Skeletal Radiol DOI 10.1007/s00256-014-2065-5
REVIEW ARTICLE
Risk of bleeding associated with interventional musculoskeletal radiology procedures. A comprehensive review of the literature Gregory B. Foremny & Juan Pretell-Mazzini & Jean Jose & Ty K. Subhawong
Received: 16 August 2014 / Revised: 11 October 2014 / Accepted: 11 November 2014 # ISS 2014
Abstract This review compiles the current literature on the bleeding risks in common musculoskeletal interventional procedures and attempts to provide guidance for practicing radiologists in making decisions regarding the periprocedural management of patients on antithrombotic therapy. The practitioner must weigh the risk of bleeding if therapy is continued against the possibility a thromboembolic occurring if anticoagulation therapy is withheld or reversed. Unfortunately, there is little empirical data to guide evidence-based decisions for many musculoskeletal interventions. However, a review of the literature shows that for low-risk procedures, such as arthrograms/arthrocenteses or muscle/tendon sheath injections, bleeding risks are sufficiently small that anticoagulants and antiplatelet therapies need not be withheld. Additionally, relatively higher-risk procedures, such as needle biopsies of bone and soft tissue, may be safely performed without holding antithrombotic therapy, provided preprocedural INR is within therapeutic range. Thus, while a patient’s particular clinical circumstances should dictate optimal individualized management, anticoagulation alone is not a general contraindication to most interventional musculoskeletal radiology procedures.
Keywords Musculoskeletal intervention . Bleeding risk . Anticoagulation . Warfarin . Arthrogram . Biopsy . G. B. Foremny : J. Jose : T. K. Subhawong (*) Department of Radiology, Jackson Memorial Hospital, University of Miami Miller School of Medicine, 1611 NW 12th Ave., JMH WW 279, Miami, FL 33136, USA e-mail: [email protected] J. Pretell-Mazzini Department of Orthopaedic Surgery-Division of Musculoskeletal Oncology, University of Miami Miller School of Medicine, 1400 NW 12th Avenue, Miami, FL 33136, USA
Arthrocentesis . Joint injection . Peritendinous injection . Bone marrow biopsy
Introduction Improved accuracy and efficacy of procedures performed under image guidance [1, 2] have increased the volume and scope of procedures performed by interventional musculoskeletal (MSK) radiologists. Image-guided arthrograms, arthrocenteses, tendon sheath injections, and biopsies are among the most common MSK interventional procedures. A variety of modalities are being utilized to perform these procedures including fluoroscopy, computed tomography (CT), ultrasound, and, more recently, magnetic resonance imaging (MRI) [3]. Not only has image guidance been shown to improve the accuracy of musculoskeletal procedures [4], it also (particularly with ultrasound) reduces the risk of bleeding complications by allowing identification of vessels in the planned needle path. In addition to inadvertent vessel injury, the other major risk for excessive bleeding is antithrombotic therapy; this is a commonly encountered subset of the patient population given that there are approximately 2.5 million people on chronic anticoagulation therapy in the United States alone [5]. The risk of bleeding, if therapy is continued, must be weighed against the possibility a thromboembolic event occurring if therapy is withheld or reversed [6]. Published guidelines for certain minimally invasive procedures support the continuation of chronic anticoagulation therapy periprocedurally [5, 7–12]. Although guidelines have been established pertaining to the perioperative management of antithrombotic therapy for major surgeries by the American College of Chest Physicians [13], American Academy of Orthopaedic Surgeons [14], and general interventional radiology procedures by the Society of Interventional Radiology (SIR) [15], there is a lack of
Skeletal Radiol
literature with respect to musculoskeletal interventional procedures. Thus, controversy may arise in optimal management of these patients. This review will focus on the literature pertaining to the bleeding risks for some of the most common musculoskeletal interventional procedures, including arthrograms/arthrocenteses, nerve or muscle/tendon sheath injections, and biopsies of bone and soft tissue, with attention to periprocedural management of patients on antithrombotic therapy, with the goal of assisting the practicing radiologist in the clinical decision-making process.
Antithrombotic strategies While it is beyond the scope of this review to discuss in detail the variety of antithrombotic agents currently in wide clinical use, those most commonly encountered can be classified as affecting either platelets (aspirin, clopidogrel, etc.), or inhibiting factors in the pathway of clot formation (warfarin, heparin, etc). Direct thrombin inhibitors, such as dabigatran, offer the ostensible advantage of obviating the need for anticoagulation monitoring, and are thought to have lower absolute rates of both fatal and traumatic intracranial hemorrhages than warfarin [16, 17]; however, these agents may increase gastrointestinal bleeding risk [18]. For patients on antiplatelet therapy, it is often advised that aspirin is held the day prior to and/or the day of the procedure; however, aspirin irreversibly inactivates platelet function, and it requires at least 3 days after stopping aspirin therapy for platelet function to return to normal, and up to 10 days for clopidogrel [19]. As outlined later in this review, however, stopping aspirin for low-risk procedures may be unnecessary. Because of a paucity of literature pertaining to the periprocedural management of oral anticoagulants in musculoskeletal interventional procedures, practice patterns vary considerably. For warfarin, current strategies span the treatment spectrum, and include selecting one of the following: (1) continuing oral anticoagulation therapy; (2) bridging, where a short-acting anticoagulant such as low molecular weight heparin (LMWH) can be used to provide temporary therapy while oral anticoagulants are held; (3) discontinuing anticoagulation therapy; or (4) reversing oral anticoagulation with fresh frozen plasma (FFP) or vitamin K [5, 15]. This variation in periprocedural management has been shown to be independent of clinical characteristics of patients [20]. Interestingly, a recent meta-analysis that included 2,321 patients undergoing implantation of cardiac rhythm devices concluded that uninterrupted warfarin therapy was associated with a lower bleeding risk than bridging heparin, without an increased risk of thromboembolic events [21]. Consensus guidelines from the American College of Chest Physicians are to continue warfarin therapy at a lower dose (INR goal of 1.3–1.5) for low-risk invasive procedures, including gynecologic and orthopedic
patients [22]. The conclusion that continuation of warfarin therapy is feasible for these minor surgeries could arguably be extended to the majority of procedures performed by the interventional musculoskeletal radiologist. Suggestions for management of anticoagulation therapy for some of the most common interventional MSK procedures are summarized in Table 1. Additionally, Table 2 provides a clear overview of the bleeding complication rate associated with these procedures. Phlebotomy It is a common policy at many institutions or imaging centers to check labs—namely platelet count and INR—before performing an MSK image-guided intervention. However, venipuncture itself is not without its own risks. A study of 4,050 out-patients (3,200 men and 850 women) on the risks of bleeding and other complications from routine phlebotomy revealed that 12.3 % of patients had minor bruising and/or hematoma at the site of venipuncture. Overall, more serious complications occurred in 3.4 % of patients, including diaphoresis with hypotension in 2.6 %, and syncope in less than 1 % [40]. It should be noted, however, that these “serious” complications would likely only be scored 1/5 in severity by the National Institutes of Health's Common Terminology Criteria for Adverse Events (CTCAE), version 3.0: Grading of Nonspecific Hematoma [41]. Direct and indirect costs to the patient aside, it may be imprudent to require labs routinely when the bleeding risk posed by phlebotomy itself exceeds the bleeding risk of the MSK interventional procedure. This reasoning justifies the separation of MSK procedures into those of low and high risk with regard to bleeding, which is done while acknowledging that this attempt at risk stratification is relative and arbitrary (i.e., what is “high” risk for the musculoskeletal radiologist may be considered moderate or low risk for the surgeon). However, for the purposes of this article, high-risk procedures differ primarily from low-risk procedures in one of two ways: either solid tissue is mechanically retrieved—specifically, bone and soft tissue biopsies require the use of a cutting needle or drill for the removal of specimens—or the procedure is performed in a location, such as the spine, where even small hematomas could have dramatic adverse consequences.
Procedures with low bleeding risk Arthrograms/arthrocentesis/joint injections Arthrograms and arthrocentesis are frequently performed by MSK radiologists under image guidance, and generally pose an almost vanishingly small bleeding risk. In a survey of 134 radiologists who had collectively performed approximately
Skeletal Radiol Table 1
Suggestions for the management of anticoagulation therapy in interventional MSK procedures
Procedure
Key articles
Labs
Medications Aspirin
Warfarin
Clopidogrel LMWH (therapeutic dose)
Arthrocentesis
25–27, 29
None routinely
Do not hold Do not hold
Do not hold Do not hold
Aspiration of fluid collection
15, 32
INR only if on warfarin
Do not hold Target INR
REVIEW ARTICLE
Risk of bleeding associated with interventional musculoskeletal radiology procedures. A comprehensive review of the literature Gregory B. Foremny & Juan Pretell-Mazzini & Jean Jose & Ty K. Subhawong
Received: 16 August 2014 / Revised: 11 October 2014 / Accepted: 11 November 2014 # ISS 2014
Abstract This review compiles the current literature on the bleeding risks in common musculoskeletal interventional procedures and attempts to provide guidance for practicing radiologists in making decisions regarding the periprocedural management of patients on antithrombotic therapy. The practitioner must weigh the risk of bleeding if therapy is continued against the possibility a thromboembolic occurring if anticoagulation therapy is withheld or reversed. Unfortunately, there is little empirical data to guide evidence-based decisions for many musculoskeletal interventions. However, a review of the literature shows that for low-risk procedures, such as arthrograms/arthrocenteses or muscle/tendon sheath injections, bleeding risks are sufficiently small that anticoagulants and antiplatelet therapies need not be withheld. Additionally, relatively higher-risk procedures, such as needle biopsies of bone and soft tissue, may be safely performed without holding antithrombotic therapy, provided preprocedural INR is within therapeutic range. Thus, while a patient’s particular clinical circumstances should dictate optimal individualized management, anticoagulation alone is not a general contraindication to most interventional musculoskeletal radiology procedures.
Keywords Musculoskeletal intervention . Bleeding risk . Anticoagulation . Warfarin . Arthrogram . Biopsy . G. B. Foremny : J. Jose : T. K. Subhawong (*) Department of Radiology, Jackson Memorial Hospital, University of Miami Miller School of Medicine, 1611 NW 12th Ave., JMH WW 279, Miami, FL 33136, USA e-mail: [email protected] J. Pretell-Mazzini Department of Orthopaedic Surgery-Division of Musculoskeletal Oncology, University of Miami Miller School of Medicine, 1400 NW 12th Avenue, Miami, FL 33136, USA
Arthrocentesis . Joint injection . Peritendinous injection . Bone marrow biopsy
Introduction Improved accuracy and efficacy of procedures performed under image guidance [1, 2] have increased the volume and scope of procedures performed by interventional musculoskeletal (MSK) radiologists. Image-guided arthrograms, arthrocenteses, tendon sheath injections, and biopsies are among the most common MSK interventional procedures. A variety of modalities are being utilized to perform these procedures including fluoroscopy, computed tomography (CT), ultrasound, and, more recently, magnetic resonance imaging (MRI) [3]. Not only has image guidance been shown to improve the accuracy of musculoskeletal procedures [4], it also (particularly with ultrasound) reduces the risk of bleeding complications by allowing identification of vessels in the planned needle path. In addition to inadvertent vessel injury, the other major risk for excessive bleeding is antithrombotic therapy; this is a commonly encountered subset of the patient population given that there are approximately 2.5 million people on chronic anticoagulation therapy in the United States alone [5]. The risk of bleeding, if therapy is continued, must be weighed against the possibility a thromboembolic event occurring if therapy is withheld or reversed [6]. Published guidelines for certain minimally invasive procedures support the continuation of chronic anticoagulation therapy periprocedurally [5, 7–12]. Although guidelines have been established pertaining to the perioperative management of antithrombotic therapy for major surgeries by the American College of Chest Physicians [13], American Academy of Orthopaedic Surgeons [14], and general interventional radiology procedures by the Society of Interventional Radiology (SIR) [15], there is a lack of
Skeletal Radiol
literature with respect to musculoskeletal interventional procedures. Thus, controversy may arise in optimal management of these patients. This review will focus on the literature pertaining to the bleeding risks for some of the most common musculoskeletal interventional procedures, including arthrograms/arthrocenteses, nerve or muscle/tendon sheath injections, and biopsies of bone and soft tissue, with attention to periprocedural management of patients on antithrombotic therapy, with the goal of assisting the practicing radiologist in the clinical decision-making process.
Antithrombotic strategies While it is beyond the scope of this review to discuss in detail the variety of antithrombotic agents currently in wide clinical use, those most commonly encountered can be classified as affecting either platelets (aspirin, clopidogrel, etc.), or inhibiting factors in the pathway of clot formation (warfarin, heparin, etc). Direct thrombin inhibitors, such as dabigatran, offer the ostensible advantage of obviating the need for anticoagulation monitoring, and are thought to have lower absolute rates of both fatal and traumatic intracranial hemorrhages than warfarin [16, 17]; however, these agents may increase gastrointestinal bleeding risk [18]. For patients on antiplatelet therapy, it is often advised that aspirin is held the day prior to and/or the day of the procedure; however, aspirin irreversibly inactivates platelet function, and it requires at least 3 days after stopping aspirin therapy for platelet function to return to normal, and up to 10 days for clopidogrel [19]. As outlined later in this review, however, stopping aspirin for low-risk procedures may be unnecessary. Because of a paucity of literature pertaining to the periprocedural management of oral anticoagulants in musculoskeletal interventional procedures, practice patterns vary considerably. For warfarin, current strategies span the treatment spectrum, and include selecting one of the following: (1) continuing oral anticoagulation therapy; (2) bridging, where a short-acting anticoagulant such as low molecular weight heparin (LMWH) can be used to provide temporary therapy while oral anticoagulants are held; (3) discontinuing anticoagulation therapy; or (4) reversing oral anticoagulation with fresh frozen plasma (FFP) or vitamin K [5, 15]. This variation in periprocedural management has been shown to be independent of clinical characteristics of patients [20]. Interestingly, a recent meta-analysis that included 2,321 patients undergoing implantation of cardiac rhythm devices concluded that uninterrupted warfarin therapy was associated with a lower bleeding risk than bridging heparin, without an increased risk of thromboembolic events [21]. Consensus guidelines from the American College of Chest Physicians are to continue warfarin therapy at a lower dose (INR goal of 1.3–1.5) for low-risk invasive procedures, including gynecologic and orthopedic
patients [22]. The conclusion that continuation of warfarin therapy is feasible for these minor surgeries could arguably be extended to the majority of procedures performed by the interventional musculoskeletal radiologist. Suggestions for management of anticoagulation therapy for some of the most common interventional MSK procedures are summarized in Table 1. Additionally, Table 2 provides a clear overview of the bleeding complication rate associated with these procedures. Phlebotomy It is a common policy at many institutions or imaging centers to check labs—namely platelet count and INR—before performing an MSK image-guided intervention. However, venipuncture itself is not without its own risks. A study of 4,050 out-patients (3,200 men and 850 women) on the risks of bleeding and other complications from routine phlebotomy revealed that 12.3 % of patients had minor bruising and/or hematoma at the site of venipuncture. Overall, more serious complications occurred in 3.4 % of patients, including diaphoresis with hypotension in 2.6 %, and syncope in less than 1 % [40]. It should be noted, however, that these “serious” complications would likely only be scored 1/5 in severity by the National Institutes of Health's Common Terminology Criteria for Adverse Events (CTCAE), version 3.0: Grading of Nonspecific Hematoma [41]. Direct and indirect costs to the patient aside, it may be imprudent to require labs routinely when the bleeding risk posed by phlebotomy itself exceeds the bleeding risk of the MSK interventional procedure. This reasoning justifies the separation of MSK procedures into those of low and high risk with regard to bleeding, which is done while acknowledging that this attempt at risk stratification is relative and arbitrary (i.e., what is “high” risk for the musculoskeletal radiologist may be considered moderate or low risk for the surgeon). However, for the purposes of this article, high-risk procedures differ primarily from low-risk procedures in one of two ways: either solid tissue is mechanically retrieved—specifically, bone and soft tissue biopsies require the use of a cutting needle or drill for the removal of specimens—or the procedure is performed in a location, such as the spine, where even small hematomas could have dramatic adverse consequences.
Procedures with low bleeding risk Arthrograms/arthrocentesis/joint injections Arthrograms and arthrocentesis are frequently performed by MSK radiologists under image guidance, and generally pose an almost vanishingly small bleeding risk. In a survey of 134 radiologists who had collectively performed approximately
Skeletal Radiol Table 1
Suggestions for the management of anticoagulation therapy in interventional MSK procedures
Procedure
Key articles
Labs
Medications Aspirin
Warfarin
Clopidogrel LMWH (therapeutic dose)
Arthrocentesis
25–27, 29
None routinely
Do not hold Do not hold
Do not hold Do not hold
Aspiration of fluid collection
15, 32
INR only if on warfarin
Do not hold Target INR
