Thrombosis Research 133 (2014) 719–724
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Population-based epidemiology of postoperative venous thromboembolism in Taiwanese patients receiving hip or knee arthroplasty without pharmacological thromboprophylaxis Po-Kuei Wu a,b,c,d, Cheng-Fong Chen b,c,d, Lien-Hsiang Chung c,d,e, Chien-Lin Liu b,d, Wei-Ming Chen b,c,d,⁎ a
Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan, ROC Department of Orthropaedics and Traumatology, Taipei Veterans General Hospital, Taipei, Taiwan, ROC c Therapeutical and Research Center of Musculoskeletal Tumor, Taipei Veterans General Hospital, Taipei, Taiwan, ROC d Orthopaedic Department School of Medicine, National Yang-Ming University e Department of Orthropaedics and Traumatology, Chia-Yi Christian Hospital, Chia-Yi, Taiwan, ROC b
a r t i c l e
i n f o
Article history: Received 16 September 2013 Received in revised form 14 January 2014 Accepted 27 January 2014 Available online 5 February 2014 Keywords: Pulmonary embolism Venous thromboembolism Deep vein thrombosis TKA THR
a b s t r a c t Introduction: Population-based evaluation on the incidence of postoperative venous thromboembolism (VTE) has not yet been reported for Asians receiving arthroplasty. In Taiwan, thromboprophylaxis was not commonly applied for patients. The population-based cohort study aimed to investigate the epidemiology, and to determine the risk factors VTE for patients receiving hip or knee replacement without pharmacological thromboprophylaxis in Taiwan. Materials and Methods: We retrospectively acquired patients’ data from National Health Insurance databases representing more than 99% of about 23 million Taiwanese citizens. The primary outcome was the incidence of composite symptomatic VTE within 28 days after receiving hip or knee replacement surgery. Results: During 2002 to 2006, there were 114,026 patients undergoing hip (n = 61,460) or knee (n = 52,566) replacement surgery. The occurrence rate of overall postoperative VTE was 0.44%. The incidence of pulmonary embolism was four in 10,000 patients receiving hip replacement or seven in 10,000 individuals undergoing knee replacement. The weekly cumulative incidence of VTE was persistently rising up to 28 days after surgery. Dramatic increase in risk of post-surgical VTE was associated with prior disease history of PE (p b 0.001 for hip replacement, p = 0.01 for knee replacement) or DVT (p = 0.004 for hip replacement, p b 0.001 for knee replacement). Prior claim of congestive heart failure was an independent risk factor associated with patients receiving knee arthroplasty (p = 0.01). Conclusion: Life-threatening PE occurred and increased cumulatively up to 28 days after hip or knee arthroplasty in Asians. Proper prophylaxis for patients with the exposure of high risks needs to be scrutinized. © 2014 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism (VTE), consisting of pulmonary embolism (PE) and deep vein thrombosis (DVT), is a common postoperative complication for patients receiving hip or knee replacement surgery [1–3]. Standard pharmacological prophylaxis such as anticoagulants, mechanical devices such as calf pump or combined intervention have been practiced for decades. Guideline recommendations for appropriate prophylaxis based on strong evidence have been published by American College of Chest Physicians (ACCP) [4], American Academy of Orthopaedic Surgeons (AAOS) [5], or the consensus of the Asia-Pacific Thrombosis Advisory Board in the countries without their own guidelines [6]. In Taiwan, postoperative thrombosis prophylaxis has not yet ⁎ Corresponding author at: 18F, 201, Section 2, Shin-Pai Road, Taipei 112, Taiwan, ROC. Tel.: +886 2 28757557x117; fax: +886 2 28757559. E-mail addresses: [email protected] (P.-K. Wu), [email protected] (W.-M. Chen). 0049-3848/$ – see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.thromres.2014.01.039
applied for patients receiving hip or knee arthroplasty that may be attributed the following reasons. Firstly, the assumption determined by previous studies indicated that occurrence of post-surgical VTE in Asian population was relatively uncommon [7,8]. Secondly, given the matters of unreliable diagnoses and silence in clinical signs, physicians may not be aware of the risk prior to manifesting severe thrombosis events such as PE [9]. In addition, due to the perception of increased post-surgical bleeding, orthopedic surgeons may concern about the use of prophylactic anticoagulant therapy [10,11]. Increasing studies reported that postoperative VTE struck Asian patients receiving hip or knee replacement surgery in a comparable rate with that in Western countries [12–14], suggesting that patients undergoing arthroplasty without VTE prophylaxis in Taiwan can be predisposed to the emerging risk of thrombosis or a higher mortality rate led by PE strikes. However, the incidences of total thrombosis events after major joint surgeries revealed great variation spanning the ranges of 0%-64% for THR, and 11%-77% for TKR [3,4,9]. A previous Asian study also showed an ethnicity-related difference in the incidence
720
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
hypertension claimed within one year prior to the date of surgery were collected. All diseases were determined by the claimed ICD-9CM codes.
of postoperative DVT. In which, a lower rate was found in Koreans (29.8%) than Chinese (45%) or other selected Asian population (46.3%) [14]. Meanwhile, validity of the prior Asian studies was concerned by the small size of population and heterogeneity of diagnoses. Taken together, reevaluating previous assumption about the rareness of VTE after major joint surgeries is warranted for Asians. A large populationbased study is required prior to determining a prophylactic policy preferable to Taiwanese population receiving major joint surgeries. A proper strategy preventing postoperative VTE needs to be determined by the balance between the costs and benefits. Considerations shall also be addressed on the patients vulnerable to adverse outcomes derived from prophylactic intervention such as bleeding, infection and joint stiffness [11]. In addition, results of population-based studies in US [15] and Europe [16,17] appeared no improvement in the rates of symptomatic or fatal PE despite anti-thrombotic prophylaxis has been introduced over a decade. Up to date, there is no study concerning about the prevalence of postoperative PE in Asians. Significant increase in cost for administering aggressive anticoagulant therapy such as low molecular weight heparin (LMWH) leads to financial difficulties of extending the benefit for reimbursement in general public [11,18]. Therefore, identification of independent risk factors associated with the occurrence of post-surgical VTE is a prerequisite to determine an adequate thromboprophylaxis guideline for clinical practice, which is not yet available in Taiwan. In the present study, we aimed to identify the incidences and the risk factors associated with postoperative VTE for Taiwanese patients receiving hip or knee replacement surgery by retrospectively evaluating on the data acquired from the National Health Insurance (NHI) database representing more than 99% population in Taiwan. The cumulative rates as well as the difference in the incidences between inpatient and outpatient visits were also calculated and compared.
Descriptive statistics were used to describe the baseline characteristics. The subjects with postoperative VTE were further stratified to four groups by the days (1–7, 8–14, 15–21 and 22–28 days) of onset since the date of surgery. Theχ2-test was performed to compare characteristics between the VTE and non-VTE groups. Logistic regression model was used for assessing the risk factors associated postoperative VTE for subjects receiving knee and hip surgery. The covariate variables included age, gender, surgery type, hospital level, and prior disease history of VTE, CHF, stroke, TIA, DM, and hypertension. We compared the unadjusted and adjusted patient characteristics between these groups by using odds ratio (OR) and 95% confidence intervals (CI). A p value b 0.05 indicated statistical significance. Data management and logistic regression model was conducted by SAS Version 8.0 (Chicago, Illinois).
Subjects and Methods
Results
Data Acquisition and Validity
Incidence of Postoperative VTE Following Hip Replacement Surgery
This was a retrospective population-based cohort study exploring the authentic database of Bureau of National Health Insurance (BNHI), Taiwan. Since March 1995, a single-payer NHI program was launched nationwide and more than 99% of Taiwanese citizens were mandatory to be insured. The registration files and original claims for reimbursement were included in the computerized database that is being maintained by the government-funded National Health Research Institute (NHRI), Taiwan. The claim files comprise outpatient and inpatient medical records on diagnosis, treatment and all management incurring expenses. Accession to explore the database requires authenticated account exclusively registered for research purposes.
During the 5-year period, data from 61,460 subjects received hip replacement surgeries were analyzed (Table 1). The total incident cases of postoperative VTE accounted for 0.27% (163/61,460) during 2002–2006, showing a significant decreasing trend over these years (p b 0.02). There were significant differences in the distribution of incidence rates in age and gender groups. Our findings revealed that higher incidence rates were found in those aged over 60 years (p b 0.01). Females were more susceptible to postoperative VTE than males (p b 0.02). No significant difference was found in the incidence rates of postoperative VTE following different types of hip replacement surgeries (p b 0.07). Patients with prior history of PE (p b 0.001), DVT (or related venous embolism and thrombosis, p b 0.001), varicose veins of lower extremity (p b 0.001), and transient ischemic attack (p b 0.01) were associated with significantly higher postoperative VTE.
Data Mining The eligible subjects were all NHI insured patients receiving hip or knee surgery during five consecutive years between January 1st, 2002 and December 3, 2006. The inclusion criteria were determined by the claimed ICD-9-CM procedure codes including total hip replacement (81.51), partial hip replacement (81.52), revision of hip replacement (81.53), total knee replacement (81.54), and revision of knee replacement (81.55). All claims required the approval of orthopedic surgeons. Meanwhile, claim records within one year prior to the date of surgery and available postoperative follow-up information for at least 28 days were required for the enrollment. For each enrolled subject, the data including demographic variables, surgery year, and surgery type were extracted and categorized. To identify the diseases associated with the incidence of postoperative VTE, subject data with prior history of VTE, congested heart failure (CHF), stroke, transient ischemic attack (TIA), diabetes mellitus (DM) and
Study Outcomes The primary outcome of the present study was the incidence of VTE determined as PE (415.1) and DVT (415.xx), or other venous embolism and thrombosis (453.xx) such as acute venous embolism, thoracic vein thrombosis in outpatient or inpatient within 28 days after surgery. All claims of PE or DVT related were centrally reviewed and confirmed by independent committees in BNHI prior to registering the approval of reimbursement in the NHI databases. Statistical Analysis
Incidence of Postoperative VTE Following Knee Replacement Surgery As summarized in Table 1, a total of 52,556 patients had knee replacement surgeries during 2002–2006 in Taiwan. Of those, the number of incident cases with postoperative VTE was 335 (0.64%). A significant decreasing trend over the 5 years was observed (p b 0.01). No significant difference was found in the VTE incidences by age or by gender. Patients who received total knee replacement surgery had a significantly higher incidence of developing postoperative VTE than those receiving a revision of replacement (p b 0.05). Moreover, significantly higher incidences were found in the subjects with prior PE (p b 0.001), DVT (p b 0.001), varicose veins of lower extremity (p b 0.05), and congestive heart failure (p b 0.001) following knee replacement surgeries.
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
721
Table 1 Patients’ characteristics and the stratified incidences of venous thromboembolism (VTE) within 28 days after hip or knee replacement surgery. Hip Replacement
Enrolled population Surgery year 2002 2003 2004 2005 2006 Age b50 years old 50-60 years old 60-69 years old 70-79 years old ≧80 years old Gender Female Male Surgery type Total replacement Partial replacement Revision of replacement Prior Disease History Pulmonary embolism DVT⁎ Varicose veins of lower extremity Congestive heart failure Stroke Transient ischemic attack Hypertension Diabetes
Knee Replacement
n (%)
Postoperative VTE n (%)
61,460
163 (0.27)
p
n (%)
Postoperative VTE n (%)
52,566
335 (0.64)
p
12,590 (28.40) 12,264 (19.95) 12,857 (20.92) 12,718 (20.69) 11,031 (17.95)
49 (0.39) 34 (0.28) 30 (0.23) 31 (0.24) 19 (0.17)
0.02
10,826 (20.60) 9,005 (17.13) 11,021 (20.97) 11,283 (21.46) 10,431 (19.84)
108 (1.00) 71 (0.79) 52 (0.47) 55 (0.49) 49 (0.47)
0.001
11,705 (19.04) 7,168 (11.66) 11,451 (18.63) 18,194 (29.60) 12,942 (21.06)
18 (0.15) 13 (0.18) 41 (0.36) 49 (0.27) 42 (0.32)
0.01
1,015 (1.93) 5,552 (10.56) 19,298 (36.71) 22,812 (43.40) 3,889 (7.40)
4 (0.39) 30 (0.54) 123 (0.60) 159 (0.70) 19 (0.49)
0.35
32,487 (52.86) 28,973 (47.14)
101 (0.31) 62 (0.21)
0.02
39,186 (74.55) 13,380 (25.45)
240 (0.61) 95 (0.71)
0.32
20,669 (33.63) 35,072 (57.06) 5,719 (9.31)
41 (0.2) 105 (0.3) 17 (0.3)
0.07
50,089 (95.29)
330 (0.66)
0.01
36 (0.06) 362 (0.59) 245 (0.4) 3,787 (6.16) 1,434 (2.33) 20,276 (32.99) 18,890 (30.74) 10,644 (17.32)
8 (22.22) 34 (9.39) 7 (2.86) 16 (0.42) 6 (0.42) 73 (0.36) 59 (0.31) 32 (0.30)
2,477 (4.71) b0.001 b0.001 b0.001 0.06 0.27 0.001 0.13 0.43
28 (0.05) 417 (0.79) 516 (0.98) 3,244 (6.17) 504 (0.96) 11,592 (22.05) 11,377 (21.64) 25,585 (48.67)
5 (0.20) 4 (14.29) 49 (11.75) 7 (1.36) 37 (1.14) 3 (0.60) 82 (0.71) 70 (0.62) 166 (0.65)
b0.001 b0.001 0.049 b0.001 1.00 0.28 0.74 0.75
⁎ Deep venous thrombosis or other venous embolism and thrombosis.
Incidence of Postoperative PE and DVT Stratified by Inpatients and Outpatients
Crude Odds Ratios of Factors Associated with Postoperative VTE for Hip or Knee Replacement Surgeries
As shown in Table 2, the incidence of postoperative PE was higher among subjects receiving surgeries of knee replacement than those having hip replacement (0.07% vs. 0.04%). Likewise, a higher incidence of DVT was also observed in those underwent knee replacement (0.57% vs. 0.22%). Similar trends concurred that postoperative VTE was found in both inpatient and outpatient visits irrespective of arthroplasty on hip or on knee. Meanwhile, inpatients revealed higher incidence rates of postoperative PE or DVT after both types of arthroplasty procedures. Furthermore, weekly cumulative increases in the incidence of VTE were sustained for up to 28 days after surgeries in both groups of hip and knee replacement (Fig. 1).
Among the subjects receiving hip replacement surgery, the unadjusted analysis showed a significantly higher risk for postoperative VTE in those aged 60–69 (OR = 2.33; 95% CI, 1.34-4.06, p = 0.003) and 70–79 (OR = 1.90; 95%CI, 1.15-3.16, p = 0.01) years compared with those who younger than 50 year old (Table 3). Female gender
Table 2 Stratified incidences of PE or DVT by inpatient or outpatient visits within 28 days after hip or knee replacement surgery.
Total postoperative VTE PE Inpatient Outpatient DVT or related Inpatient Outpatient
Hip Replacement n (%)
Knee Replacement n (%)
163 (0.27) 26 (0.04) 17 (0.03) 9 (0.01) 137 (0.22) 82 (0.13) 55 (0.09)
335 (0.64) 35 (0.07) 29 (0.06) 6 (0.01) 300 (0.57) 201 (0.38) 99 (0.19)
Abbreviations: VTE, venous thromboembolism; PE, pulmonary embolism; DVT, deep venous thrombosis.
Fig. 1. The weekly cumulative incidences of VTE occurred within 28 days after hip and knee replacement surgery.
722
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
Table 3 Crude Odds ratio (OR) of factors associated venous thromboembolism (VTE) within 28 days after hip or knee replacement surgeries. Hip Replacement
Knee Replacement
Crude OR (95%CI) Age b50 years old 50-60 years old 60-69 years old 70-79 years old Gender Male Female Surgery type Total replacement Partial replacement Revision of replacement Prior Disease History Pulmonary embolism DVT⁎ Varicose veins of lower extremity Congestive heart failure Stroke Transient ischemic attack Hypertension Diabetes
p
Crude OR (95%CI)
p
1.00 1.18 (0.58-2.41) 2.33 (1.34-4.06) 1.90 (1.15-3.16)
0.650 0.003 0.010
1.00 1.37 (0.48-3.91) 1.62 (0.60-4.40) 1.70 (0.63-4.58)
0.550 0.340 0.300
1.00 1.45 (1.06-2.00)
0.020
1.00 0.86 (0.68-1.09)
0.220
1.00 1.51 (1.05-2.17) 1.50 (0.85-2.64)
3.28 (1.35-7.94)
0.009
0.030 0.160 b0.001 b0.001 b0.001 0.050 0.260 0.002 0.130 0.430
112.94 (50.68-251.70) 49.00 (33.07-72.60) 11.51 (5.34-24.81) 1.66 (0.99-2.79) 1.60 (0.71-3.63) 1.65 (1.21-2.25) 1.28 (0.93-1.76) 1.17 (0.79-1.72)
1.00 26.33 (9.09-76.26) 24.15 (17.54-33.26) 2.17 (1.02-4.61) 1.90 (1.35-2.68) 0.94 (0.30-2.92) 1.15 (0.89-1.47) 1.04 (0.84-1.28) 0.96 (0.73-1.25)
b0.001 b0.001 0.040 b0.001 0.910 0.280 0.750 0.740
⁎ Deep venous thrombosis or other venous embolism and thrombosis.
(OR = 1.45; 95%CI, 1.06-2.00, p = 0.02) and partial replacement (OR = 1.51; 95%CI, 1.05-2.17, p = 0.03) were also significant risk factors of developing postoperative VTE. Notably, some prior diseases claimed within one year by the date of surgery were significantly associated with increased risk for post-surgical VTE, including PE (OR = 112.94; 95%CI, 50.68-251.70, p b 0.001), DVT (and related venous embolism and thrombosis, OR = 49.0; 95%CI, 33.07-72.60, p b 0.001), varicose veins of lower extremity (OR = 1.66; 95%CI, 0.99-2.79, p b 0.001), and transient ischemic attack (OR = 1.65; 95%CI, 1.21-2.25, p b 0.01). (See Table 4.) For those underwent knee replacement surgery, total replacement resulted in a significantly higher risk for developing postoperative VTE (OR = 3.28; 95% CI, 1.35-7.94, p b 0.01) compared to those having revision of replacement. Prior history of diseases such as PE (OR = 26.33; 95%CI, 9.09-76.26, p b 0.001), DVT and related thrombosis (OR = 24.15; 95%CI, 17.54-33.26, p b 0.001), varicose veins of lower extremity (OR = 2.17; 95%CI, 1.02-4.61, p b 0.05), and congestive heart failure (OR = 1.90; 95%CI, 1.35-2.68) were significant risk factors associated increased incidence of postoperative VTE. Adjusted Odds Ratio of Factors Associated for Postoperative VTE Logistic regression models were performed to adjust confounding bias for OR estimates. The adjusted OR of the patients aged 60–69 years for postoperative VTE following hip replacement was 1.87 (95%CI; 1.05-3.33, p b 0.05). Prior PE (adjusted OR = 39.52; 95%CI, 13.45-116.08, p b 0.001) and DVT (or other venous embolism
and thrombosis, adjusted OR = 35.90; 95%CI, 23.1-55.76, p b 0.01) remained as significant risk factors associated with postoperative VTE after hip replacement. In the patients received knee replacement, the age-related differences were shown no more significant. Significantly higher risk of postoperative VTE was observed in those receiving total knee arthroplasty compared with revision surgery (adjusted OR = 4.17; 95%CI, 1.7010.22, p b 0.01). Compared with the subjects receiving knee replacement surgery in district hospitals, surgeries practiced in medical centers were associated with higher risk of postoperative VTE (adjusted OR = 1.67; 95%CI 1.28-2.19, p b 0.001). Regarding to the effects of prior diseases, PE (adjusted OR = 5.91; 95%CI, 1.06-21.75, p b 0.05), DVT (or related venous embolism and thrombosis (adjusted OR = 23.73; 95%CI, 16.82-33.46, p b 0.001), and congestive heart failure (adjusted OR = 1.61, 95% CI 1.12, 2.31, p = 0.01). Discussion The NHI database covering more than 99% of Taiwanese population is highly representative on general population. Population-based data collection process provides a variety of outpatient and inpatient medical records. Additionally, information such as diagnosis, treatment and clinical adverse effects that required for longitudinal cohort studies can be tracked across time since 1996 [19]. However, some primary features may limit its applications. Firstly, diagnoses in NHI database are encoded by using ICD-9-CM coding scheme, which does not grade the severity of diseases. Secondly, some information such as BMI, smoking
Table 4 Adjusted Odds ratio of significant risk factors for venous thromboembolism (VTE) within 28 days after hip or knee replacement surgery. Hip Replacement Adjusted OR (95% CI) Age 60-69 years old Surgery type Total replacement Prior Disease History Pulmonary embolism DVT⁎ Congestive heart failure
1.87 (1.05-3.33)
Knee Replacement p
Adjusted OR (95% CI) 0.03 4.17 (1.70-10.22)
39.52 (13.45-116.08) 35.90 (23.11-55.76)
⁎ Deep venous thrombosis or other venous embolism and thrombosis.
p
b0.001 0.004
5.91 (1.60-21.75) 23.73 (16.82-33.46) 1.61 (1.12-2.31)
0.002 0.010 b0.001 0.01
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
and family history were not available. Thirdly, NHI database exclusively records the medical applications reimbursed by BNHI. Finally, automatic laboratory data such as lipid profile or D-dimer levels were not recorded from the database. Owning to these limitations, some factors potentially associated with postoperative VTE may be shed. Results of the present study might underestimate the actual rate of postoperative VTE due to the fact of that VTE occurred beyond 28 days after surgery was excluded. Additionally, the study did not collect data by linking to National Mortality File, therefore the involvement of fatal PE in the ratio of VTE was unknown. Regarding to the incidence of postoperative VTE in Asian population, the SMART study [14] demonstrated that in 386 subjects (160 total hip replacement and 226 total knee replacement) the rate of venographyconfirmed VTE was 36.5%, whereas the incidence of symptomatic VTE was 0.9%. There was no discussion about the postoperative occurrence of PE in SMART studies. In another study, Asian patients undergoing orthopedic surgery without receiving thromboprophylaxis revealed a rate of 2.3% for VTE (2,420 patients in 11 countries) [13]. The rates for symptomatic VTE were 1.4% (13/944) for knee replacement or 1.0% (4/408) for hip replacement respectively. The incidence was 0.6% (7/1068) for hip fracture surgery. In terms of symptomatic PE, the rate were 0.3% (3/944) and 0 (0/408) for hip and knee replacement respectively. The SMART study group concluded that the incidence of postoperative VTE in Asia was comparable with that in Western countries and thromboprophylaxis is required for Asian patients. It was noteworthy that the occurrence of post operative PE that may be shed in SMART and other Asian studies. In Korea, there was no sub-clinical or symptomatic PE found in two related studies for patients receiving hip and knee arthroplasty without anticoagulant therapy [20,21]. Meanwhile, high rates of thrombus resolution were reported in DVT from the studies in Korean and Taiwanese population [22]. The population-based investigation in this study revealed that the occurrence rates of postoperative PE were 4 or 7 in 10,000 hip or knee replacement surgeries. However, a study based on National Hospital Discharge Survey demonstrated a general PE incidence of 7 in 100,000 among Asians/Pacific islanders in the United States [23], which was 10 times less than that identified among Taiwanese subjects receiving knee arthroplastic surgery. For Chinese in Hong Kong, the prevalence of total PE was 33 in 100,000 [24], which is lower than that in Taiwanese receiving arthroplasty. Therefore, the results of previous studies in Asian population may lead to negligence on the risk of PE for patients receiving major joint surgeries. Compared with the prevalence of DVT/PE obtained from SMART study, our data demonstrated a higher rate of PE in patients receiving knee replacement surgery that was not found in SMART study (0.07% vs. 0). In US (California), the 90-day non-fatal PE rate was reported as 0.41% in patients received total knee replacement [17]. Despite the occurrence of postoperative PE was relatively uncommon in Asian population compared with that obtained in US, however, negligence on the PE occurred after orthopedic arthroplasty may expose patients to the risk due to high mortality of PE. The previous study [14] based on postoperative screening showed that there was no symptomatic VTE between hospital discharge and 3 months after surgery among 407 Asian patients. In contrast, our study revealed that 1/4 (for hip replacement) and 1/7 (for knee replacement) PE occurred in outpatient visits within succeeding 28 days of surgeries. Likewise, about 65% and 50% DVT (or related venous thrombosis) was accountable to outpatient visits within 28 days after hip and knee surgeries respectively. Therefore, it is noteworthy for the high proportion of postoperative VTE after hospital discharge and post-discharge follow-up is required in Taiwan. Our results concurred with previous reports [10,25–27] and the ACCP guideline [4] recommending extended pharmacological prophylaxis after hospital discharge. The results of our study demonstrated several features that have not been addressed before. Firstly, it was intriguing that type of surgery associated with significantly different incidences of postoperative VTE among the subjects receiving knee arthroplasty. The risk for
723
postoperative VTE was over 4 times higher among subjects receiving total knee replacement (adjusted OR = 4.17; 95% CI 1.70-10.22, p = 0.002) compared with those who underwent revision of replacement. Among the subjects receiving hip replacement surgery, in contrast, the trends in postoperative VTE were likely to be associated with higher incidence in the subgroups receiving partial and revision replacement surgeries despite that did not reach statistical significance while compared with the subgroup of total hip replacement. The outcomes may attribute to different types of anesthetic techniques and/or the duration consumed for different surgical practices. Jaffer and colleagues [28] illustrated evidences that an odds ratio of 3.58 (95%CI, 2.11-6.16, p b 0.001) for the occurrence of postoperative VTE was associated with anesthetic duration longer than 3.5 hours. For the patients undergoing revision surgery, however, a long median time of arthroplastic surgery (252 minutes) countered the impact of long anesthesia duration (≧3.5 hours). In fact, patients receiving regional anesthesia have been reported to be associated with less frequency of postoperative DVT [29]. In the present study, duration of anesthetic duration was not included as a covariate to compare the differences in post-surgical VTE between subgroups of total and revision replacement surgery. Further study is required to determine the conclusive involvement of anesthesia. In addition, knee replacement surgeries practiced in medical centers (university teaching hospitals or equivalent) were significantly associated with higher prevalence of postoperative VTE in comparison with the other inferior hospital levels. It may attribute to the ultimate referral of patients with more complicated situations that may synergize the development of VTE. Furthermore, in terms of the annual incidences of postoperative VTE, both groups of hip and knee replacement surgeries showed significant decreases in 2003 and a mild decreasing trend over the following years. With the limitation of retrospective study design, we were not able to track the determining factors. Finally, compared with the VTE occurrence in patients younger than 50 years old, the patients aged between 60 ~ 69 years revealed a higher rate than the other age groups among those receiving hip replacement, but the difference was not observed in the group of TKR. Concurring with the results of many other studies, claim of PE or DVT (or related venous thrombosis) within 1-year prior to surgery was significantly associated with higher occurrence rates of postoperative VTE [1,30,31]. For the Asian population, TKR and the duration of antibiotics use within 1 week prior to surgery were independent risk factors for VTE or sudden death in the SMART venography study [12]. Chronic heart failure, varicose veins of lower extremity and history of VTE were identified independent risks in SMART symptomatic VTE study [13]. For those who with prior PE or DVT (or other venous embolism and thrombosis) in one year preceding arthroplastic surgery in our study, dramatically higher adjusted ORs were associated hip replacement, implying subjects with prior PE or DVT receiving hip replacement surgeries were more vulnerable to postoperative VTE compared with the matched population underwent knee arthroplasty. Unlike the results obtained from previous SMART study, 1-year prior history of varicose veins of lower extremity preceding surgeries was not indicated as an independent risk factor for the occurrence of VTE for both groups of hip and knee replacement after adjustments. It implied that surgery-related hyper-coagulation status instead of primary venous dysfunction contributed to the higher incidence of VTE after orthopedic arthroplasty in Taiwan. Despite an entirely ideal method of thrombopropphylaxis does not exist, an effective prophylactic modality with fewer side effects, lower cost, and easy in practice and without monitoring requirement shall be seriously considered prior to surgery [11]. Given the prophylaxis rationale derived from the working group consensus in AAOS guideline, orthopedic surgeons were recommended to conduct pre-operative evaluation on all patients to assess the risks of PE and major bleeding complications. Meanwhile, ACCP argued that DVT is a valid surrogate for PE due to proven evidences showing parallel reduction of DVT and PE with administration of anti-thrombosis agents. Regarding to the
724
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
cost-effectiveness and the potential risk of bleeding, subjects receiving preferable agents for thromboprophylaxis shall be prioritized by the severity of thrombotic complications and the predisposed risks. Due to the high mortality and morbidity derived by PE, particular it was previously ignored for Asian population receiving arthroplasty, subjects predisposed to the strike of PE such as having prior history of PE of DVT in 1 year advance of surgery shall be arranged to the superior priority for preventive therapy. In Taiwan, the prevalent rate of postoperative PE among subjects receiving knee arthroplasty was two times higher than those undergoing hip surgery (0.07% vs. 0.04%), while the ratio was about 3 times compared for the incidence of DVT. Additionally, subjects with prior CHF undergoing knee replacement were more frequently struck by postoperative VTE. Total knee replacement was also an independent risk factor associated with postoperative VTE. Considering the agents applied for patients had undergone arthroplasty surgeries, AAOS’s recommendation for use of aspirin at a dose of 325 mg twice daily for six weeks in the elderly was also challenged by ACCP’s formal argument due to high risk in bleeding. Therefore selection of pharmacologic agents preventing the occurrence of postoperative VTE needs to be cautiously determined for each individual. Although anticoagulants have been entered Taiwan market since 1990s (e.g. heparin products; enoxaparin in 1998) and the concept of pharmacological prophylaxis for thromboprophylaxis was promoted for Asians [6]; however, these approaches have not yet been widely implemented in clinical practice and are far from frequently used to improve VTE prevention in Taiwan. Physicians should be encouraged to reconsider the benefits of thromboprophylaxis and apply more aggressive prophylactic approaches (including pharmacological and non-pharmacological methods) to treat patients susceptible to VTE, because treatment is usually safe and successful after excluding the patients with higher risk of major bleeding. In conclusion, we presented a population-based study representing more than 99% of 23 million Taiwanese citizens and demonstrated the incidence rates of postoperative VTE. The risk factors associated with postoperative VTE detected in this study would help predict and prevent VTE for patients in Taiwan. The occurrence of PE leading to mortality that was ignored before shall be emphasized. For the subject highly predisposed to PE after arthroplastic surgeries in Taiwan, we recommend to scrutinize an individualized prophylactic solution acquiring clinical effectiveness but minimal complication of bleeding throughout the peri-operative period. Conflicts of Interest Statement The authors declare no conflicts of interest for this study. References [1] Hitos K, Fletcher JP. Venous thromboembolism following primary total hip arthroplasty. Int Angiol 2009;28:215–21. [2] Haas SB, Barrack RL, Westrich G. Venous thromboembolic disease after total hip and knee arthroplasty. Instr Course Lect 2009;58:781–93. [3] Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126:338S–400S. [4] Colwell CW. The ACCP guidelines for thromboprophylaxis in total hip and knee arthroplasty. Orthopedics 2009;32:67–73. [5] Lachiewicz PF. Prevention of symptomatic pulmonary embolism in patients undergoing total hip and knee arthroplasty: clinical guideline of the American Academy of Orthopaedic Surgeons. Instr Course Lect 2009;58:795–804.
[6] Cohen AT. Asia-Pacific Thrombosis Advisory B. Asia-Pacific Thrombosis Advisory Board consensus paper on prevention of venous thromboembolism after major orthopaedic surgery. Thromb Haemost 2010;104:919–30. [7] Klatsky AL, Armstrong MA, Poggi J. Risk of pulmonary embolism and/or deep venous thrombosis in Asian-Americans. Am J Cardiol 2000;85:1334–7. [8] White RH, Zhou H, Romano PS. Incidence of idiopathic deep venous thrombosis and secondary thromboembolism among ethnic groups in California. Ann Intern Med 1998;128:737–40. [9] Nicolaides AN, Breddin HK, Fareed J, Goldhaber S, Haas S, Hull R, et al. Prevention of venous thromboembolism. International Consensus Statement. Guidelines compiled in accordance with the scientific evidence. Int Angiol 2001;20:1–37. [10] Rahme E, Dasgupta K, Burman M, Yin H, Bernatsky S, Berry G, et al. Postdischarge thromboprophylaxis and mortality risk after hip-or knee-replacement surgery. CMAJ 2008;178:1545–54. [11] Haas SB, Barrack RL, Westrich G, Lachiewicz PF. Venous thromboembolic disease after total hip and knee arthroplasty. J Bone Joint Surg Am 2008;90:2764–80. [12] Leizorovicz A. Epidemiology of post-operative venous thromboembolism in Asian patients. Results of the SMART venography study. Haematologica 2007;92:1194–200. [13] Leizorovicz A, Turpie AG, Cohen AT, Wong L, Yoo MC, Dans A, et al. Epidemiology of venous thromboembolism in Asian patients undergoing major orthopedic surgery without thromboprophylaxis. The SMART study. J Thromb Haemost 2005;3:28–34. [14] Piovella F, Wang CJ, Lu H, Lee K, Lee LH, Lee WC, et al. Deep-vein thrombosis rates after major orthopedic surgery in Asia. An epidemiological study based on postoperative screening with centrally adjudicated bilateral venography. J Thromb Haemost 2005;3:2664–70. [15] Katz JN, Losina E, Barrett J, Phillips CB, Mahomed NN, Lew RA, et al. Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States medicare population. J Bone Joint Surg Am 2001;83A:1622–9. [16] Howie C, Hughes H, Watts AC. Venous thromboembolism associated with hip and knee replacement over a ten-year period: a population-based study. J Bone Joint Surg Br 2005;87:1675–80. [17] SooHoo NF, Lieberman JR, Ko CY, Zingmond DS. Factors predicting complication rates following total knee replacement. J Bone Joint Surg Am 2006;88:480–5. [18] Skedgel C, Goeree R, Pleasance S, Thompson K, O'Brien B, Anderson D. The costeffectiveness of extended-duration antithrombotic prophylaxis after total hip arthroplasty. J Bone Joint Surg Am 2007;89:819–28. [19] Wen CP, Tsai SP, Chung WS. A 10-year experience with universal health insurance in Taiwan: measuring changes in health and health disparity. Ann Intern Med 2008;148:258–67. [20] Kim YH, Oh SH, Kim JS. Incidence and natural history of deep-vein thrombosis after total hip arthroplasty. A prospective and randomised clinical study. J Bone Joint Surg Br 2003;85:661–5. [21] Kim YH, Kim JS. Incidence and natural history of deep-vein thrombosis after total knee arthroplasty. A prospective, randomised study. J Bone Joint Surg Br 2002;84:566–70. [22] Wang CJ, Wang JW, Weng LH, Hsu CC, Lo CF. Outcome of calf deep-vein thrombosis after total knee arthroplasty. J Bone Joint Surg Br 2003;85:841–4. [23] Stein PD, Kayali F, Olson RE, Milford CE. Pulmonary thromboembolism in Asians/ Pacific Islanders in the United States: analysis of data from the National Hospital Discharge Survey and the United States Bureau of the Census. Am J Med 2004;116:435–42. [24] Ko PS, Chan WF, Siu TH, Khoo J, Wu WC, Lam JJ. Deep venous thrombosis after total hip or knee arthroplasty in a "low-risk" Chinese population. J Arthroplasty 2003;18:174–9. [25] Whang PG, Lieberman JR. Extended-duration low-molecular-weight heparin prophylaxis following total joint arthroplasty. Am J Orthop (Belle Mead NJ) 2002;31:31–6. [26] Prandoni P, Bruchi O, Sabbion P, Tanduo C, Scudeller A, Sardella C, et al. Prolonged thromboprophylaxis with oral anticoagulants after total hip arthroplasty: a prospective controlled randomized study. Arch Intern Med 2002;162:1966–71. [27] Arcelus JI, Kudrna JC, Caprini JA. Venous thromboembolism following major orthopedic surgery: what is the risk after discharge? Orthopedics 2006;29:506–16. [28] Jaffer AK, Barsoum WK, Krebs V, Hurbanek JG, Morra N, Brotman DJ. Duration of anesthesia and venous thromboembolism after hip and knee arthroplasty. Mayo Clin Proc 2005;80:732–8. [29] Salvati EA, Pellegrini Jr VD, Sharrock NE, Lotke PA, Murray DW, Potter H, et al. Recent advances in venous thromboembolic prophylaxis during and after total hip replacement. J Bone Joint Surg Am 2000;82:252–70. [30] White RH, Gettner S, Newman JM, Trauner KB, Romano PS. Predictors of rehospitalization for symptomatic venous thromboembolism after total hip arthroplasty. N Engl J Med 2000;343:1758–64. [31] Szucs G, Ajzner E, Muszbek L, Simon T, Szepesi K, Fulesdi B. Assessment of thrombotic risk factors predisposing to thromboembolic complications in prosthetic orthopedic surgery. J Orthop Sci 2009;14:484–90.
Contents lists available at ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
Population-based epidemiology of postoperative venous thromboembolism in Taiwanese patients receiving hip or knee arthroplasty without pharmacological thromboprophylaxis Po-Kuei Wu a,b,c,d, Cheng-Fong Chen b,c,d, Lien-Hsiang Chung c,d,e, Chien-Lin Liu b,d, Wei-Ming Chen b,c,d,⁎ a
Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan, ROC Department of Orthropaedics and Traumatology, Taipei Veterans General Hospital, Taipei, Taiwan, ROC c Therapeutical and Research Center of Musculoskeletal Tumor, Taipei Veterans General Hospital, Taipei, Taiwan, ROC d Orthopaedic Department School of Medicine, National Yang-Ming University e Department of Orthropaedics and Traumatology, Chia-Yi Christian Hospital, Chia-Yi, Taiwan, ROC b
a r t i c l e
i n f o
Article history: Received 16 September 2013 Received in revised form 14 January 2014 Accepted 27 January 2014 Available online 5 February 2014 Keywords: Pulmonary embolism Venous thromboembolism Deep vein thrombosis TKA THR
a b s t r a c t Introduction: Population-based evaluation on the incidence of postoperative venous thromboembolism (VTE) has not yet been reported for Asians receiving arthroplasty. In Taiwan, thromboprophylaxis was not commonly applied for patients. The population-based cohort study aimed to investigate the epidemiology, and to determine the risk factors VTE for patients receiving hip or knee replacement without pharmacological thromboprophylaxis in Taiwan. Materials and Methods: We retrospectively acquired patients’ data from National Health Insurance databases representing more than 99% of about 23 million Taiwanese citizens. The primary outcome was the incidence of composite symptomatic VTE within 28 days after receiving hip or knee replacement surgery. Results: During 2002 to 2006, there were 114,026 patients undergoing hip (n = 61,460) or knee (n = 52,566) replacement surgery. The occurrence rate of overall postoperative VTE was 0.44%. The incidence of pulmonary embolism was four in 10,000 patients receiving hip replacement or seven in 10,000 individuals undergoing knee replacement. The weekly cumulative incidence of VTE was persistently rising up to 28 days after surgery. Dramatic increase in risk of post-surgical VTE was associated with prior disease history of PE (p b 0.001 for hip replacement, p = 0.01 for knee replacement) or DVT (p = 0.004 for hip replacement, p b 0.001 for knee replacement). Prior claim of congestive heart failure was an independent risk factor associated with patients receiving knee arthroplasty (p = 0.01). Conclusion: Life-threatening PE occurred and increased cumulatively up to 28 days after hip or knee arthroplasty in Asians. Proper prophylaxis for patients with the exposure of high risks needs to be scrutinized. © 2014 Elsevier Ltd. All rights reserved.
Introduction Venous thromboembolism (VTE), consisting of pulmonary embolism (PE) and deep vein thrombosis (DVT), is a common postoperative complication for patients receiving hip or knee replacement surgery [1–3]. Standard pharmacological prophylaxis such as anticoagulants, mechanical devices such as calf pump or combined intervention have been practiced for decades. Guideline recommendations for appropriate prophylaxis based on strong evidence have been published by American College of Chest Physicians (ACCP) [4], American Academy of Orthopaedic Surgeons (AAOS) [5], or the consensus of the Asia-Pacific Thrombosis Advisory Board in the countries without their own guidelines [6]. In Taiwan, postoperative thrombosis prophylaxis has not yet ⁎ Corresponding author at: 18F, 201, Section 2, Shin-Pai Road, Taipei 112, Taiwan, ROC. Tel.: +886 2 28757557x117; fax: +886 2 28757559. E-mail addresses: [email protected] (P.-K. Wu), [email protected] (W.-M. Chen). 0049-3848/$ – see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.thromres.2014.01.039
applied for patients receiving hip or knee arthroplasty that may be attributed the following reasons. Firstly, the assumption determined by previous studies indicated that occurrence of post-surgical VTE in Asian population was relatively uncommon [7,8]. Secondly, given the matters of unreliable diagnoses and silence in clinical signs, physicians may not be aware of the risk prior to manifesting severe thrombosis events such as PE [9]. In addition, due to the perception of increased post-surgical bleeding, orthopedic surgeons may concern about the use of prophylactic anticoagulant therapy [10,11]. Increasing studies reported that postoperative VTE struck Asian patients receiving hip or knee replacement surgery in a comparable rate with that in Western countries [12–14], suggesting that patients undergoing arthroplasty without VTE prophylaxis in Taiwan can be predisposed to the emerging risk of thrombosis or a higher mortality rate led by PE strikes. However, the incidences of total thrombosis events after major joint surgeries revealed great variation spanning the ranges of 0%-64% for THR, and 11%-77% for TKR [3,4,9]. A previous Asian study also showed an ethnicity-related difference in the incidence
720
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
hypertension claimed within one year prior to the date of surgery were collected. All diseases were determined by the claimed ICD-9CM codes.
of postoperative DVT. In which, a lower rate was found in Koreans (29.8%) than Chinese (45%) or other selected Asian population (46.3%) [14]. Meanwhile, validity of the prior Asian studies was concerned by the small size of population and heterogeneity of diagnoses. Taken together, reevaluating previous assumption about the rareness of VTE after major joint surgeries is warranted for Asians. A large populationbased study is required prior to determining a prophylactic policy preferable to Taiwanese population receiving major joint surgeries. A proper strategy preventing postoperative VTE needs to be determined by the balance between the costs and benefits. Considerations shall also be addressed on the patients vulnerable to adverse outcomes derived from prophylactic intervention such as bleeding, infection and joint stiffness [11]. In addition, results of population-based studies in US [15] and Europe [16,17] appeared no improvement in the rates of symptomatic or fatal PE despite anti-thrombotic prophylaxis has been introduced over a decade. Up to date, there is no study concerning about the prevalence of postoperative PE in Asians. Significant increase in cost for administering aggressive anticoagulant therapy such as low molecular weight heparin (LMWH) leads to financial difficulties of extending the benefit for reimbursement in general public [11,18]. Therefore, identification of independent risk factors associated with the occurrence of post-surgical VTE is a prerequisite to determine an adequate thromboprophylaxis guideline for clinical practice, which is not yet available in Taiwan. In the present study, we aimed to identify the incidences and the risk factors associated with postoperative VTE for Taiwanese patients receiving hip or knee replacement surgery by retrospectively evaluating on the data acquired from the National Health Insurance (NHI) database representing more than 99% population in Taiwan. The cumulative rates as well as the difference in the incidences between inpatient and outpatient visits were also calculated and compared.
Descriptive statistics were used to describe the baseline characteristics. The subjects with postoperative VTE were further stratified to four groups by the days (1–7, 8–14, 15–21 and 22–28 days) of onset since the date of surgery. Theχ2-test was performed to compare characteristics between the VTE and non-VTE groups. Logistic regression model was used for assessing the risk factors associated postoperative VTE for subjects receiving knee and hip surgery. The covariate variables included age, gender, surgery type, hospital level, and prior disease history of VTE, CHF, stroke, TIA, DM, and hypertension. We compared the unadjusted and adjusted patient characteristics between these groups by using odds ratio (OR) and 95% confidence intervals (CI). A p value b 0.05 indicated statistical significance. Data management and logistic regression model was conducted by SAS Version 8.0 (Chicago, Illinois).
Subjects and Methods
Results
Data Acquisition and Validity
Incidence of Postoperative VTE Following Hip Replacement Surgery
This was a retrospective population-based cohort study exploring the authentic database of Bureau of National Health Insurance (BNHI), Taiwan. Since March 1995, a single-payer NHI program was launched nationwide and more than 99% of Taiwanese citizens were mandatory to be insured. The registration files and original claims for reimbursement were included in the computerized database that is being maintained by the government-funded National Health Research Institute (NHRI), Taiwan. The claim files comprise outpatient and inpatient medical records on diagnosis, treatment and all management incurring expenses. Accession to explore the database requires authenticated account exclusively registered for research purposes.
During the 5-year period, data from 61,460 subjects received hip replacement surgeries were analyzed (Table 1). The total incident cases of postoperative VTE accounted for 0.27% (163/61,460) during 2002–2006, showing a significant decreasing trend over these years (p b 0.02). There were significant differences in the distribution of incidence rates in age and gender groups. Our findings revealed that higher incidence rates were found in those aged over 60 years (p b 0.01). Females were more susceptible to postoperative VTE than males (p b 0.02). No significant difference was found in the incidence rates of postoperative VTE following different types of hip replacement surgeries (p b 0.07). Patients with prior history of PE (p b 0.001), DVT (or related venous embolism and thrombosis, p b 0.001), varicose veins of lower extremity (p b 0.001), and transient ischemic attack (p b 0.01) were associated with significantly higher postoperative VTE.
Data Mining The eligible subjects were all NHI insured patients receiving hip or knee surgery during five consecutive years between January 1st, 2002 and December 3, 2006. The inclusion criteria were determined by the claimed ICD-9-CM procedure codes including total hip replacement (81.51), partial hip replacement (81.52), revision of hip replacement (81.53), total knee replacement (81.54), and revision of knee replacement (81.55). All claims required the approval of orthopedic surgeons. Meanwhile, claim records within one year prior to the date of surgery and available postoperative follow-up information for at least 28 days were required for the enrollment. For each enrolled subject, the data including demographic variables, surgery year, and surgery type were extracted and categorized. To identify the diseases associated with the incidence of postoperative VTE, subject data with prior history of VTE, congested heart failure (CHF), stroke, transient ischemic attack (TIA), diabetes mellitus (DM) and
Study Outcomes The primary outcome of the present study was the incidence of VTE determined as PE (415.1) and DVT (415.xx), or other venous embolism and thrombosis (453.xx) such as acute venous embolism, thoracic vein thrombosis in outpatient or inpatient within 28 days after surgery. All claims of PE or DVT related were centrally reviewed and confirmed by independent committees in BNHI prior to registering the approval of reimbursement in the NHI databases. Statistical Analysis
Incidence of Postoperative VTE Following Knee Replacement Surgery As summarized in Table 1, a total of 52,556 patients had knee replacement surgeries during 2002–2006 in Taiwan. Of those, the number of incident cases with postoperative VTE was 335 (0.64%). A significant decreasing trend over the 5 years was observed (p b 0.01). No significant difference was found in the VTE incidences by age or by gender. Patients who received total knee replacement surgery had a significantly higher incidence of developing postoperative VTE than those receiving a revision of replacement (p b 0.05). Moreover, significantly higher incidences were found in the subjects with prior PE (p b 0.001), DVT (p b 0.001), varicose veins of lower extremity (p b 0.05), and congestive heart failure (p b 0.001) following knee replacement surgeries.
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
721
Table 1 Patients’ characteristics and the stratified incidences of venous thromboembolism (VTE) within 28 days after hip or knee replacement surgery. Hip Replacement
Enrolled population Surgery year 2002 2003 2004 2005 2006 Age b50 years old 50-60 years old 60-69 years old 70-79 years old ≧80 years old Gender Female Male Surgery type Total replacement Partial replacement Revision of replacement Prior Disease History Pulmonary embolism DVT⁎ Varicose veins of lower extremity Congestive heart failure Stroke Transient ischemic attack Hypertension Diabetes
Knee Replacement
n (%)
Postoperative VTE n (%)
61,460
163 (0.27)
p
n (%)
Postoperative VTE n (%)
52,566
335 (0.64)
p
12,590 (28.40) 12,264 (19.95) 12,857 (20.92) 12,718 (20.69) 11,031 (17.95)
49 (0.39) 34 (0.28) 30 (0.23) 31 (0.24) 19 (0.17)
0.02
10,826 (20.60) 9,005 (17.13) 11,021 (20.97) 11,283 (21.46) 10,431 (19.84)
108 (1.00) 71 (0.79) 52 (0.47) 55 (0.49) 49 (0.47)
0.001
11,705 (19.04) 7,168 (11.66) 11,451 (18.63) 18,194 (29.60) 12,942 (21.06)
18 (0.15) 13 (0.18) 41 (0.36) 49 (0.27) 42 (0.32)
0.01
1,015 (1.93) 5,552 (10.56) 19,298 (36.71) 22,812 (43.40) 3,889 (7.40)
4 (0.39) 30 (0.54) 123 (0.60) 159 (0.70) 19 (0.49)
0.35
32,487 (52.86) 28,973 (47.14)
101 (0.31) 62 (0.21)
0.02
39,186 (74.55) 13,380 (25.45)
240 (0.61) 95 (0.71)
0.32
20,669 (33.63) 35,072 (57.06) 5,719 (9.31)
41 (0.2) 105 (0.3) 17 (0.3)
0.07
50,089 (95.29)
330 (0.66)
0.01
36 (0.06) 362 (0.59) 245 (0.4) 3,787 (6.16) 1,434 (2.33) 20,276 (32.99) 18,890 (30.74) 10,644 (17.32)
8 (22.22) 34 (9.39) 7 (2.86) 16 (0.42) 6 (0.42) 73 (0.36) 59 (0.31) 32 (0.30)
2,477 (4.71) b0.001 b0.001 b0.001 0.06 0.27 0.001 0.13 0.43
28 (0.05) 417 (0.79) 516 (0.98) 3,244 (6.17) 504 (0.96) 11,592 (22.05) 11,377 (21.64) 25,585 (48.67)
5 (0.20) 4 (14.29) 49 (11.75) 7 (1.36) 37 (1.14) 3 (0.60) 82 (0.71) 70 (0.62) 166 (0.65)
b0.001 b0.001 0.049 b0.001 1.00 0.28 0.74 0.75
⁎ Deep venous thrombosis or other venous embolism and thrombosis.
Incidence of Postoperative PE and DVT Stratified by Inpatients and Outpatients
Crude Odds Ratios of Factors Associated with Postoperative VTE for Hip or Knee Replacement Surgeries
As shown in Table 2, the incidence of postoperative PE was higher among subjects receiving surgeries of knee replacement than those having hip replacement (0.07% vs. 0.04%). Likewise, a higher incidence of DVT was also observed in those underwent knee replacement (0.57% vs. 0.22%). Similar trends concurred that postoperative VTE was found in both inpatient and outpatient visits irrespective of arthroplasty on hip or on knee. Meanwhile, inpatients revealed higher incidence rates of postoperative PE or DVT after both types of arthroplasty procedures. Furthermore, weekly cumulative increases in the incidence of VTE were sustained for up to 28 days after surgeries in both groups of hip and knee replacement (Fig. 1).
Among the subjects receiving hip replacement surgery, the unadjusted analysis showed a significantly higher risk for postoperative VTE in those aged 60–69 (OR = 2.33; 95% CI, 1.34-4.06, p = 0.003) and 70–79 (OR = 1.90; 95%CI, 1.15-3.16, p = 0.01) years compared with those who younger than 50 year old (Table 3). Female gender
Table 2 Stratified incidences of PE or DVT by inpatient or outpatient visits within 28 days after hip or knee replacement surgery.
Total postoperative VTE PE Inpatient Outpatient DVT or related Inpatient Outpatient
Hip Replacement n (%)
Knee Replacement n (%)
163 (0.27) 26 (0.04) 17 (0.03) 9 (0.01) 137 (0.22) 82 (0.13) 55 (0.09)
335 (0.64) 35 (0.07) 29 (0.06) 6 (0.01) 300 (0.57) 201 (0.38) 99 (0.19)
Abbreviations: VTE, venous thromboembolism; PE, pulmonary embolism; DVT, deep venous thrombosis.
Fig. 1. The weekly cumulative incidences of VTE occurred within 28 days after hip and knee replacement surgery.
722
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
Table 3 Crude Odds ratio (OR) of factors associated venous thromboembolism (VTE) within 28 days after hip or knee replacement surgeries. Hip Replacement
Knee Replacement
Crude OR (95%CI) Age b50 years old 50-60 years old 60-69 years old 70-79 years old Gender Male Female Surgery type Total replacement Partial replacement Revision of replacement Prior Disease History Pulmonary embolism DVT⁎ Varicose veins of lower extremity Congestive heart failure Stroke Transient ischemic attack Hypertension Diabetes
p
Crude OR (95%CI)
p
1.00 1.18 (0.58-2.41) 2.33 (1.34-4.06) 1.90 (1.15-3.16)
0.650 0.003 0.010
1.00 1.37 (0.48-3.91) 1.62 (0.60-4.40) 1.70 (0.63-4.58)
0.550 0.340 0.300
1.00 1.45 (1.06-2.00)
0.020
1.00 0.86 (0.68-1.09)
0.220
1.00 1.51 (1.05-2.17) 1.50 (0.85-2.64)
3.28 (1.35-7.94)
0.009
0.030 0.160 b0.001 b0.001 b0.001 0.050 0.260 0.002 0.130 0.430
112.94 (50.68-251.70) 49.00 (33.07-72.60) 11.51 (5.34-24.81) 1.66 (0.99-2.79) 1.60 (0.71-3.63) 1.65 (1.21-2.25) 1.28 (0.93-1.76) 1.17 (0.79-1.72)
1.00 26.33 (9.09-76.26) 24.15 (17.54-33.26) 2.17 (1.02-4.61) 1.90 (1.35-2.68) 0.94 (0.30-2.92) 1.15 (0.89-1.47) 1.04 (0.84-1.28) 0.96 (0.73-1.25)
b0.001 b0.001 0.040 b0.001 0.910 0.280 0.750 0.740
⁎ Deep venous thrombosis or other venous embolism and thrombosis.
(OR = 1.45; 95%CI, 1.06-2.00, p = 0.02) and partial replacement (OR = 1.51; 95%CI, 1.05-2.17, p = 0.03) were also significant risk factors of developing postoperative VTE. Notably, some prior diseases claimed within one year by the date of surgery were significantly associated with increased risk for post-surgical VTE, including PE (OR = 112.94; 95%CI, 50.68-251.70, p b 0.001), DVT (and related venous embolism and thrombosis, OR = 49.0; 95%CI, 33.07-72.60, p b 0.001), varicose veins of lower extremity (OR = 1.66; 95%CI, 0.99-2.79, p b 0.001), and transient ischemic attack (OR = 1.65; 95%CI, 1.21-2.25, p b 0.01). (See Table 4.) For those underwent knee replacement surgery, total replacement resulted in a significantly higher risk for developing postoperative VTE (OR = 3.28; 95% CI, 1.35-7.94, p b 0.01) compared to those having revision of replacement. Prior history of diseases such as PE (OR = 26.33; 95%CI, 9.09-76.26, p b 0.001), DVT and related thrombosis (OR = 24.15; 95%CI, 17.54-33.26, p b 0.001), varicose veins of lower extremity (OR = 2.17; 95%CI, 1.02-4.61, p b 0.05), and congestive heart failure (OR = 1.90; 95%CI, 1.35-2.68) were significant risk factors associated increased incidence of postoperative VTE. Adjusted Odds Ratio of Factors Associated for Postoperative VTE Logistic regression models were performed to adjust confounding bias for OR estimates. The adjusted OR of the patients aged 60–69 years for postoperative VTE following hip replacement was 1.87 (95%CI; 1.05-3.33, p b 0.05). Prior PE (adjusted OR = 39.52; 95%CI, 13.45-116.08, p b 0.001) and DVT (or other venous embolism
and thrombosis, adjusted OR = 35.90; 95%CI, 23.1-55.76, p b 0.01) remained as significant risk factors associated with postoperative VTE after hip replacement. In the patients received knee replacement, the age-related differences were shown no more significant. Significantly higher risk of postoperative VTE was observed in those receiving total knee arthroplasty compared with revision surgery (adjusted OR = 4.17; 95%CI, 1.7010.22, p b 0.01). Compared with the subjects receiving knee replacement surgery in district hospitals, surgeries practiced in medical centers were associated with higher risk of postoperative VTE (adjusted OR = 1.67; 95%CI 1.28-2.19, p b 0.001). Regarding to the effects of prior diseases, PE (adjusted OR = 5.91; 95%CI, 1.06-21.75, p b 0.05), DVT (or related venous embolism and thrombosis (adjusted OR = 23.73; 95%CI, 16.82-33.46, p b 0.001), and congestive heart failure (adjusted OR = 1.61, 95% CI 1.12, 2.31, p = 0.01). Discussion The NHI database covering more than 99% of Taiwanese population is highly representative on general population. Population-based data collection process provides a variety of outpatient and inpatient medical records. Additionally, information such as diagnosis, treatment and clinical adverse effects that required for longitudinal cohort studies can be tracked across time since 1996 [19]. However, some primary features may limit its applications. Firstly, diagnoses in NHI database are encoded by using ICD-9-CM coding scheme, which does not grade the severity of diseases. Secondly, some information such as BMI, smoking
Table 4 Adjusted Odds ratio of significant risk factors for venous thromboembolism (VTE) within 28 days after hip or knee replacement surgery. Hip Replacement Adjusted OR (95% CI) Age 60-69 years old Surgery type Total replacement Prior Disease History Pulmonary embolism DVT⁎ Congestive heart failure
1.87 (1.05-3.33)
Knee Replacement p
Adjusted OR (95% CI) 0.03 4.17 (1.70-10.22)
39.52 (13.45-116.08) 35.90 (23.11-55.76)
⁎ Deep venous thrombosis or other venous embolism and thrombosis.
p
b0.001 0.004
5.91 (1.60-21.75) 23.73 (16.82-33.46) 1.61 (1.12-2.31)
0.002 0.010 b0.001 0.01
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
and family history were not available. Thirdly, NHI database exclusively records the medical applications reimbursed by BNHI. Finally, automatic laboratory data such as lipid profile or D-dimer levels were not recorded from the database. Owning to these limitations, some factors potentially associated with postoperative VTE may be shed. Results of the present study might underestimate the actual rate of postoperative VTE due to the fact of that VTE occurred beyond 28 days after surgery was excluded. Additionally, the study did not collect data by linking to National Mortality File, therefore the involvement of fatal PE in the ratio of VTE was unknown. Regarding to the incidence of postoperative VTE in Asian population, the SMART study [14] demonstrated that in 386 subjects (160 total hip replacement and 226 total knee replacement) the rate of venographyconfirmed VTE was 36.5%, whereas the incidence of symptomatic VTE was 0.9%. There was no discussion about the postoperative occurrence of PE in SMART studies. In another study, Asian patients undergoing orthopedic surgery without receiving thromboprophylaxis revealed a rate of 2.3% for VTE (2,420 patients in 11 countries) [13]. The rates for symptomatic VTE were 1.4% (13/944) for knee replacement or 1.0% (4/408) for hip replacement respectively. The incidence was 0.6% (7/1068) for hip fracture surgery. In terms of symptomatic PE, the rate were 0.3% (3/944) and 0 (0/408) for hip and knee replacement respectively. The SMART study group concluded that the incidence of postoperative VTE in Asia was comparable with that in Western countries and thromboprophylaxis is required for Asian patients. It was noteworthy that the occurrence of post operative PE that may be shed in SMART and other Asian studies. In Korea, there was no sub-clinical or symptomatic PE found in two related studies for patients receiving hip and knee arthroplasty without anticoagulant therapy [20,21]. Meanwhile, high rates of thrombus resolution were reported in DVT from the studies in Korean and Taiwanese population [22]. The population-based investigation in this study revealed that the occurrence rates of postoperative PE were 4 or 7 in 10,000 hip or knee replacement surgeries. However, a study based on National Hospital Discharge Survey demonstrated a general PE incidence of 7 in 100,000 among Asians/Pacific islanders in the United States [23], which was 10 times less than that identified among Taiwanese subjects receiving knee arthroplastic surgery. For Chinese in Hong Kong, the prevalence of total PE was 33 in 100,000 [24], which is lower than that in Taiwanese receiving arthroplasty. Therefore, the results of previous studies in Asian population may lead to negligence on the risk of PE for patients receiving major joint surgeries. Compared with the prevalence of DVT/PE obtained from SMART study, our data demonstrated a higher rate of PE in patients receiving knee replacement surgery that was not found in SMART study (0.07% vs. 0). In US (California), the 90-day non-fatal PE rate was reported as 0.41% in patients received total knee replacement [17]. Despite the occurrence of postoperative PE was relatively uncommon in Asian population compared with that obtained in US, however, negligence on the PE occurred after orthopedic arthroplasty may expose patients to the risk due to high mortality of PE. The previous study [14] based on postoperative screening showed that there was no symptomatic VTE between hospital discharge and 3 months after surgery among 407 Asian patients. In contrast, our study revealed that 1/4 (for hip replacement) and 1/7 (for knee replacement) PE occurred in outpatient visits within succeeding 28 days of surgeries. Likewise, about 65% and 50% DVT (or related venous thrombosis) was accountable to outpatient visits within 28 days after hip and knee surgeries respectively. Therefore, it is noteworthy for the high proportion of postoperative VTE after hospital discharge and post-discharge follow-up is required in Taiwan. Our results concurred with previous reports [10,25–27] and the ACCP guideline [4] recommending extended pharmacological prophylaxis after hospital discharge. The results of our study demonstrated several features that have not been addressed before. Firstly, it was intriguing that type of surgery associated with significantly different incidences of postoperative VTE among the subjects receiving knee arthroplasty. The risk for
723
postoperative VTE was over 4 times higher among subjects receiving total knee replacement (adjusted OR = 4.17; 95% CI 1.70-10.22, p = 0.002) compared with those who underwent revision of replacement. Among the subjects receiving hip replacement surgery, in contrast, the trends in postoperative VTE were likely to be associated with higher incidence in the subgroups receiving partial and revision replacement surgeries despite that did not reach statistical significance while compared with the subgroup of total hip replacement. The outcomes may attribute to different types of anesthetic techniques and/or the duration consumed for different surgical practices. Jaffer and colleagues [28] illustrated evidences that an odds ratio of 3.58 (95%CI, 2.11-6.16, p b 0.001) for the occurrence of postoperative VTE was associated with anesthetic duration longer than 3.5 hours. For the patients undergoing revision surgery, however, a long median time of arthroplastic surgery (252 minutes) countered the impact of long anesthesia duration (≧3.5 hours). In fact, patients receiving regional anesthesia have been reported to be associated with less frequency of postoperative DVT [29]. In the present study, duration of anesthetic duration was not included as a covariate to compare the differences in post-surgical VTE between subgroups of total and revision replacement surgery. Further study is required to determine the conclusive involvement of anesthesia. In addition, knee replacement surgeries practiced in medical centers (university teaching hospitals or equivalent) were significantly associated with higher prevalence of postoperative VTE in comparison with the other inferior hospital levels. It may attribute to the ultimate referral of patients with more complicated situations that may synergize the development of VTE. Furthermore, in terms of the annual incidences of postoperative VTE, both groups of hip and knee replacement surgeries showed significant decreases in 2003 and a mild decreasing trend over the following years. With the limitation of retrospective study design, we were not able to track the determining factors. Finally, compared with the VTE occurrence in patients younger than 50 years old, the patients aged between 60 ~ 69 years revealed a higher rate than the other age groups among those receiving hip replacement, but the difference was not observed in the group of TKR. Concurring with the results of many other studies, claim of PE or DVT (or related venous thrombosis) within 1-year prior to surgery was significantly associated with higher occurrence rates of postoperative VTE [1,30,31]. For the Asian population, TKR and the duration of antibiotics use within 1 week prior to surgery were independent risk factors for VTE or sudden death in the SMART venography study [12]. Chronic heart failure, varicose veins of lower extremity and history of VTE were identified independent risks in SMART symptomatic VTE study [13]. For those who with prior PE or DVT (or other venous embolism and thrombosis) in one year preceding arthroplastic surgery in our study, dramatically higher adjusted ORs were associated hip replacement, implying subjects with prior PE or DVT receiving hip replacement surgeries were more vulnerable to postoperative VTE compared with the matched population underwent knee arthroplasty. Unlike the results obtained from previous SMART study, 1-year prior history of varicose veins of lower extremity preceding surgeries was not indicated as an independent risk factor for the occurrence of VTE for both groups of hip and knee replacement after adjustments. It implied that surgery-related hyper-coagulation status instead of primary venous dysfunction contributed to the higher incidence of VTE after orthopedic arthroplasty in Taiwan. Despite an entirely ideal method of thrombopropphylaxis does not exist, an effective prophylactic modality with fewer side effects, lower cost, and easy in practice and without monitoring requirement shall be seriously considered prior to surgery [11]. Given the prophylaxis rationale derived from the working group consensus in AAOS guideline, orthopedic surgeons were recommended to conduct pre-operative evaluation on all patients to assess the risks of PE and major bleeding complications. Meanwhile, ACCP argued that DVT is a valid surrogate for PE due to proven evidences showing parallel reduction of DVT and PE with administration of anti-thrombosis agents. Regarding to the
724
P.-K. Wu et al. / Thrombosis Research 133 (2014) 719–724
cost-effectiveness and the potential risk of bleeding, subjects receiving preferable agents for thromboprophylaxis shall be prioritized by the severity of thrombotic complications and the predisposed risks. Due to the high mortality and morbidity derived by PE, particular it was previously ignored for Asian population receiving arthroplasty, subjects predisposed to the strike of PE such as having prior history of PE of DVT in 1 year advance of surgery shall be arranged to the superior priority for preventive therapy. In Taiwan, the prevalent rate of postoperative PE among subjects receiving knee arthroplasty was two times higher than those undergoing hip surgery (0.07% vs. 0.04%), while the ratio was about 3 times compared for the incidence of DVT. Additionally, subjects with prior CHF undergoing knee replacement were more frequently struck by postoperative VTE. Total knee replacement was also an independent risk factor associated with postoperative VTE. Considering the agents applied for patients had undergone arthroplasty surgeries, AAOS’s recommendation for use of aspirin at a dose of 325 mg twice daily for six weeks in the elderly was also challenged by ACCP’s formal argument due to high risk in bleeding. Therefore selection of pharmacologic agents preventing the occurrence of postoperative VTE needs to be cautiously determined for each individual. Although anticoagulants have been entered Taiwan market since 1990s (e.g. heparin products; enoxaparin in 1998) and the concept of pharmacological prophylaxis for thromboprophylaxis was promoted for Asians [6]; however, these approaches have not yet been widely implemented in clinical practice and are far from frequently used to improve VTE prevention in Taiwan. Physicians should be encouraged to reconsider the benefits of thromboprophylaxis and apply more aggressive prophylactic approaches (including pharmacological and non-pharmacological methods) to treat patients susceptible to VTE, because treatment is usually safe and successful after excluding the patients with higher risk of major bleeding. In conclusion, we presented a population-based study representing more than 99% of 23 million Taiwanese citizens and demonstrated the incidence rates of postoperative VTE. The risk factors associated with postoperative VTE detected in this study would help predict and prevent VTE for patients in Taiwan. The occurrence of PE leading to mortality that was ignored before shall be emphasized. For the subject highly predisposed to PE after arthroplastic surgeries in Taiwan, we recommend to scrutinize an individualized prophylactic solution acquiring clinical effectiveness but minimal complication of bleeding throughout the peri-operative period. Conflicts of Interest Statement The authors declare no conflicts of interest for this study. References [1] Hitos K, Fletcher JP. Venous thromboembolism following primary total hip arthroplasty. Int Angiol 2009;28:215–21. [2] Haas SB, Barrack RL, Westrich G. Venous thromboembolic disease after total hip and knee arthroplasty. Instr Course Lect 2009;58:781–93. [3] Geerts WH, Pineo GF, Heit JA, Bergqvist D, Lassen MR, Colwell CW, et al. Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126:338S–400S. [4] Colwell CW. The ACCP guidelines for thromboprophylaxis in total hip and knee arthroplasty. Orthopedics 2009;32:67–73. [5] Lachiewicz PF. Prevention of symptomatic pulmonary embolism in patients undergoing total hip and knee arthroplasty: clinical guideline of the American Academy of Orthopaedic Surgeons. Instr Course Lect 2009;58:795–804.
[6] Cohen AT. Asia-Pacific Thrombosis Advisory B. Asia-Pacific Thrombosis Advisory Board consensus paper on prevention of venous thromboembolism after major orthopaedic surgery. Thromb Haemost 2010;104:919–30. [7] Klatsky AL, Armstrong MA, Poggi J. Risk of pulmonary embolism and/or deep venous thrombosis in Asian-Americans. Am J Cardiol 2000;85:1334–7. [8] White RH, Zhou H, Romano PS. Incidence of idiopathic deep venous thrombosis and secondary thromboembolism among ethnic groups in California. Ann Intern Med 1998;128:737–40. [9] Nicolaides AN, Breddin HK, Fareed J, Goldhaber S, Haas S, Hull R, et al. Prevention of venous thromboembolism. International Consensus Statement. Guidelines compiled in accordance with the scientific evidence. Int Angiol 2001;20:1–37. [10] Rahme E, Dasgupta K, Burman M, Yin H, Bernatsky S, Berry G, et al. Postdischarge thromboprophylaxis and mortality risk after hip-or knee-replacement surgery. CMAJ 2008;178:1545–54. [11] Haas SB, Barrack RL, Westrich G, Lachiewicz PF. Venous thromboembolic disease after total hip and knee arthroplasty. J Bone Joint Surg Am 2008;90:2764–80. [12] Leizorovicz A. Epidemiology of post-operative venous thromboembolism in Asian patients. Results of the SMART venography study. Haematologica 2007;92:1194–200. [13] Leizorovicz A, Turpie AG, Cohen AT, Wong L, Yoo MC, Dans A, et al. Epidemiology of venous thromboembolism in Asian patients undergoing major orthopedic surgery without thromboprophylaxis. The SMART study. J Thromb Haemost 2005;3:28–34. [14] Piovella F, Wang CJ, Lu H, Lee K, Lee LH, Lee WC, et al. Deep-vein thrombosis rates after major orthopedic surgery in Asia. An epidemiological study based on postoperative screening with centrally adjudicated bilateral venography. J Thromb Haemost 2005;3:2664–70. [15] Katz JN, Losina E, Barrett J, Phillips CB, Mahomed NN, Lew RA, et al. Association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States medicare population. J Bone Joint Surg Am 2001;83A:1622–9. [16] Howie C, Hughes H, Watts AC. Venous thromboembolism associated with hip and knee replacement over a ten-year period: a population-based study. J Bone Joint Surg Br 2005;87:1675–80. [17] SooHoo NF, Lieberman JR, Ko CY, Zingmond DS. Factors predicting complication rates following total knee replacement. J Bone Joint Surg Am 2006;88:480–5. [18] Skedgel C, Goeree R, Pleasance S, Thompson K, O'Brien B, Anderson D. The costeffectiveness of extended-duration antithrombotic prophylaxis after total hip arthroplasty. J Bone Joint Surg Am 2007;89:819–28. [19] Wen CP, Tsai SP, Chung WS. A 10-year experience with universal health insurance in Taiwan: measuring changes in health and health disparity. Ann Intern Med 2008;148:258–67. [20] Kim YH, Oh SH, Kim JS. Incidence and natural history of deep-vein thrombosis after total hip arthroplasty. A prospective and randomised clinical study. J Bone Joint Surg Br 2003;85:661–5. [21] Kim YH, Kim JS. Incidence and natural history of deep-vein thrombosis after total knee arthroplasty. A prospective, randomised study. J Bone Joint Surg Br 2002;84:566–70. [22] Wang CJ, Wang JW, Weng LH, Hsu CC, Lo CF. Outcome of calf deep-vein thrombosis after total knee arthroplasty. J Bone Joint Surg Br 2003;85:841–4. [23] Stein PD, Kayali F, Olson RE, Milford CE. Pulmonary thromboembolism in Asians/ Pacific Islanders in the United States: analysis of data from the National Hospital Discharge Survey and the United States Bureau of the Census. Am J Med 2004;116:435–42. [24] Ko PS, Chan WF, Siu TH, Khoo J, Wu WC, Lam JJ. Deep venous thrombosis after total hip or knee arthroplasty in a "low-risk" Chinese population. J Arthroplasty 2003;18:174–9. [25] Whang PG, Lieberman JR. Extended-duration low-molecular-weight heparin prophylaxis following total joint arthroplasty. Am J Orthop (Belle Mead NJ) 2002;31:31–6. [26] Prandoni P, Bruchi O, Sabbion P, Tanduo C, Scudeller A, Sardella C, et al. Prolonged thromboprophylaxis with oral anticoagulants after total hip arthroplasty: a prospective controlled randomized study. Arch Intern Med 2002;162:1966–71. [27] Arcelus JI, Kudrna JC, Caprini JA. Venous thromboembolism following major orthopedic surgery: what is the risk after discharge? Orthopedics 2006;29:506–16. [28] Jaffer AK, Barsoum WK, Krebs V, Hurbanek JG, Morra N, Brotman DJ. Duration of anesthesia and venous thromboembolism after hip and knee arthroplasty. Mayo Clin Proc 2005;80:732–8. [29] Salvati EA, Pellegrini Jr VD, Sharrock NE, Lotke PA, Murray DW, Potter H, et al. Recent advances in venous thromboembolic prophylaxis during and after total hip replacement. J Bone Joint Surg Am 2000;82:252–70. [30] White RH, Gettner S, Newman JM, Trauner KB, Romano PS. Predictors of rehospitalization for symptomatic venous thromboembolism after total hip arthroplasty. N Engl J Med 2000;343:1758–64. [31] Szucs G, Ajzner E, Muszbek L, Simon T, Szepesi K, Fulesdi B. Assessment of thrombotic risk factors predisposing to thromboembolic complications in prosthetic orthopedic surgery. J Orthop Sci 2009;14:484–90.
