Journal of Surgery

Introduction

Two key historical moments for the implantation of Total Implantable Venous Access Devices (TIVADs) were that until the 1960s/70s, Cephalic Vein Cut-Down (CVC) in the deltopectoral fold was the primary route for pacemaker implantation [1], and that in 1982 Niederhuber implanted the first TIVAD via CVC in this location [2]; in both cases, a single incision was made. The other main route for TIVAD implants is the subclavian vein, following the technique described by Seldinger [3] in 1953.

However, in this case, two incisions are made, either via the infraclavicular approach described by Aubaniac [4] in 1952 or the supraclavicular approach by Yoffa [5] in 1965. Until the end of the 20^th century, the Subclavian Vein Puncture (SVP) technique using anatomical landmarks was usually carried out ‘blind’, without direct visualization of the vessel. However, from that point onward, intraoperative ultrasound was used to localize the vessel, thereby increasing success rates and decreasing complications. Before beginning to implant TIVADs in our setting, a literature review was conducted, revealing two key findings: that the success rate with the CVC technique was less than 85% in most series [6], with lack of success due to alterations or absence of the Cephalic Vein (CV), and that ultrasound imaging improved these rates [7]. Therefore, from the outset of this study, all patients scheduled for TIVAD implantation underwent preoperative ultrasound imaging of the CV. This study aims to demonstrate that a TIVAD can be implanted with a single incision via both CVC and SVP, and that the preoperative use of CV ultrasound significantly increases CVC success rate and reduces postoperative complications.

Methods

This is a retrospective observational cohort design study carried out between 2008 and 2025 in the Oncology Department of the private IMQ Zorrotzaurre Hospital in the Basque Country (Spain), which covers a population of 250,000 inhabitants. The protocol followed for the implants was the preoperative measurement of the diameter, depth, and characteristics of the cephalic vein in both upper extremities using a MyLab50 color Doppler ultrasound (Esaote). If the diameter was ≥3.2 mm, the implant was performed via CVC, as this diameter corresponds approximately to the 9.6F gauge of the catheter used in this work, a NuPort HP* System (PHS MEDICAL) with a single-chamber titanium port and silicone catheter. Otherwise, implantation was performed using the SVP technique. For this approach, a 21-gauge needle from a Micropuncture set (AngioDynamics) was first used to cannulate the vein, followed by the NuPort set. Intraoperative radiological monitoring was performed using a BV Bracelet (Philips). All patients had a preoperative anesthetic evaluation deemed suitable for the procedure and were informed about the technique to be performed; they subsequently signed the corresponding informed consent form.

Figure 1: Surgical Technique. 1A: Inferior border of clavicle (1) and incision in the deltopectoral groove (2) - 1B: Cephalic Vein (CV) dissection - DM: Deltoid Muscle- PMM: Pectoralis Major Muscle -1C: Subclavian Vein Access by Anatomical Landmarks

The most notable aspects of the surgical technique (Figure 1) are that the system is always implanted on the left side, and the incision extends from the middle third of the inferior border of the clavicle, 4-5 cm distally along the deltopectoral crease. The dissection of the CV for the CVC technique is performed in the usual manner, and for the SVP technique, the Seldinger technique is carried out using the Aubaniac approach. Once the catheter tip is positioned in the atrium under fluoroscopic guidance, a subcutaneous pocket is created to accommodate the port. Finally, the port is secured with two 3-0 Prolene sutures (Ethicon) to the anterior aspect of the pectoralis major muscle. No contrast agents, antibiotic prophylaxis or anticoagulants were used during the procedures. Right Great Saphenous Vein (RGSV) implantation was performed by dissecting the RGSV after locating it in the upper third of the inner thigh using ultrasound, introducing the catheter at that level, and then positioning and securing the port on the anterior aspect of the quadriceps (Figure 2). Surgical time in all cases was considered to run from the infiltration of local anesthesia until closure of the incision. Intraoperative complications were considered to be those occurring within the first 24 hours, and postoperative complications were considered to be those occurring up to 30 days post-implantation.

Figure 2: Implant via Great Saphenous vein; 2A: Surgical technique -2B: Radiographic Control

To determine if there were any differences in the procedures, the results were divided into two time periods: 2008–2016 and 2017– 2025. Finally, it should be noted that all ultrasound explorations and implantations were performed by the same vascular surgeon. For the statistical analysis, the baseline characteristics of the study groups used means with their corresponding Standard Deviations (SD) for continuous variables, and frequencies and percentages for categorical variables. Differences between sexes were compared according to sociodemographic, clinical variables and vein characteristics. Differences in operating time and complications between periods (2008-2016 versus 2017-2025) were compared. If the compared variable was numerical, the Wilcoxon rank-sum test was used, since data was not normally distributed, and if the variable was categorical, Chi-squared or Fisher’s exact test was used, as appropriate. Statistical significance was set at P value ≤ 0.05 and statistical analysis was performed by using RStudio® version 4.3.1.

Results

From January 2008 to December 2025, 1300 TIVADs were implanted in 1249 patients, with 51 patients (4.1%) receiving 2 or 3 TIVADs. 62.5% were women, and the mean age was 62.4

years, and significantly higher in men than in women. Colorectal cancer was the most frequent type of neoplasia in men (39.3%), while breast cancer was the most frequent type in women (44.6%). 151 patients (12.1%) had had a previous neoplasm; 63 were men, with urological cancers the most frequent (34 cases, 54.0%), and 88 were women, with gynecological cancers the most frequent (19 cases, 21.6%) (Table 1).

Table 1: General Characteristics.

	All	Male	Female	P
Gender, n (%)	1249 (100)	468 (37.5%)	781 (62.5%)	<0.0001
Age, years (mean ±  SD)	62.4± 1.6	65.9±11.0	60.4±11.4	<0.0001
Neoplasm location, n (%)
Oropharyngeal	41 (3.3)	27 (5.8)	13 (1.7)	<0.0001
Colorectal	365 (29.2)	184 (39.3)	181 (23.2)	<0.0001
Gastroesophageal	100 (8.0)	65 (13.9)	35 (4.5)	<0.0001
Hematologic	24 (1.9)	12 (2.6)	12 (1.5)	0.0333
Hepatopancreatic	102 (8.2)	55 (11.8)	47 (6.0)	<0.0001
Breast	367 (28.0)	2 (0.4)	348 (44.6)	<0.0001
Reproductive organ	82 (6.6)	13 (2.8)	69 (8.8)	<0.0001
Lung	126 (10.1)	67 (14.3)	60 (7.7)	0.0013
Urologic	36 (2.9)	32 (6.8)	4 (0.5)	<0.0001
Other	23 (1.8)	11 (2.4)	12 (1.5)	0.3194
Previous neoplasm, n (%)	151 (12.1)	63 (13.5)	88 (11.3)	0.2963

SD: Standard Deviation

Both the diameter and mean depth of the CVs were significantly larger in men. The CV was absent in 5.2% of the scans, and in 59 cases (4.5%), some anomaly was found that would prevent subsequent implantation via CVC, with crossing stenosis the most frequent, occurring in 3.5% of the scans. In 4 patients (0.3%), ultrasound studies were not performed because implantation via RGSV was chosen. Ultimately, 1074 CVs (82.8%) were candidates for CVC implantation and 222 (17.2%) for SVP implantation (Table 2).

Table 2: Anatomical Characteristics of the Cephalic Veins.

The results were classified according to the recipient vein, the side of the implant, and the implantation period. In 1084 cases (83.4%), the TIVAD was implanted via left venous access, and of these, 81.7% were via CVC. Two hundred and twelve TIVADs were implanted on the right side, and the main reasons for this were previous breast surgery with left lymphadenectomy in 118 cases (55.7%) and previous TIVAD implantation via a left-sided SVP in 51 cases (24.1%). Thirty patients were left-handed, representing 14.2% of the implants and 2.4% of the overall study population. Four TIVADs (0.3%) were implanted via the right great saphenous vein, all in women due to venous and/or lymphatic pathology in both upper extremities. Table 3 presents the distribution of surgical time and access routes by period, showing that procedures are significantly shorter in the 2017-2025 period, but access routes do not show major differences.

There were 1065 TIVADs scheduled via CVC (81.9%), but on 4 occasions (0.4%) surgery could not be performed due to a large irreversible spasm of the CV, resulting in a 99.6% success rate. Of the 231 implanted via SVP (17.5%), 227 were scheduled, and 4 were CVC reversals. On one occasion (0.4%), a significant arterial injury occurred that required endovascular repair and the implant was discontinued, resulting in a 99.6% success rate for both techniques (Table 3).

Table 3: Impalant access routes by period.

There were 27 complications (2.1%), 15 (1.2%) of which were intraoperative and 12 (0.9%) postoperative. Of the 15 intraoperative complications (1.2%), 4 (0.3%) were from CVCs and 11 (4.7%) from SVPs, of which SVP complications were significantly more frequent, primarily due to pneumothorax. Six (0.5%) patients were men and 9 (0.7%) were women. Furthermore, 10 (0.9%) complications occurred on the left side compared to 5 (2.4%) on the right, a statistically significant difference (p < 0.005).

Of the 12 postoperative complications, 5 occurred in men (0.4%) and 7 in women (0.5%), while 7 (0.6%) occurred on the left side and 5 (2.4%) on the right, also a statistically significant difference (p < 0.005). The main postoperative complication was ipsilateral subclavian DVT (1.7%), which, although more frequent with SVPs, was statistically significant. In the overall count, complications were significantly higher for SVP (Table 4). No complications occurred with RGSV implantation.

Table 4: Complications by Access Route.

Finally, no significant differences were found in complications related to the time periods (Table 5), although a significant decrease in all complications was observed in the second period.

Table 5: Complications by period

Discussion

Although authors such as Atti [6] state that CVC is the best route for pacemaker implantation and consequently for TIVAD, in the current literature, for many, such as Kilic [7], this route does not even exist. This is primarily due to two factors: first, the numerous non-surgical