Functional MRI compliance in children with attention deficit hyperactivity disorder

PURPOSE

We aimed to test the effect of prescan training and orientation in functional magnetic resonance imaging (fMRI) in children with attention deficit hyperactivity disorder (ADHD) and to investigate whether fMRI compliance was modified by state anxiety.

METHODS

Subjects included 77 males aged 6–12 years; there were 53 patients in the ADHD group and 24 participants in the healthy control group. Exclusion criteria included neurological and/or psychiatric comorbidities (other than ADHD), the use of psychoactive drugs, and an intelligence quotient outside the normal range. Children were individually subjected to prescan orientation and training. Data were acquired using a 1.5 Tesla scanner and an 8-channel head coil. Functional scans were performed using a standard neurocognitive task.

RESULTS

The neurocognitive task led to reliable fMRI maps. Compliance was not significantly different between ADHD and control groups based on success, failure, and repetition rates of fMRI. Compliance of ADHD patients with extreme levels of anxiety was also not significantly different.

CONCLUSION

The fMRI compliance of ADHD children is typically lower than that of healthy children. However, compliance can be increased to the level of age-matched healthy control children by addressing concerns about the technical and procedural aspects of fMRI, providing orientation programs, and performing on-task training. In patients thus trained, compliance does not change with the level of state anxiety suggesting that the anxiety hypothesis of fMRI compliance is not supported.Functional magnetic resonance imaging (fMRI), which uses blood oxygen level-dependent contrast, is a noninvasive procedure for imaging regional brain activity. MRI exhibits high spatial resolution; even 1.5 Tesla (T) imaging used in standard clinical practice (spatial resolution of 2–4 mm) yields robust functional signal changes (1). MRI can be performed without the ethical concerns associated with the other available imaging techniques and can thus be used in children and in healthy populations. In healthy volunteers, fMRI has produced reproducible findings across scanning sites and age groups with respect to the localization and development of cognitive processes (2). Its capacity for noninvasive imaging of the brain in vivo during cognitive processing has made fMRI an exciting tool for laboratory research, as well as clinical studies and clinical practices that involve diagnosis, follow-up, and presurgical mapping (3, 4).A disorder that attracts a great deal of attention in children is attention deficit hyperactivity disorder (ADHD). This focused attention is partly due to the high incidence (0.2%–12.2%) of ADHD, which is also the most frequent diagnosis in children referred to child psychiatry departments (5–7). From the neuropsychological point of view, ADHD is associated with deficits in executive functions (8, 9). Nevertheless, as the number of theories on the subject demonstrates, ADHD remains an unresolved issue, especially with respect to its biological basis and brain correlates (10).With its many merits, fMRI would be a valuable tool for studying the etiology, diagnosis, and follow-up of ADHD patients. However, MRI is movement-sensitive, and movement artifacts impair the diagnostic quality of the examination and can even render the scans unusable. A meta-study involving 21 000 cases, reported an overall artifact frequency of 40% (11). Normal body pulsations accounted for 7%–12% of the artifacts, but at least 10% were due to motoric unrest or restlessness. In another study, artifacts other than normal body pulsations were reported in 12.8% of the scans and 6.4% of the scans were impaired in diagnostic quality (12).The symptoms of ADHD include hyperactivity, impulsivity, and inattention (13). Of these symptoms, the first two directly challenge the immobility requirement of MRI, and the third poses a problem for the task-appropriate responses that cognitive tasks require for functional imaging. Not unexpectedly, the fMRI compliance of ADHD patients is poor. In 7–12-year-old unmedicated ADHD patients, the frequency of successful runs was 77%, and the success rate for the completion of the total fMRI battery was 50%, while the values for age-matched healthy volunteers were 96% and 88%, respectively (14).Artifacts other than those caused by organ pulsations have been associated with anxiety or anxiety-sensitivity (15). Medium-to-severe anxiety was reported in 25%–37% of adults undergoing MRI (16, 17). Up to 30% exhibited anxiety-related reactions that ranged from apprehension to a reaction level severe enough to interfere with performance (18). MRI artifacts were also related to the degree of fear and/or panic and anxiety disorders such as claustrophobia (16, 19).A group of studies rejected the contribution of anxiety and proposed another set of causal factors for the artifacts and the resulting fMRI incompliance. According to these studies, patient distress can be predicted from the degree of claustrophobia but not from anxiety sensitivity per se (16). An analytical study (12) measured state anxiety using the Spielberger State-Trait Inventory (STAI), a tool commonly used for measuring state anxiety (20). In their study, state anxiety did not account for the development of movement artifacts. The artifacts were found to be associated with prescan concerns about the technical apparatus and with the procedural aspects of imaging; these concerns were focused on the narrowness of space, noise, immobility, and scan duration. The concerns, which were rated as hardly bearable, were identified in 70.6% of all individuals developing movement artifacts.An approach for meeting the concerns about the technical and procedural aspects focuses on patient comfort and cooperation. In ADHD patients, the effect of individualized prescan preparation was investigated using operant-contingency-based procedures where immediate verbal feedback was provided on response accuracy and where positive reinforcement (verbal praise) was delivered upon criterion achievement (21). This prescan preparation reduced the extent of head movements in both ADHD patients and healthy controls. However, the approach was time-consuming and, due to the sample size (n= 4), too small to be generalizable. Another approach adopted a systematically administered prescan orientation and training program (22). The study reported an overall success rate of approximately 80% in normal children and adolescents (age range, 5–18 years). Based on these findings, the study concluded that it is feasible to conduct large-scale fMRI studies in children. To our knowledge, the effect of such prescan preparations on fMRI compliance has not yet been investigated in children with ADHD.In this study we aimed to demonstrate whether prescan training and orientation affect fMRI compliance of children with ADHD and determine whether this compliance is modified by state anxiety. The study used a well-known cognitive task in the neuropsychology literature, with well-documented activation patterns in the brain.