Risk of Perforation Is High During Corrective Reaming of Retroverted Glenoids: A Computer Simulation Study

Corrective anterior reaming is an accepted method for addressing retroversion in a biconcave retroverted (Walch classification, type B2) glenoid in anatomic total shoulder arthroplasty. However, concern still exists regarding early glenoid component failure in the setting of severe retroversion, which may be related to loss of component containment and/or violation of subchondral bone resulting from reaming. The goal of this study was to determine what characteristics of B2 glenoids are less amenable to corrective reaming by virtually implanting anatomic glenoid components.

(1) How much medial reaming is required to correct the version of a B2 glenoid to an acceptable position? (2) Are glenoids with more severe retroversion (> 25°) at higher risk of component perforation than less retroverted glenoids? (3) Is correcting to 10° of retroversion associated with greater risk as compared with reaming to 15°? (4) How does corrective reaming affect the underlying bone density on the glenoid face of B2 glenoids?

A series of 71 patients with B2 glenoids (posterior subluxation of the humeral head with posterior bone loss) with CT scans who were indicated for shoulder arthroplasty were reviewed. Forty-four of 71 glenoids (62.5%) had < 25° of native retroversion. Anatomic glenoid implants were then virtually implanted using three-dimensional CT software that allows for preoperative shoulder arthroplasty planning to correct native retroversion to 15° or 10° of retroversion using both a central peg with an inverted triangle peg configuration or a keel. The amount of reaming of the anterior glenoid required to correct retroversion, perforation of peripheral pegs, or keel was compared. Additionally, assessment of the surface area of the glenoid that had poor bone density (defined as cancellous bone under the subchondral plate) was analyzed by the software after correction.

Correction to 15° of retroversion required 5 ± 3 mm of reaming, and correction to 10° of retroversion required 8 ± 3 mm of reaming to obtain at least 80% seating. Peripheral peg perforation with correction to 15° occurred in 15 of 27 (56%) glenoids with > 25° of retroversion compared with 10 of 44 (23%) of glenoids with < 25° of retroversion (relative risk [RR], 2.4; 95% confidence interval [CI], 1.3-4.6; p = 0.006). There was no difference in perforation with keeled components. Increased correction to 10° did not increase the risk of component perforation. When correction to 15°, glenoids with higher native version (> 25°) had a greater risk of poor bone quality support (10 of 27 [37%]) when compared with glenoids with less version (four of 44 [9%]; RR, 4.1; 95% CI, 1.5-12.8; p = 0.006). Increased correction resulted in 13 of 27 (48%) glenoids with version > 25° having poor bone density versus 10 of 44 (23%) with ≤ 25° of version (RR, 2.1; 95% CI, 1.1-4.1; p = 0.028).

There is a high risk of vault perforation after corrective reaming. Glenoid retroversions > 25° are at a higher risk of having poor bone quality supporting the component.

When contemplating options for patients with severe retroversion, surgeons should consider alternatives other than corrective reaming if achieving normal glenoid version is desired.