Anales de la RANM

126 A N A L E S R A N M R E V I S T A F U N D A D A E N 1 8 7 9 NEUROTHERAPEUTICS FOR ADHD Katya Rubia An RANM. 2021;138(02): 124- 131 right anterior cingulate (16). Our meta-analysis of fMRI studies of timing functions showed consis- tently reduced activation in ADHD patients relative to healthy controls in key regions of timing such as left IFC, left inferior parietal lobe and right cerebe- llum (17). A meta-analysis of fMRI studies of working memory showed that ADHD patients relative to controls had reduced activation in bilateral middle and superior prefrontal cortex and left medial frontal cortex/anterior cingulate (18), as well as right and left IFC (16). We furthermore found in two large comparative fMRI meta-analyses of cognitive control tasks that the right IFC and striatal underactivation is disorder-specific to ADHD relative to obsessive- compulsive disorder and autism (10, 12). Overall, the fMRI meta-analyses suggest that ADHD patients have multisystem functional deficits compromising different fronto-striato-parieto-cerebellar networks that mediate several cognitive domains (14). ADHD patients have also shown to have abnormally increased activation in areas of the default mode network (15, 17). The default mode network consists of intercorrelated activation of ventromedial frontal cortex, posterior cingulate, precuneus, inferior parietal and temporal regions and is thought to reflect task-irrelevant thoughts, i.e., mind wandering (19). It has been suggested that people with ADHD have less control over their mind-wandering which intrudes into their already weak exterocep- tive attention processes, causing inattention. This immature pattern of poor activation of task-relevant networks and of decreased deactivation of the default mode network reflecting more mind-wandering has been suggested to be responsible for the poor perfor- mance in ADHD on attention-demanding higher- level cognitive tasks (14). The most consistently found dysfunctional regions, in particular right IFC, DLPFC and anterior cingulate have been used as targets for neuromodulation studies such as neurofeedback with fMRI or NIRS or brain stimulation. The last decade of neuroimaging has shown that the brain is highly plastic, not only in the develo- ping brain in childhood and adolescence, but also in adulthood (20). For example, several weeks or months of training of a particular skill in adults, for example, juggling (20), learning to meditate (21) or learning for a medical exam (22) can change the structure of specific brain regions. These insights into the neuroplastic potential of the brain make novel neuromodulation treatments, such as non-invasive brain stimulation or neurofeedback attractive clinical interventions. This applies even more to young people, who have superior neuroplasticity (14). fMRI studies of ADHD over the past decades have provided good targets for neurotherapeutics. It seems plausible that therapies that aim to reverse these key neurofunctional abnormalities could improve the disorder. FMRI-Neurofeedback or NIRS-Neurofe- edback are still very much in their childhood, with too few and very small-numbered studies in ADHD to give evidence for potential clinical effects. Non-invasive brain stimulation studies have been increasing exponentially in ADHD over the past 10 years. The majority of studies, however, have been in relatively small numbers with highly heterogenous study designs. Therefore, the findings have been inconsistent with respect to improving cognition with very little evidence, so far, on improving clinical ADHD symptoms. fMRI-neurofeedback and NIRS-neurofeedback Neurofeedback is based on operant conditioning that teaches participants to volitionally self-regulate specific regions or networks using trial and error, through real-time auditive or visual feedback of their brain activation which is typically represented on a PC in the form of a thermometer or with a videogame to make it more attractive for children. Electrophysiology (EEG)- neurofeedback has been tested in ADHD for over 45 years. There are 10 meta-analyses reviewing the evidence with the latest meta-analysis showing small to medium effect size of superiority of EEG-neurofeedback compared to non-active control groups for improving parent rated ADHD symptoms and for improving the inattention subdomain for teacher ratings; however effects are inferior to pharmacotherapy (23). Real-time fMRI neurofeedback enables participants to self-regulate the blood-oxygen level-dependent response of a targeted brain region, or network, through real-time feedback of their brain activity. fMRI-neurofeedback has superior spatial resolu- tion than EEG-neurofeedback and can target the key cortical and subcortical brain function deficits that have been established in ADHD over the past 26 years of fMRI research (14). fMRI-neurofeedback has shown some promise in improving clinical symptoms and cognition in other psychiatric disorders (24). To date, however, there are only two published fMRI-neurofeedback studies in ADHD. A small randomised controlled trial in 13 adults with ADHD asked patients to do a mental calculation task with (N = 7) and without (N = 6) fMRI-neurofeedback of the dorsal anterior cingulate in 4 weekly scans of 60 minutes (25). Both groups significantly increased anterior cingulate activation but did not differ in improvements in ADHD symptoms observed in both groups at trend-level. However, only the neurofe- edback group showed significant improvement in a sustained attention and working memory tasks, suggesting some positive effects of fMRI-neurofee- dback of dorsal anterior cingulate on cognition (25). A randomised controlled trial from our lab tested fMRI-neurofeedback of the right IFC compared to fMRI-neurofeedback of the left parahippocampal gyrus in adolescents with ADHD (26). Thirty-one boys with a clinical ADHD diagnosis had 4 hourly scans over 2 weeks, in which they did 11 runs of 8.5 min of fMRI-neurofeedback with a rocket NEUROTHERAPEUTICS IN ADHD