A comparison of thrombolytic therapy with operative revascularization in the initial treatment of acute peripheral arterial ischemia

ORIGINAL ARTICLES

A comparison of thrombolytic therapy with operative revascularization in the initial treatment of acute peripheral arterial ischemia Kenneth Ourid, MD, Cynthia K. Shortell, MD, James 1\. DeWeese, MD, Richard M. Green, MD, Charles W. Francis, MD, Michael V. U. Azodo" MD, Oscar H. Gutierrez, MD, James V. Manzione, MD, Christopher Cox, PhD, and Victor J. Marder, MD, Rochester) NY. Purpose: Despite the widespread use of intraarterial thrombolytic therapy for peripheral

arterial occlusive disease, a randomized study comparing its efficacy with that of operative intervention has never· been performed. This study evaluates the potential of intraarterial urokinase infusion to provide clinical benefits in patients with acute peripheral arterial occlusion. Methods: Patients with limb-threatening ischemia of less than 7 days' duration were randomly assigned to intraarterial catheter-directed urokinase therapy or operative intervention. Anatomic lesions unmasked by thrombolysis were treated with balloon dilation or operation. The primary end points of the study were limb salvage and survival. Results: A total of 57 patients were randomized to the thrombolytic therapy group, and 57 patients were randomized to the operative therapy group. Thrombolytic therapy resulted in dissolution of the occluding thrombus in 40 (70%) patients. Although the cumulative limb salvage rate was similar in the two treatment groups (82% at 12 months), the cumulative survival rate was significandy improved in patients randomized to the thrombolysis group (84% vs 58% at 12 months, p = 0.01). The mortality differences seemed to be primarily attributable to an increased frequency of in-hospital cardiopulmonary complications in the operative treatment group (49% vs 16%, P = 0.001). The benefits of thrombolysis were achieved without significant differences in the duration of hospitalization (median 11 days) and with only modest increases in hospital cost in the thrombolytic treatment arm (median $15,672 vs $12,253, P = 0.02). Conclusions: Intraarterial thrombolytic therapy was associated with a reduction in the incidence of in-hospital cardiopulmonary complications and a corresponding increase in patient survival rates. These benefits were achieved without an appreciable increase in the duration of hospitalization and with only modest increases in hospital cost, suggesting that thrombolytic therapy may offer a safe and effective alternative to operation in the initial treatment of patients diagnosed with acute limb-threatening peripheral arterial occlusion. (J VASe SURG 1994;19:1021-30.)

From the Departments of Surgery, Medicine (Drs. Francis and Marder), Radiology (Drs. Azodo, Manzione, and Gutierrez), and Biostatistics (Dr. Cox), University of Rochester. Supported by grants from the National Institutes of Health, National Heart, Lung, and Blood Institute, HL 40889, HL 30616, and HL 07152. Reprint requests: Kenneth Ouriel, MD, University of Rochester, 601 Elmwood Ave., Rochester, NY 14642. Copyright © 1994 by The Society for Vascular Surgery and International Society for Cardiovascular Surgery, North American Chapter. 0741-5214/94/$3.00 + 0 24/1/52316

At present, more than 100,000 operations for acute and chronic ischemia of the extremities are performed annually in the United States. 1 Although operative revascularization procedures have been associated with acceptable rates of limb salvage, a substantial risk of perioperative death has prompted some to advocate nonoperative treatment modalities in the setting of acute peripheral arterial occlusion. 2 Thrombolytic agents have been used in this regard, 1021

1022 Ouriel et al.

providing the opportunity to restore arterial perfusion in a less invasive fashion. 3·7 The widespread use of intraarterial thrombolytic therapy has been based on perceived benefits over operative treatment,8-12 yet these benefits have not been established by a controlled clinical trial. Previous reports have not provided an objective analysis of the issues, primarily as a result of the use of historical control data, nonrandomized study designs, and short length of follow-up with subjective, ill-defined end points. The question of whether intraarterial thrombolytic therapy offers any advantage over operative revascularization remains unanswered. Herein, we report the results of a prospective, randomized comparison of intraarterial thrombolytic therapy with operative revascularization in the initial treatment of acute peripheral arterial occlusion. METHODS The study was designed to assess the efficacy of thrombolytic therapy as the initial treatment of acute peripheral arterial ischemia, comparing the clinical outcome to that of operative revascu1arization. Patients meeting defined entry criteria were randomized to intraarterial thrombolytic therapy or operative revascularization. Patients assigned to the thrombolytic treatment arm underwent definitive correction of any anatomic lesions uncovered after lysis of the occluding thrombus. Patient eligibility. Eligible patients were aged 18 years and older and were admitted to the University of Rochester Medical Center with acute limb ischemia of less than 7 days' duration. Patient acquisition began on August 5, 1989, and continued through April 18, 1992. Patients with embolic or thrombotic native arterial, autogenous bypass graft, and prosthetic bypass graft occlusion were included. The ischemic process was required to be of a limb-threatening nature, such that amputation would be necessary without intervention. With the terminology and categories suggested by the Ad Hoc Committee on Reporting Standards of The Society for Vascular Surgery and the International Society for Cardiovascular SurgeryIN orth American Chapter,13 all patients were categorized within class II. Patients with a presumed embolic cause for the arterial occlusion underwent echocardiography and were excluded if mural thrombus was visualized. Patients were excluded from study if they manifested a contraindication to thrombolytic therapy, including one or more of the following: a major operative procedure within 14 days, active peptic ulcer disease, an intracranial neoplasm, or a history of a cerebrovas-

JOURNAL OF VASCULAR SURGERY June 1994

cular accident. Patients were also excluded if they had a contraindication to operative revascularization, if they were nonambulatory or otherwise had a nonfunctional extremity before the ischemic event, if the ischemic process was deemed irreversible (Society for Vascular Surgery/International Society for Cardiovascular Surgery class III), or if a contraindication to arteriography was present, including a serum creatinine level greater than 2.5 mg/dl or a history of significant allergy to contrast agents. Women of reproductive age underwent a urine pregnancy test and were excluded if the test result was positive. Informed consent was obtained in all patients by means of a printed consent form approved by the institution's Research Subjects Review Board. Randomization. Eligible patients were randomly assigned to receive thrombolytic or operative treatment according to computer generated randomization cards, the envelopes of which remained sealed until the time of entry into the study. A total of 114 patients were enrolled,S 7 into the thrombolytic arm and 57 into the operative arm. There were three nonrandomized, eligible patients who were admitted during the period of study and refused to consent to participate. There were two additional patients who were not randomized on the basis of rapidly progressing ischemia and a presumed detrimental effect on limb salvage as a result of the obligatory delays associated with arteriography. Thus 114 (96%) of 119 patients meeting the defined entry criteria were enrolled in the trial. Baseline investigations. Patients underwent standardized history taking, tabulating the presence of risk factors associated with atherosclerosis. Patients were assigned a cardiac risk index on the basis of the criteria outlined by Goldman et al. 14 The physical examination focused on a detailed evaluation of the extremities, and the laboratory assessment included plasma fibrinogen concentration. 15 Doppler ankle-brachial pressure indexes were measured in all patients. Treatment. Antiplatelet therapy was instituted with 325 mg aspirin administered orally immediately after entry into the study. Those patients assigned to receive thrombolytic therapy underwent standard intraarterial diagnostic arteriography, placing the catheter through a site distant from the segment of arterial occlusion. The catheter infusion holes were positioned within the substance of the thrombus when feasible, infusing thrombolytic agent directly into the clot. Thrombolytic therapy was initiated with an infusion of 4000 IU urokinase (Abbott Laboratories, North Chicago, Ill.) per minute at a

JOURNAL OF VASCULAR SURGERY Volume 19, Number 6

concentration of 5000 IVlml in 5% dextrose in water. The rate of infusion was reduced to 2000 IVlmin 2 hours after the institution of therapy and to 1000 IVImin 4 hours after the institution of therapy. Systemic heparin anticoagulation was not used in conjunction with the thrombolytic agent. Arteriography was repeated after 4, 12, 24, and 48 hours of therapy, and at additional times when clinically indicated. Thrombolytic therapy was terminated at a maximum of 48 hours, regardless of clinical outcome. Therapy was terminated earlier if major complications developed, if successful lysis of thrombus was documented on arteriography, or if clinical deterioration progressed during the period of infusion. Sudden clinical deterioration of the distal extremity or digits occurring as a result of dissolution and secondary embolization of thrombotic material was not considered to be a criterion for termination of therapy; rather, this event signified the need for continued infusion of thrombolytic agent to dissolve the embolized thrombus. Subsequent intervention in the form of operative revascularization or percutaneous balloon angioplasty was undertaken for failure to effect significant thrombolysis, for progressive clinical deterioration despite therapy, for residual intraluminal thrombus, or for unmasked anatomic lesions such as inflow stenotic disease, bypass graft lesions, or outflow occlusive processes. Patients with documented anatomic lesions explaining the cause of the vascular occlusion were treated with heparin after the termination of thrombolysis, until operative revascularization or percutaneous balloon angioplasty could be implemented to correct the defect. Oral anticoagulation with warfarin was instituted in all patients experiencing embolic events and in patients in whom successful thrombolysis failed to reveal the presence of a correctable anatomic lesion explaining the occlusion. Patients randomized to operative revascularization underwent diagnostic arteriography except when deemed inadvisable on the basis of delays associated with the performance of the test (three patients). The choice of operative procedure was made by the attending surgeon, and operation was undertaken as soon as was feasible in each case. Primary amputation was elected when preoperative arteriography and intraoperative exploration of distal vessels failed to demonstrate a suitable outflow site. Postoperative anticoagulation with warfarin was administered in patients requiring prosthetic bypass grafts to infragenicular arteries, in patients with embolic events, and in patients with presumed hypercoagulable states.

Ouriel et al.

1023

End points. The primary end points of the study were patient survival and limb salvage. Event-free survival was defined as the absence of death or major amputation during follow-up. Other outcome measures included major complications (hemorrhage that necessitated transfusion or operative intervention, false aneurysm formation, or other major untoward events), minor complications (fever, allergic reactions, and insignificant bleeding episodes), length of initial hospital stay, cost of initial hospitalization, and requirement for succeeding hospitalizations for limb ischemia. The ~'time to reperfusion" was calculated as the duration of thrombolytic infusion required to initially achieve flow through the occluded native artery or bypass conduit, documented by noninvasive laboratory findings or arteriography. Laboratory outcome variables included plasma fibrinogen concentration and Doppler anklebrachial pressure indexes. In-hospital interventions were tabulated and defined as follows: (1) Medical treatment alone; (2) Endovascular intervention, consisting of balloon dilation of a stenotic native arterial or graft lesion; (3) Operative thromboembolectomy; (4 ) Revision of a previously placed bypass graft, primarily comprising patch angioplasty of a stenotic lesion, or native arterial endarterectomy with or without patch angioplasty; (5) Placement of a new bypass graft; (6) Major (below-knee or more proximal) amputation. The amount of lysed thrombus was arteriographically quantitated in the patients who underwent thrombolysis. Lysis was determined to be "arteriographically successful" if contrast studies documented the dissolution of greater than 80% of the occluding thrombus, such that the maximum percent reduction in luminal diameter from residual mural thrombus amounted to less than 20% along the entire length of the involved bypass graft or native arterial segment. Statistical analysis. Previous studies documented a 40% risk of limb loss or death in patients undergoing standard, nonthrombolytic treatment of acute peripheral ischeinia. 2,16,17 The trial's clinical review committee determined that a 25 % difference in combined limb loss and mortality rate between the two treatment groups would be clinically important and the sample size determination was made on this basis. An enrollment of 56 patients per group was necessary to achieve a 5% two-tailed probability of a type I error and a 20% probability of a type II error. Accordingly, a goal of 112 patients was set for the study. All analyses were performed according to inten-

1024

JOURNAL OF VASCULAR SURGERY June 1994

Ouriel et at.

Table 1. Baseline characteristics of patients according to treatment group Thrombolytic therapy Characteristic* Demographics Age (yr) Male:female sex Right:left side Comorbid conditions Coronary artery disease Abnormal electrocardigraphy results Previous myocardial infarction Angina pectoris Diabetes mellims Hypertension Hyperlipidemia History of smoking Previous vascular surgery Cardiac risk index l4 Class 1 Class 2 Class 3 Class 4 Clinical presentation Ankle-brachial pressure index Fibrinogen concentration (mg/dl) Duration of symptoms (hours) Pain at rest Pallor Motor loss Sensory loss Site of arterial occlusiont Upper extremity Brachial artery alone Brachial and forearm arteries Lower extremity Aortoiliac segment alone Femoropopliteal segment alone Aortoiliac and femoropopliteal Femoropopliteal and tibial All three segments

(n

= 57)

Operative therapy (n

= 57)

69 ± 1.7 29:28 27:30

71 ± 1.7 25:32 26:31

29(51%) 18 (32%) 11 (19%) 17 (30%) 33 (58%) 17 (30%) 29(51%) 24 (42%)

35(61%) 16 (28%) 8 (14%) 15 (26%) 39 (68%) 19 (33%) 30 (53%) 25 (44%)

4(7%) 28 (49%) 20 (35%) 5 (9%)

3 26 23 5

0.04 379 39 51 37 11 28

± 0.02 ± 27 ± 6.3

(89%) (65%) (19%) (49%)

(5%) (46%) (40%) (9%)

0.04 ± 0.02 359 ± 41 42 ± 6.7 50 (88%) 34 (60%) 14 (25%) 29(51%)

4 (7%) 2 (4%)

4(7%) 0(0%)

4 (7%) 21 (37%) 2 (4%) 22 (39%) 2(4%)

5 (9%) 25 (44%) 0(0%) 23 (40%) 0(0%)

*No comparison between individual baseline characteristics of the thrombolytic and operative treatment arms attained statistical significance. tNo patient experienced occlusion of the forearm arteries alone, of the tibial arteries alone, or of the aortoiliac and tibial arteries without involvement of the femoropopliteal segment.

tion to treat. Baseline characteristics were compared by the chi-square test with the Pearson statistic. Fisher's exact test for discrete variables was used when the cell sizes were small. Continuous variables were assessed with the t test to compare the mean values in the groups or the Mann-Whitney statistic when the distribution of the variables deviated significantly from normal. IS The length of time to the onset of events was estimated by the Kaplan-Meier method,19 and the significance of differences between treatment effects was tested with the Peto-Prentice statistic. 20 The Cox proportional hazards regression model was used to evaluate .the effect of base-line covariates on outcome. 21 Normally distributed data were expressed with the mean ± the standard error of the mean, and nonparametric data were expressed with the median

and interquartile (twenty-fifth to seventy-fifth percentile) range. The p values were two-sided and compared the treatment groups, unless otherwise specified. The study was approved by the Research Subjects Review Board of the University of Rochester, who revi~wed the progress of the trial on an annual basis.

RESULTS Randomization created balanced treatment groups with respect to baseline characteristics (Table I). The distribution of patients within Goldman categories of cardiac risk was similar in the two treatment groups. The etiologic diagnoses for the occlusive processes did not differ between the two groups (Table II).

JOURNAL OF VASCULAR SURGERY Volume 19, Number 6

Ouriet et at.

1025

Table II. Etiologic diagnoses of acute peripheral arterial occlusion Thrombolytic therapy

Operative therapy

Diagnosis

(n = 57)

(n = 57)

Embolic occlusion Thrombotic occlusion Native artery: Atherosclerotic stenosis Popliteal aneurysm Antiphospholipid syndrome Bypass graft Aortofemoral Infrainguinal Axillofemoral

11 (19%) 46 (81%) 13 (23%) 8 (14%) 3 (5%) 2 (4%) 33 (58%) 5 (9%) 26 (46%) 2 (4%)

13 (23%) 44 (77%) 14 (25%) 10 (17%) 4(7%)

Technical details. Thrombolytic therapy was instituted in 55 (96%) patients who were randomized to the thrombolytic treatment arm group. It was not possible to administer the thrombolytic agent in two patients because of an inability to adequately position the infusion catheter in proximity to the occlusive process; these patients underwent operative bypass procedures. Arteriographically successful thrombolysis was achieved in 40 (70%) of the 57 patients. The mean duration of thrombolytic infusion was 35.6 ± 2.1 hours. The time-to-reperfusion averaged 7.6 ± 0.9 hours in 46 patients with some degree of reflow, with a range of 1 to 35 hours. An anatomic lesion responsible for the native arterial or bypass graft occlusion was uncovered in 21 (37%) patients, and each underwent correction of the lesion by means of percutaneous balloon angioplasty (2 patients) or operation (19 patients). No anatomic lesion was found in 19 (33 %) of the patients; each was given heparin and underwent long-term warfarin anticoagulation. The 17 patients (30%) with unsuccessful thrombolysis underwent either operative revascularization (15 patients) or primary amputation (2 patients). Operative intervention was undertaken in 56 (98%) patients who were randomized to the operative treatment arm group of the study; one patient died of metabolic complications before a planned operative procedure. Revascularization was attempted in 53 patients (93%), whereas primary major amputation was the only option in three (5 % ) patients in whom a suitable outflow vessel was not localized. Thromboembolectomy was performed in 16 (28%) patients, revisions of previously placed bypass grafts were performed in 9 (16%), and new bypass grafts were placed in 27 (47%). One patient with distal thrombotic occlusion underwent percutaneous balloon angioplasty of an inflow stenosis. Early results of treatment. The magnitude of

o

30 (53%) 5 (9%) 23 (40%) 2(4%)

necessary in-hospital interventions was lower in the thrombolytic treatment arm group (Table III, p "" 0.01, Pearson chi-square analysis). In-hospital cardiopulmonary complications developed less frequently in the thrombolytic group (16% vs 49%, P "" 0.001), with an increased frequency of myocardial infarction in the operative arm (16% vs 5%, P = 0.02). Thrombolytic therapy was associated with a significant improvement in 30-day event-free survival rate when compared with operation, with amputation or death in 8 (14%) versus 17 (30%) of the patients, respectively (p = 0.04). Differences in the 30-day requirement for amputation (9% vs 14%) and the 30-day mortality rate (12% vs 18%) did not attain statistical significance. Major bleeding episodes occurred more often in patients receiving thrombolytic therapy (11% vs 2%, P "" 0.06). A hemorrhagic cerebral infarction occurred after 32 hours of thrombolytic therapy in one patient, leading to uncal herniation and death. All other bleeding episodes in the thrombolytic group occurred at the site of arterial puncture. False aneurysms developed in three patients receiving thrombolytic therapy, and each required operative correction. The plasma fibrinogen concentration decreased during thrombolytic infusion, reaching a nadir of 288 ± 22 mg/dl (24% ± 7% below initial concentration) at 24 hours (p "" 0.003). Most bleeding episodes (91 %) occur~ed when the fibrinogen concentration was greater than 150 mg/dl. The only fibrinogen concentration below 100 mg/dl occurred in the patient who had a fatal hemorrhagic cerebral infarction, with a fibrinogen nadir of 87 mg/dl at 12 hours. The risk of bleeding was not associated with the duration of thrombolytic infusion, with a mean duration of infusion of 39.7 ± 3.3 hours in the patients with bleeding complications versus 34.6 ± 2.0 hours in the patients without bleeding complications. In contrast to decreased fibrinogen

JOURNAL OF VASCULAR SURGERY June 1994

1026 Ouriel et al.

Table III. In-hospital interventions Intervention Medical treatment only Endovascular intervention Thromboembolectomy Endarterectomy/graft revision Insertion of a new bypass graft Primary major amputation

concentrations in the thrombolytic treatment group, fibrinogen increased in the operative group (p = 0.001), peaking at 473 ± 69 mg/dl 24 hours (32% ± 5% over baseline) after entrance into the study. Secondary embolism to the distal vessels of the extremity occurred in five (9%) patients treated with thrombolytic therapy. The event was heralded by the reappearance of pulses in the large vessels, coinciding with an abrupt worsening of the clinical status of the distal extremity. All secondary emboli underwent complete dissolution with repositioning of the infusion catheter and continued infusion over 3.4 ± 1.1 hours (range 0.5 to 7.4 hours). The duration of hospitalization was not significantly different between the two treatment groups. Patients in both groups had a median length of stay of 11 days, with an interquartile range of 7 to 17 days in the thrombolytic group and 6 to 14 days in the operative group. Despite the similar length of hospital stay, the cost of care was greater in patients who were randomized to the thrombolytic therapy group (median $15,672; interquartile range $12,703 to $24,519 versus $12,253; interquartile range $7077 to $25,352; P = 0.02, Mann-Whitney statistic). Clinical follow-up. The length of follow-up averaged 20.5 ± 1.8 months in the thrombolytic treatment group and 17.6 ± 1.8 months in the operative treatment group. The Kaplan-Meier estimate of the event-free survival rate at 12 months was 75% in the thrombolytic group and 52% in the operative group, yielding a combined risk oflimb loss or death of 25% and 48%, respectively (p = 0.02, Fig. 1). Thrombolytic therapy resulted in a 48% reduction in the I-year risk of amputation or death when compared with operative management. Cumulative survival rate was significantly improved in the thrombolytic group, 84% versus 58% at 12 months, reflecting a 62% reduction in the I-year risk of death (p = 0.01). In contrast to these differences, the cumulative risk of amputation was 18% at 12 months in each group. Thus the observed difference in the

Thrombolytic group (n = 57)

Operative group

19 (33%) 2(4%)

1 (2%) 1 (2%) 16 (28%) 9 (16%) 27 (47%) 3 (5%)

6(11%) 10 (18%) 18 (32%) 2(4%)

(n = 57)

event-free survival rate was attributable to a difference in survival rate rather than a difference in amputation rate. There was no difference in the number of readmissions during the follow-up period, averaging 17% ± 4% annually in the thrombolytic arm and 24% ± 5% in the operative arm. Patients who were randomized to the thrombolysis group were spared operation in 36% of the cases at 12 months of follow-up by Kaplan-Meier analysis. This value is greater than the 33% frequency obtained with the estimation of the necessity of operative intervention at the time of discharge as a result of the failure of the latter method to include the censored observations of patients who died during the initial hospitalization. Most patients in the thrombolytic group who required operation underwent correction of anatomic lesions unmasked by thrombolysis during the initial hospital stay; the need for operation plateaued at a rate of 0.33% per month after the first 30 days after presentation. There were 33 deaths in the trial. Cox proportional hazards regression identified only one baseline characteristic that was, in addition to the treatment variable, predictive of death. Patients falling within Goldman classes 3 and 4 had a greater risk of death (relative risk 1.65, 95% confidence interval 1.15 to 2.32). By contrast, the risk of death was not dependent on baseline electrocardiographic changes or the administration of warfarin. Patients who experienced an in-hospital cardiopulmonary complication seemed to fare worse than patients without complications (Table IV), suggesting that the increased prevalence of cardiac and pulmonary complications in the operative group may underlie the observed survival differences. The primary patency rate of the arterial reconstruction or thrombolysed native arterial segment was not significantly different between the two treatment groups (Fig. 2). Similarly, noninvasive laboratory hemodynamic indexes were not significantly different in the thrombolytic and operative

JOURNAL OF VASCULAR SURGERY Volwne 19, Nwnber 6

1.00

-'-

Q

c

.80

Thrombolysis

._- ··0·

~

:::.; a...

0--0'--

l·o-o.-.. ~•••• .,

.60

::l

y.-....Operation _._ ..... _.... -i

0--

CI)

r····

Q)

e La.:

1027

I

.!

~

Ourie! et at.

.40

I

..... c:::

~

.20 57 57

LIJ

0

44

39

40 34

I

0

4

35 28

31 22

25 21

22

8

12

16

20

,

17

Thrombolysis Operation

I

24

Months Fig. 1. Kaplan-Meier curves for combined outcome event of death or amputation, according to assigned treatment. Numbers of patients at risk are shown at bottom of graph; error bars represent SEM at representative follow-up intervals.

Table IV. Complications of treatment Treatment group Thrombolytic Operative

In-hospital cardiopulmonary complications WithIHCC Without IHCC WithIHCC Without IHCC

Number ofpatients

12-Month mortality rate (Kaplan-Meier)

9 (16%)

51 ± 9%

48 (84%)

8 ± 3%

28 (49%) 29 (51%)

68 ± 7% 14 ± 2%

IHCC, In-hospital cardiopulmonary complications.

treatment arms, increasing from 0.04 ± 0.03 and 0.04 ± 0.02, respectively, at the time of presentation to 0.64 ± 0.17 and 0.53 ± 0.08, respectively, by the second day, and gradually decreasing to 0.47 ± 0.11 and 0.49 ± 0.12, respectively, at the 12-month follow-up examination. Embolism versus thrombosis. The short-term (in-hospital) results for embolic and thrombotic occlusion were similar in the two treatment arms (Table V). By contrast, the I-year survival rate appeared greater when patients with embolic events were treated with thrombolysis, 100% versus 51 % at 12 months. Patients with thrombotic occlusions realized a less substantial benefit; 12-month survival rates were 80% and 60% with thrombolysis and

operation, respectively. Patency rate appeared greater in patients with embolic verSUf thrombotic events, irrespective of treatment.

DISCUSSION Intraarterial thrombolysis with urokinase in patients with acute peripheral arterial occlusion yielded improved results with respect to the frequency of in-hospital cardiopulmonary complications and patient survival. The 12-month survival rate was 84% in the thrombolytic treatment arm and 58% in the operative arm. Limb salvage, by contrast, was similar in each group, averaging 82% at 1 year. The disparity in survival seemed to be attributable to an increase in the frequency of in-hospital complications in the

1028

TOURNAL OF VASCULAR SURGERY Tune 1994

Ouriel et al.

1.00

_80

o ,

~~ \------- -- ---------t _____________~~:!~~i~~

_60

________ _

Thrombolytic

.40

.20

57 57

36

31

24 27

22 18 I

I

0

4

8

12

16

0

15 16

11 8

Thrombolysis Operation

20

24

Months Fig. 2. Kaplan-Meier curves for patency rate according to assigned treatment. Data are expressed in terms of primary patency of revascularization, defined as interval between restoration of arterial continuity by means of thrombolysis or operative reconstruction and subsequent thrombotic failure of revascularized segment. Numbers of patients at risk are shown at bottom of graph; error bars represent SEM at representative follow-up intervals.

Table V. Clinical outcome in patients with embolism versus thrombosis Embolism Thrombolysis (n In-hospital outcome Successful lysis Time-to-reperfusion (hr.) Duration of infusion (hr.) Adjuvant procedure Bleeding complication In-hospital mortality Length of hospital stay (days) Hospital costs Outcome at 1 year Event-free survival Limb salvage rate Survival rate Patency rate Spared operation

= 11)

8 (72%) 8.1 ± 1.5 39.6 ± 3.4 5 (45%) 4 (36%)

o

9.6 ± 1.5 $15208 ± $3437 100% 100% 100% 92% ± 7% 55% ± 15%

Thrombosis

Operation (n

= 13)

1 (8%) 3 (23%) 11.5 ± 3.9 $16326 ± $4686 51% ± 15% 100% 51% ± 15% 89% ± 10%

operative group, complications that were primarily cardiopulmonary and appeared to function as determinants of subsequent death. Our data confirm the findings of previous uncontrolled studies of acute limb ischemia. Historically,

Thrombolysis (n

= 46)

32 (69%) 7.5 ± 1.1 35.7 ± 2.3 33 (71%) 7 (15%) 7 (15%) 15.3 ± 2.5 $23569 ± $3283 69% 78% 80% 58% 31%

± 7%

± 7% ± 6% ± 8% ± 7%

Operation (n

= 44)

4 (9%) 7 (16%) 15.1 ± 2.4 $21125 ± $2943

52% 77% 60% 51%

± 8%

± 7% ± 8% ± 8%

acute peripheral arterial occlusion has been treated with urgent operative revascularization. 2,16 Despite improvements in surgical technique and postoperative care, patient mortality and limb loss rates remained at distressingly high levels. 22 Blaisdell et al. 2

JOURNAL OF VASCULAR SURGERY Volume 19, Number 6

summarized the results of 35 surgical series of acute limb ischemia and documented perioperative mortality rates in excess of 25% and amputation rates of more than 30% in the survivors. A regimen of high-dose heparin anticoagulation and delayed operative revascularization reduced the mortality rate without an associated increase in the need for amputation. Although this approach has not been widely adopted and the data have been criticized on the basis of outmoded patient care,23 a subsequent review by Jivegaard and associates 24 corroborated the lethality of acute peripheral ischemic events, documenting a perioperative mortality rate of 18% and an amputation rate of 16%. The sequelae of acute limb ischemia were well documented by Yeager et al. 25 when they reported a I-year mortality rate of more than 35% in patients treated with operation. Corresponding follow-up data do not exist for patients treated with thrombolytic modalities, although inhospital mortality rates characteristically fall within the range of 5% to 10%.8,12,26 Thrombolytic therapy offered potential benefits compared with either anticoagulation or operative intervention, providing the opportunity to achieve limb salvage rates as high as those associated with operative revascularization without the attendant risks of a highly invasive intervention. 4,8-10,12,27,28 These theoretic advantages of thrombolytic therapy prompted an almost universal acceptance of the treatment modality without the formality of prospective, comparative studies. 5-7,9,23,29 Thus it has not been possible to gauge objectively the actual benefit from thrombolytic intervention relative to operative revascularization. This study demonstrates advantages of intraarterial thrombolytic therapy over operation in the initial treatment of acute peripheral arterial occlusion. Recognizing that many acute occlusions develop in the presence of preexistent native arterial or bypass graft lesions,9,12 we hypothesized that thrombolytic therapy might unmask the anatomic lesion responsible for the occlusive event, providing the opportunity to address the causative lesion with a directed operative or endovascular approach. In addition, the need for an invasive intervention might be circumvented in the subset of patients without an underlying anatomic lesion. In this regard, thrombolytic therapy obviated the need for operative intervention in 36% of patients over a 12-month period of follow-up. In patients with anatomic lesions responsible for the occlusive process, thrombolysis provided the opportunity to perform limited procedures directed at the correction of the focal defect.

Ouriei et ai.

lO29

Our data substantiate the widespread belief that most emboli are successfully treated with operation, with limb salvage and patency in virtually every survivor. These results, however, must be considered in light of the in-hospital mortality rate of 23% and the I-year survival rate of 51 %. By contrast, thrombolysis for embolic events achieved limb salvage and patency rates equivalent to those of operation, with a survival rate of 100% at 1 year. Thus thrombolytic treatment of embolic arterial occlusions may achieve satisfactory revascularization without the high mortality rate associated with operation. Thrombolytic therapy resulted in hemorrhagic complications in one fifth of the patients, one half of which were major. There was one case of lethal intracranial hemorrhage. These data are consistent with previous reports documenting major bleeding complications in 4% to 10%26,30-33 and intracranial bleeding in 0.5%34 of patients. The dose and choice of plasminogen activator, duration of infusion, and decrease in fibrinogen concentration have all been suggested as factors influencing the risk of bleeding,9,35,36 but other studies have failed to confirm an association between the development of hemorrhage and either the fall in fibrinogen or the duration of therapy.37-39 In summary, we found a decreased mortality rate after acute peripheral arterial occlusion initially treated with a locally administered thrombolytic agent. This observation may reflect the decreased magnitude of required interventions in the thrombolytic group, with a corresponding decrease in the rate of cardiopulmonary complications. These data support the use of intraarterial urokinase as the initial intervention in patients with acute peripheral arterial occlusion, followed by aggressive management of lesions unmasked by thrombolysis. Recommendations on the use of thrombolysis in subacute and chronic occlusions must await the results of subsequent investigations. REFERENCES 1. Rutkow 1M, Ernst <;:B. An analysis of vascular surgical manpower requirements and vascular surgical rates in the United States. J VASe SURG 1986;3:74-83. 2. Blaisdell FW, Steele M, Allen RE. Management of acute lower extremity arterial ischemia due to embolism and thrombosis. Surgery 1978;84:822-34. 3. Koltun WA, Gardiner GA, Jr., Harrington DP, Couch NP, Mannick JA, Whittemore AD. Thrombolysis in the treatment of peripheral arterial vascular occlusions. Arch Surg 1987; 122:901-5. 4. Dotter CT, Rosch J, Seaman AJ. Selective clot lysis with low-dose streptokinase. Radiology 1974;111:31-7. 5. Hess H, Ingrisch H, Mietaschk A, Rath H. Local low-dose

1030 Guriel et at.

thrombolytic therapy of peripheral arterial occlusions. N Engl J Med 1982;307:1627-30. 6. Slany J, Enzenhofer V, Karnik R. Local thrombolysis in arterial occlusive disease. Angiology 1984;35:231-7. 7. Graor ItA, Risius B, Young JR, et al. Low-dose streptokinase for selective thrombolysis: systemic effects and complications. Radiology 1984;152:35-9. 8. McNamara TO, Bomberger ItA, Merchant RF. Intra-arterial urokinase as the initial therapy for acutely ischemic lower limbs. Circulation 1991;83:Il06-Il19. 9. Sicard GA, Schier TJ, Totty WG, et al. Thrombolytic therapy for acute arterial occlusion. J VASC SURG 1985;2:65-78. 10. Comerota AJ. Intra-arterial thrombolytic therapy. In: Comerota AJ, ed. Thrombolytic Therapy. Orlando: Grune and Stratton, 1988: 125-52. 11. Persson AV, Robichaux WT, Jaxheimer EC, DiPronio E. Burst therapy: a method of administering fibrinolytic agents. Am J Surg 1984;147:531-6. 12. Gardiner GA, Jr., Sullivan KL. Catheter-directed thrombolysis for the failed lower extremity bypass graft. Semin Vasc Surg 1992;5:99-103. 13. Rutherford RB, Flanigan DP, Gupta SK, et al. Suggested standards for reports dealing with lower extremity ischemia. J VASC SURG 1986;4:80-94. 14. Goldman L, Caldera DL, Nussbaum SR, et al. Multifactorial index of cardiac risk in noncardiac surgical procedures. N Engl J Med 1977;297:845-50. 15. Clauss A. Gerinnungsphysiologische schneliomethode zur bestimmung des fibrinogens. Acta HaematoI1957;17:237-41. 16. Green RM, De Weese JA, Rob CG. Arterial embolectomy before and after the Fogarty catheter. Surgery 1975;77: 24-33. 17. Ricotta TJ, Green RM, De Weese JA. Use and limitations of thrombolytic therapy in the treatment of peripheral arterial ischemia: results of a multi-institutional questionnaire. J VASC SURG 1987;6:45-50. 18. Dixon WJ, Massey F1, Jr. Introduction to Statistical Analysis. 4th ed. New York: McGraw-Hill, 1983:394-6. 19. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457-81. 20. Peto R, Pike MC, Armitage P. Design and analysis of randomized clinical trials requiring prolonged observation of each patient: II-analysis and examples. Br J Cancer 1977; 35:1-39. 21. Cox DR. Regression models and life tables. J R Stat Soc 1972;34:187-220. 22. Abbott WM, Maloney RD, McCabe CC, et al. Arterial embolism: a 44 year perspective. Am J Surg 1982; 143 :460-4. 23. McNamara TO. Thrombolysis as an alternative initial therapy for the acutely ischemic lower limb. Semin Vasc Surg 1992;5:89-98. 24. Jiveg:iard L, Holm 1, Shersten T. The outcome in arterial thrombosis misdiagnosed as arterial embolism. Acta Chir Scand 1986;152:251-6.

JOURNAL OF VASCULAR SURGERY June 1994

25. Yeager ItA, Moneta GL, Taylor LM Jr, Hamre DW, McConnell DB, Porter JM. Surgical management of severe acute lower extremity ischemia. J VASC SURG 1992;15:38593. 26. Meyerovitz MF, Goldhaber SZ, Regan K, et al. Recombinant tissue-type plasminogen activator versus urokinase in peripheral arterial and graft occlusions: a randomized trial. Radiology 1990;175:75-8. 27. McNicol GP, Douglas AS. Treatment of peripheral vascular occlusion by streptokinase perfusion. Scand J Clin Lab Invest 1964; 16(Suppl 78):23-9. 28. Fletcher AP, Alkjaersig N, Sherry S. Maintenance of a sustained fibrinolytic state in man: induction and affects. J Clin Invest 1959;38:1096-10. 29. Krings W, Roth FJ, Cappius G, Schmidtke I. Catheter-lysis: indications and primary results. International Angiology 1985;4:117-23. 30. Graor ItA, Olin 1, Bartholomew JR, Ruschhaupt WF, Young JR. Efficacy and safety of intraarterial local infusion of streptokinase, urokinase, or tissue plasminogen activator for peripheral arterial occlusion: A retrospective review. J Vasc Med BioI 1990;2:310-15. 31. McNamara TO, Fischer JR. Thrombolysis of peripheral arterial and graft occlusions: improved results using high dose urokinase. Am J RoentgenoI1985;144:769-75. 32. Durham JD, Gellar SC, Abbott WM, et al. Regional infusion of urokinase into occluded lower-extremity bypass grafts: long-term clinical results. Radiology 1989;172:83-90. 33. Seabrook GR, Mewissen MW, Schmitt DD, et al. Percutaneous intraarterial thrombolysis in the treatment of thrombosis oflower extremity arterial reconstructions. J VASC SURG 1991;13:646-51. 34. Gardiner GA Jr, Sullivan KL. Complications of regional 'thrombolytic therapy. In: Kadir S, ed. Current practice of interventional radiology. Philadelphia: BC Decker, 1991:8791. 35. Berni GA, Bandyk DF, Zierler RE, Thiele BL, Strandness DE Jr. Streptokinase treatment of acute arterial occlusion. Ann Surg 1983;198:185-91. 36. Graor ItA, Risius B, Lucas FV, et al. Thrombolysis with recombinant human tissue-type plasminogen activator in patients with peripheral artery and bypass graft occlusions. Circulation 1986;74:Il5-I20. 37. Marder VJ. The use of thrombolytic agents: choice of patient, drug administration, laboratory monitoring. Ann Intern Med 1979;90:802-12. 38. Marder VJ. Bleeding complications of thrombolytic treatment. Am J Hosp Pharm 1990;47:S15-S19. 39. Marder VJ, Sherry S. Thrombolytic therapy: current status (2). N Engl J Med 1988;318:1585-95.

Submitted Aug. 4, 1993; accepted Oct. 18, 1993.