Purpose: To develop and test a clinical case definition for identification of generalized tonic-clonic seizures (GTCSs) by community-based health care providers.
Methods: To identify symptoms that can help identify GTCSs, patients with history of a jerky movements or rigidity in any part of the body ever in life were recruited from three sites: the community, secondary care hospital, and tertiary care hospital. These patients were administered a 14-item structured interview schedule focusing on the circumstances surrounding the seizure. Subsequently, a neurologist examined each patient and, based on available investigations, classified them as GTCS or non-GTCS cases. A logistic regression analysis was performed to select symptoms that were to be used for case definition of GTCSs. Validity parameters for the case definition at different cutoff points were calculated in another set of subjects.
Results: In total, 339 patients were enrolled in the first phase of the study. The tertiary care hospital contributed the maximal number of GTCS cases, whereas cases of non-GTCS were mainly from the community. At the end of phase I, the questionnaire was shortened from 14 to eight questions based on statistical association and clinical judgment. After phase II, which was conducted among 170 subjects, three variables were found to be significantly related to the presence of GTCSs by logistic regression: absence of stress (13.1; 4.1-41.3), presence of frothing (13.7; 4.0-47.3), and occurrence in sleep (8.3; 2.0-34.9). As a case definition using only three variables did not provide sufficient specificity, three more variables were added based on univariate analysis of the data (incontinence during the episode and unconsciousness) and review of literature (injury during episode). A case definition consisting of giving one point to an affirmative answer for each of the six questions was tested. At a cutoff point of four, sensitivity was 56.9 (47.4-66.0) and specificity, 96.3 (86.2-99.4). Among the 197 GTCS and 26 new non-GTCS patients recruited from hospitals from select SEAR Member Countries, in phase III, the sensitivity of this clinical case definition was 72% and specificity, 100%. A stratified analysis by gender in all the three phases did not show any differences between the sexes.
Conclusions: Based on these criteria, we recommend that all patients with a history of two or more episodes of jerking or rigidity of limbs, having a score of > or =4 in the case definition, be identified as having GTCSs and started on antiepileptic medications. This clinical case definition can be very useful for community-based health care providers to identify and manage cases of GTCSs in the community. This should play a major role in the reduction of treatment gap for epilepsy in developing countries.