Hemostatic Agent Injection for Treating Tract Bleeding from Tunneled Central Venous Catheter Insertion

Abstract

Purpose:

Tract hemorrhage following tunneled central venous catheter placement can be a challenging issue for some patients. We report our experience with the injection of a hemostatic agent into the central venous catheter tract for treatment of tract bleeding.

Materials and Methods:

The patient's central venous catheter exit site and surrounding area were cleaned with surgical preparation liquid and draped using an aseptic technique. The tract was infiltrated with 1% lignocaine. A hemostatic agent was injected along the tract alongside the central venous catheter using an 18G x1.5” blunt needle (Agani, Terumo, Japan) to fill the potential space around the catheter. The site was then re-dressed in a sterile manner, and an antimicrobial dressing was placed around the catheter exit site.

Results:

A total of 21 patients were treated. The technical success rate was 100%. The clinical success rate following the initial procedure was 66.6%, improving to 95% and 100% after one and two re-interventions respectively.

Conclusions:

This described technique of central venous catheter tract injection with hemostatic material assists in achieving tract hemostasis.

Introduction

Tract bleeding is a possible minor complication after insertion of a tunneled central venous catheter (CVC) due to tunnel tract trauma [1-3]. Incidence is not regularly reported and ranges widely in the literature from 0% to 42% [4-6]. Recent studies advocate for continuing anticoagulant and antiplatelet medications peri-procedurally due to the low incidence of clinically significant bleeding, competing thrombotic risks, and administrative challenges in withholding medications [7-9].

Tract bleeding causes patient distress, complicates nursing care, and repeated attempts at hemostasis increase infection risk [3, 10, 11]. Anecdotal approaches described in the literature include compression with gauze soaked with a hemostatic agent, injection of adrenaline around the tract, insertion of additional sutures, and injection of the tract with hemostatic agents [6, 10, 12-15].

This study presents a technique of injection of a hemostatic agent within the CVC tract alongside the catheter tubing to assist with hemostasis in patients who have undergone CVC insertion.

Materials and Methods

Ethics Committee approval was granted. This study was reported according to the STROBE guidelines. Patients were identified through keyword searches of the Electronic Medical Record and Radiology Information System. Patients were included in the study if they underwent hemostatic agent injection for tract bleeding following tunneled CVC insertion between July 1, 2019, and September 1, 2024.

Data collected included patient age, sex, medical comorbidities including bleeding diatheses, catheter type, indication for CVC insertion, date of insertion, most recent platelet count, activated partial thromboplastin time (APTT), international normalized ratio (INR), and anticoagulation or antiplatelet therapy before the tract injection.

Procedural data collected included the date of injection, hemostatic agent type, volume used, preceding hemostatic methods attempted, technical success, clinical success, re-interventions, and adverse events during the 3-month follow-up. Technical success was defined as visible hemostasis at the time of intervention. Clinical success was defined as ongoing resolution of tract bleeding during the three-month follow-up. Adverse events were defined in accordance with the Cardiovascular and Interventional Radiological Society of Europe guidelines [16].

Procedural technique

All dressings were removed from the CVC exit site. The exit site and surrounding area were cleaned with chlorhexidine gluconate 2% in isopropyl alcohol 70% or povidone-iodine 10%, and special care was taken to ensure that the catheter exit site from the skin on the chest wall was cleaned thoroughly. The skin site was allowed to dry and then draped with sterile drapes.

Using an aseptic technique, the tract, which was defined as the space between the CVC tubing and subcutaneous tissues of the chest, was infiltrated with 1% lignocaine local anesthetic to facilitate patient comfort. The hemostatic agent consisted of either a slurry of gelatin sponge (Gelfoam [Pfizer Inc., New York, NY, USA] or Spongostan [Ethicon Inc., Raritan, NJ, USA]) or 5,000 units recombinant thrombin (Recothrom [Baxter Healthcare Corporation, IL, USA]) in 4-6 mL of 0.9% sodium chloride solution. The hemostatic agent was drawn up into a 10-mL syringe and then injected within the tract from the skin entry site using an 18G x1.5” blunt needle (Agani, Terumo, Japan) to fill the potential space around the catheter (Figure 1a and b). The blunt needle was placed as deep as possible into the tract, and while slowly injecting material into the tract, the syringe and needle were simultaneously and slowly withdrawn with the aim of filling the entire tract space with hemostatic material. This process was repeated on both sides of the tract (Figure 1a and b). Reflux of hemostatic material out of the skin entry site was used as an indicator of the injection endpoint. The site was then re-dressed in a sterile manner, and an antimicrobial dressing (Biopatch, Johnson & Johnson, NJ, USA) was placed around the catheter exit site. A Tegaderm dressing (3M, MN, USA) was placed over the skin entry site.

Figure 1.

a – Injection of hemostatic agent along the side of the CVC line.

b – Injection of hemostatic agent along the side of the CVC line.

CVC: central venous catheter

Results

A total of 21 patients underwent hemostatic agent injection for tract bleeding. Patient demographics are shown in Table 1.

Table 1.

Patient Demographics.

Patient Characteristics	Patients (n = 21)
Age (years)
Mean	61.7
Median (Range)	64 (27–85)
Sex
Male	20 (95%)
Female	1 (5%)
Catheter Type
Permacath	17 (81%)
Hickman	4 (19%)
Indication
Dialysis	16 (76%)
Chemotherapy	4 (19%)
Plasma exchange	1 (5%)
Comorbidities
Acute renal failure	6 (29%)
Chronic renal failure	4 (19%)
End-stage renal failure	7 (33%)
Ischaemic heart disease	7 (33%)
Haematological disorders
Waldenstrom’s macroglobulinaemia	1 (5%)
Multiple myeloma	1 (5%)
Myelofibrosis	1 (5%)
Coagulation abnormalities
Platelet count <50	3 (14%)
APTT >1.5× ULN	2 (10%)
INR >1.5	1 (5%)
Anticoagulation/Antiplatelets
None	3 (14%)
Aspirin	2 (10%)
Clopidogrel	1 (5%)
Prophylactic anticoagulation	3 (14%)
Aspirin + prophylactic anticoagulation	7 (33%)
Therapeutic anticoagulation	5 (24%)

APTT: activated partial thromboplastin time; INR: international normalized ratio; ULN: upper limit of normal

In total, 20 out of 21 patients (95%) were male, with 1 female patient. The age range was 27-85 years (mean, 61.7 years; median, 64 years). Seventeen out of 21 CVCs (81%) were 14.5 Fr dual-lumen Permacaths (Palindrome Precision Chronic Catheters, Covidien, Ireland), with the remaining 4 of 21 (19%) being 12 Fr dual-lumen Hickman catheters (Bard Access Systems Inc., Utah, USA). The most common indication for insertion was dialysis (16 of 21), followed by chemotherapy (4 of 21) and plasma exchange (1 of 21).

The most common comorbidities were renal failure, whether acute (6 of 21), chronic (4 of 21), or end-stage renal failure (7 of 21), as well as ischemic heart disease (7 of 21). Three out of 21 patients had hematological disorders that may increase their bleeding risk―namely, Waldenstrom's macroglobulinemia, multiple myeloma, and myelofibrosis. Three out of 21 patients had a platelet count of <50, 2 patients had an APTT >1.5× the upper limit of normal, and 1 patient had an INR >1.5.

Three patients were not on any anticoagulation or antiplatelet therapy, as these were withheld due to thrombocytopenia. Two patients were on aspirin alone, and one patient was on clopidogrel alone. Fifteen out of 21 patients were on anticoagulation, with 5 of them on therapeutic dosage and the remaining 10 on prophylactic dosage. Out of the 10, 7 were on aspirin and prophylactic enoxaparin. One patient had their therapeutic heparin infusion continued periprocedurally, whereas two patients had their infusions paused four hours prior, and two patients on therapeutic enoxaparin received their last dose more than 12 hours prior.

Procedural results

The technical success rate was 100% (21 of 21), with all procedures achieving visible hemostasis. The clinical success rate was 66.6% (14 of 21), improving to 95% (20 of 21) and 100% (21 of 21) after one and two re-interventions, respectively. Procedural characteristics are shown in Table 2.

Table 2.

Procedure Characteristics.

Procedure Characteristic	Patients (n = 21)
Technical success	21 (100%)
Clinical success
After the initial injection	14 (66.6%)
After one re-intervention	20 (95%)
After two re-interventions	21 (100%)
The preceding hemostatic method attempted
Compression (gauze + hemostatic agent)	5 (24%)
Hemostatic sutures	4 (19%)
Lidocaine/adrenaline injection	1 (5%)
Time from catheter insertion to hemostatic injection in days
0	7 (33%)
1	2 (10%)
2-4	6 (29%)
5-7	5 (24%)
>7	1 (5%)
Hemostatic agent used for index injection
Gelfoam	19 (90%)
Spongostan	2 (10%)
Hemostatic agent used for repeated injection	Repeat procedures (n = 7)
Gelfoam	4 (57%)
Spongostan	1 (14%)
Recothrom	2 (29%)

Gelfoam (Pfizer Inc., New York, NY, USA), Spongostan (Ethicon Inc., Raritan, NJ, USA), Recothrom (Baxter Healthcare Corporation, IL, USA).

In terms of time to injection from CVC insertion, 6 of 21 patients underwent tract injection immediately after CVC insertion due to periprocedural bleeding. One patient underwent index tract injection on day 0, 8 patients by day 4, 5 patients by day 7, and one patient on day 8.

Six patients required one re-intervention, and one patient required two additional interventions. Five patients underwent secondary tract injection. One patient was managed with compression using adrenaline-soaked gauze and hemostatic sutures. The last patient required two repeat tract injections.

Regarding the preceding hemostatic methods attempted, compression with gauze soaked in tranexamic acid or adrenaline was attempted in 5 of 21 patients. Hemostatic sutures were attempted in 4 of 21 patients, and injection with lidocaine/adrenaline was attempted in one patient.

Gelfoam was used in 19 of 21 index injections, with Spongostan used in 2 of 21. For repeat tract injections, gelatin sponge was used in 5 of 7 procedures, and recombinant thrombin was used in 2 of 7 procedures.

No adverse events directly linked to hemostatic tract injection were recorded. There were no infections deemed directly related to the tract injection. Three CVCs were removed after completion of treatment, between 22 and 28 days. Two patients died during follow-up, from heart failure and end-stage renal failure.

Discussion

This study demonstrates hemostatic tract injection as a potentially effective method to control bleeding from tunneled CVC insertion, with a clinical success rate of 66.6% after one intervention, rising to 100% after one or two re-treatments. It also demonstrates its short-term safety, with no significant adverse events related to tract injection identified during the three-month follow-up.

In the literature to date, anecdotal descriptions of hemostatic agent injection have been reported in several reviews [1, 3], as well as one case report describing the use of microfibrillar collagen paste to address tract bleeding [15]. The use of hemostatic agents to seal or embolize other tracts is documented in other surgical sites, such as preventing pericatheter dialysate leak in peritoneal dialysis catheters [17], or addressing tract bleeding during organ biopsy [18-21]. Similarly, these studies demonstrate high technical success rates with minimal complications.

Notably, approximately one-third of patients required re-intervention, all of whom were either receiving anticoagulation or antiplatelet agents, or were thrombocytopenic. The patient who required two repeat injections was in acute renal failure and was therapeutically anticoagulated to support extracorporeal membrane oxygenation.

It should be noted that multiple techniques are commonly used in real-world practice in order to stop hemorrhage. This technique of injecting hemostatic agents could be helpful in cases where simple manual compression fails to stop hemorrhage. It could also be useful in patients who have a coagulopathy, are on antiplatelet agents or anticoagulation, or have thrombocytopenia. Patients with these increased risks of hemorrhage are thought to be more likely to experience treatment failure or require multiple injections, as seen in the case of the patient who was on therapeutic anticoagulation and required two repeat injections to achieve hemostasis.

Gelatin-based hemostatic agents were chosen as first-line therapies due to their ease of access and simplicity of use. Their use also allowed the entire tract cavity to be filled by adjusting the injection volumes. In cases where the initial gelatin-based agents failed to achieve hemostasis, consideration was given to the use of recombinant thrombin for patients who were coagulopathic, on anticoagulation, or on antiplatelet agents to help achieve hemostasis. Sutures were also used if the space between the catheter exit site on the chest and the catheter tubing was considered significant (e.g., >1 cm). This technique outlines the current institution's approach to managing ongoing tract hemorrhage.

It should be noted that, in the authors' opinion, patients with active tract site infections would not be recommended for this technique due to the risk of infection seeding the injectate. There is always some theoretical risk in injecting foreign material into a wound, as this may introduce bacteria that can lead to line colonization and/or sepsis. Thorough site cleaning with chlorhexidine gluconate 2% in isopropyl alcohol 70% or povidone-iodine 10%, as well as use of an aseptic technique, is required to reduce this risk. Larger studies would be necessary to further examine this.

The limitations of this study include its retrospective single-center design, small cohort size, lack of a control group, non-uniform treatment approach, and absence of a prespecified follow-up protocol. Variability in the hemostatic agents used introduced heterogeneity to the treatment approach. Delayed complications beyond three months may also be examined in future studies. The findings of this study could be further evaluated in a larger, prospective study with standardized injection agents and protocols, as well as a comparison between individual techniques (such as hemostatic sutures, prolonged compression, or adrenaline injection). Larger studies may also identify trends in cases with failure of hemostasis.

Conclusion

This study describes a technique involving hemostatic agent injection to assist in the treatment of tract bleeding from tunneled CVC insertion. This technique could be considered by interventional radiologists encountering bleeding from tunneled CVC insertion, particularly in patients who are anticoagulated, on antiplatelet therapy, or thrombocytopenic.

Author Contributions

All authors contributed to the study research and manuscript preparation and fulfil the criteria for authorship in accordance with ICMJE guidelines.

Conflict of Interest

There are no conflicts of interest.

IRB Approval

This study has obtained IRB approval from the Alfred Health ethics committee (Ref: 773/24), and the need for informed consent was waived.

Ethical Statement

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Disclaimer

Dr. Gerard s Goh, an author of this paper and a member of the Editorial Board of this journal, was not involved in the peer-review or editorial decision-making process.