In recent years, commonly performed orthopaedic procedures that once required an inpatient hospital stay of several days are now routinely performed on an ambulatory basis in appropriately selected cases1-6. This practice has been supported by institutional series reporting no significant differences in readmission rates, reoperations, or complications following ambulatory lower-extremity arthroplasty among selected patients4,6. Ambulatory total hip and knee arthroplasties also appear to provide substantial savings to the health-care system6,7.
Since 2001, the number of total shoulder arthroplasties performed in the United States has increased dramatically, with >50,000 currently being performed per year8, representing a 7% to 13% increase per year9,10. This number is increasing at a substantially higher rate than that of lower-extremity arthroplasty, making total shoulder arthroplasty one of the most commonly performed elective procedures in the elderly population today8-10. Regional anesthetic techniques and comprehensive clinical and pain management pathways have been applied to selected patients undergoing ambulatory total shoulder arthroplasty, potentially obviating the need for a postoperative inpatient hospital stay11. With the increasing number of procedures being performed annually, successful ambulatory total shoulder arthroplasty could result in a reduction in health care-related expenditures. However, the current trend for ambulatory arthroplasty must be balanced against the newly proposed Medicare bundled payment system, which does not reimburse hospitals or providers for early postoperative readmissions12,13. Given the already short hospital course following inpatient total shoulder arthroplasty, it is important to balance the goal of ambulatory surgical procedures against the potentially increased risk of complications that could be devastating, particularly in an outpatient setting. Although substantial literature has been published on the complications and risk factors for readmission following inpatient total shoulder arthroplasty, to our knowledge, no similar studies have been performed examining patients undergoing ambulatory total shoulder arthroplasty14-19. Furthermore, to our knowledge, the only existing studies of outpatient arthroplasty have been for the lower extremity3,6,7.
The purposes of the present study were to employ a national insurance database to investigate the current trends in ambulatory total shoulder arthroplasty in the United States; to characterize the rate of postoperative complications, hospital readmission, and risk factors associated with readmission following ambulatory total shoulder arthroplasty compared with a matched cohort of inpatient total shoulder arthroplasty; and to conduct a cost analysis comparing ambulatory total shoulder arthroplasty with matched inpatient total shoulder arthroplasty. We hypothesized that there has been a significant increase in the number of ambulatory total shoulder arthroplasties in recent years, that overall readmission rates and postoperative complication rates for patients undergoing ambulatory total shoulder arthroplasty would be similar to those for patients undergoing inpatient total shoulder arthroplasty, and that ambulatory total shoulder arthroplasty would result in a significant reduction in costs.
Materials and Methods
An insurance-based database, PearlDiver (http://www.pearldiverinc.com), was utilized for the study. This database contains data from several different insurers, including both Medicare and private insurers, such as UnitedHealthcare (UHC) and Humana. The patients from the present study were all taken from the Humana database within PearlDiver, as it contains the most accurate service location identifiers among the available databases. A private insurance database was chosen over the Medicare database, as private insurers are more likely to reimburse for ambulatory total shoulder arthroplasty.
The PearlDiver Humana database contains procedural volumes, patient demographic characteristics, prescription drug information, and cost data for patients with International Classification of Diseases, Ninth Revision (ICD-9) diagnoses and procedures or Current Procedural Terminology (CPT) codes. The database covers patients insured from 2007 to the second quarter of 2015, and patients can be tracked across all locations. Overall, the database contains approximately 20 million patients with orthopaedic diagnoses. All data were de-identified and were thus exempt from institutional review board approval.
Study and Control Cohorts
The database was first queried for patients who underwent total shoulder arthroplasty using CPT code 23472 and ICD-9 procedure code 81.80. Patients who underwent reverse total shoulder replacement (ICD-9 code 81.88) or shoulder hemiarthroplasty (ICD-9 code 81.81) were excluded from the analysis. As the ICD-9 code for reverse total shoulder replacement was not introduced until the fourth quarter of 2010, the analysis used data starting from the fourth quarter of 2010 through the end of 2014. Patients with a diagnosis of acute proximal humeral fracture or a diagnosis of shoulder osteonecrosis were also excluded using ICD-9 diagnosis codes, resulting in 11,747 patients who had undergone primary anatomic total shoulder arthroplasty and were available for further study. Next, patients who underwent total shoulder arthroplasty in an ambulatory setting were identified, first by selecting only those patients with a service location of an ambulatory surgical center or hospital outpatient surgical center. These patients were then confirmed to have a length of stay of 0 days and no inpatient hospital codes or inpatient admission. Patients with postoperative inpatient observation status codes, including CPT codes 99231 to 99235 and Healthcare Common Procedure Coding System (HCPCS) codes G0378 and G0379, were also excluded from the ambulatory total shoulder arthroplasty group. To reduce the effects of transient patients, that is, those who might lose insurance because of employment change or retirement in the immediate postoperative period, only patients with a minimum of 90 days of database activity were included. This filtering resulted in 706 patients undergoing ambulatory total shoulder arthroplasty, which formed the study group.
A matched control group of inpatient total shoulder arthroplasty was formed from the remaining 11,041 patients who underwent total shoulder arthroplasty with postoperative admission to an inpatient facility. Potential patients undergoing inpatient total shoulder arthroplasty were matched to patients undergoing ambulatory total shoulder arthroplasty on the basis of the following parameters: 10-year age group, sex, obesity, tobacco use, diabetes mellitus, coronary artery disease, chronic kidney disease, and a diagnosis of chronic anemia within a year prior to total shoulder arthroplasty (see Appendix). The algorithm employed matches all possible patients undergoing inpatient total shoulder arthroplasty to each patient undergoing ambulatory total shoulder arthroplasty for each matched variable. On the basis of the above parameters, 4,459 patients undergoing inpatient total shoulder arthroplasty were matched to patients undergoing ambulatory total shoulder arthroplasty.
Complications were assessed for both the patients undergoing ambulatory total shoulder arthroplasty and the matched patients undergoing inpatient total shoulder arthroplasty. The database captures all episodes of patient care in which insurance was billed; thus, complications were assessed in the office, clinic, emergency department with and without admission, and any hospital setting. Complications were assessed using ICD-9 and CPT codes (see Appendix). Medical complications were assessed within 90 days postoperatively. Shoulder-specific complications included infection within 90 days postoperatively and prosthetic dislocation, prosthetic loosening, shoulder stiffness, fracture, and early revision total shoulder arthroplasty within 1 year postoperatively. Overall medical and shoulder-specific complication rates were also determined, defined as the number of patients with at least 1 medical or shoulder-specific complication. The incidence of hospital readmission was assessed within 30 days and within 90 days postoperatively.
Risk factors and reasons for readmission within 90 days postoperatively among patients undergoing ambulatory total shoulder arthroplasty were assessed. Numerous risk factors were examined, including demographic characteristics and medical comorbidities. Demographic characteristics included sex, age, obesity (body mass index [BMI] of 30 to 39.9 kg/m2), morbid obesity (BMI of ≥40 kg/m2), tobacco or alcohol abuse, and race. Reasons for readmission were queried and were divided into medical codes (see Appendix), implant-related codes (see Appendix and additional ICD-9 codes 959.2, 996.40, 996.47, 996.77), and acute pain (ICD-9 codes 338.18-338.19 and 338.4) as the diagnoses of record for the admission(s).
Thirty-day postoperative costs were compared between patients who underwent ambulatory total shoulder arthroplasty and matched inpatient controls. All costs are expressed as actual reimbursements provided by the insurance company and therefore do not reflect the actual cost to provide the procedure. Costs were divided into itemized reimbursements and grouped reimbursements. Itemized reimbursements evaluated included surgical costs (reimbursement for the surgical CPT code); concomitant procedures; anesthesia; imaging; postoperative anesthesia care unit (PACU) observation, admission, and discharge; laboratory; inpatient and outpatient physical therapy and occupational therapy; home health other than physical therapy or occupational therapy; follow-up visits; and postoperative prescriptions. The grouped reimbursement that was compared was the reimbursement for the diagnosis-related group (DRG) billed by the hospital. The DRG includes the entire fixed reimbursement paid to the hospital, including surgical and inpatient care costs.
Data and Statistical Analysis
The trend in procedural volume over time was analyzed using a linear regression, with significance set at p < 0.05. Comparisons of demographic characteristics, comorbidities, complications, and risk factors and reasons for readmission were completed using a multivariate binomial logistic regression analysis, controlling for both patient demographic characteristics and medical comorbidities. Odds ratios (ORs) were calculated with respective 95% confidence intervals (CIs), with p < 0.05 considered significant. For cost data, the mean reimbursements per patient were compared between cohorts using a Student t test. SPSS (version 23 for Macintosh; IBM) was used for all statistical calculations.
The yearly incidence of ambulatory total shoulder arthroplasty for this single insurer increased from 108 (2011) to 224 (2014), representing a 107% increase (p < 0.0001). (The fourth quarter of 2010 was not included, as only full years were calculated.) There were no significant differences (p > 0.05) in demographic characteristics between the patients undergoing ambulatory total shoulder arthroplasty and matched inpatient controls (Table I).
A comparison of postoperative complications between patients undergoing ambulatory total shoulder arthroplasty and the matched controls is provided in Table II. In no instances were any complications present at a significantly higher rate (p > 0.05) in the patients who underwent ambulatory total shoulder arthroplasty; in fact, several complications occurred significantly more frequently in the control group: urinary tract infection (p = 0.003) and blood transfusion (p = 0.028). The rate of hospital readmission after discharge was not significantly different at 30 days or 90 days between the 2 cohorts (Table II).
Among patients undergoing ambulatory total shoulder arthroplasty, several risk factors were identified for readmission within 90 days postoperatively (Table III). Obesity (p = 0.016) and morbid obesity (p = 0.029) were the only significant independent demographic risk factors for readmission. Several medical comorbidities were identified as risk factors for readmission, including diabetes mellitus (OR, 1.5; p = 0.025), peripheral vascular disease (OR, 1.7; p = 0.025), congestive heart failure (OR, 2.0; p = 0.014), chronic lung disease (OR, 1.8; p = 0.021), depression (OR, 2.4; p < 0.0001), and chronic anemia (OR, 2.2; p = 0.001) (Table III). Risk factors for readmission after inpatient total shoulder arthroplasty were similar, but more numerous, and included age of ≥80 years (Table IV). There were no significant differences (p > 0.05) in the reasons for readmission between the patients undergoing ambulatory total shoulder arthroplasty and the control group (Table V).
Overall, ambulatory total shoulder arthroplasty had significantly lower reimbursement costs (p < 0.05) compared with matched controls in numerous itemized cost categories as well as DRG-related reimbursed costs (Table VI). In terms of itemized reimbursements, ambulatory total shoulder arthroplasty had significantly lower PACU costs (p < 0.0001), laboratory costs (p < 0.0001), physical therapy and occupational therapy costs (p < 0.0001), and narcotic prescription costs (p < 0.0001) (Table VI). Anticoagulant prescription (p = 0.014) and antiemetic prescription costs (p = 0.001) were significantly higher for patients undergoing ambulatory total shoulder arthroplasty. The DRG-related costs were significantly lower (p < 0.0001) for the ambulatory total shoulder arthroplasty group ($14,722 per patient) compared with the matched non-ambulatory total shoulder arthroplasty group ($18,336 per patient).
One method that has been demonstrated to decrease cost in lower-extremity joint arthroplasty is to identify patients who are candidates for ambulatory arthroplasty, avoiding the need for an inpatient postoperative stay and the associated costs7. If this strategy is to be viable for other procedures such as total shoulder arthroplasty, it is critical to demonstrate that there are no increases in early readmission and postoperative complications when patients are discharged directly home postoperatively. To our knowledge, there is currently a scarcity of evidence-based literature that reflects the modern prevalence and efficacy of ambulatory total shoulder arthroplasty. In the present study, the incidence of ambulatory total shoulder arthroplasty doubled over a 4-year period. Although ambulatory total shoulder arthroplasty still represents a small fraction of the total shoulder arthroplasties that were performed during the same time period, it is a rapidly developing trend that warrants investigation.
Foremost, for ambulatory total shoulder arthroplasty to be an effective practice, it must be accomplished without compromising patient safety or clinical outcomes. In the present study, we demonstrated no significant increase in postoperative complications in patients undergoing ambulatory total shoulder arthroplasty compared with matched patients undergoing inpatient total shoulder arthroplasty. In fact, rates of certain hospital-related complications, including urinary tract infection and transfusion, were higher in patients admitted postoperatively.
The reported perioperative complication rates in our matched inpatient total shoulder arthroplasty cohort are similar to those in other large database studies in the current literature17,18,20,21. The reported incidence of urinary tract infection in the 30-day perioperative period following inpatient total shoulder arthroplasty is 1.4% to 2.0%17,22,23. The use of an indwelling catheter is the most well-known risk factor for urinary tract infection24. It is much more likely that patients who are admitted following total shoulder arthroplasty have indwelling catheters inserted in the perioperative period, which increases their risk of nosocomial urinary tract infection. Our reported transfusion rates of 3.5% for ambulatory patients and 4.7% for inpatients are similar to the reported rate of perioperative allogenic blood transfusion following inpatient total shoulder arthroplasty, which ranges from 0.6% to 4.5%18,21-23. However, it is important to note that patients admitted to the hospital following total shoulder arthroplasty may be more likely to have postoperative laboratory samples drawn and thus might undergo transfusion more often than their ambulatory counterparts. Venous thromboembolism was a rare complication within 90 days following both the ambulatory and inpatient total shoulder arthroplasties in this series, which is consistent with the broader inpatient shoulder arthroplasty literature, which reports an incidence of 0.04% to 1.2%17,18,20,22,23,25. The infection rates of 3.0% in our study cohort and 2.7% in our control cohort were substantially higher than the infection rates of 0.3% to 1.8% reported in the literature, perhaps because of the inclusion of both superficial and deep infection in the definition of infection, and a more comprehensive capture of infections across all postoperative care sites18,19,22,25. Finally, the revision rates were similar to that reported by Villacis et al., who noted a 2.6% rate within 1 year25.
To ensure patient safety and to create a sustainable model for ambulatory arthroplasty, it is essential to select patients who are appropriate for ambulatory total shoulder arthroplasty. Risk factors for readmission, reoperation, and postoperative complications have been described in the inpatient total shoulder arthroplasty population. We report readmission rates of 5.7% at 30 days and 9.3% at 90 days after ambulatory total shoulder arthroplasty, which was not significantly different from the matched controls (5.0% at 30 days and 9.0% at 90 days). This is similar to reported readmission rates in other series. Schairer et al. analyzed 14,602 patients from the state inpatient databases from California, Florida, Hawaii, North Carolina, Nebraska, New York, and Utah and reported a 90-day readmission rate of 6.0% following total shoulder arthroplasty19, which is comparable with other reported rates in the literature (4.5% to 6.7%)26,27.
Current studies have focused on identifying risk factors for readmission following inpatient total shoulder arthroplasty14-19. A recent study analyzing the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database showed that readmission within 30 days following inpatient total shoulder arthroplasty was associated with a history of heart disease and hypertension14. The present study suggests that patients with morbid obesity, diabetes mellitus, vascular disease, congestive heart failure, lung disease, and depression are at increased risk for complications requiring readmission after ambulatory total shoulder arthroplasty. We found similar risk factors in patients undergoing inpatient total shoulder arthroplasty, although age of ≥80 years was a risk factor for inpatient controls but not for patients undergoing ambulatory total shoulder arthroplasty. Additionally, preoperative anemia was a significant risk factor for readmission after ambulatory total shoulder arthroplasty but not for inpatient total shoulder arthroplasty, perhaps because relevant postoperative anemia was treated during the original hospital admission for the latter group. Reviewing the medical history and identifying comorbidities of a potential candidate for ambulatory total shoulder arthroplasty are essential, as a recent study by Schairer et al. reported that >82% of readmissions following inpatient total shoulder arthroplasty were related to medical complications19. We found a similar breakdown of readmission reasons, with approximately 80% of readmissions after both ambulatory and inpatient total shoulder arthroplasty being due to medical-related complications.
One of the most frequently discussed benefits of ambulatory surgical procedures is cost savings3. A recent systematic review of the cost implications of outpatient orthopaedic surgical procedures reported a mean cost savings of 17.6% to 57.6% for ambulatory procedures compared with inpatient procedures28. To our knowledge, cost analyses of ambulatory arthroplasty have been limited to lower-extremity arthroplasty3,6,7. Lovald et al. reported a cost savings of $8,527 per patient undergoing ambulatory total knee arthroplasty over a 2-year period compared with the standard length-of-stay group6. This financial analysis has been corroborated by several smaller institutional series demonstrating similar savings3,7. We report a significant reduction in both procedural and hospital-related reimbursements, which resulted in a mean cost reduction of $3,615 per patient in DRG reimbursements.
There were several important limitations of this study that warrant mention, many of which are consistent with other studies using large, administrative databases14,29. The validity of the analysis relies on the quality of the data and accuracy of coding within the database, making miscoding and noncoding by physicians potential sources of error. During the first 90 days after the surgical procedure, patients are within a global period, possibly increasing the instances in which other minor complications were not coded despite being documented in the medical record. There are several other factors that likely impact readmission rates after ambulatory total shoulder arthroplasty that are unable to be characterized within the database, including operative time, surgical approach, hospital volume, and anesthetic technique. We employed a comprehensive algorithm to confirm that all patients included in the ambulatory total shoulder arthroplasty cohort were actually discharged home on the day of the surgical procedure; however, we could not control for patients who were discharged home and then proceeded to be admitted on their own accord to a rehabilitation facility or nursing facility. We also could not control for patients who may have been admitted following planned ambulatory surgery because of concerns about medical stability. Additionally, although we matched the study and control groups on numerous demographic and comorbidity variables and used a multivariate regression analysis to control for additional factors that could not be controlled, it is possible that there remains some selection bias in selecting healthier patients for ambulatory surgical procedures. Finally, actual reimbursement for total shoulder arthroplasty is not always consistent and may vary on the basis of contracts between providers and facilities, which is a limitation for the cost analysis.
In conclusion, in appropriately selected patients, ambulatory total shoulder arthroplasty is a viable and safe practice model. Ambulatory total shoulder arthroplasty also offers significant cost savings compared with inpatient total shoulder arthroplasty in matched patients.
Tables showing ICD-9 codes used to define demographic characteristics and comorbidities and CPT and ICD-9 codes used to define postoperative complications are available with the online version of this article as a data supplement at jbjs.org (http://links.lww.com/JBJS/C252).
Investigation performed at the Department of Orthopaedic Surgery, University of Virginia Health System, Charlottesville, Virginia
Disclosure: There was no external funding obtained for this study. On the Disclosure of Potential Conflicts of Interest forms, which are provided with the online version of the article, one or more of the authors checked “yes” to indicate that the author had a relevant financial relationship in the biomedical arena outside the submitted work (http://links.lww.com/JBJS/C251).
1. Barrington JW. Fast-track recovery and outpatient joint arthroplasty. Am J Orthop (Belle Mead NJ). 2015 ;44(10)(Suppl):S21–2.
2. Husted H. Fast-track hip and knee arthroplasty: clinical and organizational aspects. Acta Orthop Suppl. 2012 ;83(346):1–39.
3. Bertin KC. Minimally invasive outpatient total hip arthroplasty: a financial analysis. Clin Orthop Relat Res. 2005 ;435:154–63.
4. Kolisek FR, McGrath MS, Jessup NM, Monesmith EA, Mont MA. Comparison of outpatient versus inpatient total knee arthroplasty. Clin Orthop Relat Res. 2009 ;467(6):1438–42. Epub 2009 Feb 18.
5. Lovald S, Ong K, Lau E, Joshi G, Kurtz S, Malkani A. Patient selection in outpatient and short-stay total knee arthroplasty. J Surg Orthop Adv. 2014 ;23(1):2–8.
6. Lovald ST, Ong KL, Malkani AL, Lau EC, Schmier JK, Kurtz SM, Manley MT. Complications, mortality, and costs for outpatient and short-stay total knee arthroplasty patients in comparison to standard-stay patients. J Arthroplasty. 2014 ;29(3):510–5. Epub 2013 Aug 21.
7. Aynardi M, Post Z, Ong A, Orozco F, Sukin DC. Outpatient surgery as a means of cost reduction in total hip arthroplasty: a case-control study. HSS J. 2014 ;10(3):252–5. Epub 2014 Jul 12.
8. Kim SH, Wise BL, Zhang Y, Szabo RM. Increasing incidence of shoulder arthroplasty in the United States. J Bone Joint Surg Am. 2011 ;93(24):2249–54.
9. 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 ;19(8):1115–20. Epub 2010 Jun 15.
10. Jain NB, Higgins LD, Guller U, Pietrobon R, Katz JN. Trends in the epidemiology of total shoulder arthroplasty in the United States from 1990-2000. Arthritis Rheum. 2006 ;55(4):591–7.
11. Ilfeld BM, Wright TW, Enneking FK, Mace JA, Shuster JJ, Spadoni EH, Chmielewski TL, Vandenborne K. Total shoulder arthroplasty as an outpatient procedure using ambulatory perineural local anesthetic infusion: a pilot feasibility study. Anesth Analg. 2005 ;101(5):1319–22.
12. McNair PD, Luft HS. Enhancing Medicare’s hospital-acquired conditions policy to encompass readmissions. Medicare Medicaid Res Rev. 2012 ;2(2):mmrr.002.02.a03.
13. Paxton EW, Inacio MCS, Singh JA, Love R, Bini SA, Namba RS. Are there modifiable risk factors for hospital readmission after total hip arthroplasty in a US healthcare system? Clin Orthop Relat Res. 2015 ;473(11):3446–55.
14. Basques BA, Gardner EC, Toy JO, Golinvaux NS, Bohl DD, Grauer JN. Length of stay and readmission after total shoulder arthroplasty: an analysis of 1505 cases. Am J Orthop (Belle Mead NJ). 2015 ;44(8):E268–71.
15. Matsen FA 3rd, Li N, Gao H, Yuan S, Russ SM, Sampson PD. Factors affecting length of stay, readmission, and revision after shoulder arthroplasty: a population-based study. J Bone Joint Surg Am. 2015 ;97(15):1255–63.
16. Streubel PN, Simone JP, Sperling JW, Cofield R. Thirty and ninety-day reoperation rates after shoulder arthroplasty. J Bone Joint Surg Am. 2014 ;96(3):e17.
17. Waterman BR, Dunn JC, Bader J, Urrea L, Schoenfeld AJ, Belmont PJ Jr. Thirty-day morbidity and mortality after elective total shoulder arthroplasty: patient-based and surgical risk factors. J Shoulder Elbow Surg. 2015 ;24(1):24–30. Epub 2014 Aug 29.
18. Farng E, Zingmond D, Krenek L, Soohoo NF. Factors predicting complication rates after primary shoulder arthroplasty. J Shoulder Elbow Surg. 2011 ;20(4):557–63. Epub 2011 Feb 16.
19. Schairer WW, Zhang AL, Feeley BT. Hospital readmissions after primary shoulder arthroplasty. J Shoulder Elbow Surg. 2014 ;23(9):1349–55. Epub 2014 Mar 4.
20. Jiang JJ, Toor AS, Shi LL, Koh JL. Analysis of perioperative complications in patients after total shoulder arthroplasty and reverse total shoulder arthroplasty. J Shoulder Elbow Surg. 2014 ;23(12):1852–9. Epub 2014 Aug 23.
21. Padegimas EM, Clyde CT, Zmistowski BM, Restrepo C, Williams GR, Namdari S. Risk factors for blood transfusion after shoulder arthroplasty. Bone Joint J. 2016 ;98-B(2):224–8.
22. Shields E, Iannuzzi JC, Thorsness R, Noyes K, Voloshin I. Perioperative complications after hemiarthroplasty and total shoulder arthroplasty are equivalent. J Shoulder Elbow Surg. 2014 ;23(10):1449–53. Epub 2014 Apr 18.
23. 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):2099–105. Epub 2014 Dec 19.
24. Redder JD, Leth RA, Møller JK. Analysing risk factors for urinary tract infection based on automated monitoring of hospital-acquired infection. J Hosp Infect. 2016 ;92(4):397–400. Epub 2015 Dec 30.
25. Villacis D, Sivasundaram L, Pannell WC, Heckmann N, Omid R, Hatch GFR 3rd. Complication rate and implant survival for reverse shoulder arthroplasty versus total shoulder arthroplasty: results during the initial 2 years. J Shoulder Elbow Surg. 2016 ;25(6):927–35. Epub 2016 Jan 18.
26. Lyman S, Jones EC, Bach PB, Peterson MGE, Marx RG. The association between hospital volume and total shoulder arthroplasty outcomes. Clin Orthop Relat Res. 2005 ;432:132–7.
27. Mahoney A, Bosco JA 3rd, Zuckerman JD. Readmission after shoulder arthroplasty. J Shoulder Elbow Surg. 2014 ;23(3):377–81. Epub 2013 Oct 14.
28. Crawford DC, Li CS, Sprague S, Bhandari M. Clinical and cost implications of inpatient versus outpatient orthopedic surgeries: a systematic review of the published literature. Orthop Rev (Pavia). 2015 ;7(4):6177.
29. Yang S, Werner BC, Cancienne JM, Hassanzadeh H, Shimer AL, Shen FH, Singla A. Preoperative epidural injections are associated with increased risk of infection after single-level lumbar decompression. Spine J. 2016 ;16(2):191–6. Epub 2015 Jul 30.