Role of endoscopic endoprostheses in proximal malignant biliary obstruction

The management of hilar strictures is dependent upon their resectability and may therefore require a multidisciplinary approach. However, resectability rates for such tumors are reported to be in the region of 15%-20%, and, therefore, palliative therapy will be the mainstay of treatment for most patients. With the presenting symptoms being those of obstructive jaundice and the consequences of cholestasis, a significant improvement in morbidity can be obtained by achieving biliary drainage. A number of options are available, including the placement of Teflon or expandable metallic endoprostheses by either the endoscopic or percutaneous route. Some considerable debate exists as to which route of stent placement is best, and in many circumstances the decision will depend on the availability of local services. Some have suggested that success rates with percutaneous stenting are superior to those for endoscopic placement, but the latter technique may be associated with fewer complications. In competent hands, endoscopic placement does achieve a high rate of success and it should be remembered that a combined approach may further improve success rates. The debate over the use of plastic versus metallic stents is centered around the higher rates of stent occlusion/migration for plastic stents seen in some studies, although a stent change is usually possible. An additional advantage of metallic stents is that they may provide drainage of the side branches of the biliary tree through the mesh. However, possible drawbacks may be a greater difficulty in placement of a second stent where a first provides inadequate drainage, and cost issues often have to be taken into consideration. Considerable debate exists over the optimum number of stents required to achieve adequate drainage and minimize the risks of cholangitis. There is good evidence that if overfilling of the biliary tree with contrast is avoided with only the segments to be drained visualized, a single stent may be all that is required, while others argue that placement of more than one stent may improve survival. In the following review we discuss these issues, and conclude by considering success rates and complications following endoprosthesis insertion; we also discuss the prognosis of patients treated in this way.