Is the continued use of ocular pneumoplethysmography necessary for the diagnosis of cerebrovascular disease?

Abstract

The combination of duplex scanning and ocular pneumoplethysmography (OPG) has been used by many vascular laboratories for noninvasive evaluation of the carotid arteries. This study was undertaken to determine if OPG significantly improved the accuracy of duplex scanning alone. Three hundred eighty-five carotid arteries were studied in 190 patients with angiograms, duplex scans, or OPG. A total of 329 carotid arteries were examined with all three modalities. Three different criteria were used to interpret the OPG results, one each intended to deliver a high sensitivity, specificity, and overall accuracy. Depending on the criteria used, sensitivity of OPG alone for detecting hemodynamically significant (greater than 50%) stenosis ranged from 53% to 83%, with a specificity of 59% to 94%. Duplex scanning alone had a sensitivity of 87% and a specificity of 90% for similar lesions. If patients with only intracranial arterial stenosis were excluded, the sensitivity of the duplex scan rose to 91%. The addition of OPG to the duplex scan slightly increased sensitivity (range 91% to 95%; difference not significant) but significantly decreased specificity (range 56% to 84%; p less than 0.005). For detecting stenoses greater than 80%, the duplex scan alone had a sensitivity of 90% and a specificity of 88%. The addition of OPG to duplex scanning slightly increased sensitivity for these high-grade lesions but significantly decreased specificity (p less than 0.001) and overall accuracy. These results were unaffected by the presence or absence of intracranial disease. Because the addition of OPG to duplex scanning reduces specificity and accuracy without any significant increase in sensitivity, we no longer use it as part of our routine noninvasive cerebrovascular examination.