Central Venous Catheter‐Related Extravasation Due to Delayed Vascular Erosion

ABSTRACT

Central venous catheters (CVCs) are crucial in critical care but may occasionally lead to rare, serious complications such as vascular erosion and extravasation of infused solutions. Leakage of total parenteral nutrition (TPN) into the pleural space can mimic chylothorax, potentially delaying correct diagnosis and appropriate treatment. An 83‐year‐old male with COVID‐19‐related acute respiratory distress syndrome (ARDS) received TPN via a left‐sided CVC, which remained in a suboptimal position for 18 days. On day 18, he developed sudden respiratory deterioration associated with bilateral milky pleural effusions. Initially misdiagnosed as chylothorax, subsequent pleural fluid analysis revealed elevated glucose and triglyceride levels. Computed tomography demonstrated vascular erosion near the junction of the brachiocephalic vein and superior vena cava. The CVC was subsequently removed, enteral nutrition initiated, and the pleural effusions resolved. However, the patient eventually succumbed to progressive ARDS. This case highlights the risk of delayed vascular erosion associated with suboptimally positioned CVCs, highlighting the diagnostic difficulty posed by TPN extravasation mimicking chylothorax. Prompt recognition of this complication is crucial to prevent adverse outcomes.

Key Clinical Message

Delayed vascular erosion or total parenteral nutrition extravasation from a central venous catheter can mimic chylothorax, presenting as milky pleural effusion. Prevent this rare catastrophe by confirming catheter tip position in real time. Use ultrasound bubble testing or electrocardiogram‐guided placement, and ensure the tip is positioned correctly.

1. Introduction

Central venous catheters (CVCs) are indispensable tools in the management of critically ill patients, providing reliable vascular access for administering medications, total parenteral nutrition (TPN), and hemodynamic monitoring [1, 2]. Despite widespread utilization, CVCs are associated with various complications, including infections, thrombosis, and less commonly, mechanical issues such as malpositioning or vascular erosion [3]. Among these, delayed vascular erosion resulting in extravasation of infused solutions is rare but potentially life‐threatening and may remain unrecognized until significant clinical deterioration occurs [4, 5].

One particular diagnostic challenge occurs when extravasated TPN accumulates in the pleural space, radiographically and grossly mimicking chylothorax [6, 7]. Although the milky pleural effusion appearance and elevated triglyceride (TG) levels typically suggest chylothorax, biochemical analyses may be misleading when the effusion primarily comprises hyperosmolar infused solutions such as TPN or lipid‐soluble sedatives [8]. Misinterpretation under these circumstances can delay appropriate management and contribute to worse clinical outcomes.

Here, we report a fatal case involving delayed extravasation of TPN into the pleural space due to vascular erosion caused by a left‐sided CVC positioned suboptimally. The case was initially misdiagnosed as chylothorax, leading to delayed initiation of enteral feeding and prolonged reliance on the CVC. This report highlights essential diagnostic considerations, underlying pathophysiological mechanisms, and preventive strategies to mitigate such complications in clinical practice.

2. Case History/Examination

An 83‐year‐old man presented to the emergency department with dyspnea secondary to a confirmed COVID‐19 infection. Due to progressive desaturation and increasing oxygen requirements following admission, emergent endotracheal intubation was performed. A CVC (Prime‐S, 7 Fr, Triple lumen, polyurethane 16 cm; SW Medical, Seoul, Korea) was inserted through the left internal jugular vein in the emergency department for critical care management. A postprocedural chest radiograph confirmed suboptimal yet acceptable catheter positioning (Figure 1).

FIGURE 1.

Initial chest radiograph following central venous catheter (CVC) insertion and endotracheal intubation (arrowhead indicates the CVC tip located at the brachiocephalic vein–superior vena cava junction).

3. Differential Diagnosis, Investigations, and Treatment

The patient was managed for acute respiratory distress syndrome (ARDS), experiencing a fluctuating clinical course over the following 2 weeks. TPN was started on Day 2 with a commercially available three‐compartment admixture (OLIMEL N12E; Baxter). Each 1000 mL (after reconstitution) contained amino acids 75.9 g (nitrogen 12.0 g), glucose 73.3 g, and lipids 35.0 g (≈950 kcal). Chest computed tomography (CT) performed on hospital Day 11 demonstrated bilateral pulmonary infiltrates consistent with ARDS, without any additional abnormalities.

Following clinical improvement, ventilator weaning was initiated under sedation with continuous intravenous propofol. However, on hospital Day 18, a sudden increase in oxygen requirement occurred, and a chest radiograph revealed a massive right‐sided pleural effusion. Ventilator weaning was halted, and chylothorax was suspected (Figure 2). The patient was referred to the cardiothoracic surgery department, where a closed thoracostomy was performed. Approximately 2 L of milky fluid was drained from the right pleural space within the first 24 h, with persistent drainage thereafter.

FIGURE 2.

Chest radiograph showing bilateral pleural effusion on the 18th hospital day (arrowhead indicates the central venous catheter placed suboptimally).

A chest CT scan performed immediately afterward showed bilateral pleural effusions and a newly developed air‐containing artifact near the junction of the brachiocephalic vein and superior vena cava (SVC), which had not been present on prior imaging, suggesting extravasation of fluid by vascular erosion (Figure 3). Subsequently, a left‐sided thoracostomy was also performed, yielding approximately 1 L of similarly milky fluid. On the following day, the initially placed CVC was removed, and a peripherally inserted central catheter was placed through the right basilic vein, with its tip positioned in the SVC (Figure 3).

FIGURE 3.

Interval follow‐up chest computed tomography (CT) performed to evaluate massive bilateral pleural effusions. (A) Axial image (arrow indicates the central venous catheter tip located at the brachiocephalic vein–superior vena cava junction; asterisk indicates the chest tube in the right pleural space). (B) Coronal image (arrowheads denote an air artifact surrounding the vessel wall, suggestive of vascular erosion).

TPN and lipid‐soluble sedation (propofol) were discontinued, and enteral feeding was initiated via a Levine tube. Pleural fluid analysis demonstrated a glucose level of 407 mg/dL, a total protein level of 1.7 g/dL, a TG level of 687 mg/dL, a total cholesterol level of 6 mg/dL, and a lactate dehydrogenase level of 168 IU/L, findings not consistent with a diagnosis of chylothorax. We attempted to analyze the composition of the TPN fluid for comparison with the pleural fluid; however, the laboratory was unable to perform the requested analysis due to the fluid's high viscosity.

4. Outcome

Following CVC removal, no further chylous effusions were observed. Bilateral chest tubes were removed once drainage had subsided. Despite initial stabilization, the patient ultimately succumbed to progressive ARDS.

5. Discussion

Although vascular erosion is an uncommon complication associated with central venous catheterization, its reported incidence is approximately 0.17%. Despite its rarity, this complication carries a significant mortality risk. A study by Duntley et al. reported a mortality rate of 12.5% directly attributable to vascular erosion, with an additional 20%–74% of patients succumbing to secondary complications [5, 9, 10].

Symptoms caused by CVC‐related vascular erosion typically appear within an average of 3.6 days postinsertion. Delayed complications occurring after more than 7 days of catheter placement are exceedingly rare. To date, the longest duration reported in the literature before vascular erosion onset was 11 days following CVC insertion [9]. Although it is generally recommended that CVCs be replaced within approximately 2 weeks, the catheter in this case was overlooked and remained in place for 19 days [11]. Additionally, the CVC had initially been placed in a suboptimal position and was never corrected, thus increasing the risk of catheter‐related complications. Given the higher incidence of vascular erosion associated with left‐sided CVCs, right‐sided access should be preferred whenever feasible [8]. Furthermore, meticulous assessment of catheter positioning during postprocedural chest radiographs is crucial for catheters placed on the left side [10]. Female sex has been proposed as a potential risk factor for vascular erosion, possibly related to smaller vessel caliber; however, this association was not statistically significant in the largest cohort (p = 0.18), whereas older age was significantly associated with erosion (p = 0.009) [9, 10]. Prior studies have reported mortality rates of up to 20% in association with this complication [10].

Despite meeting the criteria for early EN as recommended by the European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines, the patient remained on PN [12, 13]. Early enteral nutrition should be initiated as soon as it is feasible, and maintaining a patient who could have received enteral feeding on prolonged parenteral nutrition likely became the initial trigger for this complication [13].

Biochemical analysis of pleural fluid can help distinguish chylothorax from TPN‐related extravasation. Chylothorax is supported by a pleural fluid TG concentration ≥ 110 mg/dL and/or demonstration of chylomicrons (definitive), typically with pleural cholesterol < 200 mg/dL and a cholesterol‐to‐TG ratio < 1; when results are borderline, a pleural/serum TG ratio > 1 together with a pleural/serum cholesterol ratio < 1 further supports chylothorax. Therefore, even in patients with high pleural fluid glucose—as in our case, where TPN extravasation is a concern—the lipid profile remains a key diagnostic clue, and it can also aid in differentiating pseudochylothorax, which more often shows elevated pleural cholesterol [14, 15]. Accordingly, we propose a practical diagnostic algorithm to differentiate chylothorax, pseudochylothorax, and TPN‐related extravasation in settings where chylomicron testing is unavailable (Figure 4).

FIGURE 4.

A diagnostic algorithm for milky pleural effusion. CVC, central venous catheter; TG, triglyceride; TPN, total parenteral nutrition.

This case highlights three major pitfalls. First, medical personnel continued utilizing a CVC that was not initially placed at the cavoatrial junction (CAJ). While suboptimal placement could have been suspected on chest X‐ray, definitive confirmation came via chest CT, which revealed the catheter tip located at the brachiocephalic‐SVC junction. Despite this evidence, the CVC was considered functional, and its potential contribution to complications was disregarded. Second, enteral nutrition was unnecessarily withheld for an extended period due to an initial, unconfirmed suspicion of chylothorax. This misdiagnosis likely delayed optimal management, complicating the clinical course. Third, the catheter remained suboptimally placed for 19 days, exceeding the recommended 2‐week duration. Notably, the mean onset time for extravasation from correctly positioned CVCs is approximately 3.6 days [10].

In this patient, removal of the CVC, after the incident, represented standard‐of‐care treatment. While prolonged catheter dwell time is a known risk factor for catheter‐related bloodstream infections, it remains challenging to establish a consistent association with vascular injury or extravasation, especially if the catheter tip is appropriately placed. Although angiography, autopsy, or external inspection was not performed, the absence of recurrent symptoms after CVC removal and PICC placement supported extravasation due to vascular erosion rather than puncture‐related leakage; nonetheless, we propose three plausible mechanisms for this complication.

5.1. Anatomical Factors

When the left‐sided catheter tip is positioned at the junction between the brachiocephalic vein and the SVC, the sharp angle formed between the catheter tip and the lateral wall of the SVC may cause mechanical erosion [8]. This phenomenon occurs more frequently with left‐sided CVC placements due to the anatomical configuration of the left brachiocephalic vein, which creates a sharper angle at the junction. Additionally, pulsations from the innominate artery, located posteriorly to the brachiocephalic–SVC junction, may further exacerbate mechanical stress at this site, increasing the risk of vessel wall erosion [16].

5.2. Osmolar Damage From Infused Solutions

Extravasation may also result from endothelial damage caused by hyperosmolar solutions, such as TPN or propofol, especially when these substances are infused rapidly. Hyperosmolarity‐induced endothelial injury, combined with the tensile stress exerted by high‐flow infusion, can produce sustained vessel wall damage over time. This dual mechanism—chemical irritation and mechanical strain—can increase vessel permeability and compromise wall integrity, leading to extravasation.

5.3. Mechanical Damage and Catheter Material

Direct intimal damage may occur during CVC placement, particularly when employing the Seldinger technique. The catheter tip or guidewire can tear the vessel intima, especially if excessive manipulation or trauma occurs during insertion [17]. Although such injuries are occasionally detected early, they may initially remain unrecognized. Although there may have been trauma to the intima, no hematoma or stoma was observed on external examination, as is the case in autopsy findings [6]. Subsequently, cumulative chemical and mechanical stresses or progression of vessel‐wall dissection may lead to rupture, necessitating surgical intervention in severe cases. Catheter material and stiffness may influence vessel wall injury; experimental data suggest silicone designs have lower perforation potential than polyurethane, yet similar events occur more frequently even with small silicone catheters used in pediatric patients, who have relatively thinner vessels and more fragile vascular integrity [18, 19].

These mechanisms underscore the necessity of meticulous CVC placement, careful monitoring of infusion parameters, and prompt recognition of potential complications to minimize extravasation risk and associated sequelae. Therefore, the catheter must be placed correctly from the outset. When inserting a CVC, proper tip position should be verified in real time using methods such as ECG‐guided tip localization or ultrasound bubble testing to confirm CAJ placement [20, 21].

Unrecognized complications associated with CVCs can be life‐threatening. We recently encountered a case involving pleural effusion and respiratory distress caused by delayed extravasation of TPN, occurring 18 days following CVC placement. This case underscores the critical importance of vigilance for complications related to CVC use. When an unexplained pleural effusion suddenly arises and a catheter is positioned suboptimally, clinicians must promptly consider catheter‐related extravasation to prevent further adverse outcomes.

Author Contributions

Yong Chae Jung: data curation, investigation, visualization, writing – original draft. Jun Wan Lee: conceptualization, methodology, project administration, supervision, validation, writing – review and editing.

Funding

The authors have nothing to report.

Consent

Written informed consent was obtained from the patient's guardian.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

This case report does not include any publicly available datasets. All clinical data referenced were obtained from routine patient care and have been fully anonymized to protect patient confidentiality.