442 ULTRASONOGRAPHIC TESTICULAR ECHOTEXTURE IN NELORE BULLS

Abstract

The objective of this research was to determine the minimum area representative of the testicular parenchyma echotexture in Nelore bulls, and to evaluate the presence of echotexture homogeneity between the right and left testicles and between different regions of testicle. Twenty-nine Nelore bulls raised under extensive management and free of reproductive disorders were used. The average body weight was 352.8 ± 51.0 kg and age ranged between 18 and 24 months. Echotexture was assessed using a portable ultrasonography device equipped with a 7.5-MHz linear transducer. All measurement was done by a single, experienced operator, and scanner settings that affect image attributes (i.e. proximal, distal, and total gains; focus area; brightness and contrast) were standardized to predetermined values. In the examinations, a layer of gel was applied to the transducer scanning area, which was positioned in the testicular longitudinal plane, proceeding with enough pressure to capture an image. The frozen images were obtained with minimal artifacts (refraction, dispersion, and attenuation of ultrasound waves) taking care to avoid the mediastinum testis and tunica albuginea, and recorded as .bmp files in a USB pen drive connected to the ultrasound device. Echotexture was defined in terms of mean pixel value quantified using values from 0 (black) to 255 (white) through the ImageJ 1.41 software. After the selection of a specific area of testicular parenchyma, the average tonality of the gray pixels in the selected region was calculated by the software and expressed in 256 gray-scales. Squares selections of 400 (20 × 20), 1600 (40 × 40), 3600 (60 × 60), and 6400 (80 × 80) pixels were assessed from images of the extremitas capitata, middle, and extremitas caudata regions of both testicles to determine the minimum sampling area of an image needed to represent the echotexture of testicular parenchyma. The normal distribution verification (Lilliefors test) of the variable mean pixel value was tested in each area, followed by variance analysis. The averages were compared through Tukey’s test with 5% significance. There was no significant difference between the different pixel areas assessed (P > 0.05), indicating that all were samples representative of testicular echotexture. Furthermore, the extremitas capitata, middle, and extremitas caudata regions showed no significant difference in echotexture (P > 0.05) regardless of the area analyzed, and there was no significant difference between the right and left testicles (P > 0.05) in any of the assessed areas. In conclusion, this research shows that testicular echotexture assessments can be made in Nelore bulls through selections of 400, 1600, 3600, or 6400 pixels, and that images can be captured from any region of any testicle.

Fundect - Fundação deApoio ao Desenvolvimento do Ensino, CiÊncia e Tecnologia do Estado de Mato Grosso do Sul.