Value of serum prolactin in epileptic and some non-epileptic events

ABSTRACT

A seizure or convulsion is a common problem in pediatric age group. Other events such as breath holding attacks and syncope may simulate seizures. A thorough history and accurate description by the parents or eye witness is necessary for making diagnosis of seizures. When the description is unreliable, the diagnosis becomes difficult. The adverse effects of long term antiepileptic drugs, cost of therapy as well as social implications make it essential to build a solid diagnosis before the decision of therapy. Electroencephalography [EEG] which is the common tool to confirm the diagnosis of epilepsy, could be normal in up to 40% of patients in the inter-ictal period Twenty-four hours EEG monitoring, ambulatory EEG are used in cases with diagnostic uncertainly. Such modalities are not easily available in our locality as well as our country and so, available cheaper and easily accessible alternatives are required. The aim of the study is to evaluate the diagnostic value of post-ictal serum prolactin in the diagnosis of epilepsy and its specificity to differentiate epileptic from non-epileptic events in our locality where EEG monitoring is not available. The study included sixty eight patients and twenty one healthy controls of matched age and sex. Patients were divided into two groups 45 epileptic patients and 23 non-epileptic patients [15 patients with breath-holding spells and 8 with syncope]. A thorough history, full neurological examination as well as general examination were done. Conventional EEG was carried out in the awake state. Serum urea, creatinine calcium, sodium and potassium were performed. Serum prolactin was performed by enzyme immunoassay. Two samples for prolactin assay, the first within one hour in attack and the second After 24 hours. Generalized tonic-clonic [G. T.C.] convulsions were the commonest type [57.7%], complex partial type of convulsion were the 2[nd] in frequency [17.8%], generalized tonic [G.T], atonic and partial with secondary generalization were 6.7% for each, and lastly simple partial were 4.4%. The serum level of prolactin of the epileptic patients [21.7 +/- 2.5 ng/ml] was significantly higher than that of each of the non-epileptics group [9.4 +/- 2.3 ng/ml], [p<0.0001], as well as the control group [7.2 +/- 1.6 ng/ml], [p<0.0001]. Furthermore, the serum level of prolactin was significantly higher among the non-epileptics than the control group [p<0.05]. Significant differences were also reported between first and second samples in the epileptic group [p<0.001]; while the difference was insignificant between the first and second samples in non-epileptic group. A significant negative linear correlation [r=-0.8] and [P<0.01] was detected between the serum prolactin and time lapsed after the fit in the first sample. Taking mean +/- 2SD as a cut off value, sensitivity of serum prolactin for diagnosing epileptic fits was 93.3% and specificity was 65.2%. Post-ictal serum prolactin in the first hour after the fit is a sensitive parameter for diagnosis of epileptic patients and highly specific parameter to differentiate between epileptic patients and patients with syndrome that mimic epilepsy and could build a solid diagnosis for doubtful cases of epilepsy