Factors associated with seizure in tramadol overdose: a 12-year retrospective study in Hong Kong

Early identification of patients at risk of seizure after acute tramadol overdose is crucial. Yet, current evidence has been inconclusive regarding the factors associated with the seizure. There is a lack of study on the Chinese population and the value of electrocardiogram (ECG) in risk prognostication has not been fully evaluated.

We aimed 1) to characterise the clinical presentations of isolated tramadol overdose in Hong Kong; 2) to study the frequency of seizure in such overdose and the factors associated with it using multivariable analysis; and 3) to explore the association between individual ECG parameters and seizure.

This was a retrospective observational study on consecutive patients reported to the Hong Kong Poison Information Centre for tramadol overdose from 1 January 2008 to 30 November 2020. Duplicate cases or patients with an overdose of other co-ingestants were excluded. Univariate and multivariable logistic regression was performed to identify factors significantly associated with seizures in general and seizures that occurred after emergency department (ED) arrival.

We identified 93 patients, including 54 women and 39 men, with a median age of 38 years during the study period. The majority was Chinese (90%) and only five patients had a history of seizure. The seizure occurred in 23 patients. Multivariable analysis showed that sinus tachycardia > 100 beats/min was significantly associated with a higher seizure risk (OR 27.95, 95% CI 2.22-351.84, p = 0.010) and age per 10 years increase was associated with a lower risk (OR 0.61; 95% CI 0.39-0.96, p = 0.033). No reliable predictors of seizure after ED arrival were identified. Many patients had ECG features consistent with sodium channel blockade but none of them was predictive of seizure.

A younger age and sinus tachycardia > 100 beats/min were associated with a higher seizure risk after tramadol overdose but no reliable predictors of seizure after ED arrival were identified. Individual ECG parameters were not predictive of seizure.