D. Williams et al., MORTALITY AND REBLEEDING FOLLOWING TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC STENT-SHUNT FOR VARICEAL HEMORRHAGE, Journal of gastroenterology and hepatology, 13(2), 1998, pp. 163-169
The present study investigates clinical factors associated with decrea
sed survival following Transjugular Intrahepatic Portosystemic Stent S
hunt (TIPSS). Sixty-seven patients underwent TIPSS for bleeding relate
d to portal hypertension, 42 (63%) on an urgent basis. TIPSS was succe
ssfully placed in 65 (97%) patients with no fatal procedural complicat
ions. Thirty day mortality was 21%, there being several predictive fac
tors: transfer from another institution, urgency of procedure, sepsis,
encephalopathy, higher mean serum bilirubin and low serum albumin. Ho
wever, using regression analysis, 30 day mortality was predicted indep
endently only by severe liver disease (Child-Pugh C, P = 0.003) and ol
der age (P = 0.003). When stratified by Child-Pugh class, cumulative s
urvival rates at 1 year for class A, B and C were 100, 90 and 34%, res
pectively. Only three of 25 patient deaths were due to variceal reblee
ding. Thirty (46%) patients had a total of 41 rebleeding episodes, wit
h mean time to first rebleed of 4.8 months (range, 3 days-38 months).
Cumulative rebleeding rate at 1 year was 25%. Log-rank analysis did no
t reveal a significant difference in overall survival between rebleede
rs and non-rebleeders (P = 0.125). When investigated, shunt abnormalit
ies (stenosis, occlusion) were identified in all cases of rebleeding.
Our findings confirm TIPSS can be safe and effective in the control of
refractory variceal haemorrhage. However, prognosis remains poor for
patients with advanced liver disease, particularly if older and in the
emergency setting. Vigilant surveillance and high rate of interventio
n is necessary to maintain shunt patency. Consideration could be given
to elective shunt surgery instead of TIPSS for patients with recurren
t bleeding and good prognosis liver disease.