- Email: [email protected]
Have Venous Thromboembolism Rates Decreased in Total Hip and Knee Arthroplasty?
The Journal of Arthroplasty xxx (2019) 1e6
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Have Venous Thromboembolism Rates Decreased in Total Hip and Knee Arthroplasty? Jared A. Warren, DO, ATC, CSCS a, Kavin Sundaram, MD, MSc a, Hiba K. Anis, MD a, Atul F. Kamath, MD a, Carlos A. Higuera, MD b, Nicolas S. Piuzzi, MD a, * a b
Department of Orthopedic Surgery, Cleveland Clinic, Cleveland, OH Department of Orthopedic Surgery, Cleveland Clinic Florida, Weston, FL
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 June 2019 Received in revised form 13 August 2019 Accepted 22 August 2019 Available online xxx
Background: Venous thromboembolism (VTE) is a major cause of morbidity, mortality, and healthcare costs in arthroplasty patients. In an effort to reduce VTEs, numerous strategies and guidelines have been implemented, but their impact remains unclear. The purpose of this study is to compare annual trends in 30-day VTE, deep vein thrombosis (DVT), pulmonary embolism (PE), and all-cause mortality in (1) total hip arthroplasty (THA) and (2) total knee arthroplasty (TKA). Methods: The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database identified 363,530 patients who received a TKA or THA from 2008 to 2016. Bivariate analysis was performed to assess the association between the year in which surgery was performed and demographics and comorbidities. Bimodal multivariate logistic regression models for THA and TKA were developed for 2009-2016 using 2008 as a reference. Results: Overall incidence of VTE, DVT, PE, and mortality for THA were 0.6%, 0.4%, 0.3%, and 0.2%, respectively. Based off of multivariate regression VTE, DVT, PE, and mortality rates have shown no significant (P > .05) change from 2008 to 2016 in THA patients. Overall incidence of VTE, DVT, PE, and mortality for TKA were 1.4%, 0.9%, 0.6%, and 0.1%, respectively. Multivariate regression revealed reductions when compared to 2008 for VTEs and DVTs from 2009 to 2016 (P < .05) for TKA patients. A significant reduction in PEs (P ¼ .002) was discovered for 2016, while no significant change was observed in mortality (P > .05). Conclusion: Approximately 1 in 71 patient undergoing TKA, and 1 in 167 undergoing THA developed a VTE within 30 days after surgery. Our study demonstrated that VTE incidence rates have decreased in TKA, while remaining stable in THA over the past 8 years. Further research to determine the optimal prophylaxis algorithm that would allow for a personalized, efficacious, and safe thromboprophylaxis regimen is needed. Level of Evidence: III. © 2019 Elsevier Inc. All rights reserved.
Keywords: deep vein thrombosis pulmonary embolism venous thromboembolism total hip arthroplasty total knee arthroplasty
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 https://doi.org/10.1016/j.arth.2019.08.049. Financial Support: No financial support was used for the conductance of this study. The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. * Reprint requests: Nicolas S. Piuzzi, MD, Department of Orthopedic Surgery, Cleveland Clinic, 9500 Euclid Avenue/A41, Cleveland, OH, 44195. https://doi.org/10.1016/j.arth.2019.08.049 0883-5403/© 2019 Elsevier Inc. All rights reserved.
Venous thromboembolic events are a major cause of morbidity, mortality, and healthcare costs after lower extremity total joint arthroplasty (TJA) [1e5]. The occurrence estimates of these complications are largely drawn from large national database studies from around the world. Overall, symptomatic deep vein thrombosis (DVT) and/or pulmonary embolism (PE) have been reported to occur following approximately 0.45%-5.30% of total knee arthroplasties (TKAs) and 0.24%-1.60% of total hip arthroplasties (THAs) [2e8]. Additionally, Lee et al [4] reported an incidence of 1.9% symptomatic DVTs related to TKAs in a meta-analysis of 18 randomized control trial including 1947 patients among an Asian population (Japan, Thailand, Hong Kong, India, Malaysia, Singapore, Korea, and Taiwan). Consequently, in addition to the potentially
2
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
dreadful implications to patients, economic assessments have found that venous thromboembolisms (VTEs) place a substantial economic burden on the healthcare system. Each VTE costs US$5000 after 3 months, US$10,000 after 6 months, and US$33,000 after 1 year according to Ruppert et al [1]. As a result, there have been substantial efforts nationwide to determine and implement effective strategies to prevent VTEs after TJA. Therefore, identifying risk factors may aid in the prevention of VTEs or early intervention. Several risk factors for VTE in TJA have been determined, including, but not limited to, comorbidities, advanced age, cancer, prolonged immobilization, bilateral procedures, and longer operative times [6,9e11]. In an effort to reduce VTE incidence, risk stratification models and scoring systems have been developed such as the Caprini score [12]; however, their relevance and implementation for TJA remains controversial [8,13e15]. Strategies for VTE management should be tailored to fit individual patient’s characteristics and goals [16], and several guidelines on VTE prophylaxis are available. These include the American College of Chest Physicians (ACCP) and the American Academy of Orthopedic Surgeons guidelines, perhaps the most utilized in the United States [17e22]. To denote the relevance of guidelines, a hospital in the United Kingdom reported an 80% reduction in VTE after implementing national guidelines [23]. The 2012 versions of ACCP and American Academy of Orthopedic Surgeons guidelines are more aligned than previous iterations, and provide a balance between prophylaxis efficacy and the risk of bleeding. For example, aspirin is now considered an acceptable agent according to the latest ACCP guideline [18], and its use has increased. A recent poll of over 1200 attendees of the 2016 meeting of the American Association of Hip and Knee Surgeons, showed that over 80% orthopedic surgeons use aspirin as the main VTE prophylaxis [24]. In the same vein, a retrospective analysis in the United States reported that while healthcare provider utilization of aspirin has increased (55.6% post2012 vs pre-2012 13.9%, P < .001), there has not been an apparent change in VTE risk [25]. Additionally, over the last 5-10 years early discharge and outpatient protocols have been adopted, as well as a concomitant reduction in length of stay (LOS) [26e29]. These advances in TJA care pathways may further affect the incidence of VTE and provide further grounds for VTE incidence investigations. Although the impacts of specific prevention strategies have been investigated, there remains uncertainty regarding their overall effect on the yearly VTE incidence after TJA and its trends in a larger nationwide scale. Therefore, the purpose of this study is to identify annual trends of (1) VTE, (2) DVT, (3) PE, and (4) mortality rates, following THA and TKA in the United States. Methods Database
(CPT) codes and separated into 2 treatment groups. In the primary THA group, patients were identified using the CPT code 27130 which yielded a total of 139,155 patients. Similarly, the CPT code 27447 was used to identify 224,377 patients for the primary TKA group. Cohort sizes for each year in the different treatment groups can be found in Table 1. Measured Outcomes The postoperative outcomes of interest for the study were 30day incidences of the following complications: VTE, DVT, PE, and mortality. In the NSQIP database, DVTs are defined as a new diagnosis of a blood clot or thrombus within the superficial or deep venous system which has been confirmed by duplex, venogram, or computed tomography (CT) scan. PEs were defined as a new diagnosis of a blood clot in a pulmonary artery causing partial or complete obstruction of the lung vasculature with positive imaging (CT, CT angiogram, ventilation-perfusion scan, or any other definitive imaging modality). The incidence of VTEs was determined by the occurrence of a DVT and/or PE in the same patient. Data Analysis The annual incidence for 30-day VTEs, DVTs, PE, and mortality was calculated for each of the treatment groups from 2008 to 2016. Initially, bivariate analysis was performed between the outcomes of interest and the year of surgery with Pearson’s chi-squared tests. Additionally, to assess the association between the year of surgery with patient demographics and comorbidities, bivariate analyses were performed. Pearson’s chi-squared tests or Fisher’s exact tests were performed for categorical variables, and one-way analysis of variance tests were applied for continuous variables. Specifically, the following demographic and comorbidity variables were assessed: gender, race, inpatient status, anesthesia modality, previous sepsis, American Society of Anesthesiologists classification, diabetes mellitus, smoker within 1 year, dyspnea, functional status, history of chronic obstructive pulmonary disease, history of congestive heart failure, use of hypertension medication, renal failure, use of dialysis, disseminated cancer, wound infection, use of corticosteroids, transfusions 72 hours prior to surgery, operative time as a continuous variable, total hospital stay as a continuous variable, body mass index as a continuous variable, and age as a continuous variable. To control for potential confounding variables, patient characteristics that were identified on bivariate analysis to be significantly different across the different years in the study period were used to develop individual bimodal multivariate logistic regression models. Selected demographics and comorbidities can be found in Table 2. Of note, LOS was included to account for trends toward rapid recovery programs and faster rehabilitation, with the mean LOS being 3.8 days in 2008 and decreasing to 2.61 days in 2016 for TKAs and 4.04 days to 2.58 days for THAs. Separate models were used for primary THA and primary TKA in order to identify a given year in which surgery was performed as an independent risk factor for each complication. Odds ratios (ORs) were calculated to assess the risk of complications in each year compared to 2008, the earliest year studied. All data analyses were performed using IBM SPSS Statistics 23 for Mac (IBM Corporation, Armonk, NY) and statistical significance was maintained at an a value of less than 0.05.
The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was utilized in this study. The 2016 edition of the database consists of over 250 variables collected from more than 600 institutions including intraoperative variables, preoperative risk factors, and 30-day postoperative mortality and morbidity outcomes for several major surgeries [30,31]. Data collection is performed by trained surgical clinical reviewers and data entry is regularly audited. This database has been widely used within orthopedic surgery to conduct large-scale population-based studies [32e34].
Results
Study Population
Total Hip Arthroplasty
Patients who underwent TJA from January 1, 2008 to December 31, 2016 were identified using Current Procedural Terminology
For primary THA, the overall VTE rate was 0.6%. VTE incidence remained between 0.4% and 0.9% from 2008 to 2016 with no
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
3
Table 1 Number of Cases per Year by Procedure. Arthroplasty Type
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
Total hip arthroplasty Total knee arthroplasty
1235 2668
2571 4674
3922 7086
9395 13,650
14,485 22,472
19,618 29,857
23,066 36,055
29,522 48,937
35,341 58,978
139,155 224,377
significant changes (P > .05) (Table 3). The overall DVT rate over the study period was 0.4%, with the highest DVT rate being 0.7% in 2010, after which it remained at 0.4% from 2011 to 2016 (P > .05). The overall PE rate was 0.3%, with PE incidence after primary THA was between 0.2% and 0.4% throughout the study period (P > .05). There was no significant association between year of surgery with DVT, PE, or VTE incidence on multivariate analyses (P > .05). Overall mortality rate was 0.2% over the study period. Mortality rates following primary THA reduced from 0.6% in 2008 to 0.3% in 2016 (P ¼ .030). However, after accounting for confounding variables with multivariate analysis, there was no statistically significant difference in 30-day mortality rates in 2008 compared to any of the following years in the study period (P < .05) (Table 3). Tracked incidence/adjusted ORs for the study period can be seen in Figure 1. Total Knee Arthroplasty Overall VTE rate was 1.4%. VTE rates after primary TKA reduced significantly throughout the study period. The highest VTE incidence was in 2008 (3.0%) and the lowest VTE incidence was in 2016 (1.1%, P < .001). On multivariate analysis, it was found that primary TKA patients were 59% less likely to suffer a VTE in 2016 compared to 2008 (Table 4). The overall DVT rate over the study period was 0.9%. DVT rates after TKA reduced from 2.2% in 2008 to 0.7% in 2016 (P < .001). This association remained significant after accounting for patient characteristics with multivariate analysis. Compared to primary TKA patients in 2008, those in 2016 were 64% less likely to suffer a DVT (OR 0.361, 95% confidence interval [CI] 0.273-0.478, P < .001). Overall PE rate was 0.6%. PE incidence reduced significantly from 1.0% in 2008 to 0.4% in 2016 (P < .001). After accounting for potential confounders, the reduction in PE rates remained significant for 2016; primary TKA patients were half as likely to have a PE in 2016 compared to 2008. Overall, all-cause mortality rate was 0.1%. Mortality rates after primary TKA remained between 0.1% and 0.2% throughout the study period and no significant association between year of surgery and mortality was determined on multivariate analysis either (P > .05). Discussion VTE after lower extremity TJA is a serious and costly complication [1e5]. In an effort to decrease the burden of this complication, several preventative measures have been developed; however, their collective effect on complications rates over recent years should be assessed to help guide future strategies [17e21]. Furthermore, during the past 5-10 years aspirin as a pharmacological VTE prophylaxis agent has been reported to be safe and noninferior to other agents after TJA, which has prompted increased use in the United States [25,35e37]. During that same time period LOS has been reduced and rapid recovery and outpatient protocols have been developed [26e29]. Altogether, how these changes in practice have affected the incidence of VTE after THA and TKA at a nationwide scale has not been determined. To the best of the author’s knowledge, this is the largest, current, and nationwide report on the annual incidences and trends of DVTs, PEs, VTEs, and mortality in the 30-day postoperative period after lower limb TJAs. For primary TKAs, it was found that DVTs, PEs, and VTEs have significantly reduced from 2008 to 2016. Reductions in
these complications were also observed for primary THAs, however these were not statistically significant. The lack of reduction in THA VTE might be explained by a relatively low incidence to begin with. Despite the large study population, it is important to address the study limitations. As a result of the broad inclusion criteria, the Table 2 Demographics in THAs and TKAs. Demographic
THA (n ¼ 139,155)
TKA (n ¼ 224,337)
Age, mean ± SD BMI Length of stay Gender: female Race Native American Asian African American Native Hawaiian or Pacific Islander Not reported Caucasian Anesthesia Epidural General Local Monitored anesthesia care None Other Regional Unknown Spinal Diabetes Insulin dependent Non-insulin dependent Non-diabetic Smoker Outpatient Functional status Independent Partially dependent Totally dependent Unknown History of COPD Dyspnea At rest Moderate exertion No History of ventilator use Use of hypertension medication Acute renal failure Bleeding disorder History of transfusion Previous sepsis History of CHF Dialysis Disseminated cancer Wound infection Use of corticosteroids Weight loss ASA classification 1: no disturb 2: mild disturb 3: severe disturb 4: life threatening 5: moribund None assigned
64.79 ± 11.55 30.10 ± 6.42 2.99 ± 3.23 76,793 (55.2)
66.59 ± 9.60 33.02 ± 6.99 2.99 ± 3.23 139,145 (62.1%)
556 1926 10,133 348
(0.4%) (1.4%) (7.3%) (0.3%)
17,919 (12.9%) 108,081 (77.8%)
1139 4649 16,480 781
(0.5%) (2.1%) (7.4%) (0.3%)
27,279 (12.9%) 173,270 (77.5%)
938 73,739 17 11,331 28 81 3294 15 49,697
(0.7%) (53.0%) (<0.1%) (8.1%) (<0.1%) (0.1%) (2.4%) (<0.1%) (35.7%)
2613 112,190 53 17,683 59 201 4996 14 86,550
(1.2%) (50.0%) (<0.1%) (7.9%) (<0.1%) (0.1%) (2.2%) (<0.1%) (38.6%)
3952 12,445 122,758 18,548 1385
(2.8%) (8.9%) (88.2%) (13.3%) (1.0%)
9849 30,685 183,843 19,184 2053
(4.4%) (13.7%) (81.9%) (8.5%) (0.9%)
135,028 3386 187 554 5731
(97.0%) (2.4%) (0.1%) (0.4%) (4.1%)
219,748 3216 113 1300 8056
(97.9%) (1.4%) (0.1%) (0.6%) (3.6%)
344 6376 132,435 19 78,287 82 3517 301 652 485 374 540 690 5238 353
(0.2%) (4.6%) (95.2%) (<0.1%) (56.3%) (0.1%) (2.5%) (0.2%) (0.5%) (0.3%) (0.3%) (0.4%) (0.5%) (3.8%) (0.3%)
445 13,319 210,613 12 147,252 58 5222 117 317 620 349 239 619 7827 275
(0.2%) (5.9%) (93.9%) (<0.1%) (65.6%) (<0.1%) (2.3%) (0.1%) (0.3%) (0.3%) (0.2%) (0.1%) (0.3%) (3.5%) (0.1%)
5671 74,171 56,196 2952 7 156
(4.1%) (53.3%) (40.4%) (2.1%) (<0.1%) (0.1%)
4543 110,874 104,979 3743 7 212
(2.0%) (49.4%) (46.8) (1.7%) (<0.1%) (0.1%)
THA, total hip arthroplasty; TKA, total knee arthroplasty; SD, standard deviation; BMI, body mass index; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure; ASA, American Society of Anesthesiologists score.
4
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
Table 3 Incidence of VTE and Mortality in THAs and TKAs. Procedure
Complication
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
THA (n ¼ 139,155)
VTE DVT PE Mortality VTE DVT PE Mortality
9 7 2 8 79 59 26 3
20 10 10 5 83 54 36 5
37 27 14 12 122 79 54 14
61 41 26 23 200 130 78 22
83 59 29 28 350 204 168 24
123 83 59 42 443 293 192 31
151 92 68 43 522 323 241 43
179 114 84 50 670 397 334 47
214 139 94 74 654 442 260 63
877 572 383 285 3123 1981 1389 252
TKA (n ¼ 224,337)
(0.7%) (0.6%) (0.2%) (0.6%) (3.0%) (2.2%) (1.0%) (0.1%)
(0.8%) (0.4%) (0.4%) (0.2%) (1.8%) (1.2%) (0.8%) (0.1%)
(0.9%) (0.7%) (0.4%) (0.3%) (1.7%) (1.1%) (0.8%) (0.2%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.5%) (1.0%) (0.6%) (0.2%)
(0.6%) (0.4%) (0.2%) (0.2%) (1.6%) (0.9%) (0.7%) (0.1%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.5%) (1.0%) (0.6%) (0.1%)
(0.7%) (0.4%) (0.3%) (0.2%) (1.4%) (0.9%) (0.7%) (0.1%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.4%) (0.8%) (0.7%) (0.1%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.1%) (0.7%) (0.4%) (0.1%)
P Value (0.6%) (0.4%) (0.3%) (0.2%) (1.4%) (0.9%) (0.6%) (0.1%)
.353 .329 .612 .030 <.001 <.001 <.001 .323
THA, total hip arthroplasty; TKA, total knee arthroplasty; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism.
patients included in the analysis were those who had lower limb arthroplasty for any indication. Subgroup analysis according to index diagnoses may garner a more detailed analysis of results. Additionally, several risk factors for DVT were not available for analysis including a history of DVT/PE, family history of thrombosis, positive Factor V Leiden or other hereditary coagulopathies, blood type, bilateral procedures, and recent stroke [5,6,9,10,12]. The NSQIP database does not provide information regarding the type of VTE prophylaxis or treatment/rehabilitation protocols, which may play a role in the reduction of VTE. Nevertheless, the regression analyses included several variables which are part of the commonly used Caprini scoring system, such as age, congestive heart failure, body mass index, recent sepsis, and chronic obstructive pulmonary disease [12]. It should be noted that the number of THAs and TKAs has progressively increased from 2008 to 2016, starting with 1235 THAs and 2668 TKAs in 2008. This introduced selection bias into this study, particularly for the first 3 years included in this study. As this study was performed using a large database, the potential for coding error introduces limitations; however, this should be mitigated in part by the regular audits that are performed [30]. Furthermore, NSQIP only tracked the outcomes of interest for 30 days, whereas the ACCP suggests that patients are at an increased risk of VTE for up to 90 days [18]. Despite these limitations, this study is, to the best of the authors’ knowledge, the most recent and comprehensive nationwide evaluation of VTE trends in lower extremity TJA. Similar findings have been reported in the existing literature. Dua et al [7] conducted a retrospective database study of 550,000 TKA patients and found that the incidence of DVT has decreased from 0.86% in 2001 to 0.45% in 2011. Additionally, a decrease in DVT incidence was observed from 0.55% in 2001 to 0.24% in 2011 THA patients. Our THA VTE rates were comparable to those of Dua et al [7] while our TKA VTE rates were higher with the lowest yearly incidence being double the lowest year incidence in our study.
Notably, Dua et al [7] was conducted using National Inpatient Sample, limiting the DVTs to those detected while in hospital. A retrospective review using NSQIP analyzed trends from 2006 to 2016 30-day complications and LOS in THA by grouping the years into 3 cohorts (2006-2009 [N ¼ 3873], 2010-2013 [N ¼ 45,992], and 2014-2016 [N ¼ 86,099]) [38]. The study concluded that VTE rates in THA did not change. The same group also looked at TKAs (20062009 [N ¼ 7111], 2010-2013 [N ¼ 71,943], and 2014-2016 [N ¼,142,710]) in a similar manner and concluded that DVTs decreased in the latter 2 cohorts [39]. Although our study had similar findings of decreases in DVT for TKA and not THA, our study looked at annual trends and not somewhat arbitrary groupings of years. Furthermore, this study also investigated VTE and all-cause mortality rates in addition to PE and DVT. Similarly, a retrospective single-center study in the United Kingdom found that the rate of readmissions for VTE in the 3month postoperative period following lower limb arthroplasty reduced from 0.70% in 2005 to 0.28% in 2009 [23]. Although this study reported lower extremity TJA VTE incidence, it is roughly equivalent to our findings on incidence of THA, but much lower compared to our findings for TKA. Conversely, Pedersen et al [2] reviewed 83,756 patients, 51,002 who underwent THA and 32,754 who underwent TKA between 1997 and 2011 and found that compared to 1997, the VTE risk was greater in 2005 (relative risk 1.5, 95% CI 1.0-2.2) for a pooled TJA cohort. Additionally, the study found no significant decline in VTE risk over the study period (P ¼ .79). They reported a lower 30-day cumulative VTE incidence for TKA than our study (1.1% vs 1.4%), while the cumulative incidence for VTE in THA was higher in their study (0.9% vs 0.6%). Of note, this prior study differs from our current study in that the study population was from the Danish National Registry and, importantly, it covered a different study period that was before the most recent changes to VTE guidelines had been implemented. Additionally, the accuracy of the registry may
Fig. 1. Graphical representation of incidence of VTE and mortality for (A) THA and (B) TKA.
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
5
Table 4 Adjusted Multivariate Regression Analysis of VTE and Mortality. Outcome
VTE
DVT
PE
Mortality
Year
2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016
THA (n ¼ 139,155)a
TKA (n ¼ 224,337)b
Odds Ratio (95% CI)
P Value
Odds Ratio (95% CI)
P Value
1.011 1.278 0.868 0.781 0.888 0.932 0.856 0.850 0.645 1.183 0.747 0.719 0.772 0.742 0.706 0.729 2.333 2.134 1.653 1.191 1.817 1.874 1.822 1.638 0.399 0.816 0.573 0.505 0.596 0.580 0.444 0.597
.979 .517 .695 .485 .731 .839 .652 .634 .385 .696 .480 .413 .512 .450 .374 .417 .280 .322 .496 .812 .408 .383 .403 .491 .176 .711 .277 .173 .288 .261 .094 .276
0.624 0.590 0.510 0.553 0.540 0.531 0.503 0.408 0.527 0.516 0.420 0.421 0.466 0.431 0.391 0.361 0.881 0.787 0.663 0.825 0.743 0.776 0.790 0.509 0.970 1.766 1.521 0.999 0.985 1.208 0.894 1.056
.004 .001 <.001 <.001 <.001 <.001 <.001 <.001 .001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 .632 .343 .080 .382 .173 .239 .269 .002 .967 .374 .497 .999 .980 .752 .851 .927
(0.452-2.259) (0.609-2.681) (0.427-1.762) (0.390-1.562) (0.449-1.754) (0.474-1.835) (0.437-1.680) (0.434-1.662) (0.239-1.737) (0.508-2.756) (0.332-1.679) (0.327-1.584) (0.355-1.676) (0.343-1.607) (0.328-1.521) (0.340-1.564) (0.503-10.828) (0.476-9.562) (0.388-7.031) (0.282-5.028) (0.442-7.476) (0.458-7.677) (0.447-7.432) (0.402-6.672) (0.105-1.511) (0.278-2.395) (0.210-1.564) (0.188-1.351) (0.230-1.546) (0.225-1.498) (0.172-1.147) (0.236-1.510)
(0.454-0.859) (0.438-0.794) (0.389-0.669) (0.429-0.713) (0.421-0.693) (0.415-0.679) (0.394-0.641) (0.320-0.520) (0.361-0.770) (0.364-0.732) (0.305-0.578) (0.312-0.567) (0.324-0.573) (0.324-0.573) (0.295-0.518) (0.273-0.478) (0.524-1.481) (0.481-1.290) (0.418-1.051) (0.536-1.270) (0.485-1.139) (0.508-1.183) (0.521-1.199) (0.334-0.776) (0.231-4.068) (0.504-6.187) (0.454-5.096) (0.300-3.331) (0.300-3.236) (0.374-3.904) (0.277-2.887) (0.330-3.372)
THA, total hip arthroplasty; TKA, total knee arthroplasty; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism; CI, confidence interval; BMI, body mass index; ASA, American Society of Anesthesiologists; CHF, congestive heart failure. a The model adjusted for gender, race, inpatient status, age, BMI, dyspnea, functional status, ascites, hypertension medication, wound infection, corticosteroid use, weight loss, operative time, total length of stay, anesthesia type, previous sepsis, and ASA classification. b The model adjusted for gender, race, inpatient status, age, BMI, diabetes, smoking status, dyspnea, functional status, use of a ventilator, CHF, hypertension medication, wound infection, corticosteroid use, weight loss, bleeding disorder, transfusion, operative time, total length of stay, anesthesia type, previous sepsis, and ASA classification.
be of concern due to the issues reported with coding for VTE and major bleeding in the database [40]. Another retrospective database study by Kim et al [3] found that of the 263,664 patients treated from 2008 to 2012, there was an increase in VTE incidence among those who underwent hip fracture surgery (P ¼ .018). However, no such increase was observed for TKA and THA patients. Additionally, this study reported higher incidence rates in both THA (1.6% vs 0.6%) and TKA (1.9% vs 1.4%). Notably, this study did not report VTEs after guideline changes were made in 2012 and utilized a Korean national database, which limits the applicability to the US population. A meta-analysis of 18 studies on TKAs published from 1996 to 2011 including 1947 Asian patients found that proximal DVT rates or PEs remained similar across 3 different study periods [4]. The authors found that compared to 1993-1998, DVT incidences were similar in 1999-2002 (incidence ratio 0.82, 95% CI 0.52-1.28, P > .05) as well as in 2003-2008 (incidence ratio 0.89, 95% CI 0.55-1.43, P > .05) [4]. Similarly, no difference was observed for PEs. This study reported a higher cumulative incidence of symptomatic DVTs at 1.9% (vs 0.9%) and lower incidence of symptomatic PEs at 0.1% (vs 0.6%). Although this meta-analysis has the advantage of pooled results, it is limited to an Asian population and by the relatively small sample size of the included studies. Furthermore, this prior study does not represent the same time period investigated in our current study, nor the time period in which the more recent VTE guidelines had been implemented.
Conclusion Our study of the NSQIP database demonstrates that approximately 1 in 71 patient undergoing TKA and 1 in 167 undergoing THA developed a VTE within 30 days after surgery. Over the last decade there appears to have been significant reductions in VTE incidence for primary TKA patients. However, similar reductions were not observed among the primary THA group. These findings need to be interpreted in light of implementation of multimodal approaches for VTE prophylaxis and rapid recovery programs. Our findings may indicate that the current recommendations for VTE prophylaxis are adequate for preventing VTE. Despite this, further research to determine the optimal prophylaxis algorithm that would allow for a personalized, efficacious, and safe thromboprophylaxis regimen tailored to individual needs in the setting of medical and surgical risk factors might further decrease the incidence of VTE. References [1] Ruppert A, Steinle T, Lees M. Economic burden of venous thromboembolism: a systematic review. J Med Econ 2011;14:65e74. https://doi.org/10.3111/ 13696998.2010.546465. [2] Pedersen AB, Mehnert F, Sorensen HT, Emmeluth C, Overgaard S, Johnsen SP. The risk of venous thromboembolism, myocardial infarction, stroke, major bleeding and death in patients undergoing total hip and knee replacement: a 15-year retrospective cohort study of routine clinical practice. Bone Joint J 2014;96 B:479e85. https://doi.org/10.1302/0301-620X.96B4.33209.
6
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
[3] Kim S, Ahn H, Shin SA, Park JH, Won CW. Trends of thromboprophylaxis and complications after major lower limb orthopaedic surgeries in Korea: National Health Insurance Claim Data. Thromb Res 2017;155:48e52. https://doi.org/ 10.1016/j.thromres.2017.04.023. [4] Lee WS, Kim KI, Lee HJ, Kyung HS, Seo SS. The incidence of pulmonary embolism and deep vein thrombosis after knee arthroplasty in Asians remains low: a meta-analysis knee. Clin Orthop Relat Res 2013;471:1523e32. https:// doi.org/10.1007/s11999-012-2758-9. [5] Shahi A, Bradbury TL, Guild GN, Saleh UH, Ghanem E, Oliashirazi A. What are the incidence and risk factors of in-hospital mortality after venous thromboembolism events in total hip and knee arthroplasty patients? Arthroplast Today 2018;4:343e7. https://doi.org/10.1016/j.artd.2018.02.014. [6] Masrouha KZ, Hoballah JJ, Tamim HM, Sagherian BH. Comparing the 30-day risk of venous thromboembolism and bleeding in simultaneous bilateral vs unilateral total knee arthroplasty. J Arthroplasty 2018;33:3273e3280.e1. https://doi.org/10.1016/j.arth.2018.06.002. [7] Dua A, Desai SS, Lee CJ, Heller JA. National trends in deep vein thrombosis following total knee and total hip replacement in the United States. Ann Vasc Surg 2017;38:310e4. https://doi.org/10.1016/j.avsg.2016.05.110. [8] Bateman DK, Dow RW, Brzezinski A, Bar-Eli HY, Kayiaros ST. Correlation of the Caprini score and venous thromboembolism incidence following primary total joint arthroplastydresults of a single-institution protocol. J Arthroplasty 2017;32:3735e41. https://doi.org/10.1016/j.arth.2017.06.042. [9] Newman JM, Abola MV, Macpherson A, Klika AK, Barsoum WK, Higuera CA. ABO blood group is a predictor for the development of venous thromboembolism after total joint arthroplasty. J Arthroplasty 2017;32:S254e8. https:// doi.org/10.1016/j.arth.2017.02.063. [10] Damodar D, Donnally CJ, Sheu JI, Law TY, Roche MW, Hernandez VH. A higher altitude is an independent risk factor for venous thromboembolisms after total hip arthroplasty. J Arthroplasty 2018;33:2627e30. https://doi.org/ 10.1016/j.arth.2018.03.045. [11] Yukizawa Y, Inaba Y, Kobayashi N, Kubota S, Saito T. Current risk factors for asymptomatic venous thromboembolism in patients undergoing total hip arthroplasty. Mod Rheumatol 2018;7595:1e21. https://doi.org/10.1080/ 14397595.2018.1524959. [12] Caprini JA, Arcelus JI, Hasty JH, Tamhane AC, Fabrega F. Clinical assessment of venous thromboembolic risk in surgical patients. Semin Thromb Hemost 1991;17:304e12. [13] Pannucci CJ, Bailey SH, Dreszer G, Fisher Wachtman C, Zumsteg JW, Jaber RM, et al. Validation of the Caprini risk assessment model in plastic and reconstructive surgery patients. J Am Coll Surg 2011;212:105e12. https://doi.org/ 10.1016/j.jamcollsurg.2010.08.018. [14] Bahl V, Hu HM, Henke PK, Wakefield TW, Campbell DA, Caprini JA. A validation study of a retrospective venous thromboembolism risk scoring method. Ann Surg 2010;251:344e50. https://doi.org/10.1097/SLA.0b013e3181b7fca6. [15] Shuman AG, Hu HM, Pannucci CJ, Jackson CR, Bradford CR, Bahl V. Stratifying the risk of venous thromboembolism in otolaryngology. Otolaryngol Head Neck Surg 2012;146:719e24. https://doi.org/10.1177/0194599811434383. [16] Tritschler T, Kraaijpoel N, Le Gal G, Wells P. Venous thromboembolism advances in diagnosis and treatment. JAMA 2018;3210:1583e94. [17] Venclauskas L, Llau JV, Jenny JY, Kjaersgaard-Andersen P, Jans Ø. European guidelines on perioperative venous thromboembolism prophylaxis. Eur J Anaesthesiol 2018;35:134e8. https://doi.org/10.1097/EJA.0000000000000706. [18] Falck-Ytter Y, Francis CW, Johanson NA, Curley C, Dahl OE, Schulman S, et al. Prevention of VTE in orthopedic surgery patients. Chest 2012;141: e278Se325S. https://doi.org/10.1378/chest.11-2404. [19] Mont M, Jacobs J, Boggio LN, Bozic KJ, Della Valle CJ, Goodman SB. AAOS clinical guideline on preventing venous thromboembolic disease in patients undergoing elective hip and knee arthroplasty. Am Acad Orthop Surg 2011;19:768e76. https://doi.org/10.2106/JBJS.9408edit. [20] Venous thromboembolism in over 16s: reducing the risk of hospital-acquired deep vein thrombosis or pulmonary embolism. London: National Institutes for Health and Care Excellence; 2018. https://nice.org.uk/guidance/ng89/chapter/ Recommendations. [21] Bang S, Jang MJ, Kim KH, Yhim H. Prevention of venous thromboembolism, 2nd edition: Korean Society of Thrombosis and Hemostasis Evidence-based Clinical Practice Guidelines. J Korean Med Sci 2014;29:164e71. https:// doi.org/10.3346/jkms.2014.29.2.164.
[22] Bingham JS, Salib CG, Labban K, Morrison Z, Spangehl MJ. A dedicated anticoagulation clinic does not improve postoperative management of warfarin after total joint arthroplasty. Arthroplast Today 2018;4:340e2. https:// doi.org/10.1016/j.artd.2018.04.004. [23] Faraj AA. Implementing National Institute of Clinical Excellence guidelines for venous thromboembolism prophylaxis. Am J Med Sci 2012;343:131e5. https://doi.org/10.1097/MAJ.0b013e318226664d. [24] Azboy I, Barrack R, Thomas A, Haddad F, Parvizi J. Aspirin and the prevention of venous thromboembolism following total joint arthroplasty: commonly asked questions. Bone Joint J 2017;99-B:1420e30. https://doi.org/10.1302/ 0301-620X.99B11.BJJ-2017-0337.R2. [25] Shah SS, Satin AM, Mullen JR, Merwin S, Goldin M, Sgaglione NA. Impact of recent guideline changes on aspirin prescribing after knee arthroplasty. J Orthop Surg Res 2016;11:1e9. https://doi.org/10.1186/s13018-016-0456-0. [26] Lovett-Carter D, Sayeed Z, Abaab L, Pallekonda V, Mihalko W, Saleh KJ. Impact of outpatient total joint replacement on postoperative outcomes. Orthop Clin North Am 2018;49:35e44. https://doi.org/10.1016/j.ocl.2017.08.006. [27] Burn E, Edwards CJ, Murray DW, Silman A, Cooper C, Arden NK, et al. Trends and determinants of length of stay and hospital reimbursement following knee and hip replacement: evidence from linked primary care and NHS hospital records from 1997 to 2014. BMJ Open 2018;8:e019146. https:// doi.org/10.1136/bmjopen-2017-019146. [28] Pamilo KJ, Torkki P, Peltola M, Pesola M, Remes V, Paloneva J. Fast-tracking for total knee replacement reduces use of institutional care without compromising quality. Acta Orthop 2017;89:184e9. https://doi.org/10.1080/ 17453674.2017.1399643. [29] Vehmeijer SBW, Husted H, Kehlet H. Outpatient total hip and knee arthroplasty. Acta Orthop 2017;89:141e4. https://doi.org/10.1080/17453674.2017.1410958. [30] User Guide for the 2016 ACS NSQIP Participant use data file (PUF). 2016. http://www.facs.org/~/media/files/quality%20programs/nsqip_puf_userguide _2016.ashx. [31] Cohen ME, Ko CY, Bilimoria KY, Zhou L, Huffman K, Wang X, et al. Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg 2013;217:336e346.e1. https://doi.org/10.1016/ j.jamcollsurg.2013.02.027. [32] Smith EJ, Kuang X, Pandarinath R. Comparing hospital outcomes between open and closed tibia fractures treated with intramedullary fixation. Injury 2017;48:1609e12. https://doi.org/10.1016/j.injury.2017.03.038. [33] Phan K, Cheung ZB, Vig KS, Hussain AK, Lima MC, Kim JS, et al. Age stratification of 30-day postoperative outcomes following excisional laminectomy for extradural cervical and thoracic tumors. Glob Spine J 2018;8:490e7. https://doi.org/10.1177/2192568217745824. [34] Chona D, Lakomkin N, Bulka C, Mousavi I, Kothari P, Dodd AC, et al. Predicting the post-operative length of stay for the orthopaedic trauma patient. Int Orthop 2017;41:859e68. https://doi.org/10.1007/s00264-017-3425-2. [35] Parvizi J, Ceylan HH, Kucukdurmaz F, Merli G, Tuncay I, Beverland D. Venous thromboembolism following hip and knee arthroplasty. J Bone Joint Surg 2017;99:961e72. https://doi.org/10.2106/JBJS.16.01253. [36] Faour M, Piuzzi NS, Brigati DP, Klika AK, Mont MM, Barsoum WK, et al. No difference between low- and regular-dose aspirin for venous thromboembolism prophylaxis after THA. Clin Orthop Relat Res 2019;477:396e402. https://doi.org/10.1097/CORR.0000000000000613. [37] Faour M, Piuzzi NS, Brigati DP, Klika AK, Mont MA, Barsoum WK, et al. Lowdose aspirin is safe and effective for venous thromboembolism prophylaxis following total knee arthroplasty. J Arthroplasty 2018;33:S131e5. https:// doi.org/10.1016/j.arth.2018.03.001. [38] Grosso MJ, Neuwirth AL, Boddapati V, Shah RP, Cooper HJ, Geller JA. Decreasing length of hospital stay and postoperative complications after primary total hip arthroplasty: a decade analysis from 2006 to 2016. J Arthroplasty 2019;34:422e5. [39] Sarpong NO, Boddapati V, Herndon CL, Shah RP, Cooper HJ, Geller JA. Trends in length of stay and 30-day complications after total knee arthroplasty: an analysis from 2006 to 2016. J Arthroplasty 2019;34:1575e80. https://doi.org/ 10.1016/j.arth.2019.04.027. [40] Severinsen MT, Kristensen SR, Overvad K, Dethlefsen C, Tjønneland A, Johnsen SP. Venous thromboembolism discharge diagnoses in the Danish National Patient Registry should be used with caution. J Clin Epidemiol 2010;63:223e8. https://doi.org/10.1016/j.jclinepi.2009.03.018.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Have Venous Thromboembolism Rates Decreased in Total Hip and Knee Arthroplasty? Jared A. Warren, DO, ATC, CSCS a, Kavin Sundaram, MD, MSc a, Hiba K. Anis, MD a, Atul F. Kamath, MD a, Carlos A. Higuera, MD b, Nicolas S. Piuzzi, MD a, * a b
Department of Orthopedic Surgery, Cleveland Clinic, Cleveland, OH Department of Orthopedic Surgery, Cleveland Clinic Florida, Weston, FL
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 June 2019 Received in revised form 13 August 2019 Accepted 22 August 2019 Available online xxx
Background: Venous thromboembolism (VTE) is a major cause of morbidity, mortality, and healthcare costs in arthroplasty patients. In an effort to reduce VTEs, numerous strategies and guidelines have been implemented, but their impact remains unclear. The purpose of this study is to compare annual trends in 30-day VTE, deep vein thrombosis (DVT), pulmonary embolism (PE), and all-cause mortality in (1) total hip arthroplasty (THA) and (2) total knee arthroplasty (TKA). Methods: The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database identified 363,530 patients who received a TKA or THA from 2008 to 2016. Bivariate analysis was performed to assess the association between the year in which surgery was performed and demographics and comorbidities. Bimodal multivariate logistic regression models for THA and TKA were developed for 2009-2016 using 2008 as a reference. Results: Overall incidence of VTE, DVT, PE, and mortality for THA were 0.6%, 0.4%, 0.3%, and 0.2%, respectively. Based off of multivariate regression VTE, DVT, PE, and mortality rates have shown no significant (P > .05) change from 2008 to 2016 in THA patients. Overall incidence of VTE, DVT, PE, and mortality for TKA were 1.4%, 0.9%, 0.6%, and 0.1%, respectively. Multivariate regression revealed reductions when compared to 2008 for VTEs and DVTs from 2009 to 2016 (P < .05) for TKA patients. A significant reduction in PEs (P ¼ .002) was discovered for 2016, while no significant change was observed in mortality (P > .05). Conclusion: Approximately 1 in 71 patient undergoing TKA, and 1 in 167 undergoing THA developed a VTE within 30 days after surgery. Our study demonstrated that VTE incidence rates have decreased in TKA, while remaining stable in THA over the past 8 years. Further research to determine the optimal prophylaxis algorithm that would allow for a personalized, efficacious, and safe thromboprophylaxis regimen is needed. Level of Evidence: III. © 2019 Elsevier Inc. All rights reserved.
Keywords: deep vein thrombosis pulmonary embolism venous thromboembolism total hip arthroplasty total knee arthroplasty
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 https://doi.org/10.1016/j.arth.2019.08.049. Financial Support: No financial support was used for the conductance of this study. The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the American College of Surgeons National Surgical Quality Improvement Program are the source of the data used herein; they have not verified and are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors. * Reprint requests: Nicolas S. Piuzzi, MD, Department of Orthopedic Surgery, Cleveland Clinic, 9500 Euclid Avenue/A41, Cleveland, OH, 44195. https://doi.org/10.1016/j.arth.2019.08.049 0883-5403/© 2019 Elsevier Inc. All rights reserved.
Venous thromboembolic events are a major cause of morbidity, mortality, and healthcare costs after lower extremity total joint arthroplasty (TJA) [1e5]. The occurrence estimates of these complications are largely drawn from large national database studies from around the world. Overall, symptomatic deep vein thrombosis (DVT) and/or pulmonary embolism (PE) have been reported to occur following approximately 0.45%-5.30% of total knee arthroplasties (TKAs) and 0.24%-1.60% of total hip arthroplasties (THAs) [2e8]. Additionally, Lee et al [4] reported an incidence of 1.9% symptomatic DVTs related to TKAs in a meta-analysis of 18 randomized control trial including 1947 patients among an Asian population (Japan, Thailand, Hong Kong, India, Malaysia, Singapore, Korea, and Taiwan). Consequently, in addition to the potentially
2
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
dreadful implications to patients, economic assessments have found that venous thromboembolisms (VTEs) place a substantial economic burden on the healthcare system. Each VTE costs US$5000 after 3 months, US$10,000 after 6 months, and US$33,000 after 1 year according to Ruppert et al [1]. As a result, there have been substantial efforts nationwide to determine and implement effective strategies to prevent VTEs after TJA. Therefore, identifying risk factors may aid in the prevention of VTEs or early intervention. Several risk factors for VTE in TJA have been determined, including, but not limited to, comorbidities, advanced age, cancer, prolonged immobilization, bilateral procedures, and longer operative times [6,9e11]. In an effort to reduce VTE incidence, risk stratification models and scoring systems have been developed such as the Caprini score [12]; however, their relevance and implementation for TJA remains controversial [8,13e15]. Strategies for VTE management should be tailored to fit individual patient’s characteristics and goals [16], and several guidelines on VTE prophylaxis are available. These include the American College of Chest Physicians (ACCP) and the American Academy of Orthopedic Surgeons guidelines, perhaps the most utilized in the United States [17e22]. To denote the relevance of guidelines, a hospital in the United Kingdom reported an 80% reduction in VTE after implementing national guidelines [23]. The 2012 versions of ACCP and American Academy of Orthopedic Surgeons guidelines are more aligned than previous iterations, and provide a balance between prophylaxis efficacy and the risk of bleeding. For example, aspirin is now considered an acceptable agent according to the latest ACCP guideline [18], and its use has increased. A recent poll of over 1200 attendees of the 2016 meeting of the American Association of Hip and Knee Surgeons, showed that over 80% orthopedic surgeons use aspirin as the main VTE prophylaxis [24]. In the same vein, a retrospective analysis in the United States reported that while healthcare provider utilization of aspirin has increased (55.6% post2012 vs pre-2012 13.9%, P < .001), there has not been an apparent change in VTE risk [25]. Additionally, over the last 5-10 years early discharge and outpatient protocols have been adopted, as well as a concomitant reduction in length of stay (LOS) [26e29]. These advances in TJA care pathways may further affect the incidence of VTE and provide further grounds for VTE incidence investigations. Although the impacts of specific prevention strategies have been investigated, there remains uncertainty regarding their overall effect on the yearly VTE incidence after TJA and its trends in a larger nationwide scale. Therefore, the purpose of this study is to identify annual trends of (1) VTE, (2) DVT, (3) PE, and (4) mortality rates, following THA and TKA in the United States. Methods Database
(CPT) codes and separated into 2 treatment groups. In the primary THA group, patients were identified using the CPT code 27130 which yielded a total of 139,155 patients. Similarly, the CPT code 27447 was used to identify 224,377 patients for the primary TKA group. Cohort sizes for each year in the different treatment groups can be found in Table 1. Measured Outcomes The postoperative outcomes of interest for the study were 30day incidences of the following complications: VTE, DVT, PE, and mortality. In the NSQIP database, DVTs are defined as a new diagnosis of a blood clot or thrombus within the superficial or deep venous system which has been confirmed by duplex, venogram, or computed tomography (CT) scan. PEs were defined as a new diagnosis of a blood clot in a pulmonary artery causing partial or complete obstruction of the lung vasculature with positive imaging (CT, CT angiogram, ventilation-perfusion scan, or any other definitive imaging modality). The incidence of VTEs was determined by the occurrence of a DVT and/or PE in the same patient. Data Analysis The annual incidence for 30-day VTEs, DVTs, PE, and mortality was calculated for each of the treatment groups from 2008 to 2016. Initially, bivariate analysis was performed between the outcomes of interest and the year of surgery with Pearson’s chi-squared tests. Additionally, to assess the association between the year of surgery with patient demographics and comorbidities, bivariate analyses were performed. Pearson’s chi-squared tests or Fisher’s exact tests were performed for categorical variables, and one-way analysis of variance tests were applied for continuous variables. Specifically, the following demographic and comorbidity variables were assessed: gender, race, inpatient status, anesthesia modality, previous sepsis, American Society of Anesthesiologists classification, diabetes mellitus, smoker within 1 year, dyspnea, functional status, history of chronic obstructive pulmonary disease, history of congestive heart failure, use of hypertension medication, renal failure, use of dialysis, disseminated cancer, wound infection, use of corticosteroids, transfusions 72 hours prior to surgery, operative time as a continuous variable, total hospital stay as a continuous variable, body mass index as a continuous variable, and age as a continuous variable. To control for potential confounding variables, patient characteristics that were identified on bivariate analysis to be significantly different across the different years in the study period were used to develop individual bimodal multivariate logistic regression models. Selected demographics and comorbidities can be found in Table 2. Of note, LOS was included to account for trends toward rapid recovery programs and faster rehabilitation, with the mean LOS being 3.8 days in 2008 and decreasing to 2.61 days in 2016 for TKAs and 4.04 days to 2.58 days for THAs. Separate models were used for primary THA and primary TKA in order to identify a given year in which surgery was performed as an independent risk factor for each complication. Odds ratios (ORs) were calculated to assess the risk of complications in each year compared to 2008, the earliest year studied. All data analyses were performed using IBM SPSS Statistics 23 for Mac (IBM Corporation, Armonk, NY) and statistical significance was maintained at an a value of less than 0.05.
The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was utilized in this study. The 2016 edition of the database consists of over 250 variables collected from more than 600 institutions including intraoperative variables, preoperative risk factors, and 30-day postoperative mortality and morbidity outcomes for several major surgeries [30,31]. Data collection is performed by trained surgical clinical reviewers and data entry is regularly audited. This database has been widely used within orthopedic surgery to conduct large-scale population-based studies [32e34].
Results
Study Population
Total Hip Arthroplasty
Patients who underwent TJA from January 1, 2008 to December 31, 2016 were identified using Current Procedural Terminology
For primary THA, the overall VTE rate was 0.6%. VTE incidence remained between 0.4% and 0.9% from 2008 to 2016 with no
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
3
Table 1 Number of Cases per Year by Procedure. Arthroplasty Type
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
Total hip arthroplasty Total knee arthroplasty
1235 2668
2571 4674
3922 7086
9395 13,650
14,485 22,472
19,618 29,857
23,066 36,055
29,522 48,937
35,341 58,978
139,155 224,377
significant changes (P > .05) (Table 3). The overall DVT rate over the study period was 0.4%, with the highest DVT rate being 0.7% in 2010, after which it remained at 0.4% from 2011 to 2016 (P > .05). The overall PE rate was 0.3%, with PE incidence after primary THA was between 0.2% and 0.4% throughout the study period (P > .05). There was no significant association between year of surgery with DVT, PE, or VTE incidence on multivariate analyses (P > .05). Overall mortality rate was 0.2% over the study period. Mortality rates following primary THA reduced from 0.6% in 2008 to 0.3% in 2016 (P ¼ .030). However, after accounting for confounding variables with multivariate analysis, there was no statistically significant difference in 30-day mortality rates in 2008 compared to any of the following years in the study period (P < .05) (Table 3). Tracked incidence/adjusted ORs for the study period can be seen in Figure 1. Total Knee Arthroplasty Overall VTE rate was 1.4%. VTE rates after primary TKA reduced significantly throughout the study period. The highest VTE incidence was in 2008 (3.0%) and the lowest VTE incidence was in 2016 (1.1%, P < .001). On multivariate analysis, it was found that primary TKA patients were 59% less likely to suffer a VTE in 2016 compared to 2008 (Table 4). The overall DVT rate over the study period was 0.9%. DVT rates after TKA reduced from 2.2% in 2008 to 0.7% in 2016 (P < .001). This association remained significant after accounting for patient characteristics with multivariate analysis. Compared to primary TKA patients in 2008, those in 2016 were 64% less likely to suffer a DVT (OR 0.361, 95% confidence interval [CI] 0.273-0.478, P < .001). Overall PE rate was 0.6%. PE incidence reduced significantly from 1.0% in 2008 to 0.4% in 2016 (P < .001). After accounting for potential confounders, the reduction in PE rates remained significant for 2016; primary TKA patients were half as likely to have a PE in 2016 compared to 2008. Overall, all-cause mortality rate was 0.1%. Mortality rates after primary TKA remained between 0.1% and 0.2% throughout the study period and no significant association between year of surgery and mortality was determined on multivariate analysis either (P > .05). Discussion VTE after lower extremity TJA is a serious and costly complication [1e5]. In an effort to decrease the burden of this complication, several preventative measures have been developed; however, their collective effect on complications rates over recent years should be assessed to help guide future strategies [17e21]. Furthermore, during the past 5-10 years aspirin as a pharmacological VTE prophylaxis agent has been reported to be safe and noninferior to other agents after TJA, which has prompted increased use in the United States [25,35e37]. During that same time period LOS has been reduced and rapid recovery and outpatient protocols have been developed [26e29]. Altogether, how these changes in practice have affected the incidence of VTE after THA and TKA at a nationwide scale has not been determined. To the best of the author’s knowledge, this is the largest, current, and nationwide report on the annual incidences and trends of DVTs, PEs, VTEs, and mortality in the 30-day postoperative period after lower limb TJAs. For primary TKAs, it was found that DVTs, PEs, and VTEs have significantly reduced from 2008 to 2016. Reductions in
these complications were also observed for primary THAs, however these were not statistically significant. The lack of reduction in THA VTE might be explained by a relatively low incidence to begin with. Despite the large study population, it is important to address the study limitations. As a result of the broad inclusion criteria, the Table 2 Demographics in THAs and TKAs. Demographic
THA (n ¼ 139,155)
TKA (n ¼ 224,337)
Age, mean ± SD BMI Length of stay Gender: female Race Native American Asian African American Native Hawaiian or Pacific Islander Not reported Caucasian Anesthesia Epidural General Local Monitored anesthesia care None Other Regional Unknown Spinal Diabetes Insulin dependent Non-insulin dependent Non-diabetic Smoker Outpatient Functional status Independent Partially dependent Totally dependent Unknown History of COPD Dyspnea At rest Moderate exertion No History of ventilator use Use of hypertension medication Acute renal failure Bleeding disorder History of transfusion Previous sepsis History of CHF Dialysis Disseminated cancer Wound infection Use of corticosteroids Weight loss ASA classification 1: no disturb 2: mild disturb 3: severe disturb 4: life threatening 5: moribund None assigned
64.79 ± 11.55 30.10 ± 6.42 2.99 ± 3.23 76,793 (55.2)
66.59 ± 9.60 33.02 ± 6.99 2.99 ± 3.23 139,145 (62.1%)
556 1926 10,133 348
(0.4%) (1.4%) (7.3%) (0.3%)
17,919 (12.9%) 108,081 (77.8%)
1139 4649 16,480 781
(0.5%) (2.1%) (7.4%) (0.3%)
27,279 (12.9%) 173,270 (77.5%)
938 73,739 17 11,331 28 81 3294 15 49,697
(0.7%) (53.0%) (<0.1%) (8.1%) (<0.1%) (0.1%) (2.4%) (<0.1%) (35.7%)
2613 112,190 53 17,683 59 201 4996 14 86,550
(1.2%) (50.0%) (<0.1%) (7.9%) (<0.1%) (0.1%) (2.2%) (<0.1%) (38.6%)
3952 12,445 122,758 18,548 1385
(2.8%) (8.9%) (88.2%) (13.3%) (1.0%)
9849 30,685 183,843 19,184 2053
(4.4%) (13.7%) (81.9%) (8.5%) (0.9%)
135,028 3386 187 554 5731
(97.0%) (2.4%) (0.1%) (0.4%) (4.1%)
219,748 3216 113 1300 8056
(97.9%) (1.4%) (0.1%) (0.6%) (3.6%)
344 6376 132,435 19 78,287 82 3517 301 652 485 374 540 690 5238 353
(0.2%) (4.6%) (95.2%) (<0.1%) (56.3%) (0.1%) (2.5%) (0.2%) (0.5%) (0.3%) (0.3%) (0.4%) (0.5%) (3.8%) (0.3%)
445 13,319 210,613 12 147,252 58 5222 117 317 620 349 239 619 7827 275
(0.2%) (5.9%) (93.9%) (<0.1%) (65.6%) (<0.1%) (2.3%) (0.1%) (0.3%) (0.3%) (0.2%) (0.1%) (0.3%) (3.5%) (0.1%)
5671 74,171 56,196 2952 7 156
(4.1%) (53.3%) (40.4%) (2.1%) (<0.1%) (0.1%)
4543 110,874 104,979 3743 7 212
(2.0%) (49.4%) (46.8) (1.7%) (<0.1%) (0.1%)
THA, total hip arthroplasty; TKA, total knee arthroplasty; SD, standard deviation; BMI, body mass index; COPD, chronic obstructive pulmonary disease; CHF, congestive heart failure; ASA, American Society of Anesthesiologists score.
4
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
Table 3 Incidence of VTE and Mortality in THAs and TKAs. Procedure
Complication
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
THA (n ¼ 139,155)
VTE DVT PE Mortality VTE DVT PE Mortality
9 7 2 8 79 59 26 3
20 10 10 5 83 54 36 5
37 27 14 12 122 79 54 14
61 41 26 23 200 130 78 22
83 59 29 28 350 204 168 24
123 83 59 42 443 293 192 31
151 92 68 43 522 323 241 43
179 114 84 50 670 397 334 47
214 139 94 74 654 442 260 63
877 572 383 285 3123 1981 1389 252
TKA (n ¼ 224,337)
(0.7%) (0.6%) (0.2%) (0.6%) (3.0%) (2.2%) (1.0%) (0.1%)
(0.8%) (0.4%) (0.4%) (0.2%) (1.8%) (1.2%) (0.8%) (0.1%)
(0.9%) (0.7%) (0.4%) (0.3%) (1.7%) (1.1%) (0.8%) (0.2%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.5%) (1.0%) (0.6%) (0.2%)
(0.6%) (0.4%) (0.2%) (0.2%) (1.6%) (0.9%) (0.7%) (0.1%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.5%) (1.0%) (0.6%) (0.1%)
(0.7%) (0.4%) (0.3%) (0.2%) (1.4%) (0.9%) (0.7%) (0.1%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.4%) (0.8%) (0.7%) (0.1%)
(0.6%) (0.4%) (0.3%) (0.2%) (1.1%) (0.7%) (0.4%) (0.1%)
P Value (0.6%) (0.4%) (0.3%) (0.2%) (1.4%) (0.9%) (0.6%) (0.1%)
.353 .329 .612 .030 <.001 <.001 <.001 .323
THA, total hip arthroplasty; TKA, total knee arthroplasty; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism.
patients included in the analysis were those who had lower limb arthroplasty for any indication. Subgroup analysis according to index diagnoses may garner a more detailed analysis of results. Additionally, several risk factors for DVT were not available for analysis including a history of DVT/PE, family history of thrombosis, positive Factor V Leiden or other hereditary coagulopathies, blood type, bilateral procedures, and recent stroke [5,6,9,10,12]. The NSQIP database does not provide information regarding the type of VTE prophylaxis or treatment/rehabilitation protocols, which may play a role in the reduction of VTE. Nevertheless, the regression analyses included several variables which are part of the commonly used Caprini scoring system, such as age, congestive heart failure, body mass index, recent sepsis, and chronic obstructive pulmonary disease [12]. It should be noted that the number of THAs and TKAs has progressively increased from 2008 to 2016, starting with 1235 THAs and 2668 TKAs in 2008. This introduced selection bias into this study, particularly for the first 3 years included in this study. As this study was performed using a large database, the potential for coding error introduces limitations; however, this should be mitigated in part by the regular audits that are performed [30]. Furthermore, NSQIP only tracked the outcomes of interest for 30 days, whereas the ACCP suggests that patients are at an increased risk of VTE for up to 90 days [18]. Despite these limitations, this study is, to the best of the authors’ knowledge, the most recent and comprehensive nationwide evaluation of VTE trends in lower extremity TJA. Similar findings have been reported in the existing literature. Dua et al [7] conducted a retrospective database study of 550,000 TKA patients and found that the incidence of DVT has decreased from 0.86% in 2001 to 0.45% in 2011. Additionally, a decrease in DVT incidence was observed from 0.55% in 2001 to 0.24% in 2011 THA patients. Our THA VTE rates were comparable to those of Dua et al [7] while our TKA VTE rates were higher with the lowest yearly incidence being double the lowest year incidence in our study.
Notably, Dua et al [7] was conducted using National Inpatient Sample, limiting the DVTs to those detected while in hospital. A retrospective review using NSQIP analyzed trends from 2006 to 2016 30-day complications and LOS in THA by grouping the years into 3 cohorts (2006-2009 [N ¼ 3873], 2010-2013 [N ¼ 45,992], and 2014-2016 [N ¼ 86,099]) [38]. The study concluded that VTE rates in THA did not change. The same group also looked at TKAs (20062009 [N ¼ 7111], 2010-2013 [N ¼ 71,943], and 2014-2016 [N ¼,142,710]) in a similar manner and concluded that DVTs decreased in the latter 2 cohorts [39]. Although our study had similar findings of decreases in DVT for TKA and not THA, our study looked at annual trends and not somewhat arbitrary groupings of years. Furthermore, this study also investigated VTE and all-cause mortality rates in addition to PE and DVT. Similarly, a retrospective single-center study in the United Kingdom found that the rate of readmissions for VTE in the 3month postoperative period following lower limb arthroplasty reduced from 0.70% in 2005 to 0.28% in 2009 [23]. Although this study reported lower extremity TJA VTE incidence, it is roughly equivalent to our findings on incidence of THA, but much lower compared to our findings for TKA. Conversely, Pedersen et al [2] reviewed 83,756 patients, 51,002 who underwent THA and 32,754 who underwent TKA between 1997 and 2011 and found that compared to 1997, the VTE risk was greater in 2005 (relative risk 1.5, 95% CI 1.0-2.2) for a pooled TJA cohort. Additionally, the study found no significant decline in VTE risk over the study period (P ¼ .79). They reported a lower 30-day cumulative VTE incidence for TKA than our study (1.1% vs 1.4%), while the cumulative incidence for VTE in THA was higher in their study (0.9% vs 0.6%). Of note, this prior study differs from our current study in that the study population was from the Danish National Registry and, importantly, it covered a different study period that was before the most recent changes to VTE guidelines had been implemented. Additionally, the accuracy of the registry may
Fig. 1. Graphical representation of incidence of VTE and mortality for (A) THA and (B) TKA.
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
5
Table 4 Adjusted Multivariate Regression Analysis of VTE and Mortality. Outcome
VTE
DVT
PE
Mortality
Year
2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016 2009 2010 2011 2012 2013 2014 2015 2016
THA (n ¼ 139,155)a
TKA (n ¼ 224,337)b
Odds Ratio (95% CI)
P Value
Odds Ratio (95% CI)
P Value
1.011 1.278 0.868 0.781 0.888 0.932 0.856 0.850 0.645 1.183 0.747 0.719 0.772 0.742 0.706 0.729 2.333 2.134 1.653 1.191 1.817 1.874 1.822 1.638 0.399 0.816 0.573 0.505 0.596 0.580 0.444 0.597
.979 .517 .695 .485 .731 .839 .652 .634 .385 .696 .480 .413 .512 .450 .374 .417 .280 .322 .496 .812 .408 .383 .403 .491 .176 .711 .277 .173 .288 .261 .094 .276
0.624 0.590 0.510 0.553 0.540 0.531 0.503 0.408 0.527 0.516 0.420 0.421 0.466 0.431 0.391 0.361 0.881 0.787 0.663 0.825 0.743 0.776 0.790 0.509 0.970 1.766 1.521 0.999 0.985 1.208 0.894 1.056
.004 .001 <.001 <.001 <.001 <.001 <.001 <.001 .001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 .632 .343 .080 .382 .173 .239 .269 .002 .967 .374 .497 .999 .980 .752 .851 .927
(0.452-2.259) (0.609-2.681) (0.427-1.762) (0.390-1.562) (0.449-1.754) (0.474-1.835) (0.437-1.680) (0.434-1.662) (0.239-1.737) (0.508-2.756) (0.332-1.679) (0.327-1.584) (0.355-1.676) (0.343-1.607) (0.328-1.521) (0.340-1.564) (0.503-10.828) (0.476-9.562) (0.388-7.031) (0.282-5.028) (0.442-7.476) (0.458-7.677) (0.447-7.432) (0.402-6.672) (0.105-1.511) (0.278-2.395) (0.210-1.564) (0.188-1.351) (0.230-1.546) (0.225-1.498) (0.172-1.147) (0.236-1.510)
(0.454-0.859) (0.438-0.794) (0.389-0.669) (0.429-0.713) (0.421-0.693) (0.415-0.679) (0.394-0.641) (0.320-0.520) (0.361-0.770) (0.364-0.732) (0.305-0.578) (0.312-0.567) (0.324-0.573) (0.324-0.573) (0.295-0.518) (0.273-0.478) (0.524-1.481) (0.481-1.290) (0.418-1.051) (0.536-1.270) (0.485-1.139) (0.508-1.183) (0.521-1.199) (0.334-0.776) (0.231-4.068) (0.504-6.187) (0.454-5.096) (0.300-3.331) (0.300-3.236) (0.374-3.904) (0.277-2.887) (0.330-3.372)
THA, total hip arthroplasty; TKA, total knee arthroplasty; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism; CI, confidence interval; BMI, body mass index; ASA, American Society of Anesthesiologists; CHF, congestive heart failure. a The model adjusted for gender, race, inpatient status, age, BMI, dyspnea, functional status, ascites, hypertension medication, wound infection, corticosteroid use, weight loss, operative time, total length of stay, anesthesia type, previous sepsis, and ASA classification. b The model adjusted for gender, race, inpatient status, age, BMI, diabetes, smoking status, dyspnea, functional status, use of a ventilator, CHF, hypertension medication, wound infection, corticosteroid use, weight loss, bleeding disorder, transfusion, operative time, total length of stay, anesthesia type, previous sepsis, and ASA classification.
be of concern due to the issues reported with coding for VTE and major bleeding in the database [40]. Another retrospective database study by Kim et al [3] found that of the 263,664 patients treated from 2008 to 2012, there was an increase in VTE incidence among those who underwent hip fracture surgery (P ¼ .018). However, no such increase was observed for TKA and THA patients. Additionally, this study reported higher incidence rates in both THA (1.6% vs 0.6%) and TKA (1.9% vs 1.4%). Notably, this study did not report VTEs after guideline changes were made in 2012 and utilized a Korean national database, which limits the applicability to the US population. A meta-analysis of 18 studies on TKAs published from 1996 to 2011 including 1947 Asian patients found that proximal DVT rates or PEs remained similar across 3 different study periods [4]. The authors found that compared to 1993-1998, DVT incidences were similar in 1999-2002 (incidence ratio 0.82, 95% CI 0.52-1.28, P > .05) as well as in 2003-2008 (incidence ratio 0.89, 95% CI 0.55-1.43, P > .05) [4]. Similarly, no difference was observed for PEs. This study reported a higher cumulative incidence of symptomatic DVTs at 1.9% (vs 0.9%) and lower incidence of symptomatic PEs at 0.1% (vs 0.6%). Although this meta-analysis has the advantage of pooled results, it is limited to an Asian population and by the relatively small sample size of the included studies. Furthermore, this prior study does not represent the same time period investigated in our current study, nor the time period in which the more recent VTE guidelines had been implemented.
Conclusion Our study of the NSQIP database demonstrates that approximately 1 in 71 patient undergoing TKA and 1 in 167 undergoing THA developed a VTE within 30 days after surgery. Over the last decade there appears to have been significant reductions in VTE incidence for primary TKA patients. However, similar reductions were not observed among the primary THA group. These findings need to be interpreted in light of implementation of multimodal approaches for VTE prophylaxis and rapid recovery programs. Our findings may indicate that the current recommendations for VTE prophylaxis are adequate for preventing VTE. Despite this, further research to determine the optimal prophylaxis algorithm that would allow for a personalized, efficacious, and safe thromboprophylaxis regimen tailored to individual needs in the setting of medical and surgical risk factors might further decrease the incidence of VTE. References [1] Ruppert A, Steinle T, Lees M. Economic burden of venous thromboembolism: a systematic review. J Med Econ 2011;14:65e74. https://doi.org/10.3111/ 13696998.2010.546465. [2] Pedersen AB, Mehnert F, Sorensen HT, Emmeluth C, Overgaard S, Johnsen SP. The risk of venous thromboembolism, myocardial infarction, stroke, major bleeding and death in patients undergoing total hip and knee replacement: a 15-year retrospective cohort study of routine clinical practice. Bone Joint J 2014;96 B:479e85. https://doi.org/10.1302/0301-620X.96B4.33209.
6
J.A. Warren et al. / The Journal of Arthroplasty xxx (2019) 1e6
[3] Kim S, Ahn H, Shin SA, Park JH, Won CW. Trends of thromboprophylaxis and complications after major lower limb orthopaedic surgeries in Korea: National Health Insurance Claim Data. Thromb Res 2017;155:48e52. https://doi.org/ 10.1016/j.thromres.2017.04.023. [4] Lee WS, Kim KI, Lee HJ, Kyung HS, Seo SS. The incidence of pulmonary embolism and deep vein thrombosis after knee arthroplasty in Asians remains low: a meta-analysis knee. Clin Orthop Relat Res 2013;471:1523e32. https:// doi.org/10.1007/s11999-012-2758-9. [5] Shahi A, Bradbury TL, Guild GN, Saleh UH, Ghanem E, Oliashirazi A. What are the incidence and risk factors of in-hospital mortality after venous thromboembolism events in total hip and knee arthroplasty patients? Arthroplast Today 2018;4:343e7. https://doi.org/10.1016/j.artd.2018.02.014. [6] Masrouha KZ, Hoballah JJ, Tamim HM, Sagherian BH. Comparing the 30-day risk of venous thromboembolism and bleeding in simultaneous bilateral vs unilateral total knee arthroplasty. J Arthroplasty 2018;33:3273e3280.e1. https://doi.org/10.1016/j.arth.2018.06.002. [7] Dua A, Desai SS, Lee CJ, Heller JA. National trends in deep vein thrombosis following total knee and total hip replacement in the United States. Ann Vasc Surg 2017;38:310e4. https://doi.org/10.1016/j.avsg.2016.05.110. [8] Bateman DK, Dow RW, Brzezinski A, Bar-Eli HY, Kayiaros ST. Correlation of the Caprini score and venous thromboembolism incidence following primary total joint arthroplastydresults of a single-institution protocol. J Arthroplasty 2017;32:3735e41. https://doi.org/10.1016/j.arth.2017.06.042. [9] Newman JM, Abola MV, Macpherson A, Klika AK, Barsoum WK, Higuera CA. ABO blood group is a predictor for the development of venous thromboembolism after total joint arthroplasty. J Arthroplasty 2017;32:S254e8. https:// doi.org/10.1016/j.arth.2017.02.063. [10] Damodar D, Donnally CJ, Sheu JI, Law TY, Roche MW, Hernandez VH. A higher altitude is an independent risk factor for venous thromboembolisms after total hip arthroplasty. J Arthroplasty 2018;33:2627e30. https://doi.org/ 10.1016/j.arth.2018.03.045. [11] Yukizawa Y, Inaba Y, Kobayashi N, Kubota S, Saito T. Current risk factors for asymptomatic venous thromboembolism in patients undergoing total hip arthroplasty. Mod Rheumatol 2018;7595:1e21. https://doi.org/10.1080/ 14397595.2018.1524959. [12] Caprini JA, Arcelus JI, Hasty JH, Tamhane AC, Fabrega F. Clinical assessment of venous thromboembolic risk in surgical patients. Semin Thromb Hemost 1991;17:304e12. [13] Pannucci CJ, Bailey SH, Dreszer G, Fisher Wachtman C, Zumsteg JW, Jaber RM, et al. Validation of the Caprini risk assessment model in plastic and reconstructive surgery patients. J Am Coll Surg 2011;212:105e12. https://doi.org/ 10.1016/j.jamcollsurg.2010.08.018. [14] Bahl V, Hu HM, Henke PK, Wakefield TW, Campbell DA, Caprini JA. A validation study of a retrospective venous thromboembolism risk scoring method. Ann Surg 2010;251:344e50. https://doi.org/10.1097/SLA.0b013e3181b7fca6. [15] Shuman AG, Hu HM, Pannucci CJ, Jackson CR, Bradford CR, Bahl V. Stratifying the risk of venous thromboembolism in otolaryngology. Otolaryngol Head Neck Surg 2012;146:719e24. https://doi.org/10.1177/0194599811434383. [16] Tritschler T, Kraaijpoel N, Le Gal G, Wells P. Venous thromboembolism advances in diagnosis and treatment. JAMA 2018;3210:1583e94. [17] Venclauskas L, Llau JV, Jenny JY, Kjaersgaard-Andersen P, Jans Ø. European guidelines on perioperative venous thromboembolism prophylaxis. Eur J Anaesthesiol 2018;35:134e8. https://doi.org/10.1097/EJA.0000000000000706. [18] Falck-Ytter Y, Francis CW, Johanson NA, Curley C, Dahl OE, Schulman S, et al. Prevention of VTE in orthopedic surgery patients. Chest 2012;141: e278Se325S. https://doi.org/10.1378/chest.11-2404. [19] Mont M, Jacobs J, Boggio LN, Bozic KJ, Della Valle CJ, Goodman SB. AAOS clinical guideline on preventing venous thromboembolic disease in patients undergoing elective hip and knee arthroplasty. Am Acad Orthop Surg 2011;19:768e76. https://doi.org/10.2106/JBJS.9408edit. [20] Venous thromboembolism in over 16s: reducing the risk of hospital-acquired deep vein thrombosis or pulmonary embolism. London: National Institutes for Health and Care Excellence; 2018. https://nice.org.uk/guidance/ng89/chapter/ Recommendations. [21] Bang S, Jang MJ, Kim KH, Yhim H. Prevention of venous thromboembolism, 2nd edition: Korean Society of Thrombosis and Hemostasis Evidence-based Clinical Practice Guidelines. J Korean Med Sci 2014;29:164e71. https:// doi.org/10.3346/jkms.2014.29.2.164.
[22] Bingham JS, Salib CG, Labban K, Morrison Z, Spangehl MJ. A dedicated anticoagulation clinic does not improve postoperative management of warfarin after total joint arthroplasty. Arthroplast Today 2018;4:340e2. https:// doi.org/10.1016/j.artd.2018.04.004. [23] Faraj AA. Implementing National Institute of Clinical Excellence guidelines for venous thromboembolism prophylaxis. Am J Med Sci 2012;343:131e5. https://doi.org/10.1097/MAJ.0b013e318226664d. [24] Azboy I, Barrack R, Thomas A, Haddad F, Parvizi J. Aspirin and the prevention of venous thromboembolism following total joint arthroplasty: commonly asked questions. Bone Joint J 2017;99-B:1420e30. https://doi.org/10.1302/ 0301-620X.99B11.BJJ-2017-0337.R2. [25] Shah SS, Satin AM, Mullen JR, Merwin S, Goldin M, Sgaglione NA. Impact of recent guideline changes on aspirin prescribing after knee arthroplasty. J Orthop Surg Res 2016;11:1e9. https://doi.org/10.1186/s13018-016-0456-0. [26] Lovett-Carter D, Sayeed Z, Abaab L, Pallekonda V, Mihalko W, Saleh KJ. Impact of outpatient total joint replacement on postoperative outcomes. Orthop Clin North Am 2018;49:35e44. https://doi.org/10.1016/j.ocl.2017.08.006. [27] Burn E, Edwards CJ, Murray DW, Silman A, Cooper C, Arden NK, et al. Trends and determinants of length of stay and hospital reimbursement following knee and hip replacement: evidence from linked primary care and NHS hospital records from 1997 to 2014. BMJ Open 2018;8:e019146. https:// doi.org/10.1136/bmjopen-2017-019146. [28] Pamilo KJ, Torkki P, Peltola M, Pesola M, Remes V, Paloneva J. Fast-tracking for total knee replacement reduces use of institutional care without compromising quality. Acta Orthop 2017;89:184e9. https://doi.org/10.1080/ 17453674.2017.1399643. [29] Vehmeijer SBW, Husted H, Kehlet H. Outpatient total hip and knee arthroplasty. Acta Orthop 2017;89:141e4. https://doi.org/10.1080/17453674.2017.1410958. [30] User Guide for the 2016 ACS NSQIP Participant use data file (PUF). 2016. http://www.facs.org/~/media/files/quality%20programs/nsqip_puf_userguide _2016.ashx. [31] Cohen ME, Ko CY, Bilimoria KY, Zhou L, Huffman K, Wang X, et al. Optimizing ACS NSQIP modeling for evaluation of surgical quality and risk: patient risk adjustment, procedure mix adjustment, shrinkage adjustment, and surgical focus. J Am Coll Surg 2013;217:336e346.e1. https://doi.org/10.1016/ j.jamcollsurg.2013.02.027. [32] Smith EJ, Kuang X, Pandarinath R. Comparing hospital outcomes between open and closed tibia fractures treated with intramedullary fixation. Injury 2017;48:1609e12. https://doi.org/10.1016/j.injury.2017.03.038. [33] Phan K, Cheung ZB, Vig KS, Hussain AK, Lima MC, Kim JS, et al. Age stratification of 30-day postoperative outcomes following excisional laminectomy for extradural cervical and thoracic tumors. Glob Spine J 2018;8:490e7. https://doi.org/10.1177/2192568217745824. [34] Chona D, Lakomkin N, Bulka C, Mousavi I, Kothari P, Dodd AC, et al. Predicting the post-operative length of stay for the orthopaedic trauma patient. Int Orthop 2017;41:859e68. https://doi.org/10.1007/s00264-017-3425-2. [35] Parvizi J, Ceylan HH, Kucukdurmaz F, Merli G, Tuncay I, Beverland D. Venous thromboembolism following hip and knee arthroplasty. J Bone Joint Surg 2017;99:961e72. https://doi.org/10.2106/JBJS.16.01253. [36] Faour M, Piuzzi NS, Brigati DP, Klika AK, Mont MM, Barsoum WK, et al. No difference between low- and regular-dose aspirin for venous thromboembolism prophylaxis after THA. Clin Orthop Relat Res 2019;477:396e402. https://doi.org/10.1097/CORR.0000000000000613. [37] Faour M, Piuzzi NS, Brigati DP, Klika AK, Mont MA, Barsoum WK, et al. Lowdose aspirin is safe and effective for venous thromboembolism prophylaxis following total knee arthroplasty. J Arthroplasty 2018;33:S131e5. https:// doi.org/10.1016/j.arth.2018.03.001. [38] Grosso MJ, Neuwirth AL, Boddapati V, Shah RP, Cooper HJ, Geller JA. Decreasing length of hospital stay and postoperative complications after primary total hip arthroplasty: a decade analysis from 2006 to 2016. J Arthroplasty 2019;34:422e5. [39] Sarpong NO, Boddapati V, Herndon CL, Shah RP, Cooper HJ, Geller JA. Trends in length of stay and 30-day complications after total knee arthroplasty: an analysis from 2006 to 2016. J Arthroplasty 2019;34:1575e80. https://doi.org/ 10.1016/j.arth.2019.04.027. [40] Severinsen MT, Kristensen SR, Overvad K, Dethlefsen C, Tjønneland A, Johnsen SP. Venous thromboembolism discharge diagnoses in the Danish National Patient Registry should be used with caution. J Clin Epidemiol 2010;63:223e8. https://doi.org/10.1016/j.jclinepi.2009.03.018.