Is Shoulder Replacement Really Riskier than Previously Thought?

    No, it’s not, despite what has been reported about a study published in The BMJ. We asked experienced shoulder surgeons to comment on the study methodology, the results, and the generalizability of the findings.

    An article recently published in The BMJ [1] on shoulder replacement surgery has generated significant “buzz” online, [2] having been picked up by major news outlets in the US and the UK, blogs, and social media. Alarming headlines have hinted that shoulder replacement is “riskier” than previously thought. Have the authors of this study found something new and damning about shoulder replacement surgery? Or has this article been over-hyped?

    To find out, ICJR asked experienced shoulder surgeons to comment on the study findings, its generalizability to shoulder replacement populations outside the UK, and the conversations surgeons should be having with their patients.

    Use the links below if you want to go directly to specific sections of the article.

    Increase in Shoulder Replacement
    Do the Risks Outweigh the Benefits?
    Counseling Patients Preoperatively
    Study Strengths and Weaknesses
    Are Practice Changes Warranted?

    Increase in Shoulder Replacement

    ICJR: The study reports a nearly 6-fold increase in shoulder replacement between 1998 and 2017 in the UK. Is this consistent with what you’ve seen in your practice? To what would you attribute the increase in the number of procedures being performed during the past 20 years? Have the indications for shoulder replacement changed in that time?

    Leesa M. Galatz, MD: The increase in shoulder replacement procedures reported in this study is consistent with what we have seen in the US and in many other countries. Not only has there been a dramatic increase in the numbers of replacements, but there has also been an increase in the number of surgeons performing these procedures. This is due to several factors, starting with training:

    • In 1998, there were approximately 10 to 12 fellowship positions in the US. Today, there are 46 positions in programs listed in the American Shoulder and Elbow Surgeons (ASES) match program, with more added each year.
    • Many sports medicine programs are promoted as having shoulder-specific training, and trauma surgeons are becoming interested in shoulder arthroplasty for fractures.

    Therefore, we have a tremendous number of fellowship-trained surgeons treating patients for shoulder problems of all types.

    Secondly, industry has focused substantial resources on developing shoulder platforms for joint replacement to meet the demands of this growing number of surgeons. Industry provides education in the use of shoulder implants, as well as access, providing some market-driven increase in its use across an even broader base of surgeons.

    Joseph A. Abboud, MD, and Eric M. Padegimas, MD: An analysis by our practice showed an increase in shoulder arthroplasty utilization nationwide, from 25,180 procedures in 2002 to 67,189 procedures in 2011, [3] which is consistent with prior analyses over a similar time period. [4,5] This has mostly been attributed to approval of the reverse shoulder arthroplasty by the US Food and Drug Administration in November 2003, as well as the increased activity of the older adult population in the US.

    Furthermore, the expansion of indications for reverse shoulder arthroplasty, from its inception as a treatment primarily for cuff tear arthropathy to its current utilization for fracture, non-union, primary osteoarthritis with significant glenoid deformity, and revision arthroplasty, has certainly played a role in the increasing incidence of shoulder replacement procedures.

    Jonathan D. Barlow, MD: Shoulder arthroplasty numbers have clearly increased over the last 20 years due to a number of factors. The efficacy of shoulder arthroplasty in improving quality of life has been increasingly documented, [6] as has the acceptable survivorship of implants for shoulder replacement. [7]

    The introduction and subsequent broader adoption of reverse total shoulder arthroplasty has expanded the indication of patients who can be helped with shoulder arthroplasty. Although indications for reverse total shoulder arthroplasty in particular remain controversial, the indication for the vast majority of cases of primary shoulder arthroplasty remains most commonly end-stage glenohumeral osteoarthritis.

    Robert U. Hartzler, MD, MS: The increase in shoulder replacement surgery noted in this study has been well-documented previously and is consistent with what I have observed in my practice. I agree with my colleagues that the incidence will continue to rise because of:

    • Demographic factors, such as an older and more-active population
    • Surgeon factors, including more surgeons trained to do shoulder replacements
    • Expanding indications, particularly with the widespread adoption of reverse total shoulder arthroplasty [4]

    Reverse total shoulder arthroplasty allows surgery to be reliably performed for certain conditions, such as rotator cuff deficiency or bone loss, in which anatomic total shoulder arthroplasty or hemiarthroplasty would have been poorly indicated or technically challenging.

    Back to the Top

    Do the Risks Outweigh the Benefits?

    ICJR: The study authors urge caution in recommending that patients undergo shoulder replacement based on these findings:

    • A high percentage of younger patients – specifically, men between 55 and 59 years old – had early failure of their shoulder implant, as did women age 85 and older.
    • The risk of revision was highest in the first 5 years after surgery, but decreased over time, especially in older patients.
    • The risk of serious adverse events such as myocardial infarction and pulmonary embolism was higher than previously reported: 1 in 28 at 30 days after surgery and 1 in 22 at 90 days after surgery.

    What are your thoughts on this: Do the risks of shoulder replacement outweigh the benefits? Is caution warranted?

    Dr. Abboud and Dr. Padegimas: The study authors urge caution in recommending shoulder replacement due to concerns of high revision rates in young patients (especially males) and high rates of “serious adverse events” (especially in older patients). However, we do not necessarily think that anything novel has been described here, and we believe that the concerns of the authors have been overstated.

    Firstly, the authors have defined “serious adverse events” very broadly. When looking at the data supplement for patients, for example, we can focus on males age 85 and older. The authors list 159 serious adverse events out of 637 patients, for a rate of 25.0%.

    However, 81 of the adverse events are lower respiratory tract infections, 31 are urinary tract infections, and 30 are renal dysfunction; that’s nearly 90% of the reported adverse events. Those are also very common events in males age 85 and older over the course of 90 days. It’s difficult to attribute them to the surgery in all cases.

    The more concerning complications – pulmonary embolus, myocardial infarction, and stroke – actually had a relatively low incidence: 17 combined complications, for a combined incidence rate of 2.7%. The authors express concern that these “serious adverse events” have essentially been undersold by the current orthopaedic literature.

    However, we believe this concern is unfounded based on our understanding of the literature and what we would consider to be a “serious adverse event”:

    • Pulmonary embolism: The 90-day incidence of pulmonary embolism is 0.3% in this study, compared with 0.2% (5 of 2885) in the largest orthopaedic study of pulmonary embolism. [8]
    • Myocardial infarction: The 90 day incidence of myocardial infarction is 0.3% in this study, compared with 0.3% (11 of 4019) in the largest orthopaedic study on myocardial infarction. [9]
    • Mortality: The 30-day and 90-day incidence of mortality is 0.23% and 0.5%, respectively, in this study. Both are lower than the 30- and 90-day mortality rates reported by the Danish registry: 0.7% and 1.5%, respectively. [10]

    In this study, the reported “serious adverse events” of pulmonary embolism, myocardial infarction, and mortality are essentially right in line with our current understanding.

    Secondly, the authors’ concern for revision rates centers mostly around younger male patients. Specifically, they found that males age 50 to 54 and males age 55 to 59 had lifetime revision rates of 19.3% and 23.6%, respectively. These rates were driven by later revisions, the majority occurring 10 years after the initial procedure.

    Younger patients with glenohumeral osteoarthritis are often the most difficult to treat because:

    • They generally have significant glenohumeral disease that started when they were quite young and often have a more complex presentation.
    • They are typically higher-demand patients trying to get back to work.

    It is entirely expected that this population would have a revision rate between 20% and 25%. This is consistent with what has previously been reported for shoulder arthroplasty in the younger patient.

    The study published in The BMJ has reported all the risks that are essentially consistent with what we would expect from both a medical and surgical perspective. Simultaneously, this study does not report on any of the benefits of shoulder replacement surgery, including patient-reported outcomes, return to work, pain relief, and so on.

    Furthermore, the study authors have included all indications for shoulder replacement surgery, including acute trauma, previous trauma, and tumor. Not all patients in the study, therefore, are having an elective procedure. This not only affects complication rates, but it also suggests that not all of these patients had the ability to choose non-operative management.

    Caution is always warranted when deciding on major elective surgery. Determining whether the risks outweigh the benefits is a shared decision made between the patient and the treating surgeon. The results of this study neither change these essential facts nor add anything novel to our understanding of the potential pitfalls with shoulder arthroplasty.

    Dr. Barlow: Dr. Abboud and Dr. Padegimas are correct: The relative risk of long-term (and short-term) failure of shoulder arthroplasty in young patients has been well documented in the literature. [11] This may be due to a number of factors, including the higher activity expectations in a younger population. Other, less obvious reasons, include the following:

    • Young patients are more likely to have complex shoulder issues – previous surgery, chondrolysis, previous infection, on so on – that make shoulder replacement more complex and increase the likelihood of failure.
    • Young patients who are candidates for shoulder replacement are more likely to have an underlying medical cause. One driving cause of shoulder replacement in young people is avascular necrosis, in which the humeral head dies. This can be the result of a number of medical concerns (eg, sickle cell anemia), medications (eg, steroids and transplant medications) or alcoholism, among other reasons. Each of these factors increases the rate of complications, failure, and medical issues after surgery.

    Even in cases in which there is a higher rate of complications, the benefits of surgery typically outweigh the risk in severe shoulder osteoarthritis. Although withholding appropriate medical/surgical treatment from these patients may decrease the overall complication rate associated with surgery, it would also prevent many more patients from achieving improved pain and function as a result of surgery.

    Dr. Galatz: The power of this study is the data reporting system available in the UK, the large numbers of patients, and the length of follow up.

    The risks of shoulder replacement do not outweigh the benefits, however.

    All procedures come with risks. Although this study highlights “at-risk” populations, it also reports a survival of 87% at 18 years, which is starting to approach the long-term results reported after hip and knee arthroplasty. Large numbers of patients are experiencing benefits of shoulder arthroplasty with high levels of satisfaction: Due to the advent of better techniques and implants, we are now able to treat shoulder pathologies that were observed, ignored, or treated suboptimally.

    This study gives surgeons the ability to temper our application of this technology in at-risk populations – specifically the very young because of the risk of revision and the very old because of the risk of medical complications. When we interact with our patients, one of the most powerful and helpful things we can do is to educate them. This study gives us data we can use in shared decision-making tools.

    Dr. Hartzler: Young age is a clearly established factor for complications, reoperations and revisions, and concerning radiographic outcomes after anatomic total shoulder arthroplasty and hemiarthroplasty HA, [12] a finding that has also been seen in the early experience of reverse total shoulder arthroplasty performed in Europe. [13] This is concerning for many reasons, not the least of which is that the demand for shoulder arthroplasty in younger patients is expected to increase dramatically in the coming years. [3]

    In my office, the conversation about risk in shoulder replacement surgery is much different for a young patient than for an older patient. Particularly for those younger than 55, no surgery will be offered unless the patient has clearly exhausted all non-operative measures. If surgery is unavoidable, I give strong consideration to an arthroscopic joint preservation option instead of a replacement. [14]

    The authors of the article in The BMJ have done some interesting editorializing about the findings of the study. They repeatedly write that the incidence rates for adverse events are “more common than previously described” or “surprisingly high” for surgery “previously considered lower risk.” No citations are provided to support this. Thus, a few comments about the study findings are in order.

    First, as Dr. Abboud and Dr. Padegimas note, the 30-day complication profile seen in this study, including mortality and adverse events, is essentially identical to what has been found in large US database studies. [15,16] The incidence of venous thromboembolism reported in this study is in keeping with what has previously been reported in the literature. [17] In fact, the risk of pulmonary embolism was found to be half of that found in hip and keep replacement surgery, and the mortality risk is lower than the population at large. That doesn’t make for sensational headlines, of course.

    Second, a high rate of complications has been previously documented in older adult patients undergoing shoulder replacement, [18] as was seen in the current study. What was surprising about the study was the very high rate of complications seen in patients age 75 and older – 1757 of 2677 (65%) of the 90-day serious adverse events – and the high rate of shoulder arthroplasties done in patients age 80 and older – 3557 of 15905 (22%) of arthroplasties.

    The risks of shoulder replacement certainly do NOT outweigh the benefits. While risk should not be underestimated, shoulder replacement surgery greatly improves the quality of life of properly indicated patients. [19]

    Back to the Top

    Counseling Patients Preoperatively

    ICJR: When you’re talking with patients who are candidates for shoulder replacement, what do you tell them about short- and long-term risks and benefits? Will this study change the way you counsel patients preoperatively?

    Dr. Hartzler: Preoperative counseling about surgical risk is one of the most important duties of surgeons. Even though I expect that patients facing a shoulder replacement will ultimately benefit from the operation, I make sure they know that the overall rate of complications from this surgery is as high as 10% to 15%. I tell them that short-term, serious complications are much less common, particularly in my practice where the vast majority of patients have outpatient or overnight surgery. [16]

    Counseling patients about the longevity of shoulder prostheses is made difficult by the fact that clinical and radiographic deterioration of the prosthetic joint doesn’t commonly result in revision surgery. When I talk with patients about the procedure, particularly younger patients, I tell them that even though the revision rate is generally low, the prosthesis – particularly the glenoid component in anatomic total shoulder arthroplasty – may be “wearing out” long before symptoms are bad enough to go through further surgery. Overall, the revision rates in the study published in The BMJ are similar to what has been reported from Mayo Clinic and the Australian Joint Registry: About a 1% per year cumulative revision rate.

    One strength of the article was that estimates of lifetime risk for revision and reoperation were reported. I think this data point will be very helpful in counselling patients, as it is easily understandable.

    Dr. Barlow: I have a thorough discussion about the medical and surgical risks (eg, stroke, heart attack, pulmonary issues, infection, dislocation, failure) around the time of surgery. At my institution, anesthesiologists review all patients regarding their fitness for surgery. An informed, combined decision by the surgeon and the patient that weighs the risks and benefits is critical. The risk of revision surgery is emphasized in discussions with younger patients and patients with complex shoulder issues.

    Dr. Galatz: I discuss the risks and complications of surgery with all my patients, including many of those reported here, in addition to local complications such as nerve and blood vessel injury. I also highlight the possibility of revision to younger patients. Younger patients have higher physical demands and higher expectations after surgery.

    What we don’t perceive in this study is the disability and pain that result from end-stage osteoarthritis. In the real world, patients are seeking relief from debilitating problems. I explain to my younger patients, that they are at a higher risk of revision, and that this is a quality-of-life issue. They can choose to have a pain-free shoulder at a younger age versus living with pain for another decade and then undergoing surgery at an older age when they may be facing deteriorating health and physical ability. Patients deserve this choice in the setting of appropriate indications.

    This study provides me with reliable numbers to counsel patients about the realistic possibility of revision surgery and, conversely, about the likelihood that they may not need revision surgery.

    Dr. Abboud and Dr. Padegimas: When talking with patients who are candidates for shoulder replacement, we always comprehensively discuss short-term and long-term risks:

    • Short-term surgical risks include fracture, dislocation, infection, and wound complications; short-term medical risks include myocardial infarction, pulmonary embolism, stroke, and mortality.
    • Long-term risks include mechanical failure of the implant, late infection, or rotator cuff failure (in anatomic shoulder arthroplasty and hemiarthroplasty).

    We do not often have formal conversations about the less-serious and multi-factorial medical complications such as respiratory infections or urinary tract infections, as these have many etiologies, especially in the older adult population. In the 90 days after the procedure, these medical complications may be completely unrelated to the surgery.

    Discussions about the benefits of shoulder arthroplasty depend on the specific arthroplasty performed and the specific diagnosis.

    • For anatomic shoulder arthroplasty done for primary osteoarthritis, we counsel patients that they can expect their VAS pain score to be between a 0 and a 1, their shoulder function to be around 90% to 95% of a normal shoulder, and their range of motion to be near normal in all planes by 1 year after surgery.
    • For reverse shoulder arthroplasty done for primary cuff tear arthropathy, we counsel patients that they can expect their VAS pain score to be between a 0 and a 1, their shoulder function to be around 85% to 90% of a normal shoulder, and their range of motion to be near normal in all planes (with the exception of internal rotation, which they will likely get to their lumbosacral junction) by 1 year after surgery.

    This study will not significantly affect how we counsel patients as the authors have stated nothing regarding the benefits of shoulder replacement surgery. We currently counsel patients in a comprehensive fashion on both the surgical and medical risks of shoulder arthroplasty. Furthermore, we already emphasize the risk of future revision arthroplasty in younger patients and the risk of medical complications in older patients with multiple comorbidities.

    Back to the Top

    Study Strengths and Weaknesses

    ICJR: What are the strengths and weaknesses of this study? Do you have any issues with the methodology and/or statistical analysis done? The data are from the UK; are the study findings generalizable to shoulder replacement in other countries?

    Dr. Abboud and Dr. Padegimas: The strength of this study is that the data come from a regulated National Health Service database in the UK. This database has mandated reporting for every hospital admission in order for the hospital to be reimbursed. Therefore, it is unlikely that hospital admissions were undercounted.

    In addition, the OPCS-4 codes appear to be comprehensive for both procedure and diagnosis. Similarly, the revision, reoperation, and complication codes appear to be comprehensive to capture all future events. However, there are a few methodological flaws that may invalidate a number of findings.

    The first problematic analysis is the authors’ interpretation of patients lost to follow-up. They state:

    “We assumed that censored participants, through loss to follow-up at the end of the study, contributed half to the time at risk. From this we calculated person time incidence rates for each year of follow-up as the quotient of the number of revisions (or reoperations) performed divided by the total time at risk.”

    This strategy of accounting for follow-up loss is derived from cancer epidemiology studies. We do not believe that methodology is appropriate here. Patients in a national registry who do not continue to seek care are:

    • Deceased, which is accounted for
    • Receiving ongoing care outside the country, likely an extremely small number
    • Simply not receiving ongoing care, which we interpret as a somewhat satisfied or, at the very least, a revision-free patient

    We believe the methodology on patients lost to follow-up likely skews the data toward a higher revision rate than really occurs.

    The second issue is the lack of a control group. We have some idea of how these patients who underwent shoulder arthroplasty fared relative to the general population, but we do not know how they fared relative to other patients with end-stage osteoarthritis of the shoulder.

    End-stage osteoarthritis can be a functionally debilitating condition. The limitations may affect not only patients’ quality of life, but also their overall health. This is not accounted for. Cross-over from the non-operative arm to the operative arm would be impossible to assess in a population-based study with a true control group of patients who did not undergo surgery.

    The third issue is the lack of assessment of functional outcomes, which is often a major limitation of a population-based study. The authors have assessed patient risk with regard to mortality, reoperation/revisions, and complications. They have not accounted for any preoperative change in functional outcomes that patients presumably would find unacceptable, evidenced by the fact that they elected to undergo shoulder arthroplasty.

    One final issue that we have is how the authors determined “serious adverse events.” We have fully described these concerns above, but essentially, the number of “serious adverse events” may be inflated, with the authors including lower-acuity complications such as urinary tract infections and respiratory infections that are quite common in the older adult population. The authors’ reporting on mortality, pulmonary embolism, and myocardial infarction are consistent with the results that have already been described in the shoulder arthroplasty literature.

    Regarding the generalizability of the data to other countries outside of the UK, the complication and revision rates in this study are consistent with what we have observed in the US, so we would presume that their findings are generalizable.

    What is more difficult to say is how generalizable patient expectations and patient-reported outcomes would be across different countries. Because patient expectations and patient-reported outcomes were not included in this study, we cannot speak to their generalizability.

    Dr. Hartzler: The main strengths of the study are the large number of included patients, the excellent long-term follow-up, and the fact that complications were followed up to 90 days postoperatively.

    The main weaknesses are that surgeon characteristics (particularly high versus low volume) and the type of prosthesis were not reported. The OPCS codes include “shoulder arthroplasty, hybrid shoulder replacement, and shoulder resurfacing.” Thus, the relative numbers of reverse total shoulder arthroplasty, anatomic total shoulder arthroplasty, and hemiarthroplasty procedures were not documented. These weaknesses are typical of large database studies.

    Dr. Galatz: This study was performed by a well-known group of researchers with a skilled group of statisticians. We can rely on the data. I do think it is relevant to the US.

    What we don’t learn from this study are the reasons for revisions in the higher-risk younger population. The reoperations (excluding revisions) are stratified, but a lot of information about the mode of failure and reasons for revision are not elucidated. This should drive our research going forward.

    Studies can and should focus on at-risk groups so that we can improve our delivery of care, and either rethink our indications and/or focus new designs on meeting the needs of younger patients. The risk of adverse events in patients over age 85 should sound a clear word of caution as we treat our older adult patients

    Dr. Barlow: Although the data being from the UK should allow generalizability, I believe there is some difficulty with generalizability related to the indications and surgical treatment of choice.

    For example, a 59-year-old male with a medical history of avascular necrosis after solid organ transplant who then undergoes reverse total shoulder arthroplasty will likely have a dramatically different survivorship than a healthy 59-year-old who undergoes anatomic total shoulder arthroplasty. Therefore, the data are of limited generalizability to the average patient.

    Back to the Top

    Are Practice Changes Warranted?

    ICJR: Should surgeons change practice based on the findings of this study? If so, in what ways? If not, why not?

    Dr. Barlow: If surgeons are not having a careful, detailed, and personalized discussion of risk with patients undergoing shoulder arthroplasty, this article should change their practice. Accepting and acknowledging that younger patients have a higher risk of complication and revision is important.

    For the vast majority of surgeons who are already engaging in these discussions in a modern shoulder practice, this can serve as another data point to help educate patients preoperatively.

    Dr. Hartzler: Surgeons should continue to offer shoulder replacements to patients who have failed non-operative treatment for advanced glenohumeral joint degenerative conditions. However, we need to take care to properly counsel patients about the short- and long-term risks of surgery, particularly those who are young (under age 55) or older (over age 75). We should do everything we can to improve modifiable risk factors. Surgeons and institutions who handle a low volume of shoulder arthroplasty cases should recognize that this clearly represents a risk factor for poor outcomes after surgery. [20]

    Dr. Galatz: Shoulder arthroplasty surgeons should be aware of this important information. The ability to share the data with patients as they consider treatment options allows both patients and physicians to put the risk of surgery in the context of the clinical scenario.

    I do not think we should withhold surgical treatment from various groups, but we can frame risk and set realistic expectations.

    Dr. Abboud and Dr. Padegimas: The main takeaways from this study are that there may be a high risk of revision surgery in young patients, especially males, and that the rates of medical complications increase specifically in the older adult population. Both of these points are already well-known and discussed in the literature and are part of every preoperative conversation. The numbers reported in this study are consistent with our understanding of the risks of surgery from both a surgical and a medical complication perspective.

    All patient undergoing elective shoulder arthroplasty must have autonomy and practice shared decision-making with their surgeons. This always includes a formal conversation about the risks and benefits of operative and non-operative management. Although this study may highlight that elective shoulder arthroplasty is not benign, there is nothing new here from a risk perspective and none of the potential benefits – improved quality of life, improved pain control, return to work, and so on – are discussed. Therefore, we are of the opinion that these findings would not change our practice.

    Author Information

    Joseph A. Abboud, MD, is board-certified orthopaedic surgeon specializing in the treatment of shoulder and elbow disorders at The Rothman Institute, Philadelphia, Pennsylvania. He is also the Senior Vice President of Clinical Affairs at The Rothman Institute

    Jonathan D. Barlow, MD, is board-certified orthopaedic surgeon specializing in the treatment of shoulder and elbow disorders at Mayo Clinic, Rochester, Minnesota. He was recently named Chair of the Division of Community Orthopedic Surgery at Mayo Clinic.

    Leesa M. Galatz, MD, is System Chair of Orthopaedics for the Mount Sinai Health System and Professor in the Leni and Peter W. May Department of Orthopaedic Surgery, Icahn School of Medicine at Mount Sinai, New York, New York.

    Robert U. Hartzler, MD, MS, is a board-certified orthopaedic surgeon specializing in the treatment of shoulder and elbow disorders at TSAOG Orthopaedics and Burkhart Research Institute for Orthopaedics (BRIO), San Antonio, Texas, and Baylor College of Medicine, Houston, Texas.

    Eric M. Padegimas, MD, is a clinical fellow in the Division of Shoulder and Elbow Surgery at The Rothman Institute, Philadelphia, Pennsylvania.


    Dr. Abboud has disclosed that that he receives royalties from and is a paid consultant for DJO Surgical; that he is paid presenter or speaker and receives research support from Tornier; that he receives royalties and research support from and is a paid consultant for Zimmer Biomet; and that he receives research support from Arthrex, Inc., and DePuy Synthes.

    Dr. Barlow has disclosed that he is a paid consultant for Stryker.

    Dr. Galatz has disclosed that she is a paid presenter or speaker and an unpaid consultant for Medacta.

    Dr. Hartzler has disclosed he is a paid consultant for Arthrex, Inc.

    Dr. Padegimas has no disclosures relevant to this article.


    1. Craig RS, Lane JCE, Carr AJ, Furniss D, Collings GS, Rees JL. Serious adverse events and lifetime risk of reoperation after elective shoulder replacement: population based cohort study using hospital episode statistics for England. BMJ. 2019 Feb 20;364:l298. doi: 10.1136/bmj.l298.
    2. Serious adverse events and lifetime risk of reoperation after elective shoulder replacement: population based cohort study using hospital episode statistics for England. Overview of attention for article published in British Medical Journal. February 2019. Accessed May 20, 2019.
    3. Padegimas EM, Maltenfort M, Lazarus MD, Ramsey ML, Willims GR, Namdari S. Future patient demand for shoulder arthroplasty by younger patients: national projections. Clin Orthop Relat Res. 2015 Jun;473(6):1860-7. doi: 10.1007/s11999-015-4231-z. Epub 2015 Mar 11.
    4. Kim SH, Wise BL, Zhang Y, Szabo RM. Increasing incidence of shoulder arthroplasty in the United States. J Bone Joint Surg Am. 2011 Dec 21;93(24):2249-54. doi: 10.2106/JBJS.J.01994.
    5. Day JS, Lau E, Ong KL, Williams GR, Ramsey ML, Kurtz SM. Prevalence and projections of total shoulder and elbow arthroplasty in the United States to 2015. J Shoulder Elbow Surg. 2010 Dec;19(8):1115-20. doi: 10.1016/j.jse.2010.02.009. Epub 2010 Jun 15.
    6. Boorman RS, Kopjar B, Fehringer E, Churchill RS, Smith K, Matsen FA 3rd. The effect of total shoulder arthroplasty on self-assessed health status is comparable to that of total hip arthroplasty and coronary artery bypass grafting. J Shoulder Elbow Surg. 2003 Mar-Apr;12(2):158-63.
    7. Brenner BC, Ferlic dc, Clayton ML, Dennis DA. Survivorship of unconstrained total shoulder arthroplasty. J Bone Joint Surg Am. 1989 Oct;71(9):1289-96.
    8. Sperling JW, Cofied RH. Pulmonary embolism following shoulder arthroplasty. J Bone Joint Surg Am. 2002 Nov;84-A(11):1939-41.
    9. Singh JA, Sperling JW, Cofield RH. Cardiopulmonary complications after primary shoulder arthroplasty: a cohort study. Semin Arthritis Rheum. 2012 Apr;41(5):689-97. doi: 10.1016/j.semarthrit.2011.09.003. Epub 2011 Nov 4.
    10. Amundsen A, Rasmussen JV, Olsen BS, Brorson S. Mortality after shoulder arthroplasty: 30-day, 90-day, and 1-year mortality after shoulder replacement–5853 primary operations reported to the Danish Shoulder Arthroplasty Registry. J Shoulder Elbow Surg. 2016 May;25(5):756-62. doi: 10.1016/j.jse.2015.09.020. Epub 2015 Dec 15.
    11. Sperling JW, Cofield RH, Rowland CM. Neer hemiarthroplasty and Neer total shoulder arthroplasty in patients fifty years old or less. Long-term results. J Bone Joint Surg Am. 1998 Apr;80(4):464-73.
    12. Sayegh ET, Mascarenhas R, Chalmers PN, Cole BJ, Romeo AA, Verma NN. Surgical Treatment Options for Glenohumeral Arthritis in Young Patients: A Systematic Review and Meta-analysis. Arthroscopy. 2015;31(6):1156-1166.e1158.
    13. Ek ET, Neukom L, Catanzaro S, Gerber C. Reverse total shoulder arthroplasty for massive irreparable rotator cuff tears in patients younger than 65 years old: results after five to fifteen years. J Shoulder Elbow Surg. 2013;22(9):1199-1208.
    14. Hartzler RU, Melapi S, de Beer JF, Burkhart SS. Arthroscopic Joint Preservation in Severe Glenohumeral Arthritis Using Interpositional Human Dermal Allograft. Arthroscopy. 2017.
    15. Anthony CA, Westermann RW, Gao Y, Pugely AJ, Wolf BR, Hettrich CM. What Are Risk Factors for 30-day Morbidity and Transfusion in Total Shoulder Arthroplasty? A Review of 1922 Cases. Clin Orthop Relat Res. 2015;473(6):2099-2105.
    16. Leroux TS, Basques BA, Frank RM, et al. Outpatient total shoulder arthroplasty: a population-based study comparing adverse event and readmission rates to inpatient total shoulder arthroplasty. J Shoulder Elbow Surg. 2016;25(11):1780-1786.
    17. Saleh HE, Pennings AL, ElMaraghy AW. Venous thromboembolism after shoulder arthroplasty: a systematic review. J Shoulder Elbow Surg. 2013;22(10):1440-1448.
    18. Foruria AM, Sperling JW, Ankem HK, Oh LS, Cofield RH. Total shoulder replacement for osteoarthritis in patients 80 years of age and older. J Bone Joint Surg Br. 2010;92(7):970-974.
    19. Cho CH, Song KS, Hwang I, Coats-Thomas MS, Warner JJP. Changes in Psychological Status and Health-Related Quality of Life Following Total Shoulder Arthroplasty. J Bone Joint Surg Am. 2017;99(12):1030-1035.
    20. Weinheimer KT, Smuin DM, Dhawan A. Patient Outcomes as a Function of Shoulder Surgeon Volume: A Systematic Review. Arthroscopy. 2017;33(7):1273-1281.