Comparison of Ultrasound-Guided Direct versus Ultrasound-Guided Dart Technique of Radial Artery Cannulation: A Randomized Control Study.

BACKGROUND: Three different types of cannulation method for radial artery are Direct technique, Seldinger technique, and modified Seldinger technique (Dart). Their comparative efficacy has been studied using palpatory method but not with ultrasound guidance. AIMS: We compared the efficacy of ultrasound-guided Direct and ultrasound-guided Dart technique of arterial cannulation. SETTINGS AND
DESIGN: One hundred and sixty patients posted for elective surgeries were included in prospective randomized control, single-blind study in a tertiary care center.
MATERIALS AND METHODS: The study comprised of two groups: Direct method (n = 80) and Dart method (n = 80), which were compared for the rate of successful cannulation within 5 min. The secondary objectives were time for successful cannulation, number of attempts, and rate of complications (hematoma, posterior wall puncture, and needle reinsertion) between two groups. STATISTICAL ANALYSIS: The group comparison for continuously distributed data was compared using the independent sample t-test. The Chi-square test was used for the group comparison of categorical data. Binary logistic regression was conducted to ascertain significant predictors for successful cannulation in 5 min.
RESULTS: Cannulation success rate was similar in both Direct (57.5%) and Dart (55%) groups. There was no significant difference in time for successful cannulation, number of attempts, number of needle redirection, and posterior wall puncture. However, the incidence of hematoma (Direct 22.5% [18]; Dart 8% [10]) was significant. There was better success rate of cannulation (n = 90) in patients with normal pulse and bigger radial artery lumen.
CONCLUSION: There was no significant difference between Dart and Direct technique with the use of ultrasound guidance. Copyright:

INTRODUCTION

The radial artery is the most common site used for cannulation because of its easy accessibility, dual blood supply, and low risk of complications.[1] Three types of arterial cannulation techniques have been described.[2] They are Direct, Classical Seldinger, and Modified Seldinger (Dart). Dart cannula (Integrated guidewire) is based on modified seldinger technique where guidewire is integrated in the cannula as tail. The success rate of cannulation by palpation has been found to be better in Dart technique than in Direct Technique. It has been due to its design as guidewire is already integrated in lumen and cannula is railed over it.[2] Ultrasound-guided radial arterial cannulation has been found to be associated with increased first pass success rate (76.7%) in comparison to palpation method (60%).[3] To our knowledge, the comparison of two different techniques of arterial cannulation (Direct and Dart) has never been conducted by ultrasound guidance. In this study, it was aimed to compare the rate of successful cannulation within 5 min. We hypothesise that there is no difference in the success rate between the two methods.

MATERIALS AND METHODS

This was a single center, prospective randomized controlled, single-blind and parallel study done at department of anesthesiology of a tertiary care hospital after approval from institutional review board (L No.-212/IEC/PGM/2018) on 29/12/2018. It was registered in Clinical Trial Registry of India (Reference number CTRI/2019/10/021612) on 14/10/2019. Patients were enrolled from October 2019 to June 2020. American Society of Anesthesiologist physical status (ASA) I-III patients between 18 and 90 years of age undergoing elective surgeries who required invasive arterial pressure monitoring were enrolled. Patients with arteriosclerosis, hemorrhagic shock, morbid obesity, Raynaud's disease, peripheral vascular disease, and negative modified Allen's test were excluded. Written and informed consent was obtained from them. The patients were randomly assigned into 1:1 ratio for arterial cannulation through Direct or Dart technique. Patients were randomized into two parallel groups with allocation ratio (1:1) by computer-generated randomization tables and sealed envelope technique by an investigator involved in performing the procedure. Patients were blinded to group allocation.

The patients were instructed to maintain fasting for 6 h before surgery and were prescribed alprazolam per oral the night before surgery. They were also administered alprazolam 0.25 mg and ranitidine 150 mg on the morning of surgery. The arterial pressure transducer with heparinized saline in pressure bag was assembled near operating table. In the operation room, two intravenous (i.v.) line was established, one for giving i.v. fluids in case of hypotension and another for vasopressor infusion by infusion pump with a piggy back saline through burette set. Standard ASA monitors were attached to the patients. Heart rate, noninvasive blood pressure, and continuous electrocardiogram were monitored. Femoral arterial line was secured in one of the lower limb after screening for any significant plaque or thrombus.

Patients were administered midazolam 1 mg and fentanyl 2 μg.kg− 1 i.v. After 3 min of preoxygenation, induction was carried with 2 mg.kg− 1 propofol, and endotracheal intubation was facilitated using vecuronium bromide 0.1 mg.kg− 1 body weight. Capnography was instituted after intubation.

After induction, under strict asepsis, wrist was supinated and extended to expose radial artery. Radial pulse was palpated, and character of pulse was noted as normal or feeble. Anesthesiologists performing the procedure had experience of 3 years in anesthesia. They were aware of palpatory method of radial artery cannulation using simple 20 G venous cannula. They had experience of ultrasonography (USG)-guided central venous cannulation but had no experience with USG-guided arterial cannulation. They were novice in USG-guided arterial line cannulation by both technique and were trained for each using simulator (Blue Phantom Arterial line vascular Access Ultrasound Trainer, CAE Inc. Canada) with ultrasound, performing 10 procedures each. Under sterile precautions, USG probe was held with nondominant hand and off axis plane was used to measure the lumen diameter. The probe was rotated by 90° to in axis plane to optimize the image of radial artery with maximal diameter and color Doppler was used to confirm the arterial flow. The cannula was held with dominant hand to conduct the procedure. Imaging plane, needle path, and guidewire or catheter was in same plane. The anesthesiologist who secured the arterial line was aware of the procedure used; however, the observers who recorded all events were blinded to the study protocol. The cannulation was performed using ultrasound machine (GE Health Care co, Tokyo, Japan) with a linear 6–12 MHz transducer.

Direct group

Radial artery cannulation was performed with 20 G arterial cannula (BD Medical systems, Franklin Lakes, NJ, USA) in-plain technique using linear ultrasound probe. As the needle tip was visualized inside the lumen of artery, it was further advanced 1–2 mm and catheter was railroaded over stylet with real-time imaging using only dominant hand [Figure 1a and c].

Figure 1

(a) Direct cannula needle with catheter; (b) Dart Cannula needle with integrated guidewire; (c) Direct Catheter being pushed over Cannula stylet in ultrasound guidance; (d) guidewire passing into the radial artery (hollow arrow) under ultrasound guidance

Dart group

An angiocath of 40mm 20G, with preloaded guidewire based on the principle of modified Seldinger technique (RA04020; Arrow International Inc., Reading, PA, USA/Teleflex Medical, Athlone, Ireland) was used to access the radial artery [Figure 1b and d]. The angiocath was advanced inplane caudal cranial 1 cm away from the probe. As the needle tip was visualized inside the lumen of the artery, it was NOT advanced further. A second operator advanced the integrated guidewire holding actuating lever (black tab). The catheter was advanced into the artery over the wire. Assembly of needle and guidewire was withdrawn leaving the angiocath inside.

Cannula was connected to pressure transducer by pressure monitoring line. Cannulation was confirmed by pressure waveform on the monitor. It was considered successful only if achieved within 5 min of holding the probe on to the skin surface.

Transesophageal echocardiography probes and central venous cannulation in internal jugular vein were secured in all patients. Midline sternotomy was done in all cases. After securing aortic and venous cannula cardiopulmonary bypass was initiated by clamping the arterial line and giving cardioplegia simultaneously. After the completion of surgery, de-airing the cardiac chambers, return of spontaneous cardiac contraction, the circulatory support from heart lung machine was tapered off. After coming off bypass, surgical closure of sternum was performed. Thereafter, the patient was shifted to the intensive care unit (ICU) for further management.

Our primary objective was to find the rate of successful cannulation within 5 min in both groups. The secondary objective was total number of attempts, time for successful cannulation, and incidences of complications such as hematoma, needle redirection, and posterior wall puncture. Cannulation time was defined as time from USG probe held on the wrist until arterial waveform appears on the screen. One attempt was defined, when the new skin puncture was made. When the path of needle was deviated sideways or backward to achieve cannulation without coming out of the skin, it was called needle redirection. The procedure was considered a failure when it was not achieved within 5 min, or new site was required due to hematoma or cannulation failure.

Sample size was calculated based on previous study[4] and to attain power of 90% and 0.05 two-sided Type I error rate, 73 patients per group were required. Considering 5% drop out, 80 patients were enrolled in each group.

Statistical analysis

Statistical analyses were performed using MedCalc for Windows, version 15.0 (Med Calc Software, Ostend, Belgium). The group comparison for continuously distributed data was compared using the independent sample t-test. The Chi-square test was used for the group comparison of the categorical data. Binary logistic regression was conducted to ascertain significant predictors for successful cannulation in 5 min. First, univariable regression was performed to find out the individual odds ratios and significance for each of the variables, as pulse, number of attempts, number of needle redirections, posterior wall puncture and radial artery lumen size, then multivariable logistic regression was performed to find the independent predictors of successful cannulation.

RESULTS

One hundred and sixty patients were enrolled and divided into two groups Dart and Direct. Each group had 80 participants, and there was no exclusion. All the patients were followed up and analyzed for the study. The data were collected from October 2019 to June 2020 [Figure 2]. Demographic variables such as age, gender, weight, height, body mass index, ASA physical status, type of surgery, heart rate, blood pressure (systolic blood pressure, diastolic blood pressure, and mean arterial pressure), radial artery lumen diameter, and pulse volume were comparable in both the groups [Table 1].

Figure 2

Consort flow-diagram

Table 1

Demographic variables

Parameters	Method		P
	Direct (n=80)	Dart (n=80)
Age (years)	41.9±12.1	42.0±15.6	0.746a
Gender (male/female)	44/36	49/31	0.423b
BMI (kg/m2)	23.0±3.6	23.2±6.9	0.428a
ASA grade l/ll/lll	30/17/33	31/28/21	0.068b
Surgery (cardiac/noncardiac)	24/56	29/51	0.401b
Heart rate (b/min)	85.1±15.0	87.1±16.0	0.360a
Systolic BP (mmHg)	127.6±19.6	124.8±26.7	0.319a
Diastolic BP (mmHg)	77.8±14.9	74.7±13.4	0.061a
MAP (mmHg)	94.4±15.0	91.4±16.0	0.083a
Radial artery lumen (mm)	2.19±0.47	2.11±0.48	0.310c
Pulse (normal/feeble)	71/9	68/12	0.482b

aWilcoxon test, bChi-squared test, ct-test. Direct=Direct technique, Dart=Modified seldinger technique with integrated guide wire, BMI=Body mass index, ASA=American society of anaesthesiologist, BP=Blood pressure, MAP=Mean arterial pressure

Success rate of cannulation, the primary outcome of study, with Direct method was 57.5% (46) and Dart method was 55% (44) which was not statistically significant (P = 0.750). Overall success rate of cannulation in the entire population was 56.2% (90/160). The time taken by novice anesthesiologists in Direct group was 208.4 ± 126.0 s, and in Dart group 232.1 ± 143.5 s (P = 0.42); which was not statistically significant between the groups. There was no significant difference between groups in terms of number of attempts, number of needle redirection, and posterior wall puncture. However, the difference of hematoma percentage was statistically significant (P = 0.032) [Table 2]. There was strong association of successful cannulation (n = 90) with normal pulse (P = 0.023) in the total population. Successful cannulation was associated with larger radial artery lumen size (2.25 ± 0.46 mm) in comparison to failures (2.02 ± 0.47 mm). In the successful cannulation, there was less number of attempts, needle redirection, and posterior wall puncture and no hematoma formation [Table 3].

Table 2

Primary and secondary objectives

Parameters	Method
	Direct (n=80)	Dart (n=80)	P
Successful cannulation (%)	46 (57.5)	44 (55.0)	0.750b
Time to cannulation (s)	208.4±126.0	232.1±143.5	0.426a
Number of attempts 1/2/≥3	57/17/6	45/28/7	0.124b
Number of needle redirections≤1/2/3/≥4	22/25/21/12	23/16/19/22	0.169b
Posterior wall puncture (present)	20 (25.0)	31 (38.8)	0.062b
Hematoma (present)***	18 (22.5)	8 (10.0)	0.032b

***Significant at P<0.05, aWilcoxon test, bChi-squared test. Values are presented as n (%), Direct=Direct technique, Dart=Modified seldinger technique with integrated guide wire

Table 3

Association between successful cannulation and parameters

Parameters	Successful cannulation within 5 (min)		P
	Present (n=90)	Absent (n=70)
Type of surgery
Cardiac	31 (34.4)	22 (31.4)	0.688a
Noncardiac	59 (65.6)	48 (68.6)
Pulse***
Normal	83 (92.2)	56 (80.0)	0.023a
Feeble	7 (7.8)	14 (20.0)
Number of attempts 1/2/≥3***	81/9/0	21/36/13	<0.001a
Number of needle redirections≤1/2/3/≥4***	45/24/17/4	0/17/23/30	<0.001a
Posterior wall puncture***	9 (10.0)	42 (60.0)	<0.001a
Hematoma (present)***	0	26 (37.1)	<0.001a
Radial artery lumen (mm)***	2.25±0.46	2.02±0.47	0.002b

***Significant at P<0.05, aChi square test, bt-test. Values are presented as n (%)

In the Dart group univariate analysis, number of attempts >1 had an odds ratio of 0.04 (0.01–0.12, P < 0.001), posterior wall puncture was 0.09 (0.03–0.26, P < 0.001), and for radial artery lumen was 1.21 (1.09–1.39, P = 0.002). In the multivariable model, the most significant factor was number of attempts, with an odds ratio of 0.02 (0.00–0.14, P < 0.001).

In the direct group univariate analysis, odds ratio for number of attempts >1 was 0.05 (0.01–0.17, P < 0.001), and the odds for posterior wall puncture was 0.04 (0.01–0.16, P < 0.001). In the multivariable model, the most significant factor was found to be number of attempts, with an odds ratio of 0.19 (0.04–0.81, P = 0.031).

DISCUSSION

USG guidance in vascular access is getting more popular in recent years. Various studies have shown that USG guidance for vascular access has decreased rate of failure and incidence of complications.[345] Studies have shown that using classical Seldinger or modified Seldinger technique (Dart) have increased success rate of arterial cannulation as compared to Direct technique.[2] Avoidance of impingement of needle tip on posterior wall has been cited for better results with these techniques.[6] However, other studies have shown modified Seldinger technique having no significant advantage in success rate as compared to Direct or Seldinger technique using the palpatory method.[78] USG artery cannulation has been found to be associated with increased first pass success rate in both pediatric and adult patients.[910] Needle tip, guidewire, and plastic catheter can be seen in real-time during USG-guided cannulation.[11] In our study, USG linear probe was used to image radial artery in the long-axis “In Plane” view which helped in better visualization of lumen and catheter thus avoiding any confounding factor as posterior wall impingement, as seen with palpation techniques.[12]

Novice anesthesiologists have mean time of cannulation 2.36 min (95% confidence interval [CI] =1.15–3.58) and 5.02 min (95% CI = 2.90–7.13) in out of plane and in plane, respectively.[13] Experienced anesthesiologists have shorter time of cannulation ranging from 23 s to 107 s in various studies.[1415] In our study, comprising of novice anesthesiologists had mean time of cannulation at par with other studies.

The rate of successful cannulation in Direct method (57.6%) and Dart method (55%) was not significantly (P = 0.750) different. A similar study, compared Direct and Dart technique although using palpatory method, found no significant difference in overall success rate of wire guided cannula (WGC) (71%) compared to (non-WGC) (61%).[8] In contrast another study using palpatory method for cannulation, success rate with the Seldinger technique was 82% compared with that of the Direct method of 65% (P = 0.02).[7] This difference by palpatory method was attributed to the design of cannula with guidewire which makes it easier to railroad the catheter as compared to direct technique which can end up impinging on the posterior wall of the lumen. However, this disadvantage to Direct cannula was not there when USG was used to visualize the needle tip in real time.

Beards et al. in their study of palpatory method of cannulation in ICU patients showed that use of the Direct-puncture technique was associated with a significantly higher failure rate (23%) than use of Seldinger (P < 0.001) or Modified Seldinger (P < 0.02).[2] Failure of catheter insertion with Direct technique commonly resulted from inability to advance the catheter successfully once arterial puncture was achieved, and in seldinger or modified seldinger from inability to rail road the catheter after insertion of the guidewire, thus making the procedure technique as reason for failure rates. Whereas, in our study, using USG to visualize the sheath or guidewire in real time avoided confounding factors of both the techniques.[12]

There was significant association of successful cannulation (n = 90) with normal pulse in the entire population. It did not endorse with Gerber et al., who had found significant association of successful cannulation with WGC and weak/absent pulse (P < 0.01).[8] This may be due to ease of passing cannula over guidewire in small volume or collapsed pulse, where Direct cannulation may be difficult to thread in collapsed lumen. In our study, ultrasound was used during cannulation and normal pulse had given better window and ease of cannulation.

Like Ueda et al., our cannulation success rate was more in larger lumen radial artery (2.25 ± 0.46 mm) and was significant co-variate of successful first attempt success rate.[16] The Dart group has significant association with lumen size, but not the Direct group. It may occur because of relative difference in diameter of guidewire and lumen of artery in Dart group facilitating the passage of catheter. Whereas this difference is significantly reduced in Direct group where sheath size is comparatively larger in diameter than the guidewire.

In our study, there was significant complication in the form of local hematoma (Direct 22.5%; Dart 10.0%). Hematoma was reported more with increasing number of attempts, a greater number of needle redirection, and smaller lumen of radial artery <1.87 ± 0.45 mm in total population. Only Direct group had significant relation between hematoma and bounding pulse, posterior wall puncture, and total time of cannulation, but this was not in the case with Dart group. This signifies that the use of ultrasound in real time and catheters with guidewire has less risk of trauma as compared to catheters without guidewire.

In Direct group, on univariate analysis, more than one attempt and posterior wall puncture had odds ratios <1 being 0.05 and 0.04, respectively, whereas, on multivariate analysis, the most significant factor was number of attempts, with an odds ratio of 0.19. In Dart group, on univariate analysis, more than one attempt and posterior wall puncture had odd ratio of 0.04 and 0.09, respectively, whereas, on multivariate model, the most significant factor was number of attempts with an odds ratio of 0.02. This emphasize that a greater number of attempts are going to reduce the chances of successful cannulation so our first attempt should be the best attempt.

CONCLUSION

There was no difference found in the success rate of cannulation between Dart and Direct group with the help of USG, thus accepting hypothesis. The total number of successful cannulations was 56.2% in total population which is acceptable as most of the anesthesiologists were not so well versed in ultrasound-guided radial artery cannulation.

This study has few limitations as this was a single-center study with small sample size, probably a multicenter study with greater sample size can establish the superiority of Dart method over Direct method with help of USG. The learning curve for novice anesthesiologists is steep due to difficulty in achieving hand eye coordination for USG guidance, although this was similar for both the groups. This has led to decreased overall success of cannulation. The time for cannulation, number of attempt, and needle repositioning were more in our study, which might have caused more incidence of arterial spasm and hematoma resulting in failure of cannulation. Many anesthesiologists complained about difficulty in stabilizing the Dart cannula due to its large tail. Although Dart technique can be done by a single operator, real-time guidewire placement required a second person to move the actuating lever as both the limbs of primary operator were engaged one holding cannula and the other holding USG probe. Experienced examiners could have done the procedure with better success rate and could have helped in establishing difference between the two groups which was the intention of this study.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.