Clinical characteristics and surgical management of vascular complications in patients undergoing cardiac catheterization: Interventional versus diagnostic procedures Louis M. Messina, M D , Thomas E. Brothers, M D , Thomas W. Wakefield, M D , Gerald B. Zelenock, M D , S. Martin Lindenauer, M D , Lazar J. Greenfield, M D , Lloyd A. Jacobs, M D , Elaine P. Fellows, BSN, Susan V. Grube, BSN, and James C. Stanley, M D , Ann Arbor, Mich. The purpose of this report is to define the clinical characteristics and outcome of surgical management of vascular complications after interventional cardiac catheterization and to contrast them to those after diagnostic cardiac catheterization. From October 1985 to December 1989, 101 patients were treated for 106 vascular complications after 1866 interventional and 5046 diagnostic cardiac catheterizations at the University of Michigan Medical Center. Interventional catheterizations resulted in 69 vascular complications in 64 patients (frequency 3.4%). The most common interventions included coronary angioplasty (34), of which 10 required percutaneous partial cardiopulmonary bypass, intraaortic balloon pump placement (14), and aortic val~atloplasty (11). Interventional catheterrelated complications included hemorrhage (33), arterial thrombosis (18), pseudoaneurysm formation (12), catheter embolization (2), thromboembolism (2), as well as arteriovenous fistula, pseudoaneurysm, and arterial dissection (1 each). Fifteen of these 69 patients (24%) had suffered acute myocardial infarction just before their catheterization. Surgical repair was performed under local anesthesia in 70% of patients. Major vascular reconstructions were required in 9% of patients. Three percent of the involved lower extremities had to be amputated because of complications occurring after arterial puncture. Eight percent of the patients incurring vascular complications after interventional procedures died after operation. Diagnostic catheterizations resulted in 37 vascular complications in 37 patients (frequency 0.7%). In contrast to diagnostic cardiac catheterization, vascular complications after interventional cardiac catheterization occurred more frequently, were most often due to hemorrhage at the vascular access site, and occurred in high-risk, critically ill patients. Refinements in catheter construction and insertion technique are necessary to reduce the frequency of these serious complications. In addition, optimal patient care mandates careful coordination between the responsible cardiologist, vascular surgeon, and anesthesiologist. (J V^sc SuR6 1991;13:593-600.)
The incidence of vascular complications after diagnostic cardiac catheterization is low, reflecting refinements in angiographic technique and catheter construction. The contemporary clinical practice of cardiology also includes sophisticated endovascular interventions to treat coronary artery and cardiac valvular heart disease and to support the failing heart. From the Section of Vascular Surgery, Department of Surgery, Universityof MichiganMedicalCenter, Ann Arbor. Presented at the Fourteenth AnnualMeeting of the Midwestern Vascular SurgicalSociety,Toledo, Ohio, Sept. 14, 1990. Reprint requests: LouisM. Messina,MD, 2210 TaubmanCenter, 1500 E. MedicalCenter Dr., Ann Arbor, MI 48109-0329. 24/6/27611
Such procedures are often lifesaving but carry a substantial risk of subsequent vascular complicationsY 7 These newer interventional procedures include coronary balloon angioplasty and atherectomy with and without stent placement, aortic valvuloplasty, percutaneous partial left heart bypass, and placement of intraaortic balloon pump assist devices. Concurrent administration of adjunctive anticoagulants and thrombolytic therapy further increase the hazards attending these interventions. 7x° Relatively little is known about the vascular complications after interventional cardiac catheterizationY -~4 The purpose of this report is to define the clinical characteristics and outcome of surgical man593
594
Journal of VASCULAR SURGERY
Messina et al.
Table I. Sixty-five procedures associated with vascular complications in 64 patients undergoing interventional cardiac catheterization Procedure Pcrcutaneous transluminal angioplasty (PTCA) Intraaortic balloon pump Aortic valvuloplasty PTCA/cardiopulmonary bypass + Electrophysiologic intervention AtherecmmyJ-
No. @procedures
24 (37%) 14 (22%) 11 (17%) 10 (15%) 4 (6%) 2 (3%)
*One patient also had a coronary artery stent placed. [One patient had atherectomy of left main coronary artery during partial left heart bypass.
agement of vascular complications after interventional cardiac catheterization and to contrast them to those after diagnostic cardiac catheterization. METHODS AND PATIENTS One hundred one patients underwent operation for 106 vascular complications after cardiac catheterization at the University of Michigan Medical Center from Oct. 1, 1985, to Dec. 1, 1989. The hospital records of these patients were reviewed retrospectively for the following information: (1) history of cardiovascular disease, including occurrences of acute myocardial infarction just before the catheterization; (2) risk factors, including hypertension, and peripheral vascular occlusive disease; (3) physical findings, especially documentation of precatheterization pulses and any neurologic deficits at the time of consultation; (4) the type of diagnostic or interventional procedure and site of catheterization; (5) the indication for surgical repair, the type of anesthesia used, as well as the operation performed, and (6) the postoperative complications and mortality rates. Differences in frequency of the type of complications after interventional and diagnostic catheterizations were tested by Fisher's exact test. Ap value less than 0.05 was considered significant. An odds ratio was used to compare the proportion of vascular injuries between patients undergoing interventional catheterization and patients undergoing diagnostic catheterization. Interventional cardiac catheterization Sixty-nine complications occurred in 64 patients undergoing 65 interventional cardiac catheterization procedures. During this period the Cardiac Catheterization Registry showed that 1866 interventional procedures were performed. Thus the incidence of complications requiring operation was 3.4%. The 30
men (47%) and 34 women (53%) incurring these vascular complications ranged in age from 35 to 89 years (mean, 66 years). Potential risk factors for these complications included hypertension in 41% and peripheral vascular occlusive disease in 34% of these patients. Fifteen patients (24%) suffered an acute myocardial infarction (two of whom also had a cardiac arrest) before their cardiac catheterization. Interventional procedures in this series were coronary balloon angioplasty (34, including 10 requiring percutaneous partial cardiopulmonaH bypass); intraaortic balloon pump placement (14); aortic valvuloplasty (11); coronary atherectomy (2); and electrophysiologic intervention (4) (Table I). Arterialcatheterization in these 65 interventional procedures was by a transfemoral route in 63 patients (97%) and transbrachial route in two patients (3%). Diagnostic cardiac catheterization During the same time period 37 patients undergoing diagnostic cardiac catheterization experienced 37 vascular complications that required surgical repair. During this period the Cardiac Catheterization Registry showed that 5046 diagnostic catheterizations were performed. Thus the incidence of complications requiring operation was 0.7%. Twenty men (54%) and 17 women (46%) whose age ranged from 43 to 92 years (mean age, 63 years) were in this group of patients. Risk factors potentially contributing to the development of vascular complications were hypertension in 40% and peripheral vascular occlusive disease in 35% of these patients. Five patients (14%) had a myocardial infarction, and two patients (5%) experienced a cardiac arrest just before their catheterization. The route of arterial catheterization was transfemoral in 33 patients (89%) and transbrachial in four patients (11%). RESULTS Interventional catheterization Vascular complications after interventional procedures (Table II) occurred 5.2 times (95% confidence 3.5 to 7.8) more frequently than after diagnostic procedures. These complications were hemorrhage (33), arterial thrombosis (18), pseudoaneurysm formation (12), catheter embolization (2), arterial thromboembolism (2), arteriovenous fistula and pseudoaneurysm (1), and an arterial dissection (1). Surgical repair was undertaken promptly after the diagnosis of the vascular complication. Local anesthesia was used for 70% of the operations, and general anesthesia was used for the remaining 30%.
Volume 13 Number 5 May 1991
Cardiac catheterization complications
595
Table II. Type of vascular complications occurring after diagnostic and interventional cardiac catheterizations Diagnostic ( n = 37) (no. of complications) % Hemorrhage Arterial thrombosis Pseudoaneurysm Arterial dissection Arteriovenous fistula Thromboemboli Sheared catheter embolization
(8) (10) (11) (3) (3) (2) (0)
22% 27% 30% 8% 8% 5% 0%
Interventional ( n = 69) (no. of complications) % (33) (18) (12) (1) (1) (2) (2)
48% 2 26% 17% 2% 2% 3% 3%
~Different from diagnostic catheterization by Fisher's exact test, p < 0.05.
Surgical management was individualized, taking into account the specific complication and the patient's overall condition (Table III). Hemorrhage was defined as uncontrolled bleeding or a hematoma that was expanding or whose size threatened skin viability. Local arterial repair or thrombectomy was performed in 34 (51%) of the patients. Eleven patients (16%) underwent repair of pseudoaneurysms, one of whom also had an arteriovenous fistula resection. Seven patients required local repair of large arterial and venous lacerations at the time their bypass catheters were removed. Four patients underwent arterial thromboembolectomy and vein patch angioplasty. Two patients required thromboembolectomy alone, and two other patients required retrieval of broken catheters that had embolized. Six additional patients required other, more complex repairs including femoral artery endarterectomy and thrombectomy (1); aortofemoral bypass (1); femoral artery endarterectomy and suture repair of femoral artery and vein lacerations (1); thromboembolectomy, suture repair of femoral artery lacterations, and fasciotomy (1); femoral popliteal bypass (1); resection of an infected femoral pseudoaneurysm and iliofemoral saphenous vein interposition graft (1). Thirteen postoperative complications occurred in 12 (19%) of the 64 patients undergoing operative repair of vascular injuries after interventional cardiac catheterization. Three of these 12 patients required reoperation. Eight patients had wound complications including seroma formation, lymph leak, hematoma formation, wound cellulitis, or wound edge separation. Two patients developed recurrent arterial thromboses, one of whom required a femoralpopliteal bypass. One patient exhibited a persistent lower extremity, neurologic deficit that had been present before operation. Two patients required amputation. Five patients died after operation. Only one of these deaths, a recurrent subendocardial myocardial infarction in a patient undergoing aortic
valvuloplasty was thought to be related to the surgical intervention. Three of the remaining four patients died of multisystem organ failure, and one died of intractable congestive heart failure. The patient dying of intractable congestive failure had required a balloon pump to treat cardiogenic shock after repair of an acute ventricular septal defect. Another patient died of multisystem organ failure after successful repair of a brachial artery occlusion. The patient had undergone percutaneous transluminal angioplasty for an acute coronary occlusion causing an anterior wall myocardial infarction and cardiogenic shock. A third patient died of multisystem organ failure, sepsis, and possible low-flow mesenteric ischemia after successful repair of an iliac artery perforation. The patient required an intraaortic balloon pump for cardiogenic shock after emergency coronary artery bypass grafting for an acute myocardial infarction. The fourth patient died of multisystem organ failure and sepsis after repair of an iliofemoral dissection and lower leg fasciotomy. Cardiogenic shock developed in the fourth patient as a result of acute spasm of her coronary artery bypass grafts. An above-knee amputation was performed to exclude calf muscle as a source or sepsis and acute renal tubular necrosis. Diagnostic catheterization Vascular complications requiring surgical repair after diagnostic studies (Table II) included hemorrhage (8), arterial thrombosis (10), pseudoaneurysm formation (11), arterial dissection (3), formation of an arteriovenous fistula (3) (one of whom also had a pseudoaneurysm), and acute thromboembolism (2). Surgical repair of vascular complications that occurred after diagnostic cardiac catheterization included local arterial repair or thrombectomy in 12 patients (32%), pseudoaneurysm repair in 12 patients (32%), resection of arteriovenous fistulas in three patients (8%), (one of whom also had repair of a pseudoaneurysm), arterial thromboembolectomy
596
lourna! of VASCULAR SURGERY
Messina et al.
Table III. Type of surgical procedures performed for diagnostic and interventional cardiac catheterization-related vascular complications Diagnostic ( n %
Local arterial repair/thrombectomy Resection pseudoaneurysm Resection arteriovenous fisttfla Remove arterial/venous catheters/repair lacerations Retrieve intraarterial catheter Thromboembolectomy Thromboembolectomy/patch angioplasty Saphenous vein interposition graft Other
in two patients, saphenous vein interposition grafts in two patients, and patch angioplasty and thromboembolectomy in two patients. Four additional patients required thrombectomy of a femoralpopliteal graft and profundaplasty (1); thromboembolectomy and femoral-femoral bypass (1); thromboembolectomy of femoral-popliteal graft and a distal vein graft repair (1); and repair of an iliac artery laceration (1). Operations were done under local anesthesia in 18 patients (49%), general anesthesia in 12 patients (32%), epidural anesthesia in five patients (14%), and regional anesthesia in two patients. Twelve postoperative complications occurred in 10 of the 37 patients who underwent surgical repair of vascular injuries after diagnostic cardiac catheterizations. Four of these complications required reoperation. Seven patients incurred wound complications, including seroma formation, lymph leak, wound cellulitis, or hematoma formation. Two patients suffered postoperative arterial thromboses. Two patients had neurologic deficits, although neither deficit resulted in serious disability. One patient required an amputation after suffering multiple thromboemboli to the right leg. The woman, in cardiogenic shock after an acute myocardial infarction, underwent diagnostic angiography through a right transfemoral approach and the placement of an intraaortic balloon pump assist device through the left femoral artery. The one death after diagnostic catheterization occurred in a 69-year-old man in whom a cold, mottled leg, and diffuse abdominal pain developed after diagnostic cardiac catheterization. Exploratory laparotomy revealed multiple loops of infarcted small bowel and colon as a result of artheroemboli. Eventually pulmonary and renal insufficiency developed, and the patient died. There were no postoperative myocardial infarctions.
=
37)
(12) 32% (12) 32%
3) 8% 0% 0%
2) 5% 2) 5% 2) 5% 4) 11%
Interventional %
(n =
67)
(34) 51% (11) 16% 1) 2% 7) 11%
2) 3% 2) 3% 4) 6% 0%
6) 9%
DISCUSSION
Vascular complications occur frequently after interventional cardiac catheterization procedures and often are serious injuries requiring surgical treatment on an emergency basis. Although aggressive application of percutaneous interventional cardiac catheterization procedures has been of considerable benefit to patients with coronary artery and cardiac valvular heart disease, the potential value of this treatment would be enhanced by reducing the frequency and severity of attendant vascular complications identified in this and other s t u d i e s . 4"6"7 Features that distinguish the clinical characteristics (Table II) and surgical management (Table III) of vascular complications that occurred after interventional cardiac catheterization from those after diagnostic cardiac catheterizations include (1) the frequency of complications, (2) the causes and types of complications, and (3) the high-risk nature of patients undergoing interventional cardiac catheterization procedures. Incidence
The frequency of vascular complications after interventional procedures (3.4%) was 5.2 (95% confidence limits; range, 3.5 to 7.8) times greater than that after diagnostic procedures (0.7%). The frequency of complications identified in this retrospective study are similar to those found during a recent prospective study at the authors' institution. 7 In the latter study the overall frequency of complications after diagnostic and therapeutic catheterization was 1.6% of 2400 consecutive patients undergoing cardiac catheterization during a 12-month time period.7 This frequency of overall complications is consistent with that of other series. 2-4In this recent prospective study the incidence of vascular complications was 0.6% after diagnostic catheterization, 2.6% after percutaneous transluminal angioplasty,
Volume 13 Number 5 May 1991
and 6.0% after complex percutaneous therapeutic interventions. 7 These latter interventions included coronary artery stent insertion, atherectomy, aortic valvuloplasty, placement of an intraaortic balloon pump, and partial cardiopulmonary bypass. Thus there are important differences in the frequency of complications after various interventional procedures. Of note is that concurrent use of heparin or lyric therapy, arterial sheath size _ 8F, peripheral vascular disease, and greater than three coronary risk factors predicted a significantly higher rate of postprocedure vascular complications.
Causes and types of vascular complications Atherosclerotic occlusive disease at the site of arterial catheterization is a compounding variable that on occasion contributes to the severity of the complications and necessitates major vascular reconstructions as occurred in 10% of our patients. Passage of sheaths greater than 8F or insertion of a partial left heart bypass catheters as large as 22F in femoral arteries or veins result in large lacerations, making closure more difficult, especially in the presence of calcified vessel walls or small vessels. The latter is of particular importance in women undergoing such procedures. Use of fibrinolytic agents or continuous anticoagulant therapy in patients who have incurred a recent myocardial infarction increases the risk of bleeding in these patients. 7~° Almost half the complications after percutaneous interventional cardiac procedures were due to hemorrhage at the catheterization site (Table II). Hemorrhage in this series was defined as uncontrolled bleeding, bleeding requiring transfusion, or an expanding hematoma whose size threatened the viability of the overlying tissues. Of note is that in other recent prospective studies at this institution, 26% of patients undergoing emergent catheterization for an acute myocardial infarction bled enough to require a transfusion of two units of blood. 91° This high frequency of hemorrhagic complications after interventional procedures is related directly to the greater frequency of procedure-related risk factors. Sixty percent of the procedures responsible for vascular complications reported in this study followed use of catheters or sheaths greater than 8F in size (Table IV). An additional important factor that may increase the rate of complications is the long duration of certain procedures. For example, coronary artery angioplasty in the setting of an acute myocardial infarction in the present series usually required leaving the vascular catheter and sheath in place for hours or days for later follow-up studies de-
Cardiac catheterization complications 597
Table IV. Size of catheters required for cardiac catheterization procedures Partial cardiopulrnonary bypass Aortic valvuloplasty Intraaortic balloon pump Coronary atherectomy Coronary stem Percutaneous transluminal angioplasty Electrophysiology Diagnostic angiography
22-F (7.3mm)/20F (6.7 mm) 14-F (4.7mm) l l - F (3.7mm) l l - F (3.7ram) IO-F (3.3mm) 8-F (2.7mm) 8-F (2.7mm)-lO-F (3.3mm) 7-F (2.3ram)
pending on the success rate of the initial coronary dilation. Nature of patients undergoing catheterization Patients undergoing interventional procedures were often more critically ill than those subjected to diagnostic studies. This was reflected in the 24% incidence of preprocedure acute myocardial infarction in the former group, by their 8% mortality rate after incurring these complications, and the fact that 70% of the operations for complications ofinterventional procedures were performed under local anesthesia. Although many of these complications could have been repaired more expeditiously under general anesthesia, local anesthesia was used to minimize the risk to the patient of cardiac-related complications of general anesthesia. In contrast, among patients having diagnostic catheterizations, 60% of the patients requiring operation for either pseudoaneurysms or arteriovenous fistulas underwent delayed repairs up to 5 months after the initial catheterization. Thus a greater proportion of the complications after diagnostic catheterization were repaired electively rather than on an emergency basis on acutely ill patients.
Evaluation of vascular complications Experience in the management of these of 106 vascular complications after interventional and diagnostic cardiac catheterizations has led us to a number of conclusions concerning the evaluation and treatment of these patients. Most catheterization-related vascular complications are of sudden onset, usually occurring during or immediately after the procedure and are accompanied by unequivocal symptoms and signs of extremity ischemia or hemorrhage. The diagnosis of most of these complications can be made confidently on clinical grounds. However, in certain individuals the clinical presentation may be obscured by preexisting risk factors such as peripheral vascular occlusive disease, as affected 35% of the patients in this present series. We believe that patients noted to
598 Messina et al.
have an abnormal vascular history or examination before cardiac catheterization should have anklebrachial indexes and segmental waveforms and pressures documented before their catheterization. In addition, patients undergoing high-risk interventional procedures should also have these studies documented before the study. These Doppler studies may lessen the frequency of vascular complications by identifying limbs at high risk for complications and thereby suggest to the cardiologist that an alternate site of vascular access or even a different technique is indicated. In addition, these studies may be very helpful in determining the presence and extent of a vascular complication after the catheterization especially in patients without palpable pulses. Furthermore, knowing the preoperative baseline anklebrachial index may be helpful in assessing the adequacy of repair in the operating room. A major difficulty that may arise in the initial evaluation of these patients is the distinction between a large hematoma, pseudoaneurysm, or arteriovenous fistula. Duplex color-flow imaging has proved to be a direct and accurate means to differentiate these complications, is Color-flow Doppler examination can distinguish between a large hematoma, and a pseudoaneurysm, which is characterized by this echolucent structure and a "to and fro" Doppler flow pattern, or an acute arteriovenous fistula that has continuous flow throughout the cardiac cycle. Confirmatory arteriography may be useful in the planning of treatment for chronic complications but are rarely necessary or appropriate in the acute management of these patients.
Surgical considerations The operative management of vascular complications after interventional or diagnostic catheterization is dependent on the type of injury incurred and general condition of the patient. The latter is of paramount importance to the eventual outcome. Because some of these complications require emergency repair these critically ill patients may not always have the benefit of a flail preoperative evaluation. Additionally, the growth in endovascular procedures of contemporary cardiology practice has led to an unparalleled degree of specialization. Interventionists may not have an established relationship with the patient before the cardiac emergency arises, and they may not be aware of prior conditions such as peripheral arterial occlusive disease that might contribute to catheterization-related complications. Furthermore, they may not have ongoing critical care responsibility for these patients. In this setting careful
Journal of VASCULAR SURGERY
coordination of patient management among the responsible cardiologist, vascular surgeon, and anesthesiologist is essential to an optimal clinical outcome. Certain technical maneuvers in the management of vascular complications attending these catheter injuries deserve note. Hemorrhage and hematoma formation can frequently distort normal anatomic relationships in these patients. The risk of additional injury to neurovascular structures can result from overzealous dissection in these circumstances. Similarly, extensive subcutaneous dissection will increase the incidence of local wound complications and should be avoided. Limited vascular control with the use ofintraluminal devices, such as balloon catheters, is sometimes preferred. In the case of bleeding from small vascular access wounds, hemorrhage may be controlled easily by the insertion of rigid dilators or application of a finger tip over the puncture site. Extensive proximal and distal vascular control is often not necessary in this setting. Alternatively when certain vascular complications occur such as arterial thrombosis or arteriovenous fistula formation, the dissection should provide adequate proximal and distal control as well as for a careful arterial reconstruction. Prophylactic perioperative antibiotics may be helpful to lessen the incidence of wound complications. This is particularly important when catheters or sheaths were left in place for long periods of time, or in patients experiencing tissue injury because of massive bleeding into surrounding structures. Infection and other wound complications may be reduced further by making an incision through large puncture holes in the skin so that the skin edges can be excised sharply or by making an incision remote from these large puncture sites to minimize wound contamination from them. Anesthetic technique is also an important issue in the management of patients with these vascular complications. Local anesthetics, supplemented by short acting tranquilizers, are usually sufficient for treating acute arterial occlusions. However, local anesthesia may be less satisfactory for the management of large hematomas or pseudoaneurysms, where vascular control might necessitate greater dissection, and the presence of the hematoma makes effective instillation of a local anesthetic difficult. Similarly, individuals who have had large catheters placed for partial cardiopulmonary bypass, in whom major arterial and venous lacerations may require extended vascular control and repair, may be best managed with a general, spinal, or epidural anes-
Volume 13 Number 5 May 1991
thetic. This is particularly true when the vascular injury extends above the inguinal ligament. Regardless of the type of injury, local anesthesia incurs the least risk to the patient, but on occasion the operative management of patients will require a general anesthetic. The most appropriate setting for undertaking operative repair of vascular complications after interventional and diagnostic cardiac catheterizations is in an operating room. However, exceptions to this general rule may arise. Critically ill and unstable patients, including those who have suffered an acute myocardial infarction complicated by arrhythmia or hemodynamic instability, as well as patients requiring intraaortic balloon pump assist devices, may not always be transported safely to the operating room. Nonetheless good lighting, appropriate surgical instruments, and skilled support personnel, particularly anesthesiologists, are still requirements for successful repair of the vascular complications. These requirements may sometimes be met in contemporary cardiac catheterization suites, but are uncommonly met in intensive care units. If such requirements for surgical therapy cannot be met, it may be in the patient's best interest to delay the repair until the patient can be transferred safely to the operating room. Vascular complications attending percutaneous interventional cardiac catheterizations are relatively common and will likely occur more frequently as these techniques are applied more widely in the treatment of coronary and valvular heart disease and in support of the failing heart. Reducing the incidence of these complications is important because they may be threatening to life and limb. Emergency procedures, particularly those requiring the insertion of catheters with diameters greater than 8F into women or into patients who have peripheral vascular occlusive disease, should alert the clinician to the high risk for subsequent vascular complications. Better preoperative assessment of the risk of vascular injury by formal vascular studies should reduce the incidence of these complications by identifying to the cardiologist a patient at high risk for complications and by suggesting to him a more optimal site for catheter insertion. In addition, the vascular studies will provide for a more rational basis of later clinical assessments if any injury does occur, particularly in patients who did not have palpable pulses before the catheterization. Until smaller catheters and sheaths are designed for these procedures, open catheter insertion rather than percutaneous catheter insertion may result in a
Cardiac catheterization complications 599
lower incidence of vascular compfications. '6 This recommendation would apply particularly to patients at high risk for vascular complications after intraaortic balloon pump insertion and to all patients undergoing percutaneous cardiopulmonary bypass. Since operative exposure of groin vessels to be catheterized is not standard practice, this would require that vascular surgeons expose these vessels and repair them after the catheters are withdrawn, or alternatively that cardiologists themselves be trained to accomplish this surgical procedure. Optimal management of patients undergoing interventional cardiac catheterization will require increased diligence in their care both before, during, and after the procedure, if vascular complications are to be reduced and their management improved when they do occur. We thank David Muller, M D , and Zvi Flanders, MS, o f
the Division of Cardiology, Department of Medicine, for providing information from their computer registry and their helpful comments. We also thank Cathy Blankenburg for her excellent assistance in the preparation of this manuscript. REFERENCES 1. Orcutt MB, Levine BA, Gaskill HV III, et al. Iatrogenic vascular injury. A reducible problem. Arch Surg 1985;120: 384-5. 2. Kennedy JW. Complications associated with cardiac catheterization and angiography. Cathet Cardiovasc Diagn 1982;8: 5-11. 3. Hessel SJ, Adams DF, Abrams HL. Complications of angiography. Radiology 1981;138:273-81. 4. Skillman JJ, Kim D, Balm DS. Vascular complications of percutaneous femoral cardiac interventions. Incidence and operative repair. Arch Surg 1988;123:1207-12. 5. Gardiner GA Jr, Meyerovitz MF, Stokes KR, et al. Complications of transluminal angioplasty. Radiology 1986;159: 201-8. 6. Cribier A, Savin T, Berland J, et al. Percutaneous transltmainal balloon valvuloplasty of adult aortic stenosis: report of 92 cases. J Am Coil Cardiol 1987;9:381-6. 7. Muller DW, Podd J, Shamir KJ. Vascular access site complications in the era of complex percutaneous coronary interventions. Circulation 1990;82 (suppl III):III-510. 8. McMillan I, Murie JA. Vascular injury following cardiac catheterization. Br J Surg 1984;71:832-5. 9. Topoi EJ, O'Neill WW, Langburd AB, et al. A randomized, placebo-controlled trial of intravenous recombinant tissuetype plasminogen activator and emergency coronary angioplasty in patients with acute myocardial infarction. Circulation 1987;75:420-8. 10. O'Neill W, Timmis GC, Bourdillion PD, et al. A prospective randomized clinical trial of intracoronary streptokinase versus coronary angioplasty for acute myocardial infarction. N Engl J Med 1986;314:812-8. 11. Mills JL, Wiedeman JE, Robison JG, et al. Minimizing mortality and morbidity from iatrogenic arterial injuries: the need for early recognition and prompt repair. J VASC SURG 1986;4:22-7.
600
Journal of VASCUL,M~. SURGERY
Messina et al.
12. Youkey JR, Clagett GP, Rich NM, et al. Vascular trauma secondary to diagnostic and therapeutic procedures: 1974 through 1982. A comparative review. Am J Surg 1983;146: 788-91. 13. Babu SC, Piccorelli GO, Shah PM, et al. Incidence and results of arterial complications among 16,350 patients undergoing cardiac catheterization. J VASC SURG 1989; 10:113-6. 14. Kline RM Jr, Hertzer NR, Beven EG, et al. Surgical treatment of brachial artery injuries after cardiac catheterization. J VASC SURG 1990;12:20-4.
15. Sheikh KH, Adams DB, McCann R, et al. Utility of Doppler colorflow imaging for identification of femoral arterial complications of cardiac catheterization. Am Heart J 1989; 117: 623-8. 16. Mackey W. In discussion: Skillman JJ, Kim D, Bairn DS. Vascular complications of percutaneous femoral cardiac interventions. Arch Surg 1988;123:1207-12.
Submitted Sept. 26, 1990; accepted Dec. 24, 1990.
The incidence of vascular complications after diagnostic cardiac catheterization is low, reflecting refinements in angiographic technique and catheter construction. The contemporary clinical practice of cardiology also includes sophisticated endovascular interventions to treat coronary artery and cardiac valvular heart disease and to support the failing heart. From the Section of Vascular Surgery, Department of Surgery, Universityof MichiganMedicalCenter, Ann Arbor. Presented at the Fourteenth AnnualMeeting of the Midwestern Vascular SurgicalSociety,Toledo, Ohio, Sept. 14, 1990. Reprint requests: LouisM. Messina,MD, 2210 TaubmanCenter, 1500 E. MedicalCenter Dr., Ann Arbor, MI 48109-0329. 24/6/27611
Such procedures are often lifesaving but carry a substantial risk of subsequent vascular complicationsY 7 These newer interventional procedures include coronary balloon angioplasty and atherectomy with and without stent placement, aortic valvuloplasty, percutaneous partial left heart bypass, and placement of intraaortic balloon pump assist devices. Concurrent administration of adjunctive anticoagulants and thrombolytic therapy further increase the hazards attending these interventions. 7x° Relatively little is known about the vascular complications after interventional cardiac catheterizationY -~4 The purpose of this report is to define the clinical characteristics and outcome of surgical man593
594
Journal of VASCULAR SURGERY
Messina et al.
Table I. Sixty-five procedures associated with vascular complications in 64 patients undergoing interventional cardiac catheterization Procedure Pcrcutaneous transluminal angioplasty (PTCA) Intraaortic balloon pump Aortic valvuloplasty PTCA/cardiopulmonary bypass + Electrophysiologic intervention AtherecmmyJ-
No. @procedures
24 (37%) 14 (22%) 11 (17%) 10 (15%) 4 (6%) 2 (3%)
*One patient also had a coronary artery stent placed. [One patient had atherectomy of left main coronary artery during partial left heart bypass.
agement of vascular complications after interventional cardiac catheterization and to contrast them to those after diagnostic cardiac catheterization. METHODS AND PATIENTS One hundred one patients underwent operation for 106 vascular complications after cardiac catheterization at the University of Michigan Medical Center from Oct. 1, 1985, to Dec. 1, 1989. The hospital records of these patients were reviewed retrospectively for the following information: (1) history of cardiovascular disease, including occurrences of acute myocardial infarction just before the catheterization; (2) risk factors, including hypertension, and peripheral vascular occlusive disease; (3) physical findings, especially documentation of precatheterization pulses and any neurologic deficits at the time of consultation; (4) the type of diagnostic or interventional procedure and site of catheterization; (5) the indication for surgical repair, the type of anesthesia used, as well as the operation performed, and (6) the postoperative complications and mortality rates. Differences in frequency of the type of complications after interventional and diagnostic catheterizations were tested by Fisher's exact test. Ap value less than 0.05 was considered significant. An odds ratio was used to compare the proportion of vascular injuries between patients undergoing interventional catheterization and patients undergoing diagnostic catheterization. Interventional cardiac catheterization Sixty-nine complications occurred in 64 patients undergoing 65 interventional cardiac catheterization procedures. During this period the Cardiac Catheterization Registry showed that 1866 interventional procedures were performed. Thus the incidence of complications requiring operation was 3.4%. The 30
men (47%) and 34 women (53%) incurring these vascular complications ranged in age from 35 to 89 years (mean, 66 years). Potential risk factors for these complications included hypertension in 41% and peripheral vascular occlusive disease in 34% of these patients. Fifteen patients (24%) suffered an acute myocardial infarction (two of whom also had a cardiac arrest) before their cardiac catheterization. Interventional procedures in this series were coronary balloon angioplasty (34, including 10 requiring percutaneous partial cardiopulmonaH bypass); intraaortic balloon pump placement (14); aortic valvuloplasty (11); coronary atherectomy (2); and electrophysiologic intervention (4) (Table I). Arterialcatheterization in these 65 interventional procedures was by a transfemoral route in 63 patients (97%) and transbrachial route in two patients (3%). Diagnostic cardiac catheterization During the same time period 37 patients undergoing diagnostic cardiac catheterization experienced 37 vascular complications that required surgical repair. During this period the Cardiac Catheterization Registry showed that 5046 diagnostic catheterizations were performed. Thus the incidence of complications requiring operation was 0.7%. Twenty men (54%) and 17 women (46%) whose age ranged from 43 to 92 years (mean age, 63 years) were in this group of patients. Risk factors potentially contributing to the development of vascular complications were hypertension in 40% and peripheral vascular occlusive disease in 35% of these patients. Five patients (14%) had a myocardial infarction, and two patients (5%) experienced a cardiac arrest just before their catheterization. The route of arterial catheterization was transfemoral in 33 patients (89%) and transbrachial in four patients (11%). RESULTS Interventional catheterization Vascular complications after interventional procedures (Table II) occurred 5.2 times (95% confidence 3.5 to 7.8) more frequently than after diagnostic procedures. These complications were hemorrhage (33), arterial thrombosis (18), pseudoaneurysm formation (12), catheter embolization (2), arterial thromboembolism (2), arteriovenous fistula and pseudoaneurysm (1), and an arterial dissection (1). Surgical repair was undertaken promptly after the diagnosis of the vascular complication. Local anesthesia was used for 70% of the operations, and general anesthesia was used for the remaining 30%.
Volume 13 Number 5 May 1991
Cardiac catheterization complications
595
Table II. Type of vascular complications occurring after diagnostic and interventional cardiac catheterizations Diagnostic ( n = 37) (no. of complications) % Hemorrhage Arterial thrombosis Pseudoaneurysm Arterial dissection Arteriovenous fistula Thromboemboli Sheared catheter embolization
(8) (10) (11) (3) (3) (2) (0)
22% 27% 30% 8% 8% 5% 0%
Interventional ( n = 69) (no. of complications) % (33) (18) (12) (1) (1) (2) (2)
48% 2 26% 17% 2% 2% 3% 3%
~Different from diagnostic catheterization by Fisher's exact test, p < 0.05.
Surgical management was individualized, taking into account the specific complication and the patient's overall condition (Table III). Hemorrhage was defined as uncontrolled bleeding or a hematoma that was expanding or whose size threatened skin viability. Local arterial repair or thrombectomy was performed in 34 (51%) of the patients. Eleven patients (16%) underwent repair of pseudoaneurysms, one of whom also had an arteriovenous fistula resection. Seven patients required local repair of large arterial and venous lacerations at the time their bypass catheters were removed. Four patients underwent arterial thromboembolectomy and vein patch angioplasty. Two patients required thromboembolectomy alone, and two other patients required retrieval of broken catheters that had embolized. Six additional patients required other, more complex repairs including femoral artery endarterectomy and thrombectomy (1); aortofemoral bypass (1); femoral artery endarterectomy and suture repair of femoral artery and vein lacerations (1); thromboembolectomy, suture repair of femoral artery lacterations, and fasciotomy (1); femoral popliteal bypass (1); resection of an infected femoral pseudoaneurysm and iliofemoral saphenous vein interposition graft (1). Thirteen postoperative complications occurred in 12 (19%) of the 64 patients undergoing operative repair of vascular injuries after interventional cardiac catheterization. Three of these 12 patients required reoperation. Eight patients had wound complications including seroma formation, lymph leak, hematoma formation, wound cellulitis, or wound edge separation. Two patients developed recurrent arterial thromboses, one of whom required a femoralpopliteal bypass. One patient exhibited a persistent lower extremity, neurologic deficit that had been present before operation. Two patients required amputation. Five patients died after operation. Only one of these deaths, a recurrent subendocardial myocardial infarction in a patient undergoing aortic
valvuloplasty was thought to be related to the surgical intervention. Three of the remaining four patients died of multisystem organ failure, and one died of intractable congestive heart failure. The patient dying of intractable congestive failure had required a balloon pump to treat cardiogenic shock after repair of an acute ventricular septal defect. Another patient died of multisystem organ failure after successful repair of a brachial artery occlusion. The patient had undergone percutaneous transluminal angioplasty for an acute coronary occlusion causing an anterior wall myocardial infarction and cardiogenic shock. A third patient died of multisystem organ failure, sepsis, and possible low-flow mesenteric ischemia after successful repair of an iliac artery perforation. The patient required an intraaortic balloon pump for cardiogenic shock after emergency coronary artery bypass grafting for an acute myocardial infarction. The fourth patient died of multisystem organ failure and sepsis after repair of an iliofemoral dissection and lower leg fasciotomy. Cardiogenic shock developed in the fourth patient as a result of acute spasm of her coronary artery bypass grafts. An above-knee amputation was performed to exclude calf muscle as a source or sepsis and acute renal tubular necrosis. Diagnostic catheterization Vascular complications requiring surgical repair after diagnostic studies (Table II) included hemorrhage (8), arterial thrombosis (10), pseudoaneurysm formation (11), arterial dissection (3), formation of an arteriovenous fistula (3) (one of whom also had a pseudoaneurysm), and acute thromboembolism (2). Surgical repair of vascular complications that occurred after diagnostic cardiac catheterization included local arterial repair or thrombectomy in 12 patients (32%), pseudoaneurysm repair in 12 patients (32%), resection of arteriovenous fistulas in three patients (8%), (one of whom also had repair of a pseudoaneurysm), arterial thromboembolectomy
596
lourna! of VASCULAR SURGERY
Messina et al.
Table III. Type of surgical procedures performed for diagnostic and interventional cardiac catheterization-related vascular complications Diagnostic ( n %
Local arterial repair/thrombectomy Resection pseudoaneurysm Resection arteriovenous fisttfla Remove arterial/venous catheters/repair lacerations Retrieve intraarterial catheter Thromboembolectomy Thromboembolectomy/patch angioplasty Saphenous vein interposition graft Other
in two patients, saphenous vein interposition grafts in two patients, and patch angioplasty and thromboembolectomy in two patients. Four additional patients required thrombectomy of a femoralpopliteal graft and profundaplasty (1); thromboembolectomy and femoral-femoral bypass (1); thromboembolectomy of femoral-popliteal graft and a distal vein graft repair (1); and repair of an iliac artery laceration (1). Operations were done under local anesthesia in 18 patients (49%), general anesthesia in 12 patients (32%), epidural anesthesia in five patients (14%), and regional anesthesia in two patients. Twelve postoperative complications occurred in 10 of the 37 patients who underwent surgical repair of vascular injuries after diagnostic cardiac catheterizations. Four of these complications required reoperation. Seven patients incurred wound complications, including seroma formation, lymph leak, wound cellulitis, or hematoma formation. Two patients suffered postoperative arterial thromboses. Two patients had neurologic deficits, although neither deficit resulted in serious disability. One patient required an amputation after suffering multiple thromboemboli to the right leg. The woman, in cardiogenic shock after an acute myocardial infarction, underwent diagnostic angiography through a right transfemoral approach and the placement of an intraaortic balloon pump assist device through the left femoral artery. The one death after diagnostic catheterization occurred in a 69-year-old man in whom a cold, mottled leg, and diffuse abdominal pain developed after diagnostic cardiac catheterization. Exploratory laparotomy revealed multiple loops of infarcted small bowel and colon as a result of artheroemboli. Eventually pulmonary and renal insufficiency developed, and the patient died. There were no postoperative myocardial infarctions.
=
37)
(12) 32% (12) 32%
3) 8% 0% 0%
2) 5% 2) 5% 2) 5% 4) 11%
Interventional %
(n =
67)
(34) 51% (11) 16% 1) 2% 7) 11%
2) 3% 2) 3% 4) 6% 0%
6) 9%
DISCUSSION
Vascular complications occur frequently after interventional cardiac catheterization procedures and often are serious injuries requiring surgical treatment on an emergency basis. Although aggressive application of percutaneous interventional cardiac catheterization procedures has been of considerable benefit to patients with coronary artery and cardiac valvular heart disease, the potential value of this treatment would be enhanced by reducing the frequency and severity of attendant vascular complications identified in this and other s t u d i e s . 4"6"7 Features that distinguish the clinical characteristics (Table II) and surgical management (Table III) of vascular complications that occurred after interventional cardiac catheterization from those after diagnostic cardiac catheterizations include (1) the frequency of complications, (2) the causes and types of complications, and (3) the high-risk nature of patients undergoing interventional cardiac catheterization procedures. Incidence
The frequency of vascular complications after interventional procedures (3.4%) was 5.2 (95% confidence limits; range, 3.5 to 7.8) times greater than that after diagnostic procedures (0.7%). The frequency of complications identified in this retrospective study are similar to those found during a recent prospective study at the authors' institution. 7 In the latter study the overall frequency of complications after diagnostic and therapeutic catheterization was 1.6% of 2400 consecutive patients undergoing cardiac catheterization during a 12-month time period.7 This frequency of overall complications is consistent with that of other series. 2-4In this recent prospective study the incidence of vascular complications was 0.6% after diagnostic catheterization, 2.6% after percutaneous transluminal angioplasty,
Volume 13 Number 5 May 1991
and 6.0% after complex percutaneous therapeutic interventions. 7 These latter interventions included coronary artery stent insertion, atherectomy, aortic valvuloplasty, placement of an intraaortic balloon pump, and partial cardiopulmonary bypass. Thus there are important differences in the frequency of complications after various interventional procedures. Of note is that concurrent use of heparin or lyric therapy, arterial sheath size _ 8F, peripheral vascular disease, and greater than three coronary risk factors predicted a significantly higher rate of postprocedure vascular complications.
Causes and types of vascular complications Atherosclerotic occlusive disease at the site of arterial catheterization is a compounding variable that on occasion contributes to the severity of the complications and necessitates major vascular reconstructions as occurred in 10% of our patients. Passage of sheaths greater than 8F or insertion of a partial left heart bypass catheters as large as 22F in femoral arteries or veins result in large lacerations, making closure more difficult, especially in the presence of calcified vessel walls or small vessels. The latter is of particular importance in women undergoing such procedures. Use of fibrinolytic agents or continuous anticoagulant therapy in patients who have incurred a recent myocardial infarction increases the risk of bleeding in these patients. 7~° Almost half the complications after percutaneous interventional cardiac procedures were due to hemorrhage at the catheterization site (Table II). Hemorrhage in this series was defined as uncontrolled bleeding, bleeding requiring transfusion, or an expanding hematoma whose size threatened the viability of the overlying tissues. Of note is that in other recent prospective studies at this institution, 26% of patients undergoing emergent catheterization for an acute myocardial infarction bled enough to require a transfusion of two units of blood. 91° This high frequency of hemorrhagic complications after interventional procedures is related directly to the greater frequency of procedure-related risk factors. Sixty percent of the procedures responsible for vascular complications reported in this study followed use of catheters or sheaths greater than 8F in size (Table IV). An additional important factor that may increase the rate of complications is the long duration of certain procedures. For example, coronary artery angioplasty in the setting of an acute myocardial infarction in the present series usually required leaving the vascular catheter and sheath in place for hours or days for later follow-up studies de-
Cardiac catheterization complications 597
Table IV. Size of catheters required for cardiac catheterization procedures Partial cardiopulrnonary bypass Aortic valvuloplasty Intraaortic balloon pump Coronary atherectomy Coronary stem Percutaneous transluminal angioplasty Electrophysiology Diagnostic angiography
22-F (7.3mm)/20F (6.7 mm) 14-F (4.7mm) l l - F (3.7mm) l l - F (3.7ram) IO-F (3.3mm) 8-F (2.7mm) 8-F (2.7mm)-lO-F (3.3mm) 7-F (2.3ram)
pending on the success rate of the initial coronary dilation. Nature of patients undergoing catheterization Patients undergoing interventional procedures were often more critically ill than those subjected to diagnostic studies. This was reflected in the 24% incidence of preprocedure acute myocardial infarction in the former group, by their 8% mortality rate after incurring these complications, and the fact that 70% of the operations for complications ofinterventional procedures were performed under local anesthesia. Although many of these complications could have been repaired more expeditiously under general anesthesia, local anesthesia was used to minimize the risk to the patient of cardiac-related complications of general anesthesia. In contrast, among patients having diagnostic catheterizations, 60% of the patients requiring operation for either pseudoaneurysms or arteriovenous fistulas underwent delayed repairs up to 5 months after the initial catheterization. Thus a greater proportion of the complications after diagnostic catheterization were repaired electively rather than on an emergency basis on acutely ill patients.
Evaluation of vascular complications Experience in the management of these of 106 vascular complications after interventional and diagnostic cardiac catheterizations has led us to a number of conclusions concerning the evaluation and treatment of these patients. Most catheterization-related vascular complications are of sudden onset, usually occurring during or immediately after the procedure and are accompanied by unequivocal symptoms and signs of extremity ischemia or hemorrhage. The diagnosis of most of these complications can be made confidently on clinical grounds. However, in certain individuals the clinical presentation may be obscured by preexisting risk factors such as peripheral vascular occlusive disease, as affected 35% of the patients in this present series. We believe that patients noted to
598 Messina et al.
have an abnormal vascular history or examination before cardiac catheterization should have anklebrachial indexes and segmental waveforms and pressures documented before their catheterization. In addition, patients undergoing high-risk interventional procedures should also have these studies documented before the study. These Doppler studies may lessen the frequency of vascular complications by identifying limbs at high risk for complications and thereby suggest to the cardiologist that an alternate site of vascular access or even a different technique is indicated. In addition, these studies may be very helpful in determining the presence and extent of a vascular complication after the catheterization especially in patients without palpable pulses. Furthermore, knowing the preoperative baseline anklebrachial index may be helpful in assessing the adequacy of repair in the operating room. A major difficulty that may arise in the initial evaluation of these patients is the distinction between a large hematoma, pseudoaneurysm, or arteriovenous fistula. Duplex color-flow imaging has proved to be a direct and accurate means to differentiate these complications, is Color-flow Doppler examination can distinguish between a large hematoma, and a pseudoaneurysm, which is characterized by this echolucent structure and a "to and fro" Doppler flow pattern, or an acute arteriovenous fistula that has continuous flow throughout the cardiac cycle. Confirmatory arteriography may be useful in the planning of treatment for chronic complications but are rarely necessary or appropriate in the acute management of these patients.
Surgical considerations The operative management of vascular complications after interventional or diagnostic catheterization is dependent on the type of injury incurred and general condition of the patient. The latter is of paramount importance to the eventual outcome. Because some of these complications require emergency repair these critically ill patients may not always have the benefit of a flail preoperative evaluation. Additionally, the growth in endovascular procedures of contemporary cardiology practice has led to an unparalleled degree of specialization. Interventionists may not have an established relationship with the patient before the cardiac emergency arises, and they may not be aware of prior conditions such as peripheral arterial occlusive disease that might contribute to catheterization-related complications. Furthermore, they may not have ongoing critical care responsibility for these patients. In this setting careful
Journal of VASCULAR SURGERY
coordination of patient management among the responsible cardiologist, vascular surgeon, and anesthesiologist is essential to an optimal clinical outcome. Certain technical maneuvers in the management of vascular complications attending these catheter injuries deserve note. Hemorrhage and hematoma formation can frequently distort normal anatomic relationships in these patients. The risk of additional injury to neurovascular structures can result from overzealous dissection in these circumstances. Similarly, extensive subcutaneous dissection will increase the incidence of local wound complications and should be avoided. Limited vascular control with the use ofintraluminal devices, such as balloon catheters, is sometimes preferred. In the case of bleeding from small vascular access wounds, hemorrhage may be controlled easily by the insertion of rigid dilators or application of a finger tip over the puncture site. Extensive proximal and distal vascular control is often not necessary in this setting. Alternatively when certain vascular complications occur such as arterial thrombosis or arteriovenous fistula formation, the dissection should provide adequate proximal and distal control as well as for a careful arterial reconstruction. Prophylactic perioperative antibiotics may be helpful to lessen the incidence of wound complications. This is particularly important when catheters or sheaths were left in place for long periods of time, or in patients experiencing tissue injury because of massive bleeding into surrounding structures. Infection and other wound complications may be reduced further by making an incision through large puncture holes in the skin so that the skin edges can be excised sharply or by making an incision remote from these large puncture sites to minimize wound contamination from them. Anesthetic technique is also an important issue in the management of patients with these vascular complications. Local anesthetics, supplemented by short acting tranquilizers, are usually sufficient for treating acute arterial occlusions. However, local anesthesia may be less satisfactory for the management of large hematomas or pseudoaneurysms, where vascular control might necessitate greater dissection, and the presence of the hematoma makes effective instillation of a local anesthetic difficult. Similarly, individuals who have had large catheters placed for partial cardiopulmonary bypass, in whom major arterial and venous lacerations may require extended vascular control and repair, may be best managed with a general, spinal, or epidural anes-
Volume 13 Number 5 May 1991
thetic. This is particularly true when the vascular injury extends above the inguinal ligament. Regardless of the type of injury, local anesthesia incurs the least risk to the patient, but on occasion the operative management of patients will require a general anesthetic. The most appropriate setting for undertaking operative repair of vascular complications after interventional and diagnostic cardiac catheterizations is in an operating room. However, exceptions to this general rule may arise. Critically ill and unstable patients, including those who have suffered an acute myocardial infarction complicated by arrhythmia or hemodynamic instability, as well as patients requiring intraaortic balloon pump assist devices, may not always be transported safely to the operating room. Nonetheless good lighting, appropriate surgical instruments, and skilled support personnel, particularly anesthesiologists, are still requirements for successful repair of the vascular complications. These requirements may sometimes be met in contemporary cardiac catheterization suites, but are uncommonly met in intensive care units. If such requirements for surgical therapy cannot be met, it may be in the patient's best interest to delay the repair until the patient can be transferred safely to the operating room. Vascular complications attending percutaneous interventional cardiac catheterizations are relatively common and will likely occur more frequently as these techniques are applied more widely in the treatment of coronary and valvular heart disease and in support of the failing heart. Reducing the incidence of these complications is important because they may be threatening to life and limb. Emergency procedures, particularly those requiring the insertion of catheters with diameters greater than 8F into women or into patients who have peripheral vascular occlusive disease, should alert the clinician to the high risk for subsequent vascular complications. Better preoperative assessment of the risk of vascular injury by formal vascular studies should reduce the incidence of these complications by identifying to the cardiologist a patient at high risk for complications and by suggesting to him a more optimal site for catheter insertion. In addition, the vascular studies will provide for a more rational basis of later clinical assessments if any injury does occur, particularly in patients who did not have palpable pulses before the catheterization. Until smaller catheters and sheaths are designed for these procedures, open catheter insertion rather than percutaneous catheter insertion may result in a
Cardiac catheterization complications 599
lower incidence of vascular compfications. '6 This recommendation would apply particularly to patients at high risk for vascular complications after intraaortic balloon pump insertion and to all patients undergoing percutaneous cardiopulmonary bypass. Since operative exposure of groin vessels to be catheterized is not standard practice, this would require that vascular surgeons expose these vessels and repair them after the catheters are withdrawn, or alternatively that cardiologists themselves be trained to accomplish this surgical procedure. Optimal management of patients undergoing interventional cardiac catheterization will require increased diligence in their care both before, during, and after the procedure, if vascular complications are to be reduced and their management improved when they do occur. We thank David Muller, M D , and Zvi Flanders, MS, o f
the Division of Cardiology, Department of Medicine, for providing information from their computer registry and their helpful comments. We also thank Cathy Blankenburg for her excellent assistance in the preparation of this manuscript. REFERENCES 1. Orcutt MB, Levine BA, Gaskill HV III, et al. Iatrogenic vascular injury. A reducible problem. Arch Surg 1985;120: 384-5. 2. Kennedy JW. Complications associated with cardiac catheterization and angiography. Cathet Cardiovasc Diagn 1982;8: 5-11. 3. Hessel SJ, Adams DF, Abrams HL. Complications of angiography. Radiology 1981;138:273-81. 4. Skillman JJ, Kim D, Balm DS. Vascular complications of percutaneous femoral cardiac interventions. Incidence and operative repair. Arch Surg 1988;123:1207-12. 5. Gardiner GA Jr, Meyerovitz MF, Stokes KR, et al. Complications of transluminal angioplasty. Radiology 1986;159: 201-8. 6. Cribier A, Savin T, Berland J, et al. Percutaneous transltmainal balloon valvuloplasty of adult aortic stenosis: report of 92 cases. J Am Coil Cardiol 1987;9:381-6. 7. Muller DW, Podd J, Shamir KJ. Vascular access site complications in the era of complex percutaneous coronary interventions. Circulation 1990;82 (suppl III):III-510. 8. McMillan I, Murie JA. Vascular injury following cardiac catheterization. Br J Surg 1984;71:832-5. 9. Topoi EJ, O'Neill WW, Langburd AB, et al. A randomized, placebo-controlled trial of intravenous recombinant tissuetype plasminogen activator and emergency coronary angioplasty in patients with acute myocardial infarction. Circulation 1987;75:420-8. 10. O'Neill W, Timmis GC, Bourdillion PD, et al. A prospective randomized clinical trial of intracoronary streptokinase versus coronary angioplasty for acute myocardial infarction. N Engl J Med 1986;314:812-8. 11. Mills JL, Wiedeman JE, Robison JG, et al. Minimizing mortality and morbidity from iatrogenic arterial injuries: the need for early recognition and prompt repair. J VASC SURG 1986;4:22-7.
600
Journal of VASCUL,M~. SURGERY
Messina et al.
12. Youkey JR, Clagett GP, Rich NM, et al. Vascular trauma secondary to diagnostic and therapeutic procedures: 1974 through 1982. A comparative review. Am J Surg 1983;146: 788-91. 13. Babu SC, Piccorelli GO, Shah PM, et al. Incidence and results of arterial complications among 16,350 patients undergoing cardiac catheterization. J VASC SURG 1989; 10:113-6. 14. Kline RM Jr, Hertzer NR, Beven EG, et al. Surgical treatment of brachial artery injuries after cardiac catheterization. J VASC SURG 1990;12:20-4.
15. Sheikh KH, Adams DB, McCann R, et al. Utility of Doppler colorflow imaging for identification of femoral arterial complications of cardiac catheterization. Am Heart J 1989; 117: 623-8. 16. Mackey W. In discussion: Skillman JJ, Kim D, Bairn DS. Vascular complications of percutaneous femoral cardiac interventions. Arch Surg 1988;123:1207-12.
Submitted Sept. 26, 1990; accepted Dec. 24, 1990.
