Incidence of posterior wall penetration during internal jugular vein cannulation: A comparison of two techniques using real-time ultrasound.

BACKGROUND AND AIMS: The true incidence of penetration of the posterior wall (through-and-through puncture) of the internal jugular vein (IJV) during cannulation is unknown. This may have implications if there is hematoma formation, penetration and/or inadvertent cannulation of an underlying carotid artery. This study compared the incidence of posterior wall puncture during IJV cannulation using ultrasound guidance versus traditional landmarks-guided technique.
METHODS: One hundred and seventy adult patients admitted to a gastro-liver Intensive Care Unit who required central venous lines were randomly divided into Group A: IJV cannulation using anatomical landmark-guided technique and Group B: IJV cannulation using real-time ultrasound guidance. In both groups, a second investigator followed the needle path using real-time ultrasound. The incidence of posterior wall puncture, number of attempts for successful cannulation, incidence of inadvertent arterial punctures and occurrence of complications such as hematoma formation and pneumothorax were recorded.
RESULTS: Significantly more (37/80, 46%) patients in Group A had posterior wall puncture compared to 19/90 (21%) in Group B. Incidence of arterial puncture was 8/80 (10%) in Group A, 5/90 (5.5%) in Group B. The number of attempts for venous cannulation and hematoma formation was significantly less in Group B.
CONCLUSION: Real-time ultrasound-guided IJV cannulation significantly reduces but does not wholly eliminate the incidence of posterior venous wall penetrations. It also significantly reduces the incidence of inadvertent arterial punctures and number of attempts for successful cannulation.

RCT Entities:   Population Interventions Outcomes

INTRODUCTION

The standard approach of internal jugular vein (IJV) for central venous cannulation (CVC) using visual and palpable anatomic landmarks is associated with 95% success rate,[12] but it carries the complications from inadvertent puncture of surrounding structures, including the common carotid artery due to significant variation in the degree of overlap of the IJV and carotid artery when patients are placed in a cannulating position.[3] The risk of accidental carotid artery penetration may increase with puncture of the posterior wall of the IJV.

Carotid artery puncture has an incidence ranging from 3% to 10% independent of the technique chosen or operator experience.[4] Although carotid artery puncture is usually a benign event, it can be life-threatening (inadvertent intra-arterial cannulation, stroke, haemothorax, carotid artery-IJV fistula or airway compromise due to a neck haematoma).

The true incidence of puncturing the posterior wall of the IJV during cannulation in patients is unknown and underreported. This complication is commonly noted in patients with dehydration, respiratory distress, collapsible IJV and when the IJV diameter is 4 to 5 mm.[5]

Multiple studies have demonstrated an improved success rate and a decreased complication rate for ultrasound-guided vascular access as compared to the traditional landmark technique.[678]

This study compared the incidence of posterior wall puncture with real-time ultrasound during IJV cannulation using the traditional 'blind' landmark-guided technique versus ultrasound-guided technique.

METHODS

The study was conducted after being passed by due approval from the Institutional Ethics Committee. A written, informed consent was obtained from each patient/relative for central line placements as per hospital policy. This study took place at an adult Gastroenterology and Liver Intensive Care Unit. Both surgical and medical gastroenterology patients in whom central venous lines were indicated as a part of their medical management were included in the study. The exclusion criteria included patient/relatives refusal for central venous line placement, choice of alternate site for central venous cannulation (besides IJV), presence of thrombus within the jugular vein and infection at chosen site of catheter insertion.

The patients were randomised using computer-generated random number table, and the allocation was concealed in an opaque-sealed envelope. The envelope was opened just before the central line (IJV line) placement. In both groups, Multicath 3 (Vygon)™ triple lumen catheters (outer diameter 2.5 mm, 7.5 Fr, 15 cm length) were inserted using the seldinger technique. The Sonosite M-Turbo™ (Fujifilm, Japan) portable ultrasound machine was used. The linear (vascular) high-resolution probe (5–11 MHz) was selected for visualisation of the needle path in an out of plane (transverse) view.

In both groups, central venous cannulations were performed by critical care physicians with >5-year experience in putting internal jugular central venous pressure (CVP) lines using landmark-guided technique and more than 1-year experience in real-time ultrasound-guided line placement.

In Group A, the CVC was placed using the traditional landmark-guided technique. In this, the patient was placed supine with the neck turned to the opposite site, the two heads of the sternocleidomastoid muscle were identified and the apex of the triangle formed by the two heads of the sternocleidomastoid muscles was chosen as the point of entry of needle. The investigator performing the cannulation stood on the head end of the patient while the USG probe was placed transversely (out of plane) just below the point of entry of needle by the second investigator who stood by the side of the patient. After feeling for the pulsations of the carotid artery at this point, the needle and syringe assembly was inserted just lateral to it and directed towards the ipsilateral nipple. On aspirating dark venous blood, the CVC was inserted. During this entire process, the second investigator followed the path of the needle and watched for a posterior wall puncture (through-and-through puncture) or an imminent or actual arterial puncture. The person cannulating (the first investigator) did not have access to the ultrasound images as the screen of the ultrasound machine was turned away from him/her.

In the interest of patient safety, the second investigator warned the cannulating investigator in the event of an imminent arterial puncture. Such event was considered equal to an arterial puncture and noted in data collection sheets.

In Group B, CVC was placed using real-time ultrasound guidance. Here, both the investigators, the one cannulating and the other handling the probe were able to follow in real time, the movement and the position of the needle. In this group, the patient was placed supine with the neck turned to the opposite site; the vascular probe was placed transversely at the apex of the triangle formed by the two heads of the sternocleidomastoid muscle by the second investigator. The vein was identified and centred in the middle of the screen and the needle and syringe assembly was inserted from the midpoint of the probe by the first investigator and the path of the needle tip was followed by both the investigators till successful venous cannulation was confirmed by the visualisation of the needle tip within the lumen of the vein and aspiration of venous blood in the syringe.

In both groups, the intraluminal position of the guidewire within the vein was further confirmed sonographically before dilation and insertion of the catheter.

After the procedure, the second investigator filled out standardised data collection sheets (given below). They recorded the number of times each physician penetrated the posterior wall of the IJV. If the needle was driven in at an angle and initially penetrated the anterior wall and then the side wall, this was also considered a posterior wall penetration.

The following parameters were observed and filled in the data collection sheets: (a) The position of IJV relative to carotid artery as seen just before insertion of the needle. Here, the second investigator scanned the neck and estimated the degree of overlap between the artery and vein. If there is more than 3/4 (75%) overlap, then the position was deemed anterior, if less than 3/4th was anterolateral, if no overlap then it was considered lateral; (b) Number of attempts to locate the vessel (number of times the needle required reinsertion or was withdrawn and required repositioning); (c) Posterior wall puncture (through-and-through puncture of vein; this was the primary outcome measure for the study) (d) Imminent/actual arterial puncture (e) Opening CVP post-line placement to assess volume status (f) Final position of needle before insertion of guide wire (g) Other complications, namely, haematoma formation and pneumothorax.

One hundred and seventy patients were enrolled for the study. This sample size was determined on the basis of a pilot study previously conducted in this same unit comparing the incidence of posterior wall puncture using a land mark-guided technique and a real-time ultrasound-guided technique. The result of a previously performed pilot study on thirty patients (15 per group) (8/15 [53%] in landmark guided technique had posterior wall puncture compared to 3/15 [20%] in the ultrasound guided technique) suggested that two groups of at least 80 patients would be sufficient for an adequately powered study.

RESULTS

One hundred and seventy patients were enrolled in this study. At the time of analysis, Group A had eighty patients, and Group B had ninety patients. Both group were similar with regards to gender distribution (p = 0.821). The age, coagulation parameters and opening CVP were also similar in both groups [Table 1]. Relative position of the IJV with respect to the artery on minimal compression of the neck with ultrasound probe was similar in both groups [Table 2]. The most common position in both groups was anterolateral position of vein over the artery, this was seen in 92% of patients in group A vs 86% in group B. The number of attempts at successful cannulation was significantly lesser in Group B (P = 0.00) [Table 3]. In Group A, 37/80 (46%) had inadvertent posterior wall puncture as compared to 19/90 (21%) in Group B (P = 0.00) [Table 4]. The total incidence of arterial puncture/imminent puncture was 8/80 (10%) in Group A as compared to 5/90 (5.5%) in Group B. This difference was statistically significant [Table 5]. In Group A, there were two instances of arterial puncture and six instances of imminent arterial puncture. In Group B, there was one arterial puncture and four imminent punctures. All the arterial punctures in Group B (imminent plus actual) were associated with a posterior wall puncture. The incidence of other complications (namely haematoma formation and pneumothorax) was significantly lesser in Group B as compared to Group A. 11/80 (14%) patients in Group A had hematoma formation vs 1/90 (1.1%) in group B (p = 0.003). 1 patient had a pneumothorax formation in group A and no patient had pneumothorax in group B. [Table 6]

Table 1

Age, coagulation profile, number of attempts and opening central venous pressure

Table 2

Relative position of internal jugular vein with respect to carotid artery

Table 3

Number of attempts for successful cannulation

Table 4

Incidence of posterior wall puncture of internal jugular vein

Table 5

Incidence of inadvertent arterial puncture

Table 6

Incidence of other complications

DISCUSSION

The true incidence of posterior wall puncture (through-and-through puncture) of the IJV during cannulation is unknown and underreported in literature. It is usually considered to be a benign event, and at times, a universal occurrence in patients who are dehydrated wherein the blood may be aspirated on withdrawing the needle rather than on the first pass. In the classical cannulation position (neck turned to contralateral side), the jugular vein overlaps the carotid artery in 70%–90%[91011] of the time and more so with increasing age[11] and with increasing degree of neck rotation.[1213] In our study, the anterolateral position of the IJV was seen in majority of patients. The true lateral position of vein to artery with no overlap was seen only in 3.3% of cases, similar to another study findings by Maecken et al.[9]

The use of ultrasound for real-time jugular venous cannulation has improved the success rate and reduced the complications associated with blind 'landmark' guided cannulation. Even with the use of ultrasound, complications using ultrasound guidance for needle placement have been described,[514] and lack of visualisation of the needle tip during needle advancement may contribute to inadvertent puncture of vital anatomic structures and failed vascular access attempts. Besides even with ultrasound guidance, the true incidence of posterior wall puncture is not known.

The most common approach to ultrasound guidance is a short-axis or transverse visualisation of the central vein. Short-axis technique appears to be favoured by novice sonologists and is described as being easier to perform and taking less time than the alternative longitudinal guidance approach.[15]

A study found that residents with some prior experience in ultrasound-guided vascular access, regularly had a through-and-through puncture of the IJV in a life-sized torso vascular access mannequin. The residents not only lost track of the needle and regularly penetrated the posterior vessel wall but also cannulated the carotid artery on several instances.[16] IJV cannulation by emergency physicians in a simulated model (by evaluating the path of needles that had been previously placed into vascular access tissue phantoms), the incidence of posterior vein wall penetration was found to be 34%.[17] Present study was performed on humans (previous published studies have been only on mannequins or vascular tissue phantoms).

Results of present study suggests posterior venous wall penetration (through-and-through puncture) of the IJV is a common occurrence during traditional landmark alone guided CVC placement, the use of real-time ultrasound reduces but does not totally eliminate the occurrence of this event. Indeed, the reported incidence in this study may be higher than the previous studies despite the presence of experienced personnel. This may be because, this study was performed on live individuals where the respiratory variation in vessel size and smaller vessel size in hypovolaemia could have played a role.

Accidental penetration of the posterior wall of the vessel may also be attributed to other factors which could include the angle and speed of insertion of needle, the distance from needle entry point and the transducer margin and loss of needle track during insertion. The clinical implications of such through-and-through puncture however still remain unclear. Posterior wall punctures did predispose to arterial punctures in both groups but not all patients who had posterior wall puncture had subsequent arterial puncture. Real-time ultrasound guidance for central line placement can help reduce this. In addition, special precautions can be taken in patients whose veins are collapsed or show significant variation with respiration such as reducing the angle of insertion, reducing the speed of insertion and close watch over the track of the needle tip. With respect to loss of needle track, an in-plane approach may be beneficial as it allows better visualisation of the entire needle path and may help in reducing the incidence of posterior wall puncture.

Further studies on hypovolaemic subset of patients may be required to further clarify the implications and precautions that can be taken. In addition, a larger study group with higher numbers of posterior wall punctures would have to be studied to determine the actual incidence of arterial punctures following a posterior wall puncture.

CONCLUSION

Real-time ultrasound-guided IJV cannulation significantly reduces but does not wholly eliminate the incidence of posterior venous wall (through and through) penetrations. Real-time ultrasound guidance does reduce the number of attempts for successful cannulation and significantly reduces the risk of complications associated with blind cannulations and at the same time allowed prompt diagnoses complications including arterial punctures, haematoma formation and pneumothorax and hence contributes to increased patient safety profile.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.