J Orthop Sci DOI 10.1007/s00776-015-0762-0
ORIGINAL ARTICLE
Opportunities for improvement: anticoagulation in patients undergoing total knee or hip replacement Mads Thorsberger1 · Thomas Jensen1 · Jonas Olesen1 · Peter Weeke1 · Liv Wahlsten1 · Per Jensen1 · Gunnar Gislason1 · Christian Torp‑Pedersen1 · Lars Koeber1 · Emil Fosbøl1
Received: 2 April 2015 / Accepted: 1 August 2015 © The Japanese Orthopaedic Association 2015
Abstract Background According to guidelines, initiation of pharmacological thromboprophylaxis after total knee or hip replacement is recommended and associated with improved patient outcomes. However, data on how these recommendations are followed in clinical practice are sparse. Methods All patients undergoing first-time total knee or hip replacement in Denmark 2008–2011 were identified. By cross-referencing Danish nationwide registries, quantitative use of anticoagulants administered orally and subcutaneously post-discharge was assessed by number of claimed prescriptions after surgery. Logistic regression analysis was used to identify factors associated with initiation of prophylaxis. Results A total of 50,389 patients were included in the study (median age 69, inter-quartile range 62–76). Novel oral anticoagulants were initiated in 14.7 % of the patients and heparins/fondaparinux in 2.3 % of the patients. The use of anticoagulants increased from 6.3 % in 2008 to 30.0 % of patients in 2011. Among patients initiating prophylaxis with a novel oral anticoagulant post-discharge, almost all were treated according to guidelines in terms of treatment duration. Factors significantly associated with an increased chance of prophylaxis among total hip replacement patients were: age (per 10-year increments) and female gender. Conclusions Use of pharmacological thromboprophylaxis after total knee or hip replacement was low, but increasing during the study period. This is probably due to increased availability of novel oral anticoagulants. Further initiatives
* Mads Thorsberger [email protected] 1
Department of Cardiology, Copenhagen University Hospital Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark
to increase guideline recommended use of prophylactic anticoagulation after orthopaedic surgery are warranted.
Introduction Total knee or hip replacement are associated with an increased risk of venous thromboembolic complications (i.e., deep venous thromboembolism and pulmonary embolism) and American College of Chest Physicians’ (ACCP) 2008 guidelines as well as current guidelines recommend post-surgical thromboprophylaxis for at least 10 days and up to 35 days in these patients [1, 2]. Annually, in the United States more than 800,000 patients undergo total knee or hip replacement [3] and prior studies have shown a thromboembolism event rate of 15–30 % without prophylaxis [4–7]. Anticoagulant pharmacotherapy after surgery, among other interventions (early mobilization and mechanical compression [3, 8]), lowers this associated risk [9], and low molecular weight heparins/fondaparinux along with adjusted dose warfarin have traditionally been the drugs of choice. However, a new generation of oral anticoagulants has, since 2008, gained approval for thromboprophylaxis after surgery. While randomized trials have shown the novel oral anticoagulants (NOACs) to have an effect equal to or better than traditional thromboprophylactic drugs in patients undergoing total knee or hip replacement [10–17], data regarding the use of these drugs and contemporary treatment trends are sparse. In addition, it is unknown how many patients are treated with NOACs after surgery as recommended by guidelines. To address this gap in knowledge, we used the nationwide Danish registries to examine longitudinal use of anticoagulants among patients undergoing first-time total knee or hip replacement.
13
M. Thorsberger et al.
The Danish guidelines are provided in collaboration with the Danish Ministry of Health and recommend thromboprophylaxis with dabigatran for 10 days after total knee replacement and 28–35 days after total hip replacement. Rivaroxaban is recommended for a duration of 2 weeks after total knee replacement and 5 weeks after total hip replacement. Warfarin and aspirin are not recommended as standard prophylaxis. We set out to study the following three objectives: (1) how many patients receive NOACs after discharge from surgery admission in accordance with current guideline recommendations, (2) are there temporal differences in the type of anticoagulation initiated; and (3) which factors are associated with initiation of pharmacological thromboprophylaxis?
with fractures resulting in joint replacement were not included in the study population. A prescription fill for warfarin, dabigatran, rivaroxaban, apixaban, heparins (low molecular weight and unfractionated) or fondaparinux prior to 7 days and up to 180 days before surgery for total hip replacement or total knee replacement was interpreted as treatment due to another indication (e.g., atrial fibrillation) and resulted in exclusion of the patient from the study cohort. Also, patients with a prescription fill for warfarin 7 days prior to admission for total knee replacement or total knee replacement and up to 35 days post-discharge was interpreted as treatment due to another indication (e.g. newly diagnosed atrial fibrillation or thromboembolic events). These patients were excluded from the study cohort (N = 396). Study covariates
Methods Data sources All Danish residents are provided with a unique personal registration number. In this study we used the civil registration system to link individual-level data from several national administrative registries. Any contact with a Danish hospital is recorded in the National Patient Registry [18], and discharge diagnoses have been registered according to the current International Classification of Diseases, version 10, since 1994. All contacts are recorded with one primary diagnosis and one or more secondary diagnoses, if relevant. Similarly, all surgical procedures are registered using Current Procedural Terminology (CPT) codes. The Danish Anaesthesiology Database has held patient related information pre-, per-, and post-anaesthesia for surgical procedures in Denmark since 2004, including information on body mass index (BMI) and history of tobacco use. These data were missing at a higher rate than other clinical variables. The Danish Registry of Medicinal Product Statistics holds information on all prescription fills since 1995 including dispensing date, quantity, and dose classified according to the Anatomical Therapeutic Chemical (ATC) classification system. Pharmacies in Denmark are by law mandated to register all prescriptions claimed in order for the state to manage patient reimbursement, thus yielding a high validity of information [19]. Vital status was obtained from the Danish vital status registry. Study population All patients in Denmark undergoing elective first-time total knee or hip replacement between January 1, 2008 and December 31, 2011 were included in the study. Patients
13
Based on ATC codes, we identified initiation of the following anticoagulants: dabigatran (B01AE07), rivaroxaban (B01AF01 and B01AX06), apixaban (B01AF02 and B01A), warfarin (B01AA03), heparins (B01AB) or fondaparinux (B01AX05). The NOACs, dabigatran, rivaroxaban, and apixaban, were further stratified according to prescribed dose. Duration of prophylaxis was calculated per drug by prescription filling patterns, the prescribed dose, and the quantity, as done previously [20, 21]. In Denmark, approvals of use of NOACs were: dabigatran from September 18th, 2008; rivaroxaban from January 8th, 2009; and apixaban from June 29th, 2011. All NOACs were reimbursed from date of approval. The period in which anticoagulation prescription fills were considered as initiated per indication of thrombophrophylaxis ranged from 7 days before admission to 35 days post-discharge. The Danish guidelines recommend thromboprophylaxis with dabigatran for 10 days after total knee replacement and 28–35 days after total hip replacement. Rivaroxaban is recommended for a duration of 2 weeks after total knee replacement and 5 weeks after total hip replacement. Comorbidities were identified using hospitalization diagnoses up to 5 years prior to the date of surgery. These included: deep vein thrombosis, pulmonary embolism, stroke, ischemic heart disease, peripheral arterial disease, myocardial infarction, heart failure, cancer, atrial fibrillation/atrial flutter and valvular disease. Concomitant pharmacotherapy was defined using prescription fills up to 180 days prior to the admission date for total knee or hip replacement and included: glucose lowering medications (A10), statins (C10AA), NSAID (M01A), loop diuretics (C03C), other diuretics (C02L, C03A, C03B, C03D, C03E, C03X, C07B, C07C, C07D, C08G, C02DA, C09BA, C09DA), beta blockers (C07A,
Opportunities for improvement: anticoagulation in patients undergoing total knee or hip…
C07B, C07C, C07D, C07F), calcium channel blockers (C08, C07F, C09BB, C09DB), renin–angiotensin system inhibitors (C09AA, C09BA, C09BB, C09CA, C09DA, C09DB, C09XA02, C09XA52), clopidogrel (B01AC04), aspirin (B01AC06, N02BA01), dipyridamole (B01AC07). Hypertension was defined by use of two or more antihypertensive drugs simultaneously, as done previously [22]. Factors analysed for association with initiation of NOAC or heparins/fondaparinux vs no prophylaxis included: age per 10-year increments; female gender; length of admission; a history of deep vein thrombosis, pulmonary embolism, stroke, peripheral arterial disease, ischemic heart disease, myocardial infarction, heart failure, hypertension, cancer, atrial fibrillation/atrial flutter, valvular disease; concomitant pharmacotherapy 180 days prior to admission — glucose lowering medications, statins, NSAID, loop diuretics, other diuretics, beta blockers, calcium channel blockers, renin–angiotensin system inhibitors, clopidogrel, aspirin, dipyramidole. Statistics Baseline characteristics were compared using the Kruskal–Wallis test for continuous variables and χ2 tests for categorical variables. For temporal trends, we compared binary variables and continuous variables (calendar year) using Wilcoxon rank-sum tests. Logistic regression analysis was used to identify factors associated with initiation of NOAC or heparins/fondaparinux prophylaxis vs no prophylaxis. Significant results were defined as a p value
ORIGINAL ARTICLE
Opportunities for improvement: anticoagulation in patients undergoing total knee or hip replacement Mads Thorsberger1 · Thomas Jensen1 · Jonas Olesen1 · Peter Weeke1 · Liv Wahlsten1 · Per Jensen1 · Gunnar Gislason1 · Christian Torp‑Pedersen1 · Lars Koeber1 · Emil Fosbøl1
Received: 2 April 2015 / Accepted: 1 August 2015 © The Japanese Orthopaedic Association 2015
Abstract Background According to guidelines, initiation of pharmacological thromboprophylaxis after total knee or hip replacement is recommended and associated with improved patient outcomes. However, data on how these recommendations are followed in clinical practice are sparse. Methods All patients undergoing first-time total knee or hip replacement in Denmark 2008–2011 were identified. By cross-referencing Danish nationwide registries, quantitative use of anticoagulants administered orally and subcutaneously post-discharge was assessed by number of claimed prescriptions after surgery. Logistic regression analysis was used to identify factors associated with initiation of prophylaxis. Results A total of 50,389 patients were included in the study (median age 69, inter-quartile range 62–76). Novel oral anticoagulants were initiated in 14.7 % of the patients and heparins/fondaparinux in 2.3 % of the patients. The use of anticoagulants increased from 6.3 % in 2008 to 30.0 % of patients in 2011. Among patients initiating prophylaxis with a novel oral anticoagulant post-discharge, almost all were treated according to guidelines in terms of treatment duration. Factors significantly associated with an increased chance of prophylaxis among total hip replacement patients were: age (per 10-year increments) and female gender. Conclusions Use of pharmacological thromboprophylaxis after total knee or hip replacement was low, but increasing during the study period. This is probably due to increased availability of novel oral anticoagulants. Further initiatives
* Mads Thorsberger [email protected] 1
Department of Cardiology, Copenhagen University Hospital Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark
to increase guideline recommended use of prophylactic anticoagulation after orthopaedic surgery are warranted.
Introduction Total knee or hip replacement are associated with an increased risk of venous thromboembolic complications (i.e., deep venous thromboembolism and pulmonary embolism) and American College of Chest Physicians’ (ACCP) 2008 guidelines as well as current guidelines recommend post-surgical thromboprophylaxis for at least 10 days and up to 35 days in these patients [1, 2]. Annually, in the United States more than 800,000 patients undergo total knee or hip replacement [3] and prior studies have shown a thromboembolism event rate of 15–30 % without prophylaxis [4–7]. Anticoagulant pharmacotherapy after surgery, among other interventions (early mobilization and mechanical compression [3, 8]), lowers this associated risk [9], and low molecular weight heparins/fondaparinux along with adjusted dose warfarin have traditionally been the drugs of choice. However, a new generation of oral anticoagulants has, since 2008, gained approval for thromboprophylaxis after surgery. While randomized trials have shown the novel oral anticoagulants (NOACs) to have an effect equal to or better than traditional thromboprophylactic drugs in patients undergoing total knee or hip replacement [10–17], data regarding the use of these drugs and contemporary treatment trends are sparse. In addition, it is unknown how many patients are treated with NOACs after surgery as recommended by guidelines. To address this gap in knowledge, we used the nationwide Danish registries to examine longitudinal use of anticoagulants among patients undergoing first-time total knee or hip replacement.
13
M. Thorsberger et al.
The Danish guidelines are provided in collaboration with the Danish Ministry of Health and recommend thromboprophylaxis with dabigatran for 10 days after total knee replacement and 28–35 days after total hip replacement. Rivaroxaban is recommended for a duration of 2 weeks after total knee replacement and 5 weeks after total hip replacement. Warfarin and aspirin are not recommended as standard prophylaxis. We set out to study the following three objectives: (1) how many patients receive NOACs after discharge from surgery admission in accordance with current guideline recommendations, (2) are there temporal differences in the type of anticoagulation initiated; and (3) which factors are associated with initiation of pharmacological thromboprophylaxis?
with fractures resulting in joint replacement were not included in the study population. A prescription fill for warfarin, dabigatran, rivaroxaban, apixaban, heparins (low molecular weight and unfractionated) or fondaparinux prior to 7 days and up to 180 days before surgery for total hip replacement or total knee replacement was interpreted as treatment due to another indication (e.g., atrial fibrillation) and resulted in exclusion of the patient from the study cohort. Also, patients with a prescription fill for warfarin 7 days prior to admission for total knee replacement or total knee replacement and up to 35 days post-discharge was interpreted as treatment due to another indication (e.g. newly diagnosed atrial fibrillation or thromboembolic events). These patients were excluded from the study cohort (N = 396). Study covariates
Methods Data sources All Danish residents are provided with a unique personal registration number. In this study we used the civil registration system to link individual-level data from several national administrative registries. Any contact with a Danish hospital is recorded in the National Patient Registry [18], and discharge diagnoses have been registered according to the current International Classification of Diseases, version 10, since 1994. All contacts are recorded with one primary diagnosis and one or more secondary diagnoses, if relevant. Similarly, all surgical procedures are registered using Current Procedural Terminology (CPT) codes. The Danish Anaesthesiology Database has held patient related information pre-, per-, and post-anaesthesia for surgical procedures in Denmark since 2004, including information on body mass index (BMI) and history of tobacco use. These data were missing at a higher rate than other clinical variables. The Danish Registry of Medicinal Product Statistics holds information on all prescription fills since 1995 including dispensing date, quantity, and dose classified according to the Anatomical Therapeutic Chemical (ATC) classification system. Pharmacies in Denmark are by law mandated to register all prescriptions claimed in order for the state to manage patient reimbursement, thus yielding a high validity of information [19]. Vital status was obtained from the Danish vital status registry. Study population All patients in Denmark undergoing elective first-time total knee or hip replacement between January 1, 2008 and December 31, 2011 were included in the study. Patients
13
Based on ATC codes, we identified initiation of the following anticoagulants: dabigatran (B01AE07), rivaroxaban (B01AF01 and B01AX06), apixaban (B01AF02 and B01A), warfarin (B01AA03), heparins (B01AB) or fondaparinux (B01AX05). The NOACs, dabigatran, rivaroxaban, and apixaban, were further stratified according to prescribed dose. Duration of prophylaxis was calculated per drug by prescription filling patterns, the prescribed dose, and the quantity, as done previously [20, 21]. In Denmark, approvals of use of NOACs were: dabigatran from September 18th, 2008; rivaroxaban from January 8th, 2009; and apixaban from June 29th, 2011. All NOACs were reimbursed from date of approval. The period in which anticoagulation prescription fills were considered as initiated per indication of thrombophrophylaxis ranged from 7 days before admission to 35 days post-discharge. The Danish guidelines recommend thromboprophylaxis with dabigatran for 10 days after total knee replacement and 28–35 days after total hip replacement. Rivaroxaban is recommended for a duration of 2 weeks after total knee replacement and 5 weeks after total hip replacement. Comorbidities were identified using hospitalization diagnoses up to 5 years prior to the date of surgery. These included: deep vein thrombosis, pulmonary embolism, stroke, ischemic heart disease, peripheral arterial disease, myocardial infarction, heart failure, cancer, atrial fibrillation/atrial flutter and valvular disease. Concomitant pharmacotherapy was defined using prescription fills up to 180 days prior to the admission date for total knee or hip replacement and included: glucose lowering medications (A10), statins (C10AA), NSAID (M01A), loop diuretics (C03C), other diuretics (C02L, C03A, C03B, C03D, C03E, C03X, C07B, C07C, C07D, C08G, C02DA, C09BA, C09DA), beta blockers (C07A,
Opportunities for improvement: anticoagulation in patients undergoing total knee or hip…
C07B, C07C, C07D, C07F), calcium channel blockers (C08, C07F, C09BB, C09DB), renin–angiotensin system inhibitors (C09AA, C09BA, C09BB, C09CA, C09DA, C09DB, C09XA02, C09XA52), clopidogrel (B01AC04), aspirin (B01AC06, N02BA01), dipyridamole (B01AC07). Hypertension was defined by use of two or more antihypertensive drugs simultaneously, as done previously [22]. Factors analysed for association with initiation of NOAC or heparins/fondaparinux vs no prophylaxis included: age per 10-year increments; female gender; length of admission; a history of deep vein thrombosis, pulmonary embolism, stroke, peripheral arterial disease, ischemic heart disease, myocardial infarction, heart failure, hypertension, cancer, atrial fibrillation/atrial flutter, valvular disease; concomitant pharmacotherapy 180 days prior to admission — glucose lowering medications, statins, NSAID, loop diuretics, other diuretics, beta blockers, calcium channel blockers, renin–angiotensin system inhibitors, clopidogrel, aspirin, dipyramidole. Statistics Baseline characteristics were compared using the Kruskal–Wallis test for continuous variables and χ2 tests for categorical variables. For temporal trends, we compared binary variables and continuous variables (calendar year) using Wilcoxon rank-sum tests. Logistic regression analysis was used to identify factors associated with initiation of NOAC or heparins/fondaparinux prophylaxis vs no prophylaxis. Significant results were defined as a p value
