This is the first study on a large sample comparing in-plane US-guided supraclavicular right INV approach with right IJV approach for TIVADs in adults. Our main finding is that the INV technique had a significantly higher first-attempt success rate compared with IJV, and INV approach had lower perioperative and postoperative complications, although the differences between groups about complication rates were not significant. US-guided TIVADs via the right INV approach and the right IJV approach are both safe and reliable
The methods of TIVADs implantation mainly include venous cutdown and venous puncture . At present, percutaneous puncture for TIVADs implantation via IJV and SCV are the most widely used . However, the IJV and SCV may not be the best option for many clinical situations.
TIVADs via IJV approach has a high puncture point and a large folding angle of the catheter (Figure 3). This may lead to reduction of comfort, unattractive appearance and even lead to discount and rupture of the catheter [13, 14, 15]. The SCV approach is more convenient and comfortable than IJV, but SCV approach has the possibility of occurrence of pinch-off syndrome (POS), which is the main cause of catheter rupture .
Cephalic vein cut-down technique has been used for more than 30 years . According to the previous studies, in comparison with the SCV approach, the incidence of complications of cephalic vein approach using surgical techniques for TIVAD is lower, and it is considered superior to the SCV approach [17,18, 19]. KOKETSU et al. [20, 21] also believe that TIVAD can provide safe and feasible infusion channels for patients through cephalic vein, which is worthy of promotion and application. However, surgical techniques for TIVAD also have the disadvantages of long operation time, low success rate, and great trauma.
The IJV merges with the SCV behind the sternoclavicular joint to form the INV, the bilateral INVs converge to form the superior vena cava (SVC). We know that the INV is relatively fixed and has a larger diameter than the IJV and SCV, this provides the possibility of US-guided puncture of INV safely and effectively.
With the development of ultrasound technology, ultrasound-guided INV catheterization has been gradually applied in central vena catheterization (CVC) for almost 10 years, and many studies have confirmed its safety and effectiveness [7, 8, 9]. Early studies about INV catheterization were mainly focused on catheterization for neonates, possibly because of the small diameter of IJV in newborns than the adults and is more difficult to puncture; therefore, the INV approach was adopted.
Studies have shown that the success rate of left INV catheterization in infants and newborns is higher than that in right [22, 23]. However, in adults, the left INV is deeper and more variable than the right, and it is difficult to identify the thoracic duct by ultrasound .
In this study, we chosed to puncture the right INV to avoid lymphatic leakage caused by thoracic duct injury, as reported in our previous study. [10,11]. But in Beccaria's study , 78 patients with left INV catheterization did not have thoracic duct injury. Another study showed that left INV catheterization was safe and feasible in children . Further studies are required to confirm the safety and feasibility of left INV approach of TIVADs for adult patients.
In this study, the success rate of one-time puncture in the right INV group was slightly higher than that in the right IJV group (98.64% vs 95.34%, P = 0.020) (Table 2). This was related to many anatomical advantages of right INV, such as thin subcutaneous tissue, large diameter and clear ultrasound display.
The success rate of the first puncture is related to the personal experience, besides, the ectopy of guidewire during puncture is also an important factor for the reduction of the rate of the first puncture of the internal jugular vein. In the right IJV group, the first puncture failed in 9 cases, among which 4 cases were confirmed to enter the ipsilateral SCV by DSA, and the guide wire was successfully adjusted. In the right INV group, there was no guide wire ectopy during puncture due to anatomical factor: Y-shaped anatomical morphology of IJV, SCV, and INV .
The success rate of the first puncture of INV reported in this study was higher than that of Beccaria et al  (90.18%, 257/285), which might be related to the left INV approach used in Beccaria’s study.
Perioperative complications were mainly related to mechanical injury of central vein puncture. In this study, intraoperative complications rate in the right INV group was lower than that of IJV group although the differences were not statistically significant (1.18% vs 1.43%, P＞0.05) (Table 3).
Perioperative complications were observed in 4 cases in the right INV group and in 3 casesin the right IJV group. The miss punctured artery was identified as right subclavical artery (SCA) or right common carotid artery (CCA) by ultrasound, and two patients were diagnosed as the small amount of pneumothorax by fluoroscopy during the operation. These self-limiting complications were not treated except for puncture site compression. The complete cannulation procedure was then repeated or changed the puncture site successfully.
The port-carrying time was mainly affected by the subjective factors of the patients. By the end of this study, 209 patients had their TIVADs removed autonomously or unplanned, among which 84.67% (177/209) of the patients thought that there was no need to use the infusion port in about 1 year. In addition, 5% (14/280) of patients in the right IJV group were admitted for the removal of TIVADs due to local foreign body sensation or neck discomfort, while this condition was 2.06% (7/339) in the INV group.
The overall incidence of postoperative complications in the study was 3.72% (23/619) (Table 3), which was consistent with that reported in most other studies [25, 26]. Postoperative complication rate of the right INV group was 3.54% (12/339), which was lower than that of the right IJV group (3.93%, 11/280), although the differences were not statistically significant (P= 0.799).
Postoperative complications mainly included poor healing of the incision, catheter-related infection, thrombosis, fibrin sheath formation, catheter malposition, and catheter fracture, and most of these lead to an unplanned remove after the failure of processing measures (Table 3).
It is worth noting that there is no catheter malposition or catheter rupture in the right INV group, and the mechanism for the low incidence of catheter malposition or catheter rupture are still unclear.
Compared with JIV approach, the INV approach has the advantages of low mobility, smooth catheter shape, and it avoid POS with supraclavicular approach, these are important factors that reduce catheter malposition or fracture of the catheter [27, 28, 29]. In addition, Y-shaped anatomical morphology of IJV, SCV, and INV is also a very important factor of INV approach to reduce catheter malposition.
The findings of our study have led to changes in our daily clinical practice. As a result of the high success rate of puncture, low rate of complications with right INV, we now prefer the right INV to IJV for TIVADs.
Considering its anatomical characteristics, we avoid the right INV in some conditions. Considering the need for surgery and radiotherapy, we do not allow patients with right breast cancer. Patients with not controlled local infection in the port area or the acute infection is not controlled effectively was not allowed. The catheter crosses the clavicle, and the emaciated patients are more likely to expose the catheter. The presence of coagulopathy should be considered as a relative contraindication.
Although this is a large sample comparison study with 619 enrolled patients, the study has certain limitations. First, this study was a retrospective review and some data were not accessible. Second, our incidence of postoperative complications might not be accurate, as we might have missed some TIVADs inserted at our hospital but removed in other institutions. Third, it was a monocentric study and all the TIVADs enrolled were of right INV approach or right IJV approach, the data of TIVADs with left INV approach are lacking, which may provide the further reference for clinicians.According to 2014 statistics, there were more than 400,000 cases of infusion port implantation in the United States, and about 477,300 cases of tumor were diagnosed in Germany every year, among which 125,790 cases chose TIVADs for chemotherapy. At present, the clinical application of TIVAD is increasingly extensive, which is obviously better than other long-term intravenous infusion pathways and is the best choice for tumor patients, however, there are still many problems to be solved [30, 31].
US-guided TIVADs via the right INV approach and the right IJV approach are both safe and reliable. There is a clear need for prospective and well-organized studies to confirm the feasibility and safety of the ultrasound-guided INV approach (both right and left) for TIVADs. It may stimulate future research in this area greatly.