Central venous catheterization in neonates: Comparison of complications with percutaneous and open surgical methods.

AIM: To compare the complications of two methods of placement of central venous catheters.
MATERIALS AND METHODS: One hundred neonates had percutaneously inserted central venous catheters and another 100 had the catheters placed after surgical incision and vein location.
RESULTS: No statistical difference was noted in the complication rate or efficacy
CONCLUSIONS: Both the methods are equally safe and effective.

INTRODUCTION

Central venous catheter (CVC) placement is essential in neonatal intensive care units (NICU). Although percutaneously inserted central venous catheter (PCVC) is a reasonable choice,[1] tunneled or surgically placed central venous catheters (SCVC) are used when placement of PCVC has failed. In this study, we compared the complications and outcome with percutaneous and open surgical methods for central venous catheterization in sick neonates.

MATERIALS AND METHODS

This study was carried out in two NICU centers in central part of IRAN (ST-Zahra and Shahid Beheshti hospitals, Isfahan province) from September 2007 to September 2010:

In group 1-100 patients who had failure of PCVC technique were selected as SCVC group; in group 2 - after matching for age, sex, and body weight variables, 100 patients with successful PCVC technique were selected for comparison.

Neonates with a medical diagnosis of hypercoagulable state, sepsis or babies who died due to the original disease (not attributed to CVC complications) were excluded.

In the PCVC group, catheters were placed in the right jugular vein (RJV) under anesthesia in the operating room by Seldinger's method. In the SCVC group, under local and monitoring anesthesia care (MAC), transverse incision on right sternocleido mastoid muscle was made. The right internal jugular vein was dissected free from internal carotid artery. After proximal and distal control of vein, transverse venotomy was performed and a 16 G Broviac catheter (Arrow) was inserted into the vein. The venotomy site was pursed with Prolene 6/0. The catheter end was brought out through a subcutaneous tunnel from the right subclavian region. The catheter was anchored using Prolene 3/0 suture placed on the skin at the exit site. Radiographic assessment of catheter tip location upon insertion for ensuring a safe location was performed in all patients. All intraoperative complications were recorded in the surgical notes and any postoperative event was documented in the patient's medical record. Intraoperative complications were pneumothorax, hemothorax, bleeding, arrhythmia, and air embolism. On postoperative assessment, the duration of catheterization and postoperative complications (in-hospital infection, delayed bleeding, leak from tunnel, catheter malfunction and mal-position) were recorded. For evaluation of long-term patency of RJV, 4 weeks after catheter removal, Doppler ultrasound was performed monthly for 6 months by the radiologist who was unaware of the allocation of the study participants.

Statistical analyses were performed using Chi - square or Fisher's exact test (SPSS 11, O, IL. USA). Differences in age, weight, gestational age and catheter live days were compared using Mann-Whitney U test. AA P Value less than 0.05 was considered statistically significant.

RESULTS

There were 54 females and 46 male patients in the PCVC group with a mean age of 18. 47 ± 47. 9 days and 60 female and 40 male patients in the SCVC group with a mean age of 20. 21 ± 31. 2 days (P = NS). The mean weight of patients in PCVC was 2.87±1.76 kg in the PCVC group and 2.77± 1.38 kg in the SCVC group (P = NS). The lowest weight was 700 g. The mean gestational age was 32.1±4.3 weeks in the PCVC group and 33.2±4.2 weeks in the SCVC group (PP = NS). There was no significant difference in the indications of CVC in the two groups. Catheter live days were 18.1 ± 11.9 days (6--98 days) in the PCVC group and 22.3 + 10.01 days (7-57 days) in the SCVC group. There was no significant difference in the catheter live days between the two groups. Intraoperative failure has occurred in 10 patients in the PCVC group. We have no intraoperative failure in the SCVC group (P = NS). We had no hemothorax, arrhythmia or sudden death due to air embolism in either group. Arterial puncture occurred in five patients in the PCVC group (5%). Pneumothorax occurred in one patient in the SCVC group (1%). In-hospital line sepsis which was confirmed with cultures of catheter tip occurred in one patient in the SCVC group (P = NS). Catheter dislodgement occurred in 8 patients in the PCVC group (8%) and three patients in the SCVC group (3%) (P = NS). Leak from tunnel was seen in eight patients in the PCVC group (8%) and two patients in the SCVC group (2%) (P = NS).Vein patency in 6 months of follow-up was 84% (84 patients) in PCVCs and 91% (91 patients) in the SCVC group (PP = NS).

DISCUSSION

Central venous catheters (CVC) provide many important functions for sick neonates. However the use of these catheters also carries significant risk of permanent disability or even death.[2] These catheters are placed into the superior vena cava (through jugular veins, facial vein, cephalic vein or subclavian vein) and inferior vena cava (through groin vein).[3] Although primarily the first choice of CVC is a peripherally inserted central venous catheter by a vein viewer device, but this device is not available in every NICU. The recent advent of portable ultrasound (US) machines with pediatric probes has resulted in the development of new approaches that, if correctly learned and used, should allow quicker and safer vascular access in this population.[4] However, RJV catheterization in pediatric patients is sometimes difficult even for skilled physicians who use ultrasound guidance. The main reasons for such a difficulty in performing catheterization are thought to be the small vein size and vein collapse during catheterization. Therefore in some difficult cases, surgically inserted CVC can be an optional method.[5]

In this study we compared two methods of catheter insertion into the internal jugular vein. In our study, vein patency was 91% in the open method. Although in one of the largest retrospective study in 16 Canadian tertiary pediatric centers, CVC-related thrombosis was 3.5 per 10,000 admissions,[6] but in other prospective studies,[7-12] the incidence of catheter--related thrombosis was 8-35%. Specifically, in two studies,[1112] the incidence of internal jugular vein thrombosis was 8 and 10% respectively. Although according to these studies, the catheter location is an important factor for vein thrombosis, our study showed that the method of catheter placement in a certain vein (open or percutaneously) is inconsistent as a risk factor for thrombosis. In this study, we did not use venography for evaluation of vein patency. Venography is an invasive method and requires injection of contrast material with its related complications.

In this study, we had one line-documented sepsis. Infection is the main complication of the indwelling catheter, with an incidence of 0-29%.[13] In our hospital, we place all catheterss in the operation room and under aseptic conditions and wounds were covered by sterile dressing.

In conclusion, although the open surgical catheter placement is commonly stated as a factor in the incidence of catheter-related thrombosis, our study showed that the evidence for this risk factor is inconsistent and by meticulous surgical manipulation, surgeons can place this catheter as safe as a percutaneous method.