Glenoid Bone Loss After Unidirectional Versus Combined Shoulder Instability: Letter to the Editor

Dear Editor:

I read with pleasure the article by Annette H. Yoon et al, ³ titled “Comparison of Glenoid Bone Loss After Unidirectional Versus Combined Shoulder Instability in a Military Population” and would like to offer additional commentary on the extrapolated conclusions regarding the use of the “perfect circle” technique as the primary data analysis measurement of glenoid bone loss (GBL) when drawing conclusions on its impact on the direction of shoulder instability. I hope these perspectives can provide insight into areas that may require further research and improvement.

This study aimed to compare GBL in active-duty servicemembers with unidirectional (anterior or posterior) shoulder instability versus combined-type instability, hypothesizing that total GBL and GBL in the direction of instability would be greater in those with combined-type instability compared with unidirectional instability. ³ The researchers used the perfect circle technique to measure GBL on 3.0-T magnetic resonance arthrograms and used a standard protocol in the coronal oblique plane. Contrary to their hypothesis, the study found that GBL in the direction of instability was significantly greater in the unidirectional instability groups compared with the combined-type instability group. ³

As has been noted by the authors, the use of the perfect circle technique for measuring GBL has poor interrater reliability.^2,3 The subjectivity inherent in the selection of magnetic resonance imaging (MRI) slices by different surgeons using the perfect circle technique for measuring GBL leads to considerable variability in measurements, making it a potentially unreliable method for its use in this study.

Although the authors acknowledge these shortcomings, they do not propose specific alternatives or recommendations to enhance the accuracy of GBL measurement that could improve the reliability and future reproducibility of their findings. One such alternative that has been suggested is the use of a “two-thirds glenoid height” technique that offers a more objective and reproducible method for estimating GBL. ¹ Makovicka et al ¹ pointed out that the subjectivity in the current perfect circle technique comes from the manual placement of a circle on the lower glenoid, which can differ depending on the chosen cross section of the oblique plane. ¹ This subjectivity is reflected in its moderate interrater, 0.84 (95% CI, 0.77-0.89; P < .001), and intrarater, 0.73 (95% CI, 0.63-0.81; P < .001), scores when determining glenoid circle diameter. ¹ In contrast, the two-thirds glenoid height technique employs a direct measurement of two-thirds of the glenoid height (from the supraglenoid tubercle to the most inferior aspect) to define the circle's diameter. This method reduces subjectivity and has demonstrated improved reliability, with interrater reliability of 0.85 and intrarater reliability of 0.88. ¹ Thanks to its significantly greater intrarater reliability compared with the perfect circle method (P < .001), ¹ the two-thirds glenoid height technique provides a more objective and reproducible approach for estimating GBL, regardless of the orientation of the MRI cuts that are utilized.

Given these advantages, I believe that incorporating the two-thirds glenoid height technique could enhance the objectivity and reproducibility of GBL assessments in future studies. Thank you for considering this suggestion, as I truly look forward to reading about insights into these factors.

Christopher Femino, MPHStratford, NJ, USAToms River, NJ, USA