Original Article
Objective
:
To use Cambridge Neuropsychological Test Automated Battery (CANTAB) test battery to assess the effects of discontinuation of treatment with methylphenidate on the neuropsychological performance of children with attention deficit hyperactivity disorder (ADHD) and to compare this performance with normative data.
Methods: Fifteen boys meeting criteria for Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV ADHD aged between 4.5-14.6 yrs were selected. The CANTAB test battery was conducted after discontinuation of methylphenidate for a minimum of 24 hours and was repeated one week after the recommencement of treatment.
Results: Performance differences between the unmedicated/medicated groups were found on the pattern recognition memory task (F=0.37, p=0.041) and intra/extra-dimensional (IED) Set-Shifting task [number of stages completed (z=-4.572, p=0.001) and total errors (F= 1.36, p=0.046)]. In the unmedicated group, total errors made on IED Set-Shifting correlated with a lower strategy score on the Spatial Working Memory (SWM) task (r=0.518, p=0.048). In the medicated group, greater Spatial Span Length correlated with fewer "between search" errors made on the SWM test (r=0.657, p=0.008).
Conclusion: Discontinuation of methylphenidate impairs performance on the CANTAB test battery in children with ADHD. These impairments, primarily in executive function, could be indicative of dysfunction in fronto-striatal networks, that methylphenidate can improve through manipulation of catecholaminergic pathways in the brain.
Correspondence : Peter Hoare, DM, FRCPsych, Department of Child Psychiatry, Royal Hospital for Sick Children, Edinburgh EH9 1LF, Scotland, UK
Tel : +44. (0) 131-536-0516, Fax : +44. (0) 131-536-0545, E-mail : P.Hoare@ed.ac.uk
Introduction
Attention deficit hyperactivity disorder (ADHD)1 is a common childhood behavioural disorder characterised by difficulties in inhibitory control, short attention span, behavioural and cognitive impulsivity and inappropriate restlessness. The Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV criteria divide ADHD into three subtypes: predominantly hyperactive/impulsive type; inattentive type; and combined type. ADHD is commoner in boys (9%) than girls (3.3%),2 affects children from an early age, can lead to underachievement at school3 and poor social functioning.4 Stimulant medications such as methylphenidate are effective in ameliorating the symptoms of ADHD, but there is little evidence about the effect of psycho-stimulant medication on neuropsychological performance.
Methylphenidate and dexamphetamine promote the release of the catecholamines, dopamine and noradrenaline, through direct action on dopaminergic/noradrenergic neurones, whilst also preventing the re-uptake of dopamine through their action on the dopamine transporter.5 At the low therapeutic doses needed to treat ADHD, methyl-phenidate raises extracellular dopamine several fold, but reduces the amount of dopamine released with each nerve impulse, which results in less activation of the post-synaptic dopamine receptors that drive psychomotor activity.
A dysfunction of the pre-frontal lobe and fronto-striatal pathways, which control inhibition and executive function, has been proposed as an explanation for the symptomatology seen in ADHD.6 Frontal lobe function includes the ability to orchestrate activity within sensory, perceptual and motor networks7 with the pre-frontal cortex modulating working memory e.g. the ability to retain information in order to achieve future goals.8 There is increasing evidence highlighting the modulatory effects of catecholamines in these areas resulting in a subsequent increase in cognitive performance.9
Computerised tests, such as continuous performance tests (CPTs), have been used in the assessment of ADHD.10 Improvements not attributable to practice effects occur when performance on these tests is measured after psycho-stimulant medication.11 However, because these tests only measure one parameter of cognitive ability and children with ADHD demonstrate difficulties in tasks requiring planning, working memory and inhibitory control, a test battery approach to assessment would be more informative.
The Cambridge Neuropsychological Test Automated Battery (CANTAB) is a computerised neuropsychological battery with subtests measuring motor skill, visual attention, memory and working memory. It assesses executive functions and includes measures of planning, Set-Shifting, Spatial Working Memory (SWM) and non-verbal memory span. All task stimuli are non-verbal, consisting of geometric designs or simple shapes and language proficiency is needed only to understand the instructions prior to each task. Information about neural substrates underlying brain dysfunction provided by functional magnetic resonance imaging (fMRI) and positron emission tomography (PET)12 indicate that CANTAB is sensitive to the presence of brain dysfunction in adults. CANTAB can be used in children without alteration.13,14
The majority of research using CANTAB with children has emphasised executive function or frontal lobe assessment. Unmedicated children with ADHD show deficits in tests measuring short term memory capacity, visual memory, attention and abstract reasoning, which can be ameliorated using psycho-stimulant medication.15
This study intends to establish whether the neuropsychological performance of medicated children with ADHD is affected after discontinuation of methylphenidate. It also aims to compare the neuropsychological performance of unmedicated and medicated children with available normative data.
Methods
Fifteen boys meeting criteria for DSM-IV ADHD1,3 aged between 4.5-14.6 yrs were selected from children attending psychiatric services in Edinburgh over a 12 week period. The diagnosis was made by child psychiatrists from the department of Child and Family Mental Health, Royal Hospital for Sick Children, Edinburgh.
The modified Swanson, Nolan and Pelham (SNAP)-IV Questionnaire16(Appendix 1) was completed by parents prior to each test session to provide an indication of problem behaviours whilst on and off medication. This 26 item questionnaire comprises the diagnostic criteria for ADHD and oppositional defiant disorder (ODD) according to DSM IV. The eighteen ADHD symptoms can be divided into inattentive, hyperactivity/impulsive or combined type. Each item was scored from zero to three so that the range for the inattentive items was 0-27, 0-27 for the hyperactive/impulsive and 0-24 for ODD. For convenience these were converted to percentages for each maximum. A score of 50% or above was regarded as clinically significant, a score similar to that suggested by Swanson.16 Each child completed the British Picture Vocabulary Scale (BPVS) at the initial session.17 The BPVS has been shown to be a good proxy for general intelligence.
Prior to the study, all subjects had been stabilised on methylphenidate for at least three months. Discontinuation of methylphenidate has been shown to produce no significant 'rebound' effects.18 Before the initial test session subjects were asked to discontinue their medication for a minimum of 24 hrs, approximately 5 half-lives of methylphenidate.19 The second test session was conducted after methylphenidate had been recommenced for one week.
Normative data
Whilst this study aims to establish whether methylphenidate has an effect on cognitive function in children with ADHD, it would be useful to compare our results with those of a representative group of children without psychiatric disorder. The reliability of CANTAB has been demonstrated in children without any psychiatric history.20 The normative data for present study was provided from the CANTAB database.
Cognitive tests (Table 1)
Each child was given a series of tests from the CA-NTAB battery, including the core tests. To ensure tests were presented in the same manner, a standard protocol comprising a simple explanation of each test was used. The battery contained the following tests: spatial span, SWM, pattern recognition memory, spatial recognition memory, delayed matching to sample, intra/extra-dimensional (IED) Set-Shifting, reaction time. During the first session, a motor screening test designed to familiarise the children with the touch screen computer preceded the battery.
Repeated testing
Parallel versions of the tests were used at the second test session to eliminate practice effects. Re-test reliability studies for children are not yet available, but the stability coefficients for CANTAB's measures of executive function in adults are moderate in magnitude and range from 0.6 to 0.7.21
Statistical analysis
This data was analysed using Statistical Package for the Social Sciences (SPSS).22 Outcome measures on the CANTAB for the unmedicated group, the medicated group and the normative data were compared using parametric or non-parametric tests as appropriate.
Possible associations between baseline neuropsychological performance and the cognitive measures in both the medicated and unmedicated group were investigated using Pearson's product moment correlation coefficient.
Results
BPVS: Table 2 summarises the children's scores on the BPVS. They show that the subjects had scores approximately two years behind their chronological age.
Modified SNAP Questionnaire (Table 3): Results showed the majority of children were rated in the clinically significant range using Swanson's criteria.16 Only one child, AD5, scored higher on medication rather than off medication. Comparison of the mean score on/off medication showed that there were significant differences on all three measures of ADHD on the SNAP questionnaire, but not on the ODD measure (Table 4).
CANTAB Results: Three tests (spatial recognition, spatial span, delayed matching to sample) showed no significant differences, whilst the other tests were significant (Table 5).
Spatial recognition
There were no statistical differences found between the medicated and unmedicated groups (F=1.99, p=0.4) or when these groups were compared with the normative data (F=0.91, p=0.64 and F=2.17, p=0.08) respectively.
Spatial span
There was no statistical difference for maximum forward span reached between medicated group (5.12) and normative data (5.2), F=0.26, p=0.61.
Delayed matching to sample
No differences between the medicated and unmedicated groups were found for either the percentage correct "all delays" or the percentage correct "simultaneous" (F=0.03, p=1.35) and (F=0.702, p=0.581) respectively. Comparisons between the normative data and the unmedicated or medicated groups were not done because of the large variance in the normative data.
Table 5 summarises the results for the three groups: medicated, unmedicated and normative data.
Pattern recognition
A significant difference was found between the unmedicated and medicated groups (F=0.37, p=0.041), but not between the unmedicated group and the normative data (F=0.09, p=0.64) or between the medicated group and the normative data (F=2.17, p=0.089).
Spatial Working Memory
Unmedicated children were found to display significantly poorer strategy scores when compared with the normative data (F=2.41, p=0.05) but this difference was not reduced by medication because no difference was found between medicated and unmedicated children (F=1.69, p=0.775). For the performance measure "between search error", there was no difference found between the medicated and unmedicated groups (F=0.14, p=0.57) or when either of these groups were compared with the normative data (F=3.64, p=0.6 and F=4.27, p=0.34) respectively.
Intra/extra dimensional Set-Shifting
The number of stages completed on this test was assessed non-parametrically with the Wilcoxon signed-rank sum test. This showed a significant performance difference between the medicated and unmedicated groups, 8.25 and 7.95 respectively (z=-4.572, p=0.001).
The total numbers of errors made during this task were also compared using t-tests. There was a significant difference between the unmedicated and medicated groups (F=1.36, p=0.046), but no difference between either unmedicated or the medicated group and the normative data (F=1.36, p=0.65 and F=0.563, p=0.224) respectively. In summary, children on medication performed better, completing more stages and making fewer errors.
Correlation analysis
Correlation analysis was performed to examine possible associations between age, Spatial Span Length and the modified SNAP score. As many correlations were carried out, the significance criterion was increased to p=0.01 in order to reduce the possibility of chance association. Only one association was found, namely age and Spatial Span Length (r=0.507, p=0.007).
The SWM task was compared with other CANTAB tests. In the unmedicated group, there was a correlation between strategy score on the SWM task and total errors made on the IED Set-Shifting test, namely the fewer the stages completed on the IED Set-Shifting test the lower the strategy score on the SWM test (r=-0.518, p=0.048). In the medicated group there was a correlation found between the number of "between search" errors made on the SWM task and Spatial Span Length (r=0.657, p=0.008), namely the higher the spatial length the fewer the "between search" errors.
Discussion
Modified Swanson, Nolan and Pelham questionnaire
Results showed clearly parents rated their children as more problematic after discontinuation of medication, with the exception of one child, AD5. This child's behaviour was particularly problematic on the day of testing, even though he had taken the medication.
British Picture Vocabulary Scale
It has been suggested that children with ADHD show a developmental "lag" of approximately two years behind their age matched cohorts.23 This is confirmed again in the present study, as the subjects' BPVS results were on average two years less than their chronological age.
Cambridge Neuropsychological Test Automated Battery Tests
The results indicate that methylphenidate improved the performance on the tests of IED Set-Shifting and pattern recognition. Unmedicated children also showed impairments on the SWM test. These deficits in SWM were shown to correlate with both a lower Spatial Span Length and impairment in IED Set-Shifting. Taken together, these results imply that some domains of executive function are disrupted in children with ADHD. This finding is similar to previous studies.15 Another consistent finding was the correlation between Spatial Span Length and age. This is consistent with the view that spatial memory capacity develops in a stepwise linear fashion.14,20
The better results, evident after discontinuation of medication for only 24 hrs, strongly suggest that this is due a disruption of catecholamine activity rather than any change in brain structure. SWM and IED Set-Shifting tests are both sensitive to catecholaminergic stimulation,24 and are impaired following frontal lobe damage.25
Results from the IED Set-Shifting test also provide further evidence about some deficits in executive function. During the extra-dimensional (ED) component of this test, it is necessary to integrate information from previous trials in order to identify the correct shape. This puts a greater demand on working memory and attention than previous stages of the test. This additional demand is reflected in the results showing a correlation between the fewer stages completed on the IED Set-Shifting test and a lower strategy score on the SWM test. Moreover, the present study found that unmedicated children with ADHD were able to progress beyond the ED shift stage after they had received methylphenidate, thereby providing further support for the role of medication in improving some aspects of executive functioning.
In the present study, the correlation between age and Spatial Span Length (spatial memory) was consistent across the three groups, suggesting that spatial memory and spatial working memory improve with age and that relationship is similar for children with ADHD and also normal children. However, significant differences were found between the groups on the SWM strategy score task even when age was controlled for. This suggests that SWM deficits in children with ADHD persist across the age range, but may be reduced by methylphenidate.
Previous research suggests that deficits in SWM may be due to frontal lobe dysfunction.25 The poorer results of the unmedicated children in this study support this suggestion, as they were unable to develop systematic strategies on tests of spatial working memory. Previous research has also suggested that cognitive prefrontal functions, which are necessary for strategy development, are not fully developed in children below the age of 12 years.20 Consequently, this makes it difficult to interpret the strategy results in this study, as the age range of children was wide (4.5-14.6 years).
The results do however show that methylphenidate did not affect all SWM tasks, for example performance on the Spatial Recognition Memory test did not differ between the groups.
This present study also found that certain cognitive measures, such as Spatial Span Length, are predictive of improvement on CANTAB tests following psycho-stimulant medication. For instance, medicated children with the highest Spatial Span Length made the fewest "between-search" errors on the SWM test. This is the opposite of that obtained for adults with ADHD.9 They found that adults with lower digit spans showed the most improvement after methylphenidate suggesting a dynamic model of catecholaminergic function. This supports the hypothesis that a hypo-dopaminergic state may exist in children with ADHD that is normalised by methylphenidate.26 Another explanation proposed27 is that psycho-stimulant medication may either normalise the low tonic levels of dopamine or block the elevated level of dopamine transporter.
Methodological issues
There are several issues potentially limiting the interpretation of findings in this study. It could be argued that the improvements found on the cognitive tests in medicated children are attributable to improved behaviour in general, as medicated children may be able to concentrate more easily. If this was the case, improvements would have been seen across all tests, not just with those of executive function. A general improvement on both speed and accuracy on the CANTAB tests would also have been evident.
The present results are from a small group of children all of whom had been previously established on individually tailored dose of methylphenidate. Some of the improvements may have been due to the "long-term" effects of methylphenidate, as medication was only discontinued for 24 hours in the study. However, the duration of discontinuation was at least five times the half-life of methylphenidate,19 so that this seems unlikely.
Finally, the study was limited to boys, so it is uncertain whether findings would extend to girls. The broad age span (4.5-14.6 years) could also have potentially affected data interpretation. A large prospective study focussing on changes in executive function over time would be beneficial.
Conclusion
Discontinuation of methylphenidate impairs neuro-psychological performance in children with ADHD. Deficits primarily in executive function are apparent after neuropsychological testing. These deficits are indicative of dysfunction in fronto-striatal networks. This study has provided some evidence that methylphenidate possibly improves these deficits through manipulation of catecholaminergic activity in the brain.
DSM IV. Diagnostic and Statistical Manual of Mental Disorders (4th Edition). Washington DC: American Psychiatric Association; 1994.
Szatmari P, Offord DR, Boyle MH. Correlates, associated impairments and patterns of service utilization of children with attention deficit disorder: findings from the Ontario Child Health Study. J Child Psychol Psychiatry 1989;30:205-217.
Cantwell DP, Baker L. Issues in the classification of child and adole-scent psychopathology. J Am Acad Child Adolesc Psychiatry 1988;27:521-533.
Weiss G, Hechtman L. Hyperactive Children Growing up: empirical findings and theoretical considerations (2nd ed). New York: Guildford; 1993.
Seeman P, Madras BK. Anti-hyperactivity medication: methylphenidate and amphetamine. Mol Psychiatry 1998;3:386-396.
Barkley RA. Attention-deficit/hyperactivity disorder, self-regulation, and time: toward a more comprehensive theory. J Dev Behav Pediatr 1997;18:271-279.
Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci 1986;9:357-381.
Baddeley AD. Working memory. Science 1992;255:556-559.
Mehta MA, Owen AM, Sahakian BJ, Mavaddat N, Pickard JD, Robbins TW. Methylphenidate enhances working memory by modulating discrete frontal and parietal lobe regions in the brain. J Neurosci 2000;20:RC65, 1-6.
Beck LH, Bransome ED Jr, Mirsky AF, Rosvold HE, Sarason I. A continuous performance test of brain damage. J Consult Psychol 1956;20:343-350.
Klorman R. Cognitive event-related potentials in attention-deficit disorder. J Learn Disabil 1991;24:130-140.
Robbins TW. Dissociating executive functions of the prefrontal cortex. In: Roberts AC, Robbins TW, Weiskrantz L, editors. The prefrontal cortex-Executive and cognitive Functions. New York: Oxford University Press; 1998.
Luciana M, Nelson CA. Neurodevelopmental assessment of cognitive function using the Cambridge Neuropsychological Testing Automated Battery (CANTAB): validation and future goals. In: Ernst M, Rumsey J, editors. The foundation and future of functional imaging in child psychiatry. Camgridge: Cambridge University Press; 2000.
Luciana M, Nelson CA. Assessment of neuropsychological function through use of the Cambridge Neuropsychological Testing Automated Battery: performance in 4- to 12-year-old children. Dev Neuropsychol 2002;22:595-624.
Kempton S, Vance A, Maruff P, Luk E, Costin J, Pantelis C. Executive function and attention deficit hyperactivity disorder: stimulant medication and better executive function performance in children. Psychol Med 1999;29:527-538.
Swanson J. School-based Assessments and Interventions for ADD Students. Irvine, Calif.: KC Publications; 1992.
British Picture Vocabulary Scale (BPVS). NFER-Nelson, Windsor, 1982.
Johnston C, Pelham WE, Hoza J, Sturges J. Psychostimulant rebound in attention deficit disordered boys. J Am Acad Child Adolesc Psychiatry 1988;27:806-810.
Gualtieri CT, Wargin W, Kanoy R, Patrick K, Shen CD, Youngblood W, et al. Clinical studies of methylphenidate serum levels in children and adults. J Am Acad Child Psychiatry 1982;21:19-26.
Luciana M, Nelson CA. The functional emergence of prefrontally-guided working memory systems in four-to eight-year-old children. Neuropsychologia 1998;36:273-293.
Lowe C, Rabbitt P. Test/re-test reliability of the CANTAB and IS-POCD neuropsychological batteries: theoretical and practical issues. Cambridge Neuropsychological Test Automated Battery. International Study of Post-Operative Cognitive Dysfunction. Neuropsychologia 1998;36:915-923.
SPSS Inc. Statistical Package for the Social Sciences (Version 11.5). SPSS Inc., Chicago, 2002.
Boucugnani LL, Jones RW. Behaviours analogous to frontal lobe dys-function in children with attention deficit hyperactivity disorder. Archives of Clinical Neuropsychology 1989;4:161-173.
Elliott R, Sahakian BJ, Matthews K, Bannerjea A, Rimmer J, Robbins TW. Effects of methylphenidate on spatial working memory and planning in healthy young adults. Psychopharmacology 1997;131:196-206.
Owen AM, Roberts AC, Polkey CE, Sahakian BJ, Robbins TW. Extra-dimensional versus intra-dimensional set-shifting performance following frontal lobe excisions, temporal lobe excisions or amygdalo-hippocampectomy in man. Neuropsychologia 1991;29:993-1006.
Servan-Schreiber D, Printz H, Cohen JD. A network model of catecholamine effects: gain, signal-to-noise ratio, and behavior. Science 1990;249:892-895.
Mehta MA, Goodyer IM, Sahakian BJ. Methylphenidate improves working memory and set-shifting in AD/HD: relationships to baseline memory capacity. J Child Psychol Psychiatry 2004; 45:293-305.