|Home | About | Journals | Submit | Contact Us | Français|
Clinical practice guidelines in child psychiatry recommend doing an EEG when warranted based upon a complete history and physical examination. The College of Physicians and Surgeons of Ontario published guidelines as to when an EEG is likely to provide useful information.
All the electroencephalograms ordered at a tertiary care children’s mental health centre over about a 2 year period were reviewed and compared to the guidelines published by the Ontario College of Physicians and Surgeons for ordering EEGs. The outcome of the EEGs and what the ordering physician did after receiving the results were also reviewed.
About 53% were ordered for reasons that the guidelines indicated would result in a significant probability of obtaining clinically useful information. EEG abnormalities were identified in 49% of the youth in this category. About 20% were ordered for reasons the guidelines indicated that an EEG was not likely to provide clinically useful information. EEG abnormalities were identified in 24% of the youth in this category. About 27% of EEGs were ordered for reasons not mentioned in the guidelines. EEG abnormalities were identified in 52% of those youth. Youth who had abnormal results were generally followed up with further investigations. Those youth with more severe abnormalities were often referred to a pediatric neurologist for assessment and treatment.
Children with severe mental health problems have an increased probability of having neurological problems which might have an impact on the ability to assess and treat the mental health problem.
Les directives de pratique clinique en pédopsychiatrie recommandent de faire un electro-encéphalogramme (EEG) lorsque les antécédents médicaux et l’examen physique du patient le justifient. Les directives du College of Physicians and Surgeons of Ontario précisent les circonstances dans lesquelles l’EEG est susceptible de donner des informations utiles.
Tous les EEG commandés à un centre pédopsychiatrique de soins tertiaires pendant deux ans ont été analysés et comparés aux directives du College of Physicians and Surgeons of Ontario sur les conditions de demande d’EEG. L’étude a également porté sur les résultats des EEG et sur le suivi fait par le médecin.
Environ 53% des EEG avaient été demandés pour les raisons qui, d’après les directives, donneraient fort probablement des informations cliniques utiles; des anomalies au niveau de ces EEG ont été constatées chez 49% des sujets de cette catégorie. Environ 20% des EEG avaient été demandés pour des raisons qui ne donneraient probablement pas d’informations cliniques utiles ; des anomalies ont été constatées dans 24% des sujets de cette catégorie. Enfin, environ 27% des EEG avaient été demandés pour des raisons ne figurant pas dans les directives du College of Physicians and Surgeons of Ontario; des anomalies ont été constatées dans 52% des EEG de ces sujets. Les enfants et adolescents dont les résultats étaient anormaux étaient généralement référés à un neurologue pédiatrique pour évaluation et traitement.
Les enfants souffrant de graves problèmes mentaux risquent davantage que les autres de présenter des problèmes neurologiques susceptibles d’influer sur l’évaluation et le traitement de la maladie mentale.
In 2001 the Ontario College of Physicians and Surgeons (CPSO) published guidelines for ordering electroencephalograms (EEG) (Brunet et. al. 2000). These guidelines were intended to assist physicians in deciding when the information obtained by doing an EEG would likely provide relevant and useful clinical information beyond the history and physical examination. One of the indicators was “routine screening of psychiatric patients” which was identified as an indication that was not likely to provide additional useful information.
Butros et. al. (1992) published indications for the use of an EEG in general psychiatry. The primary indication was to look for possible organic aetiologies for the person’s psychiatric condition. An organic aetiology might be considered if there is a history of a neurological disorder such as head injury or when the course of the illness is unusual because of very early onset, greater severity, or rapid progression. The presence of focal neurological signs might also be an indication to request an EEG. A history of epilepsy and episodic behavioural changes suggestive of epilepsy are other possible indication for ordering an EEG. An EEG might also be useful in differentiating between dementia and delirium and to follow the progress of Alzheimer’s dementia. Children and adolescents experiencing significant mental health disturbances frequently also have physical health co-morbidity (McDonald et.al 1998).
The Canadian Psychiatric Association (Waddell et. al. 1999) established practice guidelines for the assessment and treatment of youth with a conduct disorder. The American Academy of Child and Adolescent Psychiatry has also published practice guidelines (King et. al. 1997). The Canadian and American guidelines both emphasize the need to include a complete medical history and current physical examination as part of the diagnostic assessment. They both recommend obtaining an EEG if warranted. When conducting an assessment of a youth suspected of having mental retardation, autism, and other pervasive developmental disorders, an EEG is specifically recommended (Szymanski et. al. 1999, Volkmar et. al.1999; McDonald et. al.2006). Clinical practice guidelines for the assessment and treatment of children and adolescents with an oppositional defiant disorder (Steiner et. al. 2007), depressive disorder (Birmaher et. al. 1998), schizophrenia (McClellan et. al. 2001), substance use disorders (Bukstein et. al. 2005), and bipolar disorders (McClellan et. al. 2007) do not recommend a routine EEG.
This study was carried out in a tertiary care regional children’s mental health center that provides services to youth between birth and 18 years of age with developmental disabilities or psychiatric disturbances. At the time the study was completed, there were about 2000 referrals for outpatient services and 300 referrals for residential services each year. There were approximately 1800 active outpatients and about 200 residential admissions each year. The facility does not have an emergency room. The average length of stay for residential patients was about three months. At the time that the study was conducted there were six developmental paediatricians (the equivalent of 3 full time paediatricians) and six child and adolescent psychiatrists on staff.
One administrative assistant was assigned to book all of the EEGs for all of the physicians. A running list of EEGs that were ordered was maintained, completion of the EEG was confirmed, and receipt of the report was monitored to ensure that the results were received in a timely manner. All of the EEGs on this list from its inception in September 1999 to August 2002 were included in this study. There were 147 EEGs ordered. The majority (139) of the EEGs were done at the local children’s hospital which is a teaching hospital for a major medical school. The remainder were done in general hospitals in 4 different communities.
The casebooks of the youth who had EEGs were obtained from Clinical Records and the EEG requisition, the EEG report, and Progress Notes were reviewed to determine why the EEG was ordered, what the outcome was, and what the ordering physician did after receiving the results.
The majority (139/147) of the EEGs were done at the local children’s hospital which is a teaching hospital for a major medical school. EEGs done at that hospital were interpreted by the staff Neurologists using a modification of the Mayo Clinic classification of abnormal EEGs (Mayo Clinic, Lemieux et al). In this system “normal” means normal limits for age and state of alertness. “Essentially normal” means probably normal but contains one or more elements of questionable normality. “Dysrhythmias Grade I–IV” include five categories of varying intensity and frequency of theta-delta or rhythmic activity, including spikes or recorded seizures. Only the highest dysrhythmia grade is given. If significant abnormalities in lower grades exist, they are included within the higher grade. When EEG abnormalities are not epileptiform, they are termed “non-specific” and localized to a lobe or lobes, hemispheric, or generalized in the same manner as seizure activity is localized.
The CPSO guidelines listed a number of indications for completing an EEG which were likely to provide clinical useful information (Table 1). They also listed a number of indications for completing EEGs that would be unlikely to result in useful clinical information (Table 2). Each EEG was coded according to the one CPSO guideline indication of best fit. A number of EEGs were ordered for reasons that did not fit into any of the indications from the CPSO guidelines. These EEGs were grouped into a new category of indications that the CPSO did not mention in their guidelines (Table 3). There were many more situations mentioned by the CPSO guidelines that were not germane to this study such as seizure surgery, stupor and coma, assessment for brain death etc and these were excluded from the database. If the specific reason for ordering the EEG did not use the same terminology as one of the indicators in the guidelines, a best fit was coded based upon reading the consultation report and other clinical documentation that supported ordering the EEG. A history of epilepsy was coded as “poorly controlled epilepsy”. This was an appropriate indication according to CPSO guidelines. “Staring spells” or “blank stares” were added to the group ordered for “query seizure disorder”. “Query seizure disorder” was added to the list of indications cited by the CPSO guidelines as likely to provide useful clinical information. Hallucinations, delusions, and dissociative state were coded as “general screening of psychiatric patients”. General screening of psychiatric patients was included in the list of situations in which an EEG does not have a reasonable probability of providing useful clinical information. Youth with mental retardation (McDonald et. al. 2006, Szymanski et. al. 1999) and pervasive developmental disorders (Volkman et. al. 1999) have a much higher rate of neurological abnormalities than children with other psychiatric disorders. Referrals for EEGs where mental retardation, autism, or Asperger’s Syndrome was the indication for the EEG were put into the category (Table 3) created that included indications not mentioned in the CPSO guidelines rather than “general screening of psychiatric patients”. Clinical practice guidelines for these conditions in the psychiatric literature mention that an EEG might be a useful thing to do in the assessment of these conditions.
Approval for this quality assurance study was given by the Administrator at the children’s mental health centre pursuant to Section 182(2) of the Child and Family Services Act RSO 1990 C. 11 (last revised 2008). The University of Western Ontario Office of Research Ethics on the Use of Human Subjects approved this study and waived the need for patient consent.
A total of 147 EEGs were ordered. Of these 78 (53%) were ordered for reasons that the CPSO recognized as having a reasonable probability of obtaining clinically useful information, 29 (20%) were ordered for reasons that the CPSO did not feel would likely lead to clinically useful information, and 40 (27%) were ordered for reasons that the CPSO did not mention. Abnormal EEGs were noted in 40 (50%) of youth who had an EEG for a reason that the guidelines indicated likely would lead to useful information. Almost a quarter (N=7, 24%) of youth who had EEGs for reasons that the guidelines felt would not provide useful information had an abnormal EEG. Of those youth who had an EEG for reasons that the guidelines did not mention, 22 (55%) had an abnormal EEG. The number of EEGs ordered for reasons that the CPSO did not mention in their guidelines was statistically significant (Chi Square 26.98; df (2) p> 0.000).
If there was an abnormality in EEGs done for reasons that the CPSO felt would be useful, the abnormality tended to be more severe. For those EEGs done for reasons that the CPSO thought would not lead to useful information, if there was an abnormality, it tended to be less severe. EEGs done for indications that the CPSO did not mention showed a full spectrum of level of severity if abnormal.
Often it is very difficult to get a satisfactory EEG tracing with youth who have a difficult temperament or major behavioural problems or an intellectual compromise. Only 4 tracings were reported as unsatisfactory for interpretation by the electrencephalographer generally because of movement artefact. This speaks to the skill of the EEG technicians at the local children’s hospital where many of the EEGs were done.
The most common reasons for ordering an EEG were: query seizure (N=34), poorly controlled epilepsy (N=18), and previous abnormal EEG (N=18). The first two are reasons that the CPSO guidelines indicated would be useful reasons to order an EEG and the latter was not mentioned. Abnormal EEGs were found in 72% of youth with poorly controlled epilepsy, 50% of those with a question of epilepsy, and 61% of those with previous abnormal EEG. In most cases the abnormalities were towards the more severe end of the dysrythmias. The fourth most common reason for ordering an EEG were behavioural problems and episodic dyscontrol each with 15 EEGs ordered (each representing 10.2% of all EEGs ordered). Behaviour problems is one of the clinical concerns that the CPSO guidelines indicate that an EEG is unlikely to provide additional clinically useful data. Episodic dyscontrol is one of the situations that the CPSO guidelines indicate that an EEG would be useful. In our study 27% of the youth with behavioural problems and 20% of those with episodic dyscontrol had an abnormality in their EEG. The remaining individual indications all had fewer than 15 patients so an individual analysis was not completed.
Table 1 shows all of the indications that the CPSO felt an EEG would likely provide clinically useful information that were seen in this population as well as the outcome of all of the EEGs ordered for these reasons (49% abnormal). There were a number of other indications in the guidelines but none of our patients fit into those categories. When these children had an abnormal EEG the abnormality tended to be more severe in magnitude.
Table 2 lists all of the CPSO indicators seen in this study in the category of an EEG is unlikely to be providing useful clinical information (24.0% abnormal). Youth in this group who did have an abnormality on their EEG tended to have abnormalities toward the lower end of the severity spectrum. General screening of psychiatric patients is also in this category as are learning disabilities, attention deficit hyperactivity disorder, and language delay.
Table 3 shows all of the EEG’s that were ordered for reasons that the CPSO guidelines did not mention. Over half of the youth (56%) who had EEGs ordered for reasons in this group had abnormalities which were about equally distributed between the levels of severity. When the EEG requisition indicated “mental retardation” or “autism” as the reason for ordering the EEG it was assumed that the patient did not have a known seizure disorder.
Table 4 shows the actions undertaken by the ordering physician after receiving the results of the EEG. For 47 of the children, two interventions were undertaken, 25 had three interventions, and 7 had four interventions. If an abnormal EEG was reported, the youth likely had further investigations through neuro imaging (CT or MRI) or was referred to a paediatric neurologist or developmental paediatrician. At the time that this study was conducted, there was a paediatric neurologist at the children’s mental health centre one full day per week every week and the psychiatrists and developmental paediatricians could obtain a consultation within a few weeks.
EEGs should be ordered when the medical history and physical examination provide reasons for using an EEG for further investigations. In 2000 the Ontario College of Physicians and Surgeons published guidelines for clinical practice and facility standards regarding EEGs. We compared the EEGs ordered by all of the physicians at a regional children’s mental health centre against these guidelines.
The CPSO guidelines were developed for community-based practices. The patient population at the children’s mental health center was heavily skewed toward children with developmental disabilities. Approximately 50% of the patients at the facility have some degree of developmental delay. People with a developmental delay have a significantly higher probability of having EEG abnormalities and seizure disorders. In addition, the children who were referred to this children’s mental health centre had complicated mental health problems that were not amenable to remediation in a less restrictive situation such as the community-based practice or an outpatient department of a general hospital. Youth referred to this facility have generally exhausted the less restrictive interventions from community facilities before being referred. There is a significantly higher proportion of youth with biological predisposing, precipitating, and perpetuating factors in this population than within a community sample. For example, 16% of all children admitted to residential treatment at this children’s mental health centre have a diagnosis of a seizure disorder (unpublished data from an internal quality assurance study) whereas 1–2% of the general population in Ontario have a history of convulsions (Epilepsy Ontario website). The prevalence is higher in children and the elderly than in the general population (Epilepsy Ontario website). Therefore it is not surprising to find EEG abnormalities in youth who had an EEG done for reasons that are less likely to produce useful information in the general population study. The high frequency of EEG abnormalities in this population points out the need for child psychiatrists to be acutely aware of possible biological predisposing, precipitating, or perpetuating biological factors in a child with complex mental health problems. It also reminds us of the need to have a close working relationship with developmental paediatricians and paediatric neurologists. In some cases, unless the organic disorders are first addressed, the behavioural interventions will not be as effective.
Did the EEG abnormalities have any bearing on the treatment that the children and youth received? In some cases, it led to prescribing an anticonvulsant, changing the dose of an anticonvulsant or changing the anticonvulsant medication. In youth with mood instability or behavioural explosiveness, sometimes an anticonvulsant was prescribed. An abnormal EEG could also affect the choice of psychotropic medication to one with a lower potential for seizures as an adverse side effect. Anecdotal evidence indicates that for some youth who have an abnormal EEG and who have not responded well to behavioural interventions or psychotropic medications, adding an anticonvulsant may be useful.
Limitations of this study are that the data were collected in a retrospective chart review done a number of years after the actual EEGs were done. We were dependent on chart information and not able to clarify any questions with the physician who ordered the EEG particularly as it related to the actions taken after receiving the results. It is a relatively small study, with only 147 patients. The CPSO guidelines were developed for a primary care community based practice. With the highly complex patients in a tertiary care practice where there is a significant likelihood of the youth having a neurological problem, it is not surprising that there were a significant number of youth who had an abnormal EEG that were done for reasons that the CPSO guidelines indicated a low likelihood of providing useful information. These data are probably not transferable to the general population of children and adolescents with mental health problems because the population in this study is skewed towards children with more severe psychiatric problems and with developmental disabilities. The guidelines remain useful for a primary care practice. The study does, however, underscore the need for all physicians working with children and adolescents with mental health problems to conduct a complete medical history, particularly a neurological history, and order an EEG or a neurology consultation where indicated.
The authors have no financial relationships or conflicts to disclose.