Prospective, Blinded, Randomized Controlled Trial of Stemless Versus Stemmed Humeral Components in Anatomic Total Shoulder Arthroplasty: Results at Short-Term Follow-up.

BACKGROUND

Stemless humeral components for anatomic total shoulder arthroplasty (aTSA) have several reported potential benefits compared with stemmed implants. However, we are aware of no Level-I, randomized controlled trials (RCTs) that have compared stemless implants with stemmed implants in patients managed with aTSA. We sought to directly compare the short-term clinical and radiographic outcomes of stemless and stemmed implants to determine if the stemless implant is noninferior to the stemmed implant.

METHODS

We performed a prospective, multicenter, single-blinded RCT comparing stemless and short-stemmed implants in patients managed with aTSA. Range-of-motion measurements and American Shoulder and Elbow Surgeons (ASES), Single Assessment Numeric Evaluation (SANE), and Constant scores were obtained at multiple time points. Device-related complications were recorded. Radiographic evaluation for evidence of loosening, fractures, dislocation, or other component complications was performed. Statistical analysis for noninferiority was performed at 2 years of follow-up for 3 primary end points: ASES score, absence of device-related complications, and radiographic signs of loosening. All other data were compared between cohorts at all time points as secondary measures.

RESULTS

Two hundred and sixty-five shoulders (including 176 shoulders in male patients and 89 shoulders in female patients) were randomized and received the allocated treatment. The mean age of the patients (and standard deviation) was 62.6 ± 9.3 years, and 99% of the shoulders had a primary diagnosis of osteoarthritis. At 2 years, the mean ASES score was 92.5 ± 14.9 for the stemless cohort and 92.2 ± 13.5 for the stemmed cohort (p value for noninferiority test, <0.0001), the proportion of shoulders without device-related complications was 92% (107 of 116) for the stemless cohort and 93% (114 of 123) for the stemmed cohort (p value for noninferiority test, 0.0063), and no shoulder in either cohort had radiographic signs of loosening. Range-of-motion measurements and ASES, SANE, and Constant scores did not differ significantly between cohorts at any time point within the 2-year follow-up.

CONCLUSIONS

At 2 years of follow-up, the safety and effectiveness of the stemless humeral implant were noninferior to those of the stemmed humeral implant in patients managed with aTSA for the treatment of osteoarthritis. These short-term results are promising given the potential benefits of stemless designs over traditional, stemmed humeral components.

LEVEL OF EVIDENCE

Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.