Upper extremity arteriovenous dialysis fistula causing arterialised blood flow in internal jugular vein in patients with brachiocephalic vein occlusion

Sir,

The internal jugular vein (IJV) has become the preferred central vein for anaesthesiologists and it is also an important anatomical landmark for the upper extremity nerve blocks. However, under certain circumstances, the IJV can be mistaken for the carotid artery. The circumstances leading to the mistake include rotation of the head (which changes their relative position and overlap)[1] and, in particular, the presence of retrograde blood flow.

The reversal of the blood flow through the IJV is a consequence of occlusion of the large veins in the superior mediastinum: The brachiocephalic vein (BCV) or the superior vena cava (SVC). The causes include malignancies, external compression by mediastinal fibrosis or aneurysm or thrombosis of the thoracic aorta, use of catheters or pacemaker electrodes or drug infusions.[2] In addition to the occlusion, a prerequisite for the flow reversal is the damage of the IJV valve, which is present in about 90% of people and ensures unidirectional blood flow.[3] Of note, the IJV valve can be damaged by a central venous catheter. Regardless of its mechanism, the obstruction of SVC or its major tributaries may affect the upper limb function and lead to the development of SVC syndrome. We present here two cases of arterialised blood flow in the IJV in patients with upper extremity arteriovenous fistula (AVF) for dialysis.

In the first patient, who had a right AVF, an orientational sonographic scan was performed before insertion of a central venous line via IJV. The subclavian veins (SCVs) and IJVs were patent bilaterally but the flow in the right IJV was reversed. Subsequent imaging revealed the right BCV occlusion, apparently due to complications of the SCV dialysis catheter placement in the past. Colour Doppler ultrasonography (CDUS) of the neck and right upper arm confirmed BCV occlusion and also showed engorged veins in the upper arm with some collateral veins in the axilla. Since one BCV was obstructed, and the SVC was open, the blood flow on the side of obstruction was mainly directed retrograde through the ipsilateral IJV, sigmoid and transverse sinus, to the contralateral IJV and BCV, to finally reach the SVC [Figure 1a].

Figure 1.

Schematic drawings of the venous flow in the upper body show an obstructed brachiocephalic vein as in the patient 1 (a) and bilateral obstruction as in the patient 2 (b). The arrows indicate the directions of the venous flow. SCV, subclavian vein; BCV, brachiocephalic vein; SVC, superior vena cava; TS, transversal sinus; SS, sagittal sinus; IJV, internal jugular vein; TVs, thoracic veins; fistula, upper extremity with the arteriovenous fistula

In the second patient, who had a left AVF, a sonographic scan was performed before a nerve block. CDUS demonstrated that the dialysis access in the left upper arm and the SCVs and IJVs were patent on both sides of the body, but with retrograde blood flow in the left IJV and in the right SCV [Figure 1b]. Both BCVs were occluded, with the extension of the occlusion to the SVC. Engorged collateral veins in both axillary regions were also detected. In this case with both BCVs obstructed, the blood flow was retrograde from the subclavian vein on the side of the haemodialysis fistula through the ipsilateral IJV and dural sinuses at the skull base, to finally reach the contralateral IJV. Because the contralateral BCV was also obstructed, blood flow was retrograde through the contralateral SCV and reached the heart via the anastomoses between the axillary vein and thoracic veins. The axillary route relies on the anastomoses between the axillary vein and the venous system of the thoracic wall, such as the lateral thoracic vein. The thoracic venous circulation then reaches the right atrium via the azygos/hemiazygos pathways.[2]

Obstruction of BCV and/or SCV may seem like a non-frequent clinical entity, but it may actually be underdiagnosed as it has been described in up to 29% of patients with polytetrafluoroethylene grafts, who required haemodialysis.[4] Furthermore, haemodialysis shunts are a predisposing factor for BCV obstruction even in patients with no history of malignancy or central venous lines.[5,6] One should also not forget hypercoagulability in end-stage renal disease and its contribution to the thrombosis of vascular access.[7]

To conclude, the anaesthesiologists should not only be aware of a possibility that central venous obstruction and reversed blood flow in IJV may be present in patients with an upper extremity haemodialysis shunt, but should even expect it, in particular, in patients with end-stage kidney disease. With central venous obstruction present, the central venous line should be placed using a femoral route, and caution should be exercised to avoid confusion when using large cervical veins as anatomical landmarks for the upper extremity nerve blocks.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.