What is the angle of a banana? The difficulty in reliable assessment of hypospadias chordee

Abstract

Introduction: The degree of chordee associated with hypospadias impacts operative management. Unfortunately, poor inter-observer reliability in assessing chordee by multiple methods in vitro has been demonstrated. This variability may be related to the fact that chordee is not a discrete angle, but rather an arc-like curvature similar to that of a banana. On an attempt to improve this variability, we assessed the inter-rater reliability of a novel method of chordee measurement and compared it to measurements with a goniometer both in vitro and in vivo.

Materials and methods: In vitro assessment of curvature was performed using 5 bananas. In vivo chordee measurement was performed during 43 hypospadias repairs. On in vitro and in vivo cases, chordee was assessed independently by faculty and resident physicians. Angle assessment was performed in a standard manner with a goniometer and with a smartphone app using ruler measurements of the length and width of the arc (Summary Figure). The proximal and distal aspect of the arc to be measured was marked on the bananas, whereas the penile measurements were taken from the penoscrotal to the sub-coronal junctions.

Results: In vitro banana assessment demonstrated strong intra- and inter-rater reliability for length (0.89 and 0.88, respectively) and width measurements (0.97 and 0.96). The calculated angle demonstrated an intra- and inter-rater reliability of 0.67 and 0.67. The banana goniometer/protractor measurements were weak with an intra-rater and inter-rater reliability of 0.33 and 0.21. With hypospadias chordee, the inter-rater reliability was strong for length and width measurements (0.95 and 0.94) and 0.48 for calculated angle. The inter-rater reliability of the goniometer angle was 0.96. Further assessment of inter-rater goniometer reliability was performed relative to degree of chordee as characterized by faculty. The inter-rater reliability for ≤15°, 16-30, and ≥30° was 0.68 (n = 20), 0.34 (n = 14), and 0.90 (n = 9), respectively. When the goniometer angle was classified as ≤15, 16-30, or ≥30° by one physician, it was classified outside of this range by the other physician 23%, 47%, and 25% of the time, respectively.

Discussion: Our data demonstrate significant limitations of the goniometer for assessing chordee in vitro and in vivo. We were unable to demonstrate significant improvement in chordee assessment using arc length and width measurements to calculate radians.

Conclusions: Reliable and precise techniques for measuring hypospadias chordee remain elusive and draw into question the validity and usability of management algorithms employing discrete values.

Keywords: Goniometer; Hypospadias chordee; Inter-observer reliability; Penile curvature.