Superior vena cava perforation after central venous catheterization: the diagnostic role of progressive CVP elevation in occult injury-a case report

Abstract

Background

Central venous catheter (CVC) placement is crucial for perioperative hemodynamic management in anesthesia, and superior vena cava (SVC) injury is a rare but life-threatening complication.

Case presentation

A 63-year-old woman underwent surgery for hepatic hemangioma. During preoperative CVC placement via the left internal jugular vein (LIJV), mechanical injury resulted in SVC perforation and subsequent hemorrhage. Immediately after catheter placement, aspiration and flushing of both lumens were performed. Blood return was unobstructed with no abnormal resistance, and baseline central venous pressure (CVP) was recorded at approximately 5 cm H₂O. As CVP progressively and persistently increased with hemodynamic changes, thoracic ultrasound revealed right-sided pleural effusion. Thoracoscopy confirmed extrapericardial SVC injury, which may have resulted from either direct trauma by the guidewire at the angulated vascular junctions, or initial guidewire-induced submucosal injury that was subsequently disrupted by catheter advancement.

Conclusion

This case highlights a potential risk of SVC injury associated with the LIJV approach, particularly in patients with complex anatomy or prior catheterizations. Ultrasound monitoring should be optimized during catheter placement. Abnormally elevated CVP after catheter placement warrants high vigilance and prompt thoracic/pericardial evaluation.

Background

The superior vena cava (SVC) is vulnerable to various injuries, including iatrogenic trauma [1, 2]. Among the causes of iatrogenic SVC injury, mechanical injury associated with central venous catheter (CVC) placement is particularly notable [3]. CVC insertion is a critical procedure for monitoring and fluid management during the perioperative period and in intensive care settings. SVC injury during catheterization may lead to life-threatening complications such as hemothorax or cardiac tamponade; however, routine ultrasound guidance does not always prevent such complications [4]. The literature indicates that identifying vascular injury is more challenging under general anesthesia due to the masking of clinical symptoms and signs [5]. This paper reports a case of suspected and subsequently confirmed SVC injury following CVC placement via the left internal jugular vein (LIJV) and reviews the relevant literature. Unlike previous studies, this paper focuses on the diagnostic course of the case by tracing the sequence of events: “Progressive central venous pressure (CVP) elevation prompted bedside ultrasound examination, and the diagnosis was confirmed via thoracoscopy.” This approach aims to provide a clinically relevant reference for the rapid identification of occult major vascular injuries.

Case presentation

A 63-year-old woman (height, 165 cm; weight, 63 kg; body mass index, 23 kg/m²) was admitted for an elective laparoscopic resection of hepatic hemangiomas Upper abdominal computed tomography revealed multiple liver hemangiomas, with the largest lesions measuring approximately 73 × 77 mm in segments S3 and S6. Her medical history was notable for breast cancer, for which she had received multiple cycles of chemotherapy, with previous CVC placements via the right internal jugular vein (RIJV) or right subclavian vein (RSV).

On arrival in the operating room, her vital signs were as follows: blood pressure (BP), 125/76 mmHg; heart rate (HR), 72 bpm; temperature (T), 36.6 °C; and oxygen saturation (SpO₂), 100%. Following preoxygenation with 100% oxygen at 6 L/min, general anesthesia was induced with intravenous cipepofol (25 mg), rocuronium bromide (70 mg), and remifentanil (0.5 µg/kg/min). Anesthesia was maintained with oxygen (1.4 L/min), air (0.6 L/min), sevoflurane (1 vol%), cipepofol (0.8 mg/kg/h), and remifentanil (0.05 µg/kg/min). A left radial arterial catheter was placed for continuous BP monitoring. To facilitate controlled low central venous pressure (CLCVP) during surgery, placement of a 7-Fr double-lumen CVC (Model: 10-2361, Zhengzhou Diall MEDICAL Technology Co. Ltd., Zhengzhou, Henan, China) via the internal jugular vein (IJV) was planned.

Prior to CVC placement, ultrasound examination of both IJVs revealed that the RIJV was markedly narrower than the LIJV, with luminal diameters of 6 mm and 13 mm, respectively. In view of the history of multiple previous CVC placements on the right side, the LIJV was chosen. Under ultrasound guidance, the LIJV was punctured with an 18-gauge needle at an angle of 30°. The needle was advanced to a depth of approximately 3 cm below the skin until dark red, non-pulsatile venous blood was aspirated. The needle was stabilized, and a J-tipped guidewire (GW) was advanced through the needle. Two slight resistance changes were felt at approximately 6 cm and 11 cm during GW advancement, but it passed without requiring significant force. When the GW was inserted to a depth of about 20 cm from the skin surface, the needle was withdrawn. A dilator was then advanced over the GW to a depth of about 3 cm below the skin and subsequently removed. The catheter was inserted and fixed at 13 cm. The removed GW was inspected and showed no deformation. It indicates that the transition to GW has gone smoothly. Dark red venous blood was freely aspirated through the catheter, and the initial CVP measured approximately 5 cm H₂O. To rule out instrument-related factors, we rechecked the zero crossing, sensor height, and connections for air bubbles or kinks, finding no abnormalities.

Subsequently, over the following 30 min while awaiting the start of surgery, the patient developed tachycardia, severe hypotension (BP dropped to 70/30 mmHg), and a persistently rising stroke volume variation (SVV) (increasing from 8% to 16%), suggestive of hypovolemia. Concurrently, CVP progressively rose from the baseline of 5 cm H₂O to 16 cm H₂O, indicating a potential volume overload state. During this period, a total of approximately 750 mL of fluid was administered: 350 mL of succinylated gelatin via the central venous catheter and 400 mL of crystalloid through peripheral access. Given the resistance felt during GW placement, mechanical injury causing pericardial effusion or intrathoracic bleeding could not be ruled out. A thoracic ultrasound was performed, which revealed a large right-sided pleural effusion with no signs of pleural adhesion. The effusion was predominantly located in the dependent lung bases, leading to compression of the right lung towards the hilum and mediastinum (Fig. 1). Concurrent bedside echocardiography demonstrated reduced end-diastolic volumes of both the right and left ventricles, consistent with hypovolemia, with no evidence of pericardial effusion. Based on the presence of “signs of volume depletion,” “contradictory CVP elevation,” and “positive thoracic ultrasound findings,” the initial diagnosis was catheter-related intrapleural hemorrhage with a strong suspicion of SVC injury. Thoracic surgery consultation was obtained immediately, and thoracoscopic exploration and treatment were undertaken without delay.

Fig. 1.

Thoracic ultrasound image showing right pleural effusion with compression of the right lung toward the hilum and mediastinum

Thoracoscopy was subsequently performed by the thoracic surgeons. A thoracoscope was inserted into the right pleural cavity via the fourth intercostal space at the right midaxillary line, revealing a large amount of bloody pleural effusion and blood clots. After aspirating approximately 800 mL of blood, exploration identified a hematoma on the extrapericardial segment of the SVC, suspected to be the bleeding source. No active bleeding was observed, likely because of clot formation and tamponade. The CVC was not visualized penetrating through the SVC into the pleural space, and no pericardial effusion was detected (Fig. 2). Hemodynamic support included intravenous norepinephrine (0.05 µg/kg/min) and succinylated gelatin (600 mL/h). Surgeons applied absorbable gelatin hemostatic material to the bleeding site. Considering the CVC was correctly positioned within the vessel lumen and functional for infusion and CVP monitoring, it was decided to retain the catheter. After confirmation of the absence of ongoing bleeding or fluid leakage from the perforation site and excluding other bleeding sources, a chest tube was placed, and the thoracotomy was closed. The patient was transferred back to the ward after anesthesia recovery. Postoperative chest X-ray confirmed that the catheter was positioned within the superior vena cava, with its tip located at the junction of the superior vena cava and the right atrium.

Fig. 2.

Intraoperative thoracoscopic view showing the hematoma on the extrapericardial SVC (no active bleeding). No CVC was found penetrating the SVC into the thoracic cavity, and no pericardial effusion was detected

During the postoperative ward stay, no evidence of bleeding or respiratory distress was observed, and the CVC remained fully functional. The chest tube was removed three days later after a chest X-ray confirmed the resolution of the pleural effusion and lung re-expansion. Eight days later, the patient successful underwent laparoscopic resection of the liver hemangioma. Her postoperative recovery was uneventful, the central venous catheter was removed on postoperative day 6, and she was discharged on postoperative day 9. The key diagnostic clue in this case lies in the contradiction between the progressively rising CVP and signs of hypovolemia, suggesting that thoracic screening should be prioritized for this patient, followed by thoracoscopic confirmation and management.

Discussion

CVC insertion is a common invasive procedure in perioperative care, intensive care, and chemotherapy support, serving important clinical purposes such as fluid infusion, hemodynamic monitoring, and hemodialysis [6]. However, its widespread application is associated with a range of mechanical complications. Previous studies have shown that the overall incidence of CVC-related complications is approximately 15%–26%, with mechanical injury accounting for 5%–19% [7]. Common complications include arterial puncture, hemothorax, pneumothorax, catheter malposition, intravascular thrombosis, and catheter-related infections. In contrast, major vascular perforation, though rare, is associated with severe outcomes and a reported mortality rate as high as 12%–40% [8, 9]. Among these, SVC injury is one of the most dangerous types. Its clinical significance lies in the insidious clinical manifestations and the difficulties in localization during the early stages, which often lead to a delayed diagnosis. Such delays may result in life-threatening consequences, particularly in the setting of sudden massive hemorrhage [10]. Previous research studies have indicated that patients under anesthesia or mechanical ventilation frequently lack typical clinical symptoms. In addition, positive-pressure ventilation and changes in intrathoracic pressure may interfere with the interpretation of hemodynamic parameters, making vascular injury more likely to be overlooked [11].

SVC injury is related to multiple factors, including the type of infusate, insertion process, improper manipulation of the GW or dilator, catheter material and stiffness, vascular anatomical variations, and a history of previous catheterization [12–15]. Compared with the RIJV or SCV, the LIJV approach has a more complex anatomical pathway [16]. The LIJV drains into the SVC via the left brachiocephalic vein (LBCV), creating an angle that approaches 90°, where the GW and catheter may exert direct pressure as they traverse the junction. Previous studies have demonstrated that the left-sided approach is more prone to tangential stress at the lateral wall of the SVC, which can cause vascular intimal compression, tearing, or perforation [17–19]. In the present case, catheterization was performed through the LIJV. Two brief episodes of slight resistance were noted during GW advancement, which may correspond to the turns at the LIJV–LBCV and LBCV–SVC junctions. Although the dilator was limited to a subcutaneous depth of 3–4 cm, the ultimate perforation may have resulted from either direct trauma by the GW itself at these angulated segments, or the subsequent disruption of a GW-induced submucosal injury during catheter advancement. In addition, the physical properties of the catheter may have contributed to the mechanism of injury. Stiffer catheters, particularly those made of shape-memory polyurethane, may undergo minor displacement and continuous friction under the repetitive motion of cardiac pulsation and respiration, leading to progressive transmural erosion of the vessel wall [18]. Hemodynamic studies have indicated that at structurally complex sites turbulent flow can exacerbate wall shear stress, thereby further compromising vascular integrity and accelerating intimal injury [20]. Although the injury in the present case occurred immediately after catheter insertion, prior repeated right-sided catheterizations may still have been a contributing factor from a mechanical perspective. Chronic vascular wall fibrosis or structural remodeling related to previous catheterization could have altered stress distribution within the central venous system and indirectly increased susceptibility to injury in the left-sided SVC segment.

The diagnosis of SVC perforation under anesthesia is particularly challenging. Most catheter-related intrathoracic injuries lack specific early symptoms and initially manifest as unexplained hypotension, tachycardia, and/or decreased oxygenation. Under general anesthesia, these abnormalities are especially likely to be obscured [21]. When pleural effusion is formed by isotonic blood-like fluid, its conductivity may maintain apparently normal waveform readings on the monitor, further masking the underlying problem [22]. Traditionally, poor blood return or difficulty with aspiration after catheter placement has been regarded as an early warning sign of catheter malposition or dysfunction, warranting prompt reassessment of tip location and evaluation for mechanical complications [23]. In the present case, however, routine aspiration was smooth and the catheter lumen remained patent, and CVP continued to rise progressively. Importantly, CVP does not directly reflect intravascular volume status; rather, it represents the combined influence of right-sided cardiac preload and intrathoracic pressure [24]. Mechanical ventilation, body position, pulmonary compliance, and pleural effusion can all significantly affect the measurements [25]. When hypotension and an increased SVV indicate hypovolemia, yet CVP rises paradoxically, this pattern suggests hemodynamic incongruence among preload-related indicators, which should prompt consideration of secondary causes, particularly abnormal intrathoracic or pericardial pressure. In this case, this inconsistency provided the key diagnostic clue. After excluding measurement errors, catheter obstruction, and mechanical ventilation effects, the persistently elevated CVP suggested increased pressure within the thoracic or pericardial cavity. Subsequent bedside ultrasonography demonstrated a large right-sided pleural effusion, confirming that the elevated CVP reflected pseudo–venous hypertension caused by intrathoracic blood accumulation rather than true volume overload.

In this case, the paradoxical elevation of CVP, despite the central venous catheter being correctly positioned within the SVC lumen, can be explained by extravascular compression of the SVC or right atrium by the surrounding hematoma and pleural effusion. This compression elevates the pressure reading at the catheter tip despite reduced right heart filling, a phenomenon that can occur in conditions such as tension hemothorax, cardiac tamponade, or mediastinal effusion [26]. In other words, the dynamic trend of CVP, together with its relationship to systemic hemodynamic parameters, is more diagnostically meaningful than any isolated measurement. Interpretation of CVP trends in the context of overall volume status may substantially improve the sensitivity for early detection of occult vascular perforation.

The widespread adoption of ultrasonography has significantly reduced the mechanical complications associated with CVC placement. Randomized controlled studies have shown that ultrasound guidance decreases the number of puncture attempts, shortens procedure time, and reduces complication rates [27]. However, most studies have focused primarily on its role during puncture, lacking real-time visualization of subsequent GW and catheter advancement [28]. More recently, some researchers have proposed the concept of real-time full-path ultrasound guidance, in which the GW trajectory and final catheter position are continuously monitored during advancement, dilation, and insertion [29]. The supraclavicular transverse view, in particular, allows real-time visualization of the GW as it crosses the LBCV–SVC junction, thereby helping to prevent deviation from the vessel lumen [30].

In this case, although ultrasound guidance was employed during venipuncture, the GW advancement was not continuously monitored under ultrasound, which may have contributed to the injury. If the operator had switched to the supraclavicular window when the GW reached the level of the manubrium and confirmed its entry into the SVC rather than along the vessel wall, the complication might have been avoided. Postprocedural confirmation of catheter position is equally important. Conventional chest radiography can assess the depth in relation to the cardiac silhouette but cannot identify extravascular misplacement; combining it with bedside ultrasonography may improve diagnostic accuracy. Recent studies have also suggested that intraoperative intracavitary electrocardiography (IC-ECG) and CVP waveform analysis may serve as supplementary methods for catheter placement. As the catheter tip progressively approaches the sinus-atrial junction (SVC–RA/CAJ), the amplitude of atrial components (P wave/a wave and v wave) and their synchrony with surface ECG significantly increase. This serves as a reliable endpoint indicator and reduces reliance on postoperative chest radiographs [24, 31, 32].

For patients with confirmed or strongly suspected SVC perforation, catheter management requires extreme caution. Previous reports have indicated that blind removal may abolish the tamponade effect at the site of vascular injury and precipitate catastrophic hemorrhage [14, 15]. Conversely, in hemodynamically stable patients with contained bleeding, retaining the catheter as a monitoring and therapeutic conduit may be safer. In this case, thoracoscopic exploration revealed no active bleeding, and catheter function remained intact; therefore, the catheter was preserved for continuous CVP and chest drainage monitoring, ultimately achieving a favorable outcome. This strategy is consistent with previous reports suggesting that, when bleeding has ceased and the hemorrhage remains localized or controlled, cautious catheter retention may be preferable to immediate removal [33]. Furthermore, postoperative management should emphasize dynamic monitoring of intrathoracic pressure changes. CVP monitoring can serve not only as an early indicator of recurrent bleeding but also as a measure of drainage efficacy. A renewed increase in CVP accompanied by reduced drainage output should raise suspicion of chest tube obstruction or recurrent hemothorax. Continuous hemodynamic monitoring integrated with multimodal imaging is essential for identifying ongoing or delayed hemorrhage, preventing vascular complications, and avoiding mediastinal compression syndrome.

This case illustrates a typical diagnostic paradigm: when catheter function appears normal but CVP changes are inconsistent with the clinical assessment of intravascular volume, extravascular compression or vascular perforation should be considered. In addition, this case highlights the renewed value of CVP monitoring in the management of critically ill patients. Although CVP has long been used to assess volume status, its ability to predict fluid responsiveness is limited [34]. Nevertheless, when interpreted together with other hemodynamic parameters, such as SVV or cardiac output, and assessed in terms of temporal trends, CVP remains an important warning sign. In anesthetized or mechanically ventilated patients, abnormal CVP variations may represent the only quantifiable early signal of an impending complication. Ultrasonography should not be limited to puncture guidance alone; it should also include preprocedural vascular assessment, real-time GW tracking, and postoperative catheter localization. Standardized application of ultrasound throughout the catheterization process, especially in patients with multiple prior catheterizations, may substantially enhances procedural safety.

Finally, based on the diagnostic pathway demonstrated in this case, we propose the following clinical monitoring protocol to aid in the early detection of similar occult injuries: (1) record the baseline CVP and waveform immediately after catheter placement; (2) during subsequent monitoring, when hypotension or an elevated stroke volume variation suggests hypovolemia, assess the CVP trend simultaneously; (3) if CVP continues to rise despite correct catheter placement, perform thoracic and pericardial ultrasonography without delay; (4) when hypotension or an elevated stroke volume variation suggests hypovolemia, assess the CVP trend simultaneously The main limitation of this case report is the absence of intraoperative real-time imaging of the GW and catheter advancement, which precluded precise determination of the timing and mechanism of injury. Moreover, a single case cannot be generalized to all patient populations.

Conclusions

In summary, this case demonstrates that SVC perforation is a rare but potentially fatal complication of CVC insertion. It highlights the potential risk associated with the LIJV approach, particularly in patients with complex anatomy or a history of prior catheterizations. Ultrasound should not be confined to the initial puncture phase but should cover multiple steps of the CVC insertion process. Most importantly, this case demonstrates that a persistently elevated CVP inconsistent with the clinical volume status should be recognized as a specific early warning sign of occult vascular injury or extravascular compression, rather than merely a measurement artifact. This diagnostic insight, coupled with the proposed monitoring protocol, constitutes the key clinical contribution of this report.

Abbreviations

BP

Blood pressure

CLCVP

Controlled low central venous pressure

CVC

Central venous catheter

CVP

Central venous pressure

GW

Guidewire

HR

Heart rate

IJV

Internal jugular vein

LIJV

Left internal jugular vein

RIJV

Right internal jugular vein

RSV

Right subclavian vein

SVC

Superior vena cava

SpO₂

Oxygen saturation

SVV

Stroke volume variation

T

Temperature

Authors' contributions

GL conceived and designed the study, organized the patient case data, created the figures, and wrote the initial draft. NJ collected the patient case data, conducted follow-up assessments, performed data verification, assisted in the methodological design of the study, and conducted the literature review. DZ reviewed and revised the initial draft of the manuscript. RL critically reviewed and edited the initial draft of the manuscript and performed format proofreading. All authors have read and approved the final version of the manuscript for publication.

Funding

Not applicable.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of Guangdong Provincial People’s Hospital, Guangdong Academy of Medical Sciences, Southern Medical University. (Reference number: KY2025-880-02). Written informed consent to participate was obtained from the patient.

Consent for publication

Informed written consent was obtained from the patient for the publication of this report and any accompanying images.

Competing interests

The authors declare no competing interests.