- Email: [email protected]
Femorofemoral grafts for lower limb ischemia caused by intra-aortic balloon pump
Femorofemoral grafts for lower limb ischemia caused by intra-aortic balloon pump Mark L. Friedell, M.D., Joseph Alpert, M.D., Victor Parsonnet, M.D., Donald K. Brief, M.D., Bruce J. Brener, M.D., Robert J. Goldenkranz, M.D., and Jerome Nozick, M.D., Newark,NJ. From January 1975 to December 1985, 1454 patients had an intra-aortic balloon inserted for cardiac assistance. Eighty balloon-dependent patients had severe limb ischemia and required a femorofemoral graft (FFG) (5% of the total group of patients). Twenty-nine of the 80 patients with grafts (or 36%) left the hospital and 28 were followed up for an average of 40 months to determine late complications associated with the crossover grafts. All grafts remained patent. The 28 patients were classified into five groups according to the degree and type of lower limb ischemia. Group I consisted of 13 asymptomatic patients (46%); group II had four (14%) patients with mild claudication caused by preexisting peripheral arteriosclerosis; group III comprised four patients (14%) without preexisting disease but claudication subsequent to the FFG; group IV had five patients with irreversible ischemic sequelae before grafting ending in amputation, foot drop, or persistent paresthesia; and group V consisted of two patients with graft infection (7%). The perioperative mortality rate of the balloon-dependent patients with an FFG (64%) reflects the gravity of the cardiac condition. Placement of an FFG to relieve limb ischemia in these patients is followed by few immediate or late complications in the survivors and any persistent limb changes were related to the prolonged ischemia present before revascularization. Our data suggest that in balloon-dependent patients with limb-threatening ischemia, aggressive use of the FFG is limb-saving, durable, and allows continuation of balloon support. (J VAse SURG 1987;5:180-6.)
As a device that provides immediate cardiac assistance, the intra-aortic balloon pump (LABP) has been used widely for the past decade. The most common IABP complication is limb ischemia, which can result from the transfemoral insertion of a large-bore LABP catheter. 15 If the balloon is removed, ischemia may be reversed. Problems arise when maintenance of pump support is essential for the patient's survival in the presence of an acute ischemic limb. Management by IABP removal, thrombectomy, and ipsilateral or contralateral replacement is plagued by recurrent ischemic complications and often femorofemoral crossover grafting (FFG) is ultimately required. Because of our favorable experience with FFG in elective patients who had symptomatic chronic unilateral iliac occlusion, 6 we began using the FFG in the IABP-dependent patients with limbFrom the Division of Vascular Surgery, Newark Beth Israel Medical Center and the Department of Surgery, University of Medicine and Dentistry of New Jersey--Medical School. Presented at the Fortieth Annual Meeting of the Society of Vascular Surgery New Orleans, La. June 9-10, 1986. Reprint requests: Joseph Alpert, M.D., Division of Vascular Surgery, Newark Beth Israel Medical Center, 201 Lyons Ave., Newark, NJ 07112.
180
threatening ischemia. We have reported highly satisfactory immediate postoperative results with the ~ FFG in this group of patients. 7 The present report analyzes the subsequent course of these patients and was undertaken to determine the ultimate value,~.pf our method of management for this class of patients. PATIENTS A N D M E T H O D S From January 1975 through December 1985, 1454 patients had IABPs inserted to provide cardiac support at the Newark Beth Israel Medical Center. Eighty balloon-dependent patients (5%) had limb ischemia severe enough to warrant FFG. Of the 29 patients (36%) with FFG who survived and left the~ hospital, 2 to 100 months' follow-up was obtained for 28 patients (average 40 months). There were 16 men and 12 women whose ages ranged from 35 to 73 years (mean 55 years.) Fourteen patients (50%) underwent personal interview, physical examination, and noninvasive vascular laboratory studies. Datawere obtained on the remaining i4 patients by telephone communication with the patient, family members, or attending physician. Our technique of IABP used the Datascope system Console and Catheters (Datascope Co,, P~7'~a-
Volume 5 Number 1 Januau 1987
FFGsfor Iower limb ischemia ca~,sedby IABPs 181
CATHETER
FEMORAL A: SHEATH IABP CATH. ~tgCa~c~
Fig. 1. taemorofemoral crossover graft with intra-aortic balloon pump in place.
-anus, N.J.). The two sheath/catheter combinations used measured 12.5 Fr./12 Fr. and 11.5 Fr./10.5 Fr., respectively. The indications for IABP placement in the group were varied and included perioperative cardiac support, postinfarction angina, unstable angina, and cardiogenic shock. These indications mirrored the IABP population as a whole. Before 1980, in 11 cases, we used a Dacron slceve graft for the site of insertion of the catheter. The percutaneous technique of insertion was adopted shortly thereafter in 14 cases and became the preferred method because of its ease of insertion and removal and its freedom from infection. 8q° In four other patients, either because the artery had already been dissected or the femoral pulse was not palpable, the IABP was introduced through the exposed femoral artery. The protocol for anticoagulation of IABP patie ,~s has been the same with use of the open as well
as the percutaneous techniques of placement. If the patient had no contraindication to heparin t.hcrapy and had not had recent cardiac surgery, an intravenous bolus o f 5000 units ofheparin was given at the time of balloon inscrtion. Then a continuous intravenous heparin infusion of 1000 U/hr was administered for the length of time the patients wcre receiving IABP support. Patients who had recent cardiac surgery or in whom heparin was contraindicated received continuous intravenous infusion o f dextran 40 at the ratc of 15 to 25 ml/hr. Our technique for FFG in the balloon-dependent patient includes division of the common femoral artery just distal to the IABP catheter with an end-toend anastomosis constructed between the graft and recipient common femoral artery (Fig. 1). The catheter is placed in the proximal cut end of the femoral artery. Because subsequent LABP removal requires
Journal of VASCULAR SURGERY
182 Friedell et al.
Table I. Asymptomatic patients IABP insertion Patient
Age (yr)
Sex
lnd
I 2 3 4 5 6
58 44 56 49 68 65
F M F M F F
Wean CPB MI/CS Wean CPB Post OHS Wean CPB UA
7
51
M
Post OHS
8 9 10
60 67 55
M F M
UA MI/CS CS
11 12 13
57 48 59
M M F
PIA Wean CPB PIA
Graft material
Duration IABP (days)
Complications
Follow-up (too)
Immed. Imrned. Immed, Immed, Immed, Immed.
D D D D PTFE PTFE
5 3 4 4 4 6
Groin seromas No Groin infection Groin infection No No
3 44 100 88 52 60
Immed.
D
2
No
40
4 days Imrned. Immed.
I~FE PTFE PTFE
13 7 3
No No No
6 48 33
Immed. Immed. Immed.
PTFE PTFE PTFE
8 3 3
No No Groin infection
31 29 3
Open~closed Onset ischemia Open Open Open Open Closed Open Seld Open Seld Closed Closed Open Seld Closed Closed Closed
lnd = indication; IABP = intra-aortic ballon pump; CPB = cardiopulmonav bypass; MI = myocardial infarction; OHS = open heart surgery; UA = unstable angina; CS = cardiogenic shock; PIA = postinfarction angina; Seld = Seldinger technique; D = Dacron; PTFE = polytetrafluoroethylene.
Table II. Patients with preexisting claudication IABP insertion Age Patient (yr) Sex
Ind
Duration Open~dosed Onset ischemia Graft material IABP ( d a y s )
14 I5 16
52 50 64
F M F
PIA MUCS MI/CS
Open Open Closed
Immed. 9 days Immed.
D D ASV
8 5 6
17
73
M
PlA
Closed
Immed.
PTFE
2
Complications No No Ischemic left foot Ax-FBPG 9 days later Groin infection
Follow-up (mo) 84 63 2 10
ASV = autogenous saphenous vein; Ax-FBPG = axillo-femora[ bypass graft; for other abbreviations see Table I.
reopening the groin and oversewing the open proximal end of the femoral arteu, this method of graft placement allows separation of the IABP catheter from the prosthetic FFG to avoid reexposure of the graft and minimize infection. Eleven woven or lmitted Dacron FFGs were inserted during the period 1975 to 1981. Expanded polytetrafluoroethylene (PTFE) grafts, 6 mm in diameter, are currently preferred and have been used since 1980. In one instance, a portion of the harvested saphenous vein was available. IABP removal usually occurred 2 to 13 days (mean 5 days) after the FFG was performed. The grafts were not disturbed during balloon removal. All patients with IABPs were monitored hourly in the critical care unit. The presence or absence of limb ischemia was ascertained by clinical and Doppler ultrasound examination. The conscious patient usually complained of pain or numbness in the leg, and in more advanced stages, of loss of motor function. Evaluation was more difficult in the anesthetized or sedated patient and reliance on the Doppler findings
was critical. In every case of limb ischemia occurring after placement of the IABP, the catheter sheath was withdrawn as the first step in treatment. RESULTS During the mean observation period of 40 months all 28 grafts remained open. Observations during the follow-up period allowed assignment of the 28 patients to the following five groups: group I, 13 asymptomatic patients (46%) (Table I); group II, four patients with intermittent claudicationv before and after the surgical procedure (14%) (Table II); group III, 4 patients who were initially asymptomatic after FFG insertion but subsequently suffered claudication in the recipient limb (14%) (Table III); group IV, five patients who had irreversible ischemic tissue damage that was evident im-~ mediately despite placement of an FFG (18%) (Table IV); and group V, two patients with graft infec- ' tion (7%) (Table V). One patient in group II subsequently required
Volume 5 Number 1 January 1987
FFGs for lower limb ischemia caused by IABPs 183
Table III. Patients with postoperative claudication IABP insertion Patient
Age (yr)
Sex
Ind
18 19
57 35
M F
MI/CS Post OHS
20 21
43 59
F M
MI/CS UA
Graft material
Duration IABP (days)
Immed. Immed.
D PTFE
4 2
No No
64 60
2 days Unclear (in OR)
PTFE PTFE
7 3
No Groin infection
43 11
Open~dosed Onset ischemia Open Open Seld Closed Closed
Complications
Follow-up (too)
For abbreviations see Table I.
Table IV. Patients with permanent ischemic damage IABP insertion ge
Patient (yr) Sex
Ind
Open/closed Onset ischemia
Duration Graft mate~qal IABP (days)
-2 23 24 25
50 52 38 56
M M M M
Wean CPB UA PIA Wean CPB
Open Closed Closed Closed
Immed. 7 days (in OR) 4 days (in OR) Irnmed.
D PTFE PTFE PTFE
5 9 6 3
26
58
M
Pre OHS
Closed
Unclear (in OR)
PTFE
3
Complications Above-knee amputation Paresthesia Paresthesia; fasciotomy Foot drop, leg pain, and fasciotomy Foot drop; fasciotomy
Follow-up (too) 73 62 52 36 15
For abbreviations see Table L
Table V. Patients with graft infection IABP insertion
Age
Patient (yr) Sex
Ind
Duration Open/dosedOnset ischemia Graft material IABP (days)
27
59
F
MI/CS
Open
Immed.
D
11
28
65
F
Prophylaxis for abdominal surgery
Open
Immed.
D
4
Complications Sartorius flap; catheter irrigation 1 mo after FFG Graft removal 8 mo after FFG
Follow-up (too) 7 8
FFG = femorofemorai graft; for other abbreviations see Table I.
two femoropopliteal bypass procedures in the recip4ent limb. The same patient also had to have an FFG thrombectomy during the second femoropopliteal bypass for graft occlusion after angiography. Among the five patients in group IV, the ischemic damage resulted in an above-knee amputation and 'foot drop in two patients and paresthesia and rest pain in another two. Three patients in this group required fasciotomy. Among the patients in group V, one case of infection occurred 1 month postoperatively and was treated successfully by debridement, rotation of a sartorius muscle flap, and local catheter antibiotic irrigations. A second infection occurred 8 months after grafting and required removal of the graft and the :interposition of the endarterectomized external lilac artery between the ipsilateral hypogastric and common femoral arteries. It may also be mentioned that
six patients had superficial wound infections or seromas postoperatively without involvement of the graft.
DISCUSSION The high perioperative mortality rate of balloondependent patients with an FFG (64%) reflects the seriousness of the cardiac condition. Since placement of the FFG is a procedure that does not provoke untoward physiologic effects of clinical significance, it cannot be regarded as a significant contributing factor to death. The female/male ratio in the clinical population of the Newark Beth Israel Medical Center since 1980 in the clinical population that required FFG was 3:4, whereas the same ratio in patients undergoing cardiac surgical procedures of all kinds was 1 : 3. The most likely explanation of this disparitT (previously
Journal of VASCULAR SURGERY
184 Friedell et al.
observed by others 7) is anatomic, that is, the generally small caliber of the arterial trunks in the female body. Another factor that played an important role in the need for the use of FFG was associated athcrosclerotic lesions of the common femoral and external iliac arteries. The FFG did not adversely affect 79% of the patients in this series who remained asymptomatic or had symptoms unrelated to the FFG. The five patients in group IV incurred irreversible sequelae as a result of a protracted ischemic time before FFG. Most difficulties occurred when the insidious progression of ischemia made its recognition difficult or when the delay occurred in timing the insertion of the catheter because the patient was not expected to survive the next 48 hours. The presence of ischcmia was not recognized in three coronary artery bypass patients in whom the IABP was placed preoperatively. In two of these patients the IABP had been inserted several days preoperatively and no obvious limb ischemia was noticed up to the time of the operation. During the operation, the iliac artery containing the balloon catheter thromboscd, presumably because of the low flow hemodynamic state. In two cases the ischcmia of the limbs was noted after the intraoperative IABP placement, but the general condition of the patients was so grave (one with a cardiac index of 1,5) that survival was believed unlikely with or without an FFG. Ultimately the patients' conditions improved and the FFG was performed. Despite revascularization, the prolonged ischemia led to an above-knee amputation in one and foot drop and persistent pain in the other. Although graft infection developed in two patients (group V), the process was controlled without limb loss. These infections occurred early in the series when the open technique for IABP placement was used. There have been no graft infections since 1980 when routine use of the percutaneous approach began. The most enigmatic category of patients was group III. All four patients had documented pedal pulses and normal noninvasive studies immediately after FFG placement but during the ensuing years gradual progressive claudication developed in the recipient limb. In three of these patients treadmill testing, reactive hyperemia, and measurement of ankle pressures confirmed the presence of diminished inflow to the limb. Pulses were palpable in the contralateral foot. Because the claudication was not disabling, arteriography was not performed and the exact cause of the reduced blood flow has not been detcrmincd. One can assume that the cause was an-
atomic, that is, stenosis of an inflow artery. This assumption is supported by the circumstance that two of these patients were young women whose arteries were probably hypoplastic, The good results in these cases wcrc obtained, without the routine use of preoperative arteriography and in many of the patients significant occlusive disease was present in the donor artery. The only patient in whom angiographic study was carried out preoperatively after the completion of cardiac catheterization required an axillofemoral graft to correct steal syndrome that developed because of severe ipsilatcral iliac stenosis. A long-term follow-up of FFG patients under similar conditions was carried out by Gold et al. 11 The authors followed up nine patients who left'he hospital for a period of time ranging from 6 months to 4 years; the patients in general did well, although two required subsequent femoropopliteal bypass procedures. In three of the nine patients with severe atherosclerotic disease or with hypoplastic arteries we removed the FFG and IABP simultaneously. We do not recommend immediate removal of the FFG after IABP support is stopped. FFG is a simple, effective, and durable treatment" for limb ischemia in IABP-dependent patients. Since the advent of percutaneous IABP insertion, graft infection, which previously was the most serious complication, has not been a problem. The few patients who had arterial insufficiency as the result of this procedure can be treated successfully. Aggressive use of the FFG under the clinical conditions described appears to be indicated for the prevention of debilitating ischemic sequelae. REFERENCES
1. Mpert J, Bhaktan EK, GielchinskyI, et al. Vascular compli-~ cations of intra-aortic balloon pumping. Arch Surg 1976; 111:i190-5. 2. Pace PD, Tilney NL, Lesch M, Couch NP. Peripheral arterial complications of intra-aortic balloon counterpulsation. Surgery 1977;82:685-8. 3. McCabe JC, Abel RM, Subramanian VA, Gay WA. Complications of intra-aortic balloon insertion and counterpul~sation. Circulation 1978;57:769-73. 4. Harvey}'C, GoldsteinIE, McCabelC, Hoover EL, Gay WA, Subramanian VA. Complicationsof percutaneous intra-aortic balloon pumping. Circulation 1981;64(suppl 2):114-7. 5. Perler BA, McCabe CJ, Abbott WM, BuckelyMJ. Vascular complications of intra-aortic balloon counterputsation. Arch Surg 1983;118:957-62. 6. Brief DK, Brener BJ, Alpert J, Parsonnet V. Crossoverfemorofemoral grafts followed up 5 years or more: An analysis.. Arch Surg 1975;110:1294-9. 7. Alpert J, Parsonnet V, Goldenkranz R], et al. Limb ischemia during intraaortic balloon pumping: Indication for femoro-
Volume 5 Number 1 January 1987
femoral crossover graft. J Thorac Cardiovasc Surg 1980;79: 729-34. 8. Gonzalez M, Install E, Tremouroux J. Percutaneous intraaortic balloon pumping: initial experience. Intensive Care Med I982;81:143-7. 9. Grayzel J. Clinical evaluation of the Percor percutaneous intraaortic balloon: Cooperative study of 722 cases. Circulation 1982;66(suppl 1):223-6.
DISCUSSION
Dr. David C. Brewster (Boston, Mass.). Since its introduction in 1968, intra-aortic balloon counterpulsation has proved to be an effective means of support for the C~ling myocardium in a variety of clinical circumstances. However, complications--most commonly ipsilateral limb ischemia--are not rare and often require the attention of the vascular surgeon. Our own experience at the Massachusetts General Hospital through 1981 was reviewed and reported by Perler. He found an overall incidence of 5% ofischemia occurring after open or so-called surgical balloon insertion. Since then, as demonstrated by the series reported by Dr. Friedell et al., percutaneous methods of balloon insertion have increased substantially and with them I believe the frequency of ischemic complications, so that now we estimate approximately 10% of balloon pump patients may manifest some degree of limb ischemia. Such problems are more common in female patients with small arteries. Distal pulse deficits caused by partial obturation of small or diseased arteries by the balloon catheter are more common but often do not require surgical intervention. However, more serious limb-threatening ischemia in the truly balloon-dependent patient is much more problematic. C ;r own approach in the preoperative cardiac patient is often toward early, prompt cardiac operation and rapid weaning from the balloon in the early postoperative period. If this is not possible, we have most often chosen to transfer the balloon to the contralateral limb and restore circulation of the ischemic side by thromboembolectomy, local repair of elevated arterial plaques, and so forth. We have rarely used femorofemoral grafts in this capacity, but this approach may be beneficial in some of the circumstances outlined by Dr. Friedell and his colleagues. Certainly their good results suggest this may be a very valuable alternative. Dr. Friedell, if the contralateral iliac artery is capable of serving as the donor side of a crossover graft, why not simply transfer the balloon pump to that side) Second, and perhaps most important, we all agree that assessment of the severity of ischemia in these vasoconstricted patients, who are often on a respirator and unable to communicate, is indeed often extremely difficult. Can you elaborate on guidelines and criteria that you and your associates use to determine when surgical intervention is mandatory? Dr, Nathan P. Couch (Boston, Mass.). Despite the
FFGs for lower limb isehemia caused by IABPs 185
i0. Collier PE, Liebler GA, Park SB, Burkhotder JA, Mahcr TD, Magovern GJ. Is percutmaeous insertion of the intra-aortic balloon pump through the femoral arteu the safest technique? J VASe SUgG 1986;3:629-34. 11. Gold JP, Cohen J, Shemin RJ, et al. Femorofemoral bypass to relieve acute leg ischemia during intra-aorric balloon pump cardiac support. J VASe SURG 1986;3:35i-4.
best laid plans and in the best hands, "surge U still can beget more surgery." This is one of those examples. These authors deserve high praise for the low incidence of late leg symptoms they have achieved in these 1454 patients receiving IABPs, only 80 of whom required the femorofemoral bypass grafts. In the final analysis, 14 of the patients had some degree of late arterial occlusive disease but not necessarily all related to the balloon pump. This should be said in the context of a 64% mortality ratc for the entire group of 80 patients who had severe ischemia, at least severe enough to require femorofemoral grafting. This finding traderscores an extremely important point: a large part of the leg ischemia--at least in two thirds of the patients--was probably due to "cardiac pump failure," with just incidental but still important regional ischemia. Were any of the deaths in these patients those suspected of being due to "reflow toxicity," that is, sudden flooding of the general circulation by potassium and lactic acid? In other words, how many patients died within a few hours of declamping the graft with precipitous drops in blood pressure or cardiac output, or with acute savage arrhythmias? The second question is suggested by the fact that only 6 of the 28 patients available for follow-up required the IABP for assistance during the bypass. Has this indication for IABP been further reduced as cardioptegic methods have improved? Finally, can the caliber of the IABP catheter be reduced still further--possibly to 7 or 8 Ft.--to minimize the chance for critical iliofemoral artery- occlusion? We certainly agree that there are far more of these problems in women, especially small older women. Dr. Friedell (closing). I thank the discussants. In response to Dr. Couch's questions, the frequency of balloon pump use has been about the same during the years for weaning from cardiopulmonary bypass. Regarding the number of immediate postoperative deaths occurring af?er placement of the crossover graft, there were none. As to catheter size, if cardiac support were as efficacious with the smaller catheter, it would certainly be an advantage in reducing the occurrence ofcatheter-rdated arterial occlusion. Dr. Brewster, early in our experience, we did transfer the balloon from one side to the other but had two problems. One was that balloon support was interrupted for
186
journal of VASCULAR SURGERY
Friedell et al.
that period of time required for the transfer and, depending on technical problems with replacement, the time duration could be substantial. We also fotmd frequently that the contralateral side had atherosclerotic disease similar to the side of original balloon catheter placement. Therefore, we went directly to the femorofemoral graft as our preferred method to handle the problem.
In evaluating the sedated patient who requires mechanical ventilation, particularly the postoperative patient, we adopted a policy several years ago of the careful use of Doppler monitoring. Immediately after operation we are satisfied with a popliteal Doppler signal. However, if all pulses are absent, this is an emergent situation that requires a crossover graft.
B O U N D VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the JOURNAL OF VASCULAR SURGERY for 1987 arc available to subscribers only. They may be purchased from the publisher at a cost of S44.00 ($58.00 international) for VoL 5 (January to June) and Vol. 6 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, The C. V. Mosby Company, 11830 Westline Industrial Drive, St. Louis, MO 63146, USA. In the United States call toll free: (800)325-4177, ext. 351. In Missouri call collect: (314)872-8370, ext. 351. Subscriptions must be in force to quailS7. Bound volumes are not available in place of a regular JOURNAL subscription.
As a device that provides immediate cardiac assistance, the intra-aortic balloon pump (LABP) has been used widely for the past decade. The most common IABP complication is limb ischemia, which can result from the transfemoral insertion of a large-bore LABP catheter. 15 If the balloon is removed, ischemia may be reversed. Problems arise when maintenance of pump support is essential for the patient's survival in the presence of an acute ischemic limb. Management by IABP removal, thrombectomy, and ipsilateral or contralateral replacement is plagued by recurrent ischemic complications and often femorofemoral crossover grafting (FFG) is ultimately required. Because of our favorable experience with FFG in elective patients who had symptomatic chronic unilateral iliac occlusion, 6 we began using the FFG in the IABP-dependent patients with limbFrom the Division of Vascular Surgery, Newark Beth Israel Medical Center and the Department of Surgery, University of Medicine and Dentistry of New Jersey--Medical School. Presented at the Fortieth Annual Meeting of the Society of Vascular Surgery New Orleans, La. June 9-10, 1986. Reprint requests: Joseph Alpert, M.D., Division of Vascular Surgery, Newark Beth Israel Medical Center, 201 Lyons Ave., Newark, NJ 07112.
180
threatening ischemia. We have reported highly satisfactory immediate postoperative results with the ~ FFG in this group of patients. 7 The present report analyzes the subsequent course of these patients and was undertaken to determine the ultimate value,~.pf our method of management for this class of patients. PATIENTS A N D M E T H O D S From January 1975 through December 1985, 1454 patients had IABPs inserted to provide cardiac support at the Newark Beth Israel Medical Center. Eighty balloon-dependent patients (5%) had limb ischemia severe enough to warrant FFG. Of the 29 patients (36%) with FFG who survived and left the~ hospital, 2 to 100 months' follow-up was obtained for 28 patients (average 40 months). There were 16 men and 12 women whose ages ranged from 35 to 73 years (mean 55 years.) Fourteen patients (50%) underwent personal interview, physical examination, and noninvasive vascular laboratory studies. Datawere obtained on the remaining i4 patients by telephone communication with the patient, family members, or attending physician. Our technique of IABP used the Datascope system Console and Catheters (Datascope Co,, P~7'~a-
Volume 5 Number 1 Januau 1987
FFGsfor Iower limb ischemia ca~,sedby IABPs 181
CATHETER
FEMORAL A: SHEATH IABP CATH. ~tgCa~c~
Fig. 1. taemorofemoral crossover graft with intra-aortic balloon pump in place.
-anus, N.J.). The two sheath/catheter combinations used measured 12.5 Fr./12 Fr. and 11.5 Fr./10.5 Fr., respectively. The indications for IABP placement in the group were varied and included perioperative cardiac support, postinfarction angina, unstable angina, and cardiogenic shock. These indications mirrored the IABP population as a whole. Before 1980, in 11 cases, we used a Dacron slceve graft for the site of insertion of the catheter. The percutaneous technique of insertion was adopted shortly thereafter in 14 cases and became the preferred method because of its ease of insertion and removal and its freedom from infection. 8q° In four other patients, either because the artery had already been dissected or the femoral pulse was not palpable, the IABP was introduced through the exposed femoral artery. The protocol for anticoagulation of IABP patie ,~s has been the same with use of the open as well
as the percutaneous techniques of placement. If the patient had no contraindication to heparin t.hcrapy and had not had recent cardiac surgery, an intravenous bolus o f 5000 units ofheparin was given at the time of balloon inscrtion. Then a continuous intravenous heparin infusion of 1000 U/hr was administered for the length of time the patients wcre receiving IABP support. Patients who had recent cardiac surgery or in whom heparin was contraindicated received continuous intravenous infusion o f dextran 40 at the ratc of 15 to 25 ml/hr. Our technique for FFG in the balloon-dependent patient includes division of the common femoral artery just distal to the IABP catheter with an end-toend anastomosis constructed between the graft and recipient common femoral artery (Fig. 1). The catheter is placed in the proximal cut end of the femoral artery. Because subsequent LABP removal requires
Journal of VASCULAR SURGERY
182 Friedell et al.
Table I. Asymptomatic patients IABP insertion Patient
Age (yr)
Sex
lnd
I 2 3 4 5 6
58 44 56 49 68 65
F M F M F F
Wean CPB MI/CS Wean CPB Post OHS Wean CPB UA
7
51
M
Post OHS
8 9 10
60 67 55
M F M
UA MI/CS CS
11 12 13
57 48 59
M M F
PIA Wean CPB PIA
Graft material
Duration IABP (days)
Complications
Follow-up (too)
Immed. Imrned. Immed, Immed, Immed, Immed.
D D D D PTFE PTFE
5 3 4 4 4 6
Groin seromas No Groin infection Groin infection No No
3 44 100 88 52 60
Immed.
D
2
No
40
4 days Imrned. Immed.
I~FE PTFE PTFE
13 7 3
No No No
6 48 33
Immed. Immed. Immed.
PTFE PTFE PTFE
8 3 3
No No Groin infection
31 29 3
Open~closed Onset ischemia Open Open Open Open Closed Open Seld Open Seld Closed Closed Open Seld Closed Closed Closed
lnd = indication; IABP = intra-aortic ballon pump; CPB = cardiopulmonav bypass; MI = myocardial infarction; OHS = open heart surgery; UA = unstable angina; CS = cardiogenic shock; PIA = postinfarction angina; Seld = Seldinger technique; D = Dacron; PTFE = polytetrafluoroethylene.
Table II. Patients with preexisting claudication IABP insertion Age Patient (yr) Sex
Ind
Duration Open~dosed Onset ischemia Graft material IABP ( d a y s )
14 I5 16
52 50 64
F M F
PIA MUCS MI/CS
Open Open Closed
Immed. 9 days Immed.
D D ASV
8 5 6
17
73
M
PlA
Closed
Immed.
PTFE
2
Complications No No Ischemic left foot Ax-FBPG 9 days later Groin infection
Follow-up (mo) 84 63 2 10
ASV = autogenous saphenous vein; Ax-FBPG = axillo-femora[ bypass graft; for other abbreviations see Table I.
reopening the groin and oversewing the open proximal end of the femoral arteu, this method of graft placement allows separation of the IABP catheter from the prosthetic FFG to avoid reexposure of the graft and minimize infection. Eleven woven or lmitted Dacron FFGs were inserted during the period 1975 to 1981. Expanded polytetrafluoroethylene (PTFE) grafts, 6 mm in diameter, are currently preferred and have been used since 1980. In one instance, a portion of the harvested saphenous vein was available. IABP removal usually occurred 2 to 13 days (mean 5 days) after the FFG was performed. The grafts were not disturbed during balloon removal. All patients with IABPs were monitored hourly in the critical care unit. The presence or absence of limb ischemia was ascertained by clinical and Doppler ultrasound examination. The conscious patient usually complained of pain or numbness in the leg, and in more advanced stages, of loss of motor function. Evaluation was more difficult in the anesthetized or sedated patient and reliance on the Doppler findings
was critical. In every case of limb ischemia occurring after placement of the IABP, the catheter sheath was withdrawn as the first step in treatment. RESULTS During the mean observation period of 40 months all 28 grafts remained open. Observations during the follow-up period allowed assignment of the 28 patients to the following five groups: group I, 13 asymptomatic patients (46%) (Table I); group II, four patients with intermittent claudicationv before and after the surgical procedure (14%) (Table II); group III, 4 patients who were initially asymptomatic after FFG insertion but subsequently suffered claudication in the recipient limb (14%) (Table III); group IV, five patients who had irreversible ischemic tissue damage that was evident im-~ mediately despite placement of an FFG (18%) (Table IV); and group V, two patients with graft infec- ' tion (7%) (Table V). One patient in group II subsequently required
Volume 5 Number 1 January 1987
FFGs for lower limb ischemia caused by IABPs 183
Table III. Patients with postoperative claudication IABP insertion Patient
Age (yr)
Sex
Ind
18 19
57 35
M F
MI/CS Post OHS
20 21
43 59
F M
MI/CS UA
Graft material
Duration IABP (days)
Immed. Immed.
D PTFE
4 2
No No
64 60
2 days Unclear (in OR)
PTFE PTFE
7 3
No Groin infection
43 11
Open~dosed Onset ischemia Open Open Seld Closed Closed
Complications
Follow-up (too)
For abbreviations see Table I.
Table IV. Patients with permanent ischemic damage IABP insertion ge
Patient (yr) Sex
Ind
Open/closed Onset ischemia
Duration Graft mate~qal IABP (days)
-2 23 24 25
50 52 38 56
M M M M
Wean CPB UA PIA Wean CPB
Open Closed Closed Closed
Immed. 7 days (in OR) 4 days (in OR) Irnmed.
D PTFE PTFE PTFE
5 9 6 3
26
58
M
Pre OHS
Closed
Unclear (in OR)
PTFE
3
Complications Above-knee amputation Paresthesia Paresthesia; fasciotomy Foot drop, leg pain, and fasciotomy Foot drop; fasciotomy
Follow-up (too) 73 62 52 36 15
For abbreviations see Table L
Table V. Patients with graft infection IABP insertion
Age
Patient (yr) Sex
Ind
Duration Open/dosedOnset ischemia Graft material IABP (days)
27
59
F
MI/CS
Open
Immed.
D
11
28
65
F
Prophylaxis for abdominal surgery
Open
Immed.
D
4
Complications Sartorius flap; catheter irrigation 1 mo after FFG Graft removal 8 mo after FFG
Follow-up (too) 7 8
FFG = femorofemorai graft; for other abbreviations see Table I.
two femoropopliteal bypass procedures in the recip4ent limb. The same patient also had to have an FFG thrombectomy during the second femoropopliteal bypass for graft occlusion after angiography. Among the five patients in group IV, the ischemic damage resulted in an above-knee amputation and 'foot drop in two patients and paresthesia and rest pain in another two. Three patients in this group required fasciotomy. Among the patients in group V, one case of infection occurred 1 month postoperatively and was treated successfully by debridement, rotation of a sartorius muscle flap, and local catheter antibiotic irrigations. A second infection occurred 8 months after grafting and required removal of the graft and the :interposition of the endarterectomized external lilac artery between the ipsilateral hypogastric and common femoral arteries. It may also be mentioned that
six patients had superficial wound infections or seromas postoperatively without involvement of the graft.
DISCUSSION The high perioperative mortality rate of balloondependent patients with an FFG (64%) reflects the seriousness of the cardiac condition. Since placement of the FFG is a procedure that does not provoke untoward physiologic effects of clinical significance, it cannot be regarded as a significant contributing factor to death. The female/male ratio in the clinical population of the Newark Beth Israel Medical Center since 1980 in the clinical population that required FFG was 3:4, whereas the same ratio in patients undergoing cardiac surgical procedures of all kinds was 1 : 3. The most likely explanation of this disparitT (previously
Journal of VASCULAR SURGERY
184 Friedell et al.
observed by others 7) is anatomic, that is, the generally small caliber of the arterial trunks in the female body. Another factor that played an important role in the need for the use of FFG was associated athcrosclerotic lesions of the common femoral and external iliac arteries. The FFG did not adversely affect 79% of the patients in this series who remained asymptomatic or had symptoms unrelated to the FFG. The five patients in group IV incurred irreversible sequelae as a result of a protracted ischemic time before FFG. Most difficulties occurred when the insidious progression of ischemia made its recognition difficult or when the delay occurred in timing the insertion of the catheter because the patient was not expected to survive the next 48 hours. The presence of ischcmia was not recognized in three coronary artery bypass patients in whom the IABP was placed preoperatively. In two of these patients the IABP had been inserted several days preoperatively and no obvious limb ischemia was noticed up to the time of the operation. During the operation, the iliac artery containing the balloon catheter thromboscd, presumably because of the low flow hemodynamic state. In two cases the ischcmia of the limbs was noted after the intraoperative IABP placement, but the general condition of the patients was so grave (one with a cardiac index of 1,5) that survival was believed unlikely with or without an FFG. Ultimately the patients' conditions improved and the FFG was performed. Despite revascularization, the prolonged ischemia led to an above-knee amputation in one and foot drop and persistent pain in the other. Although graft infection developed in two patients (group V), the process was controlled without limb loss. These infections occurred early in the series when the open technique for IABP placement was used. There have been no graft infections since 1980 when routine use of the percutaneous approach began. The most enigmatic category of patients was group III. All four patients had documented pedal pulses and normal noninvasive studies immediately after FFG placement but during the ensuing years gradual progressive claudication developed in the recipient limb. In three of these patients treadmill testing, reactive hyperemia, and measurement of ankle pressures confirmed the presence of diminished inflow to the limb. Pulses were palpable in the contralateral foot. Because the claudication was not disabling, arteriography was not performed and the exact cause of the reduced blood flow has not been detcrmincd. One can assume that the cause was an-
atomic, that is, stenosis of an inflow artery. This assumption is supported by the circumstance that two of these patients were young women whose arteries were probably hypoplastic, The good results in these cases wcrc obtained, without the routine use of preoperative arteriography and in many of the patients significant occlusive disease was present in the donor artery. The only patient in whom angiographic study was carried out preoperatively after the completion of cardiac catheterization required an axillofemoral graft to correct steal syndrome that developed because of severe ipsilatcral iliac stenosis. A long-term follow-up of FFG patients under similar conditions was carried out by Gold et al. 11 The authors followed up nine patients who left'he hospital for a period of time ranging from 6 months to 4 years; the patients in general did well, although two required subsequent femoropopliteal bypass procedures. In three of the nine patients with severe atherosclerotic disease or with hypoplastic arteries we removed the FFG and IABP simultaneously. We do not recommend immediate removal of the FFG after IABP support is stopped. FFG is a simple, effective, and durable treatment" for limb ischemia in IABP-dependent patients. Since the advent of percutaneous IABP insertion, graft infection, which previously was the most serious complication, has not been a problem. The few patients who had arterial insufficiency as the result of this procedure can be treated successfully. Aggressive use of the FFG under the clinical conditions described appears to be indicated for the prevention of debilitating ischemic sequelae. REFERENCES
1. Mpert J, Bhaktan EK, GielchinskyI, et al. Vascular compli-~ cations of intra-aortic balloon pumping. Arch Surg 1976; 111:i190-5. 2. Pace PD, Tilney NL, Lesch M, Couch NP. Peripheral arterial complications of intra-aortic balloon counterpulsation. Surgery 1977;82:685-8. 3. McCabe JC, Abel RM, Subramanian VA, Gay WA. Complications of intra-aortic balloon insertion and counterpul~sation. Circulation 1978;57:769-73. 4. Harvey}'C, GoldsteinIE, McCabelC, Hoover EL, Gay WA, Subramanian VA. Complicationsof percutaneous intra-aortic balloon pumping. Circulation 1981;64(suppl 2):114-7. 5. Perler BA, McCabe CJ, Abbott WM, BuckelyMJ. Vascular complications of intra-aortic balloon counterputsation. Arch Surg 1983;118:957-62. 6. Brief DK, Brener BJ, Alpert J, Parsonnet V. Crossoverfemorofemoral grafts followed up 5 years or more: An analysis.. Arch Surg 1975;110:1294-9. 7. Alpert J, Parsonnet V, Goldenkranz R], et al. Limb ischemia during intraaortic balloon pumping: Indication for femoro-
Volume 5 Number 1 January 1987
femoral crossover graft. J Thorac Cardiovasc Surg 1980;79: 729-34. 8. Gonzalez M, Install E, Tremouroux J. Percutaneous intraaortic balloon pumping: initial experience. Intensive Care Med I982;81:143-7. 9. Grayzel J. Clinical evaluation of the Percor percutaneous intraaortic balloon: Cooperative study of 722 cases. Circulation 1982;66(suppl 1):223-6.
DISCUSSION
Dr. David C. Brewster (Boston, Mass.). Since its introduction in 1968, intra-aortic balloon counterpulsation has proved to be an effective means of support for the C~ling myocardium in a variety of clinical circumstances. However, complications--most commonly ipsilateral limb ischemia--are not rare and often require the attention of the vascular surgeon. Our own experience at the Massachusetts General Hospital through 1981 was reviewed and reported by Perler. He found an overall incidence of 5% ofischemia occurring after open or so-called surgical balloon insertion. Since then, as demonstrated by the series reported by Dr. Friedell et al., percutaneous methods of balloon insertion have increased substantially and with them I believe the frequency of ischemic complications, so that now we estimate approximately 10% of balloon pump patients may manifest some degree of limb ischemia. Such problems are more common in female patients with small arteries. Distal pulse deficits caused by partial obturation of small or diseased arteries by the balloon catheter are more common but often do not require surgical intervention. However, more serious limb-threatening ischemia in the truly balloon-dependent patient is much more problematic. C ;r own approach in the preoperative cardiac patient is often toward early, prompt cardiac operation and rapid weaning from the balloon in the early postoperative period. If this is not possible, we have most often chosen to transfer the balloon to the contralateral limb and restore circulation of the ischemic side by thromboembolectomy, local repair of elevated arterial plaques, and so forth. We have rarely used femorofemoral grafts in this capacity, but this approach may be beneficial in some of the circumstances outlined by Dr. Friedell and his colleagues. Certainly their good results suggest this may be a very valuable alternative. Dr. Friedell, if the contralateral iliac artery is capable of serving as the donor side of a crossover graft, why not simply transfer the balloon pump to that side) Second, and perhaps most important, we all agree that assessment of the severity of ischemia in these vasoconstricted patients, who are often on a respirator and unable to communicate, is indeed often extremely difficult. Can you elaborate on guidelines and criteria that you and your associates use to determine when surgical intervention is mandatory? Dr, Nathan P. Couch (Boston, Mass.). Despite the
FFGs for lower limb isehemia caused by IABPs 185
i0. Collier PE, Liebler GA, Park SB, Burkhotder JA, Mahcr TD, Magovern GJ. Is percutmaeous insertion of the intra-aortic balloon pump through the femoral arteu the safest technique? J VASe SUgG 1986;3:629-34. 11. Gold JP, Cohen J, Shemin RJ, et al. Femorofemoral bypass to relieve acute leg ischemia during intra-aorric balloon pump cardiac support. J VASe SURG 1986;3:35i-4.
best laid plans and in the best hands, "surge U still can beget more surgery." This is one of those examples. These authors deserve high praise for the low incidence of late leg symptoms they have achieved in these 1454 patients receiving IABPs, only 80 of whom required the femorofemoral bypass grafts. In the final analysis, 14 of the patients had some degree of late arterial occlusive disease but not necessarily all related to the balloon pump. This should be said in the context of a 64% mortality ratc for the entire group of 80 patients who had severe ischemia, at least severe enough to require femorofemoral grafting. This finding traderscores an extremely important point: a large part of the leg ischemia--at least in two thirds of the patients--was probably due to "cardiac pump failure," with just incidental but still important regional ischemia. Were any of the deaths in these patients those suspected of being due to "reflow toxicity," that is, sudden flooding of the general circulation by potassium and lactic acid? In other words, how many patients died within a few hours of declamping the graft with precipitous drops in blood pressure or cardiac output, or with acute savage arrhythmias? The second question is suggested by the fact that only 6 of the 28 patients available for follow-up required the IABP for assistance during the bypass. Has this indication for IABP been further reduced as cardioptegic methods have improved? Finally, can the caliber of the IABP catheter be reduced still further--possibly to 7 or 8 Ft.--to minimize the chance for critical iliofemoral artery- occlusion? We certainly agree that there are far more of these problems in women, especially small older women. Dr. Friedell (closing). I thank the discussants. In response to Dr. Couch's questions, the frequency of balloon pump use has been about the same during the years for weaning from cardiopulmonary bypass. Regarding the number of immediate postoperative deaths occurring af?er placement of the crossover graft, there were none. As to catheter size, if cardiac support were as efficacious with the smaller catheter, it would certainly be an advantage in reducing the occurrence ofcatheter-rdated arterial occlusion. Dr. Brewster, early in our experience, we did transfer the balloon from one side to the other but had two problems. One was that balloon support was interrupted for
186
journal of VASCULAR SURGERY
Friedell et al.
that period of time required for the transfer and, depending on technical problems with replacement, the time duration could be substantial. We also fotmd frequently that the contralateral side had atherosclerotic disease similar to the side of original balloon catheter placement. Therefore, we went directly to the femorofemoral graft as our preferred method to handle the problem.
In evaluating the sedated patient who requires mechanical ventilation, particularly the postoperative patient, we adopted a policy several years ago of the careful use of Doppler monitoring. Immediately after operation we are satisfied with a popliteal Doppler signal. However, if all pulses are absent, this is an emergent situation that requires a crossover graft.
B O U N D VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the JOURNAL OF VASCULAR SURGERY for 1987 arc available to subscribers only. They may be purchased from the publisher at a cost of S44.00 ($58.00 international) for VoL 5 (January to June) and Vol. 6 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Circulation Fulfillment, The C. V. Mosby Company, 11830 Westline Industrial Drive, St. Louis, MO 63146, USA. In the United States call toll free: (800)325-4177, ext. 351. In Missouri call collect: (314)872-8370, ext. 351. Subscriptions must be in force to quailS7. Bound volumes are not available in place of a regular JOURNAL subscription.