Cold Leg in Patient with High Coated Platelets: Possible Association with the Use of Rofecoxib

Diagnostic coronary angiography in the current era is a remarkably safe procedure, yet major complications do occur. Many of these complications are related to vascular access. Although discomfort at the site of access is frequent and minor hematomas are common, one of the most potentially devastating complications is acute limb ischemia. This can relate to pre-existing atherosclerosis, prothrombotic state, catheter-induced dissection, local thrombus formation, or a combination of these factors. Prolonged use of external compression devices may also increase this risk. A new concern is whether closure devices, particularly those that involve the insertion of collagen or thrombin, increase the frequency of these complications. We report the occurrence of acute limb ischemia in a patient in whom hemostasis was achieved with a 6 French (Fr) suture-mediated closure device and light manual compression. This patient presented with acute leg ischemia 20 hours after the diagnostic coronary angiography and was successfully treated by rheolytic thrombectomy and antiplatelet therapy with eptifibatide, with resultant successful limb salvage and no long-term sequelae. Case Report. A 52-year-old obese female was admitted to the hospital with chest pain and cardiac risk factors of tobacco use and dyslipidemia. A subsequent dobutamine echocardiogram demonstrated abnormal anterior wall motion at peak stress, suggestive of ischemia in the left anterior descending coronary artery distribution. The patient was enrolled in a clinical observation study to evaluate whether a newly-described platelet subpopulation, COAT (collagen- and thrombin-activated) platelets, also known as coated platelets, is associated with atherosclerosis and acute coronary syndromes. A coronary angiogram was subsequently performed via the right common femoral artery without incident. Minimal coronary artery disease was evident, and at the conclusion of the procedure, a right common femoral arteriogram was performed. This demonstrated that the access site was in the mid common femoral artery; the common femoral was a 5 mm diameter vessel and was noted to have an ulcerative 40% stenosis in the mid right external iliac artery (Figure 1). A 6 Fr Perclose™ suture-mediated closure device (Abbott Laboratories, Redwood City, California) was deployed by an experienced operator. Subcutaneous fat impeded the knot delivery, hence some ooze persisted after deployment, and 5 minutes of light manual compression was applied with complete hemostasis. The patient was discharged 6 hours later after confirmation of all distal pulses and normal right groin examination. The patient was able to walk and later in the evening, in fact, did considerable housework. Approximately 12 hours after the procedure, she awoke with severe pain in her right leg which worsened, and subsequently, she came to the emergency room. Examination revealed a cold right leg, absent femoral, popliteal and dorsalis pedis pulses, both on physical examination and confirmed by doppler. An arterial duplex scan confirmed the complete occlusion of the right common femoral artery. Heparin therapy was initiated and the patient was transferred to the cardiac catheterization laboratory. Angiography was then performed via an access site in the left common femoral artery. It demonstrated a thrombotic occlusion of the right common femoral artery from its origin, extending 4–5 cm caudal to the distal edge of the femoral head (Figure 2). The 40% ulcerative lesion noted earlier in the right external iliac artery was unchanged and was distant from the new occlusion. Then, a 90-cm 7 Fr Arrow sheath was advanced over the aortic bifurcation into the right external iliac artery using a 6 Fr internal mammary catheter and Wholey wire. Next, a Wholey™ guidewire (Mallinckrodt, Inc., Hazelwood, Missouri) was used to cross the common femoral occlusion using a 5 x 40 mm Agiltrac™ balloon catheter (Guidant Corp., Indianapolis, Indiana) for support. The latter was advanced into the right superficial femoral artery, and the Wholey wire was removed to allow injection of contrast via the balloon catheter lumen, which confirmed the catheter position in the main lumen of the vessel and that a dissection plane had not been entered. A 300 cm, 0.014 inch Mailman wire (Boston Scientific Corp., Natick, Massachusetts) was then advanced into the distal superficial femoral artery and the balloon catheter was removed. Rheolytic thrombectomy using a 6 Fr AngioJet® catheter (Possis Medical, Inc., Minneapolis, Minnesota ) was then performed beginning distal to the occlusion, and four series of aspirations were performed. Flow was restored with three-vessel runoff to the right foot. A moderate lesion remained in the mid common femoral artery, and this was ballooned with the 5 x 40 mm Agiltrac and a 6 x 40 mm Agiltrac balloon. The final result is shown in Figure 3. The right dorsalis pedis was palpable and the ankle brachial index (ABI) was 0.83. The patient was admitted to the intensive care unit, and heparin infusion was continued with maintenance of a therapeutic activated prothrombin time (aPTT). The arterial sheath was kept in place. Serial measurements of the right dorsalis pedis pressure and ankle brachial index (ABI) were obtained. The next day, the right dorsalis pedis systolic pressure had fallen to 70 mmHg and the ABI to 0.6. Duplex ultrasound showed a homogenous lesion and associated increased velocities in the right common femoral artery proximal to the arterial puncture site consistent with a flow-limiting stenosis and no evidence of intimal flap. Thrombus recurrence was the most likely explanation, therefore a 180 µg/kg bolus of eptifibatide and a 2 µg/kg/minute infusion were initiated. Within 12 hours, the right ABI had improved to 0.85 and reached 0.9 after 48 hours of eptifibatide infusion. Muscle strength and neural function were preserved. The patient was discharged on long-term aspirin and a 6-month course of clopidogrel. Follow up revealed no further complications, and her right ABI improved to 1.06 with no symptoms of ischemia. As part of a clinical study the patient was enrolled in, various platelet assays were performed on blood samples obtained before the initial diagnostic coronary angiogram. It has been described that platelets costimulated with collagen and thrombin express alpha-granule proteins on their surface, including fibrinogen, von Willebrand factor (vWF), thrombospondin, fibronectin and alpha-2-antiplasmin. These platelets, referred to as coated platelets, promote prothrombinase generation and may predispose to atherosclerosis and thrombotic events. Our patient had 63% coated platelets, one of our highest measured levels; 18 months later, coated platelets were down to 36.9%, which is within in normal range (25–40%). The patient was on rofecoxib when the initial coated platelet measurement was made; rofecoxib was discontinued 18 months prior to follow-up coated platelet measurement. Discussion. It was previously reported that peripheral vascular complications occur at a rate of 0.7% to 9% following invasive vascular procedures. Acute limb ischemia represents a small percentage of these complications. In a large series of patients using contemporary catheters for diagnostic cardiac catheterization and manual compression for hemostasis, Ammann et al. reported that the incidence of acute limb ischemia was 0.11% (8 of the 7,412 patients). They concluded that the rate of these complications was related to the catheter size, experience of the operators, and body weight of the patients.¹ A report of 1,050 consecutive patients undergoing peripheral angiography by Young et al., who exclusively used smaller-sized catheters (2 Lately, many arterial puncture closure devices (APCDs) have been introduced into clinical use in an attempt to decrease the time to hemostasis and to reduce patient discomfort during manual compression. It is estimated that these devices are now being used in 50% of patients undergoing percutaneous coronary intervention.³ These can be coagulant-mediated (VasoSeal, Datascope Corp., Montvale, New Jersey and Duett™, Vascular Solutions, Inc., Minneapolis, Minnesota), collagen sandwich-mediated (Angio-Seal™, St. Jude Medical, Inc., St. Paul, Minnesota), or suture-mediated (Perclose). The exact incidence of acute limb ischemia with APCDs is not known. In a recent meta-analysis of studies using standard manual compression versus various APCDs for hemostasis, it was shown that the relative risk of arterial leg ischemia with APCDs was 2.1 (95% CI, 0.97–4.58) compared with manual compression for hemostasis.³ The risk varies according to the closure device used. Acute limb ischemic events following the use of collagen-mediated devices usually tend to be thrombotic. On the contrary, those following the use of suture-mediated devices tend to be nonthrombotic. This was shown in the study by Boston et al., who compared complications and outcomes between patients requiring surgery after receiving a closure device and those undergoing standard manual compression. The 4 ischemic limbs in which Perclose was used showed significant intraoperative evidence of the device occluding the vessel or causing significant stenosis in an atherosclerotic vessel.⁴ They also reported that the surgical repair tends to be more complex and re-operation more frequent in the group treated with closure devices. Acute limb ischemia secondary to thrombosis alone after the use of a suture-mediated closure device has never been reported. Platelets play a major role in the primary hemostasis following vascular trauma, including arterial puncture. This was conclusively shown by O’Connor et al. in their study using iridum-111-labeled platelets before and after the arterial puncture. Their study showed very high uptake of the radioisotope in the segment of the artery between the puncture site and aortic bifurcation after the conclusion of the procedure. This was even more prominent at the site of the arterial puncture.⁵ Recently, a subpopulation of platelets called coated platelets has been described. When stimulated simultaneously with collagen and thrombin, these platelets express several alpha-granule proteins on their outer surface.^6,7 including fibrinogen, thrombospondin, vWF, fibronectin, and alpha-2-antiplasmin. These platelets may have significant prothrombotic properties due primarily to their active facilitation of a prothrombinase activity. Our patient had a very high percentage of coated platelets (63%) compared to a control group (25–40%).⁸ This high level of coated platelets may have been elevated secondary to the prothrombotic and pro-atherosclerotic effects of rofecoxib. This conclusion is supported by the return to normal of the coated platelets level after the discontinuation of rofecoxib. We hypothesize that these coated platelets may have played a major role in the initiation and maintenance of the thrombotic occlusion of the common femoral artery in our patient. A similar elevation of coated platelets has been seen in a normal platelet donor, in which the coated platelets returned to pretreatment levels once rofecoxib was discontinued (G.L. Dale, unpublished observation). The mechanism(s) for any effect of rofecoxib on coated platelet production is(are) unknown. Acute limb ischemia is a medical emergency. Unless promptly recognized and managed, it can be limb-threatening. Currently, three modalities of treatment are available for patients with acute limb ischemia. These include local thrombolytic infusion, rheolytic thrombectomy, and surgical thrombectomy. With the availability of the state-of-the-art catheter-based techniques, surgical thrombectomy can be considered the treatment of last resort, unless the required expertise is not available or percutaneous procedures are contraindicated. Intra-arterial thrombolytic infusion has been shown to result in similar acute and long-term success rates of limb salvage compared to surgical thrombectomy.⁹ However, thrombolytics are relatively contraindicated in patients with recent arterial puncture due to bleeding concerns. Rheolytic thrombectomy has also been shown to have similar success rates of limb salvage in patients with acute limb ischemia.^10–12 In their study, Silva et al. reported both acute (95%) and 6-month follow up (89%) rates of successful limb salvage, and these were comparable to those obtained with either thrombolytics or surgical thrombectomy. Most of the patients in this study were high-risk surgical candidates or had contraindications to thrombolytic therapy,¹² as did the patient in this case report. It has been reported that glycoprotein IIb/IIIa inhibitors used in conjunction with thrombolytics did not have a significant effect on the rate of successful limb salvage.13 The benefit of these drugs used in conjunction with rheolytic thrombectomy is not known. Our case report suggests that glycoprotein IIb/IIIa inhibitors do offer offer a benefit after successful rheolytic thrombectomy in some patients. The recurrence of thrombus after successful rheolytic thrombectomy combined with continued use of intravenous heparin and successful resolution after treatment with intravenous eptifibatide indicates that platelets were the major factors leading to acute limb ischemia in this patient. This observation supports our hypothesis that the high levels of coated platelets in this patient may have played a major role in the initiation and maintenance of the thrombus, leading to acute limb ischemia.

1. Ammann P, Brunner-La Rocca HP, Angehrn W, et al. Procedural complications following diagnostic coronary angiography are related to the operator’s experience and the catheter size. Catheter Cardiovasc Interv 2003;59:13–18. 2. Young N, Chi KK, Ajaka J, et al. Complications with outpatient angiography and interventional procedures. Cardiovasc Intervent Radio 2002;25:123–126. 3. Koreny M, Riedmuller E, Nikfardjam M, et al. Arterial puncture closing devices compared with standard manual compression after cardiac catheterization. JAMA 2004;291:350–357. 4. Boston US, Panneton JM, Hofer JM, et al. Infectious and ischemic complications from percutaneous closure devices after vascular access. Ann Vasc Surg 2003;17:66–71. 5. O'Connor MK, Brennan SS, Shanik DG. Indium-111 labeled platelet deposition following transfemoral angiography. Radiology 1986;158:191–194. 6. Dale GL, Friese P, Batar P, et al. Stimulated platelets use serotonin to enhance their retention of procoagulant proteins on the cell surface. Nature 2002;415:175–179. 7. Szasz R, Dale GL. Thrombospondin and fibrinogen bind serotonin-derivatized proteins on COAT-platelets. Blood 2002;100:2827–2831. 8. Szasz R, Dale GL. COAT platelets. Curr Opinion Hematol 2003;10:351–355. 9. Ouriel K, Veith FJ, Sasahara AA. A comparison of recombinant urokinase with vascular surgery as initial treatment for acute arterial occlusion of the legs. Thrombolysis or Peripheral Arterial Surgery (TOPAS) Investigators. N Engl J Med 1998;338:1105–1111. 10. Ansel GM, George BS, Botti CF, et al. Rheolytic thrombectomy in the management if limb ischemia: 30 day results from a multicenter registry. J Endovasc Ther 2002;9:395–402. 11. Kasirajan K, Gray B, Beavers FP, et al. Rheolytic thrombectomy in the management of acute and sub acute limb-threatening ischemia. J Vasc Interv Radiol 2001;12:413–421. 12. Silva JA, Ramee SR, Collins TJ, et al. Rheolytic thrombectomy in the treatment of acute limb-threatening ischemia: Immediate results and six-month follow up of the multicenter angiojet registry. Cathet Cardiovasc Diagn 1998;45:386–393. 13. Yoon HC, Miller FJ. Using a peptide inhibitor of the glycoprotein 2b/3a platelet receptor: Initial experience in patients with acute peripheral arterial occlusions. Am J Roentgenol 2002;178: 617–622.