Electrophysiology in ADHD Diagnosis, predictions, and treatment

Abstract

Study 1: The theta/beta ratio (TBR) study Significantly elevated levels of power in the QEEG theta band or the theta/beta ratio (TBR) were found in 26% of the ADHD patients compared with 2.5% of controls. Excess theta was correlated with inattention and executive problems. Finding excess theta or TBR in a patient strengthens the hypothesis that ADHD is a correct diagnosis, as this pattern is not associated with common comorbid diagnoses in child psychiatry. Our findings were in agreement with most of the recent research. Study 2: The neurofeedback (NF) study In this randomized pilot study, 30 sessions of NF were compared with stimulant medication. At evaluation, 6 to 9 months after onset of treatment, we found significant reductions of ADHD symptoms in the medication group, with large effect sizes, but no significant reductions in the NF group. We argue that NF is best suited for carefully selected groups. Study 3: Predicting acute side effects of stimulant medication In general, about 30 % of ADHD patients treated with stimulants suffer from acute side effects requiring adjustments, like regulations of doses and timing, change to an alternative drug, counseling regarding sleep patterns and food intake, or a total reliance on nonpharmacological treatments. The most common side effects (SEs) within our sample were increased irritability, anxiety, and problems with sleep and appetite. In this study of 37 patients without SEs and 33 with SEs, a logistic regression model showed that three variables contributed significantly to the prediction of SEs: an elevated ERP index of anxiety; a close-to-normal ERP component, measuring action preparation; and a fast reaction time in the VCPT task. Study 4: Predicting the clinical outcome of stimulant medication In this study (N = 98), 76% of the ADHD patients treated with stimulants were considered responders. The clinical outcome (i.e. responder or nonresponder, based on interviews and rating scales) was best predicted by electrophysiological parameters. Compared with norms, responders were characterized by excess theta—a close-to-normal ERP component related to identification of targets, and a smaller-than-normal ERP component related to inhibition. The brain dysfunctions of the responders appear to be primarily associated with prefrontal lobe hypo-activation. The nonresponders seemed to be more deviant in parietal-occipital functions. A significant moderate negative correlation was found between side effects and clinical response. Our conclusions Diagnoses in psychiatry are based on observed behavior. Supplementing these observations with variables from neurophysiology (QEEG and ERP components) and neuropsychology may increase the professional and public credibility of diagnoses like ADHD by documenting deviances in brain function. Such data are potentially helpful in making differential diagnoses, especially in cases in which data from different informants do not coincide. The same parameters are promising candidates for predictions of treatment effects. Our studies can be seen as making small contributions to the development of biomarkers for ADHD. Neurofeedback for ADHD cannot yet be considered evidence-based treatment. It seems to be best suited for carefully selected cases.