Objective: To measure the impact of strict monitoring on long term permeability of TIPS.
Methods: From November 1991 to December 2002, 208 patients (152 men and 66 women managed for the placing of a TIPS following failure with sclerotherapy or refractory ascites were included. The permeability of the prosthesis was controlled on the surgical piece. Revision of the TIPS was performed at the same time as the diagnosis of obstruction using trans-jugular angiography. The TIPS permeability was classified into 3 categories: primary permeability (duration of permeability recorded up until any intervention); assisted primary permeability (duration of continuous permeability with or without percutaneous intervention); secondary permeability (total duration of permeability with or without treated occlusion).
Results: Among the 218 patients in the study, 108 had been monitored for less than one year (49.5%), 29 for between 1 and 2 years (13.5%), 27 between 2 and 3 years (12.5%), 16 between 3 and 4 years (7.5%), 15 between 4 and 5 years (7%) and 23 for more than 5 years (10%). The mean follow-up period was of 24.1 +/- 27.2 months (median: 13.7). Current survival of these patients was of 81.2 +/- 2.9% at 1 year, 62.9 +/- 4.2% at 3 years and 52 +/- 4.9% at 5 years. Thirty-four patients were lost to follow-up (16%), after a mean follow-up of 22.9 +/- 26.7 months (median: 9.7). All these patients exhibited a permeable TIPS during their last control. One hundred and thirty-five patients (62%) had never presented obstruction of the TIPS, with a mean follow-up of 19.5 +/- 26.2 months (median: 7.4); 83 patients (38%) exhibited 117 episodes of obstruction; 24 two episodes, 7 three and 3 more than three; 4 exhibited thrombosis provoked by an invalidating encephalopathy and contraindicating transplantation, 2 of them following a first episode of thrombosis. The current primary, primary assisted and secondary permeability rates were respectively: 67.7 +/- 3 and 514 +/- 3.7%, 79.9 +/- 2.3% and 69.3 +/- 3.4%, 94.4 +/- 1.8% and 91 +/- 2.6% at 1 and 3 years (p = 0.0001, Log Rank = 65.3). Univariate analysis revealed a relationship between the onset of an episode of early or late obstruction and the patients' age at the time the TIPS was inserted, Child's score before, and the pre- and post-surgical gradient. In multivariate analysis, none of these elements reached significance.
Conclusion: Although the use of TIPS in the treatment of portal hypertension follows the course of development of all surgery towards minimally invasive methods, it should not be considered as a surgical portal shunt or as a permanent treatment: long-term permeability is only obtained with strict and regular monitoring and the possibility of rapid intervention in the case of obstruction.