Jugular venous reconstruction: A technique to minimize morbidity of bilateral radical neck dissection

Abstract

Disfiguring facial edema and elevated intracranial pressure often accompany ligation and excision of the internal jugular veins (IJV) during simultaneous or staged bilateral radical neck dissection. To minimize this morbidity, unilateral reconstruction of the IJV was undertaken at the time of second-side neck dissection in 11 patients. Five patients underwent reconstruction with the spiraled saphenous vein (SVG) and in six patients 10 mm externally supported polytetrafluoroethylene (e-PTFE) was used. Prebypass jugular venous stump pressures (JVPs) were measured and ranged from 2817 to 7554 mm Hg with mean pressures of 17 to 62 mm Hg. Two patients had simultaneous neck dissections and nine had staged dissections. The staged interval ranged from 0.3 to 33 months. Six of nine staged reconstructions were right-sided and three of nine were left-sided. Low molecular weight dextran was started intraoperatively, regional heparinization was used in all but two cases, and platelet inhibitors were continued postoperatively in all patients. Additional time in the operating room was 45 minutes (mean). All patients have been followed up 2 to 18 months postoperatively. B-mode ultrasonography, Doppler spectrum analysis, CT scan with contrast, and magnetic resonance imaging were used to assess patency. Four of five SVGs and three of six e-PTFE grafts have remained functional, which resulted in an overall patency rate of 64%. Four grafts (one SVG, three e-PTFE) failed in the immediate postoperative period. All grafts in patients who had a mean JVP >30 mm Hg remained patent, whereas those in patients with lower stump pressures thrombosed. All left-sided reconstructions thrombosed, whereas only one of six right-sided (staged reconstructions) failed. This probably represents the higher mean JVP measured on the right side (40 mm Hg) compared with the left (26 mm Hg). Clinical results were good. Of seven patients with patent grafts, three had mild facial edema on the first postoperative day, which quickly resolved. One patient (SVG plus an arteriovenous fistula) had moderate facial edema, which resolved within 3 weeks. Two of four patients with thrombosed grafts had mild facial edema. This minimal morbidity was attributed to good collateral venous drainage indicated by the low JVP. All patent grafts demonstrated phasic venous flow and SVG simulates normal jugular venomotion with respiration. These studies demonstrate that the disfiguring facial edema following bilateral radical neck dissection can be prevented by unilateral IJV reconstruction. The technique adds minimal additional operative time and results in a high overall patency rate with no increased morbidity. We believe that the SVG is preferable, especially in the contaminated wound; however, e-PTFE is a reasonable alternative. Patency correlates with JVP. These preliminary studies suggest that patients with a mean JVP ⩽30 mm Hg may not require construction, whereas those with higher JVPs can significantly benefit from maintaining jugular venous return.