PREMATURE MORTALITY

People with epilepsy have reduced life expectancy, and this is a consistent finding regardless of location. In high‐income countries (HIC), the risk of premature mortality for people with epilepsy, based on national life tables, is two to three times greater than that in the general population.¹ This figure, however, is likely to be an underestimate as data from life tables relies on death certificates, in which epilepsy is not often listed. The risk is more pronounced in low and middle‐income countries (LMIC), with a standardised mortality ratio (SMR) of 4.5 to 5.² Premature mortality is a substantial individual and societal burden.

Several factors significantly influence the risk of premature mortality. The risk is highest shortly after diagnosis, with SMR peaking in the initial years and remaining elevated afterwards. This initial peak is due to death related to underlying disease, such as cancer, stroke, infection, or degenerative conditions. Younger people, particularly those under 40, are at a higher risk of premature death. Seizure severity and type play a crucial role, with convulsive seizures strongly associated with accidents, sudden unexpected death in epilepsy (SUDEP), and status epilepticus. Treatment‐related factors also contribute to premature mortality. Antiseizure medication (ASM) polytherapy is associated with a higher risk than monotherapy, which is likely to be a surrogate for the severity of epilepsy. Non‐adherence to prescribed ASM and subsequent breakthrough seizures represent a preventable risk factor.

Causes of death are diverse and mainly driven by epilepsy‐related events and comorbidities.³ While only about 1 in 20 adult deaths in incident cohorts are directly attributed to epilepsy, this figure increases to about ten per cent in children. In prevalent cohorts, people with drug‐resistant epilepsy (DRE) are usually overrepresented. Epilepsy‐related death is responsible for over one‐third of deaths, including some directly seizure‐related and accident‐related injuries, suicide, and SUDEP. Death could also be directly related to treatment, resulting from idiosyncratic reactions and overdoses of ASM or complications of epilepsy surgery.

The presence of comorbidities significantly impacts the risk of premature mortality, as most people die of associated comorbidities, such as vascular and respiratory disorders. Autoimmune conditions, psychiatric disorders, and cancers also contribute. Accident‐related deaths and suicide are common. In HICs, status epilepticus (SE) leading to death is relatively rare among people with established epilepsy. Most epilepsy‐related deaths are sudden and unexpected without evidence of a cause of death at a post‐mortem examination: sudden unexpected death in epilepsy (SUDEP).

SUDEP AND SUDEP RISK

SUDEP is defined as the sudden, unexpected death of a person with epilepsy, with or without evidence of a seizure, excluding deaths from trauma, drowning, or SE, and without a cause found during autopsy.³ SUDEP is the leading direct epilepsy‐related cause of death and second only to stroke in terms of years of potential life lost in the United States.⁴ The incidence of SUDEP in adults is estimated at 1.2 per 1,000 persons annually, increasing to 9.7 per 1,000 in those considered for epilepsy surgery.⁵ In people with DRE, the risk is, on average, around five deaths per 1,000 persons/year.

Convulsions, when occurring more than three times annually, are a significant risk factor, increasing SUDEP odds nearly 20‐fold.⁶ Having convulsive seizures in the previous year and not sharing a bedroom with another person escalates this risk 67‐fold.⁷ SUDEP occurs even in self‐limiting epilepsy syndromes, suggesting that although the syndrome is ‘benign’, convulsions are not.

Intellectual disability (ID) increases the likelihood of SUDEP by four to five times in people with ID and epilepsy, compared to those with epilepsy but without ID. Nocturnal or sleep‐related seizures also raise the risk three to fifteenfold.⁸ Comorbid depression may further contribute to this elevated risk.⁷ While factors such as epilepsy duration and the number of ASM have been suggested, robust individual SUDEP risk stratification remains challenging. Risk inventories such as the SUDEP‐7 have limited reliability in distinguishing individuals who died suddenly from controls living with epilepsy.^9,10 Simplified inventories, including frequency of convulsions, overall seizure frequency, and ID, have improved predictive accuracy.¹¹ Bayesian methodologies, incorporating factors like alcohol excess, younger age at epilepsy onset, and family history, have also shown promise.¹²