Abstracts

Rationale: With ever expanding health costs, reduction in rehospitalization rates has been a major goal of efficient healthcare delivery. The use of continuous EEG (cEEG) monitoring for the diagnosis of non-convulsive seizures and/or status epilepticus (NCS/NCSE) has seen tremendous growth in last decade or so. While significant research efforts have been directed towards early NCS/NCSE identification among high risk individuals, the risk of rehospitalization among patients undergoing cEEG and the yield of repeat cEEG in such situations is yet to be explored. We have tried to address this question is our current study. Methods: After IRB approval, our prospectively maintained cEEG database was searched from 01/01/2015 to 12/31/2015 to identify patients = 18 years of age who were rehospitalized within 30 days of undergoing index cEEG and had a repeat cEEG monitoring during the readmission. The indications for cEEG during rehospitalization were divided into unexplained altered mental status (uAMS) or seizure like episodes (SLE). Primary outcome was electrographic seizures on repeat cEEG and secondary outcome was to find patient who develop epilepsy with in 30 days of index cEEG. Statistical tools including chi square and t-test were used as required. Results: A total of 57 patients were rehospitalized and underwent a repeat cEEG after index monitoring. This accounted for 2.3% of all unique patients monitored during 1 year study period. The mean age of study population was 61 ± 18.1 years and included 26 (45.6%) females. Thirteen (22.8%) patients had history of epilepsy. Primary etiology at time of first cEEG was structural in 38 (66.7%) [acute 20 (52.6%), remote 3 (7.9%), progressive 15 (39.5%)] patients and acute non-structural in the rest of them. Electrographic seizures were noted in 15 (26.3%) patients on index cEEG. The mean duration between two cEEG sessions across rehospitalization was 16.2 ± 7.2 days. Twenty three (40.4%) patients had suffered a SLE prompting readmission and repeat cEEG. Additional 5 (8.8%) patients rehospitalized otherwise had SLE after admission requiring cEEG. Rest underwent monitoring due to uAMS. Eighteen (31.6%) patients had a new etiology diagnosed on rehospitalization. Eleven (19.3%) patients had electrographic seizures noted on cEEG during rehospitalization. Five of them had seizures during index cEEG, 2 had epilepsy history. Rest 4 (7%) patients, without new etiology, developed epilepsy during this short period. Univariate analysis only showed SLE preceding repeat cEEG as predictor of seizures (Table 1). Conclusions: A small fraction of patients undergoing cEEG were rehospitalized in our cohort. One in five patients had electrographic seizures on rehospitalization with 7% of total study population developing epilepsy in a very short duration. SLE prior to repeat cEEG on rehospitalization seem to predict patient at risk of seizures. Larger, multicenter collaborative studies would be helpful in finding patients undergoing cEEG who are vulnerable to 30-day readmission, along with identification of patients at highest risk of developing symptomatic epilepsy. Funding: None