Article Text

Download PDFPDF

Original research
Predicting the degree of difficulty of the trans-radial approach in cerebral angiography
  1. Nickalus R Khan1,2,
  2. Jeremy Peterson1,2,
  3. David Dornbos III1,2,
  4. Vincent Nguyen1,2,
  5. Nitin Goyal3,4,
  6. Radmehr Torabi1,2,
  7. Daniel Hoit1,2,
  8. Lucas Elijovich3,4,
  9. Violiza Inoa-Acosta3,4,
  10. David Morris5,
  11. Christopher Nickele1,2,
  12. Pascal Jabbour6,
  13. Eric C Peterson7,
  14. Adam S Arthur1,2
  1. 1 Neurosurgery, University of Tennessee Health Science Center, Memphis, Tennessee, USA
  2. 2 Neurosurgery, Semmes-Murphey Neurologic and Spine Institute, Memphis, Tennessee, USA
  3. 3 Neurology, University of Tennessee Health Science Center, Memphis, Tennessee, USA
  4. 4 Neurology, Semmes-Murphey Neurologic and Spine Institute, Memphis, Tennessee, USA
  5. 5 Mid-South Imaging and Therapeutics PA, Memphis, Tennessee, USA
  6. 6 Neurological Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania, USA
  7. 7 Neurological Surgery, University of Miami, Miami, Florida, USA
  1. Correspondence to Dr Adam S Arthur, Semmes-Murphey Neurologic and Spine Institute, Memphis, TN 38120-2300, USA; aarthur{at}


Background To evaluate anatomical and clinical factors that make trans-radial cerebral angiography more difficult.

Methods A total of 52 trans-radial diagnostic angiograms were evaluated in a tertiary care stroke center from December 2019 until March 2020. We analyzed a number of anatomical variables to evaluate for correlation to outcome measures of angiography difficulty.

Results The presence of a proximal radial loop had a higher conversion to femoral access (p<0.03). The presence of a large diameter aortic arch (p<0.01), double subclavian innominate curve (p<0.01), left proximal common carotid artery (CCA) loop (p<0.001), acute subclavian vertebral angle (p<0.01), and absence of bovine aortic arch anatomy (p=0.03) were associated with more difficult trans-radial cerebral angiography and increased fluoroscopy time-per-vessel.

Conclusion The presence of a proximal radial loop, large diameter aortic arch, double subclavian innominate curve, proximal left CCA loop, acute subclavian vertebral angle, and absence of bovine aortic arch anatomy were associated with more difficult trans-radial cerebral angiography. We also introduce a novel grading scale for diagnostic trans-radial angiography.

  • catheter
  • CT Angiography
  • guidewire
  • artery

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Statistics from


There has been a recent shift from trans-femoral access (TFA) to trans-radial access (TRA) in interventional cardiology,1–6 and more recently in neurointervention.7–13 There are clear benefits to patients, including a reduction in serious access site complications, decreased length of stay, decreased cost, and improved patient satisfaction.4 5 Formal guidelines in interventional cardiology recommend radial access as the first choice for cardiac procedures.6 This recommendation is based on evidence from randomized trials showing decreased morbidity and mortality when utilizing TRA compared with TFA.6 There have been several recent studies describing the success of TRA in neurointervention.2 7 8

Despite these benefits, there are cases in which trans-radial cerebral angiography can prove difficult, requiring conversion to alternative access. The purpose of this study was to analyze the relationship between several anatomical variables and measures of angiography difficulty.


We performed a retrospective review of a prospectively collected database from our institution between December 2019 until March 2020. This database included all cases in which diagnostic trans-radial angiography was performed. At our institution, there is a subset of these patients who also had computed tomography (CT) angiograms of the aorta and neck performed prior to their catheter angiogram, which are often used to analyze the difficulty of a radial case preoperatively. The primary operators were neurointerventional fellows under direct faculty supervision with a minimum of 300 cases performed prior to their involvement with this study. We obtained Institutional Review Board (IRB) approval for this study; we also received IRB approval to waive consent because of the retrospective nature of the data.

Anatomical evaluation

Patients were included in the study if they had a CT angiogram that included the aortic arch available prior to the procedure or an arch aortogram performed during the procedure. CT angiographic data were transferred to a separate three-dimensional (3D) workstation and reconstructed for the purpose of making measurements. If an aortogram was performed, measurements were taken directly from the angiographic images. Measurements from the reconstructed CT angiographic and arch aortograms were made by two independent certified neurointerventionalists. In the event of disagreement, a third neurointerventionalist reviewed the data for consistency. The input variables analyzed are included in table 1.

Table 1

Input variables defined

The input measures listed in table 1 were next correlated with the fluoroscopy time-per-vessel, the number of catheters used during the case, and a simple operator rating scale of difficulty for trans-radial diagnostic angiography cases (table 2).

Table 2

Operator rating scale used to assess difficulty of diagnostic angiography cases at our institution at the conclusion of each case

Radial angiography technique

Our technique for trans-radial angiography includes ultrasound-guided puncture of the radial artery (either distal or proximal), followed by insertion of a 5 French (Fr) Glidesheath Slender (Terumo Interventional Systems, Terumo Medical Corporation North America, Somerset, NJ) 10 cm hydrophilic coated introducer sheath, utilizing the Seldinger technique. Proximal radial and distal brachial angiography is performed to assess for tortuosity or radial loops. A radial “cocktail” of 2.5 mg verapamil, 200 μg nitroglycerin, and 2500 IU heparin is infused over 1 min. Utilizing roadmap angiography, the radial and brachial vasculature is navigated. We use a 5 Fr Radiofocus Glidecath (Terumo Interventional Systems, Terumo Medical Corporation North America, Somerset, NJ) Simmons/Sidewinder 2 glide catheter and a 0.035 inch Radifocus Glidewire (Terumo Interventional Systems, Terumo Medical Corporation Europe, Leuven, Belgium) with an angled tip initially on all cases. If a radial loop is encountered, the 0.035 inch Glidewire is removed and a 0.035 inch Radifocus Glidewire with an angled 1.5 mm J tip “Baby J” is utilized in order to bypass the loop. If this wire is unsuccessful, a 0.035 inch Radiofocus Glidewire with an angled 3 mm J tip is attempted. If the additional wires are unsuccessful, a longer sheath is inserted.

Radial vasospasm is managed with additional intra-arterial verapamil in addition to the dose given initially. Occasionally, a warm compress or percutaneous lidocaine is used if extreme vasospasm is encountered. Rarely, in the event of severe vasospasm, the case is abandoned for another site of access. The amount of vasospasm is limited mainly by careful wire and catheter manipulation using roadmap angiography of the proximal radial and distal brachial vasculature.

Our institutional bias for catheter angiography—whether performed femorally or radially—is to perform a four-vessel angiogram, including common carotid arteries (CCAs) and vertebral arteries bilaterally, unless the clinical situation warrants evaluation of the external carotid artery (ie, evaluation for bypass or evaluation of an arteriovenous fistula). We often do not access the distal internal carotid artery unless clinically necessary due to the significant amount of atherosclerosis that is present in our local patient population.

If unsuccessful with the 5 Fr Radiofocus Glidecath Simmons/Sidewinder 2 catheter, we then use a 5 Fr Radiofocus Glidecath Berenstein catheter to catheterize the right vertebral artery or a 5 Fr Radiofocus Glidecath Simmons/Sidewinder 3 glide catheter to catheterize the left vertebral artery. Another technique we employ is to use a dual wire anchoring technique to catheterize the left vertebral artery. Two 0.018 inch Radiofocus Glidewire wires are used. The initial wire is directed distally into the left subclavian artery, allowing the catheter to be anchored in the left subclavian for access of the left vertebral artery with the other glide wire. The initial wire is then retracted and used to select the vertebral artery as well, allowing the diagnostic catheter to advance over the two 0.018 inch wires.

Statistical analysis

SPSS version 21 (Aramonk, NY) was used for statistical analysis. The Kolmogorov-Smirnov and Shapiro-Wilk tests were used to assess whether data followed the normal distribution. Categorical data that followed the normal distribution were evaluated using the analysis of variance (ANOVA) test and χ2 test. Categorical data that did not follow the normal distribution were evaluated using the Mann-Whitney U test and Kruskal-Wallis test. Continuous variables were analyzed using Pearson’s correlation test. Descriptive statistics were calculated using Microsoft Excel.


Initial clinical variables

A total of 52 patients who underwent conventional trans-radial cerebral diagnostic angiography from December 2019 through March 2020 were included in the study. Table 3 lists the overall patient and case demographic information.

Table 3

Patient and case characteristics for the entire cohort of patients (n=52)

There were two cases of radial vasospasm (4%) identified in our case series, one of which resulted in conversion to femoral access (table 4). There were no access site complications in these 52 cases. A total of three (6%) cases required conversion to femoral access (table 4) and three (6%) cases were rated as having grade 3 difficulty.

Table 4

Reasons for conversion from radial access

Correlation of anatomical variables with outcome measures of radial angiography difficulty

Table 5 provides a summary of all of the anatomical input variables compared with the outcome measures of TRA difficulty, which included fluoroscopy time-per-vessel, number of catheters used, conversion to femoral access, and operator rating scale of difficulty. (Please see the online supplementary material section for a detailed description of all the results.)

Table 5

Input variables, outcome measures, and statistical correlations*


The primary finding in this study was that specific anatomic variables correlated with several outcome measures of radial angiography difficulty. Specifically, the presence of a radial proximal loop, double subclavian-innominate curve, presence of bovine anatomy, a proximal left CCA loop, an acute subclavian-vertebral angle, and a large diameter aortic arch were correlated with outcome measures of trans-radial cerebral angiography difficulty (table 6). The results of this study have led us to propose a novel grading scale—the “TRA” grading scale.

Table 6

The trans-radial angiography “TRA” grading scale*

Tortuosity in subclavian-innominate anatomy

The presence of a double subclavian-innominate curve (Kruskal-Wallis, p<0.01) was associated with a more difficult TRA case using the operator scale of difficulty. The presence of a double subclavian-innominate curve (figure 1) often causes a loop in the catheter and reduces torquability, further causing a loss of distal catheter control. This impacts the ability to form the Simmons shape in a large diameter aortic arch or a type 2 or 3 aortic arch. The reduced torquability also makes vessel selection more difficult.

Figure 1

Double subclavian-innominate curve.

The presence of a left common carotid artery loop (figure 2A) was associated with a higher score on the operator level of difficulty scale (χ2, p<0.001) and a higher likelihood of conversion to femoral access (χ2, p<0.01). A loop in the proximal CCA makes it difficult to navigate a diagnostic catheter (especially on the left side) due to a lack of support and an inability to maintain a stable position without herniating back into the aortic arch.

Figure 2

(A, B) Proximal common carotid artery loop. This configuration makes it difficult to maintain a stable configuration of the Simmons catheter shape while obtaining vessel access, particularly for the left side.

Radial anatomy

The presence of a proximal radial loop (figure 3) was associated with a higher conversion to femoral access (p<0.03). Three of our cases (6%) had a proximal radial loop. This observed number is higher than previously reported in larger series in the cardiology literature, which ranged from 0.04% to 2.4%.14–19 These loops can often be navigated using careful technique. However, as in one of our cases (table 3), conversion to femoral access is sometimes prudent due to significant vasospasm secondary to manipulation of the radial loop.

Figure 3

(A) Proximal radial loop with accessory radial artery (arrow) at the apex of the loop. (B) Radial loop that was not able to be reduced (C).

Aortic arch anatomy

A larger diameter of the aortic arch was associated with more fluoroscopy time-per-vessel (Pearson’s coefficient, p<0.01) and an increased operator rating of difficulty (ANOVA, p<0.01). The presence of a large diameter aortic arch can make reforming a Simmons catheter more difficult in the setting of proximal tortuosity due to the catheter being less constrained and more difficult to manipulate (figure 4).

Figure 4

(A,B) A large diameter aortic arch that is not allowing the wire to progress into the descending aorta for efficient formation of the Simmons catheter.

The presence of bovine anatomy was associated with a lower number of catheters used (χ2, p=0.03), but was not associated with an overall decrease in operator rating of difficulty, number of catheters, or fluoroscopy time-per-vessel. In our experience, we are often able to directly access vessels without forming the Simmons catheter configuration when bovine anatomy is present, which we feel makes a radial case “easier”. Our study, however, is underpowered to show a statistical correlation between specific factors and ease of a case. Factors that make subtle differences in the ease of performing a diagnostic radial angiogram require a larger sample size to show statistical significance. It is far easier to show cases that result in high radiation dosages, fluoroscopy time, and conversion than it is to show subtle differences between a standard case and an “easy” case. Interestingly, the total radiation dosage was shown to be significantly lower when bovine anatomy was present.

Additional factors

The presence of an acute left or right subclavian vertebral angle (figure 5) was associated with increased fluoroscopy time-per-vessel (Pearson’s correlation, p<0.004). The acuity of the take-off angle of the left vertebral artery from the subclavian artery contributes to the degree of difficulty when accessing the proximal left vertebral artery.

Figure 5

(A, B) Acute left subclavian-vertebral angle. The presence of an acute subclavian-vertebral angle, particularly when navigating across the aortic arch from the right subclavian, increases the difficulty of vertebral catheterization.

The acuity of the take-off angle of the vertebral from the subclavian artery is more problematic on the left than on the right side, which is exemplified by cases that require exchanging a catheter that does not need to navigate the arch, such as a Berenstein catheter, to access the right vertebral artery. This configuration requires far less support than would accessing across the arch into the left vertebral artery. Interestingly, the presence of a right or left vertebral artery loop did not have the same effect as the subclavian-vertebral angle. This likely reflects that most of the diagnostic angiography performed at our institution is performed from the proximal vertebral artery, and if a loop is seen, it is purposefully not crossed for the purposes of diagnostic angiography.

Additional factors important to consider in radial angiography that did not reach significance

A type 2 or 3 aortic arch was not significantly associated with a more difficult TRA case. However, in our experience a type 2 or 3 arch can produce an acute curve at the point where the Simmons diagnostic catheter enters the vessel of interest. This configuration can lead to an increased likelihood of the catheter herniating back into the aortic arch when trying to obtain distal access. This anatomical nuance can provide a formidable obstacle when combined with a proximal curve in the subclavian or carotid. While our results do not show this, we expect this to be related to the small numbers represented in this pilot study. In fact, although statistically not significant, the fluoroscopy time-per-vessel was nearly double in a type 3 arch configuration compared with a type 1 arch configuration (10.1 min vs 4.4 min).

Lastly, it is important to consider a right aberrant subclavian artery, or “artery lusoria” (figure 6), in TRA cases. This anatomical variant, while rare, makes a trans-radial case very difficult. Due to the aberrant geometry, the formation of the Simmons shape and obtaining access to the great vessels can be nearly impossible from the right side. If this anatomical variant is noted prior to performing a TRA, alternative access is recommended, as exemplified by our single case where left radial access was attempted.

Figure 6

(A) Right aberrant subclavian artery (or “artery lusoria”). This anatomical finding makes trans-radial four vessel angiography extremely difficult to perform, as indicated by the tortuous path that the catheter must take (B) around the proximal descending aortic arch to come back and catheterize the great vessels.

Based on the results of this pilot study and our anecdotal institutional experiences, we have proposed the below grading scale for the difficulty of performing trans-radial neurointerventional angiography (table 6). This grading scale is based on results obtained from this study for the majority of the variables. Two “additional” variables included are from our clinical experience and from statistical trends within this study, which we consider important from our own anecdotal experience and would anticipate to be significant in a larger series. This scale is currently undergoing a prospective multi-institutional analysis for validation to assess its ability to accurately predict the difficulty of trans-radial cases.

Two additional factors (“type of aortic arch” and “presence of an aberrant right subclavian artery”) were included in the grading scale, which did not significantly correlate with the degree of difficulty in TRA based on this analysis. We have nonetheless included these in the scale as they are of anecdotal importance and—in our opinion—are likely not significant due to the small sample size. As we obtain a larger series, we expect these differences to impact the degree of difficulty in TRA.

Lastly, it is important that the operator consider the target vessels of interest in the choice of access and technique. A single vessel right vertebral artery access from the right radial may prove to be straightforward. However, a left-sided carotid selection with a tortuous subclavian-innominate artery, a difficult aortic arch, and a proximal CCA loop may prove to be a difficult radial case. We have several case examples below that highlight the importance of this point.

Grade I: (0 points)

A patient in the fifth decade of life presented with a hypertensive intracerebral hemorrhage and an incidental left middle cerebral artery aneurysm. Trans-radial angiography was completed to evaluate the aneurysm and search for any other potential sources of hemorrhage. Using ultrasound guidance, the radial artery was cannulated in the standard fashion as described in the methods section of this article. A radial loop was not present (figure 7). A proximal CCA loop was not present. The right vertebral artery was directly catheterized. The operator formed the Simmons shape in the descending aorta without difficulty. The fluoroscopy time-per-vessel was 5.5 min.

Figure 7

(A) No radial loop is present; the left vertebral artery is easily catheterized with a dual wire technique (B), given its straight take-off from the subclavian. (C) Type 1 aortic arch with minimal curvature of the innominate artery.

Grade II: (5 points)

A patient in the seventh decade of life presented for routine angiography to evaluate bilateral incidental middle cerebral artery aneurysms. The patient had a proximal radial loop requiring additional fluoroscopy time to gain access (figure 1). The patient also had a type 2 aortic arch with a proximal subclavian innominate curve and a large diameter aortic arch (figure 8). This forced the operator to form the Simmons shape in the ascending aorta instead of the more favorable formation in the descending aorta. This required more fluoroscopy time, but the operator was still able to access all vessels of interest.

Figure 8

Type 2 aortic arch with single innominate curve and large diameter aortic arch.

Grade III: (10 points)

A patient in the eighth decade of life presented with transient ischemic symptoms and evidence of diffuse atherosclerotic disease. The patient had a medium-size radial artery with a high bifurcation in the arm and a small kink at the antebrachial fossa (highlighted in figure 9A). This factor eventually precipitated vasospasm within the radial artery in the middle of the case, making manipulation of the catheter difficult. The right subclavian artery was tortuous with a significant double curve that also decreased the torquability of the Simmons catheter (figure 9B). The right vertebral artery originated at an acute angle, which made direct vertebral access impossible. The type 3 and large diameter arch necessitated formation of the Simmons shape within the ascending aorta (figure 9C). These factors in combination required substantially more fluoroscopy time for a single vessel angiogram, and the other vessels were not selected due to the difficulty of the case and overall indications.

Figure 9

Radial artery kink and vasospasm (A), double innominate curve (B), and large diameter aortic arch (C).

Strengths and limitations

This is a retrospective review of a prospectively maintained database. The cross-over thresholds for converting to femoral access may differ among different institutions and operators and is a limitation of this study. At our institution, a preoperative CT angiogram is present in a large number of patients; however, there could be an inherent bias in the patients who receive this study and those who do not.

The experience of the operator is also a limitation. At our institution, we have performed trans-radial angiography for several years and learned techniques that help with performing and executing angiography under difficult conditions. The results of this study may not be generalizable to institutions that are new to their radial angiography experience.

Another limitation is that we did not measure the size of the radial artery; however, we included only diagnostic angiography cases that used a small diameter sheath and no cases were aborted or altered due to an initial (without vasospasm) small caliber radial artery in this series. Lastly, we attempted to make the operator rating scale as simple as possible, leaving little to the operator’s interpretation. This is another limitation of the study. The grading scale introduced requires validation on a prospective multi-institutional basis, and this is being undertaken.


Although TRA is associated with lower morbidity, nuanced differences in patient anatomy can alter the degree of difficulty in TRA and leads subsequently to greater radiation dosage, procedural duration, and conversion to alternative access. The presence of a proximal radial loop, large diameter aortic arch, double subclavian innominate curve, proximal left CCA loop, acute subclavian vertebral angle, and absence of bovine aortic arch anatomy were associated with more difficult trans-radial cerebral angiography. A novel grading scale for diagnostic trans-radial angiography is introduced.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.


The authors wish to thank Andrew J Gienapp (Neuroscience Institute, Le Bonheur Children’s Hospital and Department of Neurosurgery, University of Tennessee Health Science Center, Memphis, TN, USA) for technical and copy editing; preparation of the manuscript, tables, and figures for publishing; and publication assistance.



  • Twitter @DornbosIII_MD, @vnnguyen, @PascalJabbourMD, @AdamArthurMD

  • Contributors All authors of this work met ICMJE criteria for authorship and made substantial contributions to the conception and design, acquisition of data, analysis and interpretation of data, drafting, critical revising, and final approval of this manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests DH serves as a consultant for Covidien/Medtronic and Microvention; and is a shareholder of Cerebrotech, Marblehead Medical, and Silver Bullett. LE serves as a consultant for Balt, Cerenovus, Medtronic, MicroVention, Penumbra, and Stryker. CN has received research support from Microvention. PJ is a consultant for Medtronic and MicroVention. ECP is a consultant for Cerenovus, Medtronic Neurovascular, Penumbra, and Stryker Neurovascular; and is a shareholder of RIST Neurovascular. ASA is a consultant for Johnson and Johnson, Medtronic, Microvention, Penumbra, Scientia, Siemens, and Stryker; receives research support from Balt, Cerenovus, Medtronic, Microvention, Penumbra, Siemens, and Stryker; and is a shareholder in Bendit, Cerebrotech, Endostream, Magneto, Marblehead, Neurogami, Serenity, Synchron, Triad Medical, and Vascular Simulations.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.