Tenecteplase in Acute Lower-Leg Ischemia: Efficacy, Dose, and Adverse Events

Tenecteplase in Acute Lower-leg Ischemia: Efficacy, Dose, and Adverse Events

Jeffrey Eaton Hull, MD, Mary Kathryn Hull, BSN, James A. Urso, MD, and H. Andy Park, MD

PURPOSE: To prospectively evaluate tenecteplase (TNK) for thrombolysis in acute lower-limb ischemia. MATERIALS AND METHODS: Forty-three consecutive limbs in 37 patients (15 male, 22 female) were treated for acute lower-limb ischemia. Group 1 included 22 limbs treated with TNK infusion of 0.25 mg/h and group 2 included 21 limbs treated with TNK at 0.125 mg/h. Technical success was defined by 95% clearing of thrombus, and clinical success was defined by Society of Interventional Radiology category for acute ischemia of ؉1. Complications were ranked by severity and relation to TNK administration. Logistic regression, Student t test, and analysis of variance were performed. RESULTS: TNK infusions averaged 24 hours in duration (SD, 13 h), with means of 20 hours in group 1 and 27 hours Technical success was achieved in 84% of limbs (36 of 43): 82% in group 1 (18 of 22) and 86% .(071. ؍ in group 2 (P The SIR ischemia category improved (ie, ؉1) in 86% of limbs (37 of 43), stayed the same .(827. ؍ in group 2 (18 of 21; P ie, category 0) in 12% of limbs (five of 43), and worsened (ie, ؊1) in 2% of limbs (one of 43). TNK-related) ,and were correlated with percentage decrease in fibrinogen level (5 ؍ complications were seen in 12% of limbs (n respectively). Initial TNK boluses ,036. ؍ P < .001, and P ,001. ؍ initial TNK bolus, and abciximab administration (P The percentage decrease .(045. ؍ of 1.5 mg or less were associated with fewer complications than boluses of 3–5 mg (P There was a 7% .(045. ؍ in fibrinogen level in group 1 was 23% (SD, 29%), compared with 7% in group 2 (SD, 20%; P .and no intracranial hemorrhages (3 ؍ incidence of major bleeding complications (n CONCLUSIONS: Treatment of acute lower-limb ischemia with TNK infusion at 0.25 mg/h and 0.125 mg/h is associated with similar success and complication rates. TNK-related complications correlated with initial TNK bolus, abciximab treatment, and percent decrease in fibrinogen level. The initial TNK bolus dose should be limited to 1.5 mg.

J Vasc Interv Radiol 2006; 17:629–636

Abbreviations: ANOVA ϭ analysis of variance, TNK ϭ tenecteplase, tPA ϭ tissue plasminogen activator

TENECTEPLASE (TNK; Genentech, and K domains. The changes have cre- strated lower bleeding complications South San Francisco, CA) is a recom- ated a drug with a longer half-life of 22 of TNK in acute myocardial infarction binant DNA thrombolytic agent ap- minutes (T domain), a 14-fold increase might also provide lower incidences of proved for treatment of acute myocar- in fibrin specificity and a twofold im- bleeding complications during treat- dial infarction. TNK is made by provement in effectiveness in arterial ment for acute lower-limb ischemia. genetic mutation of human tissue plas- thrombolysis (N domain), an 80-fold Early reports of TNK use for acute minogen activator (tPA) at the T, N, increase in resistance to plasminogen limb ischemia have been published activator inhibitor (K domain), and de- and demonstrated clinical success creased fibrinogen depletion (1–3). rates of 73%–88% in patients with pe- The clinical superiority from these ripheral arterial occlusive disease From the Vascular Center, Chippenham & Johnston- Willis Medical Center, 7101 Jahnke Road, Rich- changes has not been demonstrated in treated with an infusion dose of 0.25 mond, Virginia 23225. Received October 19, 2005; thrombolytic infusions for acute low- mg/h or 0.5 mg/h (5–8). The ideal revision requested; revision received and accepted er-limb ischemia. dose for TNK infusion has not been December 25. Address correspondence to J.E.H.; E-mail: [email protected] TNK has been found to have an established. We initially used a dose of improved safety profile compared 0.25 mg/h as described in published None of the authors have identified a conflict of interest. with tPA in the treatment of acute reports (6,8) and then empirically myocardial infarction (4), specifically a chose to reduce the infusion dose by © SIR, 2006 22% relative reduction in major bleed- half to 0.125 mg/h. In this report, we DOI: 10.1097/01.RVI.0000202751.74625.79 ing and transfusion. The demon- compare our experiences with the ad-

629 630 • Tenecteplase for Thrombolysis in Lower-limb Ischemia April 2006 JVIR

hour (5 mg in 500 mL normal saline Table 1 Patient Population solution at 25 mL/h). Patients in group 2 received TNK infused at a rate Characteristic Group 1 (n ϭ 22) Group 2 (n ϭ 21) P Value of 0.125 mg/h (5 mg in 1,000 mL nor- Age (y) 55 60 .647 mal saline solution at 25 mL/h). Other Male sex 7 (32) 8 (38) .748 than the infusion dose, there were no Grafts 16 (73) 11 (52) .177 differences in the treatment protocol Graft length (cm) 22 20 .652 between groups 1 and 2. Optional bo- Graft location* — luses of 1–5 mg TNK (0.25 mg/mL) Iliac 9 (41) 9 (43) were given before infusion. Heparin- Femoropopliteal 14 (64) 13 (62) ized saline solution or standard hepa- Distal 4 (18) 6 (29) rin (25,000 U in 250 mL 5% dextrose in * More than one possible location per patient. water) at a subtherapeutic rate (600– Note.—Numbers in parentheses are percentages. 800 U/h) was infused into sheaths during thrombolysis. Heparin boluses of 3,000–5,000 IU were usually given during catheter manipulation, angio- ministration of TNK at bolus doses of years of age, and have angiographi- plasty, and stent placement. Abcix- 0.25 mg/h and 0.125 mg/h. cally confirmed vascular occlusion of a imab, when used, was given as an native artery or graft in the lower or intravenous bolus of 0.25 mg/kg fol- MATERIALS AND METHODS upper extremity causing claudication lowed by a 12-hour infusion of 0.125 or ischemia. ␮g/kg/min. Abciximab was used at The hospital investigational review Patients were excluded in cases of operator discretion in patients who board approved this nonrandomized internal bleeding involving intracra- showed signs of acute thrombosis dur- prospective study. Data were obtained nial and retroperitoneal sites or gas- ing angiography. Patients were fol- from 43 consecutive limbs in 37 pa- trointestinal, genitourinary, or respi- lowed in the intensive care unit. Re- tients treated for peripheral artery oc- ratory tracts; superficial or surface peat angiography was performed the clusive disease from March 2003 to bleeding observed mainly at the vas- next day and every 24 hours subse- March 2005. All patients were referred cular puncture and access site or re- quently during infusion. for acute ischemia of recent onset (Ͻ14 cent surgical site; known history of left The patients were followed for ad- days). The first 22 limbs were treated heart thrombus, mitral stenosis with ditional events through completion of with TNK infusion at a rate of 0.25 atrial fibrillation, acute pericarditis, or their hospitalization and the next 30 mg/h (group 1) and the second group subacute bacterial endocarditis; preg- days. Additional follow-up to ascer- of 21 limbs received TNK at an infu- nancy; recent stroke or transient tain death and limb amputation was sion rate of 0.125 mg/h (group 2). The ischemic attack or any central nervous performed at 30 days and 6 months. lower infusion dose in group 2 was system neoplasm, bleeding, arterio- approved as a protocol change by the venous malformation, or aneurysm; Definitions hospital investigational review board. peripheral arterial graft placement less The average ages of patients were than 2 months earlier, infected grafts, Acute ischemia scores as described 55 years (SD, 13 y) in group 1 and 60 or inability to undergo thrombolytic by Rutherford and the Society of Vas- years (SD, 14 y) in group 2, which therapy; severe uncontrolled hyper- cular Surgery and International Soci- were not significantly different (P ϭ tension defined by blood pressure of ety of Cardiovascular Surgery (4) were .185). Overall, thrombolysis was per- 180/110 mm Hg or greater on re- obtained by physical and Doppler im- formed in 15 native arteries (35%), 25 peated measurements before the aging examinations (9). Infusion time bypass grafts (58%), and three previ- study; and acute leg ischemia of class was rounded to the nearest hour. ously stent-implanted arteries (7%), III or worse according to the Society of Technical success was defined as and there was no statistically signifi- Vascular Surgery and International removal of more than 95% of clot from cant difference in the number of grafts Society of Cardiovascular Surgery affected occlusion and distal runoff. treated in each group (P ϭ .254). scale for acute leg ischemia (4). Clinical success was assessed accord- The patient populations for groups ing to the SIR acute ischemia reporting ϩ 1 and 2 were similar in terms of age, Treatment Protocol scale (10) and defined by a result of 1 sex, grafts, lesion length, and lesion on the SIR ischemia scale (including location (Table 1). Cardiovascular risk All patients had diagnostic angiog- subcategories a, b, and c; Table 3). factors were also similar (Table 2). raphy of the affected limb performed Percent decrease in fibrinogen level before initiation of thrombolysis. The was defined as the admission fibrino- Inclusion and Exclusion Criteria occluded artery or graft was entered gen level minus the fibrinogen nadir with a guide wire and an infusion divided by the admission fibrinogen To be included in the study, pa- catheter of appropriate length (various level. tients were required to sign a written manufacturers) was placed within the Complications were ranked on a informed consent form before the pro- clot before thrombolytic infusion was scale of 1–3 as mild, moderate, or se- cedure, have acute ischemia of less initiated. Patients in group 1 received vere (Table 4). A severe complication than 14 days’ duration, be at least 18 TNK infused at a rate of 0.25 mg per was defined as a procedure-related Volume 17 Number 4 Hull et al • 631

kle-brachial index, postprocedural an- Table 2 Cardiovascular Risk Factors kle-brachial index, prothrombin time (maximum value measured per 12- Group 1 Group 2 hour period), preprocedural fibrino- Risk Factor (n ϭ 22) (n ϭ 21) P Value gen level, fibrinogen nadir (measured Coronary artery disease 8 (36) 10 (47) .860 each 12 hours), and percent decrease Smoking 18 (81) 15 (71) .475 in fibrinogen level. Lesion characteris- Hypertension 13 (59) 15 (71) .712 tics were lesion length and whether Hyperlipidemia 11 (50) 10 (47) .997 the occluded vessel was a native artery Diabetes 7 (32) 7 (33) .950 or bypass graft. Chronic renal failure 3 (14) 5 (24) .331 Chronic obstructive pulmonary disease 2 (9) 3 (14) .375 Carotid artery disease 2 (9) 3 (14) .375 Statistical Analysis

Note.—Numbers in parentheses are percentages. Summary statistics, logistic regression analysis, symmetric and asym- metric Student t test, and analysis of variance analysis (ANOVA) were used to evaluate for statistically signif- Table 3 SIR Acute Ischemia Scale icant differences between groups 1 and 2 and among variables (version SIR Acute Limb Ischemia Scale 8.1; Medcalc, Mariakerke, Belgium).

Score Description Total Group 1 Group 2 RESULTS Ϫ1 Worsening ischemia 1 (2) 1 (5) 0 (0) TNK infusions averaged 24 hours 0 No improvement 5 (12) 3 (14) 2 (10) Ϯ Ϯ ϩ1a Improvement with thrombolysis alone 5 (12) 2 (9) 3 (14) 14 in duration overall, 20 hours 12 ϩ1b Improvement with thrombolysis and 3 (7) 2 (9) 1 (5) in group 1, and 27 hours Ϯ 14 in group surgery 2. However, the difference in infusion ϩ1c Improvement with thrombolysis and 29 (67) 14 (64) 15 (71) times between the two groups did not endovascular procedure reach statistical significance (P ϭ .071). The average infusion doses were 5.02 Note.—There was no significant difference among groups. mg in group 1 and 3.43 mg in group 2. TNK boluses were pulsed into the occlusions of 36 limbs (84%). TNK boluses ranged from zero to 5 mg (0.25 event within 30 days of the procedure, were made with respect to all collected Ϯ including death, cerebral hemorrhage, data. Four main variable groups were mg/mL) and averaged 1.5 mg 1.2. In group 1, TNK boluses averaged 1.68 myocardial infarction, or amputation. used for statistical analysis and in- mg (SD, 1.48), and in group 2, TNK A moderate complication was defined clude main outcome variables, cardiac boluses averaged 1.24 mg (SD, 0.77), as a procedure-related event that re- risk factors, drug-related variables, without a significant difference be- quired an additional level of care. A and a combination of laboratory and ϭ mild complications was defined as an tween groups (P .224). The average lesion characteristics variables. total doses (bolus dose plus infusion event requiring treatment but not Main outcome variables were de- changing the level of care. dose) were 6.71 mg in group 1 and 4.67 fined as technical success, clinical suc- in group 2. The infusion dose and total A major bleeding complication was cess, SIR outcome criteria, complica- defined as a bleeding complication dose were significantly greater in tions, major bleeding complications, group 1 than in group 2 (P ϭ .043 and rated as moderate or severe as de- and 30-day and 6-month death and ϭ scribed earlier or requiring discontin- P 0.018). These and other drug-re- amputation rates. Cardiac risk factors lated parameters are summarized in uation of a thrombolysis procedure. included age, sex, smoking history, The relationship of each complica- Table 5. hypertension, hyperlipidemia, adult- Heparin infusions were purposely tion to the drug was ranked on a scale onset diabetes mellitus, coronary ar- of 1–4 as none, remote, possible, and subtherapeutic during thrombolysis, tery disease, chronic renal failure, even though, overall, 28 limbs (65%) probable (Table 4). The complication chronic obstructive pulmonary dis- outcomes were ranked on a scale of received heparin boluses of 3,000– ease, and carotid artery disease (Table 5,000 IU in the angiography suite. In 1–5 as resolved, improved, un- 2). For data analysis, these risk factors changed, worse, or death. group 1, 55% received heparin bo- were rated as present or absent. Drug- luses, versus 76% in group 2 (P ϭ related variables were use of heparin .224). The average maximum partial Data Analysis boluses, abciximab, TNK infusion thromboplastin time was 38 seconds dose, TNK infusion time, TNK bolus in group 1 and was significantly Quantitative and statistical compar- dose, and total TNK dose (infusion longer in group 2 at 59 seconds (P ϭ isons between the TNK 0.25 mg/h dose plus bolus dose). Laboratory .025). Seven limbs (16%) received ab- group and the TNK 0.125 mg/h group variables included preprocedural an- ciximab as a 0.25-␮g/kg bolus and a 632 • Tenecteplase for Thrombolysis in Lower-limb Ischemia April 2006 JVIR

Table 4 Complications Ranked by Severity Nadir Fibrinogen TNK Decrease in Level during Relation to Abciximab Bolus Fibrinogen Thrombolysis Rank Event Treatment TNK Used* (mg)* (%)* (mg/dL) Group 1 1 Fibrinogen decrease Blood products Probable No 5 82 64 1 Embolus Other None No 1.5 15 330 1 PTA; ruptured graft Other None No 0 18 329 3 Hematemesis Stop TNK Probable Yes 5 64 151 3 Level III ischemia Stop TNK and other Possible No 1.5 0 234 treatment 3 Compartment syndrome Stop TNK None No 0.5 30 275 Group 2 1 Allergic reaction Other drug treatment None No 0 7 291 1 Dissection Other None Yes 0.5 26 418 2 Gastrointestinal bleeding Blood products Probable Yes 3 32 389 2 Late hematoma Blood products None No 0 Ϫ6 424 3 Repair of brachial artery Other Possible Yes 1.5 Ϫ55 620 puncture

* Significant association with TNK-related complications by ANOVA. Note.—PTA ϭ percutaneous transluminal angioplasty. All adverse events were resolved.

12-hour infusion of 0.125 mg/kg/h. A Table 5 significant majority of these limbs Comparison of Drug-related Parameters (86%; six of seven) were in group 2 (P ϭ .039). Group 1 Group 2 Technical success was achieved in Parameter (n ϭ 22) (n ϭ 21) P Value 84% of limbs overall (36 of 43), with an Mean TNK infusion time SD (h) 20 Ϯ 12 27 Ϯ 14 .071 82% success rate in group 1 and an TNK infusion dose (mg) 5.02 3.43 .043* 86% success rate in group 2 (P ϭ .827; TNK bolus (mg) 1.68 1.24 .224 Table 6). The average ankle-brachial TNK total dose (mg) 6.71 4.67 .018* indexes was 0.34 before thrombolysis Patients who received heparin bolus (%) 55 76 .224 and 0.70 after thrombolysis (P ϭ Average maximum PTT (sec) 38 59 .025* .0001). Clinical success was achieved Patients who received Abciximab (%) 5 29 .039* in 82% of patients in group 1 and in Fibrinogen level before treatment (mg/dL) 359 435 .035* 90% of patients in group 2 (P ϭ .422). Fibrinogen nadir (mg/dL) 272 400 .001* Mean decrease in fibrinogen level SD (%) 23 Ϯ 29 7 Ϯ 20 .045* Outcome measures are described in Table 6. The SIR scale for results of * Statistically significant difference on Student t test. acute limb ischemia (10) showed an Note.—PTT ϭ partial thromboplastin time. improvement (ie, ϩ1) in 86% of limbs (37 of 43), stayed the same (ie, 0) in 12% of limbs (five of 43), and worsened (ie, –1) in 2% of cases (one of 43). The breakdown of results into SIR subcategories by group is given in Table 3. Table 6 Outcome Measures Ten limbs (13%) had thrombolysis with TNK and no additional proce- Group 1 Group 2 dures; five of these cases (50%) had Outcome (n ϭ 22) (n ϭ 21) P Value technical and clinical success (SIR cat- ϩ Technical success 18 (82) 18 (86) .736 egory 1a). Additional endovascular Clinical success 18 (82) 19 (90) .422 procedures were required in 30 of 43 Ankle-brachial index limbs (70%), 29 of which had a clini- Before procedure 0.36 .032 .685 cally successful outcome. Overall, After procedure 0.69 0.71 .838 therefore, 67% of limbs (29 of 43) were Major bleeding episode 1 (5) 2 (10) .537 classified as showing SIR category ϩ1c Death and/or amputation at 6 months 1 (5) 3 (14) .290 improvement: 47% were treated with Note.—Values in parentheses are percentages. angioplasty (n ϭ 20), 30% received a Volume 17 Number 4 Hull et al • 633 stent (n ϭ 13), 5% were treated with TNK (P Ͻ .001). There were no pa- rapid progression to stage III ischemia aspiration embolectomy (n ϭ 2), and tients with complications classified as in the first hour of thrombolysis re- 12% were treated with pharmacome- remotely related to TNK. Similarly, 32 quiring conversion to surgery, angio- chanical thrombolysis (n ϭ 5). Three limbs with no complications had a plasty with rupture of a vein graft limbs required surgery to complete 13% decrease in fibrinogen level, six treated with a stent, fasciotomy for successful thrombolysis (SIR category limbs with complications with no re- compartment syndrome, surgical re- ϩ1b) and these cases included a fas- lation to TNK had a 15% decrease in pair of brachial artery, anaphylactic re- ciotomy for compartment syndrome, a fibrinogen level, two limbs with pos- action to antibiotics, and a dissection. jump revision of a femoral popliteal sible relation to TNK had a 28% in- Fibrinogen levels averaged 396 bypass graft, and thromboendarterec- crease in fibrinogen level (fibrinogen mg/dL before thrombolysis and 335 tomy with patch angioplasty of the increased 56% in one of two patients in mg/dL after thrombolysis (P ϭ .001). common femoral artery. this group), and three limbs with The percentage decrease in fibrinogen Complications occurred in 26% of probable relation to TNK had a 59% level averaged 15% overall (SD, 26%), limbs overall (11 of 43), with a 27% decrease in fibrinogen level (P ϭ .001). with a significantly higher percentage incidence in group 1 (n ϭ 22) and a The finding of a relationship be- decrease in fibrinogen level in group 1 24% incidence in group 2 (n ϭ 5; P ϭ tween TNK bolus and TNK-related (23% Ϯ 29%) compared with group 2 .800). Overall, complications were as- complications prompted further anal- (7% Ϯ 20%; P ϭ .045). Patients with no sociated with increased age and use of ysis to discover if there was a TNK complications had a 13% decrease in abciximab (logistic regression, P ϭ bolus that was tolerated without a sig- fibrinogen level, patients with compli- .043 and P ϭ 0.044). ANOVA showed nificant increase in complications. cations with no relation to TNK had a only a trend toward increased compli- Findings of an ANOVA of all compli- 15% decrease in fibrinogen level, pa- cations with age, with a mean patient cations versus TNK boluses ranked tients with complications with a pos- age of 64 years in limbs with compli- from zero to 5 mg in 0.5-mg incre- sible relation to TNK had a Ϫ28% de- cations and a mean age of 55 years in ments were not statistically signifi- crease (ie, fibrinogen increased), and limbs without complications (P ϭ cant, probably as a result of small sam- patients with complications with a .077). ANOVA showed that 57% of ple sizes (P ϭ .052). However, TNK probable relation to TNK had a 59% limbs treated with abciximab had boluses of 1.5 mg or less were well- decrease (P ϭ .001). The percentage complications (n ϭ 4), compared with tolerated, with a 22% overall compli- decrease in fibrinogen level when not 19% of those that were not treated (n ϭ cation rate (eight of 36), whereas TNK ranked by likelihood of TNK as a caus- 7; P ϭ .037). boluses of 3–5 mg had a 75% compli- ative factor was not associated with The complications ranked by sever- cation rate (three of four; student t test, complications, major bleeding, or ity showed that, overall, 74% of limbs P ϭ .025). Of note, only two of eight death and amputation rates. There (n ϭ 32) had no complications, 12% limbs in the 1.5-mg bolus group had were no major bleeding complications had mild complications (n ϭ 5), 5% complications ranked as possibly in the three limbs (7%) with fibrinogen had moderate complications (n ϭ 2), TNK-related, whereas all three of four levels decreasing to less than 100 mg/ and 9% had severe complications (n ϭ limbs in the 3–5-mg bolus group were dL, even though one patient was clas- 4). On ANOVA, complication severity ranked as probably related to TNK. sified as having a minor complication had no significant correlation with car- There were three major bleeding when treated prophylactically with diovascular risk factors, main outcome complications (7%), with a 5% inci- fresh frozen plasma and cryoprecipi- variables, drug-related variables, or dence in group 1 (one of 22) and a 10% tate for an 83% decrease in fibrinogen laboratory variables. Table 4 shows incidence in group 2 (two of 21; P ϭ level from 359 mg/dL to 64 mg/dL. complications by group ranked by se- .537). ANOVA showed statistically All three limbs with fibrinogen levels verity, treatment, relationship to TNK, significant associations of male sex decreasing to less than 100 mg/dL and outcome of complication. and abciximab use with major bleed- were in group 1 (P ϭ .083). The fibrin- Complications, when rated as to the ing complications (P ϭ .026 and P ϭ ogen level before thrombolysis was likelihood of TNK as a causative factor .014, respectively). The major bleeding higher in limbs in which death or am- (ie, none, remote, possible, probable), complications were one case of upper putation had occurred at 30 days and 6 exhibited significant correlations on gastrointestinal bleeding from a previ- months, with respective means of 598 ANOVA with initial TNK bolus, use of ously unknown duodenal ulcer re- mg/dL and 538 mg/dL compared abciximab, percent decrease in fibrin- quiring transfusion of 2 U packed red with 381 mg/dL and 382 mg/dL, re- ogen level, and major bleeding (P Ͻ blood cells, a case of gastrointestinal spectively, in limbs in which death or .001, P ϭ .036, P ϭ .001, and P ϭ .004, bleeding requiring transfusion, and a amputation did not occur (P ϭ .001 respectively), but not with TNK infu- groin hematoma requiring transfu- and P ϭ .010). sion time or total dose of TNK (P ϭ sion. A minor TNK-related bleeding Abciximab was used in 16% of .279 and P ϭ .075). The average TNK complication in the form of an asymp- limbs (n ϭ 7), 5% in group 1 (one of 22) bolus was 1.4 mg in 32 patients with tomatic but precipitous decrease in fi- and 29% in group 2 (six of 21; P ϭ no complications, 0.5 mg in six pa- brinogen level of 82% (from 359 .039). Significant statistical associa- tients with complications not related mg/dL to 64 mg/dL) was treated with tions were found by ANOVA between to TNK, 1.5 mg in two patients with fresh frozen plasma and cryoprecipi- abciximab and complications, compli- complications possibly related to tate during continued TNK infusion. cation rank, major bleeding, and am- TNK, and 4.3 mg in three patients with Other complications included embo- putation at 30 days and 6 months (P ϭ complications probably related to lus requiring aspiration embolectomy, .037, P ϭ .014, P ϭ .014, P Ͻ .001, and 634 • Tenecteplase for Thrombolysis in Lower-limb Ischemia April 2006 JVIR

P Ͻ .001, respectively). Negative asso- fective thrombolysis can be achieved could possibly allow for greater levels ciations were found with technical at a dose that causes significantly less of TNK to circulate in the blood- success, but not clinical success (P ϭ fibrinogen depletion. stream. The increased resistance to .038 and P ϭ .232). The average infu- Our results showed that TNK-re- plasminogen activator inhibitor and sion time for limbs treated with abcix- lated complications were related to increased effectiveness in arterial imab was 36 hours compared with 21 TNK bolus, use of abciximab, percent- thrombolysis should make TNK more hours for limbs not treated (P ϭ .009). age decrease in fibrinogen level, and effective at the site of thrombosis. The The overall amputation rates at 30 major bleeding. The correlation with increased fibrin specificity and de- days and 6 months were 7% (three of major bleeding is expected, and the creased fibrinogen depletion may be 43) and 9% (four of 43), respectively. correlations with abciximab and per- part of a mechanism that could lead to All amputations were below the knee. centage decrease in fibrinogen level fewer bleeding complications. The death rates were zero at 30 days are discussed later. More surprising One prevalent theory suggests that and 2% (one of 43) at 6 months. One was the correlation of complications bleeding with tPA and its genetic mu- patient died after amputation. There with TNK bolus but not with infusion tations reteplase and TNK is caused by were no statistically significant differ- time, TNK dose, or TNK total dose. fibrinogenolysis in the peripheral cir- ences between groups 1 and 2 with Review of the TNK boluses showed culation with the accumulation of respect to death and amputation rates that, even though 0–5-mg boluses fragment X (11–13). Fragment X can at 30 days (P ϭ .537) or 6 months (P ϭ were allowed, 5-mg boluses were used clot and forms easily clots at sites of .290). ANOVA showed statistically in only the first three cases, two of potential bleeding that can be easily significant association of death and which had TNK-related complications lysed. If and when these clots with amputation rates with use of abcix- (hematemesis and prominent decrease fragment X break down, bleeding can imab, preprocedural fibrinogen levels, in fibrinogen level). Our protocol used occur at unintended sites. For fibrin- and preprocedural ABI. previously reported amounts for the ogenolysis to occur, the thrombolytic initial bolus (7). On further reflection, drug in the circulation needs to bind to DISCUSSION the authors thought 5 mg might be circulating DDE (a degradation prod- excessive in that a 24-hour infusion uct of thrombolysis) and circulating The current study shows that TNK would use 6 mg of TNK when infused plasminogen. Fibrinogenolysis also infusion at 0.125 mg/h provides re- at a rate of 0.25 mg/h. A systemic lytic depletes fibrinogen (14). Thrombolytic sults similar to infusion at 0.25 mg/h. state might be promoted by TNK bo- drugs with no or low fibrin specificity In turn, these results are comparable luses of this magnitude. Based on our (eg, urokinase and streptokinase) bind to the earlier reports in the literature experience with other thrombolytic poorly to DDE, producing little frag- with use of TNK at 0.25 mg/h and 0.5 agents (ie, 250,000 to 1 million U of ment X. The tissue-type plasminogen mg/h in terms of technical success, urokinase bolus with 2.4 million U in- activator drugs (eg, tPA, reteplase, clinical success, fibrinogen depletion, fused over 24 hours), we began to limit TNK) can strongly bind DDE, but less complications, and bleeding. Our tech- our boluses to 0–3 mg TNK, most so as the drug becomes more fibrin- nical and clinical success with TNK commonly 1.5 mg. Our data confirmed specific. Stewart et al (13) showed in was consistent with these reports of the concept of limiting initial boluses vitro that TNK produces less fragment TNK administered for acute limb isch- in that TNK boluses of 1.5 mg or less X than does tPA, which produces less emia. In these earlier reports, the tech- were well-tolerated, with a 22% inci- fragment X than reteplase, a pattern nical and clinical success rates were dence of complications (eight of 36), consistent with the drugs’ fibrin spec- 73%–88%, major bleeding complica- most of which (n ϭ 6) were not TNK- ificity (ie, that of TNK is greater than tion rates were 1.8%–10.4%, there were related. TNK boluses of 3–5 mg had a that of tPA, which is greater than that no intracranial hemorrhages, and fi- 75% complication rate (three of four), of reteplase). On this basis, one might brinogen levels decreased 23%–41% with all three categorized as probably expect less bleeding and fibrinogen (5–8). In group 1, we used the same TNK-related. depletion when using TNK versus tPA infusion dose of TNK (0.25 mg/h) as or reteplase. in the previous reports and had nearly Fibrinogen Depletion and Bleeding In our data, it is interesting to note identical results. Our clinical success Complications that there was significantly less fibrin- rate was 86%, major bleeding inci- ogen depletion at the lower infusion dence was 7%, there were no intracra- As noted previously, compared dose, with a 7% decrease in fibrinogen nial hemorrhages, and the percentage with tPA, TNK is a drug with a longer in group 2 compared with a 23% de- decrease in fibrinogen level averaged half-life of 22 minutes, an 80 times crease in group 1. The percent de- 23%. In group 2, at half the dose, our higher resistance to plasminogen acti- crease in fibrinogen of 15% overall in clinical success rate was 90%, major vator inhibitor, a twofold improve- our patients and a low major bleeding bleeding incidence was 10%, there ment of effectiveness in arterial throm- complication rate of 7% may support were no intracranial hemorrhages, and bolysis, increased fibrin specificity, the clinical significance of fibrin spec- percentage decrease in fibrinogen and decreased fibrinogen depletion ificity and dose with respect to fibrin- level was 7%. It was encouraging to (1–3). The longer half-life allows for ogenolysis in vivo. This notion was see that our clinical success rate re- single bolus administration for acute further supported by our data show- mained high while the decrease in fi- myocardial infarction. In infusion for ing that complications rated by likeli- brinogen level was significantly less acute lower-limb ischemia, a longer hood of TNK being the cause strongly severe in group 2, suggesting that ef- half-life may have no advantage and correlated with percent drop in fibrin- Volume 17 Number 4 Hull et al • 635 ogen, and major bleeding complica- Intracranial Hemorrhage imab with complications and poor tions. In the clinical literature, fibrino- outcomes may be related to small sam- gen depletion during thrombolysis There were no intracranial hemor- ple size and selection bias; however, with alteplase (15), reteplase (16,17), rhages in our series. However, at our further evaluation of glycoprotein IIb/ and TNK (7,8) has been reported to institution, we have seen an intra- IIIa inhibitors is needed in acute limb cause a drop in fibrinogen ranging cranial hemorrhage with TNK in a ischemia. TNK may have advantages from 23%–59%. In these reports, a con- patient treated for upper-extremity over other tPAs in terms of fibrin spec- sistent relationship between decrease embolus complicated by an asymp- ificity, which may be clinically rele- in fibrinogen, drug used, and infusion tomatic and initially undetected right vant in the treatment of acute lower- dose cannot be established. Earlier re- carotid artery embolus, confirming limb ischemia. Fibrinogen depletion ports have described an association that this complication may be seen was significantly decreased in the with bleeding complications and fi- when TNK is used for thrombolytic lower dose group and was signifi- brinogen depletion with use of alte- infusion. Intracranial hemorrhage has cantly increased in patients with com- plase and reteplase (18,19), whereas been a major source of concern in pa- plications caused by TNK, suggesting several other reports demonstrated no tients undergoing thrombolysis for a clinical link among TNK dose, fibrin- correlation (16,20,21). A recent article acute myocardial infarction, acute ogen depletion, and complications. reporting the use of reteplase de- limb ischemia, pulmonary embolism, and deep vein thrombosis (25–29). The scribes a 15% decrease in fibrinogen References level as correlating with no bleeding rate of intracranial hemorrhage during thrombolysis is reported in the range 1. Collen D, Stassen JM, Yasuda T, et al. complications, whereas a decrease of Comparative thrombolytic properties 72% was associated with major bleed- of 1%–3%. In acute myocardial infarc- of tissue-type plasminogen activator ing complications in 23% of patients tion, the rate was reported to be less and of a plasminogen activator inhibi- (19). Our data similarly show a 15% than 1% in a review of 71,073 patients tor-1-resistant glycosylation variant, in decrease in fibrinogen levels in pa- (25). In a registry of 2,454 patients a combined arterial and venous throm- tients with no complications and a treated for pulmonary embolism, 3% bosis model in the dog. Thromb Hae- 59% decrease in patients with compli- of patients had intracranial hemor- most 1994; 72:98–104. cations caused by TNK. The percent- rhage (28). The incidences of intracra- 2. Keyt BA, Paoni NF, Refino CJ, et al. A nial hemorrhage in patients treated for faster-acting and more potent form of age decrease in fibrinogen appears to tissue plasminogen activator. Proc Natl be a useful marker for the evaluation acute limb ischemia and deep vein thrombosis have ranged from 0.3% to Acad SciUSA1994; 91:3670–3674. of the performance of thrombolytic 3. Verstraete M. Third-generation throm- drugs in patients with acute lower- 2% (27,29,30). TNK has been shown to bolytic drugs. Am J Med 2000; 109:52–58. limb ischemia and may help establish cause less intracranial hemorrhage 4. Assessment of the Safety and Efficacy a link between thrombolytic drug, than alteplase in animal studies (31), of a New Thrombolytic Investigators. dose, fibrinogen depletion, and bleed- but this benefit has not been verified in Single-bolus tenecteplase compared ing complications. humans treated for acute myocardial with front-loaded alteplase in acute infarction (32). There were no intracra- myocardial infarction: the ASSENT-2 nial hemorrhages in our series or in double-blind randomised trial. Lancet Abciximab the other reports of TNK use in the 1999; 354:716–722. 5. Allie DE, Hebert CJ, Lirtzman MD, et The negative association of abcix- peripheral circulation (5–8,33). Intra- cranial hemorrhage will continue to be al. Continuous tenecteplase infusion imab use on technical success and its combined with peri/postprocedural positive associations with complica- a feared, expected, and hopefully rare platelet glycoprotein IIb/IIIa inhibition tions in the current study are confus- complication of thrombolytic therapy. in peripheral arterial thrombolysis: ini- ing. There was more abciximab use in The main limitations of this study tial safety and feasibility experience. J group 2 than in group 1, but the over- are the small sample size, lack of ran- Endovasc Ther 2004; 11:427–435. all outcomes of technical success, clin- domization, and inconsistent use of 6. Burkart DJ, Borsa JJ, Anthony JP, et al. ical success, complications, major abciximab, allowing for possible er- Thrombolysis of occluded peripheral bleeding, and SIR ischemic outcomes rors in analysis. Topics for future eval- arteries and veins with tenecteplase: a uation include lower TNK doses, use pilot study. J Vasc Interv Radiol 2002; between the groups were not signifi- 13:1099–1102. cantly different. The fact that TNK in- of glycoprotein IIb/IIIa inhibitors, al- ternative drug delivery catheters, and 7. Burkart DJ, Borsa JJ, Anthony JP, et al. fusion times (average, 36 hours) asso- Thrombolysis of acute peripheral arte- ciated with abciximab use were embolic protection devices. rial and venous occlusions with tenect- significantly longer than in patients eplase and eptifibatide: a pilot study. J who were not treated with abciximab CONCLUSION Vasc Interv Radiol 2003; 14:729–733. suggests possible selection bias. The 8. Razavi MK, Wong H, Kee ST, et al. potential benefits of abciximab re- TNK was equally effective at infu- Initial clinical results of tenecteplase ported in the literature include de- sion doses of 0.25 mg/h and 0.125 (TNK) in catheter-directed thrombo- creased thrombolysis time, decreased mg/h, with similar success and com- lytic therapy. J Endovasc Ther 2002; 9:593–598. embolization, improved patency, and plication rates. Initial boluses of TNK 9. Rutherford R, Baker J, Ernst C, et al. improved thrombolysis after initial should be limited to 1.5 mg, as higher Recommended standards for reports failure (19,22–24). In these studies, doses were associated with increased dealing with lower extremity ischemia: there were no reports of increased TNK-related complications. The con- revised version. J Vascular Surgery complications. sistent association of the use of abcix- 1997; 26:517–538. 636 • Tenecteplase for Thrombolysis in Lower-limb Ischemia April 2006 JVIR

10. Patel N, Sacks D, Patel RI, et al. SCVIR spective randomized trial evaluating istry of Myocardial Infarction 2. Ann reporting standards for the treatment of surgery versus thrombolysis for isch- Intern Med 1998; 129:597–604. acute limb ischemia with use of translu- emia of the lower extremity. The STILE 26. Ouriel K, Gray B, Clair DG, et al. minal removal of arterial thrombus. J trial. Ann Surg 1994; 220:251–266. Complications associated with the use Vasc Interv Radiol 2001; 12:559–570. 19. Hull JE, Hull MK, Urso JA. Reteplase of urokinase and recombinant tissue 11. Weitz JI. Limited fibrin specificity of with or without abciximab for periph- plasminogen activator for catheter-di- tissue-type plasminogen activator and eral arterial occlusions: efficacy and ad- rected peripheral arterial and venous its potential link to bleeding. J Vasc verse events. J Vasc Interv Radiol 2004; thrombolysis. J Vasc Interv Radiol Interv Radiol 1995; 6(suppl):19S–23S. 15:557–564. 2000; 11:295–298. 12. Owen J, Friedman KD, Grossman BA, 20. McNamara TO, Dong P, Chen J, et al. 27. Costantini V, Lenti M. Treatment of et al. Quantitation of fragment X for- Bleeding complications associated with acute occlusion of peripheral arteries. mation during thrombolytic therapy the use of rt-PA versus r-PA for periph- Thromb Res 2002; 106:V285–V294. with streptokinase and tissue plasmin- eral arterial and venous thromboem- 28. Goldhaber SZ. Modern treatment of ogen activator. J Clin Invest 1987; 79: bolic occlusions. Tech Vasc Interv Ra- pulmonary embolism. Eur Respir J 1642–1647. diol 2001; 4:92–98. 2002; 35(suppl):22s–27s. 13. Stewart RJ, Fredenburgh JC, Leslie BA, 21. Ouriel K, Katzen B, Mewissen M, et al. 29. Mewissen MW, Seabrook GR, Meiss- et al. Identification of the mechanism Reteplase in the treatment of periph- ner MH, et al. Catheter-directed responsible for the increased fibrin eral arterial and venous occlusions: a thrombolysis for lower extremity deep specificity of TNK-tissue plasminogen pilot study. J Vasc Interv Radiol 2000; venous thrombosis: report of a national activator relative to tissue plasminogen 11:849–854. multicenter registry. Radiology 1999; activator. J Biol Chem 2000; 275:10112– 22. Drescher P, Crain MR, Rilling WS. 211:39–49. 10120. Initial experience with the combination 30. Ouriel K, Veith FJ, Sasahara AA. A 14. Weitz JI, Leslie B, Ginsberg J. Soluble of reteplase and abciximab for throm- comparison of recombinant urokinase fibrin degradation products potentiate bolytic therapy in peripheral arterial with vascular surgery as initial treat- tissue plasminogen activator-induced fibrinogen proteolysis. J Clin Invest occlusive disease: a pilot study. J Vasc ment for acute arterial occlusion of the 1991; 87:1082–1090. Interv Radiol 2002; 13:37–43. legs. Thrombolysis or Peripheral Arte- 15. Bero CJ, Cardella JF, Reddy K, et al. 23. Nakada MT, Montgomery MO, Nedel- rial Surgery (TOPAS) Investigators. N Recombinant tissue plasminogen acti- man MA, et al. Clot lysis in a primate Engl J Med 1998; 338:1105–1111. vator for the treatment of lower ex- model of peripheral arterial occlusive 31. Smalling RW. Molecular biology of tremity peripheral vascular occlusive disease with use of systemic or intraar- plasminogen activators: what are the disease. J Vasc Interv Radiol 1995; terial reteplase: addition of abciximab clinical implications of drug design? 6:571–577. results in improved vessel reperfusion. Am J Cardiol 1996; 78:2–7. 16. Davidian MM, Powell A, Benenati JF, J Vasc Interv Radiol 2004; 15:169–176. 32. Dunn CJ, Goa KL. Tenecteplase: a re- et al. Initial results of reteplase in the 24. Ouriel K, Castaneda F, McNamara T, et view of its pharmacology and thera- treatment of acute lower extremity ar- al. Reteplase monotherapy and rete- peutic efficacy in patients with acute terial occlusions. J Vasc Interv Radiol plase/abciximab combination therapy myocardial infarction. Am J Cardio- 2000; 11:289–294. in peripheral arterial occlusive disease: vasc Drugs 2001; 1:51–66. 17. Drescher P, McGuckin J, Rilling WS, results from the RELAX trial. J Vasc 33. Allie DE, Hebert CJ, Lirtzman MD, et et al. Catheter-directed thrombolytic Interv Radiol 2004; 15:229–238. al. Novel simultaneous combina- therapy in peripheral artery occlusions: 25. Gurwitz JH, Gore JM, Goldberg RJ, et tion chemical thrombolysis/rheolytic combining reteplase and abciximab. al. Risk for intracranial hemorrhage thrombectomy therapy for acute criti- AJR Am J Roentgenol 2003; 180:1385– after tissue plasminogen activator cal limb ischemia: the power-pulse 1391. treatment for acute myocardial infarc- spray technique. Cathet Cardiovasc In- 18. STILE Investigators. Results of a pro- tion. Participants in the National Reg- terv 2004; 63:512–522.