Reply

Abstract

We have clearly demonstrated the efficacy of TIPS in the prevention of both gastric (GVB) and esophageal variceal (EVB) rebleeding. We agree that there is a lack of long-term data regarding endoscopic treatment of gastric varices, but believe that although the available data are promising, randomized studies in this field are required. The potential for better survival in patients who present with GVB has been noted in a previous study by our unit (Aliment Pharmacol Ther 1997;11:171–176). The increased number of patients and extended follow-up has resulted in the difference now being significant, particularly in the long term as demonstrated by a 5-year mortality of 49.5% versus 74.9% in the GVB and EVB groups, respectively. This difference in mortality is not fully understood, but one possibility is that the portal pressure gradient (PPG) may be a direct predictor of mortality as we have shown that it is significantly lower in the GVB group. Studies indicate a detrimental effect on mortality following TIPS insertion in patients with higher portal pressures, particularly shortly after the variceal bleed (Hepatology 1986;6:116–117, Gastroenterology 1999;117:626–631, J Hepatol 1995;23:123–128). All of our patients had TIPS insertion using the same technique, although some patients with GVB were treated with thrombin injections. We have recently reported our experience using this endoscopic modality (Am J Gastroenterol 2002;97:1381–1385). Variceal hemorrhage particularly from gastric varices does occur frequently at PPG of ≤12 mm Hg, a finding we have reported in the past (Aliment Pharmacol Ther 1997;11:171–176, Am J Gastroenterol 1995;90:1994–1996) and reinforced by the present study. This may be caused by the development of gastro-renal portosystemic shunts (Gastroenterology 1988;95:434–440), or the large size of the varices resulting in increased variceal wall tension (Semin Liver Dis 1986;6:318–331). Most of the gastric varices in our study originated from the fundus or cardia and were of a large size. Other factors such as the presence of red spots, variceal size, and that of gastritis may be important, although it is noteworthy that the NIEC index only applies to esophageal varices (N Engl J Med 1988;319:983–989). Our findings highlight the need to reconsider the therapeutic targets for portal pressure reduction following TIPS. We suggest, particularly for patients with EVB, that a PPG post-TIPS of <7 mm Hg is desirable to minimize variceal rebleeding. Large multicenter trials comparing TIPS with other endoscopic and pharmacological treatment modalities would be highly informative. In our experience, thrombin seems to be safe and effective, and therefore should be investigated further. In the absence of cost-effective analysis, we would currently continue to recommend TIPS as first line treatment in patients with refractory GVB in the presence of a patent portal vein.