Limitations of quantitative oculoplethysmography and of directional Doppler ultrasonography in cerebrovascular diagnosis: assessment of an air-filled OPG system.

500 consecutive patients were evaluated for extracranial disease of the internal carotid arteries by an automated, air-filled, digital oculoplethysmographic system (OPG) of the Kartchner type (Zira) and by supraorbital (SO) and supratrochlear (ST) directional Doppler ultrasonography. Cerebral arteriograms were performed in 58 patients (110 vessels), and OPG timing criteria for detecting hemodynamically significant carotid artery stenosis (60% or greater diameter reduction) were ascertained. Optimal criteria were a delay of one ocular pulse, relative to the other, of greater than 12 msec; and a delay of an ocular pulse, relative to the earlier ear (external carotid) pulse, of greater than 36 msec. These criteria correctly identified 73% of vessels with 0 to 59% stenosis and 76% of vessels with 60 to 100% stenosis. However, in 26% of the vessels, OPG was either inconclusive or inaccurate. Correct diagnosis of bilateral hemodynamically significant carotid artery stenoses was made by OPG in 6 of 9 affected patients. SO Doppler was normal in 70% of vessels with 0-59% stenosis, and abnormal in 75% of vessels with 60-100% stenosis. Corresponding percentages for ST Doppler were 95% and 44%. Abnormal Doppler responses to compression of contralateral facial branches were predictive of intracranial cross-collateralization in only 25% of patients. These results suggest that both quantitative OPG in its present form and directional Doppler studies have serious limitations as non-invasive diagnostic methods.