Arterial catheterization is a commonly performed invasive procedure in the intensive care unit that enables accurate hemodynamic monitoring and frequent blood sampling.1 Different sites can be used for arterial cannulation, such as the radial, posterior tibial, femoral, brachial, axillary, ulnar, dorsalis pedis, and temporal arteries.2 However, the radial artery is the most common site for arterial catheterization because of its superficial course, alternate arterial supply to the hand via the ulnar artery, and a low rate of complications.3 The potential complications for needle placement are thrombosis, edema, vasospasm, hematoma, and posterior wall puncture.
The recent use of ultrasound guidance for radial artery cannulation has been shown to increase the rate of cannula insertion success at the first attempt and has consequently become widely used in radial artery examinations and cannulations.1,4–6 There are 2 basic approaches in needling techniques: short-axis out-of-plane (SA-OOP) and long-axis in-plane (LA-IP) techniques.7–9 In the SA-OOP approach, the transducer is oriented transversely to the vessel, and the vessel appears as a circular anechoic structure. In the LA-IP approach, a longitudinal view of the artery is obtained, and the artery appears as a tubular anechoic structure.10
The study by Berk et al.2 showed that rate of cannula insertion success at the first attempt was 51% and 76% with the use of SA-OOP or LA-IP, respectively. It is important to note that there is a higher chance of complications with increasing numbers of attempts.11,12 In a simulated model of vascular access, Stone et al.10 reported that the LA-IP approach to ultrasound-guided vascular access was associated with improved visibility of the needletip during vessel puncture, which may help to decrease the rate of vascular complications.
The primary objective of this study was to compare the success rate on the first attempt of radial artery cannulation in the modified SA-OOP technique with the LA-IP technique. The secondary objective was to compare the ultrasonic location time, cannulation time, and the complications that occurred during arterial cannulation between the 2 techniques.
Study Design and Subjects Allocation
The local medical ethics committee approved the study protocol, and all patients provided written informed consent before their inclusion. This was an open-label, randomized, parallel study with 2 treatment arms. All patients included in the study were to undergo liver surgery or splenic resection under general anesthesia. They were all graded between American Society of Anesthesiologists I and III and were 22 to 78 years old with a weight range from 54 to 83 kg. We excluded patients with a negative Allen test, ulnar artery occlusion, prevalent atherosclerosis, hemorrhagic shock, morbid obesity, Raynaud disease, peripheral vascular disease, myocardial infarction, unstable angina, cardiogenic shock, coagulation disorders, and multiple previous radial artery interventional therapies. Using a sealed envelope method, patients were randomly assigned (allocation ratio 1:1) into 2 groups to receive either the LA-IP or modified SA-OOP approaches. The procedure in both groups was performed by the same experienced anesthesiologist, who had previously cannulated 450 radial arteries and used the ultrasound-guided technique for approximately 200 procedures.
Patients were monitored using the IntelliVue MP70 patient monitor (Philips, the Netherlands) to assess electrocardiography, heart rate, noninvasive arterial blood pressure (NIBP), and peripheral oxygen saturation. Patients were placed in supine position. The arms were abducted, and the elbows were fixed. The left median cubital vein was used to establish venous access. Patients underwent induction of general anesthesia using midazolam (0.05 mg/kg) and sufentanil (0.1 μg/kg) for analgesia and sedation. The left hand of all patients was chosen for the puncture. The wrist was extended over a roll, and the arm was raised 10 cm. Patients’ hands were positioned in dorsiflexion and fixed to the roll. Sterile preparation was performed over the skin insertion site (2 cm proximal to the wrist joint on the palmar side of the forearm) with povidone iodine, and local anesthesia (0.2 mL, 2% lidocaine) was applied. The ultrasonic probe with disposable sterile covers was connected to an ultrasound system (Terason2000+, Terason, Burlington, MA) that was used to identify the radial artery. After the arterial image was located in the transverse approach, the probe with a frequency of 18 MHz and a depth of 2 cm was adjusted to optimize the image.
Modified SA-OOP Approach
A suture was tied on the midpoint of the ultrasound probe (a developing line) and perpendicular to the long-axis as a guide (Fig. 1A). This line created a visible mark on the ultrasound view and improved accuracy in needle placement. At the beginning of the operation, the developing line was directed at the beating radial artery by adjusting the ultrasonic probe. The arterial cannula needle (BD) was placed on the contact point of the developing line and the skin using the right hand of the operator (Fig. 1B). The ultrasonic probe was moved 3 to 4 mm along the radial artery toward the heart using the operator’s left hand, and the second contact point between the developing line and skin was determined. The needle was inserted steeply downward at a 30° to 45° angle into the skin from the first contact point to underneath the second contact point according to the ultrasonic image (Fig. 1C). Entry into the artery was confirmed by visualizing the backflow of blood into the needle. The angle of the needle was then lowered from 30° to 45° to 15°, and the needle was pushed proximally for another 2 to 3 mm. The needle core was extracted, and the cannula was pushed into the radial artery. The puncture needle was fixed, and NIBP was monitored with a connected pressure transducer.
After localizing the artery in short-axis view, the probe was rotated 90°, keeping the arterial image in the middle of the screen at all times to assist in the identification of the artery in the long-axis view (Fig. 2A). The needle was inserted steeply downward at a 30° to 45° angle into the skin at the midpoint of the short axis of the ultrasonic probe (Fig. 2B). Entry into the artery was confirmed by visualizing the backflow of blood into the needle. The angle of the needle was then lowered from 30° to 45° to 15°, and the needle was pushed proximally for another 2 to 3 mm. The needle core was extracted, and the cannula was pushed into the radial artery after blood was visualized. The puncture needle was fixed, and NIBP was monitored with a connected pressure transducer.
For both techniques, the diameter of the radial artery and the depth from the skin were measured. The ultrasonic location time, cannulation time, and number of attempts in both groups were recorded. Ultrasonic location time was defined as the starting time from the ultrasonic probe contacting the skin until arterial puncture. Cannulation time was defined as the time from initial skin puncture until the catheter was placed into the radial artery. The rate of cannula insertion success on the first and second attempts, and the rate of cannula insertion failure were recorded. Cannula insertion failure was defined as >3 failures. The recorded vascular complications comprised thrombosis, hematoma, edema, and vasospasm.
Sample Size Determination
In this study, the sample size was estimated using Piantadosi method.13 Based on the results reported by Berk et al.,2 we assumed the rate of cannula insertion success on the first attempt would be 51% and 76% with the use of SA-OOP or LA-IP, respectively. Given a difference of 25% (δ = 0.25), with a type-I error 0.05 and power 0.8, the sample size for each arm was calculated as 63. While considering the dropout rate (presumably 20%), the sample size in each arm was finally determined to be 80 subjects for each arm.
Study End Points and Statistical Analysis Methods
Our primary end point was to define the success rate difference of the first cannula insertion attempt between treatment groups (modified SA-OOP versus LA-IP). The choice of this primary end point was from Berk et al.2 using the rate of cannula insertion success on the first attempt in their study.
The secondary end points were to compare the 2 treatment groups for the following:
- Rates of cannula insertion failure.
- The ultrasonic location time and cannulation time.
- The inner diameter of the radial artery and the depth from the skin to artery.
- Vascular complications in the surgery, including thrombosis, hematoma, edema, and vasospasm.
Statistical analyses were performed using SAS software (Version 9.3.1, SAS Institute, Cary, NC). Numeric end point data were tested for normality using the Shapiro-Wilk test. If normally distributed, they were presented as mean ± SD and compared by Student t tests. If not normally distributed, they were presented in the form of mean (SD). For the data not normally distributed and that could not be transformed into normal distribution, a Monte-Carlo simulation approach (bootstrap)14 and Student t test with unequal variance as sensitivity analysis15 were performed. The categorical end point variables were analyzed using the χ2 test, or Fisher exact test if the subject count in any contingency table cell was expected to be <5. Differences with P value <0.05 and 95% CIs excluding 0 were considered statistically significant. All statistical tests are 2-sided.
The subgroup analyses or adjusted analyses were not performed in this study.
One hundred sixty-four patients were randomly divided into 2 groups in this study. Patients’ baseline characteristics are shown in Table 1. One patient in the LA-IP group refused to continue with the radial artery puncture under local anesthesia because of discomfort during the establishment of venous access. This patient was excluded from the study (Fig. 3).
The results for the primary and secondary end points of this study are the followings:
- Cannula insertion success on the first attempt (primary end point) in the modified SA-OOP group was 88.9%, which is statistically significantly higher than the 73.2% in the LA-IP group (proportion difference and 95% CI is 15.7% (0.6%–30.7%), P = 0.0158, Table 2).
- Cannula insertion failures in the modified SA-OOP group and in the LA-IP group were 0% and 2.4%, respectively, that was not statistically significant (95% CI of proportion difference (−17.7 to 12.8), P = 0.50, Table 2).
- The mean values of ultrasonic location time in the modified SA-OOP group and LA-IP group were 6.2 and 15.8 seconds, respectively. This mean decrease of 9.5 seconds shows statistical significance with 95% CI of (−10.6 to −8.5 seconds), P < 0.0001, (Table 3). However, there was no statistical difference in the cannulation time between groups (mean difference with 95% CI, 3.4 seconds (−0.6 to 7.6 seconds), P = 0.1152, Table 3). Sensitivity analyses confirm these results with values very close to the values above (Table 3).
- The results were not statistically significant for the difference of depth from skin to artery when comparing the modified SA-OOP and LA-IP groups (difference of the mean with 95% CI of the difference is 0.3 mm (−0.07 to 0.57 mm), P = 0.1050, Table 4). Also there was no statistically significant difference in the inner diameter of the radial artery between groups (mean difference with 95% CI is −0.1 mm (−0.30 to 0.03 mm), P = 0.1153, Table 4). Sensitivity analyses confirm these results with values very close to the values above (Table 4).
- No vascular complications including thrombosis, edema, and vasospasm were observed in this study. We observed no statistically significant difference in hematoma between treatment groups (proportion difference with 95% CI is −3.5% (−16.1% to 9.2%), P = 0.6743, Table 5).
Arterial cannulation is the “gold standard” procedure for monitoring hemodynamic instability or severe blood loss during surgical procedures, and radial artery pressure is the most commonly monitored variable. Ultrasound guidance has been shown to improve first-pass success rate and decrease the rate of complications during vascular access compared with traditional techniques alone.16 There are 2 basic approaches for ultrasound guidance of needling techniques: SA-OOP and LA-IP, which have a lower first-pass success rate of 51% and 76%, respectively. The possible reasons for this rate are as follows: successful cannulation consists of 2 stages: puncturing the blood vessel and successful insertion of a catheter. First, the diameter of the radial artery is relatively small, and it is therefore more difficult to puncture accurately. Second, catheter insertion is not always completed successfully after the blood vessel has been punctured. One of the important reasons for this difficulty is that the needle is placed into only part of the blood vessel or blood vessel wall, which makes it difficult to insert the catheter and can result in failure. This occurs more frequently during radial artery cannulation, because the average internal diameter of the radial artery is 2.4 mm and the diameter of the needle is 1.1 mm. With the modified SA-OOP technique, we obtained a first-pass success rate of 89%, which is significantly higher than the 73% from the conventional LA-IP technique that was used in our control group. To investigate how these 2 techniques affected the number of cannula insertion attempts, we performed an exploratory analysis and found a significantly different distribution of numbers of attempts in the 2 groups (P = 0.0196, Table 6). In Table 6, we see that compared with LA-IP, the modified SA-OOP tends to have a higher rate of attempts = 1 (which is the primary end point) and lower rate of attempts = 2. We also noticed that if the first and second attempts are combined, then the success rates between the 2 groups are almost the same (P = 1.0000). Therefore, all these analyses echo the primary end point result and indicate that the modified SA-OOP approach leads to more successful first attempts by reducing the number of subjects requiring 2 or more cannula-insertions in the LA-IP approach.
Testing cannula insertion failure rate as a secondary end point is clinically significant, because any new procedure with a high failure rate will not be useful to the clinician. Although the primary goal for the modified SA-OOP procedure is to increase the first attempt success rate, we must also ensure that this procedure does not significantly increase the failure rate. Because we know that the failure rate in the traditional LA-IP method is already very low (as low as 2.4%, as shown in this study), there is little chance of statistically significantly reducing this rate. Any new procedure could potentially increase the failure rate. In this study, we saw the modified SA-OOP avoided this risk, and the failure rate even showed a statistically insignificant trend of decreasing (95% CI, −17.7% to 12.8%, P = 0.4969, Table 2).
When performing an ultrasound-guided approach for cannulation, the vessel is placed in the center of the ultrasound screen, and the needle is held so that it can be inserted into the middle point of the ultrasonic plane. In theory, this can tap into the middle of the vessel. Nevertheless, both the SA-OOP and LA-IP techniques use a visual method to determine the midpoint of the ultrasonic plane, and it can be difficult to determine the central axis of the radial artery because it is so narrow.14 Even with the use of LA-IP, the width of the ultrasound screen was only 12 mm (Fig. 2C). Our modified SA-OOP is a significant improvement, and it is simple for a developing line with a thickness equivalent to a 10.0 surgical suture (0.07-mm inner diameter) to be positioned in the center of the radial artery (Fig. 1C).
Technically, the modified SA-OOP has an advantage over the LA-IP in finding blood vessels more easily and accurately, by ascertaining the location of the artery using the developing line of the probe. The modified SA-OOP only affects the ultrasonic location process, while the cannulation procedure of the 2 methods is essentially the same. The ultrasonic location time and cannulation time are recorded and tested separately. Our study shows that modified SA-OOP required significantly less time to locate the vessel than the LA-IP technique (mean of the time decrease is 9.5 seconds). However, the cannulation time, the distance from the skin to artery, and the diameter of the radial artery were not significantly different between the 2 groups. These findings indicate that the investigator handled these common surgical steps consistently in this study. The increase of first-attempt success rate and decrease of ultrasonic location time are the direct results of technical improvement of the modified SA-OOP versus LA-IP approach.
However, reducing the rate of complications caused by arterial puncturing is also very important. In this study, we did not observe any common complications such as thrombosis, edema, and vasospasm. Although without statistical significance, we observed a slightly lower incidence of hematoma in the modified SA-OOP group, compared with the LA-IP group (15% and 18%, respectively). This could at least partly have been due to the increased rate of cannula insertion success at the first attempt using the modified SA-OOP technique. We did not observe posterior wall damage to the radial artery in either approach, but we were unable to confirm whether the posterior wall of the radial artery has been damaged using the modified SA-OOP technique. Furthermore, we ceased using the ultrasonic probe with the left hand and performed cannula insertion with both hands after visualizing blood in the needle according to our standard technique. A good choice may be to puncture the blood vessel using modified SA-OOP and view insertion using the traditional LA-IP technique.
No severe complications of radial artery cannulation, such as pseudoaneurysm formation, radial artery occlusion, hematoma of the forearm, and osteofascial compartment syndrome of the forearm occurred in any patients during the first postoperative day.
The developing line approach for vessel location is particularly suitable for the localization of small vessels such as the radial artery. This ultrasound method could improve the accuracy of vessel location, improve the success rate on the first attempt for cannulation of small blood vessels, and reduce the risk of complications occurring as a result of repeated puncture attempts.3,17
In this study, we conclude that the modified SA-OOP technique for radial artery cannulation may significantly improve the rate of cannula insertion success on the first attempt, compared with the traditional LA-IP technique.
Name: ZheFeng Quan, MD.
Contribution: This author helped in conduct of the study, data collection, data analysis and manuscript preparation, and approved the final manuscript.
Attestation: ZheFeng Quan attests to the integrity of the original data.
Name: Tian Ming, MD, PhD.
Contribution: This author helped in design and conduct of the study, data collection, data analysis and manuscript preparation, and approved the final manuscript.
Attestation: Tian Ming participated in the collect and conceive of the study, participated in its design and coordination, and helped to draft the manuscript.
Name: Ping Chi, MD, PhD.
Contribution: This author helped in design and conduct of the study, data collection, data analysis and manuscript preparation, and approved the final manuscript.
Attestation: Ping Chi attests to the integrity of the original data.
Name: YingHao Cao, MM.
Contribution: This author helped in design and conduct of the study, data collection, and data analysis.
Attestation: YingHao Cao is responsible for maintaining the study records, the original data, and the analysis reported in this manuscript.
Name: Xin Li, MM.
Contribution: This author helped in data collection and data analysis.
Attestation: Xin Li attests to the integrity of the original data.
Name: KeJun Peng, MM.
Contribution: This author attest to the integrity of the original data and the analysis.
Attestation: KeJun Peng participated in a clinical anesthesia.
This manuscript was handled by: Steven L. Shafer, MD.
1. Shiloh AL, Savel RH, Paulin LM, Eisen LA. Ultrasound-guided catheterization of the radial artery: a systematic review and meta-analysis of randomized controlled trials. Chest. 2011;139:524–9
2. Berk D, Gurkan Y, Kus A, Ulugol H, Solak M, Toker K. Ultrasound-guided radial arterial cannulation: long axis/in-plane versus short axis/out-of-plane approaches? J Clin Monit Comput. 2013;27:319–24
3. Sandhu NS, Patel B. Use of ultrasonography as a rescue technique for failed radial artery cannulation. J Clin Anesth. 2006;18:138–41
4. Ueda K, Puangsuvan S, Hove MA, Bayman EO. Ultrasound visual image-guided vs Doppler auditory-assisted radial artery cannulation in infants and small children by non-expert anaesthesiologists: a randomized prospective study. Br J Anesth. 2013;110:281–6
5. Latham GJ, Veneracion ML, Joffe DC, Bosenberg AT, Flack SH, Low DK. High-frequency micro-ultrasound for vascular access in young children - a feasibility study by the high-frequency ultrasound in kids study (husky) group. Pediatr Anesth. 2013;23:529–63
6. Brzezinski M, Luisetti T, London MJ. Radial artery cannulation: a comprehensive review of recent anatomic and physiologic investigations. Anesth Analg. 2009;109:1763–81
7. Edanaga M, Mimura M, Azumaguchi T, Kimura M, Yamakage M. [Comparison of ultrasound-guided and blindly placed radial artery catheterization]. Masui. 2012;61:221–4
8. Chittoodan S, Breen D, O’Donnell BD, Iohom G. Long versus short axis ultrasound guided approach for internal jugular vein cannulation: a prospective randomised controlled trial. Med Ultrason. 2011;13:21–5
9. Sommerkamp SK, Romaniuk VM, Witting MD, Ford DR, Allison MG, Euerle BD. A comparison of longitudinal and transverse approaches to ultrasound-guided axillary vein cannulation. Am J Emerg Med. 2013;31:478–81
10. Stone MB, Moon C, Sutijono D, Blaivas M. Needle tip visualization during ultrasound-guided vascular access: short-axis vs long-axis approach. Am J Emerg Med. 2010;28:343–7
11. Nayeemuddin M, Pherwani AD, Asquith JR. Imaging and management of complications of central venous catheters. Clin Radiol. 2013;68:529–44
12. Froehlich CD, Rigby MR, Rosenberg ES, Li R, Roerig PL, Easley KA, Stockwell JA. Ultrasound-guided central venous catheter placement decreases complications and decreases placement attempts compared with the landmark technique in patients in a pediatric intensive care unit. Crit Care Med. 2009;37:1090–6
13. Steven Piantadosi Clinical Trials: A Methodologic Perspective. 19971st ed Wiley-Interscience,
14. Robert V. Hogg, Joseph McKean, Allen T Craig Introduction to Mathematical Statistics. 20127th ed Pearson Education, Inc.
15. Zhou XH, Gao S, Hui SL. Methods for comparing the means of two independent log-normal samples. Biometrics. 1997;53:1129–35
16. Miller AH, Roth BA, Mills TJ, Woody JR, Longmoor CE, Foster B. Ultrasound guidance versus the landmark technique for the placement of central venous catheters in the emergency department. Acad Emerg Med. 2002;9:800–5
© 2014 International Anesthesia Research Society
17. Velasco A, Ono C, Nugent K, Tarwater P, Kumar A. Ultrasonic evaluation of the radial artery diameter in a local population from Texas. J Invasive Cardiol. 2012;24:339–41