How accurate is eyeball measurement of curvature? A tool for hypospadias surgery

Correction of penile curvature or "chordee" is a major component in the management of hypospadias. Accurate assessment and management of penile curvature influence both short- and long-term outcomes of surgery.

The objective of this study is to investigate the accuracy of eyeball measurement and how does it compare to objective measurement by standard goniometry (SG) and smartphone app goniometry (AG).

A Dropbox file request link was shared with paediatric urologists on various social media platforms requesting participants to upload a picture of their index finger showing what they thought 30 degrees of curvature look like using their proximal inter phalangeal joint as the point of maximal curvature., The images were assessed using SG to measure the angle of curvature. The images were also assessed using AG by the principal investigator, a physician, a scrub nurse and a paediatric urology consultant., Statistical analysis was performed using SPSS statistics software version 26 (Armonk, NY: IBM Corp). A one sample t-test and a one-way chi squared test were used to evaluate significant frequency differences. Pearson correlation was used to compare AG measurements to test intra- and inter-observer reliability and to compare AG measurements vs SG measurements. Assuming 5-degree variability in goniometer measurements and 2-degree difference between the sample and population, the number of participants needed was calculated to be 49.

Fifty-two responses were received.32.7% of respondents simulated 30° accurately (17/52). A significant proportion (23/52, 44.2%) overrepresented the degree of curvature and 23.1% (12/52) underrepresented it (p = 0.01). Compared with objective measures, eyeball estimates differed by an average of 10° ± 1.5 SE. Measurements obtained by AG were comparable to measures obtained by SG and showed excellent intra-observer and inter-observer correlation (R = 0.983, P < 0.001).

We demonstrated a significant discrepancy between eyeball assessment of curvature and objective measurements in a cohort of hypospadiologists. This can be very relevant to intraoperative decision making. The limitation of the study is the use of a simulated model rather than assessment of curvature in patients with hypospadias. Another limitation is the lack of standardization of the way the pictures were taken.

We demonstrated a tendency among hypospadiologists to overestimate or underestimate curvature by an average of 10° on eyeball assessment. The use of App Goniometry shows excellent interobserver reliability and is comparable to standard goniometry in curvature assessment.