1 INTRODUCTION
Childhood cancer burden is attracting global attention, with nearly 90% of those children from low- and middle-income countries.1 Central venous access devices are necessary for children with malignancy requiring long-term intravenous therapy and routine laboratory tests.2,3 Nowadays, totally implantable Venous access ports (TIVAPs) are the preferred choice because of their higher cost-effectiveness, fewer complications and aesthetic advantages compared with peripherally inserted central catheters (PICCs).4,5
Although there is no consensus regarding venous access approach, ultrasound-guided internal jugular vein (IJV) puncture and subclavian vein (SCV) access based on anatomical landmark for TIVAPs implantation are the most common methods in pediatric population.6,7 However, these two methods have certain drawbacks. The IJV approach has disadvantages of higher puncture point, smaller catheter radian, risk of misplacement, catheter bending and blockage, and patient discomfort due to longer catheter trajectory,7 whereas the SCV access is prone to be kinking such as pinch-off syndrome and has a relatively lower first-attempt success rate.8 However, since the initial report of ultrasound-guided brachiocephalic vein (BCV) approach in the supraclavicular region by Breschan et al.9 in 2011, it has been widely used in adults,10children,11-13 infants,14,15 and even premature infants16-18 for central venous catheters (CVCs) insertion procedure. And this approach is associated with a higher puncturing success rate, shorter cannulation time,10,19 and lower complication rate compared with other approaches.12 As such, Avanzini et al.11 recommended that BCV approach should be adopted as the first choice for long-term venous access.
While ultrasound-guided, supraclavicular BCV insertion for TIVAPs implantation in adults was proven to be safe and effective.20,21 However, its use in children has not been reported in the literature. Therefore, the goal of the present study is to describe our preliminary experience of ultrasound-guided TIVAPs placement via the right BCV approach for in pediatric patients with malignancy, aiming to evaluate its technical feasibility, safety and efficacy in this particular patient population.
2 METHODS
2.1 Study population
Electronic medical records of all pediatric patients who underwent TIVAPs implantation via ultrasound-guided insertion of the right BCV in a single institution from July 2018 to June 2021 were retrospectively reviewed. The study was approved by the institutional ethics committee (KY21012), and the written informed consent was obtained from all children’s legal guardians. The surgical decision was confirmed by the multidisciplinary board according to the expert consensus in China. Patients within the first year after this novel technique was adopted were excluded in consideration of the operator learning curve. Patients with coagulopathy (e.g. blood transfusions were performed until platelets value larger than 50 × 109/L) was corrected preoperatively and surgical contraindications (e.g. definite infection in the surgical or other sites) were excluded. There is only one brand of TIVAP devices (Babyport, 4433742, 4.5 F; B.Braun, Inc., Ile-de-France, France) in our study.
2.2 Predefined surgical protocol
2.2.1 Protocol for preoperative preparation
All surgeries were performed by two attending interventional surgeons under general anesthesia composed of muscle relaxation, tracheal intubation and positive pressure ventilation in the hybrid operating room. Both surgeons had experience with hundreds of TIVAPs implantation by ultrasound-guided BCV cannulations in adult patients. The children were placed in supine position on the operating table, with the head tilted to the opposite side, the neck and shoulder properly bolstered, and the supraclavicular area and chest wall in the surgical side fully exposed. The operators stood at the right side of the patients, taking the right BCV approach as an example. The ultrasound machine was placed on the left side of the child to optimize visualization of the screen.
2.2.2 Protocol for BCV approach
A portable ultrasound device with a 13-6 MHz linear-array transducer (M-Turbo; Sonosite, Inc., Bothell, WA, USA) was implemented to identify the BCV. Here we take the right-side access for example. Firstly, a sonographic cross-sectional view of the IJV was obtained by placing the ultrasonic probe perpendicular to the lower neck. Then, the ultrasound probe was moved caudally along the IJV until the confluence of the IJV and the ipsilateral SCV was displayed, where the BCV takes off. Finally, the optimal longitudinal view of the BCV was displayed by turning the probe slightly medially and caudally behind the clavicle. Using in-plane method, the needle was advanced from lateral to medial and into the target vessel under the real-time ultrasonographic surveillance (Figure 1A). In addition, the needle advancement was stopped immediately if the needle was no longer visualized on ultrasound.
2.2.3 Protocol for surgical procedure
Under sterile steps, the right BCV was punctured with a 21G needle after its optimum longitudinal view achieved on ultrasound screen (Figure 1A). If venous blood could be smoothly aspirated, a 0.018-inch-diameter (0.46mm) J shape guide wire was introduced. The guide wire was checked in the superior vena cava under fluoroscopy, and a 3-mm-length incision was made in the puncture site. A peelable sheath was sent into the vessel along the guide wire, and then the catheter was advanced through the sheath following the guide wire being removed. A transverse incision approximately 2-cm-length and a pocket sized to exactly accommodate the port reservoir was created on the right upper chest wall one to two fingers width below the clavicle. Accordingly, the catheter was guided to the pocket from the supraclavicular exit through a tunnel needle, and its tip was adjusted to be positioned at the cavoatrial junction under fluoroscopy (Figure 1B). Subsequently, the catheter was cut and connected to the port body, which was then placed into the pocket after confirming no obstruction and leakage via flushing. Finally, the infraclavicular incision was sutured with a 5-0 absorbable sutures, followed by blood withdrawal and fluid infusion tested again before the incision was covered with sterile dressings (Figure 1C).
2.3 Data collection and follow-up
Research data was obtained from the medical record reviewing and included preprocedural variables (e.g. basic demographics, indication for implantation, certain blood examination); procedural information (e.g. number of attempts, operative time, intraoperative complications); and procedural outcome data (e.g. postoperative complications, timing and reasons for TIVAP removal). Based on the time of occurrence, postoperative complications were divided into early (within 30 days) and late complications (after 30 days). Furthermore, complications were categorized as wound complications (e.g. wound dehiscence, delayed incision healing), mechanical complications (e.g. catheter dysfunction, catheter malposition/ fracture); and infectious complications (e.g. local infection, catheter-related bloodstream infection [CRBSI]). Operation time is calculated from beginning of puncture to incision closure. Catheter dysfunction was defined as inability of blood withdrawal with or without difficulty of fluid injection. The deadline of clinical surveillance was December 31, 2021.
2.4 Statistical analysis
Statistical analyses were performed through the SPSS software (version 25.0). All variables were tested with the Shapiro-Wilk test for normality and verified for completeness. Descriptive statistics were reported as mean ± standard deviation (range) , median (interquartile range [IQR]) and the frequency (%).
3 RESULTS
3.1 Study population
A total of 35 children who underwent TIVAPs placement were identified, with 21 males and 14 females. The patient median age at the time of surgery was 36 months (IQR: 18, 53 months), ranging from 2 to 115 months. The weight at procedure ranged from 6.5 to 38.0 kg with a median of 15.0 kg (IQR: 11.5, 17.0 kg), and only four patients were less than 10.0 kg. Intravenous chemotherapy was the only indication for TIVAPs implantation in the present study population. Underlying diseases were acute lymphoblastic leukemia (25/35, 71.4%), acute non-lymphocytic leukemia (5/35, 17.1%), hepatoblastoma (2/35, 5.7%), and retinoblastoma (2/35, 5.7%). The platelets count was elevated from the median of 102 × 109/L (IQR: 40, 233 × 109/L) on admission to that of 137 × 109/L (IQR: 70, 277 × 109/L) before surgery, and among them ten children whose platelets value less than 50 × 109/L received once or multiple blood transfusions. (Table 1)
3.2 Perioperative results
All of the 35 children’s TIVAP implantations successfully performed via right BCV approach with a success rate of 100%. Vascular access was successful by first attempt in 32 patients (91.42%), by second attempt in two cases (5.71%), and by the third attempt in one child (2.86%). There was no intraoperative conversion to the ipsilateral IJV or the contralateral BCV approach. The average time of operation was 44.63 ± 6.41 mins (range, 34-62 mins), and the fluoroscopy time ranged from 7 to 27 seconds with a median of 10 seconds (IQR: 8, 13 seconds). No procedural related complications (e.g. pneumothorax, inadvertent artery puncture) occurred. Every child began chemotherapy within 3 days after TIVAPs placement, with a median interval time from the end of surgery to initial port access of 1 day (IQR: 1, 2 days). (Table 2)
3.3 Follow-up outcomes
Three patients experienced a total of four complications, including two cases of local hematoma and two episodes of catheter dysfunction. The postoperative complication rate was 11.43%, equivalently a rate of 0.20 complications per 1000 catheter-days across the cumulative 19,723 catheter-days during the TIVAP carrying period with a mean time of 563.51 ± 208.47 days (range, 193-1014 days). Two cases of local subcutaneous hematoma were self-limited after conservative treatment with local compression and dressing changes. One of these hematomas occurred in a 2-month-old, 6.3-kg-weight infant and lasted for nearly two months (Figure 2). Both cases of catheter dysfunction were considered as intraluminal occlusion, presented as inability of blood aspiration and fluid injection. Catheter patency was restored with thrombolytic therapy using urokinase (5,000 IU/ mL) and positive pressure tube sealing for 30-60 minutes. No other complications such as wound dehiscence, catheter-related thrombosis (CRT), catheter malposition or fracture, surgical site infection, CRBSI, pinch-off syndrome and drug extravasation were observed. None of the 35 children required premature removal of the devices. A total of 11 patients (31.4%) had TIVAP removed due to the end of chemotherapy, and the remaining were still in use (Table 2).