Título

INTRA AND INTER-OBSERVER RELIABILITY OF A MODIFIED DISTRACTION TEST TO ASSESS LOWER EYELID HORIZONTAL TENSION

Objetivo

The objective of this study was to assess intra and inter observer reliability of a new method to quantify lower eyelid horizontal tension using digital image processing.

Método

A prospective study included 16 individuals aged between 25 and 50 years with no previous eyelid disease or surgery. Photographic acquisition and measurements were carried out by the same investigator in two distinct moments, and then by a different investigator in another time moment. Two pictures of the left eye were taken each time: one in primary position and a second one with inferior traction applied in the center of the lower eyelid, distracting it downward. The difference of the two measurements was considered the amount of displacement of the lower eyelid representing the eyelid horizontal tension. Measurements were processed using the Image J software. Bland and Altman concordance test was used to assess intra and inter-observer reliability.

Resultado

On intra-observer assessment, Bland and Altman concordance test evidenced a difference of 0.00025mm between the neutral picture measurements (p=0.998, SD= 0.38), with an upper concordance limit of 0.755mm and a lower concordance limit of -0.754mm. On the pictures with downward traction of the eyelid center, there was a difference of 0.113mm between measurements (p = 0.475, SD 0.61), with an upper concordance limit of 1.094mm and a lower concordance limit of 1.320mm. On inter-observer assessment, paired samples statistics evidence high agreement: mean measurements on neutral pictures were 1.74mm (SD 0,70) for the main investigator and 1.74mm (SD 0,72) for the second observer. On pictures with downward traction of the eyelid center, mean measurements were 4.79mm (SD 0,95) for the main investigator and 4,91mm (SD 0,77) for the second observer.

Conclusão

There was a high intra and inter-observer concordance with the studied method. The purposed method is simple and easily reproducible, and as far as we know is the first description of a method for quantifying lower eyelid horizontal tension based on digital image processing.