectual performance within average levels.[8 ]
[11 ]
[13 ] The latter group, however, can present deficits in specific cognitive functions.[11 ]
[14 ]
Once the diagnosis is established, patients with TSC should receive multidisciplinary clinical care. In this context, neuropsychological assessment is a valuable instrument for the identification of cognitive, behavioral, and emotional features, with the objective of studying their interrelationships.[15 ] The neuropsychological assessment can inform the development of clinical rehabilitation and educational programs.
The present study aimed to evaluate the neuropsychological profile of patients from Southern Brazil diagnosed with definite TSC, and to identify clinical and morphological factors that could have a significant impact on cognitive development in this cohort.
METHODS
Study design
We conducted an observational, analytical, cross-sectional study with prospective data collection from 2015 to 2017. The present study is part of a multiisciplinry research and was approved by the Ethics Research Committees of the university hospital of Universidade Federal do Paraná (UFPR) (CAAE: 45059915.8.0000.0096), and the Institute of Biosciences of Universidade de São Paulo (USP) (CAAE: 125729112.3.000.5464 and CAEE: 48259715.2.0000.5464), and it complied with the international ethical requirements for research involving human subjects. All of the patients and/or their families signed informed consent forms.
Study subjects
The subjects were selected from UFPR and other health and educational centers of the states of Paraná and Santa Catarina, in Southern Brazil. A definite clinical diagnosis of TSC was the only inclusion criterion. The exclusion criterion was the impossibility of answering a standard assessment of intelligence level or an adaptive behavior scale.
Neuropsychological study protocol
The standard neuropsychological assessment was composed of the following instruments: the Wechsler Intelligence Scale for Children – Fourth Edition (WISC-IV)[16 ] and for Adults – Third Edition (WAIS-III);[17 ] one version of the Snijders-Oomen nonverbal intelligence test, the SON-R 21 /2 -7[a];[18 ]the Psychological Battery for Attention Assessment (BPA);[19 ] the Rey-Osterrieth Complex Figure Test;[20 ] and the Wisconsin Card Sorting Test (WCST).[21 ] The Vineland Adaptive Behavior Scales[22 ] were administered to parents/caregivers when the participant did not answer the standard assessment due to significant cognitive impairment.
The participants were screened for autism spectrum disorder (ASD) with the Childhood Autism Rating Scale (CARS).[23 ] The cut-off score for the present study was of 30 (30 to 36: mild-to-moderate symptoms; ≥ 37: severe symptoms). A CARS validation study[23 ] conducted in Brazil comprised participants aged from 3 to 17 years; however, there is evidence the scale can be used in adults as well.[23 ]
Clinical characteristics
Data on TSC diagnosis, medical history, seizures, imaging exams, and sociodemographic status were collected using semistructured interviews with the participants and their parents/caregivers, and medical records when available.
Genetic tests were performed by extracting DNA from peripheral blood samples of 55 participants to identify the TSC1 (NM_000368.4) or TSC2 (NM_000548.3) genes (Dufner-Almeida et al., in preparation).
Statistical analysis
The statistical analysis was based on the descriptive evaluation of the data and the multiple linear regression analysis. The following Wechsler scale index scores were selected as dependent variables: full-scale intelligence quotient (IQ), verbal comprehension index, working memory index, processing speed index, and perceptual reasoning index. The following independent variables were selected: gender, age, age at diagnosis, age at the onset of the first symptoms, number of medications used at the time of the assessment, presence and type of epileptic seizures, number of neurological lesions, presence of cortical tubers, presence of subcortical lesions, presence of SEGAs, the raw scores on the copy, short-term, and long-term memory of the Rey-Osterrieth Complex Figure Test, the raw scores on the BPA –general attention index, the number of perseverative errors (standard score) on the WCST, and the interaction between age and establishment of the diagnosis. The process of variable selection, in which variables with non-significant impact were removed in each new model, was applied aiming at an appropriate model that fitted well with the data. In all models, the behavior of the residuals was satisfactory, so the validity assumptions for the linear regression model were not violated in any of the cases. The Statistica 7 (TIBCO Software Inc., Palo Alto, CA, United States) software was used for the statistical analysis, and the significance threshold was set at 0.05.
RESULTS
Clinical characteristics of the participants
A total of 62 participants with a definite TSC clinical diagnosis was assessed; 36 of them were male (58.1%; [Table 1 ]), and the median age at enrollment was of 180.0 (41.0–458.0) months (range: 3 to 38 years). The median age at diagnosis was of 24.0 (0.0–360.0) months, and at the onset of the first sign or symptom, it was of 6 (0.0–108.0) months. Epileptic seizures were the most frequent cause that led participants to first seek TSC-related medical assistance (71%), followed by hypomelanotic macules (35.5%) and fetal cardiac rhabdomyomas. The median of the difference between age at first sign/symptom onset and age at diagnosis was of 6.0 (0.0–288.0) months. Among the 61 individuals submitted to DNA testing, 55 had pathogenic variants identified, 5 (9%) in the TSC1 gene and 50 (91%) in the TSC2 gene, leading to a TSC1 :TSC2 alteration ratio of 1:10 (Dufner-Almeida et. al., in preparation).
Table 1
Clinical and demographic data of the sample
Variable
n = 62
(%)
Age: median (months)
180.0
Range
(41.0–458.0)
Gender
Male
36/62
(58.1)
Female
26/62
(41.9)
Positive family history
7/62
(11.3)
Hypomelanotic macules
60/62
(96.8)
Angiofibroma
52/62
(83.9)
Shagreen patch
26/62
(41.9)
“Confetti” skin lesions
22/62
(35.5)
Ungual fibromas
19/62
(30.6)
Fibrous cephalic plaque
17/62
(27.4)
Angiomyolipomas
22/50
(44.0)
Multiple renal cysts
10/50
(20.0)
Cardiac rhabdomyoma
17/48
(35.4)
Multiple retinal hamartomas
7/37
(19.4)
Retinal achromic patch
1/37
(2.7)
Source: Reproduced with permission from publisher.[40 ]
History of epilepsy was found in 56 (90.3%) participants, with a median age at seizure onset of 6 (0.0–144.0) months. Twenty-five (44.6%) had a history of West syndrome (WS). Among the 56 participants with epilepsy history, 38 (67.8%) had achieved seizure control at the time of the evaluation, 16 (28.6%) presented generalized onset seizures, 14 (25.0%) had focal aware seizures, 11 (19.6%) had focal to bilateral tonic-clonic seizures, and 3 (5.3%) had focal impaired awareness seizures.
The results of imaging exams were available for 61 (98.4%) participants: all of them (100%) presented subependymal nodules, 47 (77.0%), cortical tubers, 32 (52.4%), subcortical lesions, 11 (18.0%), SEGAs, and 8 (13.0%), migration lines.
Neuropsychological profile
A total of 39 participants (62.9%; age: 89–458 months) answered the standard assessment of intelligence level (Wechsler scales or SON-R). However, some patients completed only part of the other tests of the battery, as follows: the BPA was completed by 27 participants, the Rey-Osterrieth Complex Figure Test, by 33, and the WCST, by 20 participants ([Table 2 ]). The remaining 23 patients (37.1%; age: 41–262 months) were assessed with the Vineland Adaptive Behavior Scales; all presented features of global intellectual delay, suggestive of ID.
Table 2
Standard assessment results from the individuals with definite diagnosis of tuberous sclerosis complex
Wechsler scales (n = 36)
Index
Score: mean ± SD
Full-scale IQ
81.7 ± 17.6
Verbal comprehension index
83.8 ± 15.5
Perceptual reasoning index
86.5 ± 16.5
Working memory index
79.9 ± 18.1
Processing speed index
87.8 ± 18.5
SON-R 21 /2 -7[a] (n = 3)
Index
Score: mean ± SD
Total IQ
116.3 ± 17.8
Reasoning scale (raw score)
109.0 ± 12.2
Execution scale (raw score)
120.0 ± 25.2
Rey-Osterrieth Complex Figure Test (n = 33)
Index
Score: median (min.–max.)
Copy
20.0 (1.5–34.0)
Short-term memory
8.0 (0.5–60.0)
Long-term memory
7.5 (0.5–16.0)
BPA (n = 27)
Index
Score: median (min.–max.)
General attention
171.0 (25.0–313.0)
Sustained attention
61.0 (9.0–118.0)
Divided attention
50.0 (-22.0–109.0)
Alternate attention
64.0 (30.0–113.0)
WCST (n = 20)
Index
Score: mean ± SD
Total correct
60.1 ± 20.3
Total errors
79.3 ± 15.7
Perseverative errors
76.2 ± 20.2
Number of categories completed
2.9 ± 1.9
Abbreviations: BPA, Psychological Battery for Attention Assessment; IQ, intelligence quotient; max., maximum; min., minimum; SD, standard deviation; SON-R 21 /2 -7[a], a version of the Snijders-Oomen non-verbal intelligence test; WCST, Wisconsin Card Sorting Test.
Source: Reproduced with permission from publisher.[40 ]
In the analysis of the entire sample (n = 62), 31 participants (50.0%) scored as presenting ID, 8 were detected by the Wechsler scales, and 23 individuals were assessed by the Vineland Adaptive Behavior Scales. Other individuals assessed by Wechsler scales were classified as follows: 8 (12.9%) as borderline, 20 (32.2%) presented scores within average (lower average, average or upper average), and 3 (4.8%) had above-average scores (superior or very superior).
In total, 17 subjects (27.4%) had CARS results suggestive of ASD. Concerning this subgroup, 7 (41.2%) individuals presented mild or moderate symptoms, and 10 (58.5%) presented severe symptoms. The significant results of the multiple linear regression analysis are presented in [Table 3 ].
Table 3
Significant results of multiple linear regression models
Index
Variable
Estimated (β)
p -value
Full scale IQ
Presence of epileptic seizures
-26.98
0.01*
Age at diagnosis
-0.43
0.01*
General BPA (raw score)
0.11
0.01*
VCI IQ
Presence of epileptic seizures
-23.55
0.03*
Age at diagnosis
-0.43
0.01*
General BPA (raw score)
0.08
0.04*
WMI IQ
Presence of epileptic seizures
-31.25
0.00*
Age at diagnosis
-0.43
0.01*
General BPA (raw score)
0.07
0.05*
PSI IQ
Presence of epileptic seizures
-16.78
0.01*
General BPA (raw score)
0.21
0.00*
PRI IQ
ROCF – copy
0.84
0.00*
Abbreviations: BPA, Psychological Battery for Attention Assessment; IQ, intelligence quotient; PSI, processing speed index; PRI, perceptual reasoning index; ROCF, Rey-Osterrieth Complex Figure; VCI, verbal comprehension index; WMI, working memory index;.
Note: Statistically significant results (p ≤ 0.05).
Source: Reproduced with permission from publisher.[40 ]
Overall, 10 out of 25 participants with WS answered the standard assessment and demonstrated a distribution of density plots similar to the group without the syndrome regarding the full-scale IQ, verbal comprehension index, perceptual reasoning index, and working memory index results. However, in relation to the processing speed index, the results of most of the sample with WS were between 60 and 80, while in the group without WS the scores ranged from 80 to 100 ([Figure 1 ]).
Abbreviations: IQ, intelligence quotient; WS, West syndrome. Notes: 0–group without WS; 1–group with WS. Source: Reproduced with permission from publisher.[40 ]
Figure 1 Distribution of processing speed index in the groups of participants with and without WS.
DISCUSSION
The present study aimed to determine the neuropsychological profile of patients from Southern Brazil clinically diagnosed with definite TSC, and to identify the factors with a significant impact on cognitive development in this cohort. In a previous study by our group,[24 ] a high prevalence of epilepsy (95%), cognitive impairment (45%), and psychiatric manifestations was found, which is consistent with the descriptions in the literature. Intellectual impairment has been associated with a greater number of drugs used to control epilepsy and autism (p < 0.05).
Regarding clinical features, epileptic seizures were the symptom that most frequently led participants to seek medical assistance. Like our previous study,[24 ] almost the totality of the sample presented epilepsy. Subependymal nodules were the most frequent brain lesion, and radial migration lines were the least prevalent. Among the participants with genetic alterations identified (55/61), most (91%) had TSC2 pathogenic alterations. Although mutations in the TSC2 gene have been described more frequently than in the TSC1 gene among the most severe cases of TSC,[25 ] our cohort was not predominantly composed of severe cases ([Table 2 ]). However, and in agreement with our work, the TSC2 pathogenic variant is at high risk of significant developmental delays.[26 ]
In relation to cognitive characteristics, the sample followed a bimodal distribution of intellectual level, with 50% of the participants presenting ID. This bimodal pattern was pointed out by other authors.[27 ] Other studies,[28 ]
[29 ] however, have suggested a unimodal distribution. The presence of epileptic seizures, the age at diagnosis, and attention were the factors that influenced the intelligence indexes the most ([Table 3 ]).
In the present study, the frequency of subjects with epilepsy diagnosis (90.3%) was similar to that described by Jansen et al. (90%),[30 ] but slightly higher than that described by Humphrey et al. (80%).[31 ] The presence of epileptic seizures had a statistically significant negative impact on the full-scale IQ, verbal comprehension index, and working memory index ([Table 3 ]). This finding indicates that, in this group of participants, IQ tends to be lower, and that epileptic seizures may influence cognitive development. These results are in accordance with what has been pointed out by other authors regarding the presence of epilepsy.[30 ]
[32 ]
[33 ]
[34 ] A significant part of the participants with WS (16/25; 64%) showed some degree of ID, corroborating what other studies have described.[30 ]
[33 ] On the other hand, the subgroup of WS participants who answered the standard assessment (n = 10) had a performance in the Wechsler scale indexes similar to non-WS individuals who also completed the standard assessment, except with regard to the processing speed index.
A tendency towards better results in the perceptual reasoning index and the processing speed index was observed in the standard assessment. Ridler et al.[35 ] demonstrated that a group of patients with TSC showed fewer deficits in inhibition, processing speed, and visuospatial abilities in comparison to other cognitive functions. Diversely, de Vries et al.[13 ] found low performance in processing speed measures, and Jeste et al.[36 ] identified a specific deficit in non-verbal IQ in a TSC group, which could be secondary to problems in visual reception and fine motor function. The significant impact of visual perception, evaluated by the copy of the Rey-Osterrieth Complex Figure test, on the perceptual reasoning index reinforces the importance of visual information reception for posterior execution.
Age at diagnosis also had a significantly negative impact on the full-scale IQ, verbal comprehension index, and working memory index, highlighting that, the earlier the diagnosis was established and the treatment implemented, the better were the results in these domains. Van Eeghen et al.,[32 ] studying the association between epilepsy variables and development results, suggested that early diagnosis and treatment of epilepsy favored prevention of encephalopathic damages as well as improved developmental outcomes in patients with TSC.
The statistically significant positive impact observed of the general attention score on most of the dependent variables shows an association with performance in general intelligence, verbal comprehension, working memory, and information processing speed. Lezak et al.[15 ] classify attentional functions as mental activity variables that maintain the activity of other cognitive functions. This understanding reinforces the importance of care in cognitive development and in the daily activity of patients with TSC. In the present study, the lowest median score on the BPA was observed in divided attention ([Table 2 ]). A similar result was found in another study,[13 ] which described consistent impairment in divided attention in children with TSC.
The number of participants that screened positive for ASD in our cohort is consistent with the evidence in the literature.[37 ] It is important to assess ASD with caution and precision due to the high percentage of individuals with TSC that present indicative signs of the disorder.
Poor performance was identified in visual perception and short- and long-term visual memory, as well as in the working memory index (in which the subjects presented the lowest index mean). Ridler et al.[35 ] described significant impairment in different tests of long-term memory, verbal working memory, and spatial working memory in TSC patients.
In addition, the low scores on the WCST indicated a deficit in cognitive flexibility, which is part of executive functioning and contributes to involvement in complex, independent, intentional, and self-directed behaviors.[15 ] It is also worth mentioning the clinical qualitative observations. Adult participants showed difficulty in maintaining jobs or finishing graduation courses. Children demonstrated difficulty in changing inadequate behaviors, in flexibility during tasks, and in organizing their activities. These characteristics are suggestive of deficits in organization, planning, self-regulation, and goal setting. The difficulties observed in social skills, including making friends, may be related to executive functioning impairment.
Corroborating the outcomes of the present study, de Vries and Watson[38 ] emphasize neuropsychological deficits in attention and executive functions in TSC patients, even when the global intellectual abilities are within normal limits, and there are no diagnostic criteria for other developmental disorders, such as attention-deficit hyperactivity disorder, considering this clinical group. The attentional, memory, and executive function deficits described in the present research are consistent with frontal lobe impairment at the level of the third functional unit, according to Luria's theory.[39 ] Although we had limited exploration of neuroimaging data, the functional deficits observed in the current report were similar to those observed in previous[11 ] studies that revealed a high predominance of TSC cortical lesions in the frontal lobe.
For individuals diagnosed with TSC, it is recommended that visual and practical resources be available in academic and professional contexts to facilitate learning. In the school setting, the time to complete tasks should be flexible, and short-term objectives should be proposed within long-term activities. Strategies to organize daily activities should be encouraged, such as the regular use of calendars, planners, and reminders. Playtime with peers and group activities are also important for cognitive and social development.
Since TSC has a wide variety of expressions, it is difficult to propose a single neuropsychological assessment plan for the whole group. It is indicated that the first evaluation prioritizes the differential diagnosis of ID using a combination of instruments that assess functioning, and brief instruments for global cognitive performance. Further assessments of specific cognitive functions, such as attention, long- and short-term visual and verbal memory, and visual perception, should require specific tests.
The present study has certain limitations. Firstly, the small cohort of children and younger participants, including infants and toddlers. It is suggested that age-appropriate instruments for cognitive evaluation be used, as well as the performance of an investigation of social skills and other variables, such as lifestyle characteristics, familial context, and substance use. The neuropsychological assessment of individuals with TSC enables the identification of the cognitive deficits and strengths of this population. Furthermore, early assessment and intervention can encourage this group of individuals to step out of an unprivileged intellectual condition that is associated with social and academic impairment. Secondly, we did not use neuroimaging to understand injuries related to cognitive development. Finally, more accurate instruments were not used for the diagnosis of ASD and ID.
In conclusion, in the present study, we have observed clinical symptoms similar to those reported in the literature, and we have pointed out that the occurrence of epileptic seizures and a higher age at diagnosis contribute to higher impairment in the domains of cognitive development. The study reinforces the importance of early diagnosis, mainly aimed at the prevention of epileptic seizures or their rapid control. In addition, emphasis on the development of attentional skills, visual perception, and executive function is recommended.
Bibliographical Record
