Outpatient Knee/Hip Arthroplasty Yields Fewer Adverse Events
The National Surgical Quality Improvement Program (NSQIP) database contains more than a half-million records of patients who received a total knee arthroplasty (TKA), unicompartmental knee arthroplasty (UKA), or total hip arthroplasty (THA) from 2009 through 2018. Fewer than 4% of those procedures were done in an outpatient setting, but patient demand for outpatient arthroplasty is rising rapidly.
With retrospective data like that from NSQIP, the most meaningful comparisons between inpatient and outpatient procedures come through a propensity score-matched analysis. Propensity score matching pairs up patients in each group according to multiple factors thought to influence outcome. In a recent study in The Journal of Bone & Joint Surgery, Lan et al. used propensity score matching to compare inpatient and outpatient arthroplasty in terms of adverse events and readmissions.
What the Researchers Did:
- Matched each outpatient case of TKA, UKA, and THA from the database with 4 unique inpatient cases based on age, sex, ASA class, race, BMI, type of anesthesia, and history of hypertension, smoking, congestive heart failure, and diabetes
- Compared inpatient vs outpatient rates of 30-day adverse events (both minor and severe) and readmissions
- Identified risk factors for adverse events and readmissions
What the Researchers Found:
- For all 3 arthroplasty types, patients who underwent an outpatient procedure were less likely to experience any adverse event, when compared with those who underwent an inpatient procedure.
- The above adverse-event findings held true when TKAs, UKAs, and THAs were analyzed separately.
- Outpatient procedure status was an independent protective factor against the risk of adverse events.
- For all 3 procedures, readmission rates were similar among inpatients and outpatients. (The 2 most common reasons for readmission were infections and thromboembolic events.)
- Clinicians are probably (and reasonably) selecting healthier patients to undergo outpatient procedures, but 42% of the outpatient cohort had an ASA class ≥3, and 55% had a BMI ≥30 kg/m2.
In their abstract, the authors cited “increased case throughput” as one rationale for outpatient arthroplasty, but this study provides convincing evidence that adverse-event reduction is another compelling reason for certain patients to consider outpatient knee and hip procedures.
What’s New in Adult Reconstructive Knee Surgery 2021
Every month, JBJS publishes a review of the most pertinent and impactful studies published in the orthopaedic literature during the previous year in 13 subspecialties. Click here for a collection of all such OrthoBuzz specialty-update summaries.
This month, author Michael J. Taunton, MD summarizes the 5 most compelling findings from the 130 studies highlighted in the January 20, 2021 “What’s New in Adult Reconstructive Knee Surgery.”
Waiting for a Knee Replacement
–Patient wait times for joint arthroplasty, exacerbated in many places by the COVID-19 pandemic, continue to increase. As measured by the EQ-5D, the health among 12% of 2,168 patients awaiting total knee arthroplasty (TKA) in a recent cross-sectional analysis was rated as “worse than death.”1 Joint-specific function and various comorbidities were associated with these findings.
UKA vs TKA
–The multicenter randomized TOPKAT trial2 compared unicompartmental knee arthroplasty (UKA) with TKA for treating medial compartment osteoarthritis. At the 5-year follow-up, there was no between-group difference in Oxford knee scores, but UKA was more cost-effective and provided an additional 0.24 quality-adjusted life year.
Perioperative Patient Optimization
–An observational study analyzing >1,000 total joint arthroplasties3 found that implementing a “perioperative orthopaedic surgical home”—a surgeon-led screening and optimization initiative targeting 8 common modifiable comorbidities—resulted in a 1.6% 30-day readmission rate (versus 5.3% among patients not involved in the initiative).
Pain Management and Opioids
–A randomized controlled trial of >300 patients undergoing primary total knee or hip arthroplasty4 demonstrated that reducing the number of 5-mg oxycodone pills prescribed at discharge from 90 to 30 resulted in the following findings 30 days postoperatively:
- Similar between-group pain scores
- No between-group differences in patient-reported outcomes
- Significant reductions in unused opioid pills and in pain pills taken in the 30-pill group
Periprosthetic Joint Infection
–Patients undergoing primary TKA who had a history of periprosthetic joint infection (PJI) in another joint had a significantly higher risk of PJI after the primary TKA, compared with the risk among a matched cohort with no history of PJI.5
References
- Scott CEH, MacDonald DJ, Howie CR. ‘Worse than death’ and waiting for a joint arthroplasty. Bone Joint J.2019 Aug;101-B(8):941-50.
- Beard DJ, Davies LJ, Cook JA, MacLennan G, Price A, Kent S, Hudson J, Carr A, Leal J, Campbell H, Fitzpatrick R, Arden N, Murray D, Campbell MK; TOPKAT Study Group. The clinical and cost-effectiveness of total versus partial knee replacement in patients with medial compartment osteoarthritis (TOPKAT): 5-year outcomes of a randomised controlled trial. 2019 Aug 31;394(10200):746-56. Epub 2019 Jul 17.
- Kim KY, Anoushiravani AA, Chen KK, Li R, Bosco JA, Slover JD, Iorio R. Perioperative orthopedic surgical home: optimizing total joint arthroplasty candidates and preventing readmission. J Arthroplasty.2019 Jul;34(7S):S91-6. Epub 2019 Jan 18.
- Hannon CP, Calkins TE, Li J, Culvern C, Darrith B, Nam D, Gerlinger TL, Buvanendran A, Della Valle CJ. The James A. Rand Young Investigator’s Award: large opioid prescriptions are unnecessary after total joint arthroplasty: a randomized controlled trial. J Arthroplasty.2019 Jul;34(7S):S4-10. Epub 2019 Feb 4.
- Chalmers BP, Weston JT, Osmon DR, Hanssen AD, Berry DJ, Abdel MP. Prior hip or knee prosthetic joint infection in another joint increases risk three-fold of prosthetic joint infection after primary total knee arthroplasty: a matched control study. Bone Joint J.2019 Jul;101-B(7_Supple_C):91-7.
When Does I&D Beat 2-Stage Exchange in Second rTKA?
It’s hard to contemplate “conservative treatment” in the case of a revised total knee arthroplasty (rTKA) with extensive instrumentation that needs a reoperation due to periprosthetic joint infection (PJI), because all the treatment options in that scenario are pretty complex. In the January 6, 2021 issue of JBJS, Barry et al. report on a retrospective review of 87 revisions of extensively instrumented rTKAs that found that irrigation and debridement (I&D) with chronic antibiotic suppression was as effective as 2-stage exchange in preventing another reoperation for infection—and more effective in terms of maintaining knee function.
The average follow-up of the cases studied was 3.2 years, and the authors carefully defined “extensive instrumentation.” Among the 56 patients who were managed with I&D and suppression and the 31 who were managed with the initiation of 2-stage exchange (average age in both groups approximately 67 years), no significant differences were found in the rates of reoperation for infection or mortality. However, 9 of the 31 patients (29%) in the 2-stage group never underwent the second-stage reimplantation. Among those 9, 3 died prior to reimplantation and 2 underwent amputation due to failure of infection control.
Moreover, at the time of the latest follow-up, a significantly higher percentage of patients in the I&D group were ambulatory (76.8% vs 54.8% in the 2-stage group) and were able to functionally bend their knee (85.7% vs 45.2% in the 2-stage group). The authors surmise that these 2 findings are related to the soft-tissue damage and bone loss that typically occur during stage-1 removal of rTKA components.
Barry et al. conclude that in similar situations “deviating from the so-called gold standard of 2-stage exchange and accepting the modest results of I&D may be in the best interest of the patient,” as long as there are no loose implants in the existing construct. But the “sobering” mortality rates in the study (39.3% in the I&D group and 38.7% in the 2-stage group) remind us that this clinical scenario is extremely challenging for patients and surgeons, no matter which option is selected.
Click here to view an “Author Insights” video about this study with co-author Jeffrey Barry, MD.
Same-Patient Trial Design Measures Incremental Improvements in TKA Outcomes
The clinical and functional outcomes after total knee arthroplasty (TKA) are generally very favorable. The 15% to 20% of subpar patient-reported outcomes are usually related to persistent pain. Orthopaedic researchers have exhaustively investigated patient factors and technical considerations to address dissatisfaction in this minority population of TKA patients.
Meanwhile, the orthopaedic community has focused on prosthetic design in its attempts to incrementally improve outcomes for the 80% to 85% of generally satisfied TKA patients. Clearly documenting those incremental improvements often requires elegant study design. That’s what we see in the January 6, 2021 issue of The Journal, where Kim et al. report findings from a randomized trial in which 2 different knee-implant designs were compared in the same patients after primary simultaneous bilateral TKA.
Each of the 50 patients (49 of them women) received a posterior-stabilized design in 1 knee and an ultracongruent prosthesis in the other. Kim et al. selected the Forgotten Joint Score (FJS) as the primary outcome. The FJS is a 12-item questionnaire that assesses patient awareness of the artificial joint during daily activities. At 2 years, the researchers found no between-knee differences in FJS. The ultracongruent knees showed more anteroposterior laxity and less femoral rollback than the posterior-stabilized knees, but there were, again, no between-group differences in following measures:
- Range of motion
- Knee Society and WOMAC scores
- Side Preference and patient satisfaction
The ultracongruent advancement in prosthetic design does not appear to offer clinically important advantages over the posterior-stabilized design. But if additional TKA patients can be recruited into studies using clever and effective experimental designs like this one, the future is bright for more robust assessments of the incremental impact of prosthetic design on functional and clinical outcomes.
Click here to view an Infographic summarizing this study.
Marc Swiontkowski, MD
JBJS Editor-in-Chief
A Deeper Dive into Diabetes and Total Joint Replacement
There are many more “types” of diabetes than the pathophysiologic designations of Type 1 and Type 2. In the December 16, 2020 issue of The Journal of Bone & Joint Surgery, Na et al. delineate 4 different diabetes categories and determine their impact on 90-day complications and readmission rates after elective total joint arthroplasty (TJA) among Medicare patients. One premise for this investigation was that, although diabetes is a known risk factor for arthroplasty complications, alternative payment models such as the federally run Comprehensive Care for Joint Replacement (CJR) program adjust their payments only in diabetes cases where the comorbidity is coded as severe.
The authors stratified diabetes into 4 groups as follows:
- No diabetes
- Controlled-uncomplicated diabetes
- Controlled-complicated diabetes
- Uncontrolled diabetes
Among the >500,000 total knee arthroplasties (TKAs) and total hip arthroplasties (THAs) analyzed, the authors found the following when comparing data from the 3 diabetes groups with the no-diabetes group:
- The odds of TKA complications were significantly higher for those with uncontrolled diabetes (odds ratio [OR] = 1.29).
- The odds of THA complications were significantly higher for those with controlled-complicated diabetes (OR = 1.45).
- The odds of readmission were significantly higher in all diabetes groups for both TKA (ORs = 1.21 to 1.48) and THA (ORs = 1.20 to 1.70).
The authors come to 3 basic conclusions based on these findings:
- The odds of hospital readmission and complications following an elective TKA or THA are increased for Medicare beneficiaries who have diabetes.
- It would be reasonable to defer arthroplasty surgery for those with uncontrolled diabetes to allow them to achieve glycemic control.
- The Centers for Medicare & Medicaid Services should include less-severe diabetes and associated systemic complications in alternative-payment model adjustments.
Click here for an “Author Insight” video about this study from co-author Annalisa Na, PhD, DPT.
RCT: “Mini-Navigation” Improves TKA Component Alignment
Alignment is one of the most important factors affecting the longevity of total knee arthroplasty (TKA) prostheses. Among the many available tools for attaining good alignment are new palm-sized, accelerometer-based navigation systems. In the November 18, 2020 issue of The Journal of Bone & Joint Surgery, Minoda et al. deliver robust evidence from a multicenter randomized controlled trial comparing post-TKA coronal alignment with such a “mini-navigation” system versus results using conventional navigation.
The 90 TKAs analyzed (45 in each group) were performed by 8 different surgeons across 5 hospitals in Japan. The primary outcome was the number of tibial-prosthesis alignment outliers, defined as alignment error of >2° relative to neutral, based on weight-bearing radiographs 6 months after surgery. Researchers also compared absolute differences from neutral alignment for both tibial and femoral components, operative time, and complications.
Relative to conventional navigation, the mini-navigation system reduced the number of outliers in tibial alignment and hip-knee-ankle (HKA) angle. In addition, the absolute differences in the coronal alignment of the femoral and tibial prostheses were less in the mini-navigation group. Researchers found no increase in operative time with the portable navigation system, nor was there any observed increase in perioperative complications compared with conventional navigation.
Conversely, there were no significant between-group differences in postoperative Knee Society scores, but the authors note that “another study design with a longer follow-up period” is needed to evaluate the patient-centered clinical effects of prosthetic alignment using portable navigation. Both the authors and Andrew P. Kurmis, FRACS(Ortho), who commented on the study, assert that the true value of navigation systems (mini and full-sized alike) probably lies in higher reproducibility of tibial-component alignment relative to conventional navigation, regardless of the skill and experience of the surgeon.
Antibiotic-Laden Cement Lowers TKA-Revision Rates in US Veterans
We recently celebrated Veteran’s Day with the annual tradition of rightfully honoring the men and women who have served in the Armed Forces. After their active duty ends, servicemembers are eligible for care in Veterans Health Administration (VHA) hospitals around the nation. The VHA is a “closed” medical system that affords ample opportunity for population-based research.
In the November 18, 2020 issue of The Journal, Bendich et al. utilized VHA data to compare revision rates after primary total knee arthroplasty (TKA) among veterans treated with antibiotic-laden bone cement (ALBC) or plain cement. Although results of similarly designed studies focused on this question have been equivocal, antibiotic-laden cement seems to be especially effective at preventing infection in higher-risk populations, which is what the US veteran population is considered to be.
The researchers identified 15,972 primary TKAs that were implanted using Palacos bone cement between 2007 and 2015. Approximately 70% (11,231) of those cases used cement mixed with gentamicin, while 30% (4,741) utilized plain bone cement. The authors found similar patient demographics among patients treated with ALBC and those treated with plain cement, but ALBC was used more frequently in patients with higher comorbidity scores.
Overall, utilization of ALBC increased from 50.6% of the cases in 2007 to 69.4% in 2015. At a follow-up of 5 years, ALBC TKAs had a lower all-cause revision rate (5.3%) than plain-cement TKAs (6.7%) and a lower rate of revision for infection (1.9% compared to 2.6%). Even after multivariable adjustments to account for patient, surgical, and hospital factors, these revision-rate differences remained.
Bendich et al. also found that 71 TKAs needed to be implanted with ALBC to avoid 1 revision TKA. With a cost differential of $240 per case for ALBC, I think spending $17,040 ($240 × 71) is more cost-effective than 1 revision TKA, although a formal cost analysis is warranted.
In the interest of full disclosure, as an active-duty US Air Force officer, I am inherently biased, but I feel that no cost is too great to improve the health of our veterans. The authors review arguments against using ALBC, such as a theoretical risk of poor cement mechanical properties and systemic toxicity, but the findings of this study suggest that cement with antibiotics enhances treatment outcomes among these US heroes.
Click here to view the “Author Insight” interview about this study with co-author Alfred Kuo, MD, PhD.
Matthew R. Schmitz, MD
JBJS Deputy Editor for Social Media
Dashboard Depicts Surgeon-Level Value of TKA/THA
Remember when a “dashboard” referred to the display just behind a car’s steering wheel? In today’s digital universe, the word has come to mean any number of visual information displays. At the same time, the meaning of the word “value” has narrowed somewhat. In relation to health care, “value” is defined quite precisely as the quality of patient outcomes per dollar spent on healthcare services.
In the November 4, 2020 issue of The Journal of Bone & Joint Surgery, Reilly et al. explain how they created a “value dashboard” for total hip and knee arthroplasty (THA and TKA) at a tertiary-care medical center in New England. The goal: track and display the surgeon-level cost and quality of these procedures against institutional benchmarks to identify opportunities for improving value.
The 7 quality metrics that Reilly et al. used included both clinical and patient-reported outcomes, weighted by surgeons using a modified Delphi process. Average direct costs per surgeon were calculated from the medical center’s billing system, and data were collected over a 15-month period from 2017 to 2018 to ensure at least 1 year of outcomes. Six surgeons were included in the TKA value dashboard, and 5 were included in the THA dashboard.
Relative to the institutional benchmarks:
- Value for TKA by surgeon ranged from 7% below benchmark to 12% above.
- Value for THA by surgeon ranged from 12% below benchmark to 7% above.
The dashboard itself (see Figure above) displays quality, cost, and overall value so viewers can see at a glance which metrics are driving the value score for each surgeon, whose procedural volume is also depicted. The authors cite as one limitation of this study the fact that the quality metrics were weighted by local surgeons only, and they say that “ideally the weighting would be informed by a panel of national experts and several stakeholder groups,… including patients.”
Time Waits For No One—Aging Increases Costs
The cost of medical care in the United States has been shown to rise with advancing patient age, and total joint arthroplasty (TJA) is a prime example of this unsurprising phenomenon. In attempts to curtail costs and reduce variability, Medicare and other payers have introduced alternative payment models (APMs), such as the Bundled Payments for Care Improvement (BPCI) initiative. In this model’s application to TJA, when participating institutions keep the cost of the “episode” below a risk-adjusted target price, they accrue the savings as a profit, but they sustain a financial penalty if the episode costs more than the target price.
Multiple studies have suggested that APMs can negatively affect the fiscal health of institutions that care for many high-risk patients. Although increasing age has been associated with higher-cost episodes of care, age is not one of the factors that the BPCI model accounts for. Consequently, concerns have been raised that providers may practice “cost discrimination” against very old patients.
In the October 7, 2020 issue of The Journal, Petersen et al. examine how an aging population has affected a New York City orthopaedic center in terms of the BPCI model applied to TJA. The authors analyzed the relationship between patient age and cost of care among 1,662 patients who underwent primary total hip and knee arthroplasty over a 3-year period under BPCI. They then used a modeling tool to predict shifting age demographics for their local area out to the year 2040.
Petersen et al. found that under BPCI, their institution sustained a nearly $2,000-per-case loss for TJA care episodes among patients 85 to 99 years of age. Currently this loss is offset by profits realized by performing TJAs in younger patients. However, predictive modeling identified an inflection point of 2030, after which a relative increase in older patients and a decrease in younger patients will yield an overall net decrease in profits for primary TJA.
Because no one, including orthopaedic surgeons, can turn back the clock on aging, health care stakeholders must find ways either to adjust downward the cost of care for the elderly (seemingly difficult without adversely affecting outcomes) or adjust reimbursement models to account for the increased costs associated with aging. I agree with the conclusion of Petersen et al.: “The BPCI initiative and [other] novel APMs should consider age as a modifier for reimbursement to incentivize care for the more vulnerable and costly age groups in the future.”
Matthew R. Schmitz, MD
JBJS Deputy Editor for Social Media
Cost-Effectiveness of PRP for Knee OA Questioned
The cost-effectiveness analysis of platelet-rich plasma (PRP) for knee osteoarthritis by Rajan et al. in the September 16, 2020 issue of JBJS is accompanied by 105 references. That’s just one indication of the level of interest in this anti-inflammatory and pro-angiogenic orthobiologic. Current literature suggests that PRP is safe, but its clinical efficacy in musculoskeletal conditions has been hotly debated in the orthopaedic community.
Rajan et al. applied Markov decision analysis to a clinical scenario in which a 55-year-old patient with Kellgren-Lawrence grade-II or III knee osteoarthritis (OA) undergoes either a series of 3 PRP injections and a 1-year delay to total knee arthroplasty (TKA), or TKA from the outset. Their primary outcome measures were total costs and quality-adjusted life years (QALYs), organized into incremental cost-effectiveness ratios (ICERs). In Markov analyses, if one treatment costs less and produces more QALYs than a comparative treatment, it is considered to be the “dominant” approach.
The authors found that, from a health-care payer perspective, PRP (at an estimated cost of $728 per injection in 2018 US dollars) was not cost-effective if it yielded only a 1-year delay of TKA. However, from a societal perspective (which considered both lost productivity and the need for unpaid caregiving associated with TKA surgery), PRP was cheaper over a lifetime because it delayed direct and indirect costs associated with TKA. The ICER for TKA at the outset was $4,175 per QALY, which is well below the predetermined willingness-to-pay threshold of $50,000. The authors emphasize that this favorable ICER reflects the improved quality of life after TKA compared with published results of PRP injections for knee OA.
Rajan et al. do specify a clinical scenario in which PRP may have a cost-effectiveness advantage over TKA: “…in a higher-risk patient population in whom the perioperative complication rates, TKA revision rate, or postoperative functional outcomes are projected to be worse.”
Try Clinical Classroom
Recent Comments