A study on EEG findings: An experience from a tertiary care center Nepal

Abstract

Introduction

The electroencephalography (EEG) is the recording of brain’s spontaneous electrical activity over a period of time (15-20) minutes recorded from multiple electrodes placed on the scalp. It is used in seizure disorder, organicity, psychiatric conditions. There is a paucity of literature with regard to application of EEG in various conditions in our setting. The current study aims to explore the EEG findings of different cases and their associations in various clinical scenarios.

Method

It is a retrospective study of cases who underwent EEG in the EEG room of Department of Psychiatry at Patan Academy of Health Sciences (PAHS), Nepal from 2014-2017. Information was taken from EEG register and EEG reports were recorded in a data sheet. Data was analyzed using IBM SPSS version 23. Frequency distribution was studied and Chi Square test was applied for categorical variables. The only continuous variable studied was age for which mean, median and standard deviation were computed and suitable statistical tool was applied after normality test.

Result

Of the total 164 cases, mean age was 21.93 years and 51.2% were male. The common reason for EEG application was to rule out seizure disorder(80.5%) requested mostly from department of Psychiatry EEG abnormality was seen in 43.3% with EEG diagnosis of generalized epilepsy in 26.21% and slow wave (46.2%) as the most common EEG finding. Abnormal EEG detection rate was almost similar in both gender, predominant in those with clinical diagnosis of seizure disorder (47.88%). The EEG abnormality detection rate was significant in those referred from department of Psychiatry and Paediatrics.

Conclusion

EEG is a relatively inexpensive and non-invasive test for detection of electrical activity in brain. Though requested for seizure or related disorders, it can also find its place in organic pathology, monitoring treatment response among others.

Keywords: Electroencephalogram; EEG findings; Seizure; Nepal

Introduction

Electroencephalography (EEG) is the recording of electrical activity produced by the firing of neurons within the brain which is recorded for about 20 to 40 minutes from multiple electrodes placed on the scalp. It is one of the  tools to assess cerebral function which is  based on the work by Hans Berger in the 1930s.[1] The main use of this noninvasive test is in epilepsy to detect seizure activity, a common problem with estimated worldwide prevalence of 5–30 persons per 1000.[2]

The other applications are in diagnosis of coma, encephalopathy, and brain death.[3]It is also used in sleep studies and monitoring seizure during Electroconvulsive Therapy (ECT). Though gives a measure of cerebral function, the role of EEG in organic conditions is limited with the advent of newer imaging modalities like CT and MRI. The use of this neurophysiological tool has been minimum in psychiatry despite known relationship between epilepsy and psychosis and other psychiatric manifestations, especially with the temporal lobe abnormalities.[4] Despite the arguments about the role of EEG in psychiatry, patients are sent for EEG for various reasons, usually for seizure, pseudo seizure, organicity and psychosis.

There is a paucity of reported literature on EEG findings although many institutes conduct EEG in Nepal now. Patan Academy of Health Sciences (PAHS) is a tertiary care multi-specialty referral center at Lalitpur, Nepal.  It started its EEG services from 2014. The current study aims to bridge the knowledge gap and explore the EEG findings among different cases requested for EEG and their associations. Methods

This is an analytical retrospective study conducted at the Department of Psychiatry, PAHS from 2014 to 2017. The patients requested for EEG underwent a routine non-sleep deprived EEG using the international standardized 10–20 system of electrode placement. Photic stimulation and hyperventilation methods were used during the EEG recordings where age was not a barrier. All EEGs were done by one technician with the same EEG machine (16 channels RMS digital). All records were reported by one psychiatrist.

All the patients referred for EEG were considered. From the total … cases, … were excluded from the study due to incomplete data and marked artifacts. Data of 164 patients were taken from EEG register and EEG reports. A semi-structured proforma was used to record the information on the variables like age, sex, probable clinical diagnosis and reasons for referral. EEG findings and EEG diagnosis were tabulated from EEG report. EEG reports were categorized as normal and abnormal. EEG findings were categorized as Localization Related Epilepsy (LRE) and Generalized Epilepsy (GE) for focal epileptiform activity and generalized epileptic activity respectively.

Data was entered in Microsoft Excel (MS Office 2013, Microsoft Corporation, Washington, United States) and analyzed in IBM SPSS v23 for Windows (IBM Statistical Package for Social Sciences, 2015 IBM Corporation, New York, United States). Frequency distribution was studied and Chi Square test was applied for categorical variables to test for associations. The only continuous variable studied was age which was categorized in the interval of 10 and mean, median and standard deviation were computed. A suitable statistical tool for testing association was applied after Shapiro-Wilk normality test.  A p value of less than 0.05 was considered significant.

Results

The demographic characteristics are given in (Table 1). Majority of the patients belong to age group 11-20 years, comprising 29.87% (49) of the total study population. 54.26% of patients were below 20 years of age. The mean age was 24.24±12.38 years. Males were slightly more (51.2%) than females. Most of the patients (36%) did not have any clinical diagnosis mentioned at the time of requesting for EEG. Those diagnosed with seizure disorder (31.7%) were the second highest among those referred for EEG evaluation. The reason for referral for EEG was the exclusion of seizure disorder in the majority of cases (80.5%). More than half (58.5%) of the EEG were requested by the department of Psychiatry and 15.9% of the EEG requests did not contain the details of the referring departments.

Table 2 illustrates the findings of EEG where abnormality was seen in 43.3% (71). Among them, 26.21%(48) had generalized epilepsy and 14.02%(23) had localization related epilepsy. The most common abnormal pattern in EEG was slow wave (46.2%) and location was generalized (29.3%) followed by frontal lobe (5.5%).

The analysis revealed abnormalities in the age group of 1-10 years with the mean age of 18.91 years almost equally in both male (49.29%) and female (50.7%). The Chi Square test did not show any statistically significant association in the occurrence of positive EEG report between the genders (p=0.753). The Mann Whitney U test showed significance with p value of 0.001 with the patients with abnormal EEG being younger overall. The abnormality detection rate was more in seizure disorder (35.21%) and those with comorbid psychiatric disorders (12.67%). Similarly, the cases referred for exclusion of seizure had EEG abnormality in 74.64% which is statistically insignificant (p=0.09) and so are the other reasons of referral for EEG. The referral from Psychiatrists (46.47%) and Pediatricians (33.8%) had abnormal EEG which is statistically significant with p values of 0.007 and 0.000 respectively. (tables 3 and 4)

Discussion

This retrospective study analyzed the EEG findings of those who underwent EEG with different clinical diagnosis for various reason from 2014 to 2017.

The majority of cases referred for EEG in our study belonged to age group 11-20 (29.87%) with mean age of 24.24+12.38 and 89% below 20 years of age. Similar finding was reported in the studies by Shrestha et al (31.7%) [5], Bhagat et al.(29.4%) [6] and Chowdhary et al [7].This reason could be the higher number of patient in this group. Males (51.2. %) were slightly more than females (48.8%) who appeared for EEG evaluation which was in accordance to the study by Shrestha el al [5] and Molokomme and Subramaney[8]. Seizure disorder with and without psychiatric illness (40.8%) was the most common clinical diagnosis that prompted for referral for EEG evaluation whereas study by Shrestha el al[5] had EEG referral for 58.26% patients without history of seizure. The difference could be explained by the suspicion of seizure disorder. The main reason for EEG referral was similar in our study and that of and Molokomme and Subramaney[8] ie the exclusion of seizure disorder which is the reason in requesting for EEG in itself.

Compared to our study (43.3%), the study by Shrestha el al (58.3%) [5] had more EEG abnormality. The major abnormality pattern in EEG was slow waves in our study (46.2%)  and Molokomme and Subramaney(8.2%)[8]but spike and wave in (28%) Chowdhary et al[7].

In study by Shrestha el al[5] majority had abnormal EEG (25.8%) recording with seizure history. Similar finding was observed in our study with 47.88% EEG abnormality in those with seizure disorder. But the majority (55.8%)of EEG recording was normal in Bhagat et al.[6] study done among the epileptics. This difference is natural as EEG is just a cross sectional record of the brain activity and 50% of patients with epilepsy can have normal EEG which does not exclude epilepsy[9].

The seizures were not classified in both of the studies of Nepal when referred for EEG, however in our study abnormal EEG was noted in 26.21% of generalized epilepsy compared to localized (14.02%)as per EEG diagnosis. In contrary to our study, study from Bangladesh by Chowdhary et al[7] reported more (28%) localization related epilepsy than generalized (12%).Though majority of seizures were unclassified in Bhagat et al[6] too from Nepal, among the classified ones generalized seizure(75.85%) was higher than partial seizure (23.21%).The disparity between seizure type clinically and as per EEG could be the misdiagnosis of secondary generalized seizure with generalized tonic clonic seizure, lack of adequate information.

Studies from different countries have showed the range of patients with generalized seizures was 50–69%, and 31–50% had partial seizures[10],[11],[12],[13].The predominance of generalized epilepsy may be due to lack of adequate history and standardized classification.

Referral from Psychiatry had abnormal EEG in 40 % of Bangladesh study[7],46.47% in our study but only 8.2% in the African study[8]. This could be due to variation in sample size, interpreter variability. Only 1 patient each of dissociative with and without depression had abnormal EEG in our study which was similar to the study by Molokomme and Subramaney[8]with EEG abnormality in 2 mood disorder and none of the psychiatric disorder had any positive EEG finding in the study by Chowdhary et al[7].

Conclusion

The majority of patients referred for EEG fell in the young age of their life as well as the EEG finding was positive in the age group. EEG is a noninvasive test with its limitation but still can be used for causes other than seizure. Majority of the epilepsy patients give normal EEG recording due to its low sensitivity. The diagnosis does not completely rely exclusively on its reading. Abnormal EEG findings cannot be conclusive to diagnose seizure disorder as there will be normal record even in the patient with epilepsy and vice versa. Provocation methods, repeating the routine EEG, Video EEG may yield more positive results and thereby help in management of patient with epilepsy, psychiatric and organic manifestations. For the prediction of abnormality in EEG, more variables and their association with seizure/ organicity need to be looked into.

Limitation

All data were collected from a single tertiary care center of Nepal. The clinical details are not available. There can be an interpersonal variability in reporting of EEG reading which can affect the outcome of the study.

Table1: Demographic profile of patients with EEG evaluation

Table 2: EEG findings of patients with EEG evaluation

Table 3: Distribution of abnormal EEG as per demographics

Table 4: Predictors of abnormal EEG