- Email: [email protected]
Incidence and Risk Factors for Venous Thromboembolism Following Hip Arthroscopy: A Population-Based Study
Incidence and Risk Factors for Venous Thromboembolism Following Hip Arthroscopy: A Population-Based Study Zain M. Khazi, B.S., Qiang An, M.B.B.S., M.P.H., Kyle R. Duchman, M.D., and Robert W. Westermann, M.D.
Purpose: To determine the incidence of symptomatic venous thromboembolism (VTE) after hip arthroscopy (HA) using a large national database while considering several patient demographic factors. Methods: Patients 20 years old who underwent HA between 2007 and 2017 were identified within the Humana administrative claims database using relevant Current Procedural Terminology and International Classification of Diseases Ninth and Tenth Revision codes. Basic demographics, including age, gender, obesity (body mass index 30 kg/m2), oral contraceptive use, smoking history, diabetes, and chronic obstructive pulmonary disease (CLD) were recorded. Postoperative incidence of deep vein thrombosis, pulmonary embolism, and VTE was identified at 30 and 90 days postoperatively. Multivariate logistic regression analysis was performed to identify independent risk factors for VTE after HA, with statistical significance set at P < .05. Results: Overall, 9,477 patients underwent HA procedures over the study period, of whom 5,085 (53.7%) were female. The overall incidence of VTE in all patients was 0.77% (n ¼ 73) and 1.14% (n ¼ 108) at 30 and 90 days, respectively. Multivariate analysis identified age 45 (odds ratio [OR] ¼ 1.82; 95% confidence interval [CI], 1.36-2.49; P ¼ .0001), obesity (OR ¼ 1.54; 95% CI, 1.27-1.86; P < .0001), smoking (OR ¼ 1.26; 95% CI, 1.04-1.53; P ¼ .0177), diabetes (OR ¼ 1.59; 95% CI, 1.32-1.92; P < .0001), and CLD (OR ¼ 2.10; 95% CI, 1.63-2.68; P < .0001) as independent risk factors for higher incidence of VTE after HA. However, neither gender nor oral contraceptive use were risk factors for VTE after HA. Conclusions: For patients undergoing HA, the incidence of symptomatic postoperative VTE is low. This study identified age 45, obesity, tobacco use, diabetes, and CLD as independent risk factors for VTE after HA. Level of Evidence: Level III, retrospective cohort study.
V
enous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a relatively rare complication following elective orthopaedic procedures that may result in significant morbidity and mortality.1,2 The development of VTE is a complex pathophysiological process that is typically secondary to factors characterized by Virchow’s triad
Division of Sports Medicine, Department of Orthopedics and Rehabilitation, University of Iowa Hospitals and Clinics, Iowa City, Iowa, U.S.A. The authors report the following potential conflicts of interest or sources of funding: R.W.W. receives research support from Smith & Nephew. Full ICMJE author disclosure forms are available for this article online, as supplementary material. Received December 19, 2018; accepted March 24, 2019. Address correspondence to Robert W. Westermann, M.D., 2701 Prairie Meadow Dr, Iowa City, IA 52246, U.S.A. E-mail: robert-westermann@ uiowa.edu Ó 2019 Published by Elsevier on behalf of the Arthroscopy Association of North America 0749-8063/181541/$36.00 https://doi.org/10.1016/j.arthro.2019.03.054
2380
(venous stasis, endothelial injury, and hypercoagulable state).3 This scenario is most frequently encountered following significant lower extremity trauma, with a much lower incidence of VTE following elective lower extremity arthroscopic procedures.4-6 A study by Maletis et al.7 found that the incidence symptomatic VTE after knee arthroscopy was approximately 0.40%. As HA use for acetabular labral tears and femoroacetabular impingement syndrome continues to increase, determining procedure-specific rates of morbidity and mortality is critical.8 A recent systematic review by Haldane et al.9 found limited data on the incidence of VTE after HA. They found 21 studies in the literature that addressed VTE after HA, one third of which were case reports. After excluding case reports, they found the incidence of sonographically diagnosed postoperative DVT to be about 2% despite prophylaxis with aspirin, nonsteroidal anti-inflammatory drugs, or low-molecular-weight heparin.9 In addition to the incidence of postoperative VTE, determining the risk factors associated with the development of VTE after
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 35, No 8 (August), 2019: pp 2380-2384
2381
INCIDENCE AND RISK OF VTE AFTER HA
HA is also critical. However, there is a paucity of information regarding risk factors for VTE after HA. Prior studies have identified various risk factors associated with VTE after surgical procedures including oral contraception (OC) use,10 smoking,11 obesity,12 and increasing age.13 The purpose of this retrospective study is to determine the incidence of symptomatic VTE after HA using a large national database while considering several patient demographic factors. We hypothesize that the incidence of symptomatic VTE after HA will be low (<1%) and associated with female patients using OC, smokers, and obese patients.
Methods Database In this retrospective study, the Humana administrative claims database was queried using the PearlDiver research program (www.pearldiverinc.com). The database houses deidentified information on patients from a variety of insurers, including Medicare Standard Analytical files and Humana’s claims database from 2007 to 2017. Data regarding patient demographics, medical comorbidities, postoperative complications, prescription medication dosage, geographic information, and procedural volumes were queried using International Classification of Diseases Ninth and Tenth Revision (ICD-9 and ICD-10) codes, Current Procedural Terminology (CPT) codes, and National Drug Codes. For this study, the Humana Orthopedic private payer database was used. The advantages of this database include a large national patient population, ability to analyze various comorbidities and complications, and ability to longitudinally track patients. This study was granted exemption from the Institutional Review Board at the participating institution as PearlDiver uses deidentified patient information. Additionally, the PearlDiver research program does not report specific data for any category with <11 patients to maintain patient anonymity. Patient Selection A retrospective analysis of patients undergoing HA was performed from 2007 to 2017. All adult patients 20 years of age who underwent HA were identified using the CPT codes 29860, 29861, 29862, 29863, 29914, 29914, 29915, 29916, and 29999. Pediatric patients (>20) and those who were not active for at least 90 days after HA within the database were not included in the study. Patient demographics and comorbidities assessed included age (<45 vs 45 years), gender, obesity (body mass index 30 kg/ m2),14 smoking status, OC use, diabetes, and chronic obstructive pulmonary disease (CLD). Previous studies have identified increasing age (40 years) as a significant risk factor for developing DVT.13 The incidence of
VTE was determined using relevant ICD-9 and -10 codes for DVT and PE within 30 and 90 days postoperatively (Appendix 1). Previous studies have shown that the risk of VTE may persist for longer than the first postoperative month after various procedures.15,16 To capture all symptomatic VTE after HA, this study assessed the incidence of VTE at 30 and 90 days. The incidence of mortality due to VTE was also assessed. Statistical Analysis Descriptive statistics were performed to report the incidence of VTEs in patients who underwent HA. Univariate analysis was performed using Pearson’s c2-test or Fisher’s exact test to compare demographics and medical comorbidities between patients who developed VTE and those who did not. Multivariate logistic regression analysis was performed to identify independent risk factors for developing VTE after HA. Univariate analyses were performed using Microsoft Excel (Redmond, WA), and multivariate analysis was performed using the open source R tool (www.r-project.org) housed within PearlDiver, with significance set at P < .05.
Results We identified 9,477 patients who underwent HA and met our inclusion and exclusion criteria. Of these, 5,088 (53.7%) were female, 7,348 (77.5%) were 45 years, 967 (10.2%) were smokers, 1,460 (15.4%) were obese, 2,206 (23.3%) had diabetes, and 354 (3.7%) had CLD. The overall incidence of postoperative VTE in all patients who underwent hip arthroscopy (HA) was 0.77% (n ¼ 73) and 1.14% (n ¼ 108) at 30 and 90 days, respectively. Specifically, the incidence of DVTs and PEs at 30 days was 0.57% (n ¼ 54) and 0.3% (n ¼ 28), respectively; 0.82% (n ¼ 78) and 0.43% (n ¼ 41) at 90 days (Table 1). Of note, the exact incidence of concomitant DVTs and PEs was not assessed due to an insufficient number of patients (n < 11) in the concomitant cohort at 30 and 90 days. Demographic and comorbidity data in the cohort diagnosed with VTE compared with the cohort without VTE at 90 days showed a significantly higher incidence Table 1. Incidence of Venous Thromboembolism After Hip Arthroscopy DVT* PE* VTE* DVTy PEy VTEy
All, n (%) 54 (0.57) 28 (0.3) 73 (0.77) 78 (0.82) 41 (0.43) 108 (1.14)
Male, n (%) 24 (0.25) 12 (0.13) 32 (0.34) 35 (0.37) 19 (0.2) 49 (0.52)
Female, n (%) 30 (0.32) 16 (0.17) 41 (0.43) 43 (0.45) 22 (0.23) 59 (0.62)
P Value .783 .713 .67 .798 .997 .844
DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism. *Within 30 days. y Within 90 days.
2382
Z. M. KHAZI ET AL.
of smoking (16.67% vs 10.21%; P ¼ .0282), obesity (32.41% vs 15.27%; P < .0001), and diabetes (34.26% vs 23.15%; P ¼ .0066) in the cohort diagnosed with VTE. However, gender and age were not found to be significant factors in the diagnosis of VTE (Table 2). Univariate analyses for the OC and CLD patient factors were not possible due to insufficient number of patients (n < 11) in the VTE cohort. Multivariate analysis was performed to identify independent patient factors associated with the diagnosis of VTE (Table 3). In this model, independent patient factors significantly associated with postoperative VTE were age 45 (odds ratio [OR] ¼ 1.82; 95% confidence interval [CI], 1.36-2.49; P ¼ .0001), obesity (OR ¼ 1.54; 95% CI, 1.27-1.86; P < .0001), smoking (OR ¼ 1.26; 95% CI, 1.04-1.53; P ¼ .0177), diabetes (OR ¼ 1.59; 95% CI, 1.32-1.92; P < .0001), and CLD (OR ¼ 2.10; 95% CI, 1.63-2.68; P < .0001). However, gender and OC use were not independently associated with diagnosis of VTE.
Discussion Using a large administrative claims database, we found that the incidence of postoperative VTE for patients undergoing HA was 0.77% and 1.14% at 30 and 90 days, respectively. We analyzed 9,477 cases of HA and found in both univariate and multivariate models that smoking, obesity, and diabetes were significantly associated with a higher risk of developing VTE after HA. Neither gender nor OC use alone significantly increased the risk for VTE after HA. Although HA has become increasingly common, data are still lacking on the incidence and risk factors associated with symptomatic VTE after the procedure. In a recent prospective study by Mohtadi et al.,17 DVTs were sonographically confirmed in 5 of 115 patients (4.4%) who underwent HA. Of these 5 patients, 4 had a clinically symptomatic DVT. Three were diagnosed with DVT prior to the first postoperative visit (2 weeks) and received anticoagulation therapy. They also reported potential risk factors associated with the incidence of DVTs in their patients. Three of the patients were female, of whom 2 were on OCs prior to surgery. Interestingly, all of the patients who developed
Table 3. Independent Risk Factors for Thromboembolic Events After Hip Arthroscopy Age 45 Male gender Obesity Smoking Oral contraceptive use Diabetes Chronic obstructive pulmonary disease
Odds Ratio [95% CI] 1.82 [1.36- 2.49] 1.01 [0.85-1.21] 1.54 [1.27-1.86] 1.26 [1.04-1.53] 0.98 [0.54-1.67] 1.59 [1.32-1.92] 2.10 [1.63-2.68]
P Value .0001* .8913 <.0001* .0177* .9568 <.0001* <.0001*
*Significant (P < .05).
DVTs were prescribed nonsteroidal anti-inflammatory drugs prior to surgery. The authors did not observe pulmonary emboli in any patients for up to 3 months after surgery.17 Several other studies have also assessed the incidence of postoperative VTEs in patients undergoing HA. In a prospective study by Fukushima et al.,18 the incidence of DVT was found to be 6.94% (n ¼ 5) 3 days after HA via duplex ultrasonography. However, all of these cases were distal, asymptomatic DVTs. In a series of 81 patients, Salvo et al.19 reported a 3.7% incidence of symptomatic DVTs that were confirmed with the use of Doppler ultrasonography within 2 weeks after HA. In a retrospective study of 139 HAs by Alaia et al.,20 2 (1.4%) patients developed DVTs, of whom 1 also developed a concurrent PE. In both cases, the DVTs were symptomatic and diagnosed using ultrasound.20 Souza et al.21 reported 1 (0.5%) case of symptomatic DVT after HA, which resolved after pharmacologic treatment. Phillipon et al.22 had no reports of symptomatic VTE in 112 HAs after 2 years. In a meta-analysis of 1,542 cases of HA, the incidence of sonographically diagnosed postoperative VTE was about 2.3% in patients who did not receive prophylaxis.9 Furthermore, a recent meta-analysis by Bolia et al.23 assessed 4,577 cases of HA for femoroacetabular impingement syndrome to determine the incidence of postoperative VTE. After adjusting for publication, they found the incidence of DVT was 2.02%. The results of our study are similar to those reported in a prospective study of 921 patients by Domb et al.24
Table 2. Comparison of Demographic Data Between No VTE and VTE Cohort Age 45 Female gender Oral contraceptive use Smoking Obesity Diabetes Chronic obstructive pulmonary disease
VTE n, (%) (n ¼ 108) 90 (83.3) 59 (54.6) <11 18 (16.7) 35 (32.4) 37 (34.3) <11
*Significant (P < .05). VTE, venous thromboembolism.
No VTE n, (%) (n ¼ 9,369) 7,258 (77.5) 5,029 (53.7) 479 (5.1) 957 (10.2) 1,431 (15.3) 2,169 (23.2) 348 (3.7)
P Value .1409 .8435 d .0282* <.0001* .0066* d
2383
INCIDENCE AND RISK OF VTE AFTER HA
They observed the incidence of VTE after HA to be around 0.76%. Also, a study by Truntzer et al.25 assessed the incidence of various complications after HA using the PearlDiver database. Of the 2,581 cases of HA, they found the incidence of postoperative DVTs to be 0.79% at 90 days.25 The findings of Domb et al.,24 Truntzer et al.,25 and our study suggest that the incidence of VTEs after HA is relatively low when compared with other common arthroscopic procedures, such as knee arthroscopy.6 On the other hand, studies by Fukushima et al.,18 Mohtadi et al.,17 and Salvo et al.19 report a much higher incidence. This discrepancy is largely due to the distinction of sonographically diagnosed DVTs versus clinically relevant DVTs. In addition to assessing the incidence of VTE after HA, we also analyzed patient factors associated with thromboembolic events following HA, which have only been evaluated in a limited fashion previously. In our study, a multivariate logistic regression model identified older age (45), obesity, smoking history, diabetes, and CLD as independent risk factors for developing VTE after HA. After a thorough review of the current literature, potential risk factors previously reported include increased age,17,18 increased body mass index,20,26 OC use,17,19,20,27 prolonged time under traction,17 and partial weight bearing following surgery.17 However, except for Mohtadi et al.17 and Fukushima et al.,18 none of these studies statistically analyzed the significance of these risk factors with the association of VTE after HA. Mohtadi et al.17 reported that there was no significant association between VTE after HA and OC use in female patients, increased age, or prolonged time under traction. On the other hand, Fukushima et al.18 identified increased age as a significant risk factor for developing DVTs after HA. However, it is important to note that both of these studies are limited by a relatively small sample size and may have been underpowered. In our study, risk factors associated with VTE after HA were older age (45 years), obesity, tobacco use, diabetes, and CLD. However, gender and OC use were not independently associated with diagnosis of VTE. Limitations There were limitations in this study that are inherent to large national registries such as PearlDiver, which is well documented.28 For this study, the accuracy of the information is directly linked to the accuracy of the coding process because ICD and CPT codes were used to query the data. Second, it is likely that the true incidence of VTE cannot be assessed using databases because asymptomatic VTEs do not usually present to the health care system. However, the purpose of this study was to evaluate the incidence of clinically relevant VTE. Third, surgery-specific information such as anesthesia type, operative time, time under traction, and patient positioning during surgery was not
recorded in this database. Moreover, patients could not be categorized based on the American Society of Anesthesiologists classification system, a strong predictor of postoperative outcomes,29,30 due to the limitations of the database. Also, postoperative variables such as weight-bearing status, DVT prophylaxis, and postoperative rehab protocol were not available in the database. Fourth, the diagnostic modality (clinical, ultrasound, or computed tomography) for identifying the presence of thromboembolic events was not available in the database. In addition, information regarding VTE prophylaxis prior to surgery was not analyzed in this study because many patients use over-the-counter medications with antiplatelet or anticoagulant properties that would not be captured within PearlDiver. Additionally, to ensure patient anonymity, the PearlDiver research tool does not report the exact number of patients within a cohort if the occupancy of the cohort is between 1 and 10. Therefore, descriptive statistics and univariate analyses were not possible for patients with history of CLD or OC use in the VTE cohort. Similarly, it was not possible to report the exact incidence of concomitant DVTs and PEs due to insufficient patients in the concomitant cohorts at 30 and 90 days. This led to a slightly overrepresented DVT and PE cohort at 30 and 90 days. However, the incidence of VTE at 90 days was used as the final sample size for the multivariate analysis. Lastly, patient-reported outcomes are not documented in this database, and therefore comparison of these outcomes could not be performed.
Conclusions For patients undergoing HA, the incidence of symptomatic postoperative VTE is low. This study identified age 45, obesity, tobacco use, diabetes, and CLD as independent risk factors for VTE after HA.
References 1. Haas S. Deep vein thrombosis: Beyond the operating table. Orthopedics 2000;23:s629-s632 (6 Suppl). 2. Bokshan SL, DeFroda SF, Panarello NM, Owens BD. Risk factors for deep vein thrombosis or pulmonary embolus following anterior cruciate ligament reconstruction. Orthop J Sports Med 2018;6:2325967118781328. 3. Ilahi OA, Reddy J, Ahmad I. Deep venous thrombosis after knee arthroscopy: A meta-analysis. Arthroscopy 2005;21:727-730. 4. Michot M, Conen D, Holtz D, et al. Prevention of deepvein thrombosis in ambulatory arthroscopic knee surgery: A randomized trial of prophylaxis with lowmolecular weight heparin. Arthroscopy 2002;18:257-263. 5. Kanchanabat B, Stapanavatr W, Meknavin S, Soorapanth C, Sumanasrethakul C, Kanchanasuttirak P. Systematic review and meta-analysis on the rate of postoperative venous thromboembolism in orthopaedic surgery in Asian patients without thromboprophylaxis. Br J Surg 2011;98:1356-1364.
2384
Z. M. KHAZI ET AL.
6. Sun Y, Chen D, Xu Z, et al. Deep venous thrombosis after knee arthroscopy: A systematic review and meta-analysis. Arthroscopy 2014;30:406-412. 7. Maletis GB, Inacio MC, Reynolds S, Funahashi TT. Incidence of symptomatic venous thromboembolism after elective knee arthroscopy. J Bone Joint Surg Am 2012;94: 714-720. 8. Maradit Kremers H, Schilz SR, Van Houten HK, et al. Trends in utilization and outcomes of hip arthroscopy in the United States between 2005 and 2013. J Arthroplasty 2017;32:750-755. 9. Haldane CE, Ekhtiari S, de Sa D, et al. Venous thromboembolism events after hip arthroscopy: A systematic review. Arthroscopy 2018;34:321-330.e321. 10. Taube OM, Rouse MR, D’Angelo L. Oral contraceptives and venous thromboses in adolescents undergoing elective surgery: A case report, and review of the literature. J Adolesc Health 1992;13:634-636. 11. Cancienne JM, Gwathmey FW, Miller MD, Werner BC. Tobacco use is associated with increased complications after anterior cruciate ligament reconstruction. Am J Sports Med 2016;44:99-104. 12. Ahmad J, Lynch MK, Maltenfort M. Incidence and risk factors of venous thromboembolism after orthopaedic foot and ankle surgery. Foot Ankle Spec 2017;10:449-454. 13. Anderson FA Jr, Spencer FA. Risk factors for venous thromboembolism. Circulation 2003;107:I9-I16 (23 Suppl 1). 14. Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. JAMA 2012;307: 491-497. 15. Scurr JH. How long after surgery does the risk of thromboembolism persist? Acta Chir Scand Suppl 1990;556: 22-24. 16. Planes A, Vochelle N, Darmon JY, Fagola M, Bellaud M, Huet Y. Risk of deep-venous thrombosis after hospital discharge in patients having undergone total hip replacement: double-blind randomised comparison of enoxaparin versus placebo. Lancet 1996;348:224-228. 17. Mohtadi NG, Johnston K, Gaudelli C, et al. The incidence of proximal deep vein thrombosis after elective hip arthroscopy: A prospective cohort study in low risk patients. J Hip Preserv Surg 2016;3:295-303. 18. Fukushima K, Takahira N, Uchiyama K, Moriya M, Minato T, Takaso M. The incidence of deep vein thrombosis (DVT) during hip arthroscopic surgery. Arch Orthop Trauma Surg 2016;136:1431-1435.
19. Salvo JP, Troxell CR, Duggan DP. Incidence of venous thromboembolic disease following hip arthroscopy. Orthopedics 2010;33:664. 20. Alaia MJ, Patel D, Levy A, et al. The incidence of venous thromboembolism (VTE)dafter hip arthroscopy. Bull Hosp Joint Dis (2013) 2014;72:154-158. 21. Souza BG, Dani WS, Honda EK, et al. Do complications in hip arthroscopy change with experience? Arthroscopy 2010;26:1053-1057. 22. Philippon MJ, Briggs KK, Yen YM, Kuppersmith DA. Outcomes following hip arthroscopy for femoroacetabular impingement with associated chondrolabral dysfunction: Minimum two-year follow-up. J Bone Joint Surg Br 2009;91:16-23. 23. Bolia IK, Fagotti L, McNamara S, Dornan G, Briggs KK, Philippon MJ. A systematic review-meta-analysis of venous thromboembolic events following primary hip arthroscopy for FAI: Clinical and epidemiologic considerations. J Hip Preserv Surg 2018;5:190-201. 24. Domb BG, Gui C, Hutchinson MR, Nho SJ, Terry MA, Lodhia P. Clinical outcomes of hip arthroscopic surgery: A prospective survival analysis of primary and revision surgeries in a large mixed cohort. Am J Sports Med 2016;44:2505-2517. 25. Truntzer JN, Hoppe DJ, Shapiro LM, Abrams GD, Safran M. Complication rates for hip arthroscopy are underestimated: A population-based study. Arthroscopy 2017;33:1194-1201. 26. Collins JA, Beutel BG, Garofolo G, Youm T. Correlation of obesity with patient-reported outcomes and complications after hip arthroscopy. Arthroscopy 2015;31:57-62. 27. Alaia MJ, Zuskov A, Davidovitch RI. Contralateral deep venous thrombosis after hip arthroscopy. Orthopedics 2011;34:e674-e677. 28. Pugely AJ, Martin CT, Harwood J, Ong KL, Bozic KJ, Callaghan JJ. Database and registry research in orthopaedic surgery: Part I: Claims-based data. J Bone Joint Surg Am 2015;97:1278-1287. 29. Namba RS, Inacio MC, Paxton EW. Risk factors associated with deep surgical site infections after primary total knee arthroplasty: An analysis of 56,216 knees. J Bone Joint Surg Am 2013;95:775-782. 30. Bovonratwet P, Nelson SJ, Bellamkonda K, et al. Similar 30-day complications for septic knee arthritis treated with arthrotomy or arthroscopy: An American College of Surgeons National Surgical Quality Improvement Program analysis. Arthroscopy 2018;34:213-219.
INCIDENCE AND RISK OF VTE AFTER HA
2384.e1
Appendix 1. International Classification of Diseases (ICD) Codes Deep vein thrombosis
ICD-9 45340, 45341, 45342, 45382, 45383, 45384
Pulmonary embolism
41510, 41511, 41519
ICD-10 I8240, I82401, I82402, I82403, I82409, I82410, I82411, I82412, I82413, I82419, I82420, I82421, I82422, I82423, I82429, I82430 to I82499, I824Y1, I824Y2, I824Y3, I824Y9, I824Z1, I824Z2, I824Z3, I824Z9, I82621 to I82629 I2600, I2602, I2609, I2699
Purpose: To determine the incidence of symptomatic venous thromboembolism (VTE) after hip arthroscopy (HA) using a large national database while considering several patient demographic factors. Methods: Patients 20 years old who underwent HA between 2007 and 2017 were identified within the Humana administrative claims database using relevant Current Procedural Terminology and International Classification of Diseases Ninth and Tenth Revision codes. Basic demographics, including age, gender, obesity (body mass index 30 kg/m2), oral contraceptive use, smoking history, diabetes, and chronic obstructive pulmonary disease (CLD) were recorded. Postoperative incidence of deep vein thrombosis, pulmonary embolism, and VTE was identified at 30 and 90 days postoperatively. Multivariate logistic regression analysis was performed to identify independent risk factors for VTE after HA, with statistical significance set at P < .05. Results: Overall, 9,477 patients underwent HA procedures over the study period, of whom 5,085 (53.7%) were female. The overall incidence of VTE in all patients was 0.77% (n ¼ 73) and 1.14% (n ¼ 108) at 30 and 90 days, respectively. Multivariate analysis identified age 45 (odds ratio [OR] ¼ 1.82; 95% confidence interval [CI], 1.36-2.49; P ¼ .0001), obesity (OR ¼ 1.54; 95% CI, 1.27-1.86; P < .0001), smoking (OR ¼ 1.26; 95% CI, 1.04-1.53; P ¼ .0177), diabetes (OR ¼ 1.59; 95% CI, 1.32-1.92; P < .0001), and CLD (OR ¼ 2.10; 95% CI, 1.63-2.68; P < .0001) as independent risk factors for higher incidence of VTE after HA. However, neither gender nor oral contraceptive use were risk factors for VTE after HA. Conclusions: For patients undergoing HA, the incidence of symptomatic postoperative VTE is low. This study identified age 45, obesity, tobacco use, diabetes, and CLD as independent risk factors for VTE after HA. Level of Evidence: Level III, retrospective cohort study.
V
enous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a relatively rare complication following elective orthopaedic procedures that may result in significant morbidity and mortality.1,2 The development of VTE is a complex pathophysiological process that is typically secondary to factors characterized by Virchow’s triad
Division of Sports Medicine, Department of Orthopedics and Rehabilitation, University of Iowa Hospitals and Clinics, Iowa City, Iowa, U.S.A. The authors report the following potential conflicts of interest or sources of funding: R.W.W. receives research support from Smith & Nephew. Full ICMJE author disclosure forms are available for this article online, as supplementary material. Received December 19, 2018; accepted March 24, 2019. Address correspondence to Robert W. Westermann, M.D., 2701 Prairie Meadow Dr, Iowa City, IA 52246, U.S.A. E-mail: robert-westermann@ uiowa.edu Ó 2019 Published by Elsevier on behalf of the Arthroscopy Association of North America 0749-8063/181541/$36.00 https://doi.org/10.1016/j.arthro.2019.03.054
2380
(venous stasis, endothelial injury, and hypercoagulable state).3 This scenario is most frequently encountered following significant lower extremity trauma, with a much lower incidence of VTE following elective lower extremity arthroscopic procedures.4-6 A study by Maletis et al.7 found that the incidence symptomatic VTE after knee arthroscopy was approximately 0.40%. As HA use for acetabular labral tears and femoroacetabular impingement syndrome continues to increase, determining procedure-specific rates of morbidity and mortality is critical.8 A recent systematic review by Haldane et al.9 found limited data on the incidence of VTE after HA. They found 21 studies in the literature that addressed VTE after HA, one third of which were case reports. After excluding case reports, they found the incidence of sonographically diagnosed postoperative DVT to be about 2% despite prophylaxis with aspirin, nonsteroidal anti-inflammatory drugs, or low-molecular-weight heparin.9 In addition to the incidence of postoperative VTE, determining the risk factors associated with the development of VTE after
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 35, No 8 (August), 2019: pp 2380-2384
2381
INCIDENCE AND RISK OF VTE AFTER HA
HA is also critical. However, there is a paucity of information regarding risk factors for VTE after HA. Prior studies have identified various risk factors associated with VTE after surgical procedures including oral contraception (OC) use,10 smoking,11 obesity,12 and increasing age.13 The purpose of this retrospective study is to determine the incidence of symptomatic VTE after HA using a large national database while considering several patient demographic factors. We hypothesize that the incidence of symptomatic VTE after HA will be low (<1%) and associated with female patients using OC, smokers, and obese patients.
Methods Database In this retrospective study, the Humana administrative claims database was queried using the PearlDiver research program (www.pearldiverinc.com). The database houses deidentified information on patients from a variety of insurers, including Medicare Standard Analytical files and Humana’s claims database from 2007 to 2017. Data regarding patient demographics, medical comorbidities, postoperative complications, prescription medication dosage, geographic information, and procedural volumes were queried using International Classification of Diseases Ninth and Tenth Revision (ICD-9 and ICD-10) codes, Current Procedural Terminology (CPT) codes, and National Drug Codes. For this study, the Humana Orthopedic private payer database was used. The advantages of this database include a large national patient population, ability to analyze various comorbidities and complications, and ability to longitudinally track patients. This study was granted exemption from the Institutional Review Board at the participating institution as PearlDiver uses deidentified patient information. Additionally, the PearlDiver research program does not report specific data for any category with <11 patients to maintain patient anonymity. Patient Selection A retrospective analysis of patients undergoing HA was performed from 2007 to 2017. All adult patients 20 years of age who underwent HA were identified using the CPT codes 29860, 29861, 29862, 29863, 29914, 29914, 29915, 29916, and 29999. Pediatric patients (>20) and those who were not active for at least 90 days after HA within the database were not included in the study. Patient demographics and comorbidities assessed included age (<45 vs 45 years), gender, obesity (body mass index 30 kg/ m2),14 smoking status, OC use, diabetes, and chronic obstructive pulmonary disease (CLD). Previous studies have identified increasing age (40 years) as a significant risk factor for developing DVT.13 The incidence of
VTE was determined using relevant ICD-9 and -10 codes for DVT and PE within 30 and 90 days postoperatively (Appendix 1). Previous studies have shown that the risk of VTE may persist for longer than the first postoperative month after various procedures.15,16 To capture all symptomatic VTE after HA, this study assessed the incidence of VTE at 30 and 90 days. The incidence of mortality due to VTE was also assessed. Statistical Analysis Descriptive statistics were performed to report the incidence of VTEs in patients who underwent HA. Univariate analysis was performed using Pearson’s c2-test or Fisher’s exact test to compare demographics and medical comorbidities between patients who developed VTE and those who did not. Multivariate logistic regression analysis was performed to identify independent risk factors for developing VTE after HA. Univariate analyses were performed using Microsoft Excel (Redmond, WA), and multivariate analysis was performed using the open source R tool (www.r-project.org) housed within PearlDiver, with significance set at P < .05.
Results We identified 9,477 patients who underwent HA and met our inclusion and exclusion criteria. Of these, 5,088 (53.7%) were female, 7,348 (77.5%) were 45 years, 967 (10.2%) were smokers, 1,460 (15.4%) were obese, 2,206 (23.3%) had diabetes, and 354 (3.7%) had CLD. The overall incidence of postoperative VTE in all patients who underwent hip arthroscopy (HA) was 0.77% (n ¼ 73) and 1.14% (n ¼ 108) at 30 and 90 days, respectively. Specifically, the incidence of DVTs and PEs at 30 days was 0.57% (n ¼ 54) and 0.3% (n ¼ 28), respectively; 0.82% (n ¼ 78) and 0.43% (n ¼ 41) at 90 days (Table 1). Of note, the exact incidence of concomitant DVTs and PEs was not assessed due to an insufficient number of patients (n < 11) in the concomitant cohort at 30 and 90 days. Demographic and comorbidity data in the cohort diagnosed with VTE compared with the cohort without VTE at 90 days showed a significantly higher incidence Table 1. Incidence of Venous Thromboembolism After Hip Arthroscopy DVT* PE* VTE* DVTy PEy VTEy
All, n (%) 54 (0.57) 28 (0.3) 73 (0.77) 78 (0.82) 41 (0.43) 108 (1.14)
Male, n (%) 24 (0.25) 12 (0.13) 32 (0.34) 35 (0.37) 19 (0.2) 49 (0.52)
Female, n (%) 30 (0.32) 16 (0.17) 41 (0.43) 43 (0.45) 22 (0.23) 59 (0.62)
P Value .783 .713 .67 .798 .997 .844
DVT, deep vein thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism. *Within 30 days. y Within 90 days.
2382
Z. M. KHAZI ET AL.
of smoking (16.67% vs 10.21%; P ¼ .0282), obesity (32.41% vs 15.27%; P < .0001), and diabetes (34.26% vs 23.15%; P ¼ .0066) in the cohort diagnosed with VTE. However, gender and age were not found to be significant factors in the diagnosis of VTE (Table 2). Univariate analyses for the OC and CLD patient factors were not possible due to insufficient number of patients (n < 11) in the VTE cohort. Multivariate analysis was performed to identify independent patient factors associated with the diagnosis of VTE (Table 3). In this model, independent patient factors significantly associated with postoperative VTE were age 45 (odds ratio [OR] ¼ 1.82; 95% confidence interval [CI], 1.36-2.49; P ¼ .0001), obesity (OR ¼ 1.54; 95% CI, 1.27-1.86; P < .0001), smoking (OR ¼ 1.26; 95% CI, 1.04-1.53; P ¼ .0177), diabetes (OR ¼ 1.59; 95% CI, 1.32-1.92; P < .0001), and CLD (OR ¼ 2.10; 95% CI, 1.63-2.68; P < .0001). However, gender and OC use were not independently associated with diagnosis of VTE.
Discussion Using a large administrative claims database, we found that the incidence of postoperative VTE for patients undergoing HA was 0.77% and 1.14% at 30 and 90 days, respectively. We analyzed 9,477 cases of HA and found in both univariate and multivariate models that smoking, obesity, and diabetes were significantly associated with a higher risk of developing VTE after HA. Neither gender nor OC use alone significantly increased the risk for VTE after HA. Although HA has become increasingly common, data are still lacking on the incidence and risk factors associated with symptomatic VTE after the procedure. In a recent prospective study by Mohtadi et al.,17 DVTs were sonographically confirmed in 5 of 115 patients (4.4%) who underwent HA. Of these 5 patients, 4 had a clinically symptomatic DVT. Three were diagnosed with DVT prior to the first postoperative visit (2 weeks) and received anticoagulation therapy. They also reported potential risk factors associated with the incidence of DVTs in their patients. Three of the patients were female, of whom 2 were on OCs prior to surgery. Interestingly, all of the patients who developed
Table 3. Independent Risk Factors for Thromboembolic Events After Hip Arthroscopy Age 45 Male gender Obesity Smoking Oral contraceptive use Diabetes Chronic obstructive pulmonary disease
Odds Ratio [95% CI] 1.82 [1.36- 2.49] 1.01 [0.85-1.21] 1.54 [1.27-1.86] 1.26 [1.04-1.53] 0.98 [0.54-1.67] 1.59 [1.32-1.92] 2.10 [1.63-2.68]
P Value .0001* .8913 <.0001* .0177* .9568 <.0001* <.0001*
*Significant (P < .05).
DVTs were prescribed nonsteroidal anti-inflammatory drugs prior to surgery. The authors did not observe pulmonary emboli in any patients for up to 3 months after surgery.17 Several other studies have also assessed the incidence of postoperative VTEs in patients undergoing HA. In a prospective study by Fukushima et al.,18 the incidence of DVT was found to be 6.94% (n ¼ 5) 3 days after HA via duplex ultrasonography. However, all of these cases were distal, asymptomatic DVTs. In a series of 81 patients, Salvo et al.19 reported a 3.7% incidence of symptomatic DVTs that were confirmed with the use of Doppler ultrasonography within 2 weeks after HA. In a retrospective study of 139 HAs by Alaia et al.,20 2 (1.4%) patients developed DVTs, of whom 1 also developed a concurrent PE. In both cases, the DVTs were symptomatic and diagnosed using ultrasound.20 Souza et al.21 reported 1 (0.5%) case of symptomatic DVT after HA, which resolved after pharmacologic treatment. Phillipon et al.22 had no reports of symptomatic VTE in 112 HAs after 2 years. In a meta-analysis of 1,542 cases of HA, the incidence of sonographically diagnosed postoperative VTE was about 2.3% in patients who did not receive prophylaxis.9 Furthermore, a recent meta-analysis by Bolia et al.23 assessed 4,577 cases of HA for femoroacetabular impingement syndrome to determine the incidence of postoperative VTE. After adjusting for publication, they found the incidence of DVT was 2.02%. The results of our study are similar to those reported in a prospective study of 921 patients by Domb et al.24
Table 2. Comparison of Demographic Data Between No VTE and VTE Cohort Age 45 Female gender Oral contraceptive use Smoking Obesity Diabetes Chronic obstructive pulmonary disease
VTE n, (%) (n ¼ 108) 90 (83.3) 59 (54.6) <11 18 (16.7) 35 (32.4) 37 (34.3) <11
*Significant (P < .05). VTE, venous thromboembolism.
No VTE n, (%) (n ¼ 9,369) 7,258 (77.5) 5,029 (53.7) 479 (5.1) 957 (10.2) 1,431 (15.3) 2,169 (23.2) 348 (3.7)
P Value .1409 .8435 d .0282* <.0001* .0066* d
2383
INCIDENCE AND RISK OF VTE AFTER HA
They observed the incidence of VTE after HA to be around 0.76%. Also, a study by Truntzer et al.25 assessed the incidence of various complications after HA using the PearlDiver database. Of the 2,581 cases of HA, they found the incidence of postoperative DVTs to be 0.79% at 90 days.25 The findings of Domb et al.,24 Truntzer et al.,25 and our study suggest that the incidence of VTEs after HA is relatively low when compared with other common arthroscopic procedures, such as knee arthroscopy.6 On the other hand, studies by Fukushima et al.,18 Mohtadi et al.,17 and Salvo et al.19 report a much higher incidence. This discrepancy is largely due to the distinction of sonographically diagnosed DVTs versus clinically relevant DVTs. In addition to assessing the incidence of VTE after HA, we also analyzed patient factors associated with thromboembolic events following HA, which have only been evaluated in a limited fashion previously. In our study, a multivariate logistic regression model identified older age (45), obesity, smoking history, diabetes, and CLD as independent risk factors for developing VTE after HA. After a thorough review of the current literature, potential risk factors previously reported include increased age,17,18 increased body mass index,20,26 OC use,17,19,20,27 prolonged time under traction,17 and partial weight bearing following surgery.17 However, except for Mohtadi et al.17 and Fukushima et al.,18 none of these studies statistically analyzed the significance of these risk factors with the association of VTE after HA. Mohtadi et al.17 reported that there was no significant association between VTE after HA and OC use in female patients, increased age, or prolonged time under traction. On the other hand, Fukushima et al.18 identified increased age as a significant risk factor for developing DVTs after HA. However, it is important to note that both of these studies are limited by a relatively small sample size and may have been underpowered. In our study, risk factors associated with VTE after HA were older age (45 years), obesity, tobacco use, diabetes, and CLD. However, gender and OC use were not independently associated with diagnosis of VTE. Limitations There were limitations in this study that are inherent to large national registries such as PearlDiver, which is well documented.28 For this study, the accuracy of the information is directly linked to the accuracy of the coding process because ICD and CPT codes were used to query the data. Second, it is likely that the true incidence of VTE cannot be assessed using databases because asymptomatic VTEs do not usually present to the health care system. However, the purpose of this study was to evaluate the incidence of clinically relevant VTE. Third, surgery-specific information such as anesthesia type, operative time, time under traction, and patient positioning during surgery was not
recorded in this database. Moreover, patients could not be categorized based on the American Society of Anesthesiologists classification system, a strong predictor of postoperative outcomes,29,30 due to the limitations of the database. Also, postoperative variables such as weight-bearing status, DVT prophylaxis, and postoperative rehab protocol were not available in the database. Fourth, the diagnostic modality (clinical, ultrasound, or computed tomography) for identifying the presence of thromboembolic events was not available in the database. In addition, information regarding VTE prophylaxis prior to surgery was not analyzed in this study because many patients use over-the-counter medications with antiplatelet or anticoagulant properties that would not be captured within PearlDiver. Additionally, to ensure patient anonymity, the PearlDiver research tool does not report the exact number of patients within a cohort if the occupancy of the cohort is between 1 and 10. Therefore, descriptive statistics and univariate analyses were not possible for patients with history of CLD or OC use in the VTE cohort. Similarly, it was not possible to report the exact incidence of concomitant DVTs and PEs due to insufficient patients in the concomitant cohorts at 30 and 90 days. This led to a slightly overrepresented DVT and PE cohort at 30 and 90 days. However, the incidence of VTE at 90 days was used as the final sample size for the multivariate analysis. Lastly, patient-reported outcomes are not documented in this database, and therefore comparison of these outcomes could not be performed.
Conclusions For patients undergoing HA, the incidence of symptomatic postoperative VTE is low. This study identified age 45, obesity, tobacco use, diabetes, and CLD as independent risk factors for VTE after HA.
References 1. Haas S. Deep vein thrombosis: Beyond the operating table. Orthopedics 2000;23:s629-s632 (6 Suppl). 2. Bokshan SL, DeFroda SF, Panarello NM, Owens BD. Risk factors for deep vein thrombosis or pulmonary embolus following anterior cruciate ligament reconstruction. Orthop J Sports Med 2018;6:2325967118781328. 3. Ilahi OA, Reddy J, Ahmad I. Deep venous thrombosis after knee arthroscopy: A meta-analysis. Arthroscopy 2005;21:727-730. 4. Michot M, Conen D, Holtz D, et al. Prevention of deepvein thrombosis in ambulatory arthroscopic knee surgery: A randomized trial of prophylaxis with lowmolecular weight heparin. Arthroscopy 2002;18:257-263. 5. Kanchanabat B, Stapanavatr W, Meknavin S, Soorapanth C, Sumanasrethakul C, Kanchanasuttirak P. Systematic review and meta-analysis on the rate of postoperative venous thromboembolism in orthopaedic surgery in Asian patients without thromboprophylaxis. Br J Surg 2011;98:1356-1364.
2384
Z. M. KHAZI ET AL.
6. Sun Y, Chen D, Xu Z, et al. Deep venous thrombosis after knee arthroscopy: A systematic review and meta-analysis. Arthroscopy 2014;30:406-412. 7. Maletis GB, Inacio MC, Reynolds S, Funahashi TT. Incidence of symptomatic venous thromboembolism after elective knee arthroscopy. J Bone Joint Surg Am 2012;94: 714-720. 8. Maradit Kremers H, Schilz SR, Van Houten HK, et al. Trends in utilization and outcomes of hip arthroscopy in the United States between 2005 and 2013. J Arthroplasty 2017;32:750-755. 9. Haldane CE, Ekhtiari S, de Sa D, et al. Venous thromboembolism events after hip arthroscopy: A systematic review. Arthroscopy 2018;34:321-330.e321. 10. Taube OM, Rouse MR, D’Angelo L. Oral contraceptives and venous thromboses in adolescents undergoing elective surgery: A case report, and review of the literature. J Adolesc Health 1992;13:634-636. 11. Cancienne JM, Gwathmey FW, Miller MD, Werner BC. Tobacco use is associated with increased complications after anterior cruciate ligament reconstruction. Am J Sports Med 2016;44:99-104. 12. Ahmad J, Lynch MK, Maltenfort M. Incidence and risk factors of venous thromboembolism after orthopaedic foot and ankle surgery. Foot Ankle Spec 2017;10:449-454. 13. Anderson FA Jr, Spencer FA. Risk factors for venous thromboembolism. Circulation 2003;107:I9-I16 (23 Suppl 1). 14. Flegal KM, Carroll MD, Kit BK, Ogden CL. Prevalence of obesity and trends in the distribution of body mass index among US adults, 1999-2010. JAMA 2012;307: 491-497. 15. Scurr JH. How long after surgery does the risk of thromboembolism persist? Acta Chir Scand Suppl 1990;556: 22-24. 16. Planes A, Vochelle N, Darmon JY, Fagola M, Bellaud M, Huet Y. Risk of deep-venous thrombosis after hospital discharge in patients having undergone total hip replacement: double-blind randomised comparison of enoxaparin versus placebo. Lancet 1996;348:224-228. 17. Mohtadi NG, Johnston K, Gaudelli C, et al. The incidence of proximal deep vein thrombosis after elective hip arthroscopy: A prospective cohort study in low risk patients. J Hip Preserv Surg 2016;3:295-303. 18. Fukushima K, Takahira N, Uchiyama K, Moriya M, Minato T, Takaso M. The incidence of deep vein thrombosis (DVT) during hip arthroscopic surgery. Arch Orthop Trauma Surg 2016;136:1431-1435.
19. Salvo JP, Troxell CR, Duggan DP. Incidence of venous thromboembolic disease following hip arthroscopy. Orthopedics 2010;33:664. 20. Alaia MJ, Patel D, Levy A, et al. The incidence of venous thromboembolism (VTE)dafter hip arthroscopy. Bull Hosp Joint Dis (2013) 2014;72:154-158. 21. Souza BG, Dani WS, Honda EK, et al. Do complications in hip arthroscopy change with experience? Arthroscopy 2010;26:1053-1057. 22. Philippon MJ, Briggs KK, Yen YM, Kuppersmith DA. Outcomes following hip arthroscopy for femoroacetabular impingement with associated chondrolabral dysfunction: Minimum two-year follow-up. J Bone Joint Surg Br 2009;91:16-23. 23. Bolia IK, Fagotti L, McNamara S, Dornan G, Briggs KK, Philippon MJ. A systematic review-meta-analysis of venous thromboembolic events following primary hip arthroscopy for FAI: Clinical and epidemiologic considerations. J Hip Preserv Surg 2018;5:190-201. 24. Domb BG, Gui C, Hutchinson MR, Nho SJ, Terry MA, Lodhia P. Clinical outcomes of hip arthroscopic surgery: A prospective survival analysis of primary and revision surgeries in a large mixed cohort. Am J Sports Med 2016;44:2505-2517. 25. Truntzer JN, Hoppe DJ, Shapiro LM, Abrams GD, Safran M. Complication rates for hip arthroscopy are underestimated: A population-based study. Arthroscopy 2017;33:1194-1201. 26. Collins JA, Beutel BG, Garofolo G, Youm T. Correlation of obesity with patient-reported outcomes and complications after hip arthroscopy. Arthroscopy 2015;31:57-62. 27. Alaia MJ, Zuskov A, Davidovitch RI. Contralateral deep venous thrombosis after hip arthroscopy. Orthopedics 2011;34:e674-e677. 28. Pugely AJ, Martin CT, Harwood J, Ong KL, Bozic KJ, Callaghan JJ. Database and registry research in orthopaedic surgery: Part I: Claims-based data. J Bone Joint Surg Am 2015;97:1278-1287. 29. Namba RS, Inacio MC, Paxton EW. Risk factors associated with deep surgical site infections after primary total knee arthroplasty: An analysis of 56,216 knees. J Bone Joint Surg Am 2013;95:775-782. 30. Bovonratwet P, Nelson SJ, Bellamkonda K, et al. Similar 30-day complications for septic knee arthritis treated with arthrotomy or arthroscopy: An American College of Surgeons National Surgical Quality Improvement Program analysis. Arthroscopy 2018;34:213-219.
INCIDENCE AND RISK OF VTE AFTER HA
2384.e1
Appendix 1. International Classification of Diseases (ICD) Codes Deep vein thrombosis
ICD-9 45340, 45341, 45342, 45382, 45383, 45384
Pulmonary embolism
41510, 41511, 41519
ICD-10 I8240, I82401, I82402, I82403, I82409, I82410, I82411, I82412, I82413, I82419, I82420, I82421, I82422, I82423, I82429, I82430 to I82499, I824Y1, I824Y2, I824Y3, I824Y9, I824Z1, I824Z2, I824Z3, I824Z9, I82621 to I82629 I2600, I2602, I2609, I2699