The brachial artery: A critical access for endovascular procedures
Article Outline
Objective
The brachial artery is often used for coronary angiography. However, data on brachial access for aortic and peripheral interventions are limited. This study evaluated our experience with brachial artery catheterization for diagnostic arteriography and endovascular interventions.
Methods
Between August 2004 and August 2005, 2026 endovascular procedures were performed. Of these, 323 cases (16%) in 289 patients required brachial artery access, forming the basis for this study. Patients who underwent multiple interventions, but with a single access (ie, thrombolysis), were considered a single case. Demographic and clinical data were recorded in a database and analyzed using logistic regression analyses with generalized estimating equations and the Fisher exact test for nominal variables.
Results
The mean age of all patients was 66.4 years, with 57% men. Brachial access was used for diagnostic purposes in 27% and for interventions including angioplasty, stenting, and thrombolysis in 73%. The use of brachial access was considered obligatory in 40%, adjunctive in 19% (ie, endovascular repair of abdominal aortic and thoracic aortic aneurysms) and preferential to femoral access in 41%. In 91% of patients, the brachial arteries were accessed percutaneously, and 9% underwent surgical cutdown for access. In patients whose brachial artery was approached percutaneously, access was achieved in all but one (99.6% technical success rate). Hemostasis after catheterization was achieved by manual compression in 89%. Operative mortality rate was 6.2% and not related to brachial artery access. Brachial access site–related complications occurred in 21 patients (6.5%). Thirteen of these 21 patients (62%) required a surgical correction, mostly for brachial artery thrombosis or pseudoaneurysm. Patients with complications were more commonly women (odds ratio [OR], 4.7; 95% confidence interval [CI], 1.68-13.26; P = .003) and had a long interventional sheath (OR, 6.7; 95% CI, 1.53-29.07; P = .012). The risk of a brachial artery complication was not associated with thrombolysis, procedure type, vascular territory treated, or the use of heparin. No upper extremity limb or finger loss occurred.
Conclusions
Brachial artery access is necessary for complex endovascular procedures and can be achieved in most patients safely. Postprocedural vigilance is warranted because most patients with complications will require operative correction.
Endovascular diagnostic and therapeutic procedures are generally performed through the femoral artery. Some of the reasons for this generalized approach are its location, easy approach for puncture and hemostasis, and low rate of complications. Femoral puncture also allows access to virtually all of the arterial territories and affords favorable ergonomics for the operator in most instances. In some situations, however, femoral access can be difficult or even contraindicated, such as in the absence of palpable femoral pulses, severe common femoral occlusive disease, recent femoral intervention or surgery, femoral aneurysms/pseudoaneurysms, and in some cases, the presence of prosthetic material. Access through other vessels, including the brachial, radial, and ulnar, has been used when femoral access is unavailable. Considerable experience has been achieved with these approaches for endovascular interventions of the coronary arteries, with excellent technical results and a low incidence of complications.1, 2, 3
The use of brachial artery access for noncoronary interventions, however, has received less attention. Despite its utility as an adjunctive4 and sometimes obligatory technique, some endovascular surgeons are reluctant to broaden its use for fear of an increased rate of complications. Several series have documented complication rates as high as 11%.5, 6, 7
The purpose of this study was to evaluate our experience with brachial artery catheterization in diagnostic and therapeutic endovascular procedures. In particular, our goals were to assess short-term technical success, safety, and the relationship between perioperative factors and complications related to brachial artery access.
Patient and methods
We reviewed our experience with endovascular procedures that required brachial access for diagnostic or therapeutic interventions at Cleveland Clinic from August 2004 to August 2005. During these 13 months, 2026 endovascular procedures were performed in endovascular suites with dedicated imaging equipment. This study comprised 323 cases (16%) in 289 patients who required brachial artery access. Patients who underwent multiple interventions but with a single access (ie, thrombolysis) were considered a single procedure. All cases of thrombolysis lasted >24 hours (mean, 30 hours).
All demographic, clinical, and perioperative data were obtained through review of hospital and physician computerized records. Demographic data points were collected that could potentially affect the peripheral vasculature in terms of size of the artery, degree of calcification, accelerated atherosclerosis, or perfusion. The data points studied included age, sex, and comorbidities of hypertension, hyperlipidemia, diabetes mellitus, coronary artery disease, history of congestive heart failure, presence of renal disease, chronic obstructive pulmonary disease, and smoking history. The numbers of observations were slightly variable due to missing data in the patient record. Thus, not all demographic and perioperative data points totaled to 289 patients or 323 cases. The operative reports and selected radiologic imaging were reviewed. Finally, time of last follow-up was recorded as well as the presence or absence of postoperative complications. Average length of follow-up was 18.6 (SD 11) months.
Operative technique
Diagnostic angiograms and endovascular procedures were performed by the faculty of the Cleveland Clinic Department of Vascular Surgery in dedicated endovascular suites with fixed imaging. The left brachial artery was accessed in 85% because it provided a more direct route to the descending thoracic aorta, abdominal aorta, and lower extremities and avoided catheter traversal of the aortic arch. Most arteries were accessed percutaneously using a micropuncture kit (Cook Inc, Bloomington, Ind). It is our practice to puncture approximately 1 cm above the antecubital crease. A strong pulse can usually be felt at this location, and it allows for the brachial artery to be compressed against the distal humerus for hemostasis at the end of the procedure. Ultrasound (US)-guided access was not used routinely, and was reserved for 27 instances where cannulation of the artery by palpation was unsuccessful.
Primary surgical cutdown was performed selectively in 29. This decision was made preoperatively according to the surgeon's preference. Indications for cutdown included a small or difficult to palpate brachial artery, or the placement of >7F sheath, or both. Procedures that involved brachial artery access included diagnostic angiograms, abdominal or thoracic aortic stent grafting, percutaneous angioplasty/stenting, and thrombolysis.
Indications for brachial artery access were differentiated according to whether the approach was by surgeon's preference, used obligatorily, or as an adjunct. Obligatory use of the brachial artery was defined as occurring in a situation where the femoral artery was unavailable, such as nonpalpable pulse, recent femoral intervention or surgery, presence of severe common femoral occlusive disease, femoral aneurysms, and in some cases, prosthetic material. The presence of prosthetic material at the femoral artery puncture site refers to Dacron or polyethylene bypass graft material. In some patients with an iliofemoral, femorofemoral, or aortobifemoral bypass, femoral artery puncture was considered contraindicated for fear of infecting the graft or inability to achieve hemostasis after puncture rendering a higher risk of pseudoaneurysm development.
Adjunctive use of the brachial artery was defined as a case in which the femoral artery had been accessed but an additional access through the brachial artery was deemed necessary for successful completion of the procedure. All vascular territories were treated, including brachiocephalic vessels, the aorta, the visceral vessels, and the lower extremity.
Most patients received heparin. Activated clotting time (ACT) levels were not monitored routinely except for carotid stenting and endovascular repair of abdominal or thoracic aortic aneurysms, our target ACT is ≥300 seconds. The level is checked every 30 minutes.
A long interventional sheath (>10 cm) was used in 56% of the procedures. The lengths of the long sheaths used during that period of time ranged from 55 to 90 cm. If the size of the long sheath used was not indicated in the operative reports, we chose 10 cm as our cutoff between a short and long sheath because 10 cm is the size of the short sheath that we routinely use for initial access. Sheath sizes used ranged from 4F to 9F. Manual pressure was used in virtually all cases of percutaneous access. The decision to remove the sheaths was surgeon dependent, and was often a judgement based on the amount of heparin given, the time interval between when the heparin was given and the conclusion of the case, and whether the heparin was reversed with protamine.
A surgical closure was used in 29 procedures in which the artery was accessed with a surgical cutdown. In 17 of these, sheaths >5F were used. In seven instances a surgical cutdown was used to close an artery that had been accessed percutaneously.
Data analysis and statistical methods
The primary goal was to assess the relationship between perioperative factors with brachial artery complications. To assess this relationship, a logistic regression model with generalized estimating equations using an exchangeable correlation structure was fit, allowing for the control of the correlation between observations from the same patient. When these models would not converge, due to levels with all complications or without any complications, a Fisher exact test was performed. The Fisher exact test did not adjust for the correlation between surgeries on the same patient. Analyses were performed using SAS 9.1 software (SAS Institute, Cary, NC), and a significance level of 0.05 was assumed for all tests. Multivariable analyses was explored, but abandoned due to the small number of complications.
Results
The mean age of all patients undergoing brachial artery access for endovascular procedures was 66.4 years (range, 23 to 90 years), and 57% were men. Most patients were hypertensive (82%) and had a history of smoking (76%). Patients were being treated for hyperlipidemia (67%), coronary artery disease (56%), diabetes mellitus (24%), and chronic obstructive pulmonary disease (24%). Patients were receiving the following medications preoperatively: aspirin (51%), clopidogrel (23%), Coumadin (15%); (Bristol-Myers Squibb, Princeton, NJ), a β-blocker (57%), and a statin (46%). The remainder of the demographic background of the patients in this study is given in Table I.
Table I. Demographic features for 289 study patients
| Variables | Percentage or mean |
|---|---|
| Age, y | 66.4 (range,23to90years) |
| Male gender | 57 |
| Hypertension | 82 |
| Hyperlipidemia | 67 |
| Coronary artery disease | 56 |
| Smoking history | |
| Active | 29 |
| Former | 47 |
| Congestive heart failure | 18 |
| Diabetes mellitus | 24 |
| Renal insufficiency | 29 |
| COPD | 24 |
| Preoperative medications | |
| Clopidogrel | 23 |
| Aspirin | 51 |
| Coumadina | 15 |
| β-Blocker | 57 |
| Statin | 46 |
aBristol-Myers Squibb, Princeton, NJ. |
In reviewing the indications for accessing the brachial artery, we found that brachial artery access was used as a matter of preference in 131 patients (41%), was considered obligatory in another 130 (40%), and was used as an adjunctive access to facilitate cases (ie, mostly endovascular abdominal aortic and thoracic aortic aneurysm repair) in 62 (19%). The procedures were elective in 77%. Brachial access was used for diagnostic purposes in 27% and for therapeutic intervention, including angioplasty, stenting, and thrombolysis in 73%. The distribution of the types of procedures is given in Fig 1, and the vascular territories treated are given in Fig 2.
Fig 1.
The distribution of indications for brachial access. EVAR, Endovascular aneurysm repair; PTA, percutaneous transluminal angioplasty/stenting.
Fig 2.
The distribution of vascular territories treated using brachial artery access.
Percutaneous access was used for brachial artery access in 91% of the procedures and primary cutdown in 9%. One brachial artery was unable to be accessed percutaneously, yielding a technical success rate of 99.6%. In that patient the brachial artery was accessed by needle puncture, but catheter introduction was unsuccessful. An angiogram through the 4F sheath revealed a small brachial artery unsuitable for the intervention, and the procedure was then achieved through an axillary access.
The left arm was used in 85% of the cases, the right arm in 13.8%, and both arms in 1.2%. The right brachial artery was catheterized in nearly all patients to treat thoracoabdominal pathology, such as with second-stage elephant trunk procedures, thoracic endovascular aneurysm repair requiring coverage of the left subclavian artery, dissection, or rupture. In some patients the right brachial artery was catheterized for ipsilateral ischemic disease of the upper extremity or for lower extremity disease if the left brachial artery had a weak pulse or an arteriovenous fistula or graft was in place. We did not observe any difference in frequency of complications between the arm used (left [7.0%] vs right [4.6%]; P = .81). In addition, there were no complications in the four patients who underwent bilateral puncture.
Ultrasound guidance was used in 27 procedures (8%) when access was difficult because of the patient's body habitus, a small artery, or a weak pulse. Heparin and protamine were used selectively. Heparin was given in 68% of cases and protamine was used to reverse the anticoagulation in 30%.
Sheath diameter varied according to the nature of the procedure. A 4F or 5F sheath was used in 61% of cases, a 6F or 7F sheath in 31%, and 8% required an ≥8F sheath. Short sheaths (<10 cm) were used in 44%, and long sheaths, which varied in length from 55 to 90 cm, were required for the endovascular procedure in 56%. No significant difference in the complication rate was noted among the various sheath sizes used.
Manual pressure was used to achieve hemostasis upon sheath removal in 89% of procedures. In the remaining, the artery was closed surgically. Conversion to surgical closure was needed in 2.3% when the artery was initially cannulated percutaneously. The decision to surgically repair an artery that had been percutaneously accessed was largely according to surgeon preference and comfort. Influencing factors included how small the artery was felt to be and the size of the interventional sheath. There were two notable cases. In one patient, the tip of the intravascular ultrasound catheter sheared and remained in the brachial artery. A cutdown was performed and the foreign body was successfully removed. In another patient, a 5F sheath was placed for a mesenteric angiogram. Sheath placement was successful but difficult. Upon surgical cutdown, the artery was noted to have transected and required repair with an interposition vein graft.
There was no significant difference in the incidence of adverse events and the technique of hemostasis of the puncture site between manual compression and surgical closure (6.6 vs 5.6%; P = .64). Of the two patients who had complications after surgical closure, both accesses were achieved percutaneously but required open closure of the arterial wall. The first corresponds to our one technical failure, a woman with acute mesenteric ischemia where the brachial artery was cannulated by puncture but efforts were unsuccessful for catheter introduction due to a severely diseased artery. The procedure was completed through an axillary approach, and the brachial artery was closed surgically. Re-exploration was required for continuous oozing postoperatively, with evidence of bleeding from a small vein.
The second occurred in a woman with chronic mesenteric ischemia who underwent superior mesenteric artery angioplasty for in-stent restenosis. The access was gained percutaneously but required conversion after the endovascular procedure was completed owing to a brachial laceration that required placement of an interposition vein graft. The postoperative complication identified was a small antecubital hematoma managed conservatively.
We did not observe any complications in those patients who underwent a primary cutdown for artery exposure and cannulation and subsequent surgical closure.
Overall technical success for the intended endovascular procedure, whether diagnostic or therapeutic, was achieved in 94.4%. The operative mortality rate was 6.2% and not related to brachial artery access. Complications related to the brachial access occurred in 21 patients (6.5%). Pseudoaneurysm was the most common complication, followed by brachial artery thrombosis, the latter only occurring in women (Table II). Overall, 13 of these patients (62%) required a surgical procedure to address the complication. There was no median nerve dysfunction or upper extremity limb or finger loss in this experience.
Table II. Complications related to brachial artery access
| Complication | Patients (n = 21), No. (%) | Need for surgical repair, No. |
|---|---|---|
| Pseudoaneurysm | 11(52) | 5 |
| Brachial artery thrombosis | 7(33) | 7 |
| Hematoma | 3(14) | 1 |
The only demographic feature that was associated with a higher rate of complications was female gender (odds ratio [OR], 4.7, P = .003). No other preoperative features were associated with increased complication risk, including diabetes mellitus (P = .28), renal disease (P = .65), and increasing body mass index (P = .28). The relationship between complication risk and certain medications, such as anticoagulants and antiplatelets, was also studied (Table III). Interestingly, the use of aspirin was associated with lower rate of brachial artery complications (OR, 0.3; P = .015). Patients who were prescribed Coumadin preoperatively appeared to have a trend for increased complications (OR, 2.4; P = .086).
Table III. Relationship between preoperative variables and incidence of complications in 323 patientsa
| Variable | Group | No. | Complications, No (%) | OR (95% CI) | P |
|---|---|---|---|---|---|
| Female | No | 184 | 5(2.7) | 1.00(1.00-1.00) | .003 |
| Yes | 139 | 16(11.5) | 4.72(1.68-13.26) | ||
| Hyperlipidemia | No | 85 | 6(7.1) | 1.00(1.00-1.00) | .93 |
| Yes | 220 | 15(6.8) | 0.96(0.36-2.58) | ||
| CAD | Low | 139 | 10(7.2) | 1.00(1.00-1.00) | .62 |
| Intermediate | 148 | 8(5.4) | 0.74(0.28-1.94) | ||
| High | 30 | 3(10) | 1.45(0.37-5.75) | ||
| CHF | No | 256 | 16(6.3) | 1.00(1.00-1.00) | .83 |
| Yes | 57 | 4(7) | 1.13(0.36-3.55) | ||
| CVD | No | 240 | 14(5.8) | 1.00(1.00-1.00) | .43 |
| Yes | 72 | 6(8.3) | 1.50(0.55-4.11) | ||
| DM | No | 241 | 18(7.5) | 1.00(1.00-1.00) | .28 |
| Yes | 77 | 3(3.9) | 0.50(0.14-1.76) | ||
| Hypertension | No | 52 | 3(5.8) | 1.00(1.00-1.00) | .81 |
| Yes | 266 | 18(6.8) | 1.17(0.33-4.19) | ||
| Tobacco | None | 45 | 2(4.4) | 1.00(1.00-1.00) | .70 |
| Former | 151 | 10(6.6) | 1.51(0.32-7.20) | ||
| Active | 97 | 8(8.3) | 1.94(0.39-9.63) | ||
| COPD | No | 231 | 14(6.1) | 1.00(1.00-1.00) | .33 |
| Yes | 75 | 7(9.3) | 1.60(0.62-4.16) | ||
| Aspirin | No | 134 | 15(11.2) | 1.00(1.00-1.00) | .015 |
| Yes | 166 | 6(3.6) | 0.30(0.11-0.79) | ||
| Clopidogrel | No | 227 | 17(7.5) | 1.00(1.00-1.00) | .52 |
| Yes | 76 | 4(5.3) | 0.69(0.22-2.13) | ||
| Coumadinb | No | 256 | 15(5.9) | 1.00(1.00-1.00) | .086 |
| Yes | 46 | 6(13) | 2.39(0.88-6.44) | ||
| β-Blocker | No | 119 | 8(6.7) | 1.00(1.00-1.00) | .91 |
| Yes | 185 | 13(7) | 1.05(0.42-2.63) | ||
| Statin | No | 114 | 8(7) | 1.00(1.00-1.00) | .60 |
| Yes | 148 | 8(5.4) | 0.76(0.28-2.09) |
aThe total number of observations varies slightly due to missing data in the patient record. |
bBristol-Myers Squibb, Princeton, NJ. |
The only intraoperative variable that was associated with a higher rate of complications was use of a long (≥10 cm) interventional sheath (OR, 6.7; P = .012). Larger sheath sizes did not translate into a higher complication risk (P = .23), although there was a trend toward higher complication risk with 6F/7F sheaths (OR, 2.3; P = .094). The risk of a brachial artery complication was not associated with procedure type, including thrombolysis (P = .23) or the vascular territory that was treated (P = .47). Also, the use of heparin or protamine was not associated with an increased complication risk (Table IV).
Table IV. Relationship between intraoperative variables and rate of complications
| Variable | Group | No. | Complications, No. (%) | OR (95% CI) | P |
|---|---|---|---|---|---|
| Indication | Elective | 247 | 19(7.7) | 1.00(1.00-1.00) | .14 |
| Urgent | 76 | 2(2.6) | 0.33(0.08-1.43) | ||
| Vascular | Carotid | 11 | 1(9.1) | 1.00(1.00-1.00) | .47 |
| Territory | Aorta | 126 | 6(4.8) | 0.50(0.06-4.51) | |
| Viscerals | 92 | 9(9.8) | 1.09(0.13-9.40) | ||
| LExt | 94 | 5(5.3) | 0.57(0.06-5.24) | ||
| US guided | No | 296 | 17(5.7) | 1.00(1.00-1.00) | .077 |
| Yes | 27 | 4(14.8) | 2.86(0.89-9.15) | ||
| Heparin use | No | 104 | 7(6.7) | 1.00(1.00-1.00) | .92 |
| Yes | 219 | 14(6.4) | 0.95(0.37-2.43) | ||
| Protamine | No | 226 | 18(8.0) | 1.00(1.00-1.00) | .12 |
| Yes | 97 | 3(3.1) | 0.37(0.11-1.29) | ||
| Intervention | No | 109 | 4(4.6) | 1.00(1.00-1.00) | .32 |
| via BA | Yes | 214 | 16(7.5) | 1.68(0.60-4.69) | |
| Sheath diameter | 4F-5F | 186 | 8(4.3) | 1.00(1.00-1.00) | .23 |
| 6F-7F | 95 | 9(9.5) | 2.33(0.87-6.23) | ||
| ≥8F | 23 | 1(4.4) | 1.03(0.12-8.48) | ||
| Sheath length | Short | 133 | 2(1.5) | 1.00(1.00-1.00) | .012 |
| Long | 172 | 16(9.3) | 6.67(1.53-29.07) | ||
| Hemostasis | Manual | 286 | 19(6.6) | 1.00(1.00-1.00) | .89 |
| Surgical | 36 | 2(5.6) | 1.35(0.38-4.85) |
Discussion
Brachial artery access is a critical component of complex endovascular procedures. The routine use of femoral access is widespread secondary to its technical ease, wide applicability, and relative patient comfort. In our experience of >2000 endovascular cases in a 1-year period, femoral cannulation was also routine; however, brachial access was used in 16% of procedures. This may be due to the emergence of newer and more complex techniques that require multiple access points to achieve procedural success.
Brachial access is a straightforward procedure with a high success rate for percutaneous cannulation. Some technical points are worthy of mention. Our practice is to routinely use a micropuncture system with puncture just proximal to the antecubital fossa. This is visualized fluoroscopically over the olecranon process. High bifurcation of the brachial artery or high origin of the radial artery from the brachial artery is the most frequent arterial variation of the upper extremity.8 We are not aware of any instance in our series in which the radial or ulnar vessel was punctured; inadvertent puncture may have occurred, but to our knowledge this was not associated with any complications. Heparinization is instituted after placement of a sheath. Braided, flexible long sheaths are used for interventions, and manual pressure is used for hemostasis.
Technical success for percutaneous brachial access in our series was 99.6%, with only one failure. Other groups have reported similar rates. Basche et al9, 10 described a technical success rate of 99.5% for diagnostic and 94% for therapeutic procedures, respectively. Similar findings in diagnostic procedures were described by Chatziioannou et al,11 who reported a 99.52% success rate in catheterizing one of the brachial arteries for lower extremity arteriography. This group had extensive experience with this approach with 2250 brachial artery punctures. These procedures, however, used a 4F pigtail catheter without placement of a sheath.11
Complications occurred in 21 of 323 procedures (6.5%). We sought to identify factors associated with a complication and found no evidence that any of the preoperative demographic variables, other than female gender, significantly affected the incidence of brachial access related complications. Women had an 11.5% complication rate compared with 2.7% in men (P = .003). Complications were also managed differently in women compared with men: an open surgical procedure was required in 75% of women but in only 20% of men to resolve the complication. It is important to emphasize that the seven complications with brachial artery thrombosis occurred in women, and all of them required surgical correction. In general, this correlates with data in the femoral artery access literature. Applegate et al12 reported an incidence of vascular complications of 1.0% in men vs 2.0% in women (P < .05) after coronary interventions through a femoral approach.12 The largest report of vascular complications after coronary interventions, the ACC-NCDR, reported a twofold increase in the risk of any vascular complication for women compared with men.13 This may simply be due to smaller brachial artery diameter and endothelial damage from sheaths, catheters and guidewires.
An interesting finding was the positive effect of aspirin use on the incidence of complications (P = .015). Perhaps an antiplatelet effect decreased the likelihood of brachial thrombosis without an increased risk of significant bleeding. However, patients with history of Coumadin use trended toward increased complications (P = .086).
There is a general reluctance to puncture the right brachial artery due to the need to navigate through the innominate artery and arch. This may put multiple cerebral arteries (carotid and vertebral) at risk for embolization. However, we did not identify any patient with a cerebrovascular accident complication, and found no difference in the incidence of complications when we compared the two arms for brachial access. Several groups with experience using the right brachial system have suggested that it is not associated with an increased risk of cerebrovascular complications.11, 14, 15
The risk for vascular complications has been reported to be higher in those patients undergoing therapeutic interventions compared with diagnostic procedures in coronary12 and peripheral vascular cases.16 Similarly in our experience, the use of long sheaths (>10 cm long) was associated with a significant risk of puncture site complications (9.3 vs 1.5%; P = .012). The introduction and manipulation of these sheaths to reach distant targets may add additional strain on the relatively mobile brachial artery. We were surprised that we did not find a significant relationship between the incidence of complications and the sheath size. There was a trend toward increased complications in the 6F to 7F range, although not statistically significant. That trend reversed itself with the use of sheaths >8F; however, with those cases that required the use of a sheath ≥8, the sheath was introduced by primary cutdown in 11 (48%). Nonetheless, no complications occurred in the percutaneously accessed arteries with an 8F sheath in place.
The 3.4% rate of complications in diagnostic procedures in our series is within the 0.44% to 11% range reported by other groups.6, 11, 17, 18, 19 In our series, surgical intervention for local brachial artery complications in both diagnostic and therapeutic procedures was necessary in 4% of the procedures. One patient of 87 (1.1%) undergoing diagnostic angiography needed a surgical procedure for a brachial-related complication.
Ultrasound-guided cannulation has been used for several years as an adjunct for arterial and venous cannulation. In particular, with regard to internal jugular vein access, the success rate is improved, access time is decreased, and the complication rate is reduced significantly.20 We used US imaging selectively, mostly in difficult cases. We observed four complications in the 27 cases of US-guided cannulation (14.8%) vs 17 events in 296 palpation-guided (5.7%) procedures. This finding should be taken with caution, however, considering that we do not use US imaging routinely in all patients, reserving its use only for those arteries difficult to cannulate by palpation. We advocate the use of US-guided access in complicated procedures, but our data would indicate that this may serve as a warning that an increased risk of brachial artery complication may occur.
Some groups have described the successful use of closure devices for brachial punctures.21, 22 Our experience with the closure devices in the brachial location is limited, and none were used in this particular series. The currently available devices have been designed for femoral puncture closure exclusively. Most require an adequate subcutaneous tissue thickness to be deployed safely, as well as an arterial diameter that is more generous than in most brachial arteries. Brachial specific closure devices may be warranted with increased use of this access site.
Pseudoaneurysm formation complicates 0.6% to 6% of femoral arterial catheterizations.23, 24, 25 Risk factors for postcatheterization femoral pseudoaneurysm have been described and include use of anticoagulation, large sheath size, hypertension, obesity, faulty puncture technique, inadequate manual compression after catheter removal, arterial calcification, simultaneous arterial and venous cannulation, female gender, and hemodialysis.25, 26, 27 Also, interventions requiring large sheath size and use of periprocedural anticoagulation may increase pseudoaneurysm formation because they are less common after diagnostic angiography.25, 26, 27 In this series, brachial pseudoaneurysm (n = 11) was the most frequently observed complication after brachial artery access. The incidence of brachial pseudoaneurysms was 5.7% in women and 2.2% in men; this difference was not significant (P = .17). We identified two pseudoaneurysms (2.3%) after 87 diagnostic procedures. Although only five required surgical repair, four still required an intervention with US-guided thrombin injection. One patient was managed with US-guided compression, and the last one resolved spontaneously with observation. No median nerve injuries occurred in this series. The incidence of this complication after right brachial artery puncture for cardiac catheterization was 0.2% to 1.4% in the cardiology literature.28 The mechanism of injury was direct nerve compression from hematoma, direct nerve trauma, and ischemia from brachial artery occlusion. We credit the absence of median nerve injury in our series to the fact that our complications were noted early and treated accordingly in the postoperative period.
There are limitations to this study. This is a retrospective, nonrandomized study in a referral center where many of the patients are from out-of-state locations. Thus, consistent follow-up is difficult and complications may have been missed. Nevertheless, most patients did follow-up with us with a mean follow-up of >18 months. Choice of percutaneous vs cutdown brachial access (n = 29) and other technical details were left to the surgeon treating the patient, leading to inherent biases. Clinical testing was not standardized and left to the discretion of the treating physician. Thus, all patients did not have a brachial artery duplex or other noninvasive imaging; this was performed based on clinical suspicion. Lastly, conclusions regarding the use of antiplatelet and anticoagulant agents should be made with caution. Because this was a retrospective review, we did not have consistent documentation on how long oral anticoagulants and antiplatelet agents were stopped before the procedure. In addition, some patients who were known to be taking Coumadin did not have an international normalized ratio available on the day of procedure to know if they had been fully reversed.
In conclusion, brachial artery access is necessary for complex endovascular procedures and can be achieved in most patients safely. This technique expands our capability to perform complex procedures by allowing us to reach arterial targets in all territories. Female gender was associated with an increased risk of complications, as was use of a long sheath. The risk of a brachial complication is low, but postprocedural vigilance is warranted in all cases of brachial artery access since a majority of patients with a complication will require operative correction.
Author contributions
References
- . A randomized comparison of percutaneous transluminal coronary angioplasty by the radial, brachial and femoral approaches: the access study. J Am Coll Cardiol. 1997;29:1269–1275
- Brachial, radial, or femoral approach for elective Palmaz-Schatz stent implantation: a randomized comparison. Cathet Cardiovasc Diagn. 1997;41:124–130
- . Transulnar access for coronary angiography and intervention: an early review to guide research and clinical practice. J Invasive Cardiol. 2007;19:83–87
- . Brachial artery catheterization to facilitate endovascular grafting of abdominal aortic aneurysm: safety and rationale. J Vasc Surg. 2000;32:1137–1141
- . Transbrachial arteriography: techniques and complications. Cardiovasc Intervent Radiol. 1988;11:32–35
- . Complications of direct brachial artery puncture for arteriography: a comparison of techniques. Clin Radiol. 1991;44:189–191
- . Transbrachial arteriography: indications and complications. Clin Radiol. 1996;51:205–209
- . Arterial patterns of the human upper limb: update of anatomical variations and embryological development. Eur J Anat. 2003;21–28
- . Transbrachial angiography: an effective and safe approach. Vasa. 2004;33:231–234
- . The brachial artery as approach for catheter interventions--indications, results, complications. Vasa. 2004;33:235–238
- . Complications of lower-extremity outpatient arteriography via low brachial artery. Cardiovasc Intervent Radiol. 2004;27:31–34
- Vascular complications in women after catheterization and percutaneous coronary intervention 1998-2005. J Invasive Cardiol. 2007;19:369–374
- Risk of local adverse events following cardiac catheterization by hemostatic device use and gender. Invasive Cardiol. 2004;16:459–464
- . Complications of outpatient transbrachial intraarterial digital subtraction angiography. Radiology. 1987;162:125–127
- Transbrachial artery approach for selective cerebral angiography in outpatients. Am J Neuroradiol. 1988;9:334–336
- . Risk of outpatient angiography and interventional procedures: a prospective study. AJR Am J Roentgenol. 2004;183:377–381
- . Complication rates of percutaneous brachial artery access in peripheral vascular angiography. Ann Vasc Surg. 2003;17:107–110
- . Screening for occlusive vascular disease with intra-arterial DSA: preliminary experience with a high flow 4-F catheter. Radiology. 1984;153:823
- . Outpatient DSA in cerebrovascular disease using transbrachial arch injections. AJR Am J Roentgenol. 1985;145:941–947
- . Ultrasound-assisted cannulation of the internal jugular vein (A prospective comparison to the external landmark-guided technique). Circulation. 1993;87:1557–1562
- . Off-label use of Angio-Seal vascular closure device for brachial artery puncture closure-deployment modification and initial results after transbrachial PTA. Eur J Vasc Endovasc Surg. 2006;31:431–433
- . Use of a collagen-based device for closure of low brachial artery punctures. Cardiovasc Intervent Radiol. 2007;30:273–275
- . Complications of angiography. Radiology. 1981;138:273–281
- A prospective study of the incidence and natural history of femoral vascular complications after percutaneous transluminal coronary angioplasty. J Vasc Surg. 1991;13:328–333
- A prospective evaluation of surgically treated groin complications following percutaneous cardiac procedures. Am Surg. 1994;60:132–137
- . Current treatment methods for postcatheterization pseudoaneurysms. J Vasc Interv Radiol. 2003;14:697–710
- . Incidence of pseudoaneurysm after diagnostic and therapeutic angiography. Radiology. 1995;195:463–466
- . Median nerve injury: an underrecognised complication of brachial artery cardiac catheterisation?. J Neurol Neurosurg Psychiatry. 1997;63:542–546
Competition of interest: none.
PII: S0741-5214(08)01601-7
doi:10.1016/j.jvs.2008.09.017
© 2009 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
