This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
—When we began our evaluation of the cocaine test for Horner's syndrome, we assumed that the net change in anisocoria (anisocoria after cocaine instillation minus baseline anisocoria) would be the best indicator to distinguish patients with an oculosympathetic deficit from patients with pseudo-Horner's syndrome (ptosis and/or miosis). However, at the end of our study, the logistic regression analysis indicated that measuring anisocoria after cocaine instillation was just as good, if not better, in distinguishing the patients with Horner's syndrome from the normal control group. The Figure shows how this might be possible. We have plotted the initial anisocoria vs the anisocoria after cocaine instillation for both the patients with Horner's syndrome and the normal control subjects reported in our study.There was, indeed, a baseline anisocoria in many of our normal control subjects, and some overlap with the patients with Horner's syndrome (Figure). Of equal importance is that
Kardon RH, Denison CE, Brown CK, Thompson HS. The Cocaine Test and Horner's Syndrome-Reply. Arch Ophthalmol. 1990;108(12):1667-1668. doi:10.1001/archopht.1990.01070140021006