Background/aims: Maintenance of long-term patency of transjugular intrahepatic portosystemic stent-shunts (TIPSS) has proved problematic. Various prognostic variables have been assessed as predictors, but the role of diabetes mellitus, which induces vascular endothelial cell dysfunction, has not been assessed.
Methods: We analysed the records of 248 patients who underwent TIPSS between July 1991 and July 1997, followed-up through to August 1998. Patients with at least one shunt assessment by portography and available blood glucose levels were eligible (177 patients; median follow-up, 15.0 months). Fourteen patients had a pre-procedural diagnosis of diabetes (one insulin dependent, seven oral hypoglycaemic treated and six diet controlled). In another 14 patients, diabetes was diagnosed at TIPSS insertion, giving a 28/177 (15.8%) prevalence of diabetes in our patients. Fifty-nine patients were excluded from the final analysis (including five diabetics), as they either died or had early shunt insufficiency (within 1 month of stent placement), leaving 118 patients (including 23 diabetics) to be included in the final analysis.
Results: Mean age, sex distribution, median follow-up (months) and pre-shunt portal pressure gradient were comparable in the two groups (diabetics versus non-diabetics). Child-Pugh classes A and B were more common in the diabetic group (P < 0.01), and the mean inserted stent diameter was larger in the diabetic group (P < 0.05). The presence of diabetes was associated with a higher incidence of delayed shunt insufficiency (P = 0.02), but there was no evidence of an association between presence of diabetes and variceal haemorrhage post TIPSS. Kaplan-Meier analyses revealed earlier insufficiency in diabetic patients compared with those without diabetes (P = 0.04). Age, gender and presence of diabetes are included in the final logistic regression model. Individuals who have diabetes are more likely to experience shunt insufficiency independent of age and gender.
Conclusions: Diabetes mellitus is common in patients undergoing TIPSS and is associated independently with increased incidence of primary delayed shunt insufficiency.