Chest tightness and mediastinal widening

Discussion

The answer is (c): a mediastinal hematoma due to a Port-a-Cath complication and the next best step is (a): arrange an urgent CT scan of the chest. This diagnosis was supported by several nonspecific findings (including the timing of the patient's symptoms, her normal blood pressure, widened mediastinum with a particular pain quality and her decreased hemoglobin level) and was confirmed by a CT scan of her chest (Figure 2). The tip of the Port-a-Cath, which had likely abutted the wall of the left brachicephalic vein for some time (Figure 3), created local erosion. This erosion ruptured the vein wall leading to a mediastinal hematoma when pressure was applied by the pressure infusor.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2: Contrast-enhanced CT scan of the chest revealed a right mediastinal hematoma (arrow heads) compressing the superior vena cava (double arrow) and extrusion of catheter tip (arrow).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3: Contrast-enhanced CT taken 2 months before the red blood cell transfusion. On retrosepctive review, it was noted that the catheter tip (arrow) was abutting the vessel wall with anterior tension.

A mediastinal hematoma is often caused by blunt chest trauma and rupture of aortic aneurysms. It is also a rare complication of central venous catheter insertion and tends to occur within 1 week of catheter insertion.1^,2 Patients often present with sudden onset dyspnea or chest pain, and chest radiographs often show mediastinal widening and unilateral or bilateral pleural effusions.2 Abutment of the catheter tip against a vein wall is often overlooked on chest radiographs used to confirm the catheter placement. An appropriate position of the catheter tip is parallel to the superior vena cava. A retrospective review of our patient's radiograph taken 3 weeks before presentation (Figure 1) shows that the catheter tip is not in the correct location.

Differential diagnosis: Although suggestive of aortic dissection, acute development of a widened mediastinum is not specific to aortic dissection. Patients with aortic dissection classically report a sudden, intense tearing or knife-like pain; however, our patient did not report these symptoms. In addition, about 70% of patients with aortic dissection have pre-existing hypertension,3 which was also not the case for our patient. Other common but nonspecific signs of aortic dissection, such as decreased peripheral pulse and a murmur of aortic regurgitation, were absent in our patient.

Transfusion-related acute lung injury commonly manifests as dyspnea, cough, fever, hypotension and sometimes frothy sputum. These symptoms typically begin within a few hours of starting a transfusion.4^,5 The timing of our patient's symptoms were consistent with acute transfusion-related lung injury; however, the chest radiograph, which showed an absence of alveolar and interstitial infiltration and involvement of both lung fields, did not support this diagnosis.

Without ECG or myocardial enzyme evidence of an acute coronary syndrome, this diagnosis was effectively ruled out. Although we couldn't rule out a pulmonary embolus, based on our patient's initial presentation, a plausible alternative diagnosis explaining the mediastinal widening made this diagnosis less likely. The CT scan of our patient's chest, which better characterized the mediastinum, also ruled out this cause.