The demand for shoulder arthroplasty is growing at a rate higher than lower-extremity arthroplasty procedures.1–3 Given the growing demand for total shoulder arthroplasty (TSA), it has become increasingly important to focus on cost-effectiveness and efficiency. In addition, bundled payments have begun to take effect for total joint arthroplasty, which standardizes payments for 90-day episode-of-care costs.4,5 This system of bundled payments places greater financial responsibility on providers and incentivizes them to minimize costs without compromising outcomes or safety.
Traditionally, TSA has been performed as an inpatient procedure due to concerns regarding postoperative pain and intraoperative blood loss, as well as postoperative complications that closely follow surgery.6 However, recent improvements in surgical techniques and pain management have led to an overall decreased length of stay and a reduction in complications following TSA.7 With the use of novel methods for pain control and improved knowledge regarding appropriate patient selection, TSA has begun to be performed as an outpatient procedure, which has resulted in a substantial saving in overall costs.6,8,9 Although several reports suggest outpatient TSA may be performed safely in appropriately selected patients, this has yet to become commonplace, and further research is needed to understand the potential for increased complications and adverse events compared with inpatient TSA.10,11
Therefore, the purpose of this study was to examine the overall 30-day hospital readmission rates and surgical complications among patients who underwent outpatient and inpatient TSA using comorbidity-matched cohorts of patients identified in a national database. The authors hypothesized there would be no increase in 30-day hospital readmissions or surgical complications among patients who underwent outpatient TSA compared with inpatient TSA.
Materials and Methods
This study was deemed exempt from review by the authors' institutional review board. Data collected as part of the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database were used for this study. The data were prospectively collected at more than 700 hospitals across the United States and included patients undergoing “major” surgery across a variety of surgical subspecialties.12 Demographic characteristics were collected in addition to medical comorbidities and 30-day postoperative complications.
The NSQIP database included records of hospital admissions for surgical complications, with surgical complications being defined by the treating surgeon. Hospital readmissions were categorized as being either “related to surgical complications” or “not related to surgical complications.” Readmissions not related to surgical complications were largely for medical reasons that were too numerous to categorize. The NSQIP database is widely used in the orthopedic literature, and random audits by trained clinical reviewers have demonstrated less than 2% inter-rater disagreement.13,14
Patient Selection and Comorbidities
Patients who underwent primary TSA (Current Procedural Terminology codes 23472 and 23470) from 2010 to 2017 were identified. Inpatient surgeries were defined as having at least 1 overnight hospital stay, and outpatient surgeries were defined as any surgery with discharge on the same day. Patients who had sepsis on admission, prior surgery within 30 days of the index TSA, TSA performed emergently, wound infection on admission, cancer, or ascites were excluded from the study. These exclusion criteria were applied in an effort to identify complications and readmission related to TSA. Patients who had missing data also were excluded.
The American Society of Anesthesiologists (ASA) classification was used to quantify the comorbidity burden of patients. In addition, comorbidities with a frequency of 10 or greater in the propensity-matched cohort were included in the authors' descriptive analysis. These included smoking, diabetes mellitus, chronic obstructive pulmonary disease, congestive heart failure, and hypertension requiring medication.
Descriptive statistics were calculated for continuous variables and are reported as mean±SD unless stated otherwise. The authors used 1:1 propensity score matching (PSM) to create a control group of patients who underwent inpatient TSA with similar demographic characteristics and medical comorbidities compared with the outpatient TSA cohort. Demographic characteristics used in the matching algorithm included age and sex, and medical comorbidities included ASA classification, diabetes mellitus, smoking, hypertension requiring medication, chronic obstructive pulmonary disease, and congestive heart failure. On the basis of these parameters, the nearest-neighbor method was used to create matched groups based on the propensity score. An a priori power analysis showed this study had a power of 85% to detect a difference of 1% in 30-day readmission. Categorical variables are expressed as numbers and percentages. Significance was set at P<.05. Data were analyzed using Stata, version 15, software (StataCorp LLC).
Following cohort selection and 1:1 PSM, a total of 3428 patients were included in the study. This included 1714 patients who underwent inpatient TSA and 1714 patients who underwent outpatient TSA. Mean age was 64±12 years, and 51% were female. Patients had a mean body mass index (BMI) of 31±7.4 kg/m2, and the majority were ASA class 3. A total of 824 (24%) of the patients were smokers, 998 (29%) had a diagnosis of diabetes mellitus, 160 (4.7%) had chronic obstructive pulmonary disease, 19 (0.6%) had congestive heart failure, and 1940 (57%) had hypertension requiring medication (Table 1).
Patient Demographics (N=3428)
Patients who underwent outpatient surgery were 2.6 years younger on average (P<.01; 95% CI, 1.83–3.41) and were more likely to be female (54% inpatient vs 48% outpatient). Patients otherwise were similar regarding medical comorbidities included in the matching algorithm.
Comparison of 30-Day Readmission and Reoperation
Overall, 68 (2.5%) patients had a 30-day readmission of any kind following inpatient and outpatient TSA. Readmissions occurred at a mean of 15±9.1 days following surgery. Even after PSM for demographic factors and medical comorbidities, the proportion of patients with an all-cause readmission was significantly higher among those undergoing outpatient TSA vs inpatient TSA (3.4% vs 1.7%; P<.01). The proportion of patients with a readmission for a surgical complication, however, was not significantly higher among those undergoing outpatient TSA (1.9% vs 1.4%; P=.32). The percentage of patients with an unplanned return to the operating room was not significantly different between patients undergoing out-patient vs inpatient TSA (1.5% vs 0.9%; P=.11) (Table 2).
30-Day Readmission Rates and Unplanned Return to the Operating Room
Outpatient TSA has been increasingly discussed as an alternative to inpatient TSA due to the potential for cost savings by eliminating the need for inpatient hospital admission. The purpose of this study was to examine 30-day hospital readmissions and surgical complications following TSA performed on an outpatient basis vs an inpatient basis, using comorbidity-matched cohorts derived from a national database of surgical outcomes. Of the 3428 patients included in the study, there was a higher proportion of patients who had all-cause 30-day hospital admissions in the outpatient TSA cohort; however, there was no significant difference in the proportion of readmissions for surgical complications. In addition, an equal proportion of patients in both the inpatient and outpatient TSA groups had an unplanned return to the operating room.
The most notable finding of this study was the discrepancy between 30-day all-cause hospital readmissions vs 30-day hospital admissions for surgical complications. The increased proportion of all-cause admissions in the outpatient TSA group but equivalent proportion of admissions for surgical complications suggests the majority of the focus should be placed on identifying patients who are at low risk for medical issues postoperatively. When patients were matched by demographics and comorbidities, outpatient TSA did not increase the risk for complications related to the TSA surgery itself. In the future development of bundled payment models for TSA, it should be considered that although patients undergoing outpatient TSA may have elevated overall postoperative hospital admission rates, this is not necessarily resultant from surgical complications.
Several studies have examined outcomes following outpatient TSA and serve as useful comparisons with the current study. In a 2016 study by Brolin et al,15 propensity-matched cohorts of patients undergoing outpatient and inpatient TSA were studied, with 30 patients in each cohort. The authors concluded there was no increase in any complications or hospital readmission; however, no hospital admissions were reported in either group in this study.15
Given the overall low prevalence of hospital admission following TSA, larger cohorts such as the one used in the current study allow for a more detailed study of this important topic. The largest study of inpatient vs outpatient TSA to date was performed by Cancienne et al9 and used the PearlDiver administrative claims database. In their study, 706 patients who underwent outpatient TSA were matched to 4459 patients who underwent inpatient TSA, and the authors found no difference in overall 30- and 90-day hospital readmission rates.9 This is notably different from the current study, which found increased 30-day admission rates in the outpatient TSA group. The authors hypothesize this difference may be due to the increased power of the current study, which included more than twice the number of outpatient TSA compared with the study by Cancienne et al.9
It has been demonstrated previously that patient selection is of paramount importance when performing outpatient TSA. Recently, Fournier et al10 demonstrated that a patient-selection algorithm may be used to identify patients who are at low risk of perioperative complications and hospital admissions following out-patient TSA. Their criteria for outpatient TSA included age 70 years or younger; hematocrit greater than 30%; BMI of 35 kg/m2 or less; absence of chronic obstructive pulmonary disease, obstructive sleep apnea, and congestive heart failure or hypertension; and no history of deep venous thrombosis or pulmonary embolism. Even after matching cohorts in the current study for many of these comorbidities, it appears that patients undergoing outpatient TSA may have elevated all-cause postoperative hospital admissions even if the surgical complication rate is not significantly increased.
Several barriers exist to the more widespread practice of outpatient TSA, and the current study sheds light on several points that have been identified as potential concerns by surgeons. In a recent study by Brolin et al,6 surgeons were surveyed with the aim of identifying differences between surgeons performing and not performing outpatient TSA. Of 179 surgeons surveyed, only 21% performed outpatient TSA, and concern for medical complications was the primary reason that surgeons did not perform outpatient TSA.6 The current results support the idea that medical patients undergoing outpatient TSA have an increased chance of hospital admission for medical issues before 30 days and that this discrepancy persists after controlling for demographic and medical comorbidities. Thus, further understanding of the medical admissions that occur following outpatient TSA is necessary to help providers determine whether these admissions are avoidable through improved patient selection.
The results of this study should be viewed in the context of several limitations. Although the NSQIP database has been widely used in studying outcomes and complications following TSA,16,17 the rigidity of the data and anonymity of the patients limited the authors' ability to analyze patient-specific information on a more granular level. Furthermore, the categorization of hospital readmissions as being related to surgical complications involves a degree of judgment on the part of those entering data into the NSQIP registry, and specific orthopedic complications are not recorded in the NSQIP database. This study was strengthened by the large, national cohort of patients that could be analyzed over a long period. In addition, this study was powered to detect a small difference (1%) in admission rates.
Patients who underwent outpatient TSA had increased risk of all-cause 30-day hospital readmission compared with equally matched controls who underwent inpatient TSA, although readmission for surgical complications was equivalent between inpatient and outpatient TSA. Careful patient selection for outpatient TSA should be emphasized to minimize the potential for increased rates of postoperative hospital readmission.
- 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):1115–1120. doi:10.1016/j.jse.2010.02.009 [CrossRef]. PMID:20554454
- 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):591–597. doi:10.1002/art.22102 [CrossRef]. PMID:16874781
- Singh JA, Ramachandran R. Age-related differences in the use of total shoulder arthroplasty over time: use and outcomes. Bone Joint J. 2015;97-B(10):1385–1389. doi:10.1302/0301-620X.97B10.35696 [CrossRef] PMID:26430014
- Navathe AS, Troxel AB, Liao JM, et al. Cost of joint replacement using bundled payment models. JAMA Intern Med. 2017;177 (2):214–222. doi:10.1001/jamainternmed.2016.8263 [CrossRef]. PMID:28055062
- Curtin BM, Russell RD, Odum SM. Bundled payments for care improvement: boom or bust?J Arthroplasty. 2017;32(10):2931–2934. doi:10.1016/j.arth.2017.05.011 [CrossRef]. PMID:28583761
- Brolin TJ, Cox RM, Zmistowski BM, Namdari S, Williams GR, Abboud JA. Surgeons' experience and perceived barriers with outpatient shoulder arthroplasty. J Shoulder Elbow Surg. 2018;27(6S):S82–S87. doi:10.1016/j.jse.2018.01.018 [CrossRef]. PMID:29776473
- Dunn JC, Lanzi J, Kusnezov N, Bader J, Waterman BR, Belmont PJ Jr, . Predictors of length of stay after elective total shoulder arthroplasty in the United States. J Shoulder Elbow Surg. 2015;24(5):754–759. doi:10.1016/j.jse.2014.11.042 [CrossRef]. PMID:25591461
- Ilfeld BM, Wright TW, Enneking FK, et al. Total shoulder arthroplasty as an outpatient procedure using ambulatory perineural local anesthetic infusion: a pilot feasibility study. Anesth Analg. 2005;101(5):1319–1322. doi:10.1213/01.ANE.0000180199.52383.CE [CrossRef]. PMID:16243987
- Cancienne JM, Brockmeier SF, Gulotta LV, Dines DM, Werner BC. Ambulatory total shoulder arthroplasty: a comprehensive analysis of current trends, complications, re-admissions, and costs. J Bone Joint Surg Am. 2017;99(8):629–637. doi:10.2106/JBJS.16.00287 [CrossRef]. PMID:28419030
- Fournier MN, Brolin TJ, Azar FM, Stephens R, Throckmorton TW. Identifying appropriate candidates for ambulatory outpatient shoulder arthroplasty: validation of a patient selection algorithm. J Shoulder Elbow Surg. 2019;28(1):65–70. doi:10.1016/j.jse.2018.06.017 [CrossRef]. PMID:30100176
- Leroux TS, Zuke WA, Saltzman BM, et al. Safety and patient satisfaction of outpatient shoulder arthroplasty. JSES Open Access. 2018;2(1):13–17. doi:10.1016/j.jses.2017.11.002 [CrossRef]. PMID:30675561
- Molina CS, Thakore RV, Blumer A, Obremskey WT, Sethi MK. Use of the National Surgical Quality Improvement Program in orthopaedic surgery. Clin Orthop Relat Res. 2015;473(5):1574–1581. doi:10.1007/s11999-014-3597-7 [CrossRef]. PMID:24706043
- Trickey AW, Wright JM, Donovan J, et al. Interrater reliability of hospital readmission evaluations for surgical patients. Am J Med Qual. 2017;32(2):201–207. doi:10.1177/1062860615623854 [CrossRef]. PMID:26911664
- Davis CL, Pierce JR, Henderson W, et al. Assessment of the reliability of data collected for the Department of Veterans Affairs National Surgical Quality Improvement Program. J Am Coll Surg. 2007;204(4):550–560. doi:10.1016/j.jamcollsurg.2007.01.012 [CrossRef]. PMID:17382213
- Brolin TJ, Mulligan RP, Azar FM, Throckmorton TW. Neer Award 2016. Outpatient total shoulder arthroplasty in an ambulatory surgery center is a safe alternative to inpatient total shoulder arthroplasty in a hospital: a matched cohort study. J Shoulder Elbow Surg. 2017;26(2):204–208. doi:10.1016/j.jse.2016.07.011 [CrossRef]. PMID:27592373
- Lu Y, Khazi ZM, Patel BH, et al. Big data in total shoulder arthroplasty: an in-depth comparison of national outcomes databases. J Am Acad Orthop Surg. 2019. doi:10.5435/JAAOS-D-19-00173 [CrossRef] PMID:31663910
- Pugely AJ, Martin CT, Harwood J, Ong KL, Bozic KJ, Callaghan JJ. Database and registry research in orthopaedic surgery: Part I. Claims-based data. J Bone Joint Surg Am. 2015;97(15):1278–1287. doi:10.2106/JBJS.N.01260 [CrossRef]. PMID:26246263
Patient Demographics (N=3428)
|Characteristic||All patients||Inpatients (n=1714)||Outpatients (n=1714)|
|Age, mean±SD, y||64±12||66±11||63±12|
|Female, No.||1755 (51%)||830 (48%)||925 (54%)|
|BMI, mean±SD, kg/m2||31±7.4||30±6.7||31±7.1|
|ASA classification, No.|
| 1||144 (4.2%)||66 (3.8%)||78 (4.5%)|
| 2||1354 (39.5%)||860 (50.2%)||494 (28.8%)|
| 3||1754 (51.2%)||763 (44.5%)||991 (58%)|
| 4||176 (5.1%)||25 (1.5%)||151 (8.8%)|
|Smoker, No.||824 (24%)||342 (20%)||482 (28%)|
|DM, No.||998 (29%)||489 (29%)||509 (30%)|
|COPD, No.||160 (4.7%)||70 (4.1%)||90 (5.3%)|
|CHF, No.||19 (0.6%)||3 (0.2%)||16 (0.9%)|
|HTN requiring medication, No.||1940 (57%)||985 (57%)||955 (56%)|
30-Day Readmission Rates and Unplanned Return to the Operating Room
|30-day hospital readmission||68 (2.5%)||27 (1.7%)||41 (3.4%)||<.01|
|Hospital readmission for surgical complication||45 (1.6%)||22 (1.4%)||23 (1.9%)||.32|
|Unplanned return to operating room||40 (1.2%)||15 (0.9%)||25 (1.5%)||.11|