Inpatient vs outpatient arthroplasty: A in-state database analysis of 90-day complications

Abstract

Introduction

An increase in the number of policy initiatives, such as alternative payment models, have prompted healthcare providers to examine health-care expenditures while seeking to improve quality of care. Performing total joint arthroplasty (TJA) in the outpatient setting is an attractive option in driving costs down and providing psychological benefits to patients. Concerns regarding the safety and effectiveness of same-day discharge protocols warrants further investigation, especially on the state level. Due to the lack of consensus, we aimed to compare: (1) risk factors for outpatient arthroplasty and (2) incidences of postoperative complications between inpatient vs outpatient arthroplasty using an in-state database.

Methods

Patients who underwent total knee or hip arthroplasty between January 1, 2022 and December 31, 2022 were identified. Data was drawn from the Maryland State Inpatient Database (SID) and Maryland State Ambulatory Surgery and Services Database (SASD). A total of 7817 patients had TJA within this time. Patients were divided into inpatient arthroplasty (n = 1429) and outpatient arthroplasty (n = 6338). Demographic variables, medical comorbidities, and 90-day complication rates were compared between inpatient and outpatient procedures. Additional independent variables included: marital status, primary language, race, and median household income. A multivariate logistic regression analysis was performed to identify independent risk factors for complications following TJA after controlling for risk factors and patient comorbidities.

Results

Arthroplasty in the outpatient setting were more likely to be married (61.3% vs. 51.2%, p < 0.001), white (75.5% vs. 60.9%, <0.001), speak English as primary language (98.7% vs. 88.6%, p < 0.001), and have lower rates of diabetes (4.8% vs. 9.7%, p < 0.001), chronic obstructive pulmonary disease (16.3% vs. 21.8%, p < 0.001), and obesity (30.0% vs. 45.2%, p < 0.001) compared to arthroplasty in the inpatient setting, respectively. There were lower incidences of acute kidney injury (0.2 vs. 0.8%, p < 0.001) and infection (0.3% vs. 1.1%, p < 0.001) in the outpatient cohort compared to the inpatient cohort, respectively. Inpatient arthroplasty (Odds Ratio (OR) 1.98, 95% CI 1.30–3.02, p = 0.002) and hypertension (OR 2.12, 95% CI 1.23–3.64, p = 0.007) were independent risk factors for total complications following TJA.

Conclusion

Arthroplasty in the outpatient setting showed fewer complications than compared to patients in the inpatient setting. Although multiple factors should guide the decision for arthroplasty, outpatient arthroplasty may be a safe option for select, healthier patients without the increased burden of increased complications.

1. Introduction

Shifts in healthcare policy to incentivize high-quality care at a low-cost, has placed significant pressure on surgeons and hospitals to reduce the length of stay (LOS) associated with costly total joint arthroplasty (TJA) procedures.¹ This focus on quality and cost reduction has fostered remarkable advancements in surgical techniques and perioperative care, establishing outpatient arthroplasty as a viable option. These successful endeavors have even resulted in the removal of total knee arthroplasty (TKA) and total hip arthroplasty (THA) from the Centers for Medicare and Medicaid Services (CMS) inpatient-only (IPO) list.^2,3

Amidst the evolution of outpatient arthroplasty, concerns have been raised regarding its safety and suitability for these procedures. Multiple reports have described favorable outcomes of outpatient TJA, with decreased or no difference in postoperative complications compared to inpatient,4, 5, 6, 7, 8, 9, 10, 11 while others found higher rates of complications.¹² Efforts to identify patient factors that determine eligibility for these procedures have produced mixed results. However, consistent observations indicate that younger and healthier patients tend to experience better outcomes.^13,14 It should be noted that these studies largely utilized data from national databases containing data limited to a 30-day follow up period, or data from a single surgeon or institution, which place limitations on the power of the findings.

Given the increasing number of TJA performed in the outpatient setting, it is crucial to ensure surgical outcomes are not compromised in the pursuit of reducing healthcare expenditures. Our objective was to investigate the impact of Maryland's Global Budget Revenue (GBR) model on outpatient arthroplasty. As such, we aimed to compare: (1) risk factors for outpatient arthroplasty and (2) incidences of postoperative complications between inpatient vs outpatient arthroplasty using an in-state database.

2. Methods

2.1. Patient selection

Patients who underwent TJA between January 1, 2022 and December 31, 2022 were identified. Data was drawn from the Maryland State Inpatient Database (SID) and the Maryland State Ambulatory Surgery and Services Database (SASD). A total of 7817 patients had TJA within this time period. TJA included TKA and THA. Patients were divided into inpatient arthroplasty (n = 1429) and outpatient arthroplasty (n = 6338). Patients who had a length of stay of 0 days that were discharged on the same day of surgery to ambulatory surgery centers were considered outpatient procedures. Patients who had a length of stay greater than 0 days and stayed at least one night at the hospital were considered inpatient procedures. Institutional review board approval was waived for this public database study.

2.2. Variables

Demographic characteristics and medical comorbidities were compared between the inpatient and outpatient cohorts. Demographic variables included were age, sex, marital status, median household income, primary language, and race was collected. Medical comorbidities included in, including AIDS, alcohol abuse, dementia, diabetes mellitus, diabetes uncomplicated, drug abuse, hypertension, hypertension uncomplicated, COPD, obesity, hypothyroidism, and other thyroid disease were assessed. The 90-day complication rates were compared between inpatient and outpatient cohorts. Total complications included were acute kidney injury (AKI), venous thromboembolism (VTE), aseptic loosening, arthrofibrosis, dislocation/instability, infection, mechanical complication, pain, and periprosthetic fracture (PPFx). Total surgical complications were also compared for each cohort.

2.3. Statistical analysis

Patients were characterized using descriptive statistics in the study. Multiple linear regression models evaluated independent risk factors total complications following TJA after controlling for risk factors and patient comorbidities. All analyses utilized SPSS v.25 (IBM, USA). A p-value of <0.05 was set as significant.

3. Results

3.1. Patient characteristics and comorbidities

Arthroplasty in the outpatient setting were more likely to be married (61.3% vs. 51.2%, p < 0.001), white (75.5% vs. 60.9%, <0.001), speak English as primary language (98.7% vs. 88.6%, p < 0.001), and have lower rates of diabetes (4.8% vs. 9.7%, p < 0.001), chronic obstructive pulmonary disease (16.3% vs. 21.8%, p < 0.001), and obesity (30.0% vs. 45.2%, p < 0.001) compared to arthroplasty in the inpatient setting, respectively. A complete list of patient characteristics and comorbidities can be found in Table 1.

3.2. Postoperative complications

Rate of total complications was lower for the outpatient cohort (1.2% vs. 3.2%, p < 0.001). There were lower incidences of acute kidney injury (0.2 vs. 0.8%, p < 0.001) and infection (0.3% vs. 1.1%, p < 0.001) in the outpatient cohort compared to the inpatient cohort, respectively. After multivariate regression, inpatient arthroplasty (Odds Ratio (OR) 1.98, 95% CI 1.30–3.02, p = 0.002) and hypertension (OR 2.12, 95% CI 1.23–3.64, p = 0.007) were independent risk factors for total complications following TJA.

4. Discussion

The focus on value-based care in healthcare policy has placed substantial pressure on surgeons and hospitals to decrease the cost burden associated with TJA procedures. Significant advances in surgical technique and perioperative care have allowed for the transition of TJA to the outpatient setting and the subsequent removal of TKA and THA from the CMS IPO list. As the utilization of outpatient TJA continues to grow, we sought to compare the risk factors and incidences of postoperative complications between inpatient and outpatient TJA. To achieve this, we analyzed comprehensive state-wide data from Maryland, which implements a unique GBR model aimed at addressing costs, access, and quality of care. The major findings include (1) patients undergoing outpatient arthroplasty have lower rates of diabetes, COPD, and obesity compared to inpatient setting; (2) there were lower incidence of infection in the outpatient cohort compared to the inpatient cohort; (3) inpatient TJA was an independent risk factor for total complications.

We acknowledge potential limitations of this study. First, it is subject to those inherent limitations associated with the retrospective design. The number of procedures performed in the year observed were possibly affected by the limitations placed on elective procedures and promotion of outpatient arthroplasty due to the SARS-CoV-2 (COVID-19) pandemic.¹⁵ Furthermore, the indications for outpatient TJA were broadened to include older, less healthy patients during this time.¹⁶ Additionally, the follow up was limited to a 90-day period after the index surgery therefore a longer follow up would be necessary to assess longer term outcomes. The generalizability of our findings is limited due to use of a statewide database, which although ensures capture of all TJA procedures performed & may demonstrate the efficacy of the GBR model in Maryland, may not be representative of protocols and patient selection nationally. There is the possibility that there were additional factors that were unable to be accounted for with the dataset utilized that would influence outcomes. In addition to the large sample size, a strength of our study is the inclusion of multiple hospitals, surgical centers, and surgeons, which vary in their surgical techniques and perioperative protocols, and may compensate for some of the limitations in generalizability.

Outpatient TJA has demonstrated safety in the appropriately selected patients, making it critical to determine the specific factors currently associated with patients selected to undergo TJA in the outpatient setting. A common trend noted in the literature is those who are younger, have lower BMIs, and fewer comorbidities are more likely to undergo outpatient TJA compared to TJA in the inpatient setting.6, 13, 15, 17, 18 Our findings generally supported this notion, revealing significantly lower rates of diabetes, COPD, and obesity in those undergoing outpatient TJA, compared to their inpatient cohort. However, contrasting findings exist in the literature. For instance, a meta-analysis encompassing 177,792 TJ A patients from seven manuscripts found no significant differences in age or BMI between outpatient and inpatient TJA.¹¹ Additionally, a 2017 study found there to be no significant differences in rates of diabetes between cohorts.¹⁰

In a single-institution study encompassing elective, primary total joint arthroplasty (TJA) procedures conducted between June 2011 and May 2021, it was observed that patients undergoing outpatient TJA were more likely to have ASA class II or lower and were nonsmokers, in contrast to those necessitating a longer length of stay (LOS). However, it is crucial to acknowledge that the study institution had implemented explicit patient inclusion criteria and established protocols, which resulted in the exclusion of patients with specific comorbidities or limited social support from being eligible for outpatient TJA. Unfortunately, these factors could not be accounted for in the analysis.¹⁹ Outpatient arthroplasty was commonly associated with additional factors such as the absence of chronic steroid use, lower likelihood of functional dependence, absence of bleeding disorders or coagulopathy, absence of chronic dyspnea, absence of anemia, and albumin levels above 3.5 g/dL. However, it is important to note that variability exists in the literature regarding these associations.13, 18, 20, 21, 22, 23, 24 While it is clear surgeons are already implementing criteria to determine eligibility for outpatient TJA, it is difficult to draw specific conclusions. A possible explanation for the variation in findings is the heterogeneity in how “outpatient” is defined amongst the studies.

Due to ongoing safety concerns, extensive efforts have been dedicated to assessing and comparing postoperative complications between inpatient and outpatient arthroplasty. Analyzing TJAs performed between 2011 and 2013, Lovecchio et al. found that after propensity score matching, outpatients were 3 times more likely to experience medical complications within the initial 30 days following discharge. It is important to consider that some of these differences could be attributed to the utilization of a “fast-track” inpatient group for comparison purposes and the exclusion of patients with a length of stay exceeding two days.¹² Additionally, Burnett et al. demonstrated rates of any complication to be declining for inpatient and outpatient TKA from 2010 to 2020, with a greater decline in the outpatient (inpatient TKA -26% vs outpatient TKA -43%). In fact, after controlling comorbidities, outpatient TKA was independently associated with lower rates of any complication between 2018 and 2020, which was not observed between 2010 and 2012.²⁵ This points to the possibility that earlier outpatient cohorts may not be representative of current outcomes for both TKA and THA due to advances in perioperative care.^2,3 In our current study, outpatient arthroplasty demonstrated lower infection rates compared to the inpatient cohort, aligning with the well-established association between inpatient hospital stays and infection risk [28, 29, 30]. Furthermore, we identified inpatient arthroplasty as an independent risk factor for overall complications. Further supporting our findings, a meta-analysis of 20 studies involving a total of 582,790 TK A and THA patients revealed that outpatient procedures were associated with fewer adverse events, including total complications and THA readmissions.¹⁰ Many studies have consistently demonstrated similar findings of lower or no differences in infection risk between inpatient and outpatient settings,^6,8,10 with the overall complication rate reflecting this trend.5, 26, 27, 28, 29, 30, 31 The findings tentatively suggest that outpatient arthroplasty might demonstrate a comparable safety profile to that of inpatient procedures in carefully selected patients.

The outpatient approach demonstrated fewer complications when compared to inpatient arthroplasty, with inpatient status being identified as an independent risk factor for complications. These findings support the notion that outpatient arthroplasty can be safely performed on select healthier patients, without incurring the burden of increased complications. These studies will aid in clarifying any potential differences in risk associated with outpatient arthroplasty and will contribute to the refinement of current practices in this field.