Delayed Thromboembolectomy for Subacute Limb Ischemia Jonathan P. Gertler,
MD,
Dale A. Distant,
Arterial embolixation frequently requires immediate operative intervention. Occasionally, embolic events produce subacute limb ischemia that may not be recognized as thromboembolic in origin. In October 1988, a prospective policy to attempt thromboembolectomy rather than infrainguinal hypass in all patients with delayed presentation of lower limb thromboembolism was initiated at our institution. Seven limbs in five patients were identified by history, physical examination, noninvasive study, and/or angiography as ischemic due to thromboembolism, which occurred from 3 to 10 weeks prior to presentation. Six of seven limbs were studied angiographically, and all seven were treated by femoral and/or popliteal thromboembolectomy with limb salvage. Six of seven limbs were restored to normal arterial hemodynamics as assessed by intra- and postoperative noninvasive study. One limb in a patient continued to have minimal residual occlusive disease that was recognized preoperatively. In all cases, chronic and fresh thromboemboli were found at surgery and confirmed by pathology. Chronic thromboembolism seems to be an underrecognized event. Limb salvage can be achieved readily even if extensive delay in diagnosis is present. Using preoperative angiography and intraoperative noninvasive techniques to measure the success of revascularixation, as well as avoiding, when possible, complex infrainguinal reconstruction in these high-risk patients allows for an aggressive approach to limb salvage with gratifying results.
MD,
George Varughese,
MD, Brooklyn, NW
York
ower limb embolization frequently results in cataL strophic ischemia requiring urgent intervention. The conventional wisdom is that thromboembolectomy must be carried out soon after the clinical event has occurred in order to maximize retrieval of occluding thromboembolic material and limb salvage. In addition, some clinical studies suggest that early propagation of microthrombosis may lead to difficulty in retrieving thrombi, which further supports timely intervention in these patients [I]. Several reports exist [2-91, however, that describe late retrieval of embolic material and the resulting limb salvage. As the patients with thromboembolic disease shift from a rheumatic valve population, as seen earlier in this era, to an older population more prone to atherosclerotic heart and vascular disease, the clinical and diagnostic criteria used to differentiate emboli from thrombotic events may become blurred. In an aging population, coexistent peripheral occlusive disease may make single embolic events less critical due to previously established collaterals, thus making chronic or subacute ischemia the presenting symptom. Finally, a delay in seeking medical attention by an individual afflicted with subcritical embolic occlusion of a limb may cause misinterpretation of ischemic symptoms as occlusive rather than embolic in origin. Our study suggests that embolization presenting as chronic or subacute limb ischemia may be an underrecognized phenomenon, and its treatment can differ markedly from similarly severe limb ischemia based on atherosclerotic disease. Seven limbs in five patients presenting with chronic and subacute ischemia secondary to embolization were restored to normal perfusion by thromboembolectomy up to 10 weeks following the first clinically recognizable symptoms. We also review recent reports of delayed thrombcembolectomy to help estimate the efficacy of this approach. PATIENTS AND METHODS
From the Division of Vascular Surgery, State University of New York Health Science Center at Brooklyn, Brooklyn, New York. Requests for reprints should-be add&&d to Jonathan P. Gertler, MD, Division of Vascular Surgery. Box 40-DeDartment of Surnerv. State University Hospital Health &ience Center, 450 Clarkson xv& nue, Brooklyn, New York 11203. Manuscript submitted June 19,1990, and accepted in revised form October 11,199O.
p68
THE AMERICAN JOURNAL OF SURGERY
All clinical material was gathered prospectively from the Vascular Service at the State University of New York Health Science Center, Brooklyn, New York, from Gcto ber 1988 to July 1989. An aggressive policy regarding thromboembolectomy was established in July 1988 for the treatment of those patients in whom subacute ischemia was thought to be due to embolization, with or without propagation of thrombosis, based on clinical history, physical examination, noninvasive study, and angiography. Eight patients in this category were identified: one required urgent surgery; four agreed to evaluation and subsequent surgery; and the remaining three did not have limb-threatening ischemia and did not require revascularization during the period of follow-up. The data in this report are from seven limbs in five patients.
VOLUME 162 NOVEMBER 1991
Left
..-.
-_-A__----.
1 ,_------_i
183mmkjg
I _.-____
I
/
r__-_._4
184mmHQ
____
j
PVRy2
-__:
PVR=I
m _______’
ANKLE __i_______~___p..i&(
‘-----y--~
I----97mm ttd I
pI_q.&q ________~_______i_______1_ 1
_,-_____+_-____I_
I----- 1
t
Prospective data included a careful history of onset, duration and progression of symptoms, age, gender, risk factors for atherosclerosis (smoking, hypertension, diabetes mellitus, hyperlipidemia), electrocardiogram (ECG) on admission, and a thorough past medical history. All patients underwent a thorough physical examination and prospective segmental noninvasive studies with pulse volume recordings (PVR) pre-, intra-, and postoperatively. Angiography was performed in all nonemergency cases preoperatively. RESULTS Eight patients with suspected embolixation were identified based on history, examination, and noninvasive study. Three patients had claudication with multiple medical problems and deferred angiography. All were managed without anticoagulation treatment and have maintained limb viability without recurrent emboli. Three women and two men, ranging in age from 48 to 84 years, agreed to surgical care as outlined. Four of five patients (six of seven limbs) were initially associated with claudication, which progressed over time to rest pain in all cases. One patient had acute onset of rest pain as the first symptom. Tie to presentation for the patients with initial claudication and then rest ischemia ranged from 3 to 10 weeks. In the one patient experiencing rest pain at the onset of the process, progression to tissue loss prompted his seeking medical attention 4 weeks after the onset of symptoms.
1 TRANSMET
II
No patients were diabetic. Two of live patients were smokers and hypertensive. One patient had a history of smoking, hypertension, and hypercholesterolemia. Three patients had either atria1 ectopy (one patient) or atria1 fibrillation (two patients) on ECG. One patient had undergone aortic valve replacement and had had four subsequent middle cerebral artery emboli. Of note in this patient is that the onset of severe claudication progressing to rest pain coincided with her last middle cerebral artery event. Physical examination in all cases was unrevealing except for the cardiac rhythm abnormalities noted on ECG and for the fmdings of pulselessness and ischemic changes as expected in the affected liibs. In the three patients presenting with unilateral limb involvement, the contralateral limb was completely normal. In all patients, noninvasive studies were suggestive of poorly compensated superficial femoral artery occlusion with type IV to V PVR present at the transmetatarsal level. The presence of strong femoral pulses with reduced high thigh PVR in one patient suggested occlusion of both the superficial and deep femoral arteries. Postoperative PVR always revealed restoration of normal hemodynamics (FIgurea 1 and 2). Preoperative angiography was used in all but one emergency case. Segmental occlusion, clot meniscus, and poor collateralization diagnostic of emboli were present in all patients. Surgery was performed under local anesthetic in three
THE AMERICAN JOURNAL OF SURGERY
VOLUME 162 NOVEMBER 1991
469
TABLE I Summary of Individual Reports of Delayed Thromboembolectomy Therapy Age Range Reference
PI
M/F(%)
PI
71 I29 2517.5 75125
[51 191
75125
171 El
77123 *
I41
PI
l
l
(yrs) 35-71 23-79 48-67 43-92 38-70
Claudication 5 1 0 * *
Rest Pain
Anticoagulation
2 3 4 * *
*
l
l
45-88
l
l
l
*
*
Embolectomy
Direct Retrieval
7 4 2 l
1
9
7 7 5
11
Delay to Therapy
Outcome Salvage 7 4 4 19 1 11 4 4
Amputation
3 9 7 3 1
NW 2-32 1-12 l-4 1-4 3-8 >l 1-9 >I
*Notspecified.
Eight articles specifically addressing the issue of delayed thromboembolectomy were identified in a search of the recent English language medical literature [2-91. A total of 77 cases in which thromboembolectomy was performed at an interval of at least 1 week from the onset of clinical symptoms were identified from these reports. Though not all information could be obtained from all papers, a concerted effort to determine presenting symp toms, therapy rendered, outcome, demographic data, and length of delay from symptom onset to surgery was made. The results are listed in Table I. Indications for surgery, when specified, were closely divided between rest ischemia and claudication. Most surgeons utilized balloon embolectomy though at least one author stressed the importance of direct arterial exploration for older thrombo embolic material [4]. Limb salvage was achieved in 54 of 77 cases with five deaths among the patients.
of five patients (four limbs), epidural anesthetic in one of five patients (two limbs), and general anesthetic in one patient (one limb). Popliteal exploration was necessary as assessed by intraoperative PVR in two patients (three limbs) and was performed under general or epidural anesthesia. In one patient, presentation with congestive heart failure initially mandated surgery under local anesthesia. Thrombus resulted from this approach, and intraarterial urokinase (50,000 units) was then introduced and allowed to dwell for 30 minutes. Postoperatively, the patient’s limb was still ischemic, albeit improved, and, after cardiac stabilization, he was returned to the operating room for bilateral popliteal explorations under epidural anesthesia. Pre or postoperative echocardiography revealed ab normalities in all cases. Thrombus was never identified in the cardiac chambers; however, chronic atria1 fibrillation with reduced ejection fraction was diagnosed in two patients. Left ventricular hypertrophy in one patient, absent prosthetic aortic valve vegetations but poor ventricular function in one patient, and an atrioseptal defect with right to left shunt in one patient were also diagnosed. Only one patient had a history suggesting chronic mild peripheral vascular occlusive disease. All patients except this one had normal arteries that were identified at surgery with subsequent restoration of pedal pulses. In the patient with chronic mild claudication (who had had four previous strokes, a prosthetic aortic valve, and still represented a significant operative risk), surgery revealed an embolus lodged in an area of superficial femoral artery plaque. Though pedal pulses were not restored, the patient returned to her baseline state of compensated occlusive disease postoperatively. One patient underwent a four-compartment fasciotomy after popliteal embolectomy, and reperfusion led to immediate edema of the lower leg. The patient presenting with tissue loss underwent successful closed live-toe transmetatarsal amputation. All patients had anticoagulation therapy with heparin and Coumadin (Du Pont Pharmaceuticals, Wilmington, Del.) postoperatively. There have been no recurrences of limb ischemia during follow-up ranging from 11 to 20 months. Patients were evaluated with serial exam and PVR to document the level of perfusion.
The concept of delayed embolectomy is not a new one. Haimovici in 1959 [ZO]and Spencer in 1964 [II] reported good results with patients operated on for limb emboli several weeks after the onset of clinically significant events. The introduction of the balloon embolectomy catheter greatly simplified the operative approach; however, in at least one large study comparing current results with historical reviews [IZ], a significant reduction in mortality and increase in limb salvage was not noted. The advancing age and more complex presentations of vascular patients in the modern era likely contribute to this morbidity despite recent advances in perioperative care. The primary etiology in our small series, as well as in other recent reports, remains cardiac. Most patients will demonstrate an atria1 or ventricular abnormality in rhythm or anatomy. Certainly, other sources, including proximal aneurysm or atherosclerosis, iatrogenic catheter injury, prosthetic graft, and paradoxical embolus, may be considered. However, it appears that, though the population is mostly elderly with multisystem disease, most will not have associated peripheral occlusive disease. In appropriate cases, chronic limb ischemia can be treated with embolectomy alone with good success in an elderly and high-risk population, further supporting the aggres-
470
162
THE AMERICAN
JOURNAL
OF SURGERY
VOLUME
COMMENTS
NOVEMBER
1991
DELAYEDTHROMBOEMBOLECTOMY
sive approach to limb salvage espoused in recent vascular literature as well as creating a new avenue of therapy . Clinical presentation and noninvasive findings provide several clues to the diagnosis. Our patients experienced an acute onset of mild vascular symptoms that rapidly progressed, and all but one had known cardiac pathology at the time of admission. Noninvasive evaluation in several cases revealed a sudden, severe diminution of perfusion at the calf level, which implied that there was a poorly collateralized superficial femoral artery occlusion. As most patients with infrainguinal rest ischemia have more than one level of disease, these findings suggested an acute occlusion of a previously normal or insignificantly narrowed vessel or, alternatively, occlusion of both the superficial femoral artery and profunda femoris artery. Finally, all of our patients had echocardiograms that revealed either dilated or abnormal chambers or intracardiac defects. As imaging techniques for intracardiac thrombi are suboptimal at this time, fmclmg chamber abnormalities in conjunction with ischemic limbs raises the Spector of distal embolization. Even in the patient with recent onset of claudication, early diagnosis of embolism not only will facilitate limb surgery but will identify the patient at risk for subsequent, more dangerous, embolization in whom anticoagulation therapy is indicated. Diagnosis in the patient with clear indications for limb salvage surgery (i.e., rest pain or tissue loss) will be readily made by angiography. In the patient with less obvious symptoms or with claudication alone, a history of arrhythmia, cardiac chamber or valve abnormality, acute onset and asymmetric examination, or noninvasive lindings as described above that are suggestive of emboli will lead us to angiography in patients in whom normally a less-aggressive approach would be taken. The technique for thromboembolectomy in the delayed setting does not differ markedly from the routine approach. Some authors have advocated direct approaches to those emboli that are refractory to balloon removal. This was necessary in two patients; in one of these patients, the groin approach inadequately restored perfusion. For one patient, in whom popliteal embolization was suspected, we would have preferred direct exploration with access to the popliteal branches at the outset. The patient’s presenting condition in florid congestive failure with a suspected myocardial infarction precluded this, and we used a temporizing measure to allow the patient’s condition to stabilize, which extended the chance for performing limb salvage. In the setting of suspected involvement of the popliteal and its branches with thromboembolic debris, we advocate direct popliteal exploration, longitudinal arteriotomy after inspection of the vessel, direct passage of catheters, and inspection of the material returned. Although we have not routinely utilized intraoperative angiography, this approach would allow excellent angiographic visualization of the tibia1 vessels and would also facilitate the use of the popliteal artery for a bypass should the need arise. In the patient for whom revascularization via an extensive procedure is considered excessive, return of normal hemodynamics by intraopera-
THE AMERICAN
FORISCHEMIA
tive PVR at the transmetatarsal level may be sufficiently reassuring to obviate the need for intraoperative angiography. Patch graft closure to prevent popliteal artery stenosis is mandatory in this setting [ 131. Pop&al exploration may also help to diagnose popliteal aneurysm, which, when thrombosed, may mimic a popliteal embolism. The use of intraoperative lytic therapy was recently advanced by Parent et al [I#] and Quinones-Baldrich et al [IS]. A recent series of experiments by QuinonesBaldrich et al [Id] demonstrated that the no reflow phenomenon in an ischemic limb may be averted by using reperfttsate containing lytic agents and may avert a poor outcome due to microcirculatory thrombi despite the reperfusion of major vessels. We support this concept and found such an approach beneficial as a temporizing and adjunctive measure. Although no specific recommendations can be made concerning intraoperative lytic therapy, we have used a protocol similar to Quinones-Baldrich et al [ 151. Fifty thousand units of urokinase is dissolved in 100 mL heparinized normal saline introduced by catheter into the artery and allowed to dwell for 30 minutes before flow is reestablished. Postoperative evaluation of patients with limb emboli should include cardiac and aortic ultrasound. A search for an arterial based lesion may be prudent to prevent further episodes should the cardiac work-up be unrevealing, especially if no preoperative angiogram was obtained, if the pathologic examination of the embolic debris reveals atherosclerotic elements, and if the patient is a suitable candidate for further surgery [ 171. The newer noninvasive imaging modalities including deep duplex and magnetic resonance imaging may facilitate this search without extensive angiography [ 181.
CONCLUSION Our study considers seven limbs in live patients in which thromboembolectomy was successfully performed, at an interval up to 10 weeks following the onset of symp toms. This approach achieved limb salvage in all cases without the need for more extensive surgery. One patient had associated occlusive disease that rendered the outcome imperfect; however, this patient’s overall condition precluded any regional or general anesthetic. A review of recent and past literature supports the concept of delayed embolectomy with the authors reporting high rates of limb salvage and low morbidity. The patients presented in this series are patients with advanced cardiac disease: most would have been denied extensive revascularization procedures in a nonemergency setting, but all handled limb salvage with relative ease. Maintaining a high index of suspicion for thromboembolic events will help identify those patients in whom lifelong anticoagulation is necessary and who will benefit from less complex though effective arterial reconstructions. This small series of patients makes a number of useful diagnostic and therapeutic points, and these are exactly the kind of patients in whom multiple therapies are best used to achieve long-term limb salvage.
JOURNAL
OF SURGERY
VOLUME
162
NOVEMBER
1991
471
REFERENCES 1. Blaisdell FW, Steele M, Allen RF. Management of acute lower extremity &hernia due to embolism and thrombosis. Surgery 1978; 84: 822-34. 2. Morris WT. Late embolectomy. Br Med J 1972; 3: 631-3. 3. Milliken JC. Delayed arterial embolectomy. J Irish Med Assoc 1968; 61: 390-2. 4. Naqvi MA, Mackenzie DH, Allen LS, et al. Delayed arterial embolectomy. Can J Surg 1974; 17: 335-9. 5. Jarrett F, Dacumos GC, Crumeny AB, et al. Late appearance of arterial emboli: diagnosis and management. Surgery 1979; 86: 898-905. 6. I.&n BH, Giordano JM. Delayed arterial embolectomy. Surg Gynecol Obstet 1982; 155: 549-51. 7. Hammarsten J, Holm J, Schersten J. Positive and negative effects of anticoagulant treatment during and after arterial embolectomy. J Cardiovasc Surg 1978; 19: 373-9. 8. Ammann J, Seiler H, Vogt B. Delayed arterial embolectomy: a plea for a more active surgical approach. Br J Surg 1976; 63: 73-6. 9. Macgowan WAL, Mooneeram R. A review of 174 patients with arterial embolism. Br J Surg 1973; 60: 894-8. 10. Haimovici H. Late arterial embolectomy. Surgery 1959; 46: 775-86.
472
THE AMERICAN JOURNAL OF SURGERY
11. Spencer FM, Eiseman B. Delayed arterial embolectomy-a new concept. Surgery 1964; 55: 64-71. 12. Abbott WM, Maloney RD, McCarr CC, et al. Arterial embolism: a 44 year perspective. Am J Surg 1983; 143: 460-4. 13. Gupta SK, Samson RH, Veith FJ. Embolectomy of the distal part of the popliteal artery. Surg Gynecol Obstet 1981; 153: 255-7. 14. Parent FN, Bernhard VM, Pabst TS, MC Intyre KE, Hunter GC, Malone JM. Fibrinolytic treatment of residual thrombus after catheter embolectomy for severe lower limb ischemia. J Vast Surg 1989; 9: 153-60. 15. Quinones-Baldrich WJ, Baker JD, Busuttil RW. Intraoperative infusion of lytic drugs for thrombotic complications of revascularization. J Vast Surg 1989; 10: 408-17. 16. Quinones-Baldrich WJ, Chervu A, Moore WS. Skeletal muscle function after ischemia: “No reflow” vs reperfusion injury [abstract]. Presented at the Association for Academic Surgery, Louisville, Kentucky, 1989. 17. Williams GM, Harrington D, Burdick J, White RI. Mural thrombus of the aorta. Ann Surg 1981; 194: 737-44. 18. Langsfeld M, Neptune J, Hershey FB. The use of deep duplex scanning to predict hemodynamically significant aortoiliac stenosis. J Vast Surg 1988; 7: 359-9.
VOLUME 162 NOVEMBER 1991
MD,
Dale A. Distant,
Arterial embolixation frequently requires immediate operative intervention. Occasionally, embolic events produce subacute limb ischemia that may not be recognized as thromboembolic in origin. In October 1988, a prospective policy to attempt thromboembolectomy rather than infrainguinal hypass in all patients with delayed presentation of lower limb thromboembolism was initiated at our institution. Seven limbs in five patients were identified by history, physical examination, noninvasive study, and/or angiography as ischemic due to thromboembolism, which occurred from 3 to 10 weeks prior to presentation. Six of seven limbs were studied angiographically, and all seven were treated by femoral and/or popliteal thromboembolectomy with limb salvage. Six of seven limbs were restored to normal arterial hemodynamics as assessed by intra- and postoperative noninvasive study. One limb in a patient continued to have minimal residual occlusive disease that was recognized preoperatively. In all cases, chronic and fresh thromboemboli were found at surgery and confirmed by pathology. Chronic thromboembolism seems to be an underrecognized event. Limb salvage can be achieved readily even if extensive delay in diagnosis is present. Using preoperative angiography and intraoperative noninvasive techniques to measure the success of revascularixation, as well as avoiding, when possible, complex infrainguinal reconstruction in these high-risk patients allows for an aggressive approach to limb salvage with gratifying results.
MD,
George Varughese,
MD, Brooklyn, NW
York
ower limb embolization frequently results in cataL strophic ischemia requiring urgent intervention. The conventional wisdom is that thromboembolectomy must be carried out soon after the clinical event has occurred in order to maximize retrieval of occluding thromboembolic material and limb salvage. In addition, some clinical studies suggest that early propagation of microthrombosis may lead to difficulty in retrieving thrombi, which further supports timely intervention in these patients [I]. Several reports exist [2-91, however, that describe late retrieval of embolic material and the resulting limb salvage. As the patients with thromboembolic disease shift from a rheumatic valve population, as seen earlier in this era, to an older population more prone to atherosclerotic heart and vascular disease, the clinical and diagnostic criteria used to differentiate emboli from thrombotic events may become blurred. In an aging population, coexistent peripheral occlusive disease may make single embolic events less critical due to previously established collaterals, thus making chronic or subacute ischemia the presenting symptom. Finally, a delay in seeking medical attention by an individual afflicted with subcritical embolic occlusion of a limb may cause misinterpretation of ischemic symptoms as occlusive rather than embolic in origin. Our study suggests that embolization presenting as chronic or subacute limb ischemia may be an underrecognized phenomenon, and its treatment can differ markedly from similarly severe limb ischemia based on atherosclerotic disease. Seven limbs in five patients presenting with chronic and subacute ischemia secondary to embolization were restored to normal perfusion by thromboembolectomy up to 10 weeks following the first clinically recognizable symptoms. We also review recent reports of delayed thrombcembolectomy to help estimate the efficacy of this approach. PATIENTS AND METHODS
From the Division of Vascular Surgery, State University of New York Health Science Center at Brooklyn, Brooklyn, New York. Requests for reprints should-be add&&d to Jonathan P. Gertler, MD, Division of Vascular Surgery. Box 40-DeDartment of Surnerv. State University Hospital Health &ience Center, 450 Clarkson xv& nue, Brooklyn, New York 11203. Manuscript submitted June 19,1990, and accepted in revised form October 11,199O.
p68
THE AMERICAN JOURNAL OF SURGERY
All clinical material was gathered prospectively from the Vascular Service at the State University of New York Health Science Center, Brooklyn, New York, from Gcto ber 1988 to July 1989. An aggressive policy regarding thromboembolectomy was established in July 1988 for the treatment of those patients in whom subacute ischemia was thought to be due to embolization, with or without propagation of thrombosis, based on clinical history, physical examination, noninvasive study, and angiography. Eight patients in this category were identified: one required urgent surgery; four agreed to evaluation and subsequent surgery; and the remaining three did not have limb-threatening ischemia and did not require revascularization during the period of follow-up. The data in this report are from seven limbs in five patients.
VOLUME 162 NOVEMBER 1991
Left
..-.
-_-A__----.
1 ,_------_i
183mmkjg
I _.-____
I
/
r__-_._4
184mmHQ
____
j
PVRy2
-__:
PVR=I
m _______’
ANKLE __i_______~___p..i&(
‘-----y--~
I----97mm ttd I
pI_q.&q ________~_______i_______1_ 1
_,-_____+_-____I_
I----- 1
t
Prospective data included a careful history of onset, duration and progression of symptoms, age, gender, risk factors for atherosclerosis (smoking, hypertension, diabetes mellitus, hyperlipidemia), electrocardiogram (ECG) on admission, and a thorough past medical history. All patients underwent a thorough physical examination and prospective segmental noninvasive studies with pulse volume recordings (PVR) pre-, intra-, and postoperatively. Angiography was performed in all nonemergency cases preoperatively. RESULTS Eight patients with suspected embolixation were identified based on history, examination, and noninvasive study. Three patients had claudication with multiple medical problems and deferred angiography. All were managed without anticoagulation treatment and have maintained limb viability without recurrent emboli. Three women and two men, ranging in age from 48 to 84 years, agreed to surgical care as outlined. Four of five patients (six of seven limbs) were initially associated with claudication, which progressed over time to rest pain in all cases. One patient had acute onset of rest pain as the first symptom. Tie to presentation for the patients with initial claudication and then rest ischemia ranged from 3 to 10 weeks. In the one patient experiencing rest pain at the onset of the process, progression to tissue loss prompted his seeking medical attention 4 weeks after the onset of symptoms.
1 TRANSMET
II
No patients were diabetic. Two of live patients were smokers and hypertensive. One patient had a history of smoking, hypertension, and hypercholesterolemia. Three patients had either atria1 ectopy (one patient) or atria1 fibrillation (two patients) on ECG. One patient had undergone aortic valve replacement and had had four subsequent middle cerebral artery emboli. Of note in this patient is that the onset of severe claudication progressing to rest pain coincided with her last middle cerebral artery event. Physical examination in all cases was unrevealing except for the cardiac rhythm abnormalities noted on ECG and for the fmdings of pulselessness and ischemic changes as expected in the affected liibs. In the three patients presenting with unilateral limb involvement, the contralateral limb was completely normal. In all patients, noninvasive studies were suggestive of poorly compensated superficial femoral artery occlusion with type IV to V PVR present at the transmetatarsal level. The presence of strong femoral pulses with reduced high thigh PVR in one patient suggested occlusion of both the superficial and deep femoral arteries. Postoperative PVR always revealed restoration of normal hemodynamics (FIgurea 1 and 2). Preoperative angiography was used in all but one emergency case. Segmental occlusion, clot meniscus, and poor collateralization diagnostic of emboli were present in all patients. Surgery was performed under local anesthetic in three
THE AMERICAN JOURNAL OF SURGERY
VOLUME 162 NOVEMBER 1991
469
TABLE I Summary of Individual Reports of Delayed Thromboembolectomy Therapy Age Range Reference
PI
M/F(%)
PI
71 I29 2517.5 75125
[51 191
75125
171 El
77123 *
I41
PI
l
l
(yrs) 35-71 23-79 48-67 43-92 38-70
Claudication 5 1 0 * *
Rest Pain
Anticoagulation
2 3 4 * *
*
l
l
45-88
l
l
l
*
*
Embolectomy
Direct Retrieval
7 4 2 l
1
9
7 7 5
11
Delay to Therapy
Outcome Salvage 7 4 4 19 1 11 4 4
Amputation
3 9 7 3 1
NW 2-32 1-12 l-4 1-4 3-8 >l 1-9 >I
*Notspecified.
Eight articles specifically addressing the issue of delayed thromboembolectomy were identified in a search of the recent English language medical literature [2-91. A total of 77 cases in which thromboembolectomy was performed at an interval of at least 1 week from the onset of clinical symptoms were identified from these reports. Though not all information could be obtained from all papers, a concerted effort to determine presenting symp toms, therapy rendered, outcome, demographic data, and length of delay from symptom onset to surgery was made. The results are listed in Table I. Indications for surgery, when specified, were closely divided between rest ischemia and claudication. Most surgeons utilized balloon embolectomy though at least one author stressed the importance of direct arterial exploration for older thrombo embolic material [4]. Limb salvage was achieved in 54 of 77 cases with five deaths among the patients.
of five patients (four limbs), epidural anesthetic in one of five patients (two limbs), and general anesthetic in one patient (one limb). Popliteal exploration was necessary as assessed by intraoperative PVR in two patients (three limbs) and was performed under general or epidural anesthesia. In one patient, presentation with congestive heart failure initially mandated surgery under local anesthesia. Thrombus resulted from this approach, and intraarterial urokinase (50,000 units) was then introduced and allowed to dwell for 30 minutes. Postoperatively, the patient’s limb was still ischemic, albeit improved, and, after cardiac stabilization, he was returned to the operating room for bilateral popliteal explorations under epidural anesthesia. Pre or postoperative echocardiography revealed ab normalities in all cases. Thrombus was never identified in the cardiac chambers; however, chronic atria1 fibrillation with reduced ejection fraction was diagnosed in two patients. Left ventricular hypertrophy in one patient, absent prosthetic aortic valve vegetations but poor ventricular function in one patient, and an atrioseptal defect with right to left shunt in one patient were also diagnosed. Only one patient had a history suggesting chronic mild peripheral vascular occlusive disease. All patients except this one had normal arteries that were identified at surgery with subsequent restoration of pedal pulses. In the patient with chronic mild claudication (who had had four previous strokes, a prosthetic aortic valve, and still represented a significant operative risk), surgery revealed an embolus lodged in an area of superficial femoral artery plaque. Though pedal pulses were not restored, the patient returned to her baseline state of compensated occlusive disease postoperatively. One patient underwent a four-compartment fasciotomy after popliteal embolectomy, and reperfusion led to immediate edema of the lower leg. The patient presenting with tissue loss underwent successful closed live-toe transmetatarsal amputation. All patients had anticoagulation therapy with heparin and Coumadin (Du Pont Pharmaceuticals, Wilmington, Del.) postoperatively. There have been no recurrences of limb ischemia during follow-up ranging from 11 to 20 months. Patients were evaluated with serial exam and PVR to document the level of perfusion.
The concept of delayed embolectomy is not a new one. Haimovici in 1959 [ZO]and Spencer in 1964 [II] reported good results with patients operated on for limb emboli several weeks after the onset of clinically significant events. The introduction of the balloon embolectomy catheter greatly simplified the operative approach; however, in at least one large study comparing current results with historical reviews [IZ], a significant reduction in mortality and increase in limb salvage was not noted. The advancing age and more complex presentations of vascular patients in the modern era likely contribute to this morbidity despite recent advances in perioperative care. The primary etiology in our small series, as well as in other recent reports, remains cardiac. Most patients will demonstrate an atria1 or ventricular abnormality in rhythm or anatomy. Certainly, other sources, including proximal aneurysm or atherosclerosis, iatrogenic catheter injury, prosthetic graft, and paradoxical embolus, may be considered. However, it appears that, though the population is mostly elderly with multisystem disease, most will not have associated peripheral occlusive disease. In appropriate cases, chronic limb ischemia can be treated with embolectomy alone with good success in an elderly and high-risk population, further supporting the aggres-
470
162
THE AMERICAN
JOURNAL
OF SURGERY
VOLUME
COMMENTS
NOVEMBER
1991
DELAYEDTHROMBOEMBOLECTOMY
sive approach to limb salvage espoused in recent vascular literature as well as creating a new avenue of therapy . Clinical presentation and noninvasive findings provide several clues to the diagnosis. Our patients experienced an acute onset of mild vascular symptoms that rapidly progressed, and all but one had known cardiac pathology at the time of admission. Noninvasive evaluation in several cases revealed a sudden, severe diminution of perfusion at the calf level, which implied that there was a poorly collateralized superficial femoral artery occlusion. As most patients with infrainguinal rest ischemia have more than one level of disease, these findings suggested an acute occlusion of a previously normal or insignificantly narrowed vessel or, alternatively, occlusion of both the superficial femoral artery and profunda femoris artery. Finally, all of our patients had echocardiograms that revealed either dilated or abnormal chambers or intracardiac defects. As imaging techniques for intracardiac thrombi are suboptimal at this time, fmclmg chamber abnormalities in conjunction with ischemic limbs raises the Spector of distal embolization. Even in the patient with recent onset of claudication, early diagnosis of embolism not only will facilitate limb surgery but will identify the patient at risk for subsequent, more dangerous, embolization in whom anticoagulation therapy is indicated. Diagnosis in the patient with clear indications for limb salvage surgery (i.e., rest pain or tissue loss) will be readily made by angiography. In the patient with less obvious symptoms or with claudication alone, a history of arrhythmia, cardiac chamber or valve abnormality, acute onset and asymmetric examination, or noninvasive lindings as described above that are suggestive of emboli will lead us to angiography in patients in whom normally a less-aggressive approach would be taken. The technique for thromboembolectomy in the delayed setting does not differ markedly from the routine approach. Some authors have advocated direct approaches to those emboli that are refractory to balloon removal. This was necessary in two patients; in one of these patients, the groin approach inadequately restored perfusion. For one patient, in whom popliteal embolization was suspected, we would have preferred direct exploration with access to the popliteal branches at the outset. The patient’s presenting condition in florid congestive failure with a suspected myocardial infarction precluded this, and we used a temporizing measure to allow the patient’s condition to stabilize, which extended the chance for performing limb salvage. In the setting of suspected involvement of the popliteal and its branches with thromboembolic debris, we advocate direct popliteal exploration, longitudinal arteriotomy after inspection of the vessel, direct passage of catheters, and inspection of the material returned. Although we have not routinely utilized intraoperative angiography, this approach would allow excellent angiographic visualization of the tibia1 vessels and would also facilitate the use of the popliteal artery for a bypass should the need arise. In the patient for whom revascularization via an extensive procedure is considered excessive, return of normal hemodynamics by intraopera-
THE AMERICAN
FORISCHEMIA
tive PVR at the transmetatarsal level may be sufficiently reassuring to obviate the need for intraoperative angiography. Patch graft closure to prevent popliteal artery stenosis is mandatory in this setting [ 131. Pop&al exploration may also help to diagnose popliteal aneurysm, which, when thrombosed, may mimic a popliteal embolism. The use of intraoperative lytic therapy was recently advanced by Parent et al [I#] and Quinones-Baldrich et al [IS]. A recent series of experiments by QuinonesBaldrich et al [Id] demonstrated that the no reflow phenomenon in an ischemic limb may be averted by using reperfttsate containing lytic agents and may avert a poor outcome due to microcirculatory thrombi despite the reperfusion of major vessels. We support this concept and found such an approach beneficial as a temporizing and adjunctive measure. Although no specific recommendations can be made concerning intraoperative lytic therapy, we have used a protocol similar to Quinones-Baldrich et al [ 151. Fifty thousand units of urokinase is dissolved in 100 mL heparinized normal saline introduced by catheter into the artery and allowed to dwell for 30 minutes before flow is reestablished. Postoperative evaluation of patients with limb emboli should include cardiac and aortic ultrasound. A search for an arterial based lesion may be prudent to prevent further episodes should the cardiac work-up be unrevealing, especially if no preoperative angiogram was obtained, if the pathologic examination of the embolic debris reveals atherosclerotic elements, and if the patient is a suitable candidate for further surgery [ 171. The newer noninvasive imaging modalities including deep duplex and magnetic resonance imaging may facilitate this search without extensive angiography [ 181.
CONCLUSION Our study considers seven limbs in live patients in which thromboembolectomy was successfully performed, at an interval up to 10 weeks following the onset of symp toms. This approach achieved limb salvage in all cases without the need for more extensive surgery. One patient had associated occlusive disease that rendered the outcome imperfect; however, this patient’s overall condition precluded any regional or general anesthetic. A review of recent and past literature supports the concept of delayed embolectomy with the authors reporting high rates of limb salvage and low morbidity. The patients presented in this series are patients with advanced cardiac disease: most would have been denied extensive revascularization procedures in a nonemergency setting, but all handled limb salvage with relative ease. Maintaining a high index of suspicion for thromboembolic events will help identify those patients in whom lifelong anticoagulation is necessary and who will benefit from less complex though effective arterial reconstructions. This small series of patients makes a number of useful diagnostic and therapeutic points, and these are exactly the kind of patients in whom multiple therapies are best used to achieve long-term limb salvage.
JOURNAL
OF SURGERY
VOLUME
162
NOVEMBER
1991
471
REFERENCES 1. Blaisdell FW, Steele M, Allen RF. Management of acute lower extremity &hernia due to embolism and thrombosis. Surgery 1978; 84: 822-34. 2. Morris WT. Late embolectomy. Br Med J 1972; 3: 631-3. 3. Milliken JC. Delayed arterial embolectomy. J Irish Med Assoc 1968; 61: 390-2. 4. Naqvi MA, Mackenzie DH, Allen LS, et al. Delayed arterial embolectomy. Can J Surg 1974; 17: 335-9. 5. Jarrett F, Dacumos GC, Crumeny AB, et al. Late appearance of arterial emboli: diagnosis and management. Surgery 1979; 86: 898-905. 6. I.&n BH, Giordano JM. Delayed arterial embolectomy. Surg Gynecol Obstet 1982; 155: 549-51. 7. Hammarsten J, Holm J, Schersten J. Positive and negative effects of anticoagulant treatment during and after arterial embolectomy. J Cardiovasc Surg 1978; 19: 373-9. 8. Ammann J, Seiler H, Vogt B. Delayed arterial embolectomy: a plea for a more active surgical approach. Br J Surg 1976; 63: 73-6. 9. Macgowan WAL, Mooneeram R. A review of 174 patients with arterial embolism. Br J Surg 1973; 60: 894-8. 10. Haimovici H. Late arterial embolectomy. Surgery 1959; 46: 775-86.
472
THE AMERICAN JOURNAL OF SURGERY
11. Spencer FM, Eiseman B. Delayed arterial embolectomy-a new concept. Surgery 1964; 55: 64-71. 12. Abbott WM, Maloney RD, McCarr CC, et al. Arterial embolism: a 44 year perspective. Am J Surg 1983; 143: 460-4. 13. Gupta SK, Samson RH, Veith FJ. Embolectomy of the distal part of the popliteal artery. Surg Gynecol Obstet 1981; 153: 255-7. 14. Parent FN, Bernhard VM, Pabst TS, MC Intyre KE, Hunter GC, Malone JM. Fibrinolytic treatment of residual thrombus after catheter embolectomy for severe lower limb ischemia. J Vast Surg 1989; 9: 153-60. 15. Quinones-Baldrich WJ, Baker JD, Busuttil RW. Intraoperative infusion of lytic drugs for thrombotic complications of revascularization. J Vast Surg 1989; 10: 408-17. 16. Quinones-Baldrich WJ, Chervu A, Moore WS. Skeletal muscle function after ischemia: “No reflow” vs reperfusion injury [abstract]. Presented at the Association for Academic Surgery, Louisville, Kentucky, 1989. 17. Williams GM, Harrington D, Burdick J, White RI. Mural thrombus of the aorta. Ann Surg 1981; 194: 737-44. 18. Langsfeld M, Neptune J, Hershey FB. The use of deep duplex scanning to predict hemodynamically significant aortoiliac stenosis. J Vast Surg 1988; 7: 359-9.
VOLUME 162 NOVEMBER 1991
