predict the individual who will finish first. Thats a lot harderto predict. I have had people ask when they finish early if theywere the first; I sometimes say yes to more than one of themas they leave because Im empathetic and it seems so importantsomehow to their egos.

I also have yet, in this session, to have the inevitable com-plaint that these test questions do, in fact, stink, and the com-plainer would be happy to supply us with the correct answersto the inappropriate or just plain poor-quality questions. Idont know for sure, but Id bet those guys are usually going tobe the ones to not pass the test, somehow.

Correction: I just now received that specific complaint.Now I can relax. The remedy I offer the complaining individ-ual is to have them document their concerns, in pencil (nopens allowed, please, and no scissors or belts in the test envi-ronment) on the feedback sheet, which is guaranteed to beeyeballed by the authorities after the test is in the can. I hopethis is actually true. I have seen several pages of specificshanded in by one person in a past session, which was some-what of a record. Most diplomates are content to get out ofhere as fast as possible and are just glad to get it overwith.Mostof these folks will pass this test anyway and forget about it foranother 10 years.

Although I can only personally vouch for the neuroradiol-ogy COQ/MOC, I thought the test itself was pretty well done,fair and balanced, and actually a learning experience. This isbecause when I went home and looked up some of the things Isaw on the test, I found I was sometimes wrong (yes, it hap-pens) and I actually learned the right answer. Belatedly, but Istill passed.

By now you have undoubtedly heard about the infamoustrue or false questions, right? When I took the test as one ofthe first responders 2 years ago, neither the ABR as tester nor Ias tested knew these questions constituted a minefield thatwould continue to be a danger for future diplomates evenwhen warned specifically to look out for them. In brief, T orF questions typically have 4 or 5 choices as answers, each witha T and an F box in front. It is honest-to-goodness com-plete human nature to only bother the check off the Ts andleave the Fs blank. You cannot help yourself. The only prob-lem is that an F left unchecked will be counted as incorrect.I didnt figure that out until halfway through 42 questionswhen I first took the test. To compound the problem, many ofthese are blocked so that once they are left, the test-taker cango back and look at them, and scream, but cannot changethem in any way. Bummer, but I still passed the examination(must have done really well on the spine questions). Word tothe wise, but Ill bet youre still going to do this to at least oneof these questions when you take the test.

All right, now. Its almost over for this session and I onlyhave 2 more sessions to go tomorrow. The CAQ is obviouslyan important certification to have, and to maintain; it may beeven more important in the future, if hospital privileges orreimbursements ever require them. As much of a pain as allthis rigmarole is, you should support it and go get tested. It hasactually been very heartening to see some of the giants of neu-roradiology, in their seventh and eighth decades, being tor-tured at the test centers along with the younger generation.Since I am one of those old guys, I guess Im glad I set anexample, griping and moaning all the way. Im also real glad I

actually did pass the test, given how much I screwed up thetrue or false questions.

F. Reed Murtagh, MD

COMMENTARY

Recommendations for AnticoagulatedPatients Undergoing Image-GuidedSpinal Procedures

Anticoagulated patients often need image-guided spinalprocedures for CSF harvest, myelography, vertebroplasty,vertebral biopsies, or epidural injections. The risk of spinalhematoma is increased in anticoagulated patients who un-dergo lumbar puncture or neuraxial anesthesia. Any proce-dure involving needle manipulation or biopsy with potentialtransgression of the subarachnoid, subdural, or epidural vas-culature, however, likely carries a similar risk. This risk is in-creased, often substantially, by the use of multiple anticoagu-lants and the intensity of anticoagulation. It is crucial thatradiologists who perform spinal procedures be familiar withthe common anticoagulant and antiplatelet medications.

Radiologists are increasingly being asked to provide fluo-roscopically assisted access to the neuraxial system.Whether aroutine lumbar puncture, epidural steroid injection, spinal bi-opsy, or themore unusual C12 cervical puncture, there is thepotential for bleeding complications. Most of the case reportsinvolving spinal hematomas following lumbar puncture, highcervical myelogram, and epidural injection (as well as thoserelated to neuraxial anesthesia) are reported in the anesthesiaand surgical literature.1-4 Large series consistently note thatthe risk of spinal hematoma is potentiated by the concomitantadministration of anticoagulant and/or antiplatelet therapyand difficult and/or traumatic spinal instrumentation.5,6 Neu-rologic compromise typically presents as a sensory or motordeficit or bowel/bladder dysfunction, not severe radicularback pain. Because of delays in the diagnosis, neurologic re-covery is poor in most cases. Thus, radiologists must be awareof the risk factors and diagnosis of spinal bleeding.

Much of the information related to postprocedure spinalhematomas in anticoagulated patients is derived from cases ofspinal hematoma associated with neuraxial anesthesia and an-esthesia. Formal recommendations have been put forth by theAmerican Society of Regional Anesthesia and Pain Medicine,but correlative recommendations by the radiology commu-nity are currently not available.7 In hopes of facilitating themanagement of patients presenting to radiologists for spinalprocedures in the setting of anticoagulant or antiplatelet ther-apy, we offer a focused, readily accessible set of guidelines forperforming spinal procedures on anticoagulated patients.

DiscussionLiterature is available regarding recommendations formanag-ing patients with medication-induced coagulopathies and isreviewed below (Table). Patients typically receive these medi-cations for chronic antithrombotic therapy in the prevention

468 Commentary AJNR 27 Mar 2006 www.ajnr.org

of stroke or myocardial ischemia, thromboprophylaxis fol-lowing surgery, or treatment of acute thromboembolism orcoronary syndrome. The intensity and duration of anticoagu-lation affect the risk of spontaneous, as well as procedural-related spinal bleeding.8 Although less common than needleplacement for injection or biopsy, radiologists may also berequested to assist with placement of an indwelling neuraxialcatheter, such as a spinal drainage catheter. In these cases,significant anticoagulant medications should not be adminis-tered until the catheter or drain is removed.

Anticoagulant TherapyWarfarin. Chronic warfarin therapy increases the risk of

spinal hematoma following lumbar puncture. The addition ofagents that affect different parts of the clotting mechanismlikely increase the risk for spinal hematoma and do so withoutfurther elevation of the prothrombin time (PT) or interna-tional normalized ratio (INR).9 These medications includeheparin, nonsteroidal anti-inflammatory drugs (NSAIDs),and antiplatelet agents. Warfarin should be discontinued inanticipation of the spinal procedure and normalization of theINR documented preprocedure. If a spinal procedure is per-formed on a patient with an INR1.2, close neurologic test-ing of motor and sensory function should be performed for atleast 24 hours to ensure prompt recognition and treatment ofspinal hematoma. In emergent cases, the injection of vitaminK or transfusion of fresh frozen plasma may counteract theeffects of warfarin.10

Heparin. There is no contraindication to spinal punc-ture in patients receiving subcutaneous heparin as a pro-phylaxis for deep venous thrombosis providing the totaldose is10,000 U.11 Higher dosing may result in sustainedprolongation of the activated partial thromboplastin time(aPTT). These patients are managed similar to those whoare systemically heparinized. Delaying the scheduled hepa-rin injection until after the puncture may reduce the risk ofspinal hematoma. The risk of bleeding is likely increased indebilitated patients on prolonged therapy. Patients receiv-ing heparin for longer than 4 days need to have a plateletcount assessment because of the potential for heparin-in-duced thrombocytopenia.12

Systemic heparinization represents an increased risk for

spinal bleeding.8 Heparin infusion should be discontinuedand aPTT normalized before the procedure. A subsequentdose of intravenous heparin should not be administered for atleast an hour after the procedure.11,13 The combined use ofother anticoagulants with unfractionated heparin may in-crease the risk of spinal hematoma. These include antiplate-lets, low-molecular-weight heparin (LMWH), and oral anti-coagulants.

LMWH. LMWH is the recommended thromboprophylac-tic agent followingmajor orthopedic and general surgical pro-cedures.14 It is important that there be a number of dosingregimens for LMWH, including low-dose (thromboprophy-lactic) and high-dose (therapeutic) applications. There aremany pharmacologic differences between standard unfrac-tionated heparin and LMWH, including prolonged half-lifeand irreversibility with protamine.15,16 Early postoperativedosing, twice-daily dosing, and traumatic needle placementwere identified as risk factors for spinal hematoma associatedwith neuraxial anesthesia. Because significant anticoagulantactivity persists for 12 hours after low-dose injection (and 24hours for a high-dose injection), these time intervals should beobserved before a spinal procedure. Likewise, the first post-procedural LMWHdose should be administered 1824 hourslater, to allow for adequate hemostasis.

Thrombolytic TherapyData are not available to clearly define how long spinal punc-ture should be avoided following termination of thrombolyt-ic/fibrinolytic therapy; however, significant defects in hemo-stasis are present for longer than 24 hours. Patients who haverecently had or that are likely to receive thrombolytic/fibrino-lytic therapy should bewarned against receiving a spinal punc-ture except in very unusual circumstances. Likewise, patientsshould be questioned before starting thrombolytic/fibrino-lytic therapy whether there has been a recent spinal proceduresuch as lumbar puncture. This will allow for appropriatemon-itoring in cases where the drugmust be administered. Originalguidelines recommended avoidance of thrombolytic drugs for10 days following puncture of noncompressible vessels.17 Incertain cases, measurement of fibrinogen level (one of the lastclotting factors to recover) may be helpful in monitoring apatient who underwent or will undergo a spinal procedure.7

Recommended guidelines for performing spinal procedures in anticoagulated patients

Warfarin Discontinue chronic warfarin therapy 45 days before spinal procedure and evaluate INR. INR should be within thenormal range at time of procedure to ensure adequate levels of all vitamin K-dependent factors.

Antiplatelet medications No contraindications with aspirin or NSAIDs. Thienopyridine derivatives (clopidogrel and ticlopidine) should bediscontinued 7 days and 14 days, respectively, prior to procedure. GP IIb/IIIa inhibitors should be discontinued toallow recovery of platelet function prior to procedure (8 hours for tirofiban and eptifibatide, 2448 hours forabciximab).

Thrombolytics/fibrinolytics There are no available data to suggest a safe interval between procedure and initiation or discontinuation of thesemedications. Follow fibrinogen level and observe for signs of neural compression.

LMWH Delay procedure at least 12 hours from the last dose of thromboprophylaxis LMWH dose. For treatment dosingof LMWH, at least 24 hours should elapse prior to procedure. LMWH should not be administered within 24hours after the procedure.

Unfractionated SQ heparin There are no contraindications to neuraxial procedure if total daily dose is less than 10,000 units. For higherdosing regimens, manage according to intravenous heparin guidelines.

Unfractionated IV heparin Delay spinal puncture 24 hours after last dose, document normal aPTT. Heparin may be restarted 1 hour followingprocedure.

Note:NSAIDs indicates nonsteroidal antiinflammatory drugs; GP IIb/IIIa, platelet glycoprotein receptor IIb/IIIa inhibitors; INR, international normalized ratio; LMWH, low-molecular-weightheparin; aPTT, activated partial thromboplastin time. Adapted from: Horlocker TT, Wedel DJ, Benzon H, et al. Regional anesthesia in the anticoagulated patient: defining the risks(the second ASRA Consensus Conference on Neuraxial Anesthesia and Anticoagulation). Reg Anesth Pain Med 2003;28:17297.

EDITO

RIALS

AJNR Am J Neuroradiol 27:46771 Mar 2006 www.ajnr.org 469

Antiplatelet TherapyThe antiplatelet medications include a diverse group of agentsin terms of their effects on platelet function; therefore, it is notpossible to extrapolate between the various groups of drugsregarding spinal procedures. These agents include NSAIDs,thienopyridine derivatives, and GP IIb/IIIa antagonists.

NSAIDsThe use of NSAIDs alone does not seem to increase the risk ofspinal hematoma from spinal puncture. At this time, there donot seem to be specific concerns related to timing of spinalpuncture in relation to the dosing of NSAIDs or postproce-dure monitoring.18,19

Thienopyridine DerivativesThis class of antiplatelet agents works by inhibiting adenosinediphosphateinduced platelet aggregation. These drugs affectboth primary and secondary platelet aggregation as well asplatelet-fibrinogen binding.20 The agents in this class includeclopidogrel (Plavix) and ticlopidine (Ticlid). The patientshould be carefully assessed for other factors thatmight lead tobleeding such as easy bruising/bleeding, female sex, and in-creased age.7 The addition of other medications affecting dif-ferent clotting mechanisms will likely increase the chance forspinal hematoma.

GP IIb/IIIaReceptor AntagonistsThese agents affect platelet-fibrinogen and plateletvon Wil-lebrand factor binding to inhibit platelet aggregation. Thesemedications are often given concomitantly with aspirin andheparin. This class of antiplatelet drugs includes abciximab(ReoPro), eptifibatide (Integrilin), and tirofiban (Aggrastat).Normal platelet aggregation is usually achieved 8 hours afterdiscontinuation of tirofiban and eptifibatide and 2448 hoursafter discontinuing abciximab.

The true risk of spinal hematoma in patients on thienopy-ridine derivatives or GP IIb/IIIa antagonists is unknown.Management is based on labeling precautions and prior expe-rience. The concomitant use of aspirin with these agents mayincrease the risk for spinal hematoma. The GP IIa/IIIb antag-onists have a profound effect on platelet aggregation and spi-nal puncture should be avoided until platelet function hasrecovered.21 Of note, these agents are contraindicatedwithin 4weeks of surgery. There is not a definitive test, including bleed-ing time, that can guide antiplatelet therapy.

ConclusionThe increased vigilance over venous thromboembolism andintroduction of more efficacious antiplatelet agents has intro-duced a degree of complexity into the performance of spinalprocedures. The presence and continued evolution of anti-platelet agents, various heparin derivatives and thrombolytictherapy requires a thorough investigation of a patients med-

ication history. Continued surveillance of the literature will benecessary to stay abreast of the newer agents that are sure toappear, aswell as any changes in the recommendations regard-ing agents currently in use. The guidelines referenced in thetable and can be accessed on-line at www.asra.com.

References1. Reitman CA, Watters W. Subdural hematoma after cervical epidural steroid

injection. Spine 2002;27:E174762. AghiM,Valery-Coumans J, Brisman JL. Subarachnoidhematoma, hydroceph-

alus, andasepticmeningitis resulting fromahighcervicalmyelogram. J SpinalDisord Tech 2004;17:34851

3. Stoll A, SanchezM. Epidural hematoma after epidural block: implications forits use in pain management. Surg Neuro 2002;57:23540

4. Diaz FG, Yock DH Jr, Rockswold GL. Spinal subarachnoid hematoma afterlumbar puncture producing acute thoracic myelopathy: case report. Neuro-surgery 1978;3:40406

5. Vandermeulen EP, Van Aken H, Vermylen J. Anticoagulants and spinal-epi-dural anesthesia. Anesth Analg 1994;79:116577

6. Horlocker TT, Wedel DJ.Neuraxial block and lowmolecular weight heparin:balancing perioperative analgesia and thromboprophylaxis. Reg Anesth PainMed 1998;23:16477

7. Horlocker TT,Wedel DJ, BenzonH, et al. Regional anesthesia in the anticoag-ulated patient: Defining the risks: the second ASRA consensus conference onneuraxial anesthesia and anticoagulation. Reg Anesth Pain Med 2003;28:17297

8. Stafford-Smith M. Impaired haemostasis and regional anaesthesia. Can J An-aesth 1996;43:R12941

9. Harder S, Thurmann P. Clinically important drug interactions withanticoagulants: an update. Clin Pharmacokinet 1996;30:41644

10. Shields RC, McBane RD, Kuiper JD, et al. Efficacy and safety of intravenousphytonadione (vitamin K1) in patients on long-term oral anticoagulant ther-apy.Mayo Clin Proc 2001;76:26066

11. Liu SS, Mulroy MF. Neuraxial anesthesia and analgesia in the presence ofstandard heparin. Reg Anesth Pain Med 1998;23:15763

12. Hirsh J, Raschke R, Warkentin TE, et al.Heparin: mechanism of action, phar-macokinetics, dosing considerations, monitoring, efficacy, and safety. Chest1995;108:258S75S

13. Ruff RL, Dougherty JH Jr. Complications of lumbar puncture followed byanticoagulation. Stroke 1981;12:87981

14. GeertsWH, Pineo GF, Heit JA, et al. Prevention of venous thromboembolism:the seventh ACCP conference on antithrombotic and thrombolytic therapy.Chest 2004;126(3 suppl):338S400S

15. Cosmi B, Hirsh J. Low molecular weight heparins. Curr Opin Cardiol 1994;9:61218

16. Holst J, LindbladB, BergqvistD, et al.Protamine neutralization of intravenousand subcutaneous low-molecular-weight heparin (tinzaparin, Logiparin): anexperimental investigation in healthy volunteers. Blood Coagul Fibrinolysis1994;5:795803

17. Majerus PW, Broze GJ, Miletich JP, et al. Goodman and Gilmans the phamaco-logical basis of therapeutics. 8th ed. New York: McGraw-Hill;1993:132231

18. CLASP (Collaborative Low-Dose Aspirin Study in Pregnancy) CollaborativeGroup.CLASP: a randomized trial of low-dose aspirin for the prevention andtreatment of preeclampsia among 9364 pregnant women. Lancet 1994;343:61929

19. Horlocker TT, Bajwa ZH, Ashraft Z, et al. Risk assessment of hemorrhagiccomplicationsassociatedwithnonsteroidal antiinflammatorymedications inambulatory pain clinic patients undergoing epidural steroid injection.AnesthAnalg 2002;95:169197

20. Taniuchi M, Kurz HI, Lasala JM. Randomized comparison of ticlopidine andclopidogrel after intracoronary stent implantation in a broad patient popula-tion. Circulation 2001;104:53943

21. Shlansky-GoldbergR.Platelet aggregation inhibitors foruse inperipheral vas-cular interventions: what can we learn from the experience in the coronaryarteries? J Vasc Interv Radiol 2002;13:22946

Kennith F. Layton, MD,David F. Kallmes, MD, andTerese T. Horlocker, MD

470 Commentary AJNR 27 Mar 2006 www.ajnr.org