search text   Advanced Search

Post-publication Peer Reviews to:

ELECTRONIC ARTICLE: Christian M. Korff and Douglas R. Nordli, Jr Paroxysmal Events in Infants: Persistent Eye Closure Makes Seizures Unlikely Pediatrics 2005; 116: e485-e486 [Abstract] [Full text] [PDF]

P3Rs: Submit a response to this article

P³Rs published:

Eye Closure During Paroxysmal Events: Unjustified And Potentially Misleading Conclusions

Aubrey J Cunnington   (29 October 2005)

Eye Closure During Paroxysmal Events: Unjustified And Potentially Misleading Conclusions 29 October 2005
Aubrey J Cunnington, Specialist Registrar in Paediatrics Whittington Hospital, London UK Send letter to journal: Re: Eye Closure During Paroxysmal Events: Unjustified And Potentially Misleading Conclusions acunning{at}doctors.org.uk Aubrey J Cunnington	Korff and Nordli present an interesting study of the proportion of infants with seizures who have persistent eye closure throughout the event (1). They use this data to draw conclusions about the usefulness of this observation to rule out seizures as the cause of a paroxysmal event, in other words, to use this as a diagnostic test. Their results do not justify these conclusions and under some circumstances these conclusions could be very misleading. This retrospective study examined only infants known to have seizures. There was no comparison with infants having paroxysmal events due to causes other than seizures. Even if the infants studied were typical of those having paroxysmal events due to seizures (which is unlikely) the only characteristic of the diagnostic test which can be calculated from the data is the sensitivity. In order to know whether persistent eye closure increases or decreases the likelihood of an infant having had a seizure as the cause of a paroxysmal event, it is essential to know what proportion of infants have persistent eye closure both with AND without a seizure (2). If the proportion of infants having a non-ictal paroxysmal event with persistent eye closure is less than 6.6% (for which the likelihood ratio is 1) then the conclusion of the study is misleading. Under these circumstances persistent eye closure would increase the likelihood of the event being due to a seizure. This study does suggest that persistent eye closure may be interesting to evaluate properly as a diagnostic test for the cause of a paroxysmal event in infancy. This would need to be done in a large prospective study of an unselected population of infants having paroxysmal events. Until this is done it is inadvisable to use persistent eye closure as a diagnostic tool to rule out seizures. References 1. Korff CM, Nordli DR. Paroxysmal events in infants: persistent eye closure makes seizures unlikely. Pediatrics. 2005;116:e485-e486 2. Guyatt G, Rennie D. Users' guide to the medical literature. Essentials of evidence-based clinical practice. Chicago, IL: American Medical Association Press; 2002 Conflict of Interest: None declared

	Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2008 American Academy of Pediatrics. All rights reserved.