Theta                              Absolute Beta                     Relative Beta

Figure 1: This figure shows the average brain activity (quantitative EEG – QEEG) of 275 children with ADHD, compared to a control group. On the left the increased theta EEG activity (p<.0001) can be seen, in the middle the absolute beta EEG activity (p<.0001) and on the left the decreased relative beta EEG activity (p<.0001). This deviant brain activity has a fronto-central localization. This pattern is found in almost all ADHD studies.

If one takes a look at the individual data, however, (see figure 2) a completely different picture emerges. In figure 2 the individual data of 36 random ADHD patients is represented drawn from this same Brain Resource International Brain Database. In the table below the following is represented Theta – Red; Alpha – Yellow and Beta – light blue. The quantative EEG data of these children – or QEEG’s – are compared to a normative database of more than 5000 healthy controls enabling an individual comparison. It can clearly be seen that indeed 47% of the ADHD patients have an increased theta activity. However, 5,6% of the patients show a decreased beta activity and 22% show an increased beta activity. The increased beta can be explained by the presence of beta spindles which occur in 20% of the cases of the ADHD children and which demand a different treatment strategy. The inter-individual variability within a behaviorally homogenous group such as ADHD patients thus is quite considerable.

Figure 2: This figure shows the EEG data of 36 random children (4-digit ID codes) with ADHD from the same dataset as in figure 1. This time, however, the individual data are represented. Indeed some children show an increased theta EEG activity (47%), but only 5,6% of the children show a decreased beta EEG activity and 22% show an increased beta EEG activity. Note the contrast between the individual and group data.

Considering the fact that Ritalin does not have a clinically meaningful effect in 20-40% of the ADHD patients (Swanson et al., 1993; Gordon, 2007) is seems plausible that the cause of it is to be found in the interindividual variability of the brain functioning, as shown.

The above mentioned group of ADHD children is part of a large scale clinical investigation, and therefore all of these children were treated with a stimulant such as Ritalin. Consequently we tested the prior mentioned hypothesis. A group of 50 of those children were divided into different groups according to their individual EEG phenotype, as seen in figure 3. It turned out that only the group with frontal slow activity (frontal theta) responded well to treatment with stimulant medication (Methylphenidate: Ritalin), as measured by the improvement on a continuous performance test (CPT), whereas the other groups did not show any improvement on the CPT as a result of medication (Arns et al. 2008).

Comparable studies of ADHD have shown that patients who responded well to stimulant medication such as Ritalin indeed showed excess slow brain activity frontally (Delta en Theta: Clarke et al., 2002; Satterfield et al., 1972; Suffin & Emory, 1995). This is clearly a well identifiable ADHD sub-type that responds well to medication. This knowledge can very well be applied to personalizing ADHD treatment not only with respect to pharmacotherapy but also for EEG Biofeedback or Neurofeedback treatment.

Figure 3: In the figure above the prevalence of the occurrence of the different EEG Phenotypes is shown for a group of ADHD children and a matched control group. It can clearly be seen that ADHD and control group differ particularly on op Frontal Slow, Slow Alpha Peak Frequency, Low Voltage en Frontal Alpha. This demonstrates that there is a great diversity of EEG patterns observed within a group of patients, but also within a group of ‘healthy’ controls. According to the literature these ADHD patients will all respond favourably to different medications, the Frontal Slows will respond best to Ritalin (Arns et al., 2008) and the Frontal Alpha will respond better to an SSRI (Suffin & Emory, 1995).

To Read the full Research report EEG PHENOTYPES PREDICT TREATMENT OUTCOME TO STIMULANTS IN CHILDREN WITH ADHD

For more background information on EEG Based Pesonalized Medicine also see: http://www.brainclinics.com/personalized-medicine

The Concept of Phenotypes

Prior studies using statistical analysis of electroencephalography (EEG) have documented clusters of EEG/quantitative EEG (QEEG) features within psychiatric populations (John, Prichep, & Almas, 1992). Experience over the past 30-plus years with a large number of clinical EEGs and, more recently, decades worth of clinical and research experience with QEEG, as well as a review of the field’s literature, have shown that a limited set of EEG patterns can characterize the majority of EEG variance.

In the perspective I’d like the reader to consider, these proposed EEG/EEG groupings might be considered as phenotypes, based on genetics. There is an indirect linkage between genetics and behavior, with an intermediate step. This intermediate step is one that is based on the expression of the genetics and other factors and constitutes the bridging between the person’s genetics and behavior: the phenotype. These phenotypic EEG/QEEG divergence patterns constitute reliable indices of brain function and predict response to therapy.

It should be noted that these phenotypic patterns are not isomorphic with the established Diagnostic and Statistical Manual of Mental Disorders (DSM) categories, although the phenotypes have powerful implications for therapeutic interventions, both with medication and with neurofeedback. One phenotype may be seen in a wide variety of DSM groupings, from posttraumatic encephalopathy, to affective and attentional related DSM groupings, to many more. These ideas have been published elsewhere (Johnstone, Gunkelman, & Lunt, 2005), although these concepts seem to have especially important perspective implications for this special issue of Biofeedback focusing on subtypes of specific DSM categories.(1)

The very concept of an EEG pattern’s being a subtype of a specific disorder seems foundationally flawed to me because of the lack of specificity of the pattern for the DSM grouping. The theta-beta ratio being increased for age may be a metric that is sensitive to attention deficit hyperactivity disorder (ADHD), but the same increase in this ratio also may be seen in a wide variety of other clinical presentations as well as in the absence of ADHD, so the lack of specificity remains a problem. I’d like to invite the reader to transcend the DSM’s limited perspective through the use of phenotypes.

Enhancing Neurofeedback Through the Use of Phenotypes

Behaviorally based neurofeedback interventions have been used with great effectiveness in the hands of good clinicians and practitioners, as evidenced by our field’s ever growing efficacy literature. This efficacy literature is based on actual clinical outcome data and provides support for the rapidly growing list of neurofeedback applications that can claim efficacy based on the jointly adopted Association for Applied Psychophysiology and Biofeedback (AAPB)/International Society for Neuronal Regulation (ISNR) efficacy template (La Vaque et al., 2002).

Recently, there have been several good publications detailing the state of the efficacy literature in biofeedback and neurofeedback, including the book by Yucha and Gilbert (2004), Evidence-Based Practice in Biofeedback and Neurofeedback, as well as several white papers on specific disorders, in a series sponsored by the AAPB and ISNR (Monastra et al., 2005; Moss, LaVaque, & Hammond, 2004).

Transcend the DSM Using Phenotypes

The literature on medication response prediction suggests that a phenotypic perspective may help enhance our efficacy (Suffin & Emory, 1995). This was also suggested in the outcome improvement reported by Wright and Gunkelman (1998) when he added the QEEG approach to guide neurofeedback.

The presence of genetically linked EEG patterns provides a solidly data-based set of observations on which to propose an initial list of phenotypic patterns. One EEG pattern with genetic links is the low-voltage fast EEG (Gunkelman, 2001). This low-voltage fast pattern was characterized as a phenotype in a recently published study of phenotypic patterns in alcoholism (Enoch, White, Harris, Rohrbaugh, & Goldman, 2002) and by others who have identified the genetic link to Gene 4’s regulation over gamma-aminobutyric acid receptors (Bierut et al., 2002).

Another EEG pattern with genetic links has been identified in some cases of idiopathic epilepsy (Haug et al., 2003). The paroxysmal epileptiform bursts seen in the EEG in these clinical cases may achieve many hundreds of microvolts, occasionally exceeding 400 to 600 ÌV, with spikes and slow components emerging from a relatively normal background EEG. In a survey of genetic factors in epilepsy, Kaneko, Iwasa, and Okada (2002) showed that the most common human genetic epilepsies display a complex pattern of inheritance and that the identities of the specific genes are largely unknown, despite recent advances in the science of genetics. They also show that the genetic markers of certain types of epilepsy have been identified, such as those with neurodegenerative characteristics and a small number of familial idiopathic epilepsies (Haug et al., 2003). A similar pattern of seizures and EEG phenotype is seen in a group of subjects with benign childhood epilepsy with centrotemporal spikes, found in three children with de novo terminal deletions of the long arm of Chromosome 1q. This suggests that this chromosomal location could be a potential site for a candidate gene (Vaughn, Greenwood, Aylsworth, & Tennison, 1996).

Characterizations linking genomic information, intermediate EEG phenotypes, and behavioral manifestation are likely to have important implications for therapeutics. The International Brain Database (Brain Resource Company) includes the EEG, event related potential, neuropsychological test measures, and genetic testing, thus making it unique in allowing an integrative approach that combines neurophysiology, neuroanatomy, cognition, and genetics. The M.I.N.D. Center at the University of California, Davis, uses the phenomic approach in working with pervasive developmental disorder/autism to avoid treating all clients alike in such a heterogeneous population and to help clarify research on clinical therapeutic effects.

Subtypes Versus Phenotypes

Although many studies have shown subtypes within a DSM-identified disorder, such as the work of Prichep et al. (1993) with obsessive-compulsive disorder and that of Chabot, Merkin, Wood, Davenport, and Serfontein (1996) identifying subgroups of ADHD, these subgroups do not have diagnostic specificity because they can be seen in other disorders. These EEG clusters did, however, predict treatment efficacy using medication in both the ADHD and obsessive-compulsive disorder studies. Suffin and Emory’s (1995) article identified frontal theta in attentional problems, similar to the Chabot et al. findings in ADHD, but the same pattern was also identified in affective disorders. The behavioral grouping did not predict the EEG/QEEG pattern, nor did behavior groupings predict the proper pharmaceutical treatment, although the EEG patterns did predict the effective drug intervention in both studies.

The medication intervention, when clinically based on prospectively identifying the phenotype, was the basis for the highly effective intervention in treatment refractory depression in recent pilot work (S. Suffin, personal communication, 2003) at the Sepulveda Veterans Affairs in Los Angeles. The application of these principles was also at the heart of the doubling of the clinical efficacy in a recent study using QEEG to design the neurofeedback intervention for the attention deficit disorder/ADHD population tested, as compared to behaviorally based neurofeedback interventions (Wright & Gunkelman, 1998).

In an updated and abbreviated review of prior work (Johnstone et al., 2005), the Appendix offers proposed phenotypic patterns as well as a listing of the associated neurofeedback interventions. A reading of the original article is advised for more detail, especially with respect to medication response prediction.

Conclusion: An Introductory Framework for EEG/QEEG Phenotypes

It must be remembered that phenotypes may coexist, and the various combinations are too variable to be handled in this limited presentation. This appendix should not be construed as a replacement for professional assistance in designing a neurofeedback intervention, nor in any way can this be used to fully characterize an individual’s EEG/QEEG.

Appendix: Electroencephalography (EEG)/Quantitative EEG Phenotypes

Note

1. The material is adapted from Johnstone, Gunkelman, and Lunt (2005), Clinical database development: Characterization of EEG phenotypes. Clinical EEG and Neuroscience, 2, 99-107, and is used with the permission of Clinical EEG and Neuroscience.

References

Bierut, L. J., Saccone, N. L., Rice, J. P., Goate, A., Foroud, T., Edenberg, H., et al. (2002). Defining alcohol-related phenotypes in humans. The Collaborative Study on the Genetics of Alcoholism. Alcohol Research and Health, 26, 208-213.

Chabot, R. J., Merkin, H., Wood, L. M., Davenport, T. L., & Serfontein, G. (1996). Sensitivity and specificity of QEEG in children with attention deficit or specific developmental learning disorders. Clinical EEG, 27, 26-34.

Chabot, R. J., & Serfontein, G. (1996). Quantitative electroencephalographic
profiles of children with attention deficit disorder. Biological Psychiatry, 40, 951-963.

Enoch, M. A., White, K. V., Harris, C. R., Rohrbaugh, J. W., & Goldman, D. (2002). The relationship between two intermediate phenotypes for alcoholism: Low voltage alpha EEG and low P300 ERP amplitude. Journal of Studies on Alcohol, 63, 509-517.

Gunkelman, J. (1998). Evaluating the frontal lobes in affective and attentional disorders with QEEG and EP—the electrophysiology of frontal lobe disconnection syndrome: Implications for neurotherapy. Journal of Neurotherapy, 3(1-2).

Gunkelman, J. (2001). Low voltage or absolute power. Journal of Neurotherapy, 5(1-2), 107-110.

Haug, K., Warnstedt, M., Alekov, A. K., Sander, T., Ramirez, A., Poser, B., et al. (2003). Mutations in CLCN2 encoding a voltage-gated chloride channel are associated with idiopathic generalized epilepsies. Nature Genetics, 33, 527.532.

John, E. R., Prichep, L. S., & Almas, M. (1992). Subtyping of psychiatric patients by cluster analysis of QEEG. Brain Topography, 4, 321-326.

Johnstone, J. (2005a, September). QEEG patterns: Subgroups, profiles, phenotypes. Paper presented at the 13th annual meeting of the International Society for Neuronal Regulation, Denver, CO.

Johnstone, J. (2005b, April). Symposium on quantitative EEG, neurofeedback, and cognition: Quantitative EEG profile as phenotype. Paper presented to the Neuropsychiatric Institute, UCLA, Los Angeles, CA.

Johnstone, J., Gunkelman, J., & Lunt, J. (2005). Clinical database development: Characterization of EEG phenotypes. Clinical EEG and Neuroscience, 2, 99-107.

Kaneko, S., Iwasa, H., & Okada, M. (2002). Genetic identifiers of epilepsy. Epilepsia, 43(Suppl. 9), 16-20.

La Vaque, T. J., Hammond, D. C., Trudeau, D., Monastra, V., Perry, P., & Lehrer, P. (2002). Template for developing guidelines for the evaluation of the clinical efficacy of psychophysiological interventions. Applied Psychophysiology and Biofeedback, 27, 273-281.

Monastra, V., Lynn, S., Linden, M., Lubar, J. F., Gruzelier, J., & Vaque,T.J.(2005). Electroencephalographic biofeedback in the treatment of Attention Deficit/Hyperactivity Disorder, Applied Psychophysiology and Biofeedback, 30(2), 95-114.

Moss, D., LaVaque, T. J., & Hammond, D. C. (2004). Introduction to white paper series—Guest editorial. Applied Psychophysiology and Biofeedback, 19, 151-152.

Niedermeyer, E., & Lopes da Silva, F. (1993). EEG patterns and genetics. In E. Niedermeyer & F. Lopes da Silva, Electroencephalography: Basic principles, clinical applications and related fields (3rd ed., pp. 192195). Baltimore: Lippincott, Williams & Wilkins.

Prichep, L. S., & John, E. R. (1992). QEEG profiles of psychiatric disorders. Brain Topography, 4, 249-257.

Prichep, L. S., Mas, F., Hollander, E., Liebowitz, M., John, E. R., Almas, M., et al. (1993). Quantitative electroencephalographic subtyping of obsessive-compulsive disorder. Psychiatry Research, 50(1), 25-32.

Suffin, S. C., & Emory, W. H. (1995). Neurometric subgroups in attentional and affective disorders and their association with pharmacotherapeutic outcome. Clinical Electroencephalography, 26, 76-83.

Vaughn, B. V., Greenwood, R. S., Aylsworth, A. S., & Tennison, M. B. (1996). Similarities of EEG and seizures in del(1q) and benign rolandic epilepsy. Pediatric Neurology, 15, 261-264.

Wright, C., & Gunkelman, J. (1998, September). QEEG evaluation doubles the rate of clinical success: Series data and case studies. Abstract presented at the sixth annual conference of the Society for the Study of Neuronal Regulation, Austin, TX.

Yucha, C., & Gilbert, C. (2004). Evidence-based practice in biofeedback and neurofeedback. Wheat Ridge, CO: Association for Applied Psychophysiology and Biofeedback.

Biofeedback Association for Applied Psychophysiology & Biofeedback
Volume 34, Issue 3, pp. 95-98 www.aapb.org