Glenoid component loosening is a common cause of failure for anatomic total shoulder arthroplasty (TSA). These authors sought to determine the clinical significance of progression of radiolucency around the central peg (central peg osteolysis, CPO) of the glenoid component as determined on standardized AP and axillary x-rays.
Initial radiographic and clinical follow-up was between 1 and 3 years after surgery. All patients had an intact rotator cuff, and had either a non-augmented central pegged glenoid or augmented central pegged all-polyethylene glenoid component ((APG, Depuy-Synthes (60%); Steptech APG (23%), Depuy-Synthes; Affiniti, Wright-Tornier (16%)). From this group, 104 patients that had a Penn Shoulder Score (PSS) less than 70 points at initial follow-up were eliminated because the authors wanted to evaluate the relationship of CPO on patients with high shoulder scores at first follow up, essentially asymptomatic patients. Another 199 patients were eliminated as they did not have a minimum interval of 2 years between the initial clinical and radiographic follow-up and their final follow-up. Patients that had a revision surgery in this 2-year interval time period were included for analysis. This left a final cohort of 73 shoulders.
For these patients an average patient age at surgery was 65±7 and 63% were male. Median initial follow
up was 14 months [IQR 12-25] and final median follow-up was 56 months [IQR 47-69]. There were 4 revision surgeries and 17 had a "PSS failure", a drop in the Penn Shoulder Score (PSS) > 11.4 (the minimal clinically important difference, MCID).
True AP (Grashey view) radiographs demonstrating central peg grading (CPG) scale for (A) Grade 1 osteolysis (Central Peg Osteolysis), (B) Grade 2 bone growth to the edge of the flanges, and (C) Grade 3 bone growth within the flanges.
Example of CPG progression of (D) Grade 3 to (D`) Grade 1.
Example of CPG progression from (E) Grade 1 to (E`) worsening of Grade 1. Notice the advance of early sclerotic edges to radiolucency around the central peg and peripheral pegs (E`).
Example of early radiograph with (F) Grade 3 and no change of CPG with stable (F`) Grade 3 CPG.
Example of stable early (G) Grade 1 CPO. Notice (G`) shows no progression of CPO from (G) and stable peripheral peg appearance.
They found that central peg osteolysis (CPO) at final follow-up, central peg grade (CPG) progression, and worse PSS at follow-up were associated with revision surgery (p<0.05).
They found younger age at surgery, CPO at final follow-up, CPG progression, and greater glenoid component retroversion at final follow-up to be associated with clinical failure (PSS failure or revision surgery) (p<0.05).
They also found differences among glenoid implant types with a higher CPO rate for Steptech at early follow-up (p=0.022), but no difference at the late follow-up (p=0.30), however, there was a significant difference in radiographic follow-up times in the group (p=0.0005). APG implants had 67.4 ± 21.2 months and Affiniti implants had 64.8 ± 23.3 months follow-up, whereas Steptech only had 44.8 ± 9.5 months follow-up.
Multivariate analysis found only CPG progression to be associated with clinical failure (p<0.001).
Comment: This is an important study that demonstrates the relationship between progressive bone loss around the central peg and clinical failure (revision or a drop in clinical score > MCID). It is sobering indeed to note that in the hands of highly experienced surgeons the revision rate in this cohort was 5% (4/73) and that the revision or clinical failure rate was 25% (18/73). CPG progression was noted in 37% (27/73) of the shoulders.
So now the challenge is to identify the factors associated with CPG and clinical failure. The authors looked for these factors in this study, but they did not find significant association of failure with age at surgery, sex, superior migration, posterior displacement of the humeral head in relation to the plane of the scapula, posterior decentering of the humeral head on the glenoid, or glenoid version.
