Optimal radiologic position of an umbilical venous catheter tip as determined by echocardiography in very low birth weight newborns

Abstract

OBJECTIVE: To compare chest X-ray with echocardiogram (ECHO) in the localization of an umbilical venous catheter (UVC) tip in very low birth weight infants (VLBW). Secondary objectives determined the association between techniques for tip placement by the vertebral body level on X-ray, as well as the length of the thoracic inferior vena cava-right atrial (TIVC-RA) junction by ECHO.

STUDY DESIGN: Prospective, sequentially enrolled, masked, single regional perinatal center study. Shortly after birth, one or more anterior-posterior X-rays were ordered by the clinical team to verify that the UVC tip was fixed in the central right atrium (cRA) or at the TIVC-RA junction. An echocardiogram was performed as soon as possible after the last X-ray and UVC tip location was interpreted by a pediatric cardiologist. The pediatric radiologist and cardiologist were masked with regard to each other’s reading.

RESULTS: The newborns (n = 51) were 27 (±3) weeks by gestational age with birth weights of 1029 (±288) grams (mean±SD). The radiologist read 50 UVC tips (98%) in the cRA or TIVC-RA junction and 1 (2%) in the LA. The cardiologist read 22 (43%) in the cRA or TIVC-RA, 21 (41%) in the LA and 8 (16%) tips could not be located in the heart. When the UVC tip was interpreted by X-ray as located in the TIVC-RA junction 8/29 (28%) were in the LA by echocardiogram. There was no agreement between vertebral level and tip position in the TIVC-RA junction, RA or LA. The TIVC-RA junction measured 6±1 mm and correlated with birth weight r = 0.54 (p < 0.001).

CONCLUSION: In VLBW newborns, placement of the UVC tip into the cRA or TIVC-RA junction by X-ray does not avoid misplacement in the left atrium as demonstrated by echocardiography. For VLBW infants, it is suggested that echocardiography may be helpful in verifying that the original placement or migration of the UVC tip into the LA has not occurred.

Keywords

Umbilical venous catheter functional echocardiography ultrasound central line neonate

1 Introduction

Umbilical venous catheter (UVC) placement has been commonly used for vascular access in critically ill newborns of all birth weights immediately after birth for many decades [1, 2]. Although there are formulas that can help the clinician approximate the catheter insertion length [3 –5], the standard of care is then to provide radiologic proof that the UVC tip is in an effective and safe position [6]. Previous studies, combining enrollment of preterm and term newborns, have used echocardiography to indicate that the central right atrium (cRA) and/or the thoracic inferior vena cava-RA (TIVC-RA) junction are two safe locations for the UVC tip, but there is no international agreement [7, 8]. Importantly, there have been no studies evaluating these locations with a focus on very low birth weight infants, where the optimal tip location may be very restricted due to the smaller size of these patients and the possible limitations of radiologic interpretations.

The path that an electively placed umbilical catheter takes is via the ductus venosus, the very short abdominal and thoracic inferior vena cava segment into the TIVC-RA junction or central RA. In the fetus, the preferential anatomical distribution of oxygenated umbilical venous blood flows along this path, then is directed by the Eustachian valve principally into the left atrium via the foramen ovale. After birth a UVC tip could also take the same course [9]. Therefore, we hypothesized that in very low birth weight infants, the tip of a UVC considered acceptable in one of the two optimal positions by X-ray, could be actually located in unwanted positions, particularly the left atrium. The primary objective of this study was to determine the association of the X-ray versus echocardiography when the UVC tip was interpreted by X-ray as being either in the central right atrium or TIVC-RA junction. Secondary outcomes included: 1) determination of the length of the TIVC-RA junction by echocardiography in all study patients; and 2) whether the vertebral body location of the UVC tip on X-ray was a reliable tool.

2 Methods

2.1 Study design

This was a single center, prospective, sequentially enrolled, masked study at a Regional Perinatal Center. This clinical study was approved by the Institutional Review Board of Stony Brook University. Patients with birth weights from 400 to 1500 grams requiring an umbilical venous catheter met inclusion criteria. Exclusion criteria were met if there was an abnormality of the diaphragm or abdominal wall, complex congenital heart disease, situs inversus or a UVC tip in the liver by X-ray. Informed consent was obtained from the parents for all patients.

The UVC (polyurethane umbilical vessel catheter, Covidien, Mansfield, MA) was inserted by a member of the clinical team using standard techniques for measurement of appropriate length. [3 –5]. The UVC was sutured into the Wharton’s jelly before X-rays were taken. The clinical team members who inserted the UV catheter obtained one or more anterior-posterior (AP) chest X-ray views, which always included the upper part of the abdomen. All patients were centered appropriately for the X-rays. No lateral X-rays were used for positioning the catheter tip, since they are not routinely used and have also not been shown to be reliable for determining exact UV tip location especially in preterm neonates [10]. When the clinical team was satisfied with the final position of the catheter in the central RA or the TIVC-RA junction by AP X-ray, the line was fixed with an Umbilical Vessel Catheter Anchor (Covidien, Mansfield, MA) adhered to the patient’s abdomen within several minutes after the X-ray was taken. The final position of the catheter tip by X-ray, prior to the echocardiogram, was officially read by one senior pediatric radiologist using conventional landmarks [11]. The radiologist also assigned the vertebral level of the catheter tip that was later analyzed in association with the echocardiogram results. The radiologist was masked to the echocardiogram readings until the end of the study. The abdominal girth was measured just above the umbilicus immediately after the UVC was fixed in place as well as at the time when the echocardiogram was obtained. Abdominal girth variations should be considered as a cause of UVC migration [12].

Echocardiography was performed by one senior pediatric cardiologist or an experienced echocardiography technician as soon as possible after the last placement X-ray by the clinical team was done. All echocardiograms were interpreted by the same pediatric cardiologist. Echocardiography was done with Ultra Diagnostic Ultrasound System (Zonare Medical Systems Inc., Mountain View, CA) using imaging transducer C10-3 or C9-3 at 9 mhz. Three standard pediatric echocardiographic views were used to assess UVC tip placement: subxiphoid short-axis, apical 4-chamber and parasternal short-axis [13]). The pediatric cardiologist was masked to the official X-ray readings until the end of the study. The time interval between the X-ray and echocardiogram was recorded as well as the best view to see the tip. If the UVC tip was in the LA by echocardiography, only the clinical team was made aware of the tip position by the principal investigator.

One of the secondary objectives of the study was to define and measure the length of the TIVC-RA junction by echocardiography. By X-ray, the TIVC-RA junction was predefined as being at or just above the diaphragm as opposed to centrally in the RA. By echocardiography, this junction appears as a tubular structure, contiguous with the abdominal part of the inferior vena cava bordered by the Eustachian valve on one side and the septum primum on the other side. The TIVC-RA junction was measured as the distance from a line tangent to the diaphragm where the IVC entered the thorax to a parallel line starting at the base of the atrial septum. This zone is depicted in Fig. 1.

Fig.1

Schematic representation of the thoracic IVC-RA junction with helpful cardiac landmarks identified. The junction length is measured between the two parallel lines in millimeters by echocardiography in a subxiphoid short-axis view. Abbreviations: SVC, superior vena cava; RAA, right atrial appendage; RA, right atrium; LA, left atrium; EV, eustachian valve; IVC-RA, inferior vena cava-right atrial junction; IVC, inferior vena cava.

2.2 Study sample

All patients were sequentially enrolled if they met inclusion and exclusion criteria. Figure 2 shows the enrollment process for the study patients. Analyses of study data were carried out after 51 patients were enrolled in the study period of June 2014 to July 2015. The sample size was a convenience sample based on the number of babies available. This sample of 51 patients is representative of the larger number of eligible patients (n = 57) with regard to the primary outcome, with a margin of error of±4.5%.

Fig.2

Selection of the study population.

2.3 Statistical analysis

Descriptive statistics were obtained for all study variables. Means (with standard deviations), medians (with range), and proportions are reported as appropriate to the data type (i.e. continuous or categorical). A paired samples t-test was used to determine if there was a difference in the abdominal girth measurements at the time of the last X-ray and prior to the echocardiogram. Linear regression analysis was used to determine the correlation between birth weight and the length of the thoracic IVC (Pearson correlation coefficients are reported). All tests of significance are two-tailed and evaluated at the level of p < 0.05. Data analyses were conducted using IBM SPSS Statistics, Version 21.

3 Results

Table 1 shows the clinical characteristics of the enrolled patients. The results of a paired samplest-test demonstrate that abdominal girth measurements at the time of the last X-ray and prior the echocardiogram are not significantly different (p = 0.07).

Table 1
Demographic characteristics of study patients (n = 51)

Gestational age (weeks) 27±3^b

Birth weight (g) 1029±288^b

Female 51% (26)

Respiratory support

Intubated 49% (25)

NIMV or NCPAP 45% (23)

Room Air 6% (3)

Time between birth and study X-ray (hours) 22±20^b

Time for echo study (mins) 7±3^b

Time between X-ray and echo (hours) 4 (1.5–14)^c

Abdominal circumference difference (mm)^a 2±5^b

Length of thoracic IVC-RA junction (mm) 6±1^b

^aAbdominal circumference difference between time of X-ray and time of echocardiogram. ^bValues are mean±SD; ^cValues are median (25% – 75%).

Figure 3 shows an X-ray with a UVC tip at the level of the thoracic IVC-RA junction, as read by the masked pediatric radiologist. The abdominal girth was 20 cm at the time of X-ray and 20 cm at the time of echocardiography. The UVC tip was best seen in the LA on parasternal short axis view on echocardiogram. In this view, the aortic root is readily visualized, providing a reliable marker just anterior to the LA. The best views for locating the UVC as per the masked pediatric cardiologist was the parasternal short axis and the subxiphoid short axis views, 65% and 20% of the time respectively. The time to complete the ECHO search for the UV tip was 7±3 minutes (mean±SD).

Fig.3

Parasternal short axis echocardiographic view of the UV catheter tip in the left atrium in a very low birth weight infant. On the chest X-ray the catheter tip appears to be in the thoracic inferior cava /right atrial junction. Abbreviations: RA, right atrium; LA, left atrium; UVC, umbilical venous catheter; TIVC-RA, thoracic inferior vena cava-right atrial junction.

Table 2 demonstrates the primary outcome using the condition that the UVC tip was at the correct position radiographically if the tip was either in the central RA or at the thoracic IVC-RA junction for all the study patients. There were eight patients wherein the catheter was not intracardiac, and therefore assumed to be at or just below the diaphragm. This was confirmed based on subsequent X-ray review after enrollment of all patients. Under this radiologic condition of acceptability, approximately 40% of the catheter tips were found in the left atrium by echocardiography. There were 29 patients that had UVC tips that met acceptable conditions as defined as only in the thoracic IVC-RA junction by screening X-ray (Table 3; based on echocardiography: 16 weremalpositioned (8 in the LA and 8 in the RA), 7 in the TIVC-RA junction and 6 were not visualized.

Table 2

Echocardiographic location of the UVC tip when screening X-ray is interpreted as acceptable in the central right atrium or the thoracic IVC-right atrial junction

UV tip location	By	By
	echocardiogram	X-ray
	(%)	(%)
Central right atrium or thoracic	22 (43)	50 (98)
IVC-right atrium junction
Left atrium	21 (41)	1 (2)
Not found by echocardiography	8 (16)	0
Total subjects	51	51

Table 3

Echocardiographic location of the UVC tip when screening X-ray is interpreted as acceptable in the thoracic IVC-right atrial junction alone

UV Tip Location	By Echocardiogram (%)	By X-ray (%)
Thoracic IVC- right	7 (24)	29 (100)
atrium junction
Right atrium	8 (28)
Left atrium	8 (28)
Not found by	6 (20)
echocardiography
Total subjects	29	29

When we examined the association between X-ray and echocardiogram, specifically for those babies with birth weights ≤1000 grams (n = 20), we found different results compared to the primary outcome. Two subjects were not included in the next analysis as the catheter tip was not intracardiac by echocardiogram. If the UVC tip was in the central RA or TIVC-RA junction by X-ray, 6/18 (33%) were in the LA by echocardiogram. Under these conditions, if the UVC tip was only at the TIVC-RA junction by X-ray, 0/6 (0%) were in the LA byechocardiogram.

Figure 4 shows that using the vertebral body location on x-ray as a marker of where the UVC tip should be, was not reliable, compared to an echocardiogram.

The TIVC-RA junction based on the a priori definition of this zone was 7±1 mm (mean±SD) for neonates with birth weights of 1001 to 1500 grams and 6±1 mm for babies from 510 to 1000 grams. There was a correlation between the length of TIVC-RA and birth weight, r = 0.54 (p < 0.001).

Fig.4

Depiction of UVC tip location via echocardiography in relation to thoracic vertebral level as seen on AP chest X-ray, (n = 43).

4 Discussion

This was a prospective, masked study with consecutive enrollment, that focused specifically on very low birth weight infants. The study provides newevidence in this particular population that when a UVC tip is thought to be in the central RA or IVC-RA junction radiographically, there are an appreciable number of UVC tips actually in the left atrium. Elective UVC placement has been a mainstay of vascular access for critically ill newborns for decades. Safety concerns regarding UVC have previously focused primarily on thrombosis, arrhythmia, atrial perforation, hemorrhage, pericardial tamponade and infection [6 , 15]. However, little is known as to whether inadvertent and undetected location of the UVC tip in the left atrium could cause serious morbidity due to systemic circulatory disturbances in very low birth weight infants. As a start, to address this potential safety concern, the present study used either the central RA or the TIVC-RA junction location of the UVC tip by X-ray as clinically acceptable [7 , 16]. The present study also defined and demonstrated that placement of the UVC tip in the TIVC-RA junction by an AP screening X-ray will reduce but not eliminate inadvertent malposition of the UVC tip in the LA as proven by echocardiography.

There are no authoritative guidelines as to where a UVC tip catheter should be initially placed and what the frequency of surveillance of that tip should be, in case of unexpected UVC migration [12]. The present study has shown that the optimal zone for the UVC tip in the TIVC-RA junction is very small in length and, not unexpectedly, directly correlates with birth weight. Specifically, in very low birth weight infants, this zone has been defined a priori and measured in this study population to be approximately 5-8 mm in length. One of the first studies to formally evaluate X-ray versus echocardiographic location of the UVC tip was written by Ades and coworkers [7]. UVC tips were considered in a clinically acceptable position if they were located in the RA, thoracic IVC-RA junction or thoracic inferior vena cava. Fifty three patients were enrolled with birth weights of 550 to 4320 grams; 45% of the UVC tips were in the left atrium. The authors mentioned that prematurity was a significant risk factor for catheter placement in the left atrium. However, the distribution of birth weights in the enrolled patients was not described. In a retrospective study, 51 patients were studied who had a UVC in place for 24 hours or more; in a subset of patients <1000 grams, 9 out of 22 patients had catheters in the left atrium. This past study did not define what position of the UV tip by X-ray was considered acceptable prior to echocardiography as it was retrospective [17]. In contrast, the present prospective study found that in the subset of neonates weighing ≤1000 g at birth, no catheter was in the LA if the X-ray showed that the UV tip was in the thoracic IVC-RA junction. A caveat to this interpretation is that this subpopulation was small.

From the present study, it appears that the X-ray zone where the UVC tip will be less likely to be found in the LA by echocardiography, is the thoracic IVC-RA junction. There is no clear boundary between the thoracic IVC and the RA by X-ray in VLBW infants and there are no previous measurements made by echocardiography in VLBW infants. The thoracic IVC-RA junction was anatomically defined as the distance from a tangent to the diaphragm where the IVC entered the thorax to a parallel line starting at the base of the atrial septum. For the very low birth weight infant the zone was approximately 5–8 mm and the length of the zone correlated with the birth weight of the baby. It is not surprising that with this small zone that using the vertebral body level of the UV tip was not a reliable method for ruling out a tip misplaced into the left atrium, as previously published [7, 18].

There are some limitations of the present study. First, in the design of this study, one assumed that one could test the association of the X-ray screening tool to tip placement by the echocardiogram. For the most part this was true but in 8 out of 51 patients, the tip of the UVC could not be seen byechocardiogram. Based on the review of these eight X-rays after enrollment was completed, the tips were very close to the diaphragm and the masked radiologist read that the tip was in an acceptable position. Therefore between the density of the diaphragm and the liver just below it, this may be the reason why the echocardiographer could not see the tip. In a study using evaluation of UVC tip position by ultrasound, views of the catheter tip could be seen in the hepatic vein or liver parenchyma [19]. Comparing ultrasound versus echocardiogram for UV tip location has not been studied in the newborn. The tiny length of the abdominal IVC above the ductus venosus and below the diaphragm anatomically could not be seen by echocardiogram.

Another limitation of this study was that the X-ray and echocardiography results were only compared once. Therefore, it is not known whether UV tip migration was possible between the study X-ray and echocardiogram or later in the hospital course. However, the combination of suturing the UVC into the Wharton’s jelly along with fixation of the UVC outside the abdomen with a UVC anchor immediately after the X-ray, indicated tip migration could only occur if there was a change in abdominal girth [12]. The abdominal circumference between the X-ray and echocardiogram in the present study only changed by an average of 2.5 mm. Therefore it is unlikely that significant migration occurred. Over time, since the thoracic IVC-RA zone in the VLBW infants is approximately 5–8 mm, migration of the tip into an undesirable zone may occur with major changes in abdominal girth. There is a possibility that changes in lung expansion could affect UVC tip migration although this factor was not explored in this study.

5 Conclusion

In very low birth weight infants, the best location to position the UV tip by X-ray is at the thoracic RA-IVC junction; this reduces but does not eliminate inadvertent misplacement into the LA. Two studies have shown that routine use of an UVC for blood sampling and infusions can affect cerebral oxygenation in premature babies but the location of the UVC tips in these studies were not confirmed by echocardiography [20, 21]. Only a large multicenter quality improvement study could help answer the question as to whether there can be adverse systemic effects (e.g. blood clots or air bubbles) of a UV catheter tip in the LA. With the increasing use of echocardiography and ultrasonography by neonatologists, these modalities could be utilized in the proper placement of the UVC tip more frequently in the future, especially in very low birth weight newborns [22, 23]. Imaging intracardiac central lines in pediatrics is a straightforward technique that is a well-established standard in echocardiography. This study utilized a targeted, multi-plane imaging approach to limit misdiagnoses of line position within the heart. This protocol could potentially be learned and adopted by a non-cardiologist in a bed-side setting. In the course of our study, it was learned that the parasternal short axis view could be obtained conveniently out of the sterile field and demonstrates the aortic root landmark rather easily. This technique, when inserting the UVC and verifying the catheter tip position, could be easily taught to novice users.

Disclosure statement

All authors have no financial relationships or conflicts of interest to disclose for this study to disclose for this study. Research involving human subjects were conducted in accordance with the ethical standards of all applicable national and institutional committees and the World Medical Association’s Helsinki Declaration.

Footnotes

Acknowledgments

We would like to thank the families of our neonatal intensive care unit patients and nurses and practitioners for their participation in this study.

References

Diamond

, Allen

Jr , Thomas

Jr . Erythroblastosis fetalis – treatment with exchange transfusion. N Engl J Med 1951;244(2):39–49.

Weber

, DeLuca

, Shannon

. Normal and abnormal position of the umbilical artery and venous catheter on the roentgenogram and review of complications. Am J Roentgenol Radium Ther Nucl Med 1974;120(2):361–7.

Gupta

, Peesay

, Ramasethu

. Simple measurements to place umbilical catheters using surface anatomy. J Perinatol 2015;35:476–80.

Shukla

, Ferrara

. Rapid estimation of insertional length of umbilical catheters in newborns. AMA Am J Dis Child 1986;140(8):786–8.

Verheij

, te Pas

, Smits-Wintjens

, Sramek

, Walther

, Lopriore

. Revised formula to determine the insertion length of umbilical vein catheters. Eur J Pediatr 2013;172(8):1011–5.

Raval

, Gonzalez

, Bhat

, Pearlman

, Stefano

. Umbilical venous catheters: Evaluation of radiographs to determine position and associated complications of malpositioned umbilical venous catheters. J Perinatol 1995;12(3):201–4.

Ades

, Sable

, Cummings

, Cross

, Markle

, Martin

. Echocardiographic evaluation of umbilical venous catheter placement. J Perinatol 2003;23(1):24–8.

Oestreich

. Umbilical vein catheterization–appropriate and inappropriate placement. Pediatr Radiol 2010;40(12):1941–9.

Schlesinger

, Braverman

, DiPietro

. Pictorial Essay. Neonates and umbilical venous catheters: Normal appearance, anomalous positions, complications, and potential aid to diagnosis. AJR AM J Roentgenol 2003;180(4):1147–53.

10.

Butler

, Al-Assaf

, Tarrant

, Ryan

, El-Khuffash

. Using lateral radiographs to determine umbilical venous catheter tip position in neonates. Ir Med J 2014;107(8):256–8.

11.

Brodsky

, Martin

. Neonatology Review. 2nd Ed. Brodsky

and Martin

, eds. 2010. p. 135.

12.

Salvadori

, Piva

, Filippone

. Umbilical venous line displacement as a consequence of abdominal girth variation. J Pediatr 2002;141(5):737.

13.

Lai

W.W.

, Geva

, Shirali

G.S.

, Frommelt

P.C.

, Humes

R.A.

, Brook

M.M.

, et al. Guidelines and standards for performance of a pediatric echocardiogram: A report from the Task Force of the Pediatric Council of the American Society of Echocardiography. J Am Soc Echocardiogr 2006;19:1413–30.

14.

Moeckel

, Cresalia

, Vachharajani

. Umbilical vein catheterization. Neoreviews 2013;14(8):e416–e8.

15.

Green

, Yohannan

. Umbilical arterial and venous catheters: Placement, use, and complications. Neonatal Netw 1998;17(6):23–8.

16.

Said

, Rais-Bahrami

. Umbilical Vein Catheterization. In: MacDonald

MG RJ

, editor. Atlas of Procedures in Neonatology. 5th ed. Lippincott Williams and Wilkins; Philadelphia, PA: 2013. pp. 173–81.

17.

Harabor

, Soraisham

. Rates of intracardiac umbilical venous catheter placement in neonates. J Ultrasound Med 2014;33(9):1557–61.

18.

Greenberg

, Movahed

, Peterson

, Bejar

. Placement of umbilical venous catheters with use of bedside real-time ultrasonography. J Pediatr 1995;126(4):633–5.

19.

Hoellering

, Koorts

, Cartwright

, Davies

. Determination of umbilical venous catheter tip position with radiograph. Pediatr Crit Care Med 2014;15(1):56–61.

20.

Huning

, Horsch

, Roll

. Blood sampling via umbilical vein catheters decreases cerebral oxygenation and blood volume in preterm infants. Acta paediatr 2007;96(11):1617–21.

21.

Bray

, Stucchi

, Fumagalli

, et al. Blood withdrawal and infusion via umbilical catheters: Effect on cerebral perfusion and influence of ibuprofen. Biol Neonate 2003;84(3):187–93.

22.

Fleming

, Kim

. Ultrasound-guided umbilical catheter insertion in neonates. J Perinatol 2011;31(5):344–9.

23.

Pulickal

, Charlagorla

, Tume

, Chhabra

, Narula

, Nadroo

. Superiority of targeted neonatal echocardiography for umbilical venous catheter tip localization: Accuracy of a clinician performance model. J Perinatol 2013;33(12):950–3.

Gestational age (weeks)	27±3^b
Birth weight (g)	1029±288^b
Female	51% (26)
Respiratory support
Intubated	49% (25)
NIMV or NCPAP	45% (23)
Room Air	6% (3)
Time between birth and study X-ray (hours)	22±20^b
Time for echo study (mins)	7±3^b
Time between X-ray and echo (hours)	4 (1.5–14)^c
Abdominal circumference difference (mm)^a	2±5^b
Length of thoracic IVC-RA junction (mm)	6±1^b