Annals of Pharmacotherapy2015, Vol. 49(6) 735 –739© The Author(s) 2015Reprints and permissions: sagepub.com/journalsPermissions.navDOI: 10.1177/1060028015579424aop.sagepub.com

Review Article

Introduction

Heparin-induced thrombocytopenia (HIT) is an immune-mediated, prothrombotic adverse reaction that occurs after exposure to negatively charged polyanions such as unfrac-tionated heparin (UFH) and low-molecular-weight hepa-rin.1,2 When these polyanions form a multimolecular complex with platelet factor 4 (PF4), the result is antibody formation.3 That is to say, HIT antibodies are not directed toward the heparin molecule alone but rather toward the heparin complex with positively charged PF4. HIT IgG antibodies together with heparin-PF4 complex cross-link platelet FcγIIa (IgG) receptors, resulting in platelet activa-tion and release of additional PF4.4,5 With continued hepa-rin exposure, there is additional formation of antigenic complexes, propagating a sequence of platelet activation, aggregation, and generation of procoagulant, platelet-derived microparticles.6,7 This process results in platelet consumption and thrombin generation, leading to both thrombocytopenia and potentially HIT-associated venous and arterial thrombosis.8-11

In acute HIT, IgG antibodies can cross-react with PF4 bound to platelets without the need for the presence of additional heparin. During this period, when the sera of patients are incubated with platelets, the platelets are acti-vated even without the presence of heparin.12 Additionally, approximately 50% of patients who did not experience HIT-associated thrombosis at the time of diagnosis devel-oped delayed onset HIT, occurring within a month after cessation of heparin.6,13 This occurrence demonstrates why stopping heparin is not sufficient to prevent progression of HIT and describes the rationale for initiating an alternative anticoagulant to counter the effects of the autoimmune cascade.

579424 AOPXXX10.1177/1060028015579424Annals of PharmacotherapyMiyares and Davisresearch-article2015

1Jackson Memorial Hospital, Miami, FL, USA

Corresponding Author:Marta A. Miyares, Pharmacy Department, Jackson Memorial Hospital, 1611 N W 12 Avenue, East Tower Basement Room, 069, Miami, FL 33136-109, USA. Email: mmiyares@jhsmiami.org

Direct-Acting Oral Anticoagulants as Emerging Treatment Options for Heparin-Induced Thrombocytopenia

Marta A. Miyares, PharmD1, and Kyle A. Davis, PharmD1

AbstractObjective: To review the evidence for the use of the direct-acting oral anticoagulants (DOACs) in adult patients with heparin-induced thrombocytopenia (HIT). Data Source: A PubMed search (1950-February 2015) was collected using the terms heparin-induced thrombocytopenia, with dabigatran, rivaroxaban, or apixaban, or heparin-induced thrombocytopenia and target-specific anticoagulants, or heparin-induced thrombocytopenia and direct-acting oral anticoagulants, or heparin-induced thrombocytopenia and new oral anticoagulants. Study Selection and Data Extraction: All English-language articles were reviewed for inclusion. The references of included articles were reviewed for additional data. Data Synthesis: HIT is an immune-mediated, prothrombotic adverse reaction that requires not only discontinuation of heparin but also initiation of an alternative nonheparin anticoagulant to counter the effects of the autoimmune cascade. Pharmacotherapeutic management with argatroban is unpredictable and problematic. The DOACs display predictable pharmacokinetic and pharmacodynamic profiles and exhibit no interaction with platelet factor 4. Currently, the DOACs are approved by the Food and Drug Administration for venous thromboembolism, yet have limited evidence in both in vitro and clinical HIT studies. Conclusions: Though dabigatran, rivaroxaban, and apixaban have been used in case reports, currently data are not yet sufficient to recommend clinical use of these agents in the management of HIT. Future trial results may further substantiate management of HIT with use of the DOACs.

Keywordsapixaban, dabigatran, rivaroxaban, heparin-induced thrombocytopenia, direct oral anticoagulants

at UTHSC Library on November 12, 2015aop.sagepub.comDownloaded from

736 Annals of Pharmacotherapy 49(6)

Because patients with HIT are at high risk of developing thrombosis, and it is not always possible to initially identify these patients, it is recommended to treat all patients pro-phylactically with a nonheparin anticoagulant while await-ing lab confirmation or exclusion of the diagnosis. Because low-molecular-weight heparin has a high rate of interacting with HIT antibodies, it is contraindicated in patients with HIT. Treatment options that have been used for HIT include argatroban, lepirudin, desirudin, danaparoid, bivalirudin, and fondaparinux. Whereas 2 of the direct thrombin inhibi-tors, desirudin and bivalirudin, may be used off label for HIT, only bivalirudin has a labeled indication for HIT asso-ciated with percutaneous coronary intervention. According to the 2012 Antithrombotic Therapy and Prevention of Thrombosis guidelines, in patients with HIT with thrombo-sis or isolated HIT who have normal renal function, the use of argatroban, lepirudin, or danaparoid are recommended over other nonheparin anticoagulants. In patients with con-firmed isolated HIT or HIT with thrombosis, initiation of a vitamin K antagonist (VKA; eg, warfarin) is recommended after the platelet count normalizes, and because of its slow onset of action, it should be overlapped with a nonheparin anticoagulant for a minimum of 5 days. HIT-related throm-bosis should be treated for at least 3 months with a VKA. In patients with HIT and thrombosis who have renal insuffi-ciency, argatroban is recommended over other nonheparin anticoagulants.14 As lepirudin and danaparoid are not avail-able in the United States, limited options remain (ie, argatroban).

Argatroban is a synthetic, direct, highly-selective throm-bin inhibitor that produces an increase in nitric oxide. It reversibly binds to the active thrombin site of free and clot-associated thrombin. Argatroban has a half-life elimination of 39 to 51 minutes and attains steady-state levels within 1 to 3 hours after initiation of the continuous intravenous infusion. Because argatroban increases the activated partial thromboplastin time and international normalized ratio (INR) in a dose-dependent manner, the combined effect of argatroban and VKA on the INR should be considered; dis-continuing argatroban when the INR exceeds 4. After remeasuring the INR within 4 to 6 hours after discontinuing argatroban, if it is below the target range, argatroban may be restarted. This process should be repeated until the target INR is achieved with VKA alone.

As argatroban is used primarily for the niche indication of HIT, prescribers’ experience with this medication is lim-ited. Consequently, clinicians’ lack of understanding in monitoring argatroban can potentially increase the risk for bleeding or thrombosis as a result of suboptimal dosing. Hursting et al15 demonstrated that in a subgroup of patients who were transitioned from argatroban to warfarin, 7 of 16 new thrombotic episodes occurred on the day after argatro-ban was discontinued. Of the patients who had an adverse event during transition, 70% of the patients received <5 days of treatment with argatroban, which was attributed to

misinterpretation of a high INR as indicative of therapeutic anticoagulation with warfarin.15 Therefore, proper manage-ment and a rapid diagnosis of HIT are imperative. Additionally, the cost of this anticoagulant is considerable, and if used inappropriately, can bear significant financial implications. Use of direct-acting oral anticoagulants (DOACs) can potentially alleviate many of these concerns because no monitoring is required, and there is ease of con-version between other anticoagulants and DOACs.

DOACs have reached the market and received Food and Drug Administration (FDA) approval for various indica-tions, including nonvalvular atrial fibrillation, deep-vein thrombosis, pulmonary embolism, and postoperative venous thromboprophylaxis. These drugs display predict-able pharmacokinetic and pharmacodynamic profiles.16 Rivaroxaban and apixaban directly inhibit activated factor X, whereas dabigatran is a direct thrombin inhibitor, similar to argatroban. Because of their structure, rivaroxaban and dabigatran are not expected to have an interaction with PF4.17

Objective

The objective of this article is to review the evidence for the use of dabigatran, rivaroxaban, and apixaban in adults with HIT.

Data Sources and Selection

An English language literature search was performed with MEDLINE/PubMed from January 1950 to February 28, 2015, using the search terms heparin-induced thrombocyto-penia, with dabigatran, rivaroxaban, or apixaban, or hepa-rin-induced thrombocytopenia and target-specific anticoagulants, or heparin-induced thrombocytopenia and direct-acting oral anticoagulants, or heparin-induced thrombocytopenia and new oral anticoagulants to identify relevant articles. The references in the retrieved articles were also reviewed to identify other pertinent articles.

Dabigatran

The use of dabigatran in HIT has been described both in vitro and clinically. Krauel et al17 evaluated the interaction of dabigatran with the PF4/heparin complex and its effect on platelet activation via anti-PF4/heparin antibodies in vitro. Dabigatran elicited no effect on PF4 or PF4/heparin complex binding to platelets. Additionally, antibody bind-ing to the PF4/heparin complex was uninfluenced by the presence of dabigatran. Finally, dabigatran had no effect on platelet activation via anti-PF4/heparin antibodies, a funda-mental process in the development of HIT. The authors concluded that dabigatran may represent an alternative treatment option for patients with a history of HIT who require anticoagulation therapy.17

at UTHSC Library on November 12, 2015aop.sagepub.comDownloaded from

Miyares and Davis 737

Currently, 2 case reports (Table 1) exist evaluating the clinical use of dabigatran in HIT. The first reported case involved the use of dabigatran in a patient with nonvalvular atrial fibrillation who underwent coronary artery bypass surgery. A 70-year-old man with a significant history of vas-cular disease presented with a complaint of chest pain. On admission, the patient was initiated on heparin, which was continued for 6 days while the patient was evaluated for surgery. Following surgery, the patient developed thrombo-cytopenia, and by postoperative day 2, the platelet count had diminished by 29%. Because the patient had nonvalvu-lar atrial fibrillation and there was suspicion for HIT, anti-coagulation therapy was initiated with dabigatran 150 mg twice daily on postoperative day 2. On postoperative day 7, the enzyme immunoassay heparin PF4 antibody test recorded a positive result (1.583 on the reactivity index). On postoperative day 8, the patient was transitioned from dabigatran to warfarin and discharged. At discharge (day 8), the patient’s platelet count had increased from a nadir of 80 × 109/µL to 144 × 109/µL. The patient did not demonstrate any evidence of thrombosis during his hospital stay. Although this case report is promising, the outcome should be taken with caution because the patient had a 4T pretest probability of “low,” and a confirmatory serotonin release assay was not performed.18

The second of these cases involved enoxaparin-induced HIT in a 67-year-old female patient who underwent ortho-pedic surgery following a left femoral fracture. The patient received thromboprophylaxis for 6 days prior to her proce-dure. Five days following the procedure, the patient devel-oped symptoms of a deep-vein thrombosis in her left thigh, which was subsequently confirmed with Doppler sonogra-phy. Two weeks following initiation of enoxaparin therapy, the platelet count had diminished from 173 × 109/µL to 32 × 109/µL. A diagnosis of HIT with thrombosis was made, and enoxaparin was discontinued. The patient was initiated on dabigatran 110 mg twice daily. Three days after initiation of dabigatran, the platelet count recovered to 236 × 109/µL. By day 10 of treatment, the patient expressed improvement of symptoms, and imaging revealed recanalization of the thrombosis.19 Although these cases are promising, only short-term outcomes of dabigatran therapy are discussed.

Rivaroxaban

Of the 3 currently approved DOACs, rivaroxaban is the most studied in the setting of HIT. Two studies have examined the effect of rivaroxaban on PF4 and platelets in vitro. Walenga et al20 assessed the comparative interactions of rivaroxaban, UFH, enoxaparin, fondaparinux, and argatroban with HIT

Table 1. Patient Cases of DOAC Therapy in HIT.18,19,21,22,25

Trial Study Type nIndication for

AnticoagulationDiagnostic Test/

Diagnosis Treatment Regimen

HIT Phase DOAC Initiated Outcomes

Fieland and Taylor18

Case report 1 Nonvalvular atrial fibrillation status post-CABG

PF4 antibody/HIT Dabigatran 150 mg twice daily

Acute No thrombotic or bleeding events reported

Mirdamadi19 Case report 1 Status post–orthopedic surgery

Clinical findings and lab results/HIT with thrombosis

Dabigatran 110 mg twice daily

Acute Recanalization of thrombosis reported

Hantson et al21

Case report 1 Status post–orthopedic surgery

PF4 antibody/HIT with thrombosis

Rivaroxaban 15 mg twice daily ×21 days; 20 mg daily thereafter

Acute Recanalization of thrombus reported

Ng et al22 Case series 3 Hemodialysis; status post–embolectomy for critical limb ischemia; pulmonary embolism, and deep-vein thrombosis

PF4 antibody/HIT with thrombosis (2); isolated HIT (1)

Rivaroxaban 10 mg daily transitioned to warfarin; rivaroxaban 15 mg twice daily ×3 weeks followed by 20 mg daily; 15 mg twice daily

Acute No thrombotic or bleeding events reported

Sharifi et al25 Case series 12 Not specified Not specified Apixaban 5 mg twice daily (n = 3); dabigatran 150 mg twice daily (n = 2); rivaroxaban 20 mg daily (n = 7)

Subacute No thrombotic or bleeding events reported

Abbreviations: CABG, coronary artery bypass graft; DOAC, direct-acting oral anticoagulants; HIT, heparin-induced thrombocytopenia; PF4, platelet factor 4.

at UTHSC Library on November 12, 2015aop.sagepub.comDownloaded from

738 Annals of Pharmacotherapy 49(6)

antibodies. Unlike UFH and enoxaparin, rivaroxaban did not cause platelet activation or aggregation in the presence of HIT antibodies. Additionally, rivaroxaban did not interact with PF4, nor did it trigger the release of PF4 from platelets.20 The second of these in vitro studies echoed similar results. Rivaroxaban exhibited no effect on PF4 or anti-PF4/heparin antibody–mediated platelet interaction.17

Several case reports (Table 1) have illustrated the use of rivaroxaban in acute HIT, and there is currently a prospec-tive, multicenter, single-arm cohort study evaluating its use in this arena. The successful use of rivaroxaban in HIT with thrombosis was described in a 36-year-old man who under-went orthopedic surgery 4 days following a traumatic fall. On admission, the platelet count was 166 × 109/µL, which increased to 280 × 109/µL following the procedure. During this period, the patient received nadroparin for prevention of venous thromboembolism. Nine days following initia-tion of nadroparin, the platelet count began to decrease sig-nificantly, reaching a nadir of 25 × 109/µL by day 12. Following positive PF4 and heparin antibody tests, nad-roparin was discontinued, and fondaparinux was initiated. The patient was noted to have a 4T pretest probability of “high,” and 3 days later, Doppler ultrasonography revealed a significant radial artery thrombosis. A diagnosis of HIT-induced thrombosis was made based on these findings, and rivaroxaban 15 mg twice daily was initiated, followed by 20 mg daily at day 22. The platelet count increased 4 days fol-lowing initiation of rivaroxaban and normalized at 10 days. Recanalization of the thrombus was noted at 1 month and reconfirmed at 2 months.21

The authors of 1 case series describe the use of rivaroxa-ban in 3 patients with acute HIT, 2 of whom experienced HIT with thrombosis. The 2 patients with thrombosis devel-oped right lower-limb deep-vein thrombosis as a result of heparin exposure, whereas the third patient developed HIT on receiving enoxaparin for the treatment of a pulmonary embolism. HIT was confirmed in each patient because each tested positive for anti-PF4/heparin antibodies. Rivaroxaban was initiated at platelet counts of 69 × 109/µL, 20 × 109/µL, and 28 × 109/µL, respectively. Two of the 3 patients received chronic therapy with rivaroxaban, and 1 patient was bridged to warfarin because she was on dialysis. There were no reports of thrombotic complications or bleeding episodes associated with rivaroxaban; however, these patients were only followed for 1 to 2 weeks.22

A single arm cohort study evaluating the incidence of new symptomatic venous and arterial thromboembolism in a cohort of patients with suspected and confirmed HIT at 30 days is currently under way. The study will enroll patients with an intermediate or high 4T pretest probability to receive rivaroxaban 15 mg twice daily. Patients with con-firmed HIT and no thrombosis will continue to receive 15 mg twice daily until platelet recovery (platelet count ≥150 × 109/µL or to baseline platelet count); thereafter, patients will receive rivaroxaban 20 mg once daily until study day 30. In

those HIT-positive patients with thrombosis, rivaroxaban 15 mg twice daily will be administered for a minimum of 21 days or until platelet recovery. Thereafter, rivaroxaban 20 mg daily will be administered for a minimum of 3 months. Measured outcomes will include major thrombotic and bleeding events occurring within 30 days of therapy initia-tion. The targeted sample size is 200 patients, and results are expected in 2016. These results should confirm the effi-cacy and safety of rivaroxaban in HIT that has been demon-strated previously in case-based and in vitro studies.23

Apixaban

Apixaban, another FDA-approved DOAC, has also been assessed in HIT. One in vitro study evaluated apixaban’s effect on the serotonin release assay and on platelet aggre-gation in HIT. As determined by the serotonin release assay, apixaban did not initiate HIT antibody–mediated platelet aggregation. The platelet aggregation assay mirrored these results, confirming the absence of HIT antibody–mediated aggregation. Similar to the other DOACs, apixaban’s lack of in vitro interactions with HIT antibodies and ease of use make it a potential anticoagulant for the management of HIT.24

Dabigatran, Rivaroxaban, and Apixaban

The use of apixaban, dabigatran, and rivaroxaban in HIT has been evaluated in 1 small case series of 12 patients (Table 1). On suspicion for HIT, heparin products were dis-continued, and a continuous infusion of argatroban was ini-tiated. Patients were then transitioned to DOAC therapy with dabigatran 150 mg twice daily (n = 2), rivaroxaban 20 mg daily (n = 7), or apixaban 5 mg twice daily (n = 3) for a minimum of 3 months. Of the 12 patients included in the assessment, 2 experienced HIT with deep-venous thrombo-sis prior to the initiation of NOAC therapy. Following an average follow-up of 16.2 months, no hemorrhagic or thrombotic events were reported in any patient. Results of this study further substantiate the use of chronic DOAC therapy in the setting of HIT following acute treatment with argatroban.25

Discussion

The management of HIT is associated with many chal-lenges. Patient-specific dosing and monitoring with arg-atroban and misinterpretation when transitioning to warfarin has resulted in significant adverse events.15 Currently, the DOACs are FDA approved for venous thromboembolism, yet have limited evidence in both in vitro and clinical HIT studies. Though most of the data are derived from case and cohort studies based on the lack of in vitro interaction with PF4, it is unlikely that a large randomized controlled trial

at UTHSC Library on November 12, 2015aop.sagepub.comDownloaded from

Miyares and Davis 739

will be performed because of the low prevalence of HIT. Currently, rivaroxaban is the most studied of these agents, and an ongoing study of rivaroxaban will further validate the role of the DOACs in this arena.

Conclusion

Though dabigatran, rivaroxaban, and apixaban have been used in case reports, currently data are not yet sufficient to recommend clinical use of these agents in the management of HIT. Future trial results may further substantiate man-agement of HIT with the use of the DOACs.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

1. Amiral J, Bridey F, Dreyfus M, et al. Platelet factor 4 com-plexed to heparin is the target for antibodies generated in heparin-induced thrombocytopenia. Thromb Haemost. 1992;681:95-96.

2. Maccarana M, Lindahl U. Mode of interaction between plate-let factor 4 and heparin. Glycobiology. 1993;3:271-277.

3. Greinacher A, Pötzsch B, Amiral J, Dummel V, Eichner A, Mueller-Eckhardt C. Heparin-associated thrombocytopenia: isolation of the antibody and characterization of a multimo-lecular PF4- heparin complex as the major antigen. Thromb Haemost. 1994;71:247-251.

4. Kelton JG, Smith JW, Warkentin TE, Hayward CP, Denomme GA, Horsewood P. Immunoglobulin G from patients with heparin-induced thrombocytopenia binds to a complex of heparin and platelet factor 4. Blood. 1994;8311:3232-3239.

5. Chong BH, Fawaz I, Chesterman CN, Berndt MC. Heparin-induced thrombocytopenia: mechanism of interaction of the heparin-dependent antibody with platelets. Br J Haematol. 1989;732:235-240.

6. Warkentin TE, Hayward CP, Boshkov LK, et al. Sera from patients with heparin-induced thrombocytopenia generate platelet-derived microparticles with procoagulant activity: an explanation for the thrombotic complications of heparin-induced thrombocytopenia. Blood. 1994;8411:3691-3699.

7. Warkentin TE, Sheppard JI. Generation of platelet-derived microparticles and procoagulant activity by heparin-induced thrombocytopenia IgG/serum and other IgG platelet ago-nists: a comparison with standard platelet agonists. Platelets. 1999;105:319-326.

8. Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular weight heparin or unfractionated heparin. N Engl J Med. 1995;332:1330-1335.

9. Warkentin TE, Roberts RS, Hirsh J, Kelton JG. An improved definition of immune heparin-induced thrombocytopenia

in postoperative orthopedic patients. Arch Intern Med. 2003;163:2518-2524.

10. Greinacher A, Farner B, Kroll H, Kohlmann T, Warkentin TE, Eichler P. Clinical features of heparin-induced thrombocyto-penia including risk factors for thrombosis: a retrospective analysis of 408 patients. Thromb Haemost. 2005;94:132-135.

11. Warkentin TE, Kelton JG. A 14-year study of heparin induced thrombocytopenia. Am J Med. 1996;101:502-507.

12. Socher I, Kroll H, Jorks S, Santoso S, Sachs UJ. Heparin-independent activation of platelets by heparin-induced throm-bocytopenia antibodies: a common occurrence. J Thromb Haemost. 2008;6:197-200.

13. Lubenow N, Eichler P, Lietz T, Farner B, Greinacher A. Lepirudin for prophylaxis of thrombosis in patients with acute isolated heparin-induced thrombocytopenia: an analysis of 3 prospective studies. Blood. 2004;104:3072-3077.

14. Linkins LA, Dans AL, Moores LK, et al; American College of Chest Physicians. Treatment and prevention of heparin-induced thrombocytopenia: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2, suppl):e495S-530S.

15. Hursting MJ, Lewis BE, Macfarlane DE, Hursting MJ, Lewis BE, Macfarlane DE. Transitioning from argatroban to warfa-rin therapy in patients with heparin-induced thrombocytope-nia. Clin Appl Thromb Hemost. 2005;11:279-287.

16. Mavrakanas T, Bounameaux H. The potential role of new oral anticoagulants in the prevention and treatment of thromboem-bolism. Pharmacol Ther. 2011;130:46-58.

17. Krauel K, Hackbarth C, Fürll B, Greinacher A. Heparin-induced thrombocytopenia: in vitro studies on the interac-tion of dabigatran, rivaroxaban, and low-sulfated heparin, with platelet factor 4 and anti-PF4/heparin antibodies. Blood. 2012;119:1248-1255.

18. Fieland D, Taylor M. Dabigatran use in a postoperative coronary artery bypass surgery patient with nonvalvular atrial fibrillation and heparin-PF4 antibodies. Ann Pharmacother. 2012;46:e3.

19. Mirdamadi A. Dabigatran, a direct thrombin inhibitor, can be a life-saving treatment in heparin-induced thrombocytopenia. ARYA Atheroscler. 2013;9:112-114.

20. Walenga JM, Prechel M, Jeske WP, et al. Rivaroxaban—an oral, direct Factor Xa inhibitor—has potential for the manage-ment of patients with heparin-induced thrombocytopenia. Br J Haematol. 2008;143:92-99.

21. Hantson P, Lambert C, Hermans C. Rivaroxaban for arterial thrombosis related to heparin-induced thrombocytopenia. Blood Coagul Fibrinolysis. 2015;26:205-206.

22. Ng HJ, Than H, Teo EC. First experiences with the use of rivaroxaban in the treatment of heparin-induced thrombocy-topenia. Thromb Res. 2015;135:205-207.

23. Linkins LA, Warkentin TE, Pai M, et al. Design of the riva-roxaban for heparin-induced thrombocytopenia study. J Thromb Thrombolysis. 2014;38:485-492.

24. Walenga JM, Prechel M, Hoppensteadt D, et al. Apixaban as an alternate oral anticoagulant for the management of patients with heparin-induced thrombocytopenia. Clin Appl Thromb Hemost. 2013;19:482-487.

25. Sharifi M, Freeman W, Bay C, Sharifi M, Breed C, Schwartz F. Treatment of heparin-induced thrombocytopenia with new oral anticoagulants. J Am Coll Cardiol. 2014;63:A2096.

at UTHSC Library on November 12, 2015aop.sagepub.comDownloaded from