The Optimal Depth of Central Venous Catheter for Infants Less Than 5 kg

Abstract

To avoid fatal complications of central venous catheterization such as cardiac tamponade, the tip of the central venous catheter (CVC) should be placed outside of the cardiac chamber. To suggest a guideline for a proper depth of CVC in infants, we measured the distance from the skin puncture site to the junction between superior vena cava and right atrium (SVC-RA junction) by using transesophageal echocardiography (TEE). Fifty infants less than 5 kg undergoing surgery for congenital heart disease were enrolled in this prospective study. After the induction of general anesthesia, CVC was inserted via the right subclavian vein. After the tip of the CVC was placed at the SVC-RA junction using TEE guidance, the length of the CVC inserted beneath the skin was measured. The measured distance had a high correlation with the patient’s height, weight, and age (r = 0.88, 0.76, and 0.64, respectively). In infants smaller than 5 kg, the following guideline can avoid intraatrial placement of the CVC: a depth between 40 and 45 mm for infants 2.0–3.0 kg in weight, 45–50 mm for those 3.0–3.9 kg, and 50–55 mm for those more than 4.0 kg.

In Brief

IMPLICATIONS: To create a guideline to avoid intracardiac placement of central venous catheters in infants, we examined the distance from skin to superior vena cava and right atrial junction by using transesophageal echocardiography. The recommended depth of the central venous catheter inserted through right subclavian vein is between 40–55 mm for infants less than 5 kg.

Placement of a central venous catheter (CVC) is frequently needed during pediatric anesthesia and intensive care. Its malposition not only causes faulty central venous pressure measurement but also causes fatal complications such as thrombosis, arrhythmia, cardiac perforation, and cardiac tamponade (^1–6). To avoid these fatal complications, the CVC is recommended to be placed in the distal superior vena cava (SVC), outside of the cardiac chamber (^4–6).

Although malposition-related complications are more common and more serious in infants (^7–9), there are no evidence-based guidelines for the proper depth of the CVC for small infants. The subclavian vein, along with the internal jugular vein, is one of the most commonly used routes of the CVC for pediatric patients (^9,10). Thus, to suggest a guideline for the proper length of a CVC inserted through the subclavian vein in infants, we measured the distance from the skin puncture site to the SVC-right atrium (RA) junction by using transesophageal echocardiography (TEE).

Methods

After obtaining hospital IRB approval and informed parental consent, patients less than 5 kg in weight who were scheduled for elective repair of ventricular septal defect or atrial septal defect were enrolled in this prospective study. Patients with left-sided SVC or any extra-cardiac vascular abnormality were excluded. After the induction of general anesthesia, central venous catheterization was performed using the Seldinger technique. All the catheterizations were performed by one of four anesthesiologist staff members with more than 1 year of experience in pediatric anesthesia. Patients were positioned in slight head down position with a rolled towel placed transversely under the shoulder. An infraclavicular approach as described in our previous report (¹¹), was performed on the right subclavian vein (RSCV). Each anesthesiologist was allowed to attempt to advance the CVC only twice. If two anesthesiologists failed to advance the CVC into the SCV-RA junction, another vein was chosen for central catheterization and the patient was excluded from the study. While one of the anesthesiologists performed the CVC cannulation, another anesthesiologist observed the SVC-RA junction using TEE. After the tip of the CVC was located at the SCV-RA junction, defined as the superior border of crista terminalis in bicaval view, the length of the CVC beneath the skin was measured using the depth indicator on the CVC and an aseptic paper ruler. When the tip of the CVC reached the SCV-RA junction, the length of the CVC beneath the skin was considered to be equal to the distance from the skin puncture site to the SVC-RA junction.

All data are expressed as mean ± sd or median (range). Plots of distance from the skin to the SVC-RA junction versus age, weight, or height were made and simple linear regression analysis was performed by the least-squares method. Multiple linear regression analysis was performed in standard and forward selection to identify independent factors affecting the distance from the skin to the SVC-RA junction. Variables with the value of P < 0.1 from correlation analysis were entered into regression analysis. Statistical significance was taken as P < 0.05. All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS Windows version 11.0; SPSS Inc., Chicago, IL).

Results

Sixty infants were initially enrolled. Ten patients were excluded during the study because of failure in subclavian vein puncture (n = 6), and advancing the tip of the CVC to the SVC-RA junction (n = 4). As a result, only 50 patients’ data were analyzed. Patient demographic characteristics of the patients are in Table 1.

Table 1: Demographic Patient Characteristics

The distance from the skin puncture site to the SVC-RA junction (54.1 ± 4.7; 41–62 mm) showed a high correlation with the patient’s height, weight, and age (r = 0.88, 0.76, and 0.64, respectively). Fig. 1 shows the scatter plot of the distance from the skin puncture site to the SVC-RA junction versus age, weight, or height. In standard forward multiple regression analysis, the distance from the skin puncture site to the SVC-RA junction could be predicted from height and weight as the formula: distance from skin to SVC-RA junction (mm) = 11.6 + (0.70 × height in cm) + (1.14 × weight in kg) (r² = 0.81; P < 0.0001).

Figure 1.: The plots of distance versus age (a), weight (b), and height (c). a) Age (month): distance = 49.2 + (2.3 × age); r ² = 0.40, P < 0.1. b) Weight (kg): distance = 35.8 + (4.7 × wt); r ² = 0.58, P < 0.001. c) Height (cm): distance = 8.5 + (0.8 × ht); r ² = 0.77, P < 0.0001. Distance = the distance from skin to superior vena cava and right atrial junction.

Based on these results, we made simple and practical guidelines (Table 2). When our guidelines were applied to the present data, in 49 patients (98%), the CVC was placed in the SVC above the RA. In the one other patient, the tip of CVC was placed 1 mm inside the RA.

Table 2: Suggested Depth of CVC Through RSCV

Discussion

Central venous catheterization can be performed through various veins, such as the femoral vein (¹²), internal jugular vein (¹³), and brachial vein (¹⁴). The subclavian vein is one of the most frequently used central venous routes in pediatric patients. The subclavian vein’s skin puncture site is less likely to become infected than the other veins’ puncture sites (¹⁵), and the patients are free to move their arms and heads (¹⁰). The RSCV is preferred to the left subclavian vein because of the higher risk of chylothorax (¹⁶) or vascular perforation (¹⁷) with the eft subclavian vein.

The proper depth of a right subclavian catheter is between 16–19 cm in adult patients (^16,18,19). However, in pediatric patients, commonly used equations or guidelines are rare (^20,21). A guideline suggested by Andropoulos et al. (²⁰) is comparable to ours. They suggested a depth of 4 cm for patients less than 3 kg and 5 cm for patients between 3–5 kg. When their guidelines are applied to our results, in 30% (15 of 50) of the patients, the predicted depths were shallower than the measured depth by more than 8 mm. A too-shallow position of the CVC can be associated with serious complications such as phlebitis or thrombosis (^1,3). The difference between their results and ours comes from the way the study was approached. Their guideline is from a mixture of results mostly from the right internal jugular vein catheter and some RSCV catheters. As the distance from skin puncture site to the SVC-RA junction is shorter for the right internal jugular vein in comparison with the RSCV (¹⁸), the equation derived from his data gives a shallower depth than ours. In addition, the central venous anatomy of a small infant is far different from that of large children or adults (^22,23), therefore applying an equation derived using the data from a pool of patients consisting of patients over a large age range may cause an improper result.

Our study was limited to infants <5 kg undergoing RSCV catheter insertion, therefore indicating a practical guideline for depth of a CVC that should be used for infants when using the RSCV. Using our guideline, in 98% of the present patients, the CVC was placed at a proper position (within 5 mm short of SVC-RA junction). For the other 2%, the CVC was only 1 mm deep into the RA, the possibility for the tip of a CVC to contact with the wall of the cardiac chamber is not likely at this position. Therefore, following our guideline, almost every CVC can be placed at a safe position.

There are limitations to our study. We measured the distance from the skin puncture site to the SVC-RA junction, as we defined the ideal position of a CVC as “distal SVC outside RA.” But the ideal position of the tip of a CVC can be “distal SVC outside of pericardial reflection” (⁴). To meet this definition, the length of the pericardium that covers the distal SVC should have been considered in our study. However, it was not possible to achieve the location of the pericardial reflection on TEE. Therefore, our results should be a guideline for “the distance from the skin puncture site to the SVC-RA junction” or “the maximum allowed depth” rather than “the proper depth” of the CVC.

From our study, we conclude that, the maximum depth of the CVC allowed for the RSCV catheter to avoid its intracardiac placement is between 40–55 mm for infants less than 5 kg.

The statistical analysis of the present work was supported by the research fund of Hanyang University (HY-2003-T). We also thank all members of the Laboratory of Statistical Data Analysis of Hanyang University College of Natural Sciences for their expert statistical analysis.