An analysis of MRI findings in patients referred with fits

Results: There were 54.5% (n=126) males and 45.5% (n=105) were females. 115yrs age group consisted of 39.4% (n=91). Majority, 67 % had generalized tonic clonic (GTC) seizures. There were 225 newly diagnosed fits and 6 follow-up cases. MRI findings were abnormal in 57.6% (n=133).Findings in 36.8% were cerebral tumors, out of which 48.9% were gliomas and 16.3% were meningiomas. >15 year group had neoplasms in 48.5% compared to 8.1% in 1-15 year group. Regarding co-morbidities, 9% (n=21) had congenital/perinatal conditions versus 8.2% (n=19) with malignancy


Introduction
Fits are clinical manifestations of temporary disturbance in the electrical activity of the brain.It is a common symptom, experienced by 4% of all people at least once in life time 1 .Recurrent fits are known as 'seizure disorders' or 'epilepsy' 2 .In Sri Lanka seizure disorders or epilepsy is the commonest chronic disabling neurological condition, with a reported incidence of 50 per 100,000 per annum and crude prevalence ranging from 9-11 cases per 1000 3 .Even though it is a disease seen in all age groups, prevalence of seizures is higher in the 5-20 years age group 5,8 .
Fit is a common indication for Magnetic resonance imaging (MRI) of brain worldwide, since it is a proven reliable modality to investigate the underlying cause.The first episode of seizure is regarded as an indication for MRI, according to the current internationally accepted medical practice 4 .MRI is a non-invasive and highly sensitivity imaging modality in neurology.However it is expensive and not widely available.Therefore, in order to provide maximum benefit to the population, it is important to assess its efficacy and usefulness in diagnosis of the underlying cause of epilepsy, in different demographic and clinical categories.No such evaluation has been performed in Sri Lanka.Our aim is to assess variations in MRI findings related to various age groups, special risk factors, different co-morbidities, seizure type etc.This information is useful in forming imaging guidelines and clinical management protocols.

Materials & Methods
A retrospective, descriptive hospital based cross sectional study was conducted at

Age and Sex Distribution
Age distribution is summarized in Figure 1.
In the study population MRI findings were abnormal in 57.6% (n=133) while 42.5% (n= 98) of MRI's were negative.With P<0.05, there is a statistically significant difference between the patients with fits, having a positive finding versus those having a negative MRI. MRI findings among males and females is demonstrated in Figure 2 Fig. 2

Fig. 1. Age distribution
There is no statistically significant difference in males with fits having a positive finding compared to females with fits having a positive MRI finding in the study population, P>0.05 .
As shown in Figure 4, those presenting with GTC have a higher chance (61.9%) having positive findings on MRI (CI±7.65)while all cases of multi-focal fits have negative MRI.

Age Groups and MRI Findings
There is a statistically significant difference in detecting a positive MRI finding in >15 years age group compared to 1-15 years age group (P<0.05).It is demonstrated in Figure 5.

Discussion
An abnormal brain MRI was found in 57.6% of our study population.A study by the neuro imaging unit in Florida reveals that, in patients with first seizure, an epileptogenic lesion was identified by MRI in 14% while in intractable epilepsy MRI identifies the pathologic substrate in 82-86% 4 .There is a difference in the proportions of patients who had a positive MRI in the two study populations.This is due to the fact that, in their setup the first episode of seizure was indicated for MRI 4 whereas in ours, patients either present late or undergo other imaging modalities 3,4 .
Reported prevalence of seizures is more in the 5-20 year age group 5,8 .In our study also the commonest age group was years.
Cause of most seizures is not detected and considered idiopathic (70%) 5 .In our study an abnormality was not detected in 42.5% and the >15 year age group have a higher chance of having a positive finding on MRI.
MRI is 100% sensitive in detecting an epileptogenic neoplasm 6,9 .Cerebral neoplasms is the most frequent (36.8%)MRI diagnosis in our study group.Glioma is the commonest neoplasm followed by Meningioma.This trend was more marked in > 15 years age group.Patients presenting with generalized tonic clonic (GTC) seizures had a higher frequency of lesion detection on MRI compared to other seizure types.There was no statistically significant difference regarding the presence of underlying cause for epilepsy between male and female patients.
Awareness of these possible correlations between presenting clinical patterns and MRI findings is useful for radiologists to identify epileptogenic lesions.There were some limitations in our study.We could not review histopathology reports to confirm the radiological diagnosis as specimens are sent to different hospitals.Another limitation is that these MRI scans have been done with a 1T system which has less sensitivity compared to higher field systems.
In conclusion, performing MRI in patients with fits significantly contribute to the detection of underlying cause.We also noted a variation of MRI findings in relation to seizure type, age group etc. which is helpful in MRI interpretation.

Table 1
illustrates the MRI findings, Table2analyzes types of neoplasms and MRI findings with the age distribution are shown in