Primary outcome measures

doi:10.1136/bmj.b3368

Published 18 August 2009, doi:10.1136/bmj.b3368
Cite this as: BMJ 2009;339:b3368

Letters

Antihistamines for anaphylaxis?

Primary outcome measures

The first 100% of the full text of this article appears below.

Andreae and Andreae recommend that future trials of antihistaminesin anaphylaxis should use a primary outcome of "whether antihistaminesresult in reduced mortality."¹

Given that death from anaphylaxis is rare, and that fewer thanhalf of deaths from anaphylaxis occur before reaching emergencymedical care, this outcome is not useful.

Perhaps we should first establish, using prospective observationalstudies in both hospital and community settings, the clinicaloutcomes with a simple protocol of supportive care includingadrenaline and fluid resuscitation.

The results of such studies may remove the need for a clinicaltrial of antihistamines. Alternatively, they may suggest alternative,clinically relevant primary outcomes for a placebo controlledtrial of antihistamines—for example, the incidence ofadverse effects or delayed phase ("biphasic") reactions, orboth.

Cite this as: BMJ 2009;339:b3368

Simon G A Brown, professor, emergency medicine¹

¹ Royal Perth Hospital, University of Western Australia, Perth, WA 6000, Australia

simon.brown@uwa.edu.au

Competing interests: None declared.