How to choose a humeral component in shoulder arthroplasty: ASES Podcast 155

Hat's off to Drs Chalmers and Waterman for yet another outstanding ASES podcast: Episode 155, in which they were joined by Drs Athwal, Cuff and Hatzidakis to discuss fixation of humeral arthroplasty components: "standard" length stems (100-150 mm), short stems (60-90 mm) and stemless designs. 

Here are some takeaways from that presentation coupled with some additional thoughts from the U.W..

(1) The evolution in humeral components is attributed to (a) mitigation of complications (periprosthetic fracture, stress shielding from high canal-filling ratios, early loosening from certain ingrowth surfaces), (b) matching competitor features, and (c) responding to market forces.

The Standard Stem

(2) A standard length stem uses the canal as an alignment guide. It can be stabilized in the canal by (a) a bony ingrowth surface, (b) cement, or (c) reaming to a tight diaphyseal fit, reaming that, in any case, removes cortical bone asymmetrically and unpredictably. [7]. As discussed later in this post, the standard stem can also be stabilized in the canal by impaction grafting. Much of the stress-shielding literature conflates two different things — stem length and canal fill — and they are not the same. Proximal stress shielding tracks with the filling ratio — how much of the canal the implant occupies and how much load it diverts away from proximal bone — far more than with the absolute length of the stem. [9,10]. A long stem that fills the canal shields bone; a long stem that sits loosely in the canal, transferring load through impacted cancellous bone, need not. 

(3) Osteoporosis or altered humeral anatomy may drive the use of a longer stem for better fixation and durability.

The Short Stem

(4) The short stem depends on loading at the metaphyseal level — in bone of variable shape and quality — making both fixation and orientation challenging.

CT of the proximal humerus showing the thin metaphyseal cortex 

The short stem may not have enough leverage to reduce loosening.

Canal filling can lead to stress shielding

And periprosthetic fracture

The absence of the control provided by a standard length stem 

can lead to malpositioning of a short stem

(5) Stemless implants depend on fixation at the anatomic neck, where a thin cortical shell surrounds cancellous bone. Bone density at this level is of variable quality, especially in patients in the arthroplasty age range. While some surgeons claim “100% stemless,” that claim is not realistic for every patient having shoulder arthroplasty. 

AP radiograph with the anatomic-neck fixation zone annotated

The big question about stemless is how a surgeon decides, in a given patient, whether good fixation can be achieved. Preoperative imaging may give some clues, but it comes down to the intraoperative findings. Some have advocated the “thumb” test;

a more reliable approach is to insert the trial nucleus and determine whether it is solid. When it is not solid because of “mushy” cancellous bone, the options are limited. In our experience, adding cancellous graft at the anatomic neck does not rescue fixation: the site is already failing in cancellous bone, and loose graft packed against soft cancellous bone has nothing rigid to lock against. 

The ability to improve fixation with longer fins is likewise limited; upsizing the trial to gain purchase can risk the lateral cortex.

 

It is also worth noting that even when stemless fixation is achieved, adaptive bone changes are common; at short-to-mid term, however, they have not reliably correlated with loosening or worse function.[6,8]

So in the real world, the stemless-loving surgeon needs a Plan “B” when trial fixation is unstable. Conversion to a stemmed component is the most attractive option. One caveat: if the problem is mushy cancellous bone, fixation of a short stem can itself be challenging — a fatter stem versus bone graft, and the same question of how to assure durability in the desired position.

Avoiding Trouble with the Stemless

Be aware of common technical mistakes - assure complete head resection and avoid excessive varus or valgus. 

Insufficient head resection

Varus cut

The free-hand cut is key.

Revision

(6) Two concerns arise when the humeral component needs revision. First, removal of a well-fixed implant can risk fracture of the tuberosity, bone loss, and shaft fracture, and may require a humeral osteotomy or window. Second, although some humeral implants are “convertible” (the humeral fixation system stays in place), retaining the implant only makes sense if it is well fixed at an appropriate height and acceptable version. One podcast participant reported that 25–40% of nominally convertible stems cannot actually be converted in practice. Taken together, the routine use of “convertible” implants seems unattractive.

Malrotation

Too low or too high position of the stem

Another approach

(7) Our experience is that secure, safely revisable humeral component fixation can be achieved with a smooth, standard-length stem set at a small filling ratio and fixed with impaction autografting — using bone from the resected humeral head that one of our fellows named “God’s Own Glue” (see "Procrustean Method") [2,3]. The graft is impacted between a smooth stem and the endosteal cortex, where it stabilizes the implant at the time of surgery, rather than relying on a bony ingrowth surface, cement, or a tight diaphyseal press fit. [4]. A low filling ratio protects the humerus from stress shielding. Because the stem is deliberately undersized, load is transferred through grafted cancellous bone rather than bypassed down a canal-filling implant — addressing stress shielding at the variable that drives it. [9,10]. 

In 48 ream-and-run and 78 total shoulder arthroplasties using a smooth, standard-length impaction-grafted stem, two-year radiographs showed adaptive changes that were generally minor and not associated with component shift or subsidence. Inserted this way, a smooth standard-length stem offers secure, bone-preserving fixation providing results can serve as a basis for comparison for other component designs and fixation methods. [5]. 

In a consecutive single-surgeon series of 458 anatomic total shoulder arthroplasties using this construct (mean follow-up 9.2 years; 114 shoulders beyond ten years), Simple Shoulder Test scores improved from 3.3 to 9.2 and were sustained — never declining by more than the MCID — past ten years. The overall revision rate was 2.6% (12 of 458). None of the revisions were performed for a humeral-component cause: there was no humeral component loosening, no periprosthetic fracture, and no stem-related failure.[11] 

 A low filling ratio protects the humerus from stress shielding

 

Two year followup

Six year followup

Impaction grafting with a low filling ratio avoids incomplete seating of the humeral stem (humerus captivus)

Impaction grafting a standard stem in revising a failed short stem

Impaction grafting a thin long stem  in humeral deformity

Impaction grafting after fracture fixation

Impaction grafting a thin standard stem enables easy, safe removal followed by new implant insertion should revision be necessary.

Thus "revisability" does not require "convertibility"

Returning to the observation stated at the beginning of this post: humeral component evolution is attributed to (a) mitigation of complications, (b) matching competitor features, and (c) responding to market forces. 

Impaction grafting of a thin, smooth stem is an approach for addressing  periprosthetic fracture, stress shielding from high canal-filling ratios, and early loosening. It is not driven by market forces. Impaction grafting can be used with any standard length stem (anatomic or reverse) combined with freely available autograft from the patient’s own humeral head that would otherwise be discarded.

Should we be thinking straight?

Beauty in Simplicity

Tundra Swan

Union Bay Natural Area

Disclosure. The author has no financial relationship with any manufacturer of the orthopaedic devices discussed in this post.

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis

Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133//

References of interest

1. The ASES Podcast (American Shoulder and Elbow Surgeons), Episode 155: standard, short, and stemless humeral components and convertible designs. Available at: https://www.youtube.com/watch?v=q_s_g8oQgPg

2. Razfar N, Reeves JM, Langohr GDG, Willing R, Athwal GS, Johnson JA. Comparison of proximal humeral bone stresses between stemless, short stem, and standard stem length: a finite element analysis. J Shoulder Elbow Surg. 2016;25(7):1076–1083. doi:10.1016/j.jse.2015.11.011. PMID 26810016.

3. Reeves JM, Langohr GDG, Athwal GS, Johnson JA. The effect of stemless humeral component fixation feature design on bone stress and strain response: a finite element analysis. J Shoulder Elbow Surg. 2018;27(12):2232–2241. doi:10.1016/j.jse.2018.06.002. PMID 30104100.

4. Synnott S, Langohr GDG, Reeves JM, Johnson JA, Athwal GS. The effect of humeral implant thickness and canal fill on interface contact and bone stresses in the proximal humerus. JSES Int. 2021;5(5):881–888. doi:10.1016/j.jseint.2021.05.006.

5. Aibinder WR, Uddin F, Bicknell RT, Krupp R, Scheibel M, Athwal GS. Stress shielding following stemless anatomic total shoulder arthroplasty. Shoulder Elbow. 2023. doi:10.1177/17585732211058804. PMID 36895609.

6. Raiss P, Edwards TB, Deutsch A, Shah A, Bruckner T, Loew M, Boileau P, Walch G. Radiographic changes around humeral components in shoulder arthroplasty. J Bone Joint Surg Am. 2014;96(7):e54. doi:10.2106/JBJS.M.00378. PMID 24695931.

7. Denard PJ, Raiss P, Gobezie R, Edwards TB, Lederman E. Stress shielding of the humerus in press-fit anatomic shoulder arthroplasty: review and recommendations for evaluation. J Shoulder Elbow Surg. 2018;27(6):1139–1147. doi:10.1016/j.jse.2017.12.020. PMID 29422391.

8. Sheth MM, Kahsai EA, Yang J, Whitson AJ, Matsen FA III, Hsu JE. What is the clinical importance of radiographic changes around the humeral component in anatomic shoulder arthroplasty? A minimum 4-year follow-up study. J Shoulder Elbow Surg.2025;34(8):1877–1885. doi:10.1016/j.jse.2024.11.024. PMID 39800107.

9. Denard PJ, Hsu JE, Whitson A, Neradilek MB, Matsen FA III. Radiographic outcomes of impaction-grafted standard-length humeral components in total shoulder and ream-and-run arthroplasty: is stress shielding an issue? J Shoulder Elbow Surg.2019;28(11):2181–2190. doi:10.1016/j.jse.2019.03.016. PMID 31272887.

10. Kim SC, Park JH, Bukhary H, Yoo JC. Humeral stem with low filling ratio reduces stress shielding in primary reverse shoulder arthroplasty. Int Orthop. 2022;46(6):1341–1349. doi:10.1007/s00264-022-05383-4.

11. Lee M, Chebli C, Mounce D, Bertelsen A, Richardson M, Matsen FA III. Intramedullary reaming for press-fit fixation of a humeral component removes cortical bone asymmetrically. J Shoulder Elbow Surg. 2008;17(1):150–155. doi:10.1016/j.jse.2007.03.032. PMID 18029200.

12. Boorman RS, Hacker SA, Lippitt SB, Matsen FA III. A conservative broaching and impaction grafting technique for humeral component placement and fixation in shoulder arthroplasty: the Procrustean method. Tech Shoulder Elbow Surg.2001;2(3):166–175. doi:10.1097/00132589-200109000-00004.

13. Hacker SA, Boorman RS, Lippitt SB, Matsen FA III. Impaction grafting improves the fit of uncemented humeral arthroplasty. J Shoulder Elbow Surg. 2003;12(5):431–435. doi:10.1016/s1058-2746(03)00053-3. PMID 14564262.

14. Lucas RM, Hsu JE, Gee AO, Neradilek MB, Matsen FA III. Impaction autografting: bone-preserving, secure fixation of a standard humeral component. J Shoulder Elbow Surg. 2016;25(11):1787–1794. doi:10.1016/j.jse.2016.03.008. PMID 27262410.

15. Lee et al. Stress shielding effects of short stem alignment and bone density in reverse shoulder arthroplasty. J Orthop Res. 2026. doi:10.1002/jor.70140.

16. Vasiliadis AV, Giovanoulis V, Lepidas N, Bampis I, Servien E, Lustig S, Gunst S. Stress shielding in stemmed reverse shoulder arthroplasty: an updated review. SICOT J. 2024;10:37. doi:10.1051/sicotj/2024029.

17. Ritter D, Raiss P, Denard PJ, Werner BC, Müller PE, Woiczinski M, Wijdicks CA, Bachmaier S. Volumetric humeral canal fill ratio effects primary stability and cortical bone loading in short and standard stem reverse shoulder arthroplasty: a biomechanical and computational study. J Imaging. 2024;10(12):334. doi:10.3390/jimaging10120334.

18. Kramer M, Olach M, Zdravkovic V, Manser M, Raiss P, Jost B, Spross C. The effects of length and width of the stem on proximal humerus stress shielding in uncemented primary reverse total shoulder arthroplasty. Arch Orthop Trauma Surg. 2024. doi:10.1007/s00402-023-05129-w.

19. John PB, Nageswaran S. Mechanobiological evaluation of solid and multiple porous humeral stem architectures in reverse shoulder arthroplasty based on design and materials: a finite element study. Front Bioeng Biotechnol. 2026;13:1675726. doi:10.3389/fbioe.2025.1675726.

20. Takayama K, Ito H. Association between the canal filling ratio and bone resorption in trabecular metal stems in reverse total shoulder arthroplasty: a radiographic analysis using tomosynthesis. JSES Int. 2024;8(5):1077–1086. doi:10.1016/j.jseint.2024.05.010.

Are patients having shoulder arthroplasty doing better over the last decade, if so, why?

This post examines two questions of importance to the surgeons and patients interested in shoulder arthroplasty:

(1) Over the past decade, is there evidence that patient-reported outcomes (PROs) after shoulder arthroplasty have gotten better by a clinically significant amount — that is, by at least the minimal clinically important difference (MCID)?

(2) If they have, is that improvement associated with the surgeon, the technology, the shoulder, or the patient?

Based on the available evidence, the answers are: (1) yes, modestly; and (2) the gain is not attributable to new technology or to any change in the shoulders we treat. The largest lever on the outcome is the patient, while the decade’s trend itself most plausibly reflects accumulated surgical experience and care.

A word on the measure. Surrogates — such as radiographic component position, glenoid version, three-dimensional planning accuracy, and data on implant survivorship drawn from registries — are not the outcomes of primary interest. The important outcome is what the patients report. A technology that improves a surrogate but does not move a PRO beyond its MCID has not improved the outcome for the patient.

Now the detail.

1. Are patient-reported outcomes improving over time?

Anatomic TSA (aTSA). In a single-institution series of 1,899 arthroplasties (2008–2018), the 5-year ASES improvement rose by 1.65 points per year of index surgery on univariable analysis and by 2.20 points per year after adjustment for patient factors [1]. The per-year increment is small — well under one MCID — and reaches clinical significance only when accumulated across the decade. The authors could not attribute the gain to any single factor [1].

Reverse TSA (RSA). Some of the most relevant data come from the U.K. National Joint Registry (24,411 RSAs, 2013–2021): mean 6-month Oxford Shoulder Score (OSS) improvement rose from 15.84 to 20.29 — about half a point per year, roughly one MCID across the whole period [2].

In the Danish Shoulder Arthroplasty Registry (2,867 arthroplasties for osteoarthritis, 2006–2015), the adjusted WOOS score rose by about 10 points [3] — short of the WOOS MCID of 12.3 [16].

So.....both aTSA and RSA show measured, year-over-year PRO improvement. The increments are small per year and reach clinically meaningful magnitude only when accumulated across the decade — about one MCID for RSA across the period, and a larger accumulated difference for aTSA, whose adjusted per-year gain compounds over ten years. These are population- and cohort-level shifts: a patient operated late in the decade does modestly better, on average, than one operated early.

Survivorship bias. A patient-reported outcome plotted against time after surgery is available only for survivors: to contribute a 6-month OSS or a 5-year ASES, a patient must be alive, unrevised, and reachable. Those who die, are revised, or move their care to another surgeon before the minimum follow-up never enter the average — which is therefore selected for better outcomes.

Which way this biases the trend is not obvious. Perioperative mortality and early reoperation  fell across this same decade [2], so fewer patients were removed by death in the later years; if anything the later cohorts are less selected, and across-year survivorship is unlikely to be what produced the rising trend. The selection that matters is not across calendar years but across follow-up length — the late-timepoint averages rest on progressively smaller, more selected subsets of the original cohorts.

Implant durability is vulnerable to the same problem in a more specific way. When revision is estimated with a standard Kaplan–Meier survival curve, a patient who dies is treated as still at risk of a revision that can no longer occur. This overestimates the cumulative revision rate, and it overestimates most where mortality is highest — the older patients who receive reverse  replacements [15]. Comparisons of durability across groups with different mortality are confounded for the same reason: part of any apparent difference reflects who lived long enough to ever reach a revision, not how the implant performed. Revision rates in older, higher-mortality groups should accordingly be read as upper bounds rather than firm figures [15].

The usual adjustments for age, comorbidity, and surgeon volume operate only among the patients who actually reported a score; they cannot recover those missing because they died or never returned. So even where adjustment strengthens the trend [1], survivorship bias remains in the data. It matters most where follow-up completion is low and uneven, as it was in the National Joint Registry [2].

So two cautions in reading these data. The within-period findings — smoking [9], unemployment and education [10], resilience and mental health [11], and the absence of any effect of glenoid shape [5] or technology [4] — compare patients operated on in the same period, so the across-year attrition that inflates the temporal trend cannot account for them. The year-over-year trend is the vulnerable part: surgeon experience, technique, rehabilitation, and perioperative care all improved over the same years, so a dataset of this kind cannot apportion the gain among them; the measured improvement is best taken as the most that can be claimed, not a precise figure.

2. The surgeon, the technology, the shoulder, or the patient — what is the change associated with?

The surgeon. Surgeon volume and fellowship training have a well-documented effect on revision, reoperation, complications, and length of stay — and a weak-to-absent effect on the size of the PRO improvement patients realize. In the National Joint Registry, surgeons averaging at least 10.4 shoulder replacements a year had lower revision risk (the hazard of revision roughly doubled below that threshold) along with fewer reoperations, fewer serious adverse events, and shorter stays, but no patient-reported outcome was assessed [12]. Fellowship training shows the same pattern: shoulder-and-elbow and sports-medicine fellowship-trained surgeons have significantly lower complication rates at 90 days, 1 year, and 5 years, again with no PRO assessed [14]. The most direct test comes from the Australian registry: a surgeon’s two-year revision rate showed no clinically relevant correlation with PROMs, and all surgeons produced similar postoperative Oxford scores [13]. Revision rates and PROs measure two different facets of the result — surgeon factors move the first; patient factors move the second.

The technology. Advanced technology does not appear to be related to patient-reported outcomes. A systematic assessment of three-dimensional planning, patient-specific instrumentation, navigation, stemless humeral components, and augmented glenoid components found that no individual technology was associated with a statistically or clinically significant improvement in PROs [4]. A single-surgeon consecutive series of 389 anatomic arthroplasties reaches the same conclusion from the other side: implant choice and surgical technology do not predict outcome [5]. The durable result there came from fundamentals — subscapularis management, a conservative neck cut, a securely seated all-polyethylene glenoid, soft-tissue balancing — not new technology.

Cross-linked polyethylene is the one material change with an apparent benefit: in the Australian registry it had a lower anatomic revision rate than conventional polyethylene [6]. But the benefit is on durability — reduced wear-particle osteolysis and loosening — not on comfort and function; no study has shown that it improves PROs. And the advantage appears smaller in contemporary practice, where glenoid component design and fixation, not polyethylene type, emerge as the dominant predictor of revision [7].

The shoulder. Glenoid morphology — the usual proxy for a “worse” arthritic shoulder — does not predict outcome: in a 389-case series, B2 and B3 glenoids met or exceeded the outcomes of other glenoid types, and symptomatic glenoid loosening occurred in 1 of 389 shoulders [5]. Nor has the case mix drifted toward milder disease. In the Australian registry, the reverse diagnosis mix barely changed across the decade — osteoarthritis fell only slightly as a share of reverse procedures (48.2% to 44.7%) while cuff arthropathy rose by almost the same margin (36.2% to 39.7%) — if anything a drift toward classic cuff-deficient disease, the opposite of an “easier shoulders” explanation [8]. A shift in shoulder characteristics does not appear to be driving the PRO gain.

The patient. The data point to the patient — and to the judgment applied in selecting and preparing the patient. Three patient-side domains stand out. Modifiable risk: in an analysis of 14,465 arthroplasties, smoking was independently associated with increased surgical complications [9]; cessation, along with glycemic, nutritional, and opioid optimization, appears to improve the result. Socioeconomic context: in a nationwide cohort of 2,292 arthroplasties, after adjustment for age, sex, diagnosis, implant, and comorbidity, unemployment was associated with a WOOS deficit of roughly 14 to 19 points at one year and lower educational attainment with a further 5 to 8 points [10] — differences that exceed the MCID. Resilience and mental health: in 399 anatomic arthroplasties, greater resilience and better mental health were associated with better outcomes [11].

So among the candidate factors, the patient is the strongest predictor of the outcome a given arthroplasty achieves — selection, optimization, socioeconomic context, and resilience each move the PRO, several beyond the MCID. Whether the patient mix shifted across the decade in a way that would explain the time trend is a separate question, and these data do not show that it did.

Conclusion

The evidence answers both opening questions.

First, patient-reported outcomes after shoulder arthroplasty have improved over the past decade — modestly, and by an amount that reaches the MCID only when accumulated across the period, not within any single year.

Second, two different questions hide inside "what is the change associated with," and they have different answers.

(A) The decade-long trend is real but cannot be apportioned by the available data; it is most plausibly the product of accumulated surgeon experience, refinements in technique, rehabilitation, and perioperative care. The case mix did not drift toward easier shoulders [8], and no individual technology was associated with the improved outcomes over the decade [4].

(B) The other question — what determines the result for an individual patient — has a clearer answer. The available evidence has not shown a patient-reported benefit from any technology assessed, nor from any change in the shoulders we treat. Surgeon volume and training affect revision and complication rates, but not, as far as the data show, the patient-reported result. That leaves the patient: selection, optimization of modifiable risk, socioeconomic context, and resilience.

The greatest demonstrated effect on the outcome therefore lies in choosing and preparing the patient; the experience the surgeon brings to the procedure most likely matters too, though these data cannot show by how much.

Market forces are clearly driving orthopaedic companies to develop new implants and technologies that strengthen their competitive position, even though evidence that these have a major impact on patient outcomes is lacking.

What this analysis does offer, though, is actionable intelligence: the surest routes to improving outcomes for our patients are (1) thoughtful patient selection and optimization and (2) education that enables surgeons to improve their craft of shoulder arthroplasty.

Will the future be better?

Broad-billed Hummingbird

Tucson

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis

Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133//

References

[1] Mathew JI, Nicholson AD, Finocchiaro A, Okeke L, Dines DM, Dines JS, Taylor SA, Warren RF, Gulotta LV. Outcomes of shoulder arthroplasty by year of index procedure: are we getting better? J Shoulder Elbow Surg. 2022;31(2):245-251. doi:10.1016/j.jse.2021.08.024. PMID 34592407.

[2] O’Malley O, Davies A, Taghavi Azar Sharabiani M, Rangan A, Sabharwal S, Reilly P. Are we getting better over time? Clinical and patient-reported outcomes for reverse shoulder arthroplasty: a National Joint Registry cohort study. BMJ Open. 2025;15(9):e096084. doi:10.1136/bmjopen-2024-096084. PMID 40953878.

[3] Rasmussen JV, Amundsen A, Sørensen AKB, Klausen TW, Jakobsen J, Jensen SL, Olsen BS. Increased use of total shoulder arthroplasty for osteoarthritis and improved patient-reported outcome in Denmark, 2006-2015: a nationwide cohort study from the Danish Shoulder Arthroplasty Registry. Acta Orthop. 2019;90(5):489-494. doi:10.1080/17453674.2019.1633759. PMID 31240980.

[4] Schiffman CJ, Prabhakar P, Hsu JE, Shaffer ML, Miljacic L, Matsen FA 3rd. Assessing the value to the patient of new technologies in anatomic total shoulder arthroplasty. J Bone Joint Surg Am. 2021;103(9):761-770. doi:10.2106/JBJS.20.01853. PMID 33587515.

[5] Schiffman CJ, Chin, Whitson AJ, Hsu JE, Matsen FA 3rd. Anatomic shoulder arthroplasty: a single surgeon’s consecutive series of 458 patients (389 anatomic TSAs, mean 9.2-year follow-up). University of Washington SHORE Registry. [Manuscript under review.]

[6] Page RS, Alder-Price AC, Rainbird S, Graves SE, de Steiger RN, Peng Y, Holder C, Lorimer MF, Gill SD. Reduced revision rates in total shoulder arthroplasty with crosslinked polyethylene: results from the Australian Orthopaedic Association National Joint Replacement Registry. Clin Orthop Relat Res. 2022;480(10):1940-1949. doi:10.1097/CORR.0000000000002293. PMID 35901440.

[7] Gill DRJ, Corfield S, Harries D, Page RS. Modeling highly crosslinked polyethylene vs. non-highly crosslinked polyethylene glenoid revision rates for anatomic shoulder arthroplasty in osteoarthritis including differing polyethylene glenoid fixation designs. Semin Arthroplasty JSES. 2024;34(4):843-853. doi:10.1053/j.sart.2024.06.003.

[8] Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR). Hip, Knee & Shoulder Arthroplasty: 2025 Annual Report. Adelaide: Australian Orthopaedic Association; 2025.

[9] Althoff AD, Reeves RA, Traven SA, Wilson JM, Woolf SK, Slone HS. Smoking is associated with increased surgical complications following total shoulder arthroplasty: an analysis of 14,465 patients. J Shoulder Elbow Surg. 2020;29(3):491-496. doi:10.1016/j.jse.2019.07.012. PMID 31519425.

[10] Jensen ML, Valsamis EM, Madrid AS, Olsen BS, Rasmussen JV. Association between socioeconomic status and patient-reported outcome at 1 year after shoulder arthroplasty for osteoarthritis or cuff-tear arthropathy: a nationwide cohort study of 2,292 arthroplasties. Acta Orthop. 2025;96:45-51. doi:10.2340/17453674.2024.42700.

[11] Levins JG, Dasari SP, Quinlan NJ, Whitson AJ, Matsen FA 3rd, Hsu JE. Anatomic shoulder arthroplasty: the correlation between patient resilience, mental health, and outcome. J Shoulder Elbow Surg. 2024;33(6S):S9-S15. doi:10.1016/j.jse.2024.03.008. PMID 38548096.

[12] Valsamis EM, Collins GS, Pinedo-Villanueva R, Whitehouse MR, Rangan A, Sayers A, Rees JL. Association between surgeon volume and patient outcomes after elective shoulder replacement surgery using data from the National Joint Registry and Hospital Episode Statistics for England: population based cohort study. BMJ. 2023;381:e075355. doi:10.1136/bmj-2023-075355. PMID 37343999.

[13] Hoskins W, Bingham R, Corfield S, Harries D, Harris IA, Vince KG. Do the revision rates of arthroplasty surgeons correlate with postoperative patient-reported outcome measure scores? A study from the Australian Orthopaedic Association National Joint Replacement Registry. Clin Orthop Relat Res. 2024;482(1):98-112. doi:10.1097/CORR.0000000000002737. PMID 37339166.

[14] Harkin WE, Saad Berreta R, Turkmani A, Williams T, Scanaliato JP, McCormick JR, Nicholson GP, Garrigues GE. How fellowship training affects complication rate after shoulder arthroplasty: a nationwide assessment. J Shoulder Elbow Surg. 2025;34(2):499-506. doi:10.1016/j.jse.2024.05.014. PMID 38944372.

[15] Lacny S, Wilson T, Clement F, Roberts DJ, Faris PD, Ghali WA, Marshall DA. Kaplan-Meier survival analysis overestimates the risk of revision arthroplasty: a meta-analysis. Clin Orthop Relat Res. 2015;473(11):3431-3442. doi:10.1007/s11999-015-4235-8. PMID 25804881.

[16] Nyring MRK, Olsen BS, Amundsen A, Rasmussen JV. Minimal clinically important differences (MCID) for the Western Ontario Osteoarthritis of the Shoulder Index (WOOS) and the Oxford Shoulder Score (OSS). Patient Relat Outcome Meas. 2021;12:299-306. doi:10.2147/PROM.S316920. PMID 34588833.

Why are surgeons performing reverse total shoulders rather than anatomic arthroplasties for cuff intact arthritis?

 In the prior post we reviewed the evidence that - in contrast to reverse arthroplasty (RSA) - anatomic total shoulder arthroplasty (aTSA) is less expensive and provides patients with better comfort and function, fewer serious complications, and safer revision options for complications should they occur.

Paradoxically, however, the proportion of aTSAs being performed for cuff-intact arthritis is dropping precipitously.  

We can speculate on possible reasons for this paradox.

(1) Surgeons may perceive that a lower revision rate for RSA is a positive factor for the patient, when in fact the lower RSA revision rate is due in large part to the fact that some of the most common and serious RSA complications are often not revisable (e.g. pain and poor function, displaced acromial/spine fractures). 

(2) Surgeons may perceive that the RSA is easier to perform. This is, of course, due to the fact that few surgeons have training/experience in performing a basic aTSA, not that the operation is of itself more challenging.

(3) Industry influence and conflicts of interest preferentially motivate the more expensive/profitable RSA option.

(4) Recent "innovations" targeting the use of preoperative planning to achieve high levels of "accuracy and precision" - that may be clinically irrelevant - can make the aTSA unnecessarily complex, expensive and daunting. 

The solution may lie in assuring that shoulder surgeons are well trained in both aTSA and RSA.  This requires that organizations such as AAOS and ASES provide hands-on educational opportunities and that training programs assure that their fellows and residents have a meaningful experience in both.  Interestingly in our most recent round of interviews for our fellowship, a number of applicants reported they had never seen, much less performed, an aTSA.

Below is the basic approach I use for anatomic arthroplasty presented at the amazing Nice Shoulder Course of Pascal Boileau.  These steps may be helpful for surgeons wishing to build their aTSA skills.

Keeping it simple

House Finch

Matsen yard

Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/