Rad/ol Oncol 1996: .30: 165-70.
Color doppler for the evaluation of puncture site complications after percutaneous coronary interventions
Andreja Černe, Igor Kranjec, Mirta Koželj, Jadranka Buturovic Ponikvar
Department of Cardiology, University Medical Center, Ljubljana, Slovenia
Femora/ «cress size complications (FASC) q/ier /oercuta/ieous comncuy inte/ven/io/is (PC!) may be./re//ue/it, //i//ic»/t to i//e/itifv, c//i// associate// with sig/iific«/i/ rnorbiV/iVv. To rc'/ine our catheterization technique, we pe/jor/ned //up/ex co/or .//mi' imaging (DCF!) i/i 42 «onsecu/ive /oa/ients u/ide/going diagnos/ic coronary arteriography (Gmu/o A) or PC/ (Grou/o B). In all /oa/ients, arteria/ (6-9F) and venous/e;/iiora/ sheaths (8F) were inserted /oy the Se/di/iger technique /rol/miW by' the /V bo/us oj' heparin (3000-10000 U). I/i Gmu/o A, the sheaths wee removed immediately q/ter the /orocedu/e. /n Grou/o B, he/oarin i/i/iisio/i (7000 U/hr) was started a^ie/' PC/, «/id the sheaths were /'ernoved the ./o/lowing morning. DCF/ was per/ormed the day q/ier catheterization usi/ig the 7.5 MHz /inear /ohased-army transduce,: A/worma/ /ohysica/ (g/ooi/i swe//i/ig, /iew /on»/) a/id ///trasou/i// fi/idi/igs (hematoma, /oseudoaneu/ryvm, AV ./istu/a) were recorded i/i l l /oa/ie/i/s (26.2 %) «/id 15 /o«/ie/its /35.7 %), res/oective/y //! patients with groin swe//ing or /iew hmi/s, DCF/ revea/e// hematomas, /oseiidoaneuiysnis, or AV./is/ul«s. FASC corre/ated with age >65 years (/o <0.05), sheath size (/o <O.Ol), «u// he/o«ri/i //os«ge (/o <O.O/). /n co/ic/usion, FASC occur coyie/i q/ier PC/, and can be easi/y ide/itif/ed /ov DCF/. Risk ./actors i/ic/ude advanced age, /urge sheaths, and more aggressive he/oari/i ad//ii/iis-/ration. Ear/y //erection using DCF/ wil/ minimize the nw//oidity associated with FASC.
Key words: angioplasty, transluminal, percutaneous coronary-adverse effects; coronary angiography-adverse effects; femoral vein-ultrasonography.
Introduction
Femoral access site complications (FASC) following percutaneous coronary interventions (PCI) add morbidity to the procedures and increase cosls by prolonging hospital stay. FASC have been reported with increasing frequency, partly due to older patients with more extensive peripheral vascular disease, as well as owing to the widespread use of anticoagulant/fibrinolytic therapy, and larger size vascular sheaths.1-11 Hematoma is the most common finding and represent a simple collection of blood wilh no communication with the punclured
Correspondence to: Andreja Cerne, M.D., Department of Cardiology, University Medical Center. Zaloska 7, 1000 Ljubljana, -Slovenia.
UDC: 616.132.2-173.75-06:616.1473
vessel.12 Pseudoaneurysm results from leakage of blood into the soft tissues around the femoral artery, with subsequent fibrous encapsulation and failure of the defect in the vessel wall to heal.11 AV fistula is an abnormal communication between the artery and the vein, occurring when a needle tract crosses bolh the artery and the vein and is dilated during catheterization.I4' 15 While hematoma is almost always spontaneously resorbed, pseudoaneurysm and AY fistula represent potentially unstable lesions, which may predispose to more serious sequelae. They may give origin to emboli; and may cause local infection, vascular thrombosis, or pressure effects and might rupture or cause high-output cardiac failure."' Several previous reports emphasized the limitations of physical findings, such as pulsatile groin mass, enlarging ecchymosis, abnormal puncture site pain and a new audible bruit in Ihe clinical setting of peripheral vascular inju-
166
CemeA elal.
ries.17 ls Nowadays, duplex color flow imaging (DCFI) is believed to be the leading noninvasive, highly specific and sensitive modality lor the diagnosis, planning of therapy and follow-up of polential FASC. l<'~2-, It allows lor simultaneous 2D imaging of anatomic structures and Doppler study of blood tlow, thus eliminating ihe need for invasive radiological procedures.
In order to refine our currenl catheterization technique. FASC were prospectively evaluated in our group of patients following diagnostic coronary arteriography or PCI. The aims of this study were threefold: 1) to detecl the frequency and nature of femoral access site complications in our group of patients: 2) to access ihe utilily of clinical signs in the diagnosis of vascular complications as verified by DCFI findings: 3) lo identify clinical and procedural factors that would predicl the likelihood of these complications in our catheterizalion laboratory.
Materials ancl methods
Cafhefe/vzafion p/r;ced/i/'e.v
42 conseculive patients undergoing diagnostic coronary arteriography (Group A, 22 palienls) or PCI with PTCA and slenting (Group B, 20 palienls) al ihe Department of Cardiology, University Medical Center. Ljubljana, were enrolled in the prospective study between October 12, 199.5 and February 20, 1996. Cardiac calheterizations were performed by lour skilled interventional cardiologists. In ali patients, arterial (Group A. 6F: Group B, 9F) and venous femoral sheaths (8F) were inserted using the Judkins approach by lhe double-wall Seldinger technique. followed by IV adminislralion of a heparin bolus (Group A, 3000 U: Group B. 10000 U). In Group A, lhe sheaths were removed immediately afler the diagnoslic procedure. In Group B. heparin infusion ( 1000 U/hr) was slarted after PCI and lhe sheaths were removed the following morning 4 hrs alter the discontinuation of heparin. The patienls were maintained on bed resl for 6 (Group A) or 12 hrs (Group B) thereafter.
C/inica/ derec'lion o{ vascular co//ip/ic.,alion.v
Following lhe removal of shealhs each patient was examined for evidence of prolonged bleeding, subcutaneous ecchymosis, groin swelling, access sile pain. vessel thromhosis and new femoral hruil.
Va.sr/i/a/' imaging
DCFI was routinely performed in ali patients with the Acuson computed color Doppler sonography system (Acuson 128. Acuson Corporation, Mountain View, Canada) employing the 7.5 MHz linear phased-array lransducer. The greatest diameter and the sile of the peripheral vascular complications were recorded. An echo-free mass with no communication wilh the vascular structures was considered to be a hematoma. Pseudoaneurysm was identified as an exlravascular hypoechogenic cavity communicating with the femoral arlery by a sinus tracl. displaying "to and fro" signal on pulsed Doppler examination (Figure I a). AV fistula was character-
Rffii

Figure la. A longitudinal color and pulsed Doppler image of the pseudoaneurysm (PA) at the ventral aspect of the superficial femoral artery (SFA) showing the cavity communicating with the artery by a sinus tract (arrow). and characteristic "to and fro" waveform.
ized by demonslraling a communication between the femoral artery and vein with a unidirectional conlinuous AV shunt represenling typically, high diastolic flow in lhe arlerial waveform proximal to lhe lislula and increased turbulent llow with arterial pulsalion in the draining vein (Figure I b).
C/inica/ dala
The recorded baseline clinical variables included palient age, gender. heighl, weight, body mass index (BMI), pulse pressure during catheterization, risk factors for alherosclerolic disease and the presence of femoral and aono-iliac artery disease. The procedural parameters included the type of cath-
Color dopplerfor the evaluation ofjmncture site complications after percutaneous coronary interventions 167
Figure lh. A transverse color and pulsed Doppler image ol' the AV fistula between the common lemoral artery (CFA) and the common femoral vein (CFV) .showing a communicating channel (arrow). and continuous. unidirectional. turbulent waveform.
eterization procedure, the size of indwelling vascular sheaths. the number of consecutive ipsilateral punctures. and the length of bed rest prior to ambulating. Thc amount of heparin administered during and after the procedure was also documented.
Statistics
Numeric continuous data were analyzed as thc mean ±1 SD. and categorical parameters were expressed as proportions. Differences in the distribution of selected characteristics hetween patients with DCFI-detected vascular complications and those without them were examined by the two-tailed Student t-test
and the Chi-square test for continuous and categorical variables, respectively. Univariate logistic regression was then performed to identify individual factors predicting the likelihood of vascular complications. Statistical significance was defined as a p value <0.05. Ali analyses were performed with the SPSS release 6.0 statistical package.
Results
Diagnosis
In 42 consecutive patients undergoing diagnostic coronary arteriography or PCI no major FASC, such as severe groin/abdominal hemorrhage, access site infection, sepsis or limb loss were observed. Abnormal physical ancl ultrasound findings were observed in 11 (26.2 %) and 15 (35.7 %) patients, respectively (Table I a). In 11 patients clinically suspected of having FASC, 7 hematomas, 1 pseudoaneurysm, rnid 3 AV fistulas were confirmed by DCFI. 4 (9.5 %) additional patients with no local signs of FASC were found to have a significant vascular pathology on routine DCFI examination. One from 8 hematomas (12.5 %), 2 from 3 pseucloaneurysms (66.6 %), and I of 4 AV fistulas (25.0 %) would be missed solely on clinical grounds.
Clinical assessment
Physical findings at the access site in 15 patients with the DCFl-conlirmed FASC are detailed in Table I b. In patients with large groin swelling measuring '2.7 cm, DFCI revealed 4 hematomas, but 1 pseudoaneurysm and I AV fistula. Though femoral
Table I a. Clinical and DFCI findings in patients following diagnostic coronary arteriography (Group A) and PCI (Group B).
Patients		Swelling	2 7 cm	New bruit Hematoma	Pseudoaneurysm AV fistula
Group A (n Group B (n Total (n	= 22) = 20) = 42)	4 patients 2 patients 6 patients	(18.2 %) (IO.O 'k) (14 %)	2 patients (9.1 %,) 4 patients (20.0 %,) patients (15.0 %) 4 patients (18.2 %) 5 patients (12 %) 8 patients (19.0 %,)	1	patient (4.5 %,) 1 patient (4.5 %) 2	patients (10.0 90) 3 patients (15.0 %) 3	patients (7.1 %) 4 patients (9.5 %)
Tahlc lh. Physical findings at the access site in 1.5 patients with DCFl-confirmed diagnosis of lemoral artery injury following diagnostic coronary arteriography or PCI.
Patients (n=l5)	Hematoma	Pseudoaneurysm	AV fistula
Ecchyrnosis	6/8 patients	1/3 patient	0/4 patient
Groin swelling	4/8 patients	1/3 patient	1/4 patient
Pulsation	0/8 patient	1/3 patient	0/4 patient
Tenderness	8/8 patients	3/3 patients	4/4 patients
New bruit	1/8 patients	1/3 patient	3/4 patients
168
Ceme A el a!.
arterial pulsations were transmitted from the femoral artery in all patients. a more market! pulsation that extended over the femoral artery boundaries was observed in only I patient with pseudoaneu-rysm. New bruits were detected in both patients with pseudoaneurysm (33.3 o/,) and AV fistula (35.0 %). but they were also recorded in 1 (12.5 %) patient with hematoma. The bruit associated with a pseudoaneurysm was systolic in nature; on the other hand. the bruits present in patients with AV fistula were described as continuous in 2 patients, and as "long-systolic" in I patient. Subcutaneous ecchy-mosis was associated with hematoma in 6, and with pseudoaneurysm in I patient. Mikl to moderate tenderness at the local puncture site was present in all patients.
DCF! characteristics
A total of 15 FASC included 8 hematomas (53.3 %), 3 pseudoaneurysms (20.0 %) ;md 4 AV fistulas (26.6 %). Most of hematomas were graded as small and measured 2.6 ± 0.6 cm in diameter. Pseudoaneurysms had an average dimension of 2.5 ± 0.5 cm and were located within I cm from the involved arterial segment. 2 pseudoaneurysms occurred at the ventral aspect of the superficial femoral artery, and one at the dorsal aspect of the common femoral artery in close proximity to its bifurcation. AV fistulas were located in superficial femoral artery in 3 cases. and in the common femoral artery in I case. They were considered to be hemodinamicaly unimportant, since no signs of venous hypertension or
proximal arterial dilatation were observed during DCFI examination.
Risk factors
FASC were more likely to result from PCI as compared to diagnostic coronary angiography (45.0 % versus 27.2 %, p < 0.05). A comparison of baseline and procedural variables between patients with DCFl-confirmed vascular complications and patients without complications are shown in Table 2. The studied groups did not differ as concerns sex, obesity. underlying predisposing risk factors and distribution of aorto-iliac artery disease. In both groups, blood coagulation tests before catheterization were within normal range. Pre-treatment with aspirin or dipyridamol did not influence the risk of FASC in this study. Among all patients. the strongest predictors of FASC were advanced age (p < 0.05), the use of large-size indwelling sheaths (p < 0.01) and higher dosage of heparin (p < O.O1). Univariate logistic regression was used to identify the clinical and procedural factors predicting the likelihood of FASC (Table 3). A dose-dependent regimen of heparin administration remained a statistically significant risk factor in all entities. White advanced age significantly contribute to hematoma formation, the use of large-size sheaths predicted the likelihood of AV fistula formation.
Management
All peripheral vascular lesions were managed non-operatively. Transfusion of 3 U of blood was ad-
Table 2. DFCI confirmed FASC as a function of clinical and procedural characteristics in patients undergoing diagnostic-coronary arteriography or PCI.
Characteristic-	Complications	Controls	p
	(11 = 15)	(n = 27)	
Age (years)	65.4±I0.0	56.2±II.6	0.0I3
Age > 6.5 years	40.0 %	25.9 %	0.029
Male gender	66.6 %	59.2 %	NS
BMI (kg/ m')	26.5 ±2.8	25.9 ±4.3	NS
Obesity (BMI >25 kg/m2)	40.0%	37.0 %	NS
Hypertension	66.6	62.9%	NS
Pulse pressure (mm Hg)	65.4 ±19.9	65.2 ±I9.9	NS
Diabetes mellitus	3.3 %	7.4%	NS
Hypercholesterolemia	53.3 %	48.I %	NS
Current smoking	33.3 %	59.2%	NS
Peripheral vascular disease	26.6%	25.9%	NS
Aspirin or dipyridamole before catheterization	80.0 %	85.I %	NS
Heparin during catheterization (10.000 U vs 3.000 U)	73.3 %	29.6%	0.006
Continuous heparin overnight (1000 U/h)	73.3%	29.6%	0.006
Sheath dimension (9 F vs 6 F)	66.6%	33.3 %	0.005
Consecutive number of ipsilateral arterial puncture	2.I±I.8	I.8 ±I.2	NS
Color dopp/crfor the era/nation ofptrncture .site complications after perciifaneous coronary interventions 169
ministered to only I patient with moderate hemorrhage into the retroperitoneal space. Most hematomas resorbed within a few days or weeks. depending on their size. One pseudoaneurysm showed spontaneous resolution as documented by the follow-up DCFI. Two stable pseudoaneurysms were successfully treated by DCFI-guided external compression. All AV fistulas appeared hemodinamicaly unimportant and were followed-up closely by DCFI.
Discussion
The reported incidence of FASC after diagnostic coronary arteriography varies widely, from less than 1 % to as much as 20 7-"> When PCI are included the figure is higher (6.1-24 %). possibly owing to the introduction of new devices. increased anticoagulation and larger sheath size.4-''- 11 Furthermore, owing to the difference in definition there is a wide discrepancy in the reported incidence of FASC. in particular those of hematoma (0.6-12.3 %) and bleeding (0.1-20.0 %).'••->■' Our 26.2 % incidence of FASC detected clinically ranks among the highest reported rates. It was mostly contributed to small stable hematomas, that would he excluded in other studies. The incidence of pseudoaneurysms (2.3 % clinical detection; 7.1 % DCFI) compares favorably with previously reported (0.5-5.2 clinical detection: 5.8-9.0 % DCFl).1- '-1 Our incidence of AV fistulas (7.1 % clinical detection: 9.5 % DCFI) was higher as compared to other studies (0.05-0.1 % clinical detection. 0.3-5.2 % DCFI)'" 2' and may be related to our common practice to perform the double-wall Seldinger technique and routine ipsilateral arterial and venous catheterization. However, in contrast to other reporls, no major FASC. such as severe hemorrhage. access sile infeclion. sepsis. limb loss or even death were observed in our study.
Our current methods for the detection of FASC included a routine clinical and DCFI examination. Clinically. FASC were suspected in 11 patients, bul DCFl-conlirmed in 15 patients. Four patients (9.5 %) with significant vascular pathology would be missed
on clinical grounds alone. Accordingly, we believed that clinical examination alone is not sufficient in order to detecl all FASC. Although several physical signs have been used to detect FASC, clinical examination alone may not reliably distinguish be-Iween different entities. The rale of diagnostic accuracy of any single physical sign detected in our study was low. In patienls with groin swelling, DCFI revealed hematomas. as well as pseudoaneurysms rnid AV lislulas. Although pulsatility of the mass appeared to be suggestive of pseudoaneurysm, it was present in only one case. As expected, new bruits suggested both pseudoaneurysm and in AV listula, bul simple hematoma did also cause a bruit due to extrinsic compression of the neighboring artery. Furthermore, a quality of bruit did not reliable identify its etiology.
The early recognition and prompl treatment of unstable vascular lesions are necessary to prevent further morbidity."'- Apart from progressive expansion and pain no objective criteria exist that would prospectively identify those pseudoaneury-sms that will ruplure and those that will spontaneously thrombose."' B In the present study, two stable pseudoaneurysms were successfully treated by DCFl-guided manual compression. Thus, we obviated the need for surgical interventions, and prevented spontaneous rupture in those patients followed conservatively.
The incidence of FASC was significantly higher alter PCI than after diagnostic coronary arteriography as confirmed previously."- (HS FASC were most closely associated with the use of large-diameter sheaths and higher amount of heparin administration. Significantly higher complication rale observed in patients of advanced age may be due to a somewhal delayed process of healing, and seems to implicale a rigidity and poor retractability of underlying vessels.
Conclusion and future outlook
FASC occur frequently alter PCI and can be easily identified by DCFI. Diagnosis based on clinical examination alone is rather inaccurate. An exact
Table 3. Univariale logislic regression for Ihe identification of individual clinical and procedural factors predicting the likelihood of hematoma. pseudoaneurysm and AV fistula.
Characteristic Hematoma Pseudoaneurysm AV fistula
Advanced age p<().()3 NS NS
Periprocedural regimen of heparin p <0.05 p <0.04 p <0.02 Sheath dimension_NS_NS_p<0.04_
170
Cerne A et a/.
diagnosis facilitates the choice of treatment and prevent.s further morbidity. Based on our initial experience, DCFl-guided compression seems to be a useful alternative to surgical closure of pseudoan-eurysm. Risk factors, identified in our study in-elude advanced age, large sheaths, and large amount of heparin administration. Monitoring of adjunctive anticoagulant therapy, and careful patients selection may reduce the morbidity and increase the safety of these procedures. It seems advisable to use a single-wall puncture technique, to avoid routine venous puncture, and to alternate the access site with repeat catheterization. As indicated by our study, meticulous attention to catheterization details will minimize complications related to the access, while early detection and management will reduce the morbidity associated to complications.
References
1.	Gwost J, Stoebe T, Chesler E, Weir EK. Complications of coronary arteriography from the collaborative study of coronary artery surgery (CASS). 0>n//at/»n 1979: 59:1 10.5-11.
2.	Skillrnan J.I. Kirn D. Baim DS. Vascular complications of percutaneous femoral cardiac interventions. Arch Surg 1988: 123: 1207-12.
3.	Babu SC. Picorrellil GO, Shash PM, Stein JH. Clauss RH. lncidence and results of arterial complications among 16,350 patients undergoing cardiac catheterization. ./ V/sc Surg 1989: 10: 1 L3-6T
4.	Kaufman J. Moglia R, Lacy C. Dinerstein C. Moreyra A. Peripheral vascular complications from percutaneous transluminal angioplasty: A comparison with trans-fernoral cardiac catheterization. /V/edta/ Sciences 1989: 297: 22-5.
5.	Oweida SW. Roubin GS. Smith RB, Salam AA. Posl-calhelerizalion vascular complications associated with percutaneous transluminal coronary angioplasty. .1 li/.sc Surg 1990: 12: 310-5.
6.	Muller DW, Podd J. Shamir KJ. Vascular access site complications in the era of complex interventions. C/r-rnlat/on 1992: 82: 5 1O.
7.	Mullet DW, Shamir KJ, Ellis SG, Topol EJ. Peripheral vascular complications alter conventional and complex percutaneous coronary procedures. 11/// ./ Card/»/ 1992: 69: 63-8.
8.	Fransson SG. Nylander E. Vascular injury following cardiac catheterization. coronary angiography, and coronary angioplasty. Ez/mpean Hew/ Journal 1994: 15: 232-5.
9.	Ricci MA, Trevisani GT, Pilcher DB. Vascular complications of catheterizations. The American ./»urna/ o( Sugerv 1994: 167: 357-78.
10. Bogart DB, Bogart M, Miller JT, Farrar MW, Barr K. Montgomery MA. Femoral artery catheterization complications: A study of 503 consecutive patients. Ca///e/ Caj///»va.vc D/ag» 1995: 34: 8-13.
1 1. Omoigui NA, Califf RM, Pieper K, Keller G. O' Hane-sian MA, Berdan LG, Mark DB et al. Peripheral vascular complications in the coronary angioplasty versus excisional atherectomy trial (CAVET). 711CC 1995: 26: 922-30.
12.	Altin RS, Flicker S, Naidech HJ. Pseudoaneurysm and arteriovenous lislula after femoral artery catheterization: Association with low femoral punctures. 11/R 1989: 152: 629-31.
13.	Mitchell DG. Femoral artery pseudoaneurysm: diagnosis with conventional duplex and color Doppler US. Radiology 1987; 165: 687-90.
14.	Heystraten FMJ. Fast JH. Arteriovenous fistula as a complication of the Seldinger procedure of the femoral artery and vein. Diagn linag 1983: 52: 197-20 l.
15.	Thadani U. Pratt AE. Profunda femoral arteriovenous fistula after percutaneous arterial and venous catheterization. Br Hear/ .1 1971: 33: 803-5.
16.	Mills JL. Wiedman JE, Robinson JG, Hallet JW. Minimizing mortality and morbidity from iatrogenic arterial injuries: the need for early recognition and prompt repair. ./ Have Surg 1986: 4: 22-7."
17.	Messina JL. Clinical characteristics and surgical management of vascular complications in patients undergoing cardiac catheterization. ./ \iivc Siiig 1989: 13: 593-600.
18.	Cohen GL Chan KL. Physical examination and echo Doppler study in the assessment of femoral arterial complications following cardiac catheterization. C«//;ev Gird/mar Diagn 1990: 21: 137-43.
19.	Reneman RS. Cardiovascular applications of multi-gate pulsed Doppler systems. Ul/ra.vo»»// in Me// and Biol 1986; 12: 357-70.
20.	Coughlin BF, Pausther DM. Peripheral pseudoaneurysm: evaluation with duplex US. Radiology 1988: 168: 339-42.
21.	Igidbashian VN, Mitchell DG, Middleton WD. Schwartz RA, Goldberg B. latrogenic femoral arteriovenous fistula: diagnosis with color Doppler imaging. Rad/o/ogy 1989: 170: 749-51.
22.	Mitchell DG. Femoral artery pseudoaneurysm: diagnosis with conventional duplex and color Doppler US. Radiolog- 1987: 165: 687-90.
23.	Sheikl KH. Utility of Doppler color llow imaging for identification of femoral arterial complications of cardiac catheterization. 11iii Hear/ .1 1989; 117: 623-8.
24.	Kirsh JD, Reading CC. Charboneau W. Ultrasound-guided compression and repair of postangiographic femoral arteriovenous fistulas. Mayo C/in Pmc 1993: 68: 612-3.