Inter and intra-rater reliability of accuracy of testicular volume evaluation: a simulation study

Abstract

Background: The measurement of testicular volume is a practical and inexpensive method of pubertal staging in adolescents. Traditionally, volume is measured using orchidometers or calipers, the most widely used of which is the Prader Orchidometer. However, there is conflicting evidence of its use with regards to reliability in different hands. Objectives: To determine the accuracy of the Prader orchidometer in measuring testicular volume, by comparing the resultant measurement with the actual testicular volumes. Information on inter-observer and intra-observer variation was obtained for 16 investigators. Furthermore, the effect of factors such as clinical experience, gender and training in using an orchidometer on the accuracy of testicular volume measurement were investigated. Methods: A total of 16 investigators were asked to measure a range of 4 testicular volumes, attached via prosthesis to child-sized mannequins. The volume of each testis was measured independently using the Prader Orchidometer. A proportion of investigators also repeated testicular volume measurements, blinded by measurements obtained by other investigators. Results: The mean testicular volume measured by 16 examiners for 4 sizes at 20ml, 10ml, 5ml and 3ml. There were 4 male investigators and 12 female investigators. Fischer’s exact test found males to be significantly more accurate at measuring testicular volume than females (P =0.04). Chi-squared values were measured at each volume and showed no significant difference between those with clinical paediatric endocrine experience compared with those with no experience. Inter-observer reliability was measured using coefficient of variation due to our sample size of 16. This showed no significant difference between investigators measuring each testicular volume. However, it showed that larger volume testicles were more likely to be accurately measured compared to smaller volumes. Results showed that the 20ml volume had the smallest variability of 0.109 and 0.130 for right and left respectively ad the 3ml volume had a higher variation at 0.456 and 0.559 for right and left testicles respectively. Interestingly, variation was higher at all volumes in the left testicle. Furthermore, the 5ml volume has an unexpectedly low variability which may be attributed to its link with puberty and the need for endocrinologists to be skilledat identifying this particular volume due it’s clinical relevance. 7 people were trained vs 9 untrained in using orchidometers. Orchidometer training R20 A 3 B 10 C 5 D L20 A 3 B 10 C 5 D % accuracy trained 3/7 42.9 2/7 28.6 5/7 71.4 2/7 28.6 6/7 85.7 FE: 4/7 57.1 FE: 6/7 85.7 FE: 2/7 28.6 FE: %accuracy not trained 7/9 77.8 2/9 22.2 3/9 33.3 2/9 22.2 5/9 55.6 3/9 33.3 3/9 33.3 2/9 22.2 FE 0.302 1.00 0.315 1.00 0.308 0.615 0.060 1.000 Fischer’s exact: does having orchidometer training affect accuracy of results- not significant. Conclusion In conclusion, it was found that measurement of testicular volumes on paediatric testicular models is subject to great/little inter/intra-observer variation. The data suggests that there is a clinical need for formal training in testicular volume measurement as there is currently no established training in place. As a result, the innovative, custom-made testicular models for this study could have a role in simulation training as an educational tool.