Home Stroke & neuro Paramedics miss many non-convulsive seizures
Stroke & neuro
Paramedics miss many non-convulsive seizures
Prehospital recognition and differentiation of seizure presentations by paramedics and ambulance nurses: A scoping review
Seizure · 2026-10-30 · Added 14 November 2026
An Australian scoping review of 18 studies found paramedic recognition of seizure presentations remains suboptimal, with sensitivity of 54.0–58.5% and non-recognition of non-convulsive seizures reaching 53.0–81.4%.
- Study type
- Systematic review
- Design
- Scoping review (PRISMA-ScR) of 18 studies across five databases
- Participants
- 18 studies (16 paramedic, 2 ambulance nurse) from 1,264 citations
- Main finding
- Sensitivity 54.0–58.5% and specificity 95.9–96.6% against variable reference standards; recognition rates 56.8–89.7%; non-convulsive seizures missed in 53.0–81.4% of cases
- Caveat
- Included studies used variable reference standards and EMS workforce configurations, limiting pooled interpretation
- Bottom line
- Hypothesis-generating
Why it matters in Australia (High relevance). Diagnostic accuracy of seizure recognition is a crew-level problem everywhere, and the review covers paramedic practice directly comparable to Australian scope.
Learning goal
Recognise non-convulsive seizure presentations in the field to reduce missed diagnoses.
More
- Topic: Stroke & neuro
- Daily scan: 14 November 2026
- Podcast: Paramedic Papers Daily, 14 November 2026
Related papers on Stroke & neuro
-
Most fatal brain bleeds die before neurosurgery
A Finnish–New Zealand population cohort of 12,912 patients with aneurysmal subarachnoid haemorrhage found 70.5% of 30-day deaths occurred without neurosurgical unit admission, with higher proportions in districts farther from a neurosurgical centre.
-
Ultrasound added to stroke scales for LVO triage
A German exploratory reanalysis of the 20-patient SONAS trial found portable non-imaging brain perfusion ultrasound combined with prehospital stroke scales sorted suspected stroke into likely-LVO, likely-non-LVO and unassessable groups better than either method alone.
-
Stroke triage app adopted but not embedded
A Dutch mixed-methods evaluation of the Stroke Triage App across two EMS regions found good initial adoption (median 120 app triages/month) but inconsistent routine integration, with workflow fit and limited perceived added value the main barriers.
-
M-DIRECT stroke triage scale revalidated
A Spanish prospective study of 1,950 prehospital code-stroke patients revalidated the M-DIRECT scale five years after implementation, showing 77% sensitivity and 76% specificity for identifying endovascular-treatment candidates, with a positive score more than tripling the odds of EVT.