Children with congenital and acquired cardiac disease often will require extended central venous access before and after surgical interventions and for the treatment of chronic cardiac failure. These populations are at substantial risk for catheter-associated blood stream infections and vessel thrombosis, as well as the inherent risks of repeated invasive procedures and exposure to sedation. The small size of these infants and their complex venous anatomy may also present technical challenges for successful placement of central venous lines. In many instances, sparing of the larger venous vessels such as the femoral and internal jugular vein is desirable. This attenuates the risk of catastrophic vessel occlusion and preserves these sites for possible future cardiac catheterisations. This is particularly important with single-ventricle patients who may undergo multiple surgeries and staged repairs. There is extensive and long-term experience with the use of peripherally inserted central catheters in neonates and infants.Reference Trotter1–Reference Dolcourt and Bose4 Peripherally inserted central catheters are well-tolerated for extended periods of time and may avoid disruption and damage to the larger central venous vessels. Successful percutaneous insertion of a peripherally inserted central catheter may also eliminate the need for more invasive cut-down procedures.
Several models of peripherally inserted central catheter insertion programmes exist in paediatric centres.Reference Gamulka, Mendoza and Connolly5–Reference Tan, Hess and Diaz7 These models vary from a bedside nurse programme in the neonatal intensive care unit,Reference Linck, Donze and Hamvas6 to a “hybrid” model of nurses placing peripherally inserted central catheters in radiology with radiology supportReference Gamulka, Mendoza and Connolly5 or an all interventional radiology physician programme.Reference Tan, Hess and Diaz7 Before the implementation of bedside placement and a nurse-led peripherally inserted central catheter service at The Children’s Hospital and The Heart Institute in Denver, the peripherally inserted central catheters placed in children with congenital cardiac disease were performed so by attending cardiac intensivists, neonatologists, paediatric cardiac fellows, or interventional radiologists, and performed by a variety of techniques. At present, at The Children’s Hospital, the peripherally inserted central catheter team is exclusively led and staffed by dedicated registered nurses. Here, we present a 36-month review of the team’s performance for peripherally inserted central catheter placement in neonates and infants in the cardiac intensive care unit.
Materials and methods
Patients and study design
After receiving institutional review board approval, a review was conducted on data gathered prospectively on the 161 consecutive peripherally inserted central catheter insertions in neonates and infants, 12 months of age and under, admitted to the cardiac intensive care unit from January, 2006 to December, 2008 for peri-operative or medical care. The cardiac intensive care unit is a 16-bed intensive care unit dedicated to the care of acutely ill neonates, infants, children, and young adults with congenital or acquired cardiac disease. The peripherally inserted central catheters placed were used for multiple types of therapy and infusates including vasoactive medications, parenteral nutrition, and long-term antibiotic therapy. All insertions were performed at the bedside by a peripherally inserted central catheter nurse, who as part of the hospital-wide peripherally inserted central catheter team is dedicated to line placement, monitoring, and care of these lines. Insertions were performed by direct visualisation, palpation, or ultrasound.
A database and spreadsheet was developed and maintained by the primary investigators and data collection was initiated at the time of insertion and completed once the catheter was removed. Variables collected prospectively included patient characteristics, underlying cardiac diagnoses, indication for catheter placement, procedural specifics including vein used, catheter characteristics, and other insertion-related complications. Ongoing catheter monitoring and removal information was also collected.
Educational requirements for registered nurses before peripherally inserted central catheter clinical training included didactic and practical components for insertion and care and management of the catheters by a qualified instructor as approved and recommended by the Association of Vascular Access.Reference Manly and Williams8 Successful insertions were performed by the nurse with the preceptor present, until competency was achieved as agreed upon by both clinicians.
Description of the catheter placement
All procedures were performed bedside in the cardiac intensive care unit with the necessary cardiopulmonary monitoring. Sedation was provided as necessary by the attending physician or the bedside nurse following the American Society of Anesthesia Guidelines. The decision to use sedation and the choice of sedation was at the discretion of the cardiac intensive care physician. At the time of peripherally inserted central catheter placement, the nurse and the physician devised a sedation plan for each patient. If propofol was used, the physician was present to titrate the infusion to affect. When intermittent sedation was used, this was given at the discretion of the peripherally inserted central catheter nurse and the bedside nurse.
The veins were located primarily by palpation or visualisation. Ultrasound was provided by a SonoSite 180™ with a 38-millimetre linear probe of 5–10 megahertz (SonoSite, Bothell, Washington, United States of America) was available for part of this study period. A majority of catheter placements were performed in the upper extremity although a few lower extremity insertions were attempted. After the location and quality of veins were assessed and a site chosen, the non-intubated infant was swaddled and an estimation of catheter length was determined by following the pathway of the vein from insertion site to one finger-breadth above the nipple line. Infants requiring mechanical ventilation were not swaddled. The extremity was prepared and draped in a sterile fashion and maximal sterile barrier precautions9 were followed. A modified Seldinger techniqueReference Pettit10–Reference Royer13 was used in all of the procedures. The vein was accessed using 22- or 24-gauge peripheral intravenous catheters. A guide wire of 0.014 or 0.018 inch was then advanced through the cannula into the vein but advanced no further than the proximal portion of the extremity. The site was then anaesthetised subcutaneously with a small amount of 1% lidocaine solution without epinephrine buffered with sodium bicarbonate ratio of 10:1. The catheter was then trimmed by using a Trimming Tool™Reference Pettit14 (Becton Davis, Franklin Lakes, NJ, USA) according to the previously taken measurement. By using a two-piece peel away 5-centimetre dilator, the vein was dilated just enough to ensure that the larger portion of the dilator was indwelling. The inner portion of the sheath dilator was removed and the catheter advanced. Techniques used to facilitate placement included slow advancement, retracting the stylet 0.5–1 centimetre into the body of the catheter or using a “float-in” flushing technique with a syringe of normal saline. If resistance was encountered during advancement, the extremity and/or patient’s head was repositioned. Ultrasound in some instances was used to scan the ipsilateral internal jugular region to check whether or not the catheter was malpositioned.
Portable X-ray was obtained bedside to confirm peripherally inserted central catheter placement. The peripherally inserted central catheter nurse was responsible for repositioning the catheter if necessary. Central access was defined as the catheter tip residing in the superior caval vein, cavoatrial junction, or inferior caval vein. Dressing changes were performed by the peripherally inserted central catheter team, time permitting, otherwise this was performed by the primary bedside nurse. The peripherally inserted central catheter nurses were also available for occlusion management or other problems related to site care. The endpoint for tracking catheter outcomes was upon catheter removal or discharge date if the patient was discharged or transferred with the catheter in place for the purpose of the study.
Results
A total of 161 procedures were attempted with successful catheter placement in 125 cases, in a total of 105 patients, for an overall success rate of 78% (125 of 161). Demographic data (Table 1) are based on the first insertion in cases of multiple insertions. The median age at the time of peripherally inserted central catheter placement was 7 days with a mean age of 37 days, range: 1–265 days. In all, 74% of the 105 patients were neonates, defined as greater than or equal to 28 days of age. The median weight at time of insertion was 3.3 kilograms with a mean weight of 3.5 plus or minus 1.1 kilograms with a range from 1.8 to 7.6 kilograms. Cyanotic cardiac lesions accounted for 76% of the patient’s diagnoses. The most common diagnosis was hypoplastic left cardiac syndrome found in 29% of patients, followed by the transposition of the great arteries in 13% of patients.
Table 1 Patient characteristics.
Table 2 outlines procedural information for the 161 total procedures attempted. The basilic and cephalic veins were the most commonly accessed veins. An ultrasound was used for visualisation in 23% of the cases. A SonoSite S-ICU™ (SonoSite, Bothell, Washington, United States of America) using a 25-millimetre linear probe of 6–13 megahertz) was implemented in October, 2008. With this newer technology used in the last 3 months of the study, the rate for successful peripherally inserted central catheter insertion increased to 90%. Patients received intravenous sedation during 57% of the procedures, and a titrated continuous propofol infusion was used in 3% of the procedure for five patients. A total of 38% patients with 61 procedures did not require any sedation and were managed with comfort measures including swaddling and a sucrose-dipped pacifier. Central placement was achieved in 75% of the insertions. About 29 malpositioned catheters in jugular veins, contralateral subclavian veins or right atrium required manipulation to achieve successful central placement.
Table 2 Procedural and catheter information.
SCV, superior caval vein
The mean number of attempts needed for successful placement was 2.4 plus or minus 1.3. Procedural complications included inadvertent arterial access in five patients, that is, 3% of the total procedures attempted. One of these neonates experienced a transient arterial compromise after arterial access. He was taken for exploration and underwent a sympathectomy, but found to have an intact brachial artery. He has since fully recovered with no neurovascular or functional sequelae. Of note, all the inadvertent arterial punctures occurred before the adoption of ultrasound use. The other procedural complications noted were documented in two non-intubated patients who experienced desaturations related to hypoventilation with the administration of sedation. These patients required transient non-invasive support but did not decompensate further and did not require intubation. Catheters remained in place for up to 71 days, with a mean dwell time of 17 days and a median time of 30 days. The peripherally inserted central catheter team was able to provide same day insertion service in 60% of the referrals and another 30% of the insertions were performed within 24 hours of the referral. The remaining peripherally inserted central catheter lines were placed within 3 days after the referral was received.
Table 3 outlines catheter performance for the 125 successfully placed catheters. Successful treatment included completion of therapy, home discharge, and transfer to another facility, or expiration with the catheter in place. This was achieved in 72% of the time (90 catheters). This left a total of 35 instances in which early catheter removal was required. The most frequent complications were catheter occlusions in seven (6%) of patients, six (5%) broken or leaking catheters, and five (4%) unplanned catheter removals. Four patients developed catheter-associated bloodstream infections with Staphylococcus species, none were identified as Staphylococcus aureus. The infection rate was 1.9 incidences per 1000 line days. This was below the hospital’s internal target rate for catheter-associated bloodstream infections for intensive care patients.
Table 3 Catheter performance and reasons for removal.
A total of 36 of the 161 peripherally inserted central catheter insertion attempts were unsuccessful. The most common reasons as cited by the peripherally inserted central catheter nurses were difficulty with advancing guide wires and/or dilators in 16 instances, or difficulty in accessing the vein in 11 instances.
Discussion
This study describes the adoption and performance of a nursing led, bedside placement peripherally inserted central catheter team in a population of critically ill neonates and infants with congenital and acquired cardiac disease. The neonate and infant with either congenital or acquired cardiac disease may present particular challenges to successful placement and maintenance of central venous lines. The peripherally inserted central catheter team in our cardiac intensive care unit experienced few complications as well as a low incidence of catheter-associated bloodstream infections. The team was also able to respond to referrals in a rapid and efficient manner. All the insertions were performed at the bedside in the cardiac intensive care unit. This permitted patients to remain under close monitoring and reduced the inherent risks of transporting a critically ill patient to the radiology suite. Continuing evaluation should show an improvement as the nurses gain experience.
The peripherally inserted central catheter team depended primarily on palpation techniques to achieve successful placement. The SonoSite 180™ was available for vein visualisation for the majority of the study, but found to be more useful on larger infants because of the size of the probe and the resolution of the image. Our success rate increased dramatically with the introduction of the SonoSite S-ICU™ late in the study used for vein visualisation. A majority of our patients suffered from cyanotic cardiac disease and were prostaglandin dependent. They, therefore, often required placement of peripherally inserted central catheters within the first few hours and days of life. This acuity as well as their weight at the time of insertion may speak to the challenges of placing peripherally inserted central catheters in these patients. In fact, as shown by the high number of neonates with hypoplastic left cardiac syndrome, these may be considered among the most tenuous of patients in the cardiac intensive care unit.
Organisations such as the Association for Vascular Access,15 Infusion Nursing Society,16 and Society of Interventional RadiologyReference Lewis, Allen and Burke17 have released position statements or guidelines stating that the distal third of the superior caval vein or the cavoatrial junction are the recommended tip locations for central lines. The national association of neonatal nurses guidelines for practiceReference Pettit and Wyckoff18 states that catheter tips should reside in the superior caval vein and this could also be included for the inferior caval vein. Centrally placed tips are associated with lower rates of complications such as thrombosis, catheter occlusion, and leaking at the catheter insertion site.Reference Racadio, Doellman, Johnson, Bean and Jacobs19–Reference James, Bledsoe and Hadaway21 From the author’s experience it was not always possible to obtain central access in neonates and infants through the bedside techniques of palpation, visualisation, and ultrasound guidance. However, the ability to obtain central placement at the bedside compared favourably in our model to others who collected initial X-ray location data.Reference Fricke, Racadio, Duckworth, Donnelly, Tamer and Johnson22, Reference Chaturvedi, Bithal, Dash, Chauhan and Mohanty23 One quarter of the peripherally inserted central catheters placed did not meet the previously mentioned criteria for central venous location. This must be considered a limitation of bedside peripherally inserted central catheter placement. Ultimately these devices were not removed and were successfully utilised after the therapy or infusates were adjusted accordingly.Reference Thiagarajan, Bratton, Gettmann and Ramamoorthy24
Other models for peripherally inserted central catheter placement in paediatrics have utilised interventional radiologists and fluoroscopy,Reference Tan, Hess and Diaz7 bedside registered nurses in a neonatal intensive care unit,Reference Linck, Donze and Hamvas6 and a hybrid model of registered nurses placing peripherally inserted central catheters with the support of radiology.Reference Gamulka, Mendoza and Connolly5 These models provide a similar success rate for placement as well few complications. In the author’s opinion, an important benefit of bedside placement is maintaining critical care monitoring in a safe environment of an often fragile or unstable patient. Related to this, another benefit of our bedside insertion model is eliminating transportation of a potentially unstable patient out of the cardiac intensive care unit. Finally, this bedside model also avoids the cost of fluoroscopy that other models may incur. Of course, in the absence of a direct comparison of cost, it is difficult to say which of these models is financially preferable. Future research should include a cost analysis of different models.
A limitation of this study is that it is reflective of the experience of a single centre. There was no analysis of data before the implementation of an exclusive bedside nursing led peripherally inserted central catheter insertion model; thus it is not possible to state that the dedicated bedside insertion team led by nurses is necessarily better than other models. It also remains to be seen if with the increased use of bedside ultrasound to assist insertion, whether the incidence of complications such as inadvertent arterial puncture can be further reduced.
Conclusion
A nursing led peripherally inserted central catheter team can provide safe and efficient peripherally inserted central catheter insertions in the critically ill neonate and infant with congenital and acquired cardiac disease. More experience is needed to determine the degree of improvement that may occur with the use of the upgraded portable ultrasound; however, it is possible to build a successful programme of bedside peripherally inserted central catheter insertion lead by dedicated and trained nurses in a congenital cardiac centre.
Acknowledgements
The authors would like to acknowledge Dunbar Ivy, MD, and Annie Harrier, RN, for their support of the Children’s Hospital Heart Institute’s PICC team and Jeannie Zuk, PhD, BSN, for their review on the manuscript. The authors declare that there is no financial support for this study.
