Reliability of clinical measurements used in the determination of facial indices

Abstract

The purpose was to investigate the index by which facial height/width ratio is optimally determined and the reliability of anthropometrical measurements. The following measurements: zygion-zygion, sellion-gnathion, supraorbitale-gnathion, and interpupillary distance (IPD) were performed twice by the senior surgeon, and trainee surgeon, during an average interval of 18.6 days, on 50 patients (23 male and 27 female). The IPD was measured with a digital pupillometer (PM-600 - Nidek Co, Aichi, Japan), while the other parameters were measured with an anthropometrical Tessier Ruler. The Pearson correlation coefficient was determined for the intra-observer reliability (test-retest) and the intra-class co-efficiency was used to determine inter-observer reliability. The mean measurement error for the IPD with the digital pupillometer was less than 1mm. For all other facial measurements with the Tessier Ruler, horizontal and vertical, the error was more than 2mm. The intra- and inter-observer reliabilities (in vertical dimension) were better for the measurements of supraorbitale-gnathion, than those for sellion-gnathion. The Pearson correlation coefficient for the classical facial index was from 0.874 to 0.912 for the IPD/supraorbital-gnathion index. Measuring these distances with an anthropometrical ruler is not highly reliable, although the distance supraorbitale-gnathion can be measured more precisely than the distance sellion-gnathion. However, IPD measurements with a digital pupillometer are very reliable. These findings, together with the visibility of these parameters in the frontal view, lead to the introduction of a new facial index: IPD/supraorbitale-gnathion.