Vascular complications of cardiac catheterization Richard L. McCann, M D , Lewis B. Schwartz, M D , and Karen S. Pieper, MS, Durham, N. C. Femoral artery injury after diagnostic and therapeutic cardiac catheterization is a frequently encotmtered clinical problem. This study reviews 150 femoral artery injuries that occurred in 16,350 patients over a 5-year period. A prospective computerized cardiac catheterization data bank was used to determine significant risk factors associated with the occurrence of vascular injury. Logistic regression analysis revealed that congestive heart failure, female gender, and percutaneous transluminal coronary angioplasty or valvuloplasty procedure were significantly associated with occurrence of vascular injury. Greater age, smaller body surface area, and smaller weight were also associated with injury. Recommendations for management include an aggressive approach to obstruction even if local vascular reconstruction is required. False aneurysms are studied by ultrasonography, and small ones may be observed expectantly. Larger aneurysms and aneurysms that persist beyond a few days are managed by use of general or regional anesthesia and by gaining proximal control of the iliac artery. Arteriovenous fis~las resulting from catheterization are rarely hemodynamically significant and often close spontaneously. Thus repair should be delayed. (l Vase Sva~ 1991;14:375-81.)
With the introduction and widespread use of fibrinolytic therapy for acute myocardial ischemia and greater use of percutaneous transluminal coronary angioplasty, a marked increase has taken place in the number of iatrogenic peripheral vascular injuries in patients undergoing diagnostic and therapeutic cardiac catheterization.~'2 The population undergoing cardiac catheterization has a high incidence Of systemic atherosclerotic disease, including severe peripheral vascular disease. ylany patients undergoing catheterization are given anticoagulants or treated with fibrinolytic agents or both. A significant number of these procedures are performed under urgent or frankly emergency conditions, often in patients whose condition is unstable. Therapeutic catheterization may be lengthy and often requires repeated access over a period of time. All of these factors predispose to a significant, and to some extent unavoidable, number of peripheral vascu-
From the Departments of Surgery (Drs. McCann and Schwartz), and Medicine, Division of Cardiology (Ms. Pieper), Duke University Medical Center, Durham. Presented at the Fifteenth Annual Meeting of the Southern Association for Vascular Surgery, Palm Springs, Calif., 1an. 23-26, 1991. Reprint requests: Richard L. McCann, MD, P. O. Box 2990, Duke University Medical Center, Durham, NC 27710. 24/6/30062
lar complications associated with cardiac catheterization. Currently the preferred access site for both diagnostic and therapeutic catheterization is the femoral artery. The implications of femoral artery injury in these patients is significant. In those patients with acute myocardial infarction or unstable coronary ischemia, the anesthesia, blood loss, and general stress of a surgical procedure is potentially hazardous. Femoral artery injury may result in acute blood loss through hemorrhage or acute ischemia from intimal dissection or thrombosis. Either of these conditions may aggravate the preexisting cardiac injury. The resulting leg ischemia may be limb threatening and thus may mandate an attempt at vascular reconstruction. Furthermore, since walking exercise is frequently an important component of cardiac rehabilitation, even non-limb-threatening ischemia can be very important in this population if it limits their ability to exercise. Therefore efficient management of vascular complications of cardiac catheterization procedures can contribute significantly to the total care of these patients.
PATIENTS A N D M E T H O D S From January 1985 until December 1989, 150 patients required 152 operative interventions for a vascular complication directly related to diagnostic or 375
Journal of VASCULAR SURGERY
376 21/lcCann, Schwartz, and Pieper
Table I. Vascular injury rates 1985
1986
1987
1988
1989
No. of catheterizations No. of injuries Rate
2958 20 0.7%
4793 26 0.5%
5387 36 0.7%
5735 37 0.6%
5799 29 0.5%
No. of patients No. of patients injured Rate
2695 20 0.7%
3580 36 1.0%
3463 36 1.0%
3223 29 0.9%
X2 = 2.15
p = NS 3389 25 0.7%
×2= 2.96
therapeutic cardiac catheterization. One hundred forty-six of these patients were 12 years of age or older and comprised the group for statistical analysis. Simple hematomas not requiring vascular repair and other injuries exclusive of the vascular system were not considered in this review. During this period 24,820 cardiac catheterizations in 16,350 patients were prospectively recorded in the Duke Cardiovascular Data Bank. This data base was used to determine risk factors associated with vascular injury requiring surgical repair. The surgical data were reviewed to determine important factors in the management of these vascular injuries. The types of injuries encountered, the diagnostic evaluation, and the outcome have been tabulated to formulate a cohesive approach to this increasingly encountered clinical problem. STATISTICAL M E T H O D S The univariate and joint relationships between the clinical factors of interest and the two outcomes, vascular complications and in-hospital death, were evaluated by means of logistic regression models (SAS, Cary, N.C.). Stepwise regression techniques were used to determine which set of variables jointly best predict vascular complications or in-hospital death. All categorical variables are reported as N or percents. All continuous variables are reported as 25th percentile/50th percentile/75th percentile. RESULTS
The risk of requiring surgical repair for vascular injury during this period was 0.60% per catheterization or 0.89% per patient. The incidence by year computed per patient and per catheterization is shown in Table I. The incidence did not vary significantly during the period of this review. Factors from the cardiac catheterization data base that might be associated with the occurrence of a vascular complication are listed in Table II. Significant risk factors in this analysis include history of peripheral
p =NS
vascular disease, female gender, percutaneous transluminal coronary angioplasty, valvuloplasty and congestive heart failure. Advanced age, lighter weight, shorter height, and smaller body surface area were significantly associated with the occurrence of vascular complication from cardiac catheterization. In a multivariable model, only age, congestive heart failure, and body surface area are significant variables. Once these three factors have been adjusted for, no other variable contains significant additional predictive value. Epidemiologic variables associated with atherosclerotic vascular disease were distributed equally between patients in whom vascular complications developed and those who avoided them. These include diabetes, hyperlipidemia, history of smoking, hypertension, family history of heart disease, and history of significant cerebrovascular disease. Also noted from the univariate analysis was the suggestion that patients with more severe manifestations of atherosclerotic disease were more likely to suffer vascular complications from cardiac catheterization. These include the presence of congestive heart failure and the requirement for therapeutic catheterization through percutaneous angioplasty or valvuloplasty. Older patients were more likely to suffer vascular injury as were patients who were smaller and shorter (perhaps cachectic). These concepts were confirmed by the multivariable analysis that demonstrated that age, congestive heart failure and smaller body surface area were significant variables. Four types of vascular procedures were required, and these are analyzed separately to determine the significant clinical factors associated with each (Table III). Fifty-six patients required simple Fogarty thrombectomy or embolectomy for acute thrombosis of the access vessel or distal embolization from this site or from the catheter itself. Acute false aneurysm at the puncture site required femoral artery repair in 64 patients. Formal vascular reconstruction with a graft or patch, usually of venous material harvested locally, was required in 18 patients, and arteriove-
Volume 14 Number 3 September 1991
Vascular complications of cardiac catheterization
377
Table II. A. Univariate analysis of patients undergoing cardiac catheterization With vascular complications (%)
Discrete variables Diabetes Hyperlipidemia
Without vascular complications (%)
19 32 60 55 55 9 35 19 60 34
Smoker
Hypertension Family history of heart disease History of cerebral vascular disease Congestive heart failure History of peripheral vascular disease Female
Percutaneous transhiminal coronary angioplasty Valvuloplasty
p value
19 35 66 51 51 8 15 11 31 24
9
0.953 0.429 0.172 0.417 0.320 0.633 < 0.001 0.008 < 0.001 < 0.001
0.4
< 0.001
Table II. B. Univariate analysis of patients undergoing cardiac catheterization Continuous variables Age (yr) Body surface area (m2) Weight (kg) Height (cm)
25th/50th/75th
25th/50th/75th
percentile
percentile
51/60/66 1.8/1.9/2.1
57/65/71
1.6/1.7/1.9 56/66/80
< 0.001 < 0.001 < 0.001 < 0.001
69/79/89 165/173/179
160/163/173
Table III. Significant clinical factors associated with each procedure Thromboembolectomies No.
Gender (% female) Diabetes (%) Myocardial infarction (%) Lyric Rx (%)
Percutaneous transluminal coronary angioplasty (%) Emergency (%) Mortality rate (%) Age (25th/50th/75th) Ejection fraction
False aneurysm
Graft~patch
Arteriovenous fistula
Total
56 57 21 48 18 21
64 66 14 28 22 33
18 50 28 72 39 72
8 38 25 25 13 50
146 60 19 41 22 34
88 32
83 3 61/68/77 0.45/0.55/0.63
83 6 51/60/64 0.40/0.51/0.60
0 13 51/61/70 0.39/0.43/0.55
80 15
53/64/69 0.29/0.44/0.53
nous fistulas that were surgically ligated developed in eight patients. The clinical factors associated with each of these surgical procedures are listed in Table III. Age and gender were similar among the four groups. Approximately 20% of the patients had diabetes, and no significant difference in mean ejection fraction was determined at catheterization. A suggestion was made that patients requiring formal vascular repair had more severe cardiovascular disease requiring more aggressive therapy in that 65% of these patients were being treated for acute myocardial infarction. In addition, 35% required fibrinolytic therapy, and 65% underwent percutaneous transluminal coronary. The catheterization was classified as emergent in 85% of the patients requiring formal vascular repair.
The hospital mortality rate of 15 % is indicative of the significance of systemic vascular disease present in this population. Statistically significant factors associated with in-hospital deaths included ejection fraction, urgent surgery, the use of lytic therapy, and acute myocardial infarction. It is significant to note that all deaths were directly attributable to the cardiac disease mandating catheterization. Neither the surgical repair of vascular injury nor the anesthesia required for this surgery contributed to any of the deaths. Five deaths occurred in patients who have had more than local anesthesia. These occurred an average of 18 days after operation, and in none did the operative procedure appear to influence the hospital course.
Journal of VASCULAR SURGERY
378 McCann, Schwartz, and Pieper
Fig. 1. A, (left)Groin duplex ultrasonography with color flow Doppler imaging 2 days after cardiac catheterization. There is a relatively long passage (singlearrow) leading from the femoral artery to the cavity (arrowhead). Flow in this passage is indicated by the blue color. 13, (right) Follow-up 2 days later after treatment with only reduced activity shows no flow within the passage (double arrow) and complete thrombosis of the mass. All formal vascular reconstructions were successful at limb salvage, and all arteriovenous fistulas were ligated with maintenance of vessel patency in all arteries and all but one vein. One patient with repair of an acute false aneurysm had recurrence of the lesion 2 weeks after operation requiring repeat repair. One patient required amputation after Fogarty thrombectomy failed to restore adequate circulation to the limb. This patient subsequently died of her severe coronary artery disease during the same hospitalization. DISCUSSION
This study demonstrates that peripheral vascular injury requiring surgical repair remains a frequent compl!cation of diagnostic and therapeutic cardiac catheterization. Previous studies have shown a similar incidence of 0.5% to 1.0%. T M Despite the potential cardiovascular instability of these patients, vascular repair of groin injuries is well tolerated and should be aggressively pursued to maximize limb salvage and function. In contrast to previous series, the present study found a similar incidence of thrombotic or obstructive complications compared
to bleeding or pseudoaneurysm formation. This may be related to our aggressive approach, particularly with respect to lilac thrombosis, which may produce non-limb-threatening ischemia but which may jeopardize the function of the limb and thus the rehabilitation of the patient. The techniques for treatment of patients with peripheral vascular injuries from cardiac catheterization have evolved during the years of this review. Acute ischemia of the leg caused by thrombosis at the puncture site or embolism from the catheter is usually clinically obvious. An accurate and complete vascular examination before the catheterization is very helpful in determining whether change has occurred. Occlusion of the iliac artery usually results in a pale and pulseless foot but only rarely produces rest pain. If not treated appropriately, however, short distance claudication is likely to result as well as loss of this side as an access site should repeat catheterization be necessary in the future. If a patient's superficial femoral artery becomes occluded, symptoms are often more severe. This may occur if the puncture site is distal to the femoral bifurcation. The most severe ischemia occurs in patients with preexisting superfi
Volume 14 Number 3 September 1991
Vascular complications of cardiac catheterization 379
f
I I i ~ U la I ( t i i l y { : l l i1~1 IL
femoral
Talse
aneurysm
Fig. 2. Drawing demonstrates the approach to treatment of patients with acute femoral false aneurysms. A short transverse lower abdominal incision is made, and the distal external iliac artery is exposed just proximal to the inguinal ligament, and a vascular clamp is applied for proximal control. Next, incision is made directly over the palpable mass and the aneurysm entered and back-bleeding controlled with a digital pressure. This method avoids excessiveblood loss if control of the vessel is not obtained before entering the aneurysm.
cial femoral artery occlusion in whom the deep femoral artery is occluded by thrombus or local intimal dissertion after catheterization. These patients often are admitted with severe, acute, painful ischemia requiring an emergency procedure. With careful physical examination and knowledge of the status before the catheterization, angiography is seldom, if ever, needed. Groin exploration under local anesthesia is performed promptly if ischemia is acute and limb threatening and, more electively, for lesser degrees ofischemia. The puncture site is always exposed, and it is often convenient to enlarge this opening transversely and use it as the access site for passage of a Fogarty catheter or for inspection of the luminal surface. If local intimal dissection has occurred contributing to the thrombosis, then endarterectomy with or without vein patch angioplasty should be considered. If a significant local injury has occurred, simple thrombectomy may be ineffective and may result only in repeat thrombosis as soon as anticoagulation is stopped after operation. To obtain adequate vascular repair it is often necessary to extend the dissection to the distal external iliac by dividing the inguinal ligament. This is preferable to a femorofemoral crossover graft because the latter will make subsequent catheterization difficult. These procedures are often performed with the patient under local anesthesia and saphenous vein segments being
harvested from the local area. A synthetic graft may be used if local vein is not available, but the potential for infection is high in thesc traumatized wounds; often significant hematoma and injury to the local tissues predisposes to infection. When the puncture site fails to seal, a perivascular hematoma forms with a central area that remains fluid. This has traditionally been referred to as a false aneurysm, but since there is no fibrous aneurysmal wall, only a rim ofthrombus, this lesion might better be termed a pulsating hematoma. It is the persistence of the liquid blood in the center that distinguishes these lesions from bland, clinically insignificant perivascular hematoma. We think that failure of the puncture site to seal is often related to improper or difficult compression technique because of body habitus or inaccurate puncture of a vessel other than the common femoral. These lesions may be distinguished from simple hematoma on clinical evaluation, because whereas a hematoma may transmit a pulse, these lesions are perceptibly pulsatile in a radial direction on palpation of the lateral edges of the mass. The clinical impression can easily and accurately be confirmed by color flow Doppler and duplex ultrasound examination. This evaluation has proved to be more accurate and helpful than other radiographic imaging modalities including angiography. 5"6 We recommend this evaluation for all of these lesions
380 McCann, Schwartz, and Pieper
Fig. 3. Duplex ultrasound and color flow Doppler imaging of an arteriovenous fistula. Continuous flow in the artery (top) is shown and flow in the vein (bottom) from the site of the fistula toward the heart. The Doppler signal shows diastolic augmentation of the flow waveform. because it helps to direct surgical therapy. 5 Many of these lesions recognized early after catheterization will seal spontaneously and will not need surgical therapy. Sonographic characteristics suggesting spontaneous resolution include small size of the fluid center and a long "neck" or passage between the puncture site and the cavity. Fig. 1 shows such a lesion with a relatively long passage between the surface of the arterial wall and the fluid cavity. This lesion spontaneously thrombosed with observation within 48 hours, and no surgical therapy was required. The risk of these lesions is that the outer
Journal of VASCULAR SURGERY
shell of hematoma will break down, resulting in frank rupture into the tissues of the leg or the retroperitoneum. During the period of this study we know of three patients with unrecognized false ancurysms that ruptured, two of which were fatal. Our current recommendations for small lesions are reduced activity, serial ultrasound evaluation, and repair if resolution or improvement is not evident after 3 or 4 days of observation. Indications for surgical repair include presence of a lesion so large that pressure necrosis of the skin is threatened, continued transfusion requirement, or failure to resolve after a period of observation. Our preferred technique for repair of these lesions is different from that previously reported.~ Some have recommended direct entry to the cavity with digital control of the defect in the artery as the most direct method of approaching these lesions. However, it is our feeling that acute blood loss may be poorly tolerated by this patient population with limited cardiac reserve. We prefer preliminary proximal control of the distal external iliac artery through a separate transverse abdominal wall incision (Fig. 2). Only after proximal control is achieved is the aneurysm entered. M'ter evacuation of the hematoma, the slight back-bleeding is easily controlled with digital pressure until the vessel is sufficiently exposed, so that the defect in the vessel can be sutured. Usually not more than one or two sutures are required. Care must be taken to place these sutures in the vessel wall rather than in what appears to be the wall of the false aneurysm, or recurrence is likely. Because of the hematoma mass, the requirement for two incisions, and often the large size of the patient, local anesthesia is seldom effective, and we have insisted on regional or general anesthesia. This has proved to be safe even in the face of acute myocardial ischemia and infarction but does often require invasive monitoring. Our experience with arteriovenous fistula is also contrary to that previously reported. 1,2 These lesions occur when both vessels arc punctured simultaneously. The fistulas usually are small and not hemodynamically significant, at least in the immediate postcatheterization period. Physical findings may be minimal consisting of a thrill or bruit that was not present before the catheterization. Again, the ultrasound examination is extremely helpful in making this diagnosis (Fig. 3). Most of these fistulas will close spontaneously, and, because they are not hemodynamically significant, we recommend delay of at least 6 weeks before considering repair. If the fistula does not resolve spontaneously during that period, however, repair may be considered to prevent progression
Volume 14 Number 3 September 1991
Vascular complications of cardiac catheterization 381
o f the shunt to h e m o d y n a m i c significance and to obviate the potential for local infection at the site o f the fistula.
REFERENCES 1. Oweida SW, Roubin GS, Smith RB, Salam AA. Postcatheterization vascular complications associated with percutaneous transluminal coronary angioplasty. J VAse SURG 1990;12: 310-5. 2. Babu SC, Piccorelli GO, Shah PM, Stein lit, Clauss RH. Incidence and results of arterial complications among 16,350 patients undergoing cardiac catheterization. J VAsc SURG i989;10:113-6. 3. Kaufman I, Moglia R, Lacy C, Dinerstein C, Moreyra A.
Peripheral vascular compfications from percutaneous transluminal coronary angioplasty: a comparison with transfemoral cardiac catheterization. Am J Med Sci 1989;297:22-5. 4. Lilly MP, Reichman W, Sarazen AA Jr, Carney WI Jr. Anatomic and cfinical factors associated with complications of transfemoral arteriography. Ann Vase Surg 1990;4:264-9. 5. Sheikh KH, Adams DB, McCann RL, Lyerly HK, Sabiston DC, Kisslo l- Utility of Doppler color flow imaging for identification of femoral arterial complications of cardiac catheterization. Am Heart J 1989;117:623-8. 6. Hayek ME, Ludwig MA, Fischer K, Sisler C. The use of scintiangiography with technetium 99m in the diagnosis of traumatic pseudoaneurysm. J VAsc SURG 1988;7:409-13. Submitted Feb. 2, 1991; accepted Apr. 9, 1991.
SOCIETY FOR VASCULAR SURGERY LIFELINE FOUNDATION GRANT AWARD The Lifeline Foundation of the Society for Vascular Surgery invites grant applications for fimding of meritorious research by young surgical investigators. The awards are intended for surgeons who have completed their formal surgical education in general surgery and who have completed or are in an advanced training program in vascular surgery. To be considered for selection a candidate: 1. Should be certified by the American Board of Surgery or have completed the requirements for certification 2. Should submit an application within three years of completion of an approved residency training program 3. Must have either a faculty appointment in an approved medical school in the United States or Canada or have received an academic appointment within the guidelines of the applicant's institution Grant awards are not intended to supplement salary, which will remain the responsibility of the institution in which the awardee holds an appointment. The awardee is expected to devote a significant amount of time to the funded project. A progress report will be presented by the investigators during the annual meeting of the Society for Vascular Surgery. A grant awards committee will review competitive applications. It is anticipated that two grants will be awarded annually totaling $50,000 each to include indirect costs. Each award will be for one year with the option to extend for an additional year. Grant applications may be obtained from: The Lifeline Foundation Society for Vascular Surgery Thirteen Elm St. Manchester, MA 01944
The deadline for receiving applications in the Foundation office is January 15, 1992. Funds will be awarded by July 1, 1992.
With the introduction and widespread use of fibrinolytic therapy for acute myocardial ischemia and greater use of percutaneous transluminal coronary angioplasty, a marked increase has taken place in the number of iatrogenic peripheral vascular injuries in patients undergoing diagnostic and therapeutic cardiac catheterization.~'2 The population undergoing cardiac catheterization has a high incidence Of systemic atherosclerotic disease, including severe peripheral vascular disease. ylany patients undergoing catheterization are given anticoagulants or treated with fibrinolytic agents or both. A significant number of these procedures are performed under urgent or frankly emergency conditions, often in patients whose condition is unstable. Therapeutic catheterization may be lengthy and often requires repeated access over a period of time. All of these factors predispose to a significant, and to some extent unavoidable, number of peripheral vascu-
From the Departments of Surgery (Drs. McCann and Schwartz), and Medicine, Division of Cardiology (Ms. Pieper), Duke University Medical Center, Durham. Presented at the Fifteenth Annual Meeting of the Southern Association for Vascular Surgery, Palm Springs, Calif., 1an. 23-26, 1991. Reprint requests: Richard L. McCann, MD, P. O. Box 2990, Duke University Medical Center, Durham, NC 27710. 24/6/30062
lar complications associated with cardiac catheterization. Currently the preferred access site for both diagnostic and therapeutic catheterization is the femoral artery. The implications of femoral artery injury in these patients is significant. In those patients with acute myocardial infarction or unstable coronary ischemia, the anesthesia, blood loss, and general stress of a surgical procedure is potentially hazardous. Femoral artery injury may result in acute blood loss through hemorrhage or acute ischemia from intimal dissection or thrombosis. Either of these conditions may aggravate the preexisting cardiac injury. The resulting leg ischemia may be limb threatening and thus may mandate an attempt at vascular reconstruction. Furthermore, since walking exercise is frequently an important component of cardiac rehabilitation, even non-limb-threatening ischemia can be very important in this population if it limits their ability to exercise. Therefore efficient management of vascular complications of cardiac catheterization procedures can contribute significantly to the total care of these patients.
PATIENTS A N D M E T H O D S From January 1985 until December 1989, 150 patients required 152 operative interventions for a vascular complication directly related to diagnostic or 375
Journal of VASCULAR SURGERY
376 21/lcCann, Schwartz, and Pieper
Table I. Vascular injury rates 1985
1986
1987
1988
1989
No. of catheterizations No. of injuries Rate
2958 20 0.7%
4793 26 0.5%
5387 36 0.7%
5735 37 0.6%
5799 29 0.5%
No. of patients No. of patients injured Rate
2695 20 0.7%
3580 36 1.0%
3463 36 1.0%
3223 29 0.9%
X2 = 2.15
p = NS 3389 25 0.7%
×2= 2.96
therapeutic cardiac catheterization. One hundred forty-six of these patients were 12 years of age or older and comprised the group for statistical analysis. Simple hematomas not requiring vascular repair and other injuries exclusive of the vascular system were not considered in this review. During this period 24,820 cardiac catheterizations in 16,350 patients were prospectively recorded in the Duke Cardiovascular Data Bank. This data base was used to determine risk factors associated with vascular injury requiring surgical repair. The surgical data were reviewed to determine important factors in the management of these vascular injuries. The types of injuries encountered, the diagnostic evaluation, and the outcome have been tabulated to formulate a cohesive approach to this increasingly encountered clinical problem. STATISTICAL M E T H O D S The univariate and joint relationships between the clinical factors of interest and the two outcomes, vascular complications and in-hospital death, were evaluated by means of logistic regression models (SAS, Cary, N.C.). Stepwise regression techniques were used to determine which set of variables jointly best predict vascular complications or in-hospital death. All categorical variables are reported as N or percents. All continuous variables are reported as 25th percentile/50th percentile/75th percentile. RESULTS
The risk of requiring surgical repair for vascular injury during this period was 0.60% per catheterization or 0.89% per patient. The incidence by year computed per patient and per catheterization is shown in Table I. The incidence did not vary significantly during the period of this review. Factors from the cardiac catheterization data base that might be associated with the occurrence of a vascular complication are listed in Table II. Significant risk factors in this analysis include history of peripheral
p =NS
vascular disease, female gender, percutaneous transluminal coronary angioplasty, valvuloplasty and congestive heart failure. Advanced age, lighter weight, shorter height, and smaller body surface area were significantly associated with the occurrence of vascular complication from cardiac catheterization. In a multivariable model, only age, congestive heart failure, and body surface area are significant variables. Once these three factors have been adjusted for, no other variable contains significant additional predictive value. Epidemiologic variables associated with atherosclerotic vascular disease were distributed equally between patients in whom vascular complications developed and those who avoided them. These include diabetes, hyperlipidemia, history of smoking, hypertension, family history of heart disease, and history of significant cerebrovascular disease. Also noted from the univariate analysis was the suggestion that patients with more severe manifestations of atherosclerotic disease were more likely to suffer vascular complications from cardiac catheterization. These include the presence of congestive heart failure and the requirement for therapeutic catheterization through percutaneous angioplasty or valvuloplasty. Older patients were more likely to suffer vascular injury as were patients who were smaller and shorter (perhaps cachectic). These concepts were confirmed by the multivariable analysis that demonstrated that age, congestive heart failure and smaller body surface area were significant variables. Four types of vascular procedures were required, and these are analyzed separately to determine the significant clinical factors associated with each (Table III). Fifty-six patients required simple Fogarty thrombectomy or embolectomy for acute thrombosis of the access vessel or distal embolization from this site or from the catheter itself. Acute false aneurysm at the puncture site required femoral artery repair in 64 patients. Formal vascular reconstruction with a graft or patch, usually of venous material harvested locally, was required in 18 patients, and arteriove-
Volume 14 Number 3 September 1991
Vascular complications of cardiac catheterization
377
Table II. A. Univariate analysis of patients undergoing cardiac catheterization With vascular complications (%)
Discrete variables Diabetes Hyperlipidemia
Without vascular complications (%)
19 32 60 55 55 9 35 19 60 34
Smoker
Hypertension Family history of heart disease History of cerebral vascular disease Congestive heart failure History of peripheral vascular disease Female
Percutaneous transhiminal coronary angioplasty Valvuloplasty
p value
19 35 66 51 51 8 15 11 31 24
9
0.953 0.429 0.172 0.417 0.320 0.633 < 0.001 0.008 < 0.001 < 0.001
0.4
< 0.001
Table II. B. Univariate analysis of patients undergoing cardiac catheterization Continuous variables Age (yr) Body surface area (m2) Weight (kg) Height (cm)
25th/50th/75th
25th/50th/75th
percentile
percentile
51/60/66 1.8/1.9/2.1
57/65/71
1.6/1.7/1.9 56/66/80
< 0.001 < 0.001 < 0.001 < 0.001
69/79/89 165/173/179
160/163/173
Table III. Significant clinical factors associated with each procedure Thromboembolectomies No.
Gender (% female) Diabetes (%) Myocardial infarction (%) Lyric Rx (%)
Percutaneous transluminal coronary angioplasty (%) Emergency (%) Mortality rate (%) Age (25th/50th/75th) Ejection fraction
False aneurysm
Graft~patch
Arteriovenous fistula
Total
56 57 21 48 18 21
64 66 14 28 22 33
18 50 28 72 39 72
8 38 25 25 13 50
146 60 19 41 22 34
88 32
83 3 61/68/77 0.45/0.55/0.63
83 6 51/60/64 0.40/0.51/0.60
0 13 51/61/70 0.39/0.43/0.55
80 15
53/64/69 0.29/0.44/0.53
nous fistulas that were surgically ligated developed in eight patients. The clinical factors associated with each of these surgical procedures are listed in Table III. Age and gender were similar among the four groups. Approximately 20% of the patients had diabetes, and no significant difference in mean ejection fraction was determined at catheterization. A suggestion was made that patients requiring formal vascular repair had more severe cardiovascular disease requiring more aggressive therapy in that 65% of these patients were being treated for acute myocardial infarction. In addition, 35% required fibrinolytic therapy, and 65% underwent percutaneous transluminal coronary. The catheterization was classified as emergent in 85% of the patients requiring formal vascular repair.
The hospital mortality rate of 15 % is indicative of the significance of systemic vascular disease present in this population. Statistically significant factors associated with in-hospital deaths included ejection fraction, urgent surgery, the use of lytic therapy, and acute myocardial infarction. It is significant to note that all deaths were directly attributable to the cardiac disease mandating catheterization. Neither the surgical repair of vascular injury nor the anesthesia required for this surgery contributed to any of the deaths. Five deaths occurred in patients who have had more than local anesthesia. These occurred an average of 18 days after operation, and in none did the operative procedure appear to influence the hospital course.
Journal of VASCULAR SURGERY
378 McCann, Schwartz, and Pieper
Fig. 1. A, (left)Groin duplex ultrasonography with color flow Doppler imaging 2 days after cardiac catheterization. There is a relatively long passage (singlearrow) leading from the femoral artery to the cavity (arrowhead). Flow in this passage is indicated by the blue color. 13, (right) Follow-up 2 days later after treatment with only reduced activity shows no flow within the passage (double arrow) and complete thrombosis of the mass. All formal vascular reconstructions were successful at limb salvage, and all arteriovenous fistulas were ligated with maintenance of vessel patency in all arteries and all but one vein. One patient with repair of an acute false aneurysm had recurrence of the lesion 2 weeks after operation requiring repeat repair. One patient required amputation after Fogarty thrombectomy failed to restore adequate circulation to the limb. This patient subsequently died of her severe coronary artery disease during the same hospitalization. DISCUSSION
This study demonstrates that peripheral vascular injury requiring surgical repair remains a frequent compl!cation of diagnostic and therapeutic cardiac catheterization. Previous studies have shown a similar incidence of 0.5% to 1.0%. T M Despite the potential cardiovascular instability of these patients, vascular repair of groin injuries is well tolerated and should be aggressively pursued to maximize limb salvage and function. In contrast to previous series, the present study found a similar incidence of thrombotic or obstructive complications compared
to bleeding or pseudoaneurysm formation. This may be related to our aggressive approach, particularly with respect to lilac thrombosis, which may produce non-limb-threatening ischemia but which may jeopardize the function of the limb and thus the rehabilitation of the patient. The techniques for treatment of patients with peripheral vascular injuries from cardiac catheterization have evolved during the years of this review. Acute ischemia of the leg caused by thrombosis at the puncture site or embolism from the catheter is usually clinically obvious. An accurate and complete vascular examination before the catheterization is very helpful in determining whether change has occurred. Occlusion of the iliac artery usually results in a pale and pulseless foot but only rarely produces rest pain. If not treated appropriately, however, short distance claudication is likely to result as well as loss of this side as an access site should repeat catheterization be necessary in the future. If a patient's superficial femoral artery becomes occluded, symptoms are often more severe. This may occur if the puncture site is distal to the femoral bifurcation. The most severe ischemia occurs in patients with preexisting superfi
Volume 14 Number 3 September 1991
Vascular complications of cardiac catheterization 379
f
I I i ~ U la I ( t i i l y { : l l i1~1 IL
femoral
Talse
aneurysm
Fig. 2. Drawing demonstrates the approach to treatment of patients with acute femoral false aneurysms. A short transverse lower abdominal incision is made, and the distal external iliac artery is exposed just proximal to the inguinal ligament, and a vascular clamp is applied for proximal control. Next, incision is made directly over the palpable mass and the aneurysm entered and back-bleeding controlled with a digital pressure. This method avoids excessiveblood loss if control of the vessel is not obtained before entering the aneurysm.
cial femoral artery occlusion in whom the deep femoral artery is occluded by thrombus or local intimal dissertion after catheterization. These patients often are admitted with severe, acute, painful ischemia requiring an emergency procedure. With careful physical examination and knowledge of the status before the catheterization, angiography is seldom, if ever, needed. Groin exploration under local anesthesia is performed promptly if ischemia is acute and limb threatening and, more electively, for lesser degrees ofischemia. The puncture site is always exposed, and it is often convenient to enlarge this opening transversely and use it as the access site for passage of a Fogarty catheter or for inspection of the luminal surface. If local intimal dissection has occurred contributing to the thrombosis, then endarterectomy with or without vein patch angioplasty should be considered. If a significant local injury has occurred, simple thrombectomy may be ineffective and may result only in repeat thrombosis as soon as anticoagulation is stopped after operation. To obtain adequate vascular repair it is often necessary to extend the dissection to the distal external iliac by dividing the inguinal ligament. This is preferable to a femorofemoral crossover graft because the latter will make subsequent catheterization difficult. These procedures are often performed with the patient under local anesthesia and saphenous vein segments being
harvested from the local area. A synthetic graft may be used if local vein is not available, but the potential for infection is high in thesc traumatized wounds; often significant hematoma and injury to the local tissues predisposes to infection. When the puncture site fails to seal, a perivascular hematoma forms with a central area that remains fluid. This has traditionally been referred to as a false aneurysm, but since there is no fibrous aneurysmal wall, only a rim ofthrombus, this lesion might better be termed a pulsating hematoma. It is the persistence of the liquid blood in the center that distinguishes these lesions from bland, clinically insignificant perivascular hematoma. We think that failure of the puncture site to seal is often related to improper or difficult compression technique because of body habitus or inaccurate puncture of a vessel other than the common femoral. These lesions may be distinguished from simple hematoma on clinical evaluation, because whereas a hematoma may transmit a pulse, these lesions are perceptibly pulsatile in a radial direction on palpation of the lateral edges of the mass. The clinical impression can easily and accurately be confirmed by color flow Doppler and duplex ultrasound examination. This evaluation has proved to be more accurate and helpful than other radiographic imaging modalities including angiography. 5"6 We recommend this evaluation for all of these lesions
380 McCann, Schwartz, and Pieper
Fig. 3. Duplex ultrasound and color flow Doppler imaging of an arteriovenous fistula. Continuous flow in the artery (top) is shown and flow in the vein (bottom) from the site of the fistula toward the heart. The Doppler signal shows diastolic augmentation of the flow waveform. because it helps to direct surgical therapy. 5 Many of these lesions recognized early after catheterization will seal spontaneously and will not need surgical therapy. Sonographic characteristics suggesting spontaneous resolution include small size of the fluid center and a long "neck" or passage between the puncture site and the cavity. Fig. 1 shows such a lesion with a relatively long passage between the surface of the arterial wall and the fluid cavity. This lesion spontaneously thrombosed with observation within 48 hours, and no surgical therapy was required. The risk of these lesions is that the outer
Journal of VASCULAR SURGERY
shell of hematoma will break down, resulting in frank rupture into the tissues of the leg or the retroperitoneum. During the period of this study we know of three patients with unrecognized false ancurysms that ruptured, two of which were fatal. Our current recommendations for small lesions are reduced activity, serial ultrasound evaluation, and repair if resolution or improvement is not evident after 3 or 4 days of observation. Indications for surgical repair include presence of a lesion so large that pressure necrosis of the skin is threatened, continued transfusion requirement, or failure to resolve after a period of observation. Our preferred technique for repair of these lesions is different from that previously reported.~ Some have recommended direct entry to the cavity with digital control of the defect in the artery as the most direct method of approaching these lesions. However, it is our feeling that acute blood loss may be poorly tolerated by this patient population with limited cardiac reserve. We prefer preliminary proximal control of the distal external iliac artery through a separate transverse abdominal wall incision (Fig. 2). Only after proximal control is achieved is the aneurysm entered. M'ter evacuation of the hematoma, the slight back-bleeding is easily controlled with digital pressure until the vessel is sufficiently exposed, so that the defect in the vessel can be sutured. Usually not more than one or two sutures are required. Care must be taken to place these sutures in the vessel wall rather than in what appears to be the wall of the false aneurysm, or recurrence is likely. Because of the hematoma mass, the requirement for two incisions, and often the large size of the patient, local anesthesia is seldom effective, and we have insisted on regional or general anesthesia. This has proved to be safe even in the face of acute myocardial ischemia and infarction but does often require invasive monitoring. Our experience with arteriovenous fistula is also contrary to that previously reported. 1,2 These lesions occur when both vessels arc punctured simultaneously. The fistulas usually are small and not hemodynamically significant, at least in the immediate postcatheterization period. Physical findings may be minimal consisting of a thrill or bruit that was not present before the catheterization. Again, the ultrasound examination is extremely helpful in making this diagnosis (Fig. 3). Most of these fistulas will close spontaneously, and, because they are not hemodynamically significant, we recommend delay of at least 6 weeks before considering repair. If the fistula does not resolve spontaneously during that period, however, repair may be considered to prevent progression
Volume 14 Number 3 September 1991
Vascular complications of cardiac catheterization 381
o f the shunt to h e m o d y n a m i c significance and to obviate the potential for local infection at the site o f the fistula.
REFERENCES 1. Oweida SW, Roubin GS, Smith RB, Salam AA. Postcatheterization vascular complications associated with percutaneous transluminal coronary angioplasty. J VAse SURG 1990;12: 310-5. 2. Babu SC, Piccorelli GO, Shah PM, Stein lit, Clauss RH. Incidence and results of arterial complications among 16,350 patients undergoing cardiac catheterization. J VAsc SURG i989;10:113-6. 3. Kaufman I, Moglia R, Lacy C, Dinerstein C, Moreyra A.
Peripheral vascular compfications from percutaneous transluminal coronary angioplasty: a comparison with transfemoral cardiac catheterization. Am J Med Sci 1989;297:22-5. 4. Lilly MP, Reichman W, Sarazen AA Jr, Carney WI Jr. Anatomic and cfinical factors associated with complications of transfemoral arteriography. Ann Vase Surg 1990;4:264-9. 5. Sheikh KH, Adams DB, McCann RL, Lyerly HK, Sabiston DC, Kisslo l- Utility of Doppler color flow imaging for identification of femoral arterial complications of cardiac catheterization. Am Heart J 1989;117:623-8. 6. Hayek ME, Ludwig MA, Fischer K, Sisler C. The use of scintiangiography with technetium 99m in the diagnosis of traumatic pseudoaneurysm. J VAsc SURG 1988;7:409-13. Submitted Feb. 2, 1991; accepted Apr. 9, 1991.
SOCIETY FOR VASCULAR SURGERY LIFELINE FOUNDATION GRANT AWARD The Lifeline Foundation of the Society for Vascular Surgery invites grant applications for fimding of meritorious research by young surgical investigators. The awards are intended for surgeons who have completed their formal surgical education in general surgery and who have completed or are in an advanced training program in vascular surgery. To be considered for selection a candidate: 1. Should be certified by the American Board of Surgery or have completed the requirements for certification 2. Should submit an application within three years of completion of an approved residency training program 3. Must have either a faculty appointment in an approved medical school in the United States or Canada or have received an academic appointment within the guidelines of the applicant's institution Grant awards are not intended to supplement salary, which will remain the responsibility of the institution in which the awardee holds an appointment. The awardee is expected to devote a significant amount of time to the funded project. A progress report will be presented by the investigators during the annual meeting of the Society for Vascular Surgery. A grant awards committee will review competitive applications. It is anticipated that two grants will be awarded annually totaling $50,000 each to include indirect costs. Each award will be for one year with the option to extend for an additional year. Grant applications may be obtained from: The Lifeline Foundation Society for Vascular Surgery Thirteen Elm St. Manchester, MA 01944
The deadline for receiving applications in the Foundation office is January 15, 1992. Funds will be awarded by July 1, 1992.
