Abstract
Background
Understanding outcome expectations is a key component of pre-operative counseling before shoulder arthroplasty. However, the capacity of surgery to restore the shoulder back to normal and how long this may take remains unclear. The purpose of this study is to evaluate shoulder arthroplasty patients who have rated their shoulder as normal, characterize the overall time taken to achieve this level, the time intervals where improvement occurs, and define the plateau in improvement.
Methods
A retrospective query of our institution's shoulder and elbow surgery repository identified patients treated with primary anatomic total shoulder arthroplasty (aTSA) or reverse total shoulder arthroplasty (rTSA) between November 2006 and April 2024. Patients were included if they had a post-operative Single Assessment Numeric Evaluation (SANE) score available to review. SANE scores and additional patient-reported outcome measure data were evaluated at all routine post-operative follow-up intervals: 3 months, 6 months, and yearly from 1 year to 10 years. Those patients who reported a SANE score of ≥95 were designated as having returned to normal level of function. The percentage of patients who returned to normal was compared between those treated with aTSA and rTSA. A similar subanalysis comparing aTSA and rTSA patients treated for glenohumeral osteoarthritis (OA) with an intact rotator cuff was also conducted.
Results
A total of 1,399 aTSA and 1,505 rTSA patients met inclusion criteria, with 714 aTSA (51.0%) and 562 rTSA (37.3%) achieving a post-operative SANE score ≥95. aTSA patients returned to normal at a higher rate (P < .001) but did not return to normal faster than rTSA patients (1 year vs. 1 year; P = .607). However, in the OA subanalysis, rTSA patients returned to normal faster than aTSA patients (0.5 year vs. 1 year; P = .020), despite aTSA patients (52.3%) having a higher rate of returning to normal than rTSA patients (42.8%; P < .001). Of those who returned to normal, at least 70% of patients did so within 1 year. Similar trends were observed in the OA cohort. There was a consistent plateau in improvement at three years across all cohorts.
Conclusion
Patients treated with aTSA have a greater chance of rating their shoulder as normal when compared to rTSA, and typically both achieve this level by 1 year. This is true when comparing all indications as well as OA with an intact rotator cuff. However, rTSA patients treated for OA with an intact rotator cuff have a faster return to normal than aTSA patients.
Keywords: Patient reported outcome measures, Anatomic total shoulder arthroplasty, Reverse total shoulder, Post-operative adaption, Post-operative recovery, Speed of recovery
Patient satisfaction following total shoulder arthroplasty has been shown to be strongly influenced by defined expectations pre-operatively and strongly related to subjective measures of symptoms and function post-operatively.4 Thus, an understanding of the recovery characteristics of subjective outcome measures is a key component of pre-operative counseling. Levy et al10 reported on the speed of recovery of pain, function, and motion between anatomic total shoulder arthroplasty (aTSA) and reverse total shoulder arthroplasty (rTSA) total shoulder arthroplasty up to 2 years post-operatively. aTSA patients reached plateaus in recovery faster and more predictably than rTSA patients. While the majority of improvement in pain and function occurred by 6 months and in motion by 1 year, recovery in rTSA patients was more variable and sustained over the study's 2-year period.
More recently, Mahendraraj et al11 similarly observed a greater degree of variability in American Shoulder and Elbow Surgeons (ASES) score improvement at 1- and 2-year follow-up when comparing rTSA to aTSA. Grubhofer et al7 demonstrated aTSA patients achieve 90% of improvement in pain by 6 weeks, 96% of ASES score improvement by 6 months, and 92% of Single Assessment Numeric Evaluation (SANE) score improvement by 6 months compared to 82%, 85%, and 89%, respectively, in rTSA patients.
In addition, return to work, return to sport, and even the recovery from sleep disturbance after shoulder arthroplasty have been characterized.8,9,19 However, these studies have largely focused on when the majority of overall improvement occurs irrespective of a benchmark end goal. Thus, none have specifically investigated one of the most frequent questions patients ask when setting expectations, which revolves around the possibility their shoulder will return to a level they perceive to be normal and, subsequently, how much time that will take. The purpose of this study is to evaluate patients that return to normal following total shoulder arthroplasty including overall time taken to return to normal, time intervals at which the most improvement occurs, and presence of any improvement plateaus. We hypothesize patients undergoing aTSA return to normal sooner than those undergoing rTSA, but may display similar trends in achieving to their new normal in longer term follow-up.
Methods
A retrospective query of our institution's shoulder and elbow surgery repository identified patients treated with primary aTSA or rTSA of a single manufacturer (Enovis, Austin, TX, USA) by a single fellowship trained shoulder and elbow surgeon between November 2006 and April 2024. Patients were included if they had a post-operative SANE score available to review at a minimum 3-month follow-up. Patients undergoing revision arthroplasty or who did not have a SANE score available to review were excluded.
SANE scores were collected at all routine post-operative follow-up intervals: 3 months, 6 months, and yearly from 1 year to 10 years, as well as at most recent follow-up if beyond 10 years. The time point for which patients returned to normal was defined as the point in routine post-operative follow-up where they first rated their operative shoulder as a SANE score of 95 or greater. This definition of a ‘New Normal’ following shoulder arthroplasty has been previously established by Beleckas et al.3
Patient demographics (including age, gender, body mass index, and hand dominance), comorbidities, and surgical indications were also collected. Patient-reported outcome measures (PROMs; including ASES score, Simple Shoulder Test, visual analog scale [VAS] function, and VAS pain), goniometer-measured shoulder range of motion, and provider-assessed strength were collected and analyzed to reveal trends in the differences between the cohorts of those reaching a SANE ≥95 and those who failed to reach that threshold, as well as between aTSA and rTSA cohorts reaching a new normal. The percentages of those patients who returned to normal by each follow-up time point were plotted and compared between those who underwent aTSA or rTSA.
Using previously described methodology, interval changes in the SANE score occurring across follow-up intervals (ie, 3 to 6 months vs. 6 months to 1 year) were compared to evaluate for a plateau in improvement after surgery.10 Any plateaus in maximal improvement were defined as the follow-up point at which no statistically significant improvement was observed in relation to adjacent time intervals.
A similar subanalysis of post-operative SANE scores comparing aTSA and rTSA patients who were treated for glenohumeral osteoarthritis (OA) with intact rotator cuff was also conducted.
Statistical analysis
Data were checked for normality, and appropriate descriptive statistics including frequencies with percentages, medians with interquartile ranges, and means with standard deviations were calculated for all variables. Mann-Whitney U, Chi-square, and Fisher's exact tests were performed to compare patient characteristics by type of surgery and SANE score. Cochran's Q with post hoc pairwise comparisons was used to evaluate changes in SANE score at each post-operative time point for aTSA and rTSA patients. Data were analyzed using statistical package for the social sciences (version 29; IBM Corp., Armonk, NY, USA). All tests were two-tailed and a priori α level set at a P value less than or equal to .05 was used to determine statistical significance.
Results
A total of 1,399 aTSA and 1,505 rTSA patients met inclusion criteria, and of those 714 aTSA (51.0%) and 562 rTSA (37.3%) achieved a post-operative SANE score ≥95 (Fig. 1). Of those who were excluded with a maximal post-operative SANE score of less than 95, there was a higher proportion of patients receiving rTSA compared to the cohort reaching a new normal (57.9% vs. 44.0% P < .001). Moreover, the cohort who failed to reach a new normal post-operatively had a marginally lower body mass index (27.7 vs. 28.2 P = .046), greater percentage of cigarette smokers (11.4% vs. 8.3% P = .024), as well as lower baseline PROM scores for SANE, Simple Shoulder Test, VAS function and lower starting range of motion and strength (Supplementary Table SI).
Figure 1.
A Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) flow diagram displaying number of and reason why patients were excluded. aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty; SANE, Single Assessment Numeric Evaluation.
Comparing aTSA and rTSA patients with a post-operative SANE score of 95 or greater at most recent follow-up, rTSA patients were slightly older (74 years vs. 70 years; P < .001) with a greater percentage of females in the rTSA cohort (59.1% vs. 45.2%; P < .001) and greater percentage of surgery on the dominant arm (63.2% vs. 56.0%; P = .010). Additionally, the surgical indications differed significantly, with aTSA patients being treated for OA (94%) and the majority of rTSA patients being treated for cuff tear arthropathy or massive cuff tear (51.8%; P < .001) (Supplementary Table SII).
Assessing the return to normalcy for all indications, a higher percentage of aTSA patients returned to normal (51.0% vs. 37.3%; P < .001); however, both groups took 1 year to return to normal (1 year vs. 1 year; P = .607) (Table I). Across all time points of post-operative follow-up, a greater percentage of aTSA patients reported normal (P < .022) (Fig. 2; Table II). The plateaus for reaching normal are observed graphically at 3 years for both aTSA and rTSA cohorts (Fig. 2) as well as demonstrated statistically in Table III.
Table I.
Return to normal∗.
| Outcome | n | aTSA |
n | rTSA |
P value |
|---|---|---|---|---|---|
| Median (IQR) or N (%) | Median (IQR) or N (%) | ||||
| All patients | |||||
| Achieve SANE ≥95 at any time | 1,399 | 714 (51.0) | 1,505 | 562 (37.3) | <.001 |
| Time to SANE ≥95 (yr) | 714 | 1 (0.5-2) | 562 | 1 (0.5-2) | .607 |
| Patients with osteoarthritis | |||||
| Achieve SANE ≥95 at any time | 1,282 | 671 (52.3) | 491 | 210 (42.8) | <.001 |
| Time to SANE ≥95 (yr) | 671 | 1 (0.5-2) | 210 | 0.5 (0.5-1) | .020 |
aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty; IQR, interquartile range.
Bold values indicate statistical significance (P < .05).
Defined as a Single Assessment Numeric Evaluation (SANE) score ≥95.
Figure 2.
Of all primary total shoulder arthroplasties, cumulative percentage of patients who reached a Single Assessment Numeric Evaluation (SANE) score ≥95 over time. Both overall cohorts (any surgical indication) as well as cohorts indicated for only cuff-intact glenohumeral osteoarthritis are plotted in the figure. aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Table II.
Cumulative percentage of overall aTSA and rTSA patients who achieved SANE ≥95 (N = 2,904).
| Time point | aTSA (n = 1,399) N (%) |
rTSA (n = 1,505) N (%) |
P value |
|---|---|---|---|
| 3 mo | 114 (8.1) | 90 (6.0) | .022 |
| 6 mo | 331 (23.7) | 265 (17.6) | <.001 |
| 1 yr | 510 (36.5) | 397 (26.4) | <.001 |
| 2 yr | 585 (41.8) | 481 (32.0) | <.001 |
| 3 yr | 639 (45.7) | 525 (34.9) | <.001 |
| 4 yr | 663 (47.4) | 539 (35.8) | <.001 |
| 5 yr | 680 (48.6) | 549 (36.5) | <.001 |
| 6 yr | 693 (49.5) | 555 (36.9) | <.001 |
| 7 yr | 703 (50.3) | 559 (37.1) | <.001 |
| 8 yr | 710 (50.8) | 561 (37.3) | <.001 |
| 9 yr | 714 (51.0) | 562 (37.3) | <.001 |
| 10 yr | 714 (51.0) | 562 (37.3) | <.001 |
SANE, Single Assessment Numeric Evaluation; aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Bold values indicate statistical significance (P < .05).
Table III.
Cumulative return to normal (SANE ≥95) among overall aTSA and rTSA patients by time interval.
| Variable | P value | Plateau (mo/yr) |
|---|---|---|
| aTSA (N = 1,399) | 3 yr | |
| SANE ≥95 | ||
| 3 mo post-op to 6 mo post-op | <.001 | |
| 6 mo post-op to 1 yr post-op | <.001 | |
| 1 yr post-op to 2 yr post-op | <.001 | |
| 2 yr post-op to 3 yr post-op | <.001 | |
| 3 yr post-op to 4 yr post-op | .078 | |
| 4 yr post-op to 5 yr post-op | .211 | |
| 5 yr post-op to 6 yr post-op | .339 | |
| 6 yr post-op to 7 yr post-op | .462 | |
| 7 yr post-op to 8 yr post-op | .607 | |
| 8 yr post-op to 9 yr post-op | .769 | |
| 9 yr post-op to 10 yr post-op | 1.00 | |
| rTSA (N = 1,505) | ||
| SANE ≥95 | 3 yr | |
| 3 mo post-op to 6 mo post-op | <.001 | |
| 6 mo post-op to 1 yr post-op | <.001 | |
| 1 yr post-op to 2 yr post-op | <.001 | |
| 2 yr post-op to 3 yr post-op | <.001 | |
| 3 yr post-op to 4 yr post-op | .243 | |
| 4 yr post-op to 5 yr post-op | .405 | |
| 5 yr post-op to 6 yr post-op | .617 | |
| 6 yr post-op to 7 yr post-op | .739 | |
| 7 yr post-op to 8 yr post-op | .868 | |
| 8 yr post-op to 9 yr post-op | .934 | |
| 9 yr post-op to 10 yr post-op | 1.00 |
SANE, Single Assessment Numeric Evaluation; aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty; post-op, post-operative. Bold values indicate statistical significance (P < .05).
The greatest improvement in the percentage of patients achieving a SANE score of 95 or greater occurred, in both cohorts, between 3 to 6 months (+15.5% aTSA/+11.6% rTSA), followed by 6 to 12 months (+12.8% aTSA/+8.8% rTSA). Beyond that time interval, the percentage of patients reaching normal dropped with each consecutive year (Fig. 3). Both cohorts demonstrated similar improvements across all follow-up intervals through 10 years post-operatively with 1 exception — the 1 year to 2 year follow-up interval demonstrated a higher percentage of rTSA patients returning to normal (14.9% vs. 10.5%; P = .017) (Fig. 3). Of note, among patients who achieved normal, 71.5% of aTSA and 70.6% of rTSA patients achieved normal within the first year (Fig. 3).
Figure 3.
In patients within the overall cohort who return to normal, percentage of patients who reached a Single Assessment Numeric Evaluation (SANE) score ≥95 by follow-up time point. The red text and box highlights the percentage of patients returning to normal within 1 year following the date of surgery. The red asterisk indicates a statistically significant difference between rTSA and aTSA cohorts at that time point. aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Among patients treated for OA with an intact rotator cuff, aTSA patients returned to normal at a significantly higher rate (52.3% vs. 42.8%, P < .001); however, rTSA patients returned to normal faster than aTSA patients (0.5 year vs. 1 year; P = .020) (Table I). Furthermore, there was no significant difference in the rate of return to normal appreciated in the first 2 years post-operatively (Fig. 2; Table IV). Similar to the overall cohorts, both rTSA OA and aTSA OA cohorts demonstrate statistically significant improvement plateaus at 3 years post-operatively (Table V). The OA cohorts also demonstrated similar improvements across all follow-up intervals up to 10 years post-operatively between aTSA and rTSA (Fig. 4). The same follow-up intervals also yielded the greatest improvement in the percentage of patients achieving a SANE score of 95 or greater, with first being between 3 to 6 months (+15.9% aTSA/+15.1% rTSA), followed by the interval of 6 to 12 months (+13.0% aTSA/+11.0% rTSA). Also similar to overall cohorts, over 70% of both aTSA for OA (71.1%) and rTSA (77.1%) reached post-operative normalcy within 1 year (Fig. 4).
Table IV.
Cumulative percentage of OA patients who achieved SANE ≥95 (N = 1773).
| Time point | aTSA (n = 1,282) N (%) |
rTSA (n = 491) N (%) |
P value |
|---|---|---|---|
| 3 mo | 106 (8.3) | 39 (7.9) | .344 |
| 6 mo | 310 (24.2) | 113 (23.0) | .606 |
| 1 yr | 477 (37.2) | 162 (33.0) | .098 |
| 2 yr | 546 (42.6) | 187 (38.1) | .085 |
| 3 yr | 599 (46.7) | 201 (40.9) | .028 |
| 4 yr | 623 (48.6) | 206 (42.0) | .012 |
| 5 yr | 640 (49.9) | 209 (42.6) | .006 |
| 6 yr | 650 (50.7) | 210 (42.8) | .003 |
| 7 yr | 660 (51.5) | 210 (42.8) | .001 |
| 8 yr | 667 (52.0) | 210 (42.8) | .001 |
| 9 yr | 671 (52.3) | 210 (42.8) | <.001 |
| 10 yr | 671 (52.3) | 210 (42.8) | <.001 |
OA, Osteoarthritis; SANE, Single Assessment Numeric Evaluation; aTSA, Anatomic Total Shoulder Arthroplasty; rTSA, Reverse Total Shoulder Arthroplasty.
Bold values indicate statistical significance (P < .05).
Table V.
Cumulative return to normal (SANE ≥95) among aTSA and rTSA patients with OA by time interval.
| Variable | P value | Plateau (mo/yr) |
|---|---|---|
| aTSA (N = 1,282) | ||
| SANE ≥95 | 3 yr | |
| 3 mo post-op to 6 mo pos-op | <.001 | |
| 6 mo post-op to 1 yr post-op | <.001 | |
| 1 yr post-op to 2 yr post-op | <.001 | |
| 2 yr post-op to 3 yr post-op | <.001 | |
| 3 yr post-op to 4 yr post-op | .069 | |
| 4 yr post-op to 5 yr post-op | .198 | |
| 5 yr post-op to 6 yr post-op | .449 | |
| 6 yr post-op to 7 yr post-op | .449 | |
| 7 yr post-op to 8 yr post-op | .596 | |
| 8 yr post-op to 9 yr post-op | .762 | |
| 9 yr post-op to 10 yr post-op | 1.00 | |
| rTSA (N = 491) | ||
| SANE ≥95 | 3 yr | |
| 3 mo post-op to 6 mo pos-op | <.001 | |
| 6 mo post-op to 1 yr post-op | <.001 | |
| 1 yr post-op to 2 yr post-op | <.001 | |
| 2 yr post-op to 3 yr post-op | .046 | |
| 3 yr post-op to 4 yr post-op | .475 | |
| 4 yr post-op to 5 yr post-op | .668 | |
| 5 yr post-op to 6 yr post-op | .886 | |
| 6 yr post-op to 7 yr post-op | 1.00 | |
| 7 yr post-op to 8 yr post-op | 1.00 | |
| 8 yr post-op to 9 yr post-op | 1.00 | |
| 9 yr post-op to 10 yr post-op | 1.00 |
OA, Osteoarthritis; SANE, Single Assessment Numeric Evaluation; aTSA, Anatomic Total Shoulder Arthroplasty; rTSA, Reverse Total Shoulder Arthroplasty.
Bold values indicate statistical significance (P < .05).
Figure 4.
In patients with osteoarthritis who returned to normal, percentage of patients who reached a Single Assessment Numeric Evaluation (SANE) score ≥95 by follow-up time point.The red text and box highlights the percentage of patients returning to normal within 1 year following the date of surgery. aTSA, anatomic total shoulder arthroplasty; rTSA, reverse total shoulder arthroplasty.
Discussion
This study evaluated trends in the speed to return to normal in patients undergoing primary aTSA or rTSA. A greater percentage of aTSA patients were able to achieve normal (ie, reporting a SANE score of ≥95) than rTSA, when considering all surgical indications as well as the single indication of glenohumeral OA with an intact rotator cuff. Among those who returned to normal, over 70% of these patients did so within the first year following surgery across all investigated cohorts. Interestingly, among patients treated for OA, rTSA returned to normal at a faster pace than aTSA, reaching normal by 6 months on average. This is contrary to our hypothesis, as we had theorized that maintenance of anatomical joint structure may expedite the post-operative return of a subjective normal function. However, similar plateaus and intervals of greatest improvement in the speed of return to normal suggest a similar process of adaption to a new normal likely occurs in both rTSA and aTSA patients.
Beleckas et al3 has previously investigated patients who rated the surgical shoulder normal following rTSA and aTSA. In this series, aTSA patients outperformed rTSA patients in the ability to perform higher-demand activities, noting superior post-operative motion and greater ability to return to sport and work.3 This was observed for all indications as well as those patients treated for glenohumeral OA with an intact rotator cuff. Additionally, a higher percentage of the aTSA patients returned to normal (40%) than rTSA patients (26%) in both all indications cohort as well as in the OA cohort (41% vs. 35%),2 similar to the results of our study. Our study expands on this prior work, evaluating not only whether patient achieve a normal but also the pace at which at which this occurs.
With the expanding indications for rTSA and the widespread use of the procedure, it is becoming increasingly common for some surgeons to select rTSA in patients with OA and an intact rotator cuff. As a result, recent studies have focused on aTSA and rTSA comparisons in hopes to identify specific demographics, wear patterns, or other patient characteristics that may impact overall outcomes when treating OA patients. For instance, despite theoretical concern of a degenerative rotator cuff in older patients, it has been shown that patients over 75 and 80 years old can have very successful outcomes with aTSA.2,12 Glenoid wear pattern in OA also plays an important role when determining aTSA vs. rTSA. Ardebol et al2 demonstrated that patients with OA and an intact rotator cuff do well with either aTSA or rTSA regardless of wear pattern but noted that the rTSA cohort inherently demonstrated a much larger proportion of B2 and B3 glenoids.2 Aleisawi et al attempted to control for this, by observing OA with an intact rotator cuff in only B2 and B3 glenoids. They found that while B2 and B3 consisted of more patients undergoing rTSA, the B2 and B3 glenoids that underwent aTSA had slight differences in outcomes and complication rates based on differences in technique such eccentrically reaming, noncorrective reaming, and posterior augments.1 Our study adds to the literature on outcomes comparing aTSA to rTSA for OA, by demonstrating that aTSA patients have a higher rate of returning to normal, while patients undergoing rTSA actually achieved normal faster. Additionally, over 70% of OA patients reach normal within 1 year.
The nuance of feeling that an operative joint has returned to normal may be more complex than what improvements in patient reported outcomes or improvement in pain can define. In other total joint arthroplasty (TJA), while no study has explicitly investigated return to ‘normal’ function, recovery of functional outcomes has also been investigated. For example, von Rottkay found similar activity, functionality and quality of life metrics among total hip arthroplasty patients compared to healthy controls once patients reach 1 year post-operatively,14 which correlates well with the estimated 1 year to return to normal found in our overall cohorts of aTSA and rTSA (Table I). Moreover, among both total hip and total knee arthroplasty patients, Ekhtiari et al found that the greatest improvements in PROMs post-TJA occur in the first six months, similar to our periods of greatest improvements across cohorts being 3 to 6 months. Orange et al13 also found the greatest clinically meaningful improvements in physical function at 3 to 6 months post-operatively among total knee arthroplasty patients. And while differences in recovery are expected with different joint arthroplasty, especially comparing lower extremity TJA to upper extremity shoulder arthroplasty, similarity in temporal trends of post-operative recovery may point to similar mental accommodation of patients adjusting to life with a complete joint replacement.
In our study, the rate at which patients returned to normal was greater in aTSA than rTSA across all time points when all indications were considered and across all time points after 2 years for patients with OA. Similarly, prior studies have shown that recovery of pain, SANE, ASES, and other functional outcomes is generally faster and superior in aTSA compared to rTSA at nearly all time points within the first two years after surgery, when all indications are considered.7,10
However, in our overall cohorts, the time to return to normal was comparable between procedures (Table I), with 71.5% of aTSA and 70.6% of rTSA patients reaching normal within a year (Fig. 3). This suggests greater similarity than difference in recovery timelines among those who achieved this outcome. The discrepancy from prior literature may reflect the multifactorial nature of “return to normal,” making direct comparison with other functional outcomes less appropriate. This, along with the absence of any single explanatory variable, may account for the similar trends in return-to-normal recovery observed in both cohorts.
There is likely a similar process of adaption to a new post-operative normal occurring for all shoulder arthroplasty patients, which helps to explain the similar recovery timeline. The power of this common adaptive psychology should not be underestimated, as mental health including anxiety and depression, as well as other psychosocial factors such as pre-operative opioid use and lower patient confidence in expected improvement have already been associated with inferior post-operative outcomes in shoulder surgery.5,6,18,20 Resilience and the ability to recover from stress, has also been investigated and similarly found to have improved pain relief following shoulder surgery, although its isolated influence on SANE scores were less reliable.17 And while no mental health parameters were assessed in the scope of this study, the observed similarity in temporal outcomes among aTSA and rTSA patients in this study may suggest the psychological experience for those undergoing either procedure and having an excellent outcome is similar.
Although this study provides important guidance in setting expectations of the recovery timeline pre-operatively and demonstrates potentially nuanced time tables and trends of improvement for return to subjective normal shoulder function after reverse and aTSA, it is not without its limitations. First, there is likely an inherent selection bias in selection of aTSA or rTSA among patients treated for OA. Additionally, matching the cohorts was intentionally avoided, as the primary aim was not to compare the end-point superiority of rTSA vs. aTSA, but to describe recovery patterns in the average patient achieving an excellent outcome (SANE ≥95). Matching on variables such as age, sex, or comorbidities would have created highly selected, homogeneous groups, which may not reflect the broader clinical population. By retaining the natural variation present in each cohort, our findings are more representative of real-world patient populations and thus more applicable for setting patient expectations in routine practice. Moreover, the collected PROM data were prospectively collected from either an in-person office visit or a scheduled email survey at the time of routine follow-up but retrospectively reviewed in this study. There may naturally be discrepancies between PROM surveys completed in the office vs. at home, and as more patients reaching longer term follow-up maintained PROM follow-up via online survey rather than in-person clinic visits since they were doing so well, there may be a bias in the longer outcome data. Moreover, other measures of assessing normal function were not formally assessed, including simply asking the dichotomous yes or no survey item of ‘does your shoulder feel normal?’ which may limit the interpretation of these results. Additionally, alternative measures of what may be suggested as “normal” after total shoulder arthroplasty were also not considered here such as rate of return to sport and rate of return to work. However, our high threshold of SANE ≥95 reasonably captures most people who would rate their normal as normal by a different, straightforward survey method. Studies have explored patients' perceived benefit after arthroplasty including measured Substantial Clinical Benefit and Patient Acceptable Symptomatic State.16,15 While these correlate with patient improvement after shoulder arthroplasty, they do not directly measure the feeling of returning to normal after shoulder arthroplasty. In reality, the end point of normality is a complex, subjective outcome for patients, which warrants additional investigation in order characterize these excellent outcomes following shoulder arthroplasty and optimize outcomes for all patients undergoing aTSA and rTSA.
Conclusion
Patients treated with aTSA have a greater chance of rating their shoulder as normal when compared to rTSA, and typically both achieve this level by 1 year. This is true when comparing all indications as well as OA with an intact rotator cuff. However, rTSA patients treated for OA with an intact rotator cuff have a faster return to normal than aTSA patients.
Disclaimers
Funding: No funding was disclosed by the authors.
Conflicts of interest: Jonathan C. Levy, MD is a paid consultant for Enovis, Stryker, Exactech, and Globus Medical. He receives royalties from Enovis, Stryker, Exactech, Globus Medical, and Innomed. He receives research support from Enovis and Stryker. All the other authors, their immediate families, and any research foundation which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
Footnotes
This study received Category 4 Exemption status per 45 CFR 46.116(e)(1) on 05/21/2024 (Study #2024-049).
All work was performed at the Levy Shoulder to Hand Center at the Paley Orthopedic and Spine Institute (9960 N Central Park Blvd. Suite 150A, Boca Raton, FL 33428, USA).
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jseint.2026.101628.
Supplementary data
References
- 1.Aleisawi H., Kruse C., Nucci N., Alturki N., Abdel Khalik H., Athwal G.S., et al. Outcomes of anatomic versus reverse shoulder arthroplasty for B2 & B3 glenoids with an intact rotator cuff: an updated systematic review and proportional meta-analysis. Shoulder Elbow. 2025 doi: 10.1177/17585732251359590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ardebol J., Flores A., Kiliç A., Pak T., Menendez M.E., Denard P.J. Patients 75 years or older with primary glenohumeral arthritis and an intact rotator cuff show similar clinical improvement after reverse or anatomic total shoulder arthroplasty. J Shoulder Elbow Surg. 2024;33:1254–1260. doi: 10.1016/j.jse.2023.10.021. [DOI] [PubMed] [Google Scholar]
- 3.Beleckas C.M., Schodlbauer D.F., Mousad A.D., Levy J.C. Evaluation of new normal after shoulder arthroplasty: comparison of anatomic vs. reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2025;34(6S):S43–S49. doi: 10.1016/j.jse.2025.02.010. [DOI] [PubMed] [Google Scholar]
- 4.Chen A.L., Bain E.B., Horan M.P., Hawkins R.J. Determinants of patient satisfaction with outcome after shoulder arthroplasty. J Shoulder Elbow Surg. 2007;16:25–30. doi: 10.1016/j.jse.2006.04.013. [DOI] [PubMed] [Google Scholar]
- 5.Colasanti C.A., Lin C.C., Anil U., Simovitch R.W., Virk M.S., Zuckerman J.D. Impact of mental health on outcomes after total shoulder arthroplasty. J Shoulder Elbow Surg. 2023;32:980–990. doi: 10.1016/j.jse.2022.10.028. [DOI] [PubMed] [Google Scholar]
- 6.Gil J.A., Goodman A.D., Mulcahey M.K. Psychological factors affecting outcomes after elective shoulder surgery. J Am Acad Orthop Surg. 2018;26:e98–e104. doi: 10.5435/jaaos-d-16-00827. [DOI] [PubMed] [Google Scholar]
- 7.Grubhofer F., Muniz Martinez A.R., Ernstbrunner L., Haberli J., Selig M.E., Yi K., et al. Speed of recovery of the most commonly performed shoulder surgeries. JSES Int. 2021;5:776–781. doi: 10.1016/j.jseint.2021.03.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Küffer J., Taha M.E., Hoffmeyer P., Cunningham G. Return to sport after shoulder arthroplasty: a systematic review. EFORT Open Rev. 2021;6:771–778. doi: 10.1302/2058-5241.6.200147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lalehzarian S.P., Agarwalla A., Liu J.N. Return to work following shoulder arthroplasty: a systematic review. World J Orthop. 2022;13:837–852. doi: 10.5312/wjo.v13.i9.837. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Levy J.C., Everding N.G., Gil C.C., Jr., Stephens S., Giveans M.R. Speed of recovery after shoulder arthroplasty: a comparison of reverse and anatomic total shoulder arthroplasty. J Shoulder Elbow Surg. 2014;23:1872–1881. doi: 10.1016/j.jse.2014.04.014. [DOI] [PubMed] [Google Scholar]
- 11.Mahendraraj K.A., Carducci M.P., Galvin J.W., Golenbock S.W., Grubhofer F., Jawa A. Reassessing the minimum two-year follow-up standard after total shoulder arthroplasty-Is one year sufficient? Shoulder Elbow. 2021;13:527–533. doi: 10.1177/1758573220922845. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mousad A.D., Beleckas C.M., Lack B., Schodlbauer D.F., Levy J.C. Anatomic and reverse total shoulder arthroplasty for osteoarthritis: outcomes in patients 80 years old and older. J Shoulder Elbow Surg. 2025;34:2122–2129. doi: 10.1016/j.jse.2024.12.014. [DOI] [PubMed] [Google Scholar]
- 13.Orange G.M., Hince D.A., Travers M.J., Stanton T.R., Jones M., Sharma S., et al. Physical function following total knee arthroplasty for osteoarthritis: a longitudinal systematic review with meta-analysis. J Orthop Sports Phys Ther. 2025;55:1–11. doi: 10.2519/jospt.2024.12570. [DOI] [PubMed] [Google Scholar]
- 14.von Rottkay E., Rackwitz L., Rudert M., Nöth U., Reichert J.C. Function and activity after minimally invasive total hip arthroplasty compared to a healthy population. Int Orthop. 2018;42:297–302. doi: 10.1007/s00264-017-3541-z. [DOI] [PubMed] [Google Scholar]
- 15.Simovitch R.W., Elwell J., Colasanti C.A., Hao K.A., Friedman R.J., Flurin P.H., et al. Stratification of the minimal clinically important difference, substantial clinical benefit, and patient acceptable symptomatic state after total shoulder arthroplasty by implant type, preoperative diagnosis, and sex. J Shoulder Elbow Surg. 2024;33:e492–e506. doi: 10.1016/j.jse.2024.01.040. [DOI] [PubMed] [Google Scholar]
- 16.Simovitch R., Flurin P.H., Wright T., Zuckerman J.D., Roche C.P. Quantifying success after total shoulder arthroplasty: the substantial clinical benefit. J Shoulder Elbow Surg. 2018;27:903–911. doi: 10.1016/j.jse.2017.12.014. [DOI] [PubMed] [Google Scholar]
- 17.Song D.J., McDermott E.R., Homeier D., Tennent D.J., Aden J.K., Ernat J.J., et al. Does resilience change in patients undergoing shoulder surgery? A retrospective comparative study utilizing the brief resilience scale. Clin Orthop Relat Res. 2025;483:1049–1059. doi: 10.1097/corr.0000000000003368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Vajapey S.P., Cvetanovich G.L., Bishop J.Y., Neviaser A.S. Psychosocial factors affecting outcomes after shoulder arthroplasty: a systematic review. J Shoulder Elbow Surg. 2020;29:e175–e184. doi: 10.1016/j.jse.2019.09.043. [DOI] [PubMed] [Google Scholar]
- 19.Vegas A., Garcia J.R., Glener J., Levy J.C. Improvement in sleep disturbance following anatomic and reverse shoulder arthroplasty. J Bone Joint Surg Am. 2023;105:1450–1457. doi: 10.2106/jbjs.23.00169. [DOI] [PubMed] [Google Scholar]
- 20.Wong S.E., Zhang A.L., Berliner J.L., Ma C.B., Feeley B.T. Preoperative patient-reported scores can predict postoperative outcomes after shoulder arthroplasty. J Shoulder Elbow Surg. 2016;25:913–919. doi: 10.1016/j.jse.2016.01.029. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.