Ultrasound-Guided Percutaneous Central Venous Access in Low Birth Weight Infants: Feasibility in the Smallest of Patients

Introduction

Central venous catheter (CVC) placement is a vital and common procedure performed in children. Particularly for infants requiring critical care, central venous access is critical in order to administer fluids, provide nutrition, and obtain blood for laboratory analysis. However, catheter placement is technically challenging in this population, particularly for neonates with active medical or surgical disease. The traditional method of CVC placement in small neonates in whom previous attempts at peripheral venous access have failed is surgical cutdown over the facial vein or saphenous vein with insertion of the catheter under direct visualization. Prior studies have suggested that percutaneous catheter placement into the internal jugular vein is possible in infants,^1,2 but little is known regarding the feasibility in premature or low birth weight neonates, and the practice patterns of pediatric surgeons and anesthesiologists still vary widely according to preference and familiarity.

The ultrasound-guided (UG) approach to percutaneous CVC venipuncture is well established in adults and is widely recommended as the contemporary technique of choice for internal jugular cannulation.^3,4 Ultrasound guidance allows real-time visualization of the target vessel, allowing the provider to appropriately direct the needle for venipuncture and avoid adjacent structures such as the carotid artery. However, although UG percutaneous catheterization has been recommended by some in children,^5,6 its feasibility has not been fully established in neonates or found to be of overall benefit in the pediatric population generally.^7,8

The purpose of this study is to report our experience with UG CVC placement in underweight neonates and describe modifications to common technique that facilitate uncomplicated catheter placement.

Subjects and Methods

A retrospective review was performed of infants under 3.5 kg who underwent attempted UG CVC placement between August 2012 and November 2013. Data collected included infant weight, vascular access site, diameter of cannulated vein (in mm), and complications.

We use a modified Seldinger technique for UG placement of CVCs in the internal jugular vein. Procedures are typically performed in the operating room or neonatal intensive care unit with the infant intubated, anesthetized, and paralyzed. All infants underwent UG CVC placement of a standard 3.0-French or 4.2-French CVC system (Bard Access Systems, Inc., Salt Lake City, Utah).

The UG approach was performed on all infants with the M-Turbo^® ultrasound system (SonoSite, Inc., Bothell, WA). The infant is positioned supine, and a sterile field is created, including the neck to the angle of the mandible and the chest inferior to the clavicles. A 13.6-MHz linear probe on the ultrasound system is placed in a sterile sleeve and used to identify the carotid artery and internal jugular vein in axial cross-section (Fig. 1A). The vein is typically traced as inferiorly as the clavicle, and the ultrasound transducer will allow maximizing diameter. Under continuous ultrasonic scanning, an introducer needle is passed through-and-through the vein under direct visualization with minimal negative pressure applied to the attached 3-mL slip-tip syringe (Fig. 1B). The needle is then slowly withdrawn, still with minimal negative pressure, until return of venous blood is noted. The syringe is removed, and a guidewire is placed through the needle into the vein (Fig. 2). The prepackaged 0.025-inch-diameter “J” wire included in the Bard sets is used in all infants weighing greater than 2.5 kg. A 0.018-inch-diameter angled glidewire (Radiofocus^® GLIDEWIRE^®; Boston Scientific Inc., Natick, MA) is used in infants less than 2.5 kg.

OPEN IN VIEWER

FIGS. 1–3.

Demonstration of the ultrasound-guided Seldinger technique in an infant. (1A) The right carotid artery (CA) and internal jugular vein (IJV) are located using ultrasound. (1B) Venipuncture is performed with minimal negative pressure applied to the syringe. (2A) The IJV is fully compressed by the needle with artifact (arrowheads) during puncture. (2B) The needle is slowly withdrawn until blood return is noted; then a guidewire is inserted. (3A) After the needle is removed, the guidewire can be faintly seen in the venous lumen. (3B) A tunneled catheter can be placed over the guidewire in the usual fashion.

Once the needle is removed over the wire (Fig. 3), fluoroscopic snapshots are used to confirm proper positioning of the guidewire. The appropriately sized catheter is then tunneled, and an introducer sheath is threaded over the wire. The catheter is inserted through the introducer, with the final positioning again confirmed by fluoroscopy.

Results

Twenty infants underwent 21 UG CVC placements during the study period. The mean patient weight was 2.4 kg (range, 1.4–3.4 kg). The right internal jugular vein was the access site for 16 infants who underwent 17 UG CVC placements, whereas the left internal jugular vein was the target for 3 infants. The average size of the target vessel as measured by ultrasound was 4.0 mm (range, 3.5–5.0 mm). One infant experienced an inadvertent removal of his CVC on postoperative Day 7, returned to the operating room, and underwent a successful UG CVC in the same right internal jugular vein. There were no other placement- or line-related complications in the group.

Discussion

In this study we establish the feasibility of ultrasound guidance for CVC placement in low birth weight neonates and describe modifications to the common technique that facilitate uncomplicated catheter placement. This is germane to contemporary pediatric surgical practice, as surgical requests for central venous access in the neonatal intensive care unit population are frequent and challenging. Peripherally inserted central catheters are often a first-line choice for these small infants, but many critically ill neonates of this size do not have peripheral or scalp veins amenable for venipuncture. It is widely recognized that CVC placement in infants presents a unique set of technical challenges. ⁹ However, trends toward increasing use of ultrasound guidance in children have been largely based on studies outside of this population. Here, we report the results of a series of 21 UG CVC placements in infants under 3.5 kg with a mean weight of 2.4 kg.

Percutaneous CVC placement in infants avoids the need for surgical cutdown, but real-time visualization of the vascular structures is crucial. Although the UG approach has been described in children, the benefits and limitations of its use are not well characterized. In a randomized trial of anatomic landmark versus the UG technique for tunneled CVC placement in children, ultrasound guidance was found to decrease the number of attempts at venous cannulation but had no effect on overall complications. ² However, the average patient weight in their cohort was over 30 kg, making the conclusions not likely generalizable to much smaller children. The dimensions of the infant neck and diminutive size of the vessels are generally thought to preclude a landmark-based approach due to a high risk of complications; thus, ultrasound has been a critical enabling technology to permit the percutaneous technique. The two prior series that have described its use in neonates and infants^1,10 both reported preliminary feasibility in cohorts with a median weight of 2.9 kg. A case report of successful UG CVC placement in an infant under 1 kg has been reported, ¹¹ but the technique has not been explored in a systematic fashion in premature and very low birth weight neonates.

Our modifications to the Seldinger technique in this cohort are important to highlight. We initiate the venipuncture by traversing all the way through the vein prior to applying negative pressure to the syringe. In our experience, drawing back on the syringe during needle entry will cause the vein to collapse as the needle punctures the front wall, leading to the vessel pulling away from the bevel and losing access. By passing all the way through and only applying gentle negative pressure as the needle is withdrawn, aspiration of blood can be noted without disruption of the relationship between the needle and the vessel, allowing a guidewire to be passed. This is only permitted by the use of ultrasound and the direct imaging visualization of the finder needle passing through the vein.

In conclusion, UG CVC placement is a safe and efficient approach to central venous access in infants as small as 1.4 kg. Our experience supports the use of a UG percutaneous technique as the initial approach in infants who require central venous access.