Concerns Regarding the Use of Oral Anticoagulants (Rivaroxaban and Apixaban) for Venous Thromboembolism Prophylaxis in Plastic Surgery Patients

Morales et al1 recommend oral factor Xa inhibitors to reduce the risk of venous thromboembolism (VTE) in plastic surgery patients. The authors1 explain that institutional review board approval was not obtained for their study because the investigators are in private practice. Institutional review board approval is not limited to academic institutions.2 Medical journals insist on institutional review board approval to protect patients from unnecessary risks.3 Institutional review boards accredited by the Association for the Accreditation of Human Research Protection Programs are available to surgeons in private practice.2 I am also in private practice and obtained institutional review board approval for my investigation of VTE in plastic surgery patients using diagnostic ultrasound.4 The Declaration of Helsinki guidelines are meant for international authors.3 American investigators need to follow the regulations of the Department of Health and Human Services.3 Both sets of guidelines insist on informed patient consent and oversight by an independent review board. Surprisingly, published studies1,5-7 of plastic surgery patients treated with rivaroxaban including a prospective randomized study,5 have had no institutional review board approval. Retrospective studies analyzing existing data may receive an exemption,8 but this waiver is still needed. Morales et al1 report that informed consent was obtained. However, it is not clear that written consent for participation in a study of factor Xa inhibitors was obtained in addition to the usual surgical consent forms. Were patients informed of the specific risks associated with these medications and their right to decline to take part in the study? The authors1 claim that enoxaparin lowers the VTE risk after abdominoplasty. In fact, the referenced study9 found no risk reduction among abdominoplasty patients who did not also undergo belt lipectomies. Abdominoplasty patients experienced a 5% risk of VTE despite administration of enoxaparin, plus a significantly increased risk of bleeding.9 The study by Hatef et al9 included preoperative and intraoperative doses of enoxaparin, possibly increasing the bleeding risk. However, other studies of patients who started taking rivaroxaban after surgery also document an increased bleeding risk.5,7 Enoxaparin, rivaroxaban, and apixaban are approved by the FDA for VTE prophylaxis in patients undergoing knee or hip replacement, not plastic surgery operations such as large-volume liposuction or body contouring procedures.1 The authors1 suggest that their use is simply an off-label application. Off-label use of botulinum toxin or fillers is typically harmless; this is not true for anticoagulation. No antidote is available yet for these oral anticoagulants,1 unlike enoxaparin or unfractionated heparin. The authors1 expect a reversal agent to be available soon. However, there are no still no studies demonstrating a reduced risk of VTE using either rivaroxaban or apixaban in patients undergoing plastic surgical procedures. Patronella, a coauthor,1 stopped using rivaroxaban because of a 2.8% hematoma rate, which he considers unacceptable.7 This relevant observation is not mentioned in this article.1 The authors1 discontinued using rivaroxaban during their study and started using apixaban instead. Morales et al1 recognize that their study is inadequately powered to determine whether there is a difference in efficacy (or even if there is any efficacy in view of the lack of a control group) or safety between medications. When using a medication off-label, the application must be based on sound medical evidence as part of the practice of medicine.10 It is not clear that the patient characteristics and procedures are similar to approved uses. A young woman undergoing an outpatient breast augmentation is much different from an elderly patient undergoing a hip replacement in hospital. Ultrasound findings suggest that the pathophysiology of VTE in plastic surgery patients differs from joint replacement patients.11 Unlike joint replacement patients, deep venous thromboses typically do not develop intraoperatively during plastic surgery procedures.11 The authors double the usual prophylactic dose of apixaban for patients with body mass indices >30 kg/m2, possibly meeting the requirements of an investigational use.10 Such a dosage adjustment may exceed the scope of off-label use, and require an approved study protocol. Hematomas are not trivial and should not be regarded as an acceptable trade for an unproven VTE risk reduction.12 Today, dozens of law firms solicit patients who experience bleeding after taking rivaroxaban, alleging that patients are not informed of the risk of uncontrolled hemorrhage. If plastic surgery patients are informed that (1) there is no proven benefit in reducing VTE risk, (2) bleeding risk is increased, and (3) there is no antidote, how many patients would agree to take oral anticoagulants, particularly if they are offered a risk-free alternative? When given the choice, >90% of patients prefer ultrasound surveillance to routine anticoagulation.4,11 Bleeding, hematomas, blood transfusions, and unplanned hospital admissions should be rare events after elective cosmetic surgery. In my study of 551 consecutive body contouring cases that included 167 abdominoplasties, none of whom received chemoprophylaxis, only one VTE was encountered (0.2%).13 There were no hematomas. The authors1 use general endotracheal anesthesia for all cases and anticoagulate routinely, without individual risk stratification. Even breast augmentation patients are anticoagulated.1 Empirical evidence strongly suggests that the anesthesia method is related to VTE risk.12 Intravenous anesthesia without paralysis of the calf muscle pump is preferred.12-14 Chemoprophylaxis is not recommended for non-risk-stratified plastic surgery patients.14 The authors1 believe that Caprini scores have received “formal validation.” However, their validity has been challenged.15 Caprini scores do not consider the type of anesthesia. Recently, Shaikh et al16 were unable to find a significant difference in VTE risk in patients compared by Caprini scores. Remarkably, there were no VTEs among all 36 patients with Caprini scores >10.16 No significant treatment benefit has been reported in risk-stratified plastic surgery patients.14,17,18 Studies consistently find that individual risk stratification has a false positive rate of 97%.16,17,19 These findings should not come as a surprise. Caprini scores do not correlate with known relative risk values.15 The authors1 repeat the old controversy about increased risk when combining abdominoplasty with other procedures. In fact, Hester et al20 did not find a significant difference in the VTE rate in patients undergoing combined procedures. The authors1 recommend ultrasound surveillance for future studies without referencing existing studies using this device.4,11 They dismiss ultrasound imaging as financially prohibitive.1 Three ultrasound scans cost about $200 per patient.15 If one considers the medical costs incurred by the 20 study patients with complications1 (not counting the distress to the patient and surgeon), the financial impact of ultrasound surveillance seems trivial. Clinical signs of deep venous thromboses are notoriously unreliable.11 It is too late to wait until a dangerous thrombosis has propagated before initiating diagnostic studies. By embracing new technology, we can truly improve patient safety and avoid unnecessary risks. The author declared no potential conflicts of interest with respect to the research, authorship, and publication of this article. The author received no financial support for the research, authorship, and publication of this article.