The Journal of Arthroplasty 30 (2015) 1927–1930
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Perioperative Outcomes Following Unilateral Versus Bilateral Total Knee Arthroplasty Linda I. Suleiman, MD a, Adam I. Edelstein, MD a, Rachel M. Thompson, MD a, Hasham M. Alvi, MD a, Mary J. Kwasny, ScD b, David W. Manning, MD a a b
Department of Orthopaedic Surgery, Northwestern University, Feinberg School of Medicine, Chicago, Illinois Department of Preventative Medicine, Division of Biostatistics, Northwestern University, Feinberg School of Medicine, Chicago, Illinois
a r t i c l e
i n f o
Article history: Received 2 March 2015 Accepted 19 May 2015 Keywords: knee arthroplasty complication total knee arthroplasty simultaneous outcomes
a b s t r a c t Simultaneous bilateral total knee arthroplasty (SB-TKA) is potentially a cost saving manner of caring for patients with bilateral symptomatic knee arthritis. We performed a retrospective analysis using the 2010–2012 American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) to evaluate the risk of perioperative complication following SB-TKA. Demographic characteristics, comorbidities, and 30-day complication rates were studied using a propensity score-matched analysis comparing patients undergoing unilateral TKA and SB-TKA. A total of 4489 patients met the inclusion criteria, of which 973 were SB-TKA. SB-TKA was associated with increased overall complications (P = 0.023), medical complications (P = 0.002) and reoperation (OR 2.12, P = 0.020). Further, total length of hospital stay (4.0 vs 3.4 days, P b 0.001) was significantly longer following bilateral surgery. © 2015 Elsevier Inc. All rights reserved.
Total knee arthroplasty (TKA) has long been shown to be both safe and efficacious for the treatment of degenerative conditions of the knee: relieving pain, restoring function, correcting deformity, and providing measurable improvements in quality of life [1–4]. Additionally, from a population health perspective, TKA has been demonstrated to be a cost effective treatment for symptomatic knee arthritis with estimates between ten and twenty thousand dollars per Quality Adjusted Life Year (QALY) [5,6]. In the United States in relationship to the rise in obesity, an aging population, and expanding surgical indications, there is a near exponential increase in symptomatic knee osteoarthritis and the subsequent demand for TKA [7]. Frequently patients present with bilateral symptomatic disease and deformity, which cause patients and surgeons to give consideration to simultaneous bilateral total knee arthroplasty (SB-TKA). For these patients, uncomplicated SB-TKA has been advocated as economically advantageous and efficient compared to the total cost and time of recovery for two separate unilateral TKA surgeries [8]. General cost of care for SB-TKA has been shown to be reduced by 18% to 26% and total hospital length of stay shortened from 6 to 4 days compared with staged bilateral TKA (StgB TKA). In
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.05.039. Reprints requests: David W. Manning, MD, 676 N St. Clair St, Suite 1350, Chicago, IL, 60611. http://dx.doi.org/10.1016/j.arth.2015.05.039 0883-5403/© 2015 Elsevier Inc. All rights reserved.
keeping with a perceived decreased work effort and cost for SB-TKA compared to StgB TKA, several third party payers, including Medicare and Medicaid, decrease reimbursement for the surgeon and hospital by 50% if a second total knee arthroplasty is performed within 90 days of the first total knee [9]. Although economic factors have been shown to favor SB-TKA over staged procedures, concern exists regarding potential increased rates of peri-operative morbidity and mortality [8–13]. While historical reports of SB-TKAs showed outcomes comparable to a unilateral total knee arthroplasty with respect to hospital length of stay and wound complications, [10–12] more recent analyses have shown an increased rate of perioperative complications with SB-TKA. Compared with unilateral TKA, there are reported increased risks of pulmonary embolism, need for blood transfusion, and higher rates of mortality with SB-TKAs [1]. In one study, patients older than 70 years of age with pre-existing pulmonary conditions had a 3-fold higher risk of postoperative complications when undergoing SB-TKA as compared to unilateral TKA [13]. In addition, post-operative complications are now considered indicators of quality, are publicly reported, and will be linked to reimbursement, thus potentially creating a negative fiscal incentive to perform SB-TKA [14]. Lastly, post-operative complications likely impose a significant but infrequently quantified cost burden on the American health care system [15]. Given the controversy regarding indications, outcomes, and potential risks and benefits of SB-TKA we sought to utilize the American College of Surgeons-National Quality Improvement Program (ACS-NSQIP) dataset to perform a population based assessment
1928
L.I. Suleiman et al. / The Journal of Arthroplasty 30 (2015) 1927–1930
comparing medical and surgical complication risk in patients undergoing unilateral TKA and SB-TKA.
Table 1 Demographics and Baseline Characteristics for the Unilateral and Simultaneous Bilateral Total Knee Arthroplasty Cohorts.
Methods We performed a retrospective analysis using the 2010–2012 ACSNSQIP dataset. Details of the dataset have been described previously [16]. In brief, the program includes 394 participating institutions across the United States and captures over 200 perioperative variables. Postoperative morbidity and mortality outcomes are tracked for 30 days regardless of discharge status. Participating institutions include academic medical centers, community hospitals and surgical centers that participate at their own discretion. Tracked variables include patient demographics and comorbid factors; intraoperative variables; laboratory values; and 30-day complications. These variables are prospectively collected and analyzed by dedicated clinical reviewers at each participating site in a standardized fashion. Each site employs surgical clinical reviewers who are rigorously trained to collect data through chart review and, in certain cases, discussion with the treating surgeon and/or patient, allowing for robust, quality data with demonstrated inter-rater reliability [16,17]. The NSQIP database was queried for patients who underwent either unilateral or simultaneous bilateral primary total knee arthroplasty by Current Procedural Terminology (CPT) codes variables in NSQIP (27447). Patient demographic characteristics, preoperative comorbidities, laboratory data, and 30-day perioperative outcomes were recorded. All variables were used as defined in the NSQIP User Guide. Outcomes of interest included post-operative 30-day complications, reoperation, hospital length of stay, readmission, and mortality. Overall complications were divided into surgical and medical. Variables with over 50% missing data were excluded. Propensity scores estimated with logistic regression using demographics, comorbidity data, and laboratory values were used to match unilateral to bilateral TKA patients in order to reduce confounding between cohorts. Calculated propensity scores for each patient were obtained using a logistic regression model and matched pairs were created using a 1:4 variable ratio, parallel, balanced nearest neighbor approach. Baseline demographics, comorbidities, and lab values were compared across the groups using generalized linear models adjusting for matched group with a log-link. Outcomes were then compared between these matched cohorts using generalized linear models with logit link for binomial outcomes or log-logit for length of stay. A clustered proportional hazards model was fit for time to discharge. Demographics are presented as percentages for categorical variables, and means and standard deviations for continuous variables. Measures of risk for outcomes are presented as odds ratios or hazard ratios (for time to discharge), and least square means for total hospital days. Estimates are presented with corresponding 95% confidence intervals. All analyses were run in SAS v9.4 (Cary, NC), and the level of significance was set at the nominal P b 0.05. Results There were 43,393 total knee arthroplasties in the ACS-NSQIP 2010–2012 dataset, of which 1105 (2.6%) were SB-TKAs. A total of 973 SB TKAs were matched at an average of 1:3.7 with 3516 unilateral TKA cases. (Table 1) After matching, there was no significant difference in demographic, laboratory, or comorbidity variables between unilateral TKA and SB-TKA groups. There was no significant difference between matched cohorts in rates of surgical complications (P = 0.267), including peri-prosthetic infection (P = 0.982), or readmission rate (P = 0.675) or 30-day mortality (P = 0.925). However, SB-TKA was associated with increased overall complications (OR 1.48, P = 0.023), medical complications (OR 1.88, P = 0.002) and reoperation risk (OR 2.12, P = 0.020) compared to unilateral TKA (Table 2). Further, total length of hospital stay (4.0 vs 3.4 days, P b 0.001) was significantly longer following SB-TKA.
N age ASA class 1 2 3 4 Bleeding disorder Race White Black/AA Asian Other/unknown General Anesthesia Diabetes IDDM NIDDM Current smoker Functional status Independent History COPD Hypertension medication Steroid use BMI Underweight Healthy weight Overweight Obese class I Obese class II Obese class III
Unilateral
Bilateral
3516 64.3 ± 10.1
973 64.0 ± 8.6
112 (3.2) 1931 (54.9) 1412 (40.2) 50 (1.4) 63 (1.8)
33 (3.4) 532 (54.7) 390 (40.1) 13 (1.3) 17 (1.8)
3005 (85.5) 175 (5.0) 98 (2.8) 238 (6.7) 2354 (67.1)
830 (85.3) 54 (5.6) 34 (3.5) 55 (5.6) 659 (67.7)
87 (2.5) 414 (11.8) 283 (8.1)
26 (2.7) 118 (12.1) 81 (8.3)
3436 (97.7) 75 (2.1) 2242 (63.8) 105 (3.0)
944 (97.0) 20 (2.1) 618 (63.5) 26 (2.7)
14 (0.4) 332 (9.4) 1003 (28.5) 1010 (28.7) 650 (18.5) 507 (14.4)
3 (0.3) 84 (8.6) 286 (29.4) 279 (28.7) 184 (18.9) 137 (14.1)
P 0.204 0.925
0.925 0.494
0.493 0.895
0.793 0.187 0.878 0.880 0.604 0.954
ASA, American Society of Anesthesiologists; AA, African American; IDDM, insulin-dependent diabetes mellitus; NIDDM, non-IDDM; COPD, chronic obstructive pulmonary disease; BMI, body mass index.
Table 3 shows the “raw” or absolute percentage of postoperative complications comparing unilateral versus bilateral total knee arthroplasty. Of note, there was a positive association between increasing ASA class and post-operative morbidity risk in both the unilateral TKA and SB-TKA groups. Discussion Total knee arthroplasty has demonstrated reproducible benefits in quality of life for patients with symptomatic knee arthritis. The demand for TKA has been projected to increase dramatically over the coming decades and the manner in which these surgeries are implemented will significantly impact the health care system in the areas of cost and quality [18]. For patients with bilateral knee involvement, the timing of surgical intervention is of paramount importance. Despite controversy regarding the safety of SB-TKA, it has comprised 4%–6% of all TKAs performed annually in the United States over the last several decades [6]. Advocates of SB-TKA cite multiple proposed benefits: decreased cost, quickened recovery, avoidance of multiple
Table 2 Odds Ratios for 30-Day Outcomes Comparing Simultaneous Bilateral to Unilateral Procedures. Outcome
Odds Ratio, 95% Confidence Interval, P Value
Overall Complications (N = 211) Surgical Complications (N = 56) Superficial wound infection (N = 34) Deep/joint space infection (N = 14) Medical Complications (N = 162) Reoperation (N = 60) Length of stay (Hazard ratio) Readmission (N = 52) Mortality (N = 5)
1.48 (1.09, 2.02) P 0.69 (0.34, 1.43) P 0.62 (0.24, 1.63) P 0.99 (0.27, 3.54) P 1.88 (1.34, 2.65) P 2.12 (1.25, 3.58) P 1.31 (1.22, 1.41) P 0.86 (0.41, 1.81) P 0.90 (0.10, 8.10) P
= 0.023 = 0.267 = 0.268 = 0.982 = 0.002 = 0.020 b 0.001 = 0.675 = 0.925
L.I. Suleiman et al. / The Journal of Arthroplasty 30 (2015) 1927–1930 Table 3 Absolute Percentages for 30-Day Outcomes Comparing Simultaneous Bilateral to Unilateral Procedures. Outcome
Unilateral
Bilateral
Overall Complications (N = 211) Surgical Complications (N = 56) Superficial wound infection (N = 34) Deep/joint space infection (N = 14) Medical Complications (N = 162) Return to OR (N = 60) Readmission (N = 52) 30 day mortality rate (N = 5)
150 (4.3%) 47 (1.3%) 29 (0.8%) 11 (0.3%) 107 (3.0%) 38 (1.1%) 43 (4.4%) 4 (0.1%)
61 (6.3%) 9 (0.9%) 5 (0.5%) 3 (0.3%) 55 (5.7%) 22 (2.3%) 9 (3.8%) 1 (0.1%)
anesthetic events, and equivalent functional outcomes [19]. Radiographic outcomes have also been shown to be equivalent between simultaneous and staged procedures [20]. These proposed benefits must be considered within the context of potential risk associated with prolonged anesthesia time, greater surgical stress, and a more intensive rehabilitation course. Prior investigators have identified increased rates of cardiovascular events, thromboembolic events, peri-operative blood transfusion, and mortality in patients undergoing SB-TKA [21]. There is currently no prospective, randomized controlled trial that exists in the joint arthroplasty literature that can guide physicians and their patients to proceed with simultaneous bilateral vs unilateral TKA. However, the Consensus Conference on Bilateral Total Knee Arthroplasty, a multidisciplinary group of internists, surgeons, and anesthesiologists agreed that same-day SB-TKA was associated with higher risk of complications [21]. In addition, they recommend further investigation regarding the safety and efficacy of SB-TKA. Several single center series have demonstrated comparable risk profiles and functional recovery rates for unilateral TKA and SB-TKA [19,22–24]. Powell et al compared pain levels and recovery of ambulatory skills in 59 matched pairs of unilateral TKA and SB-TKA patients. They found comparable pain scores after the first 48 h after surgery and only a 36-h delay in ambulatory milestones [22]. Kim et al found comparable complication and mortality rates between cohorts of 2385 SB-TKAs and 719 unilateral TKAs at a single Korean hospital [19]. Poultsides et al studied 17,959 TKA cases from a single institution and found that the overall rates of deep infection and reoperation for infection were comparable between unilateral and SB-TKA [23]. However, other single center series and a multi-center study comparing SB-TKA to unilateral TKA have shown increased complication rates [25–27]. One study compared 428 SB-TKA and 3239 unilateral TKA cases and found a three-fold increase in rates of pulmonary embolism after SB-TKA compared to unilateral TKA (0.61% vs 1.87%, P b 0.001) [26]. Memtsoudis et al analyzed the Nationwide Inpatient Sample, which is the largest all-payer database in the United States, encompassing over 1000 hospitals of varying size and teaching status in 38 states. Their sample included over 670,000 TKA admissions with 6.52% being bilateral cases. An unadjusted analysis revealed an increased rate of perioperative complications (9.45% vs 7.07%, P b 0.0001) and mortality (0.30% vs 0.14%, P b 0.0001) for SB-TKA compared to unilateral TKA despite younger age and lower comorbidity burden in the bilateral group [27]. Our adjusted analysis of the 2010–2012 NSQIP dataset corroborates previous findings of an increased complication rate in SB-TKA. The significant difference in complication profile found in our analysis of comorbidity-matched cohorts of SB-TKA and unilateral TKA patients was driven predominantly by differences in post-operative medical complications rates. We also found SB-TKA to be associated with increased risk for reoperation, although there was no statistically significant difference in overall surgical complications. Contrary to other investigations we found no relationship between SB-TKA and increased peri-prosthetic infection or mortality at 30 day follow-up. There were no statistical differences in surgical complications, readmissions, and
1929
mortality; this may represent a type II error due to the low rates of occurrence for these outcomes. Our study did also find a positive association between increasing ASA class and post-operative morbidity risk in both the unilateral TKA and SB-TKA groups. This result is consistent with previously reported data that there is an increase in postoperative complications with an ASA score of 3 or 4, older age and preoperative cardiovascular disease [28]. Of note, 61% of patients in both the unilateral and bilateral groups were obese, consistent with recent trends for arthroplasty patients. Over the last twenty years the rate of obesity has tripled among Medicare total knee patients; a BMI greater than 40 has been shown to increase the risk for postoperative complications following joint arthroplasty [29,30]. Findings for this majority obese cohort may not be applicable to all arthroplasty patients. We also found a statistically significant difference of 0.6 days in hospital length of stay. This difference must be considered in the context of a second admission for staged bilateral arthroplasty. With the continued rollout of the Affordable Care Act, health care delivery is intended to be increasingly value-driven. Pursuing a strategy that minimizes risk may ultimately offset cost benefits of simultaneous procedures. As complications are associated with significant cost, minimizing risk may be a cost saving endeavor. If nothing else, surgeons and patients should be equipped with adequate information when discussing the risks and benefits of various surgical options. Surgeon selection bias is a legitimate concern for any retrospective series examining differences in surgical outcomes. In this case, potential surgeon selection bias would make it more difficult to statistically identify a difference in complication risk, as surgeons would likely select healthier patients (ie: lower risk patients) for SB-TKA. While the potential for bias can never be truly eliminated in a population-based analysis, here we attempt to limit confounding by use of a propensity scorematching algorithm. After matching, baseline differences were eliminated for all tracked variables. Additionally, use of a comprehensive database such as NSQIP helps to reduce the potential for untracked variables to bias results. Various limitations persist. Participating NSQIP institutions may not represent all settings where arthroplasty surgery is performed. The database tracks patients for only 30 days and it is possible that the incidence of complications in one or both study populations may be greater at longer term follow-up than identified in this study. Lastly, the database does not include information on the clinical, functional, or cost impact of the identified complications, data on functional outcomes or hospital and surgeon factors, thus precluding our ability to control for these factors. We believe this analysis adds to the existing literature that indicates an increased risk of post-operative complications with SB-TKA compared to unilateral TKA. Our findings do not however, accurately describe risk associated with staged bilateral TKA over time. We caution against the unsupported supposition that risk of staged bilateral TKA over time is simply additive. We are unable to accurately assess the risk associated with a second future surgery, which could plausibly be lower or higher depending upon any number of variables. While the health care system continues to be redefined in the areas of value, quality and equitable reimbursement, we as physicians should advocate for patient safety, optimal outcome and the avoidance of risk. This investigation clearly links SB-TKA to heightened risk and should assist physicians and patients perform accurate relative risk assessment when considering surgical management for bilateral symptomatic knee arthritis.
References 1. Fu D, Li G, Chen K, et al. Comparison of clinical outcome between simultaneous-bilateral and staged-bilateral total knee arthroplasty: a systematic review of retrospective studies. J Arthroplasty 2013;28(7):1141. 2. Rodriguez JA, Bhende H, Ranawat CS. Total condylar knee replacement: a 20-year followup study. Clin Orthop Relat Res 2001;388:10. 3. Gill GS, Joshi AB, Mills DM. Total condylar knee arthroplasty. 16- to 21-year results. Clin Orthop Relat Res 1999;367:210.
1930
L.I. Suleiman et al. / The Journal of Arthroplasty 30 (2015) 1927–1930
4. Thadani PJ, Vince KG, Ortaaslan SG, et al. Ten- to 12-year followup of the InsallBurstein I total knee prosthesis. Clin Orthop Relat Res 2000;380:17. 5. Manning D. Simultaneous Bilateral Total Knee Arthroplasty: What Is the Value? Commentary on an article by Susan M. Odum, PhD, and Bryan D. Springer, MD: "In-Hospital Complication Rates and Associated Factors After Simultaneous Bilateral Versus Unilateral Total Knee Arthroplasty". J Bone Joint Surg Am 2014;96(13):e114. 6. Odum SM, Troyer JL, Kelly MP, et al. A cost-utility analysis comparing the cost-effectiveness of simultaneous and staged bilateral total knee arthroplasty. J Bone Joint Surg Am 2013;95(16):1441. 7. Hootman JM, Helmick CG. Projections of US prevalence of arthritis and associated activity limitations. Arthritis Rheum 2006;54(1):226. 8. Macario A, Schilling P, Rubio R, et al. Economics of one-stage versus two-stage bilateral total knee arthroplasties. Clin Orthop Relat Res 2003;414:149. 9. Reuben JD, Meyers SJ, Cox DD, et al. Cost comparison between bilateral simultaneous, staged, and unilateral total joint arthroplasty. J Arthroplasty 1998;13(2):172. 10. Soudry M, Binazzi R, Insall JN, et al. Successive bilateral total knee replacement. J Bone Joint Surg Am 1985;67(4):573. 11. Ritter MA, Meding JB. Bilateral simultaneous total knee arthroplasty. J Arthroplasty 1987;2(3):185. 12. Morrey BF, Adams RA, Ilstrup DM, et al. Complications and mortality associated with bilateral or unilateral total knee arthroplasty. J Bone Joint Surg Am 1987;69(4):484. 13. Fabi DW, Mohan V, Goldstein WM, et al. Unilateral vs bilateral total knee arthroplasty risk factors increasing morbidity. J Arthroplasty 2011;26(5):668. 14. Hackbarth G, Reischauer R, Mutti A. Collective accountability for medical care–toward bundled Medicare payments. N Engl J Med 2008;359(1):3. 15. Maradit Kremers H, et al. Determinants of direct medical costs in primary and revision total knee arthroplasty. Clin Orthop Relat Res 2013;471(1):206. 16. Henderson WG, Daley J. Design and statistical methodology of the National Surgical Quality Improvement Program: why is it what it is? Am J Surg 2009;198(5 Suppl):S19. 17. Shiloach M, Frencher SK, Steeger JE, et al. Toward robust information: data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg 2010;210(1):6.
18. Nwachukwu BU, Bozic KJ, Schairer WW, et al. Current Status of Cost Utility Analyses in Total Joint Arthroplasty: A Systematic Review. Clin Orthop Relat Res 2014; 473(5):1815. 19. Kim YH, Choi YW, Kim JS. Simultaneous bilateral sequential total knee replacement is as safe as unilateral total knee replacement. J Bone Joint Surg (Br) 2009; 91(1):64. 20. Kilincoglu V, Unay K, Akan K, et al. Component alignment in simultaneous bilateral or unilateral total knee arthroplasty. Int Orthop 2011;35(1):43. 21. Memtsoudis SG, Hargett M, Russell LA, et al. Consensus statement from the consensus conference on bilateral total knee arthroplasty group. Clin Orthop Relat Res 2013; 471(8):2649. 22. Powell RS, Pulido P, Tuason MS, et al. Bilateral vs unilateral total knee arthroplasty: a patient-based comparison of pain levels and recovery of ambulatory skills. J Arthroplasty 2006;21(5):642. 23. Poultsides LA, Memtsoudis SG, Vasilakakos T, et al. Infection following simultaneous bilateral total knee arthroplasty. J Arthroplasty 2013;28(8 Suppl):92. 24. Cohen RG, Forrest CJ, Benjamin JB. Safety and efficacy of bilateral total knee arthroplasty. J Arthroplasty 1997;12(5):497. 25. Luscombe JC, Theivendran K, Abudu A, et al. The relative safety of one-stage bilateral total knee arthroplasty. Int Orthop 2009;33(1):101. 26. Yeager AM, Ruel AV, Westrich GH. Are bilateral total joint arthroplasty patients at a higher risk of developing pulmonary embolism following total hip and knee surgery? J Arthroplasty 2014;29(5):900. 27. Memtsoudis SG, Ma Y, Della Valle AG, et al. Perioperative outcomes after unilateral and bilateral total knee arthroplasty. Anesthesiology 2009;111(6):1206. 28. Koh IJ, Kim GH, Kong CG, et al. The Patient's Age and American Society of Anesthesiologists Status Are Reasonable Criteria for Deciding Whether to Perform Same-Day Bilateral TKA. J Arthroplasty 2014;30(5):770. 29. Cram P, Lu X, Kates SL, et al. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991-2010. JAMA 2012;308(12):1227. 30. Alvi HM, Mednick RE, Krishnan V, et al. The Effect of BMI on 30 Day Outcomes Following Total Joint Arthroplasty. J Arthroplasty 2015 [in press].
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Perioperative Outcomes Following Unilateral Versus Bilateral Total Knee Arthroplasty Linda I. Suleiman, MD a, Adam I. Edelstein, MD a, Rachel M. Thompson, MD a, Hasham M. Alvi, MD a, Mary J. Kwasny, ScD b, David W. Manning, MD a a b
Department of Orthopaedic Surgery, Northwestern University, Feinberg School of Medicine, Chicago, Illinois Department of Preventative Medicine, Division of Biostatistics, Northwestern University, Feinberg School of Medicine, Chicago, Illinois
a r t i c l e
i n f o
Article history: Received 2 March 2015 Accepted 19 May 2015 Keywords: knee arthroplasty complication total knee arthroplasty simultaneous outcomes
a b s t r a c t Simultaneous bilateral total knee arthroplasty (SB-TKA) is potentially a cost saving manner of caring for patients with bilateral symptomatic knee arthritis. We performed a retrospective analysis using the 2010–2012 American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) to evaluate the risk of perioperative complication following SB-TKA. Demographic characteristics, comorbidities, and 30-day complication rates were studied using a propensity score-matched analysis comparing patients undergoing unilateral TKA and SB-TKA. A total of 4489 patients met the inclusion criteria, of which 973 were SB-TKA. SB-TKA was associated with increased overall complications (P = 0.023), medical complications (P = 0.002) and reoperation (OR 2.12, P = 0.020). Further, total length of hospital stay (4.0 vs 3.4 days, P b 0.001) was significantly longer following bilateral surgery. © 2015 Elsevier Inc. All rights reserved.
Total knee arthroplasty (TKA) has long been shown to be both safe and efficacious for the treatment of degenerative conditions of the knee: relieving pain, restoring function, correcting deformity, and providing measurable improvements in quality of life [1–4]. Additionally, from a population health perspective, TKA has been demonstrated to be a cost effective treatment for symptomatic knee arthritis with estimates between ten and twenty thousand dollars per Quality Adjusted Life Year (QALY) [5,6]. In the United States in relationship to the rise in obesity, an aging population, and expanding surgical indications, there is a near exponential increase in symptomatic knee osteoarthritis and the subsequent demand for TKA [7]. Frequently patients present with bilateral symptomatic disease and deformity, which cause patients and surgeons to give consideration to simultaneous bilateral total knee arthroplasty (SB-TKA). For these patients, uncomplicated SB-TKA has been advocated as economically advantageous and efficient compared to the total cost and time of recovery for two separate unilateral TKA surgeries [8]. General cost of care for SB-TKA has been shown to be reduced by 18% to 26% and total hospital length of stay shortened from 6 to 4 days compared with staged bilateral TKA (StgB TKA). In
One or more of the authors of this paper have disclosed potential or pertinent conflicts of interest, which may include receipt of payment, either direct or indirect, institutional support, or association with an entity in the biomedical field which may be perceived to have potential conflict of interest with this work. For full disclosure statements refer to http://dx.doi.org/10.1016/j.arth.2015.05.039. Reprints requests: David W. Manning, MD, 676 N St. Clair St, Suite 1350, Chicago, IL, 60611. http://dx.doi.org/10.1016/j.arth.2015.05.039 0883-5403/© 2015 Elsevier Inc. All rights reserved.
keeping with a perceived decreased work effort and cost for SB-TKA compared to StgB TKA, several third party payers, including Medicare and Medicaid, decrease reimbursement for the surgeon and hospital by 50% if a second total knee arthroplasty is performed within 90 days of the first total knee [9]. Although economic factors have been shown to favor SB-TKA over staged procedures, concern exists regarding potential increased rates of peri-operative morbidity and mortality [8–13]. While historical reports of SB-TKAs showed outcomes comparable to a unilateral total knee arthroplasty with respect to hospital length of stay and wound complications, [10–12] more recent analyses have shown an increased rate of perioperative complications with SB-TKA. Compared with unilateral TKA, there are reported increased risks of pulmonary embolism, need for blood transfusion, and higher rates of mortality with SB-TKAs [1]. In one study, patients older than 70 years of age with pre-existing pulmonary conditions had a 3-fold higher risk of postoperative complications when undergoing SB-TKA as compared to unilateral TKA [13]. In addition, post-operative complications are now considered indicators of quality, are publicly reported, and will be linked to reimbursement, thus potentially creating a negative fiscal incentive to perform SB-TKA [14]. Lastly, post-operative complications likely impose a significant but infrequently quantified cost burden on the American health care system [15]. Given the controversy regarding indications, outcomes, and potential risks and benefits of SB-TKA we sought to utilize the American College of Surgeons-National Quality Improvement Program (ACS-NSQIP) dataset to perform a population based assessment
1928
L.I. Suleiman et al. / The Journal of Arthroplasty 30 (2015) 1927–1930
comparing medical and surgical complication risk in patients undergoing unilateral TKA and SB-TKA.
Table 1 Demographics and Baseline Characteristics for the Unilateral and Simultaneous Bilateral Total Knee Arthroplasty Cohorts.
Methods We performed a retrospective analysis using the 2010–2012 ACSNSQIP dataset. Details of the dataset have been described previously [16]. In brief, the program includes 394 participating institutions across the United States and captures over 200 perioperative variables. Postoperative morbidity and mortality outcomes are tracked for 30 days regardless of discharge status. Participating institutions include academic medical centers, community hospitals and surgical centers that participate at their own discretion. Tracked variables include patient demographics and comorbid factors; intraoperative variables; laboratory values; and 30-day complications. These variables are prospectively collected and analyzed by dedicated clinical reviewers at each participating site in a standardized fashion. Each site employs surgical clinical reviewers who are rigorously trained to collect data through chart review and, in certain cases, discussion with the treating surgeon and/or patient, allowing for robust, quality data with demonstrated inter-rater reliability [16,17]. The NSQIP database was queried for patients who underwent either unilateral or simultaneous bilateral primary total knee arthroplasty by Current Procedural Terminology (CPT) codes variables in NSQIP (27447). Patient demographic characteristics, preoperative comorbidities, laboratory data, and 30-day perioperative outcomes were recorded. All variables were used as defined in the NSQIP User Guide. Outcomes of interest included post-operative 30-day complications, reoperation, hospital length of stay, readmission, and mortality. Overall complications were divided into surgical and medical. Variables with over 50% missing data were excluded. Propensity scores estimated with logistic regression using demographics, comorbidity data, and laboratory values were used to match unilateral to bilateral TKA patients in order to reduce confounding between cohorts. Calculated propensity scores for each patient were obtained using a logistic regression model and matched pairs were created using a 1:4 variable ratio, parallel, balanced nearest neighbor approach. Baseline demographics, comorbidities, and lab values were compared across the groups using generalized linear models adjusting for matched group with a log-link. Outcomes were then compared between these matched cohorts using generalized linear models with logit link for binomial outcomes or log-logit for length of stay. A clustered proportional hazards model was fit for time to discharge. Demographics are presented as percentages for categorical variables, and means and standard deviations for continuous variables. Measures of risk for outcomes are presented as odds ratios or hazard ratios (for time to discharge), and least square means for total hospital days. Estimates are presented with corresponding 95% confidence intervals. All analyses were run in SAS v9.4 (Cary, NC), and the level of significance was set at the nominal P b 0.05. Results There were 43,393 total knee arthroplasties in the ACS-NSQIP 2010–2012 dataset, of which 1105 (2.6%) were SB-TKAs. A total of 973 SB TKAs were matched at an average of 1:3.7 with 3516 unilateral TKA cases. (Table 1) After matching, there was no significant difference in demographic, laboratory, or comorbidity variables between unilateral TKA and SB-TKA groups. There was no significant difference between matched cohorts in rates of surgical complications (P = 0.267), including peri-prosthetic infection (P = 0.982), or readmission rate (P = 0.675) or 30-day mortality (P = 0.925). However, SB-TKA was associated with increased overall complications (OR 1.48, P = 0.023), medical complications (OR 1.88, P = 0.002) and reoperation risk (OR 2.12, P = 0.020) compared to unilateral TKA (Table 2). Further, total length of hospital stay (4.0 vs 3.4 days, P b 0.001) was significantly longer following SB-TKA.
N age ASA class 1 2 3 4 Bleeding disorder Race White Black/AA Asian Other/unknown General Anesthesia Diabetes IDDM NIDDM Current smoker Functional status Independent History COPD Hypertension medication Steroid use BMI Underweight Healthy weight Overweight Obese class I Obese class II Obese class III
Unilateral
Bilateral
3516 64.3 ± 10.1
973 64.0 ± 8.6
112 (3.2) 1931 (54.9) 1412 (40.2) 50 (1.4) 63 (1.8)
33 (3.4) 532 (54.7) 390 (40.1) 13 (1.3) 17 (1.8)
3005 (85.5) 175 (5.0) 98 (2.8) 238 (6.7) 2354 (67.1)
830 (85.3) 54 (5.6) 34 (3.5) 55 (5.6) 659 (67.7)
87 (2.5) 414 (11.8) 283 (8.1)
26 (2.7) 118 (12.1) 81 (8.3)
3436 (97.7) 75 (2.1) 2242 (63.8) 105 (3.0)
944 (97.0) 20 (2.1) 618 (63.5) 26 (2.7)
14 (0.4) 332 (9.4) 1003 (28.5) 1010 (28.7) 650 (18.5) 507 (14.4)
3 (0.3) 84 (8.6) 286 (29.4) 279 (28.7) 184 (18.9) 137 (14.1)
P 0.204 0.925
0.925 0.494
0.493 0.895
0.793 0.187 0.878 0.880 0.604 0.954
ASA, American Society of Anesthesiologists; AA, African American; IDDM, insulin-dependent diabetes mellitus; NIDDM, non-IDDM; COPD, chronic obstructive pulmonary disease; BMI, body mass index.
Table 3 shows the “raw” or absolute percentage of postoperative complications comparing unilateral versus bilateral total knee arthroplasty. Of note, there was a positive association between increasing ASA class and post-operative morbidity risk in both the unilateral TKA and SB-TKA groups. Discussion Total knee arthroplasty has demonstrated reproducible benefits in quality of life for patients with symptomatic knee arthritis. The demand for TKA has been projected to increase dramatically over the coming decades and the manner in which these surgeries are implemented will significantly impact the health care system in the areas of cost and quality [18]. For patients with bilateral knee involvement, the timing of surgical intervention is of paramount importance. Despite controversy regarding the safety of SB-TKA, it has comprised 4%–6% of all TKAs performed annually in the United States over the last several decades [6]. Advocates of SB-TKA cite multiple proposed benefits: decreased cost, quickened recovery, avoidance of multiple
Table 2 Odds Ratios for 30-Day Outcomes Comparing Simultaneous Bilateral to Unilateral Procedures. Outcome
Odds Ratio, 95% Confidence Interval, P Value
Overall Complications (N = 211) Surgical Complications (N = 56) Superficial wound infection (N = 34) Deep/joint space infection (N = 14) Medical Complications (N = 162) Reoperation (N = 60) Length of stay (Hazard ratio) Readmission (N = 52) Mortality (N = 5)
1.48 (1.09, 2.02) P 0.69 (0.34, 1.43) P 0.62 (0.24, 1.63) P 0.99 (0.27, 3.54) P 1.88 (1.34, 2.65) P 2.12 (1.25, 3.58) P 1.31 (1.22, 1.41) P 0.86 (0.41, 1.81) P 0.90 (0.10, 8.10) P
= 0.023 = 0.267 = 0.268 = 0.982 = 0.002 = 0.020 b 0.001 = 0.675 = 0.925
L.I. Suleiman et al. / The Journal of Arthroplasty 30 (2015) 1927–1930 Table 3 Absolute Percentages for 30-Day Outcomes Comparing Simultaneous Bilateral to Unilateral Procedures. Outcome
Unilateral
Bilateral
Overall Complications (N = 211) Surgical Complications (N = 56) Superficial wound infection (N = 34) Deep/joint space infection (N = 14) Medical Complications (N = 162) Return to OR (N = 60) Readmission (N = 52) 30 day mortality rate (N = 5)
150 (4.3%) 47 (1.3%) 29 (0.8%) 11 (0.3%) 107 (3.0%) 38 (1.1%) 43 (4.4%) 4 (0.1%)
61 (6.3%) 9 (0.9%) 5 (0.5%) 3 (0.3%) 55 (5.7%) 22 (2.3%) 9 (3.8%) 1 (0.1%)
anesthetic events, and equivalent functional outcomes [19]. Radiographic outcomes have also been shown to be equivalent between simultaneous and staged procedures [20]. These proposed benefits must be considered within the context of potential risk associated with prolonged anesthesia time, greater surgical stress, and a more intensive rehabilitation course. Prior investigators have identified increased rates of cardiovascular events, thromboembolic events, peri-operative blood transfusion, and mortality in patients undergoing SB-TKA [21]. There is currently no prospective, randomized controlled trial that exists in the joint arthroplasty literature that can guide physicians and their patients to proceed with simultaneous bilateral vs unilateral TKA. However, the Consensus Conference on Bilateral Total Knee Arthroplasty, a multidisciplinary group of internists, surgeons, and anesthesiologists agreed that same-day SB-TKA was associated with higher risk of complications [21]. In addition, they recommend further investigation regarding the safety and efficacy of SB-TKA. Several single center series have demonstrated comparable risk profiles and functional recovery rates for unilateral TKA and SB-TKA [19,22–24]. Powell et al compared pain levels and recovery of ambulatory skills in 59 matched pairs of unilateral TKA and SB-TKA patients. They found comparable pain scores after the first 48 h after surgery and only a 36-h delay in ambulatory milestones [22]. Kim et al found comparable complication and mortality rates between cohorts of 2385 SB-TKAs and 719 unilateral TKAs at a single Korean hospital [19]. Poultsides et al studied 17,959 TKA cases from a single institution and found that the overall rates of deep infection and reoperation for infection were comparable between unilateral and SB-TKA [23]. However, other single center series and a multi-center study comparing SB-TKA to unilateral TKA have shown increased complication rates [25–27]. One study compared 428 SB-TKA and 3239 unilateral TKA cases and found a three-fold increase in rates of pulmonary embolism after SB-TKA compared to unilateral TKA (0.61% vs 1.87%, P b 0.001) [26]. Memtsoudis et al analyzed the Nationwide Inpatient Sample, which is the largest all-payer database in the United States, encompassing over 1000 hospitals of varying size and teaching status in 38 states. Their sample included over 670,000 TKA admissions with 6.52% being bilateral cases. An unadjusted analysis revealed an increased rate of perioperative complications (9.45% vs 7.07%, P b 0.0001) and mortality (0.30% vs 0.14%, P b 0.0001) for SB-TKA compared to unilateral TKA despite younger age and lower comorbidity burden in the bilateral group [27]. Our adjusted analysis of the 2010–2012 NSQIP dataset corroborates previous findings of an increased complication rate in SB-TKA. The significant difference in complication profile found in our analysis of comorbidity-matched cohorts of SB-TKA and unilateral TKA patients was driven predominantly by differences in post-operative medical complications rates. We also found SB-TKA to be associated with increased risk for reoperation, although there was no statistically significant difference in overall surgical complications. Contrary to other investigations we found no relationship between SB-TKA and increased peri-prosthetic infection or mortality at 30 day follow-up. There were no statistical differences in surgical complications, readmissions, and
1929
mortality; this may represent a type II error due to the low rates of occurrence for these outcomes. Our study did also find a positive association between increasing ASA class and post-operative morbidity risk in both the unilateral TKA and SB-TKA groups. This result is consistent with previously reported data that there is an increase in postoperative complications with an ASA score of 3 or 4, older age and preoperative cardiovascular disease [28]. Of note, 61% of patients in both the unilateral and bilateral groups were obese, consistent with recent trends for arthroplasty patients. Over the last twenty years the rate of obesity has tripled among Medicare total knee patients; a BMI greater than 40 has been shown to increase the risk for postoperative complications following joint arthroplasty [29,30]. Findings for this majority obese cohort may not be applicable to all arthroplasty patients. We also found a statistically significant difference of 0.6 days in hospital length of stay. This difference must be considered in the context of a second admission for staged bilateral arthroplasty. With the continued rollout of the Affordable Care Act, health care delivery is intended to be increasingly value-driven. Pursuing a strategy that minimizes risk may ultimately offset cost benefits of simultaneous procedures. As complications are associated with significant cost, minimizing risk may be a cost saving endeavor. If nothing else, surgeons and patients should be equipped with adequate information when discussing the risks and benefits of various surgical options. Surgeon selection bias is a legitimate concern for any retrospective series examining differences in surgical outcomes. In this case, potential surgeon selection bias would make it more difficult to statistically identify a difference in complication risk, as surgeons would likely select healthier patients (ie: lower risk patients) for SB-TKA. While the potential for bias can never be truly eliminated in a population-based analysis, here we attempt to limit confounding by use of a propensity scorematching algorithm. After matching, baseline differences were eliminated for all tracked variables. Additionally, use of a comprehensive database such as NSQIP helps to reduce the potential for untracked variables to bias results. Various limitations persist. Participating NSQIP institutions may not represent all settings where arthroplasty surgery is performed. The database tracks patients for only 30 days and it is possible that the incidence of complications in one or both study populations may be greater at longer term follow-up than identified in this study. Lastly, the database does not include information on the clinical, functional, or cost impact of the identified complications, data on functional outcomes or hospital and surgeon factors, thus precluding our ability to control for these factors. We believe this analysis adds to the existing literature that indicates an increased risk of post-operative complications with SB-TKA compared to unilateral TKA. Our findings do not however, accurately describe risk associated with staged bilateral TKA over time. We caution against the unsupported supposition that risk of staged bilateral TKA over time is simply additive. We are unable to accurately assess the risk associated with a second future surgery, which could plausibly be lower or higher depending upon any number of variables. While the health care system continues to be redefined in the areas of value, quality and equitable reimbursement, we as physicians should advocate for patient safety, optimal outcome and the avoidance of risk. This investigation clearly links SB-TKA to heightened risk and should assist physicians and patients perform accurate relative risk assessment when considering surgical management for bilateral symptomatic knee arthritis.
References 1. Fu D, Li G, Chen K, et al. Comparison of clinical outcome between simultaneous-bilateral and staged-bilateral total knee arthroplasty: a systematic review of retrospective studies. J Arthroplasty 2013;28(7):1141. 2. Rodriguez JA, Bhende H, Ranawat CS. Total condylar knee replacement: a 20-year followup study. Clin Orthop Relat Res 2001;388:10. 3. Gill GS, Joshi AB, Mills DM. Total condylar knee arthroplasty. 16- to 21-year results. Clin Orthop Relat Res 1999;367:210.
1930
L.I. Suleiman et al. / The Journal of Arthroplasty 30 (2015) 1927–1930
4. Thadani PJ, Vince KG, Ortaaslan SG, et al. Ten- to 12-year followup of the InsallBurstein I total knee prosthesis. Clin Orthop Relat Res 2000;380:17. 5. Manning D. Simultaneous Bilateral Total Knee Arthroplasty: What Is the Value? Commentary on an article by Susan M. Odum, PhD, and Bryan D. Springer, MD: "In-Hospital Complication Rates and Associated Factors After Simultaneous Bilateral Versus Unilateral Total Knee Arthroplasty". J Bone Joint Surg Am 2014;96(13):e114. 6. Odum SM, Troyer JL, Kelly MP, et al. A cost-utility analysis comparing the cost-effectiveness of simultaneous and staged bilateral total knee arthroplasty. J Bone Joint Surg Am 2013;95(16):1441. 7. Hootman JM, Helmick CG. Projections of US prevalence of arthritis and associated activity limitations. Arthritis Rheum 2006;54(1):226. 8. Macario A, Schilling P, Rubio R, et al. Economics of one-stage versus two-stage bilateral total knee arthroplasties. Clin Orthop Relat Res 2003;414:149. 9. Reuben JD, Meyers SJ, Cox DD, et al. Cost comparison between bilateral simultaneous, staged, and unilateral total joint arthroplasty. J Arthroplasty 1998;13(2):172. 10. Soudry M, Binazzi R, Insall JN, et al. Successive bilateral total knee replacement. J Bone Joint Surg Am 1985;67(4):573. 11. Ritter MA, Meding JB. Bilateral simultaneous total knee arthroplasty. J Arthroplasty 1987;2(3):185. 12. Morrey BF, Adams RA, Ilstrup DM, et al. Complications and mortality associated with bilateral or unilateral total knee arthroplasty. J Bone Joint Surg Am 1987;69(4):484. 13. Fabi DW, Mohan V, Goldstein WM, et al. Unilateral vs bilateral total knee arthroplasty risk factors increasing morbidity. J Arthroplasty 2011;26(5):668. 14. Hackbarth G, Reischauer R, Mutti A. Collective accountability for medical care–toward bundled Medicare payments. N Engl J Med 2008;359(1):3. 15. Maradit Kremers H, et al. Determinants of direct medical costs in primary and revision total knee arthroplasty. Clin Orthop Relat Res 2013;471(1):206. 16. Henderson WG, Daley J. Design and statistical methodology of the National Surgical Quality Improvement Program: why is it what it is? Am J Surg 2009;198(5 Suppl):S19. 17. Shiloach M, Frencher SK, Steeger JE, et al. Toward robust information: data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg 2010;210(1):6.
18. Nwachukwu BU, Bozic KJ, Schairer WW, et al. Current Status of Cost Utility Analyses in Total Joint Arthroplasty: A Systematic Review. Clin Orthop Relat Res 2014; 473(5):1815. 19. Kim YH, Choi YW, Kim JS. Simultaneous bilateral sequential total knee replacement is as safe as unilateral total knee replacement. J Bone Joint Surg (Br) 2009; 91(1):64. 20. Kilincoglu V, Unay K, Akan K, et al. Component alignment in simultaneous bilateral or unilateral total knee arthroplasty. Int Orthop 2011;35(1):43. 21. Memtsoudis SG, Hargett M, Russell LA, et al. Consensus statement from the consensus conference on bilateral total knee arthroplasty group. Clin Orthop Relat Res 2013; 471(8):2649. 22. Powell RS, Pulido P, Tuason MS, et al. Bilateral vs unilateral total knee arthroplasty: a patient-based comparison of pain levels and recovery of ambulatory skills. J Arthroplasty 2006;21(5):642. 23. Poultsides LA, Memtsoudis SG, Vasilakakos T, et al. Infection following simultaneous bilateral total knee arthroplasty. J Arthroplasty 2013;28(8 Suppl):92. 24. Cohen RG, Forrest CJ, Benjamin JB. Safety and efficacy of bilateral total knee arthroplasty. J Arthroplasty 1997;12(5):497. 25. Luscombe JC, Theivendran K, Abudu A, et al. The relative safety of one-stage bilateral total knee arthroplasty. Int Orthop 2009;33(1):101. 26. Yeager AM, Ruel AV, Westrich GH. Are bilateral total joint arthroplasty patients at a higher risk of developing pulmonary embolism following total hip and knee surgery? J Arthroplasty 2014;29(5):900. 27. Memtsoudis SG, Ma Y, Della Valle AG, et al. Perioperative outcomes after unilateral and bilateral total knee arthroplasty. Anesthesiology 2009;111(6):1206. 28. Koh IJ, Kim GH, Kong CG, et al. The Patient's Age and American Society of Anesthesiologists Status Are Reasonable Criteria for Deciding Whether to Perform Same-Day Bilateral TKA. J Arthroplasty 2014;30(5):770. 29. Cram P, Lu X, Kates SL, et al. Total knee arthroplasty volume, utilization, and outcomes among Medicare beneficiaries, 1991-2010. JAMA 2012;308(12):1227. 30. Alvi HM, Mednick RE, Krishnan V, et al. The Effect of BMI on 30 Day Outcomes Following Total Joint Arthroplasty. J Arthroplasty 2015 [in press].
