Quantitative and qualitative evaluation of residual deposits in implanted intravenous ports informs maintenance strategies

Abstract

This study aimed to evaluate the distribution and composition of intraluminal deposits in implanted intravenous ports removed from oncology patients, with a focus on assessing the effectiveness of current nursing maintenance protocols. Ports removed after completion of cancer treatment were analyzed. Each catheter was segmented into proximal, middle, and distal thirds for regional analysis. Total protein concentrations were measured using the bicinchoninic acid (BCA) assay. Thrombosis-related proteins—fibrin, fibrinogen, thrombin, and P-selectin—were identified by Western blot analysis. A total of 46 patients were included. Three distinct patterns of protein residue distribution were identified between the injection chamber and the catheter. Eleven patients exhibited comparable protein levels in both regions, but only seven demonstrated adequate irrigation. Eighteen patients showed higher protein accumulation in the catheter, while seventeen had greater deposits in the injection chamber. The distal catheter segment consistently exhibited the highest protein concentration, particularly in ports implanted for more than five years. Western blot analysis confirmed the presence of thrombogenic proteins, and the absence of β-actin indicated minimal cellular debris. These findings suggest that mechanical flushing plays a more critical role than pharmacologic flushing in deposit clearance. Proteinaceous deposits persist despite adherence to current protocols, with a notable presence in areas of suboptimal flow, such as the chamber base and distal catheter. These findings suggest a link between flow dynamics and deposit accumulation, indicating that revisions to nursing guidelines—including standardized insertion, increased flushing volumes, and simulation-based competency training—may help support improved port function and potentially lower complication risks.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-53383-w.

Introduction

Implanted intravenous ports are essential for oncology patients, providing reliable vascular access for chemotherapy, nutritional support, transfusions, and imaging procedures^1–4. To ensure long-term functionality, routine flushing and maintenance are critical components of post-treatment care. However, because implanted intravenous ports cannot be directly visualized by nursing staff or monitored using real-time imaging, there are currently no definitive methods to assess puncture orientation or evaluate intraluminal irrigation status⁵. Current nursing guidelines typically recommend flushing with 10 mL of normal saline followed by 10 mL of heparin solution (50 IU/mL)⁵. However, emerging evidence suggests that this protocol may be inadequate, as accumulation of blood components within ports has been observed despite adherence to recommended practices^6–8. Reported malfunction rates range from 0.54% to 2.98%, often necessitating port revision^9–11. These findings raise concerns about the true efficacy of current maintenance strategies.

Previous studies have demonstrated that intraluminal protein deposits commonly form within the injection chamber of implanted intravenous ports^6–8. These deposits consist of biological substances, primarily fibrin, albumin, and other plasma proteins, that accumulate on the inner surfaces of the catheter lumen.^12,13 From the perspective of irrigation efficacy, two major factors should be considered: irrigation orientation and flushing volume. After adequate irrigation, heparinized solution is typically administered to prevent intraluminal blood clot formation. Therefore, sufficient irrigation to minimize residual proteins followed by heparin lock solution is recommended by current maintenance guidelines (5). Ex vivo simulations have shown that the intraluminal flow pattern within implanted ports consists of a double vortex within the injection chamber, followed by laminar flow along the catheter (14-16). To achieve optimal irrigation efficacy, nursing staff must identify the implanted port, puncture the non-coring needle at the center of the injection chamber, and maintain the optimal orientation (180°) between the needle opening and the chamber (e) (Fig. 1A,B)¹⁵. However, implanted ports are positioned subcutaneously, and needle insertion is guided by palpation rather than direct visualization. This may introduce variability in puncture angle and depth among nursing staff, potentially disrupting optimal flow dynamics (Fig. 1C,D). In real-world clinical settings, deviations from the ideal irrigation conditions—such as off-center puncture or improper orientation between the non-coring needle opening and the injection chamber—may lead to suboptimal irrigation flow. Consequently, blood components, medications, or protein residues may accumulate within the lumen, thereby compromising irrigation efficacy. Flushing volume is another critical factor. Previous research indicates that flushing with approximately 20 times the internal volume of the port is required to effectively remove protein deposits under ideal conditions¹⁶. In clinical practice, however, variations in needle puncture among nursing staff may further disrupt irrigation flow, suggesting that larger flushing volumes may be necessary to achieve adequate cleaning.

Fig. 1.

Intraluminal flow patterns in implanted ports and the impact of individual variation in non-coring needle puncture technique. A. Lateral view: Intraluminal flow pattern within the injection chamber under ideal irrigation conditions, demonstrating a double-vortex formation. The flow within the catheter is laminar. B. Top view: Intraluminal flow pattern within the injection chamber under ideal irrigation conditions, showing the characteristic double-vortex flow. The catheter exhibits laminar flow. C. Lateral view: Illustration of individual variations in non-coring needle puncture technique. (Becton, Dickinson and Company, NJ, USA). D. Top view: Illustration of individual variations in non-coring needle puncture technique. (Becton, Dickinson and Company, NJ, USA).

From the perspective of catheter locking, heparinized solution is recommended by maintenance guidelines to prevent blood clot formation. However, the literature indicates that clot formation at the catheter tip has been reported in nearly all implanted ports, regardless of heparin use⁶. The vacuum effect created during needle withdrawal, combined with silicone diaphragm rebound, may draw blood into the catheter, resulting in heparin–blood mixing and potential clot formation¹⁷. Given the short half-life of heparin (approximately 1.5 hours)¹⁸, its anticoagulant effect may be insufficient between maintenance intervals, particularly during long-term port use¹⁹.

To gain further insight into the mechanisms underlying port malfunction, this study performed a detailed analysis of proteinaceous deposits in explanted ports. The catheter was segmented into proximal, middle, and distal sections to evaluate the regional distribution of intraluminal protein residues. Western blot analysis was also performed to identify key thrombosis-related proteins. The aim of this study is to provide both quantitative and qualitative evidence to guide future improvements in port maintenance protocols, including irrigation strategies and standardized procedural training.

Materials and methods

Study design and ethical consideration

Study design

This prospective study, conducted from January 2022 to April 2024, enrolled 46 oncology patients who had completed treatment and were scheduled for elective port removal. All devices had served as secure vascular access and received standard maintenance throughout their indwelling period. Removal was performed in the operating room using a standardized safety protocol. While catheter fracture is a rare complication, its clinical management differs significantly from biological issues like intraluminal protein deposits or infection. Because fractured fragments may migrate into the central veins, heart, or pulmonary artery—potentially precipitating intravascular thrombosis or ventilation–perfusion mismatch—emergent endovascular retrieval via the femoral vein is mandatory to prevent lethal embolic events.^20–22 Consequently, our institutional protocol incorporates intraoperative fluoroscopy prior to and during the removal procedure. This imaging ensures the device remains intact, definitively rules out catheter fracture, and verifies the complete extraction of the system.

Ethical consideration

This study was approved by the Institutional Review Board of Chang Gung Memorial Hospital (IRB No. 201509029B0, 201901829A3, and 202200949B0) and was conducted in accordance with the principles of the Declaration of Helsinki. All participants provided written informed consent prior to port removal and inclusion in this study. This work was supported by Chang Gung Memorial Hospital, Linkou, Taiwan (grant numbers: CMRPG3M1281 and CMRPG5G0131). All participants provided written informed consent prior to port removal and inclusion in this study.

Port implantation technique

For SVC port placement, the coracoid process of the scapula served as the primary anatomical landmark. Following sterile preparation and local anesthesia, a 2-cm incision was made over the deltopectoral groove to expose the target vasculature, such as the cephalic vein or the deltoid branch of the thoracoacromial vein.^10,23–25 The cannulation strategy was determined by vessel morphology: a direct cutdown was performed for vessels with sufficient caliber and a linear course, while wire-guided assistance—with or without puncture—was utilized for tortuous or small-caliber vessels. In cases of complex venous anatomy, intraoperative venography was employed to clarify the cannulation route. If the cephalic vein or deltoid branch was inaccessible, the internal jugular vein (IJV) served as a rescue route via ultrasound-guided high-neck puncture.^10,23–25

For IVC port implantation, the anterior superior iliac spine (ASIS) and pubic tubercle were utilized as landmarks. This approach required two separate incisions: a subinguinal incision for the exploration of the great saphenous vein and catheter insertion, and a second incision near the ASIS to secure the port housing to the rectus abdominis fascia.^10,23–25

Port maintenance

The maintenance protocol for implanted intravenous ports at our institution was standardized as follows.⁵ Nursing staff performed a sterile skin preparation of the area surrounding the port site, followed by the application of sterile draping over the pocket. After manual palpation to stabilize the device, a non-coring (Huber) needle was used to puncture the injection chamber perpendicularly. To optimize flow dynamics, the needle opening was oriented at a 180 degree angle relative to the catheter outflow tract whenever possible.¹⁵ Following needle placement, a blood withdrawal test was performed to confirm patency and ensure the port was functional for clinical use. After each administration, the system was flushed with 10 mL of normal saline (push–pause technique) and locked with 10 mL of heparinized solution (50 IU/mL) to prevent intraluminal protein deposits and thrombosis. For patients who have completed their treatment course, port irrigation is scheduled at three-month intervals as part of routine outpatient maintenance services

Subjects and sampling

Sample collection and preparation

After surgical removal, each port was immediately cleaned of external tissue and placed in an ice bucket. The sample was delivered to the laboratory within one hour. Before disassembly, a full photographic record of the intact port was taken. Key data including port brand, serial number, implantation and removal dates, and duration of use were carefully documented. All extraluminal tissue was completely excised and cleared from the device prior to its disassembly. Ports were then disassembled into three components: catheter, locking nut, and injection chamber. Care was taken to prevent fluid leakage from the lumen. (Supplementary Figure 1 A, B) The catheter was examined on a light box (Supplementary Figure 1 C) to assess residual intraluminal fluid, which was aspirated and quantified. Photographs were taken of the catheter’s internal surfaces, including the tip and bottom. (Supplementary Figure 1 D) The injection chamber was disassembled (Supplementary Figure 1 E), and the silicone diaphragm was cut along its edge to expose the inner surface, which was photographed. Fluids from within the chamber were aspirated for subsequent protein analysis. The catheter was also examined for fibrin sheaths and then cut into proximal, middle, and distal segments for protein quantification. (Supplementary Figure 1 F)

Patient grouping

The catheter was the only component of the implanted system positioned intravascularly. For this study, catheters were categorized into two groups based on their constituent manufacturing materials. The intraluminal flow pattern within the implanted port is characterized by a double vortex within the injection chamber, transitioning to laminar flow through the catheter, from the proximal connection site to the distal tip^14,15. Because protein deposition likely varies along the irrigation flow path, the catheter was partitioned into three equal segments—proximal, middle, and distal thirds—for independent protein analysis. Furthermore, we compared protein accumulation between the injection chamber and the catheter. Based on this distribution, patients were stratified into three distinct cohorts: the comparable distribution group, the catheter-predominant deposition group, and the chamber-predominant deposition group.

Protein extraction (Bicinchoninic acid assay) and quantification

To analyze protein deposits, the port chamber and catheter components were digested with 0.1% Trypsin-EDTA (Cat# 15400054, Gibco, Canada) and subjected to a 15-minute water bath sonication at 37 °C to facilitate the detachment of residual proteins. Total proteins were then extracted using ice-cold Tissue Protein Extraction Reagent (Cat# 78510, Thermo Fisher Scientific, USA), supplemented with a protease inhibitor cocktail at a 1:100 (v/v) dilution (Cat# P8340, Sigma-Aldrich, USA). Protein concentrations were measured using the bicinchoninic acid (BCA) assay with the Pierce BCA Protein Assay Kit (Cat# 23227, Thermo Fisher Scientific, USA). For the assay, 10 μL of each protein sample and bovine serum albumin (BSA) standards were loaded onto a microplate, followed by 200 μL per well of BCA working solution prepared by mixing Reagents A and B at a 50:1 ratio. The microplate was incubated at 37 °C for 30 minutes, cooled to room temperature, and absorbance was measured at 562 nm using a microplate reader (TECAN-Infinite M200 PRO, Switzerland). Protein concentrations were calculated using a standard curve derived from the BSA standards.

Western blot analysis

Due to limited protein yield, only 16 out of 46 port chamber samples (34.8%) contained sufficient material for Western blot analysis. Equal amounts of total protein (35 μg) from each sample were separated using 8% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) at 80 V for 2.5 hours, then transferred onto a 0.45 μm PVDF membrane (Immobilon-P, Cat# IPVH85R, Merck Millipore, Ireland) at 100 V and 4 °C for 2 hours. Membranes were pre-activated with methanol prior to transfer. Blots were blocked for 1 hour at room temperature in 7% non-fat milk/TBST, followed by overnight incubation (16 hours at 4 °C) with the following primary antibodies: mouse monoclonal anti-P-selectin (1:100; Cat# sc-8419, Santa Cruz Biotechnology, USA), rabbit polyclonal anti-fibrinogen (1:5000; Cat# A0080, Dako, USA), mouse monoclonal anti-fibrin (1:500; Cat# MABS2155, clone 59D8, Merck Millipore, Germany), rabbit polyclonal anti-thrombin (1:500; Cat# ab92621, Abcam, UK), and mouse monoclonal anti-β-actin (1:200; Cat# sc-47778, Santa Cruz Biotechnology, USA). After washing with TBST, membranes were incubated for 2 hours at room temperature with horseradish peroxidase-conjugated secondary antibodies: either goat anti-mouse IgG (1:5000; Cat# AP124P, Millipore, Germany) or donkey anti-rabbit IgG (1:2000; Cat# NA934, Cytiva Life Sciences, USA). Protein bands were visualized using the UVP ChemStudio imaging system (UVP ChemStudio PLUS Touch, Analytik Jena, Germany) in conjunction with Western Lightning Plus ECL substrate (Cat# NEL105, PerkinElmer, USA).

Statistical analysis

All data are reported as mean ± standard error of the mean (SEM). Statistical analysis was performed using GraphPad Prism (version 8.0.2). Continuous variables are presented as mean ± standard deviation, whereas categorical variables are reported as number (percentage). Categorical variables were compared using the chi-square test or Fisher’s exact test when appropriate. For non-parametric data, Mann–Whitney U and Kruskal–Wallis tests with Dunn’s post hoc comparisons were applied. For datasets with sample sizes ≥30 and normal distributions, parametric tests such as unpaired two-tailed t-tests and one-way ANOVA with Tukey’s multiple comparisons were used. A p-value of less than 0.05 was considered statistically significant.

Results

This study investigated the characteristics and protein residue accumulation within implanted intravenous ports used in oncology patients receiving chemotherapy. A total of 46 patients were enrolled (Table 1), with an average port implantation duration of 4.3 ± 2.1 years, ranging from 0.1 to 10.2 years. The average number of port chamber punctures was 23.6 ± 12.9 per patient, ranging from 1 to 71. Gross findings showed 91.3% had tip clot formation, 21.7% had a catheter fibrin sheath, and 56.5% showed intraluminal fibrin deposits. The specifications of the implanted ports used in the study are detailed in Supplementary Table 1.

Table 1.

Characteristics of the study cohort.

Variables	N (%) / Mean± SD	No. of port with silicone catheter	No. of port with polyurethane catheter	p-value
Case No.	46 (100%)	38 (82.6%)	8 (17.4%)	-
Gender
Male	21 (45.7%)	18 (39.1%)	3 (6.5%)	0.7098
Female	25 (54.3%)	20 (43.5%)	5 (10.9%)
Age (years):	61.7 ± 12.8	62.3±11.8	59.0±11.5	0.7132
< 60 years	14 (30.4%)	12 (26.1%)	2 (4.3%)
≥ 60 years	32 (69.6%)	26 (56.5%)	6 (13.1%)
Underlying cancer
Head and neck	4 (8.7%)	2 (4.3%)	2 (4.3%)	0.2181
Pulmonary	6 (13.0%)	6 (13.0%)	0 (0%)
GI (Upper & Lower)	21 (45.7%)	18 (39.1%)	3 (6.5%)
Others 1	15 (32.6%)	12 (26.1%)	3 (6.5%)
Stage
Stage I	8 (17.4%)	7 (15.2%)	1 (2.2%)	0.8941
Stage II	14 (30.4%)	11 (23.9%)	3 (6.5%)
Stage III	15 (32.6%)	12 (26.1%)	3 (6.5%)
Stage IV	9 (19.6%)	8 (17.4%)	1 (2.2%)
Reason of removals
Chemotherapy completion	41 (89.0 %)	33 (71.1%)	8 (17.4%)	0.9467
Catheter fracture	1 (2.2%)	1 (2.2%)	0 (0%)
Catheter malfunction	1 (2.2%)	1 (2.2%)	0 (0%)
Catheter blockage	1 (2.2%)	1 (2.2%)	0 (0%)
Port migration	1 (2.2%)	1 (2.2%)	0 (0%)
Poor wound healing	1 (2.2%)	1 (2.2%)	0 (0%)
Puncture no. --chemotherapy	11.2 ± 7.3	11.6 ± 7.5	9.6 ± 6.4	0.6641
1-12	36 (78.3%)	29 (63.1%)	7 (15.2%)
13-33	10 (21.7%)	9 (19.5%)	1 (2.2%)
Puncture No -- Treatment	11.6 ± 7.4	12.0 ± 7.6	9.6 ± 6.4	0.4088
1-12	33 (71.7%)	26 (56.5%)	7 (15.2%)
13-35	13 (28.3%)	12 (26.1%)	1 (2.2%)
Puncture No -- Maintenance	12.0 ± 9.6	11.5 ± 8.6	14.4 ± 14.0	0.4525
0-12	28 (60.9%)	22 (47.9%)	6 (13.0%)
13-47	18 (39.1%)	16 (34.7%)	2 (4.4%)
Time of placement	4.3 ± 2.1	4.1 ± 1.9	5.1 ± 2.6	0.4429
0.1 - 5.0 years	29 (63.0%)	25 (54.3%)	4 (8.7%)
5.1 - 10.2 years	17 (37.0%)	13 (28.3%)	4 (8.7%)
Puncture –Total	23.6 ± 12.9	23.5 ± 11.4	24.0 ± 19.4	0.1278
1-20	22 (47.8%)	16 (34.8%)	6 (13.0%)
21-71	24 (52.2%)	22 (47.8%)	2 (4.4%)
Gross presentation
Tip clot	42 (91.3%)	35 (76.1%)	7 (15.2%)	0.8013
Catheter fibrin sheath	10 (21.7%)	9 (19.6%)	1 (2.1%)
Fibrin deposit	26 (56.5%)	21 (45.7%)	5 (10.8%)

¹ Others ^: GU malignancy, Hematologic malignancy, Breast malignancy, Double primary malignancy

The total protein residues quantified within the injection chambers and catheters are summarized in Fig. 2A, with overall trends illustrated in Fig. 2B. Based on the relative distribution of these residues, three distinct protein deposition patterns were identified: (1) comparable levels in both the chamber and catheter, (2) higher deposition in the catheter than in the chamber, and (3) higher deposition in the chamber than in the catheter. In comparable distribution group (n=11), seven patients (15.2%) were characterized by low protein deposition in both the chamber and the catheter (91.8 ± 37.6 µg/cm³ vs. 84.8 ± 34.8 µg/cm³, p = 0.6657, Fig. 2 C,D). The remaining four patients (8.7%) exhibited high protein residues across both regions. In catheter-predominant deposition group (n=18), these patients demonstrated significantly higher protein accumulation in the catheter compared to the injection chamber (295.5 ± 72.8 µg/cm³ vs. 44.4 ± 16.4 µg/cm³,p < 0.0001; Figs. 2E,F). In chamber-predominant deposition group (n=17), the cohort exhibited a sharp decline in the protein concentration gradient from the port to the catheter, with significantly higher accumulation concentrated in the chamber (915.4 ± 438.7 µg/cm³ vs. 40.4 ± 11.4 µg/cm³, p < 0.0001; Figs. 2G and 2H).

Fig. 2.

Protein residues and distribution patterns in the injection chamber and catheter. A. Protein residues normalized to volume in injection chambers and catheters (n = 46; µg/cm³). B. Variation trends in protein residue distribution between injection chamber and catheter for the entire cohort (n = 46). C. Protein residues in patients with comparable distribution (n = 11; µg/cm³). D. Variation trends between chamber and catheter in patients with comparable distribution (n = 11). E. Protein residues in patients with catheter-predominant deposition (n = 18; µg/cm³). F. Variation trends between chamber and catheter in patients with catheter-predominant deposition (n = 18). G. Protein residues in patients with catheter-predominant deposition (n = 17; µg/cm³). H. Variation trends between chamber and catheter in patients with catheter-predominant deposition (n = 17).

Further analysis across implantation durations showed no significant difference in total protein residue between the two groups (360.8 ± 179.2 µg/cm³ vs. 199.6 ± 70.4µg/cm³, p=0.6687, Fig. 3A). To examine whether the duration of port implantation affected protein deposition, patients were divided into two groups: 0.1–5 years and 5.1–10.2 years. In the shorter-duration group, the port chamber exhibited significantly higher protein levels than the catheter (490.3 ± 264.2 µg/cm³ vs. 102.2 ± 27.5 µg/cm³, p = 0.0379, Fig. 3B). However, in the longer-duration group, protein levels were comparable between the chamber and catheter (185.4 ± 100.1 µg/cm³ vs. 233.9 ± 77.2 µg/cm³, p =0.0851) This indicates that port abnormalities are more likely to originate in the chamber during the early implantation period, while the catheter becomes more affected over time.

Fig. 3.

Total protein deposition in implanted ports and segmental distribution within catheters across different implantation periods.A. Total protein deposition in implanted ports (sum of injection chamber and catheter deposits) across different implantation time windows, normalized to volume (μg/cm³). B. Protein residues normalized to volume in injection chambers and catheters across different implantation time windows (μg/cm³); Protein residues in patients with short utilization period, ie. 0.1 to 5 years, had more protein residues. (p=0.0379). C. Protein residues normalized to volume across catheter segments (proximal, middle, and distal) (μg/cm³). Protein residues in distal and proximal segments was 287.1 ± 65.8 µg/cm³ and 55.9 ± 15.4 µg/cm³, respectively. (p=0.0006) Protein residues in distal and middle segments was 287.1 ± 65.8 µg/cm³ and 109.8 ± 31.3 µg/cm³, respectively. (p=0.0114). D. Protein residues in catheter segments (proximal, middle, distal), normalized to volume, across different implantation time windows (μg/cm³).

Further dissection of the catheter into proximal, middle, and distal sections showed that the distal section had the highest protein residue levels (287.1 ± 65.8 µg/cm³), followed by the middle (109.8 ± 31.3 µg/cm³) and proximal (55.9 ± 15.4 µg/cm³) sections, suggesting that the distal portion is the most prone to residue buildup and potential malfunction. (Fig. 3C) Furthermore, in both duration groups, the distal catheter section consistently showed the highest protein levels, suggesting that protein accumulation progresses distally over time. In the 0.1–5 year group, the distal section had significantly higher residue than the middle and proximal regions (215.3 ± 73.0 µg/cm³ vs. 58.9 ± 18.5 µg/cm³ vs. 32.9 ± 10.2 µg/cm³, p = 0.0387 and p= 0.0026, respectively, Fig. 3D). Similarly, in the 5.1–10.2 year group, the distal section again had the highest levels and showed significantly higher than proximal third (409.7 ± 124.5 µg/cm³ vs. 95.3 ± 36.6 µg/cm3, p=0.0094) and further confirming this distal accumulation trend.

To determine if the number of punctures affected protein buildup, patients were categorized based on 1–12 and 13–33 puncture events. No significant difference was observed in total protein residues between the groups (331.0 ± 146.4 µg/cm³ vs. 193.9 ± 76.9 µg/cm³, p= 0.7038, Fig. 4A). Nevertheless, the distal catheter section continued to show the highest protein levels in both groups. In the 1–12 puncture group, the distal section had significantly more protein than the middle and proximal sections (261.2 ± 69.1 µg/cm³ vs. 82.0 ± 21.2 µg/cm³, p =0.0004; 82.0 ± 21.2 vs. 47.3 ± 14.8 µg/cm³, p= 0.016, Fig. 4B). In the 13–33 puncture group, the same pattern was observed, although the difference was not statistically significant. Overall, more frequent punctures were associated with a general trend toward increased protein residue, particularly at the distal end of the catheter.

Fig. 4.

Total protein deposition in implanted ports and segmental distribution within catheters across different numbers of punctures. A. Total protein deposition in implanted ports (sum of injection chamber and catheter deposits) across two chemo-puncture frequency groups (1–12 vs. 13–33 punctures), normalized to volume (μg/cm³). B. Protein residues in catheter segments (proximal, middle, distal), normalized to volume, across the same puncture frequency groups (1–12 vs. 13–33 punctures), μg/cm³.

As previously described, 56.5% of patients had visible intraluminal fibrin deposits. To further investigate, we conducted a qualitative analysis to identify the protein components. Only 16 patients had sufficient protein material for molecular analysis of the residues. Thrombogenic molecules—including P-selectin, thrombin, fibrinogen, and fibrin—were sporadically detected. Interestingly, β-actin, which was used as a control marker for cellular cytoskeleton, was absent in cases 8, 9, and 13. This finding suggests that the residues were composed primarily of blood-soluble proteins rather than cellular components (Fig. 5).

Fig. 5.

Western blot analysis of implanted port deposits proteins extracted from port deposits were analyzed by western blot with (35) μg of total protein loaded per lane.

Discussion

Port malfunction is frequently associated with intraluminal obstruction resulting from the accumulation of blood components, protein aggregates, or medication precipitates^26,27. Such obstructions can compromise vascular access and potentially delay critical oncological interventions, often necessitating revision surgery for device replacement. Therefore, a deeper characterization of the nature and distribution of these deposits is essential for improving maintenance strategies and preventing avoidable complications. While port complications are often attributed to operator experience^9,10,24, catheter tip localization²⁸, and patient-specific clinical characteristics²⁹, recent evidence suggests that these outcomes correlate more strongly with the surgical implantation method than with the choice of entry vessel.²³ Collectively, these factors may lead to suboptimal device positioning, which subsequently impairs the efficacy of standard maintenance protocols and predisposes the system to intraluminal failure. In this study, we conducted a comprehensive evaluation of proteinaceous residues in explanted intravenous ports, including devices removed for reasons other than the completion of chemotherapy to capture a diverse range of clinical presentations. Given that the intraluminal space is accessible exclusively via non-coring needle puncture—provided structural integrity is maintained—we hypothesized that irrigation efficacy remains independent of external factors such as migration or impaired wound healing. Notably, we included one case of catheter fracture where the device demonstrated a patent blood withdrawal test and normal injection functionality, yet the patient reported pain during administration. A tiny perforation was identified at the junction between the locking nut and the proximal catheter segment; however, this was deemed to have a negligible impact on overall irrigation efficacy, and the device was retained for analysis. In this study, we conducted a comprehensive evaluation of proteinaceous residues in explanted intravenous ports. Under experimental conditions, a central puncture of the silicone diaphragm with proper needle alignment and adequate irrigation volume is reported to generate a double-vortex flow pattern within the injection chamber, facilitating residue clearance^15,16. However, our findings suggest that these idealized conditions may not be consistently achieved in routine clinical practice, leading to the observed patterns of protein deposition.

This study incorporated both quantitative and qualitative analyses to assess protein deposition in implanted ports. Quantitative analysis focused on protein extraction and effectively recovered most deposited proteins, leaving only minimal residual material that did not affect measurement accuracy (Supplementary Figure 3A, B). Residual protein retention was could be associated with surface smoothness, as the inner surfaces of implanted ports exhibited a mean surface roughness (Ra) of approximately 10 μm, which may allow proteins to be retained within microscopic surface irregularities. Marked interpatient variability in protein deposition was observed, potentially attributable to difference in irrigation efficacy following administration of protein-containing medications, cumulative effects of suboptimal catheter maintenance, or variability in nursing practices. Qualitative analysis was limited to samples with sufficient protein yield (≥35 μg per well), resulting in eligibility for 16 patients (34.8%). Although samples with minimal deposition were excluded, even small protein residues may promote bacterial adhesion and biofilm formation, whereas samples with substantial deposition may provide insight into strategies for optimizing catheter maintenance.

An implanted port is a closed system comprising an injection chamber and a catheter. The intraluminal irrigation flow pattern consists of a double-vortex flow within the injection chamber, followed by sequential clearance along the catheter with laminar flow.^14,15 Effective irrigation volume is approximately 20 times the intraluminal volume of an implanted port in ex vivo models¹⁶. Optimal needle positioning may facilitate proximal-to-distal protein removal, whereas suboptimal orientation may reduce irrigation efficiency and increase residual protein concentration. The observed variation in protein residue between the injection chamber and the catheter appears to reflect differences in puncture technique and irrigation method. From a fluid dynamics perspective, only central puncture directly opposite the chamber outlet is through to minimize the formation of recirculation zones where protein accumulation typically occurs (Supplementary Figure 1E). Furthermore, continuous irrigation with normal saline under ideal orientation may generate sufficient vortex strength and shear forces to effectively eliminate residues. As puncture orientation deviates from the ideal position, the recirculation zone appears to enlarge proportionally, and may decrease vortex strength accordingly. In this study, only seven patients (15.2%) exhibited low protein residues in both the injection chamber and catheter (Fig. 2C), suggesting ideal irrigation. In contrast, four patients (8.7%) exhibited high protein levels in both regions, likely reflecting suboptimal irrigation techniques. The push–pause flushing method, commonly used in clinical practice, may fail to sustain a strong double-vortex flow, thereby impairing longitudinal clearance of intraluminal protein (Fig. 2D). Slight deviations in needle orientation (39.1%) resulted in increased chamber recirculation and higher protein accumulation within the catheter (Figs. 2E, 2F), whereas marked deviations (34.7%) produced expanded recirculation zones and pronounced protein accumulation within the chamber (Figs. 2G, 2H). These findings highlight the critical role of puncture technique and irrigation volume in determining intraluminal flow dynamics and protein residue distribution.

The linear geometry of the catheter supports laminar flow, which is generally associated with more consistent flushing. (Supplementary Figure 1F). Within the catheter, a distinct gradient of protein deposition was observed, with the highest concentration in the distal third, followed by the middle and proximal segments. This gradient likely reflects both a dilution effect during irrigation and retrograde blood reflux during needle withdrawal. The vacuum created by needle removal and rebound of the silicone diaphragm may draw blood into the catheter tip, promoting distal clot formation¹⁷. This deposition pattern was consistent across patient subgroups with different implantation durations and puncture frequencies (Figs. 3D and 4B). From the perspective of the implantation period, no statistically significant difference in total protein accumulation was observed between shorter and longer-duration devices. (p=0.6687) Notably, a significant predominance of chamber deposition was found exclusively in patients whose ports had been implanted for less than five years (p =0.0379; Fig. 3A). This localized accumulation in the early years may reflect the high-intensity utilization of the device during the active phases of oncological treatment, where frequent access and high-volume infusions are most common. Although all ports were functional at the time of removal, more than half (56.5%) demonstrated visible fibrin deposits. This finding suggests that current maintenance protocols may prevent overt clinical malfunction but may be insufficient to fully eliminate intraluminal deposit. For further characterization, both intraluminal fluids and solid deposits were recovered for analysis. Due to protein consumption during quantitative testing and the requirement for ≥35 μg of protein per well for qualitative analysis, only 16 patients (34.8%) were eligible for Western blot analysis. Thrombogenic proteins—including P-selectin³⁰, thrombin, fibrinogen, and fibrin^7,31—were sporadically detected, supporting the role of thrombosis in residue formation. The presence of thrombosis-related proteins may suggest that standard heparin lock solutions are insufficient to fully prevent the formation of intraluminal residues. This phenomenon may be attributable to suboptimal irrigation, as previously discussed, or to irrigation intervals that exceed the pharmacological half-life of heparin¹⁸. Although a growing body of literature suggests that normal saline is as effective as heparin in preventing catheter occlusion^32,33, further investigation into these dynamics is warranted. Interestingly, the absence of β--actin—a marker of the cellular cytoskeleton³⁴—in several samples is consistent with the hypothesis that deposits are primarily composed of blood-soluble proteins rather than cellular debris (Fig. 5). This finding may indicate that non-thrombotic mechanisms also contribute to protein accumulation. Given that late complications, such as malfunction and infection, are frequently associated with intraluminal deposits^35,36, these qualitative findings may offer insight into the pathophysiological mechanisms of port failure and could serve as a reference for optimizing maintenance and medication strategies

This study has several limitations. The study was conducted at a single center with a relatively small cohort, which may limit generalizability. Only a subset of ports yielded sufficient protein for molecular characterization, and tightly adherent deposits may not have been fully recovered due to surface smoothness. In addition, flow dynamics and needle orientation were inferred from residue distribution patterns rather than measured in real time. Despite these limitations, the findings have important implications for nursing practice. Adherence to current flushing guidelines alone may be insufficient to prevent subclinical residue accumulation. Operator-dependent factors—including puncture angle, flushing technique, and irrigation volume—play a critical role in maintenance outcomes. These findings suggest that implementing standardized nursing protocols and simulation-based training may provide a framework for reinforcing optimal flushing practices.⁶. Simulation training and competency-based assessment may enhance consistency in port access and irrigation techniques. Revising nursing protocols to incorporate port geometry, flushing volume, and needle orientation may help standardize maintenance practices and reduce subclinical residue accumulation. Collectively, these improvements may enhance the safety, functionality, and longevity of implanted ports, ultimately improving outcomes for patients requiring long-term intravenous access.

Conclusion

This study suggests that proteinaceous deposits may persist in implanted intravenous ports despite compliance with current maintenance protocols. These deposits—primarily composed of blood-soluble proteins—appear to accumulate most heavily in areas with suboptimal flow, particularly the injection chamber base and distal catheter segment. The results indicate that the mechanical efficacy of irrigation, could be a primary determinant of intraluminal residue clearance, potentially more so than pharmacologic flushing alone. From a nursing perspective, these findings point toward a potential need to re-evaluate port maintenance strategies. Consideration could be given to prioritizing needle orientation, assessing irrigation volumes, and implementing evidence-based simulation training. Integrating flow dynamics into nursing education and refining flushing protocols to account for procedural variability may help support long-term port function and patient safety. Furthermore, while thrombosis contributes to residue formation, the prevalence of blood-soluble proteins suggests that non-thrombotic mechanisms merit further investigation to inform more comprehensive maintenance strategies.

Author contributions

Conceptualization: Ching-Yang Wu and Jui-Ying Fu. Methodology: Yen Chu, Elsie Khai-Woon Toh, and Ya-Ying Hsu. Validation and data curation: Ching Feng Wu and Po-Chun Chang. Formal analysis: Jui-Ying Fu, Yen Chu, and Ching-Yang Wu. Investigation: Elsie Khai-Woon Toh and Ya-Ying Hsu. Writing – original draft: Jui-Ying Fu and Ching-Yang Wu. Writing – review and editing: Yen Chu, Po-Chun Chang, and Ching Feng Wu. Funding acquisition and supervision: Ching-Yang Wu. All authors read and approved the final manuscript.

Funding

Chang Gung Memorial Hospital,Linkou,CMRPG3M1281,CMRPG5G0131.

Data availability

The datasets generated and analyzed during the current study are not publicly available because they contain sensitive personal medical information. In accordance with Taiwan regulations, the data may only be shared with permission from the Institutional Review Board of Chang Gung Memorial Hospital. Qualified researchers may request access to the de-identified dataset by submitting an IRB-approved proposal to the Chang Gung Memorial Hospital IRB.

Declarations

Competing interests

The authors declare no competing interests.