It is a prospective study conducted in a 20-bedded mixed ICU of a tertiary care center. As arterial line placement is universal in the ICU for hemodynamically unstable patients, a waiver of consent was applied and approved by Institutional Review Board (IRB). After the Clinical trial registry of India (CTRI) registration, patients were randomized to either DP or USG groups by a computer-generated random number from February 2020 to February 2021. Inclusion criteria were adult patients (age ≥ 18 years) admitted to the ICU requiring invasive arterial pressure monitoring. Exclusion criteria include patients admitted with a pre-existing arterial catheter and cannulated with other than a 20-gauge (G) cannula for radial and dorsalis pedis artery [Fig. 1]. Trainee intensivists who had at least 20 successfully USG arterial cannulations performed the procedures. The trainee intensivist selected the site for cannulation upon their discretion.

Figure 1.

Consort diagram for the study. USG Group: ultrasound-guided group, DP Group: direct palpation group.

(0.44MB).

DP method: For radial and dorsalis pedis arterial cannulations, the artery was palpated and punctured at the site of maximal pulsation. Then a 20 G BD Venflon™ Pro IV Cannula (32 mm in length) was advanced over the needle until a flash of blood was seen in the hub of the cannula. For femoral artery cannulation, the artery was palpated and punctured at the site of maximal pulsation, and a 16 G (5 French) ARROW® REF CV-50016 cannula (20 cm in length) was placed by Seldinger technique.

Pre-procedure scan- For all procedures, the artery was assessed by scanning through the transverse and longitudinal axis. Vessel size, distance to the skin, and vessel course were taken into consideration for the selection of the artery [Fig. 2].

Figure 2.

Ultrasound and clinical images with linear transducer (13-6 MHz) position corresponds to each other in upper and lower panel. White arrow showing vessel course and location of artery. A) Radial artery in in-plane view, B) dorsalis pedis artery in in-plane view, C) femoral artery in out-of-plane view.

(0.4MB).

Procedure scan- For radial and dorsalis pedis artery, the artery was identified using USG with a linear transducer in longitudinal axis view after preparing the transducer with a sterile cover. Approximately 0.5 cm distal to the probe, a 20 G BD Venflon™ Pro IV Cannula was introduced in-plane and advanced at 15−30° to the skin until the tip of the needle was in contact with the anterior wall of the artery. The needle was advanced until blood appeared as a flash in the hub. Then the catheter was advanced over the needle. For femoral artery cannulations, the artery was assessed using a linear transducer in transverse plane. Then, the out-of-plane technique was used for catheter insertion. A 16 G ARROW® REF CV-50016 cannula was inserted using the Seldinger technique. (USG machine - Sonosite Edge II, Bothell, WA, Linear probe [13-6 MHz]) [Fig. 3].

Figure 3.

Ultrasound image with linear probe (13-6 MHz) showing radial artery cannulation procedure. A) Pre-procedural Scan- White arrow showing radial artery course B) Procedural Scan- White arrow showing radial artery cannulation with 20 G cannula with needle in situ. C) Post-procedural scan- White arrow showing catheter position after removal of needle.

(0.18MB).

Post-procedure scan- It was performed in all patients to confirm catheter position, distal flow, thrombosis, and hematoma formation.

For the traditional DP palpation technique, time is taken from palpation of a patient's artery to arterial line placement, i.e., when the catheter was successfully placed into the vessel was noted. In the ultrasound group, the linear probe was prepared with a sterile probe cover and gel. Time was calculated from pre-procedural scan until the catheter was successfully placed into the vessel. The total time taken for insertion was calculated in seconds. Cannulation attempts were defined as the number of needle tips completely withdrawn from the skin. If there was a failure to cannulate on a third attempt, the cannulation site was changed.

Additional data collection included demographic profile, attempts for a successful cannulation, complications (hematoma, thrombosis, infection, and vasospasm), number of cannulae used, vitals before and after cannulations, and vasoactive inotropic score (VIS) for patients in septic shock21 [Table1].

The sample size was estimated using G-Power software®. Under the following assumptions: P1 (first attempt success rate with USG technique) = 0.50, P2 (first attempt success rate with DP technique) = 0.30, α (level of significance) = 0.05, 1-β (power of the study) = 0.80, and one-sided independent test for proportion, the sample size is 74 in each group (USG group and direct palpation group). Appropriate statistical measures and tests are used for analysis. A P-value <.05 was considered statistically significant. A statistical package for social science version 23 (SPSS-23 IBM, USA®) was used.

In our study, USG arterial cannulations showed a high first-pass success rate in all three arterial cannulation sites (radial, dorsalis pedis, femoral).

In a meta-analysis by Bhattacharjee S et al., USG radial artery cannulation is associated with a higher first-attempt success rate of cannulation in comparison to digital palpation [OR (95% CI) 2.76 (186, 4.10); P < .001].18,19 Similarly, a meta-analysis by Wan-Jie Gu et al. published in 2016 including 546 patients showed a first pass success rate of 48.5% vs 30.3% for USG and DP group, respectively.24 Levin et al. reported that ultrasound guidance improved the first-pass success rate (USG, 62% vs. DP, 34%) and decreased the number of attempts.25 Our study results resembles the findings of the aforementioned studies.

In an earlier study in operation theatre settings in neurosurgical patients by Anand et al., ultrasound guidance showed a trend towards an improved first-attempt success rate over the direct palpation technique (76.7% vs. 60%, P-value .267), though this was not statistically significant.22 In contrast, our study showed that the first pass success rate was significantly higher with USG compared with DP.

One recent randomized controlled trial reported that ultrasound guidance increased the procedural success rate when compared with the landmark technique for femoral artery cannulation (USG 96% vs landmark technique 78%, P-value .004).26 In our study, USG femoral artery cannulation showed a higher first-pass success rate compared with the DP technique.

A systematic review by Flumignan et al. reported that real-time B-mode ultrasound guidance might improve the time needed for a successful procedure for radial artery catheterization compared to palpation.27 A study comparing ultrasound guidance vs. palpation method for dorsalis pedis artery cannulation showed a statistically similar total procedure time (sum of both assessment and cannulation time, P = .8882).22 In our study, USG arterial cannulation also reduced the time significantly to cannulate in all three arterial cannulation sites (radial, dorsalis pedis, femoral) when compared to the DP palpation technique.

In our study, out of 201 patients, 130 patients were on vasopressor support (69 in the ultrasound group and 61 in the DP group). A VIS higher than 10 represents a predictor of poor outcome, prolonged ICU stay, and an increased number of days on mechanical ventilation.28,29 In our study, out of 130 patients on vasopressor support, 63 patients (USG group = 34 vs. DP group = 28) had VIS more than 10. Invasive BP measurement in this group of critically ill patients may aid in the best titrating vasopressor dose. First-attempt success rate and time for cannulation at all arterial cannulation sites (radial, dorsalis pedis, femoral) were significantly higher in the USG group when compared to the DP group in patients with VIS of more than 10 (Fig. 6). This is an important new finding of our study.

USG arterial cannulation is particularly helpful in patients who are obese.30 In our study, 64 patients had BMI ≥ 25. In our study, the first-attempt success rate was significantly higher in the ultrasound group when compared to the DP group. In addition, time for cannulation at all three sites (radial, dorsalis, femoral) were significantly reduced in the ultrasound group. This is an important new finding of our study.

Compared to the palpation technique that uses anatomical landmarks, USG arterial cannulation reduces the number of attempts resulting in a lower complication rate.31,32 Hematoma formation was an uncommon complication in our study, without clinical relevance. To the best of the authors’ knowledge, this is the first study in ICU settings to compare all arterial site cannulation.

This was a single-center study with small sample size. A large sample size would have allowed researchers to determine the average values of data better and avoid errors from testing a small number of possible atypical samples. The study population was randomized at technique allocation only but not at individual arterial sites. This study looked only for immediate complications and not for long-term complications. Operator expertise should also be taken under consideration, given that the results may not be replicated in less experienced hands.

By virtue of inclusion criteria operator experience, trainee intensivists who had cannulated at least 20 arterial lines under ultrasound guidance reduced bias. Data generation at 3 different arterial sites was not done previously in any studies. Comparison of both techniques in VIS > 10 and BMI ≥ 25 subsets of the population was a unique part of our study.