To determine the best clinical method for identifying peripheral vestibular nystagmus, by comparing eye movement examination with optic fixation, and with fixation removed using Frenzel's glasses, infra-red video-Frenzel's goggles or an ophthalmoscope, with results of electronystagmography.

One hundred patients referred for electronystagmography from the audiovestibular medicine clinic at Queen Alexandra Hospital, Portsmouth, were examined immediately before undergoing electronystagmography.

Video-Frenzel's goggles were highly effective at detecting peripheral vestibular nystagmus, with a sensitivity of 85 per cent (95 per cent confidence interval, 62.1–96.8 per cent) and a specificity of 65 per cent (53.5–75.3 per cent), compared with electronystagmography. Ophthalmoscopy had comparable sensitivity to Frenzel's glasses (used in the dark), i.e. 26.3 per cent (9.1–51.2 per cent) compared with 31.6 per cent (12.6–56.6 per cent), respectively. Frenzel's glasses as normally used in ENT clinics (i.e. in dim lighting) were ineffective, with a sensitivity of just 10 per cent (1.2–31.7 per cent).

Video-Frenzel's goggles should be used in all clinics with substantial numbers of balance-impaired patients. Traditional Frenzel's glasses have no place in clinical practice unless formal black-out facilities are available.