I would like to congratulate Fraser et al. on a well-conducted study from which we are able to draw a narrow but strong conclusion: for the displaced anatomic neck fracture with involvement of the lesser and greater tuberosities, a reverse total shoulder arthroplasty (TSA) yields improved outcome at a 2-year follow-up across a population of patients. It is important to acknowledge that results are superior at 2 years and superior in this injury pattern, not generally superior postoperatively or for other broad definitions of proximal humeral fracture. There was a mean difference in the Constant score of 13.4 points overall and 18.7 points for C2 fractures for this modest timeframe and in these fracture patterns. Both exceeded the minimal clinically important difference of 10 points, according to the article by Fraser et al.
This article is well organized and written, easy to understand, and comprehensive. The strengths of the investigation include a well-powered cohort for the primary outcome variable, the Constant score, although it was underpowered for secondary outcomes and subgroup analysis. Additionally, the inclusion and exclusion criteria were well established, and I like that Fraser et al. stuck to the controversial treatment groups of the OTA B2 and C2 fractures and, at that, only those with severely angulated and displaced parameters of >45° valgus or >30° varus on an anteroposterior radiographic view or >45° angulation on a scapular Y radiographic view. Although there was a high percentage of excluded patients (54% [146 of 270 patients] with OTA/AO B2 and C2 fractures), most exclusions were due to the patients not meeting displacement criteria, ultimately leaving 60 patients in the open reduction and internal fixation (ORIF) group and 64 patients in the reverse TSA group.
It was outstanding that the physiotherapists performing measurements were blinded to the procedure during the measurement of outcome parameters and were different from those physiotherapists rendering treatment. Further, in this study, the same surgeons performed the arthroplasties and ORIF, which has the advantage of consistency of technique and protocol in each treatment arm. However, the possible downside is that this circumstance may also introduce bias for 1 treatment arm given that the operating surgeons were likely shoulder surgeons by training.
We should not conclude that, for every patient and every surgeon, improved results can be expected with one modality over another. The randomized controlled trial (RCT) helps to give us a guideline for a population of patients treated by a population of surgeons; therefore, it should be obvious that within those populations there exist patients who will outperform the mean in both treatment groups, and the same is true for the surgeon. This RCT does not tell us what the optimal outcome is at the onset of treatment for any individual patient, nor even whether optimal reverse TSA is better than optimal ORIF; therefore, we must be circumspect in counseling our patients, taking into consideration variables that could make them better candidates for a particular technique or that may make the surgeon more facile with another technique.
What would not be appropriate is to conclude that reverse TSA is better than ORIF for proximal humeral fractures. We must beware; it is only a 2-year study. If there was a 25% revision rate for reverse TSA at 5 to 10 years, then the superior results would be reversed, and we would be reinventing another wheel in orthopaedics again. I commend the authors for an excellent study, and I am pleased to learn that this study group is in the midst of mining longer-term results that would elevate our understanding further regarding this very important fracture for which excellent treatment results have been hard to find.