Background Many neurointerventionalists have transitioned to transradial access (TRA) as the preferred approach for neurointerventions as studies continue to demonstrate fewer access site complications than transfemoral access. However, radial artery spasm (RAS) remains one of the most commonly cited reasons for access site conversions. We discuss the benefits, techniques, and indications for using the long radial sheath in RAS and present our experience after implementing a protocol for routine use.
Methods A retrospective review of all patients undergoing neurointerventions via TRA at our institution from July 2018 to April 2020 was performed. In November 2019, we implemented a long radial sheath protocol to address RAS. Patient demographics, RAS rates, radial artery diameter, and access site conversions were compared before and after the introduction of the protocol.
Results 747 diagnostic cerebral angiograms and neurointerventional procedures in which TRA was attempted as the primary access site were identified; 247 were performed after the introduction of the long radial sheath protocol. No significant differences in age, gender, procedure type, sheath sizes, and radial artery diameter were seen between the two cohorts. Radial anomalies and small radial diameters were more frequently seen in patients with RAS. Patients with clinically significant RAS more often required access site conversion (p<0.0001), and in our multivariable model use of the long sheath was the only covariate protective against radial failure (OR 0.061, 95% CI 0.007 to 0.517; p=0.0103).
Conclusion In our experience, we have found that the use of long radial sheaths significantly reduces the need for access site conversions in patients with RAS during cerebral angiography and neurointerventions.
Data availability statement
Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
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.
Given the overwhelming data supporting transradial access (TRA) as a safer alternative to transfemoral access (TFA) in coronary interventions, neurointerventionalists have begun to transition away from femoral access. Recent studies in the neurointerventional literature have confirmed the safety benefits of TRA over TFA with significant decreases in access site complications.1–5 The key difference between TRA and TFA is the smaller size of the radial artery as compared with the femoral artery. This is advantageous in preventing access site complications, but also presents a new challenge as the smaller radial artery is prone to spasm. Large series have shown the incidence of radial artery spasm (RAS) to be as high as 50%, which can lead to unwanted crossovers to TFA.6–9
The new thin-walled radial artery vascular sheaths are designed with a thinner wall and smaller outer diameter than the conventional femoral sheaths to accommodate the smaller arterial diameter. However, the typical length of the radial artery sheath is approximately 10 cm, which is significantly shorter than the average radial artery length. Vasospasm is only problematic when it is located in the radial artery as the caliber of the brachial artery is significantly larger than a guide catheter. Recently, long radial sheaths have been introduced that are designed to span the entire length of the radial artery, thereby preventing RAS from contacting the guide catheter. In this study, we report our experience incorporating these long radial sheaths into our daily practice and describe the benefits associated with routine use of long radial sheaths as compared with the short radial sheaths.
Following institutional review board approval, a retrospective review of all neuroendovascular procedures at the University of Miami/Jackson Health System from July 2018 to April 2020 was performed. The consent process was waived due to the retrospective nature of the study and patient data were stripped of all identifying information. For inclusion, TRA must have been attempted as the primary access site for the procedure. Demographic, procedural, radiographic, and clinical data were recorded. The results were compared between two groups: procedures completed before and after a long radial sheath protocol was introduced. Radiographic RAS was defined as any arterial spasm seen on the radial artery angiogram. Clinical RAS was defined as any spasm that required further measures to counteract it, such as administration of additional antispasmodic agents or access site conversion.
Long radial sheath placement protocol
The long radial sheath protocol was introduced at our institution in a stepwise protocol as a means to counteract RAS. Specifically, we evaluated all patients with a prior radial angiogram to determine if they had previously experienced RAS. If so, the long sheath was placed directly for their next transradial procedure. For those patients without a prior angiogram, we initially placed a short radial sheath. If RAS was encountered during arterial cannulation or seen on the radial angiogram then we exchanged the short sheath for the long. If RAS was not observed during access then we continued with the procedure. However, if RAS was subsequently encountered during the procedure we would again exchange for the long sheath. Online supplementary figure S1 displays a flow diagram for our protocol.
Standard transradial access techniques
Our technique for TRA has been described previously.1 3 4 10–20 Briefly, the patient is placed supine on the angiography table and a pulse oximeter is placed on the ipsilateral thumb to monitor hand perfusion. A support board is placed inferior to the hand and lateral to the right knee and padding is placed underneath the entire length of the arm terminating just distal to the fingertips so that the wrist remains level with the hip. For standard TRA, the hand is taped in a slightly supinated position to retract the thenar eminence, allowing access to the radial artery just proximal to the radial styloid. In distal TRA (dTRA) the hand is secured in a neutral position to expose the anatomic snuffbox.
Ultrasound is used in all cases without exception after prospective trial data demonstrated that it reduces procedure times.21 22 Once appropriate backflow of blood through the sheath is identified, a radial artery angiogram is performed in all cases. Antispasmodic agents (usually 2.5 mg of verapamil and 200 μg of nitroglycerin) are given through the sheath in an effort to prevent RAS, and 65–70 units/kg of intravenous heparin is administered to protect against post-procedural radial artery occlusion.1 4 11 12 14 16 18 23
Long sheath exchange technique
Access and positioning are as described above. A short 10 cm 5, 6 or 7 French (F) radial sheath (Terumo Glidesheath Slender) is placed and the radial angiogram is reviewed for spasm risk and radial anomalies. Specifically, the length and width of the exposed radial artery between its origin from the brachial artery and the tip of the sheath is assessed. If clinical RAS is experienced during radial artery cannulation or significant radiographic RAS is seen on the angiogram, we proceed with exchanging for a long sheath. As with all radial artery angiograms, it is important to evaluate for radial artery anomalies such as a radial loop, a high-bifurcation radial origin, or radial tortuosity. Although unusual, they present unique technical challenges and may preclude the use of a long sheath. If a radial loop is present, microwire access to the brachial artery must be obtained and the loop straightened prior to placing the long sheath.
Once the decision has been made to exchange for a long sheath, a radial roadmap is obtained through the short sheath and a long 0.025 inch hydrophilic guidewire packaged with the Merit 23 cm Ideal Sheath or Terumo 16 cm Glidesheath is navigated into the brachial artery. The long sheath is then exchanged over the wire for the short sheath with the introducer. The introducer and wire are then removed and back bled. A repeat radial artery angiogram is then performed to confirm placement of the sheath and to ensure that the radial artery no longer contains any segments of spasm. Online supplementary figure S2 displays this. Systemic heparinization is then administered at this point.
We prefer the Merit Prelude Ideal 23 cm radial sheath (Merit Medical, South Jordan, UT) because the longer length consistently spans the entire length of the radial artery. The longest Glidesheath Slender is 16 cm (Terumo, Somerset, NJ) and often leaves a portion of the proximal radial artery uncovered and thus remains at risk of RAS. In particular, the 23 cm sheath is beneficial for covering the entire length of the radial artery for dTRA as the anatomic snuffbox is often 3–5 cm more distal than standard TRA.
Alternative long sheath placement technique
Rather than always exchanging for the long sheath, we place it during the initial access for select cases when RAS is encountered during arterial cannulation or previous radial angiograms have demonstrated RAS. This has the advantage of faster procedure time, fewer steps, and less equipment used. This strategy is particularly useful for patients undergoing a TRA intervention that have had a prior TRA diagnostic angiogram. In these patients, the radial artery anatomy is known prior to puncture because we routinely obtain a radial artery angiogram as part of the diagnostic angiogram. If the radial artery is small or if spasm was encountered during the diagnostic angiogram, we routinely proceed directly with a long radial sheath for the intervention.
In cases where the patient does not have a prior radial angiogram, it is important to understand that there is a small chance that there is an underlying radial anomaly such as a radial loop or high-bifurcation radial origin. Unlike short sheaths, long sheaths often extend past radial loops so it is important to be vigilant for the presence of a radial anomaly until a contrast injection is obtained to confirm correct placement into the brachial artery. There are two strategies for this. The first is to simply insonate the length of the radial artery with the ultrasound prior to puncture. The operator can thus confirm that the radial artery connects to the brachial artery at the level of the elbow and no radial loop or high radial takeoff exists.24 The second strategy is to advance the microwire under fluoroscopy until it is past the elbow. Given the relative infrequency of radial anomalies (13.8%), if the wire goes smoothly one can place the long radial sheath at this point. It is important to recognize that the wire can go smoothly even in the presence of a high-bifurcation radial origin or a radial artery loop. Contrast injection immediately following sheath placement is needed to confirm that the sheath is in the proximal radial or distal brachial artery prior to introducing the guide catheter.
If there is any question or concern with the appearance of the wire on fluoroscopy, the long sheath is advanced partially into the first 10 cm of the radial artery, the introducer is then removed and the sheath is back bled. An angiogram is performed to confirm the radial artery anatomy. In the absence of any radial anomalies, the introducer is then replaced and the sheath is advanced completely over the microwire.
Continuous variables with non-parametric and parametric distributions were represented as median and mean estimates, respectively, with their associated interquartile range or standard deviation. Comparisons of means/distributions of normally continuous variables were carried out using the pooled or Aspin-Welch-Satterthwaite t-test, while non-parametric distributions were compared using the Wilcoxon rank-sum test. Categorical variables were presented as frequency and percent. Statistical analyses of categorical variables were carried out using χ2 and Fisher’s exact t-tests, as appropriate. Univariate logistic regression was utilized to identify significant covariates associated with the likelihood of radial access failure. A multivariable model, adjusted for all significant covariates and potential confounders (p<0.20), was utilized to assess the relationship between the use of a long radial sheath and the likelihood of radial access failure. P values ≤0.05 were considered statistically significant. Statistical analysis was performed with SAS 9.4 (Cary, NC).
Overall patient characteristics
A total of 747 diagnostic cerebral angiograms and neurointerventional procedures in which TRA was attempted as the primary access site were identified; 247 were performed after implementation of the long sheath protocol described above. Table 1 displays the pertinent characteristics of patients undergoing neurointerventional procedures before and after implementation of the protocol. Following the introduction of the long sheath protocol, there was a significant reduction in access site conversions due to clinical RAS (p<0.0001). No significant differences in age, gender, procedure type, sheath sizes, and radial artery diameter were seen between the two cohorts.
Clinical and radiographic RAS
When comparing those patients with RAS to the rest of the cohort, no significant differences in procedure type, sheath size, or periprocedural complications were identified. Although patients with clinical RAS trended towards being younger this did not reach significance (p=0.0761). As expected, radial anomalies and small radial artery diameters were associated with higher rates of radiographic and clinical RAS. Clinical RAS was associated with access site conversion (p<0.0001) whereas radiographic RAS was not (p=0.6161). Online supplementary table S1 displays these results.
Patient characteristics following establishment of the long sheath protocol
When only evaluating those patients after the introduction of our long sheath protocol, we found that that the long sheath was used more often in females (p=0.0003) with smaller radial artery diameters (p<0.0001) and higher rates of clinical (p<0.0001) and radiographic RAS (p=0.0004). The 6F long sheath was also more frequently used than the 6F short sheath (p=0.0027), and higher rates of long sheath usage during interventions were seen but did not reach significance (p=0.0511). Online supplementary table S2 displays these results.
Effect of long sheath protocol on access site conversion in clinical RAS
Evaluating those patients with clinical RAS, implementation of the long sheath protocol significantly decreased access site crossover (p<0.0001). Radial failure among these patients was not associated with the procedure type, sheath size, age or radial diameter. In univariable logistic regression, larger radial diameter and use of the long sheath were the only significant covariates found to be protective against access site conversion. In the multivariable model, use of the long sheath was the only covariate that remained significant (OR 0.061, 95% CI 0.007 to 0.517; p=0.0103). Table 2 displays these results.
The main finding of this study is that implementation of a long sheath protocol to address RAS significantly lowers crossover rates due to clinical RAS in diagnostic cerebral angiography and neurointerventional procedures performed via TRA. Given the impressive results on lowering crossover rates, we now use this protocol for all TRA procedures performed at our institution.
Rationale for use of long radial sheaths
Although the use of the radial artery for endovascular procedures is associated with significantly lower complications as compared with the femoral artery, radial artery spasm can often preclude a transradial approach. Rates of radial artery spasm range between 6.8% and 50% in some large series. A systematic review evaluating 7197 patients undergoing TRA in coronary interventions reported an incidence of RAS of 14.7% and found that it did not differ based on the use of a 5F or 6F sheath.7 Young age, female gender, small radial artery diameter, and unsuccessful first attempts at radial artery cannulation have all been identified as independent risk factors for the development of RAS.7 In addition, anatomical variants such as high-bifurcation radial origins, radial loops and radial tortuosity are also associated with increased risk of spasm.25 Although trapped catheters are rare, even when the catheter is removed the radial artery often remains significantly spasmed, precluding further radial access and necessitating conversion to femoral access.11
Previous studies have demonstrated that hydrophilic coating of sheaths and catheters significantly reduces RAS.8 26–28 However, with the use of a short radial sheath, the proximal portion of the radial artery is exposed and directly contacts a diagnostic or interventional catheter without hydrophilic coating. As the catheters are navigated into position in the great vessels, the radial artery endothelium experiences repetitive friction forces that lead to RAS. Figure 1A,B demonstrate this. While there are certainly radial arteries that are large enough to accommodate most systems, as catheter sizes increase, this scenario becomes increasingly relevant. Thus, even if the distal portion of a guide catheter has hydrophilic coating, it provides no benefit as that portion is quickly advanced into one of the great vessels, leaving the uncoated proximal portion of the catheter in contact with the exposed radial artery.
The use of a long radial sheath obviates all of this by providing a stable hydrophilic conduit from the skin to the brachial artery, at which point the large diameter of the brachial artery renders spasm a non-issue (figure 1C). Thus, the radial artery never comes into contact with the dynamic guide catheter. Many patients have a sufficiently large radial artery (>2 mm) to allow a case to be performed with a short sheath without event. However, in our experience RAS can occur even when the radial artery diameter appears large enough to have a low risk of spasm and has been seen even with 5F systems. As such we have developed a very low threshold for use of these long sheaths in our practice. Specifically, we have found three clinical scenarios where use of a long sheath is particularly helpful.
Clinical scenarios where long sheath is helpful: diagnostic angiograms
As we have moved the majority of our diagnostic practice to dTRA, we have utilized the 5F 23 cm Merit Ideal sheath with increasing frequency. The snuffbox puncture site is 3–5 cm more distal than traditional TRA which means that, once placed, the sheath will terminate more distally leaving more exposed endothelium in the proximal radial artery for spasm to occur. For patients with a large caliber radial artery on ultrasound, we will often still begin with a short radial sheath, but convert to the long sheath if the radial artery is small or spasm is encountered during cannulation. In addition, if we placed a short sheath and the radial angiogram reveals a small radial artery or severe spasm distal to the sheath, we immediately exchange for a long sheath.
Left radial access for neurointerventions is significantly less common than in interventional cardiology. Nonetheless, for many left vertebral interventions the preferred access site is the left radial artery. As we have described, the preferred approach to the left radial artery is via the snuffbox approach with the arm bent and positioned over the patient’s abdomen.10 29 While this allows for a significantly more ergonomic setup for the operator, the distal puncture site again leads to more exposed radial artery at the end of a short sheath. A long radial sheath directly addresses this issue.
Clinical scenarios where long sheath is helpful: radial artery anomalies
Radial anomalies are also an indication for a long sheath in our practice. Radial loops and high-bifurcation radial origins have higher risk of RAS because both anomalies result in increased radial artery surface area exposed to diagnostic catheters without hydrophilic coating.25 Figure 2A,B display a radial loop and a high-bifurcation radial origin, respectively. As such, both are ideal cases for placement of long radial sheaths. If a radial loop is present, wire access to the brachial artery must be obtained and the loop straightened prior to placing the long sheath.
Clinical scenarios where long sheath is helpful: interventions
The rationale and anatomy described in the above section applies for interventions as well. Short sheaths suffice as long as the radial artery distal to the sheath is of sufficient caliber to accommodate an interventional catheter without development of RAS. We now use 6F Prelude Ideal 23 cm sheaths (Merit, South Jordan, UT) in all our 6F interventions if prior radial angiograms demonstrated any RAS in order to reduce the risk of crossover. This has become even more relevant as we have begun to use dTRA for interventions as well.
We have also found an increasingly larger role for the long 7F 23 cm radial sheaths. Prospective data have found rates of radial artery occlusion (RAO) for these 7F sheaths to be very low (<5%) and placement of them allows for a radial arterial line to be provided for anesthesia during our procedure, further reducing patient punctures and procedure times.30 This is particularly helpful during procedures requiring bilateral radial access, such as complex posterior circulation aneurysms or balloon test occlusions, in which anesthesia does not have access to a radial artery for invasive blood pressure monitoring.10
There are currently two thin-walled radial sheaths available: the Glidesheath Slender (Terumo, Somerset, NJ) and the Ideal sheath (Merit, South Jordan, UT), and both exist in 5F, 6F, and 7F sizes. The Terumo slender sheaths (Terumo, Somerset, NJ) are available in 10 cm and 16 cm lengths, and the Merit radial sheaths are available in 7 cm, 11 cm, and 23 cm lengths (Merit, South Jordan, UT). While we prefer the hydrophilic coating of the Terumo Glidesheath Slender, only the 23 cm sheath consistently covers the entire length of the radial artery (figure 1C). Thus, at present our preferred radial sheath is the Glidesheath Slender for short sheath cases, and the 23 cm Merit Ideal sheath for long sheath cases. As always, we rely heavily on the radial artery angiogram to make decisions regarding sheath choice and length.
RAS remains one of the most common complications associated with TRA in neuroendovascular procedures. The use of long 23 cm hydrophlic radial sheaths may prevent RAS as they protect the full length of the radial artery and diminish the repetitive friction forces of the catheter against the radial artery. In this study, we found significantly reduced rates of access site conversions for patients with RAS during cerebral angiography and neurointerventions after introducing a long sheath placement protocol. Our findings warrant further prospective studies.
Data availability statement
Data are available upon reasonable request. All data relevant to the study are included in the article or uploaded as supplementary information.
Roberto Suazo for designing the figures displayed in this manuscript.
Twitter @KatBerryMD, @Starke_neurosurgery
Contributors All authors contributed significantly to this manuscript.
Funding RMS's research is supported by the NREF, Joe Niekro Foundation, Brain Aneurysm Foundation, Bee Foundation, and by National Institute of Health (R01NS111119-01A1) and (UL1TR002736, KL2TR002737) through the Miami Clinical and Translational Science Institute, from the National Center for Advancing Translational Sciences and the National Institute on Minority Health and Health Disparities. He also has an unrestricted research grant from Medtronic and has consulting and teaching agreements with Penumbra, Abbott, Medtronic, InNeuroCo and Cerenovus. The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.