The overall sensitivity of the Wood’s lamp, compared to the gold standard slit lamp, to establish the diagnosis of common corneal and conjunctival abnormalities was found to be an abysmal 52%.

Topical antibiotic use for bacterial conjunctivitis was associated with improved rate of clinical cure compared to placebo. For early and late clinical cure with topical antibiotics, the NNT was 9 and 11, respectively.

In less experienced sonographers, ocular ultrasound for retinal detachment was specific but not sensitive.  It's useful to rule it in, but don't use it as the sole means to rule it out.

CT was specific for open globe but not sensitive.  The best CT predictors of open globe were change in globe contour and vitreous hemorrhage.  Open globe is a clinical or surgical diagnosis.  CT alone can't rule it out.

This study found 1% tetracaine for 24 hours appeared to be safe for treating corneal abrasion pain.