• cirrhosis of the liver;
  • endoscopic sclerotherapy;
  • extrahepatic portal venous obstruction (EHO);
  • hepatic functional status;
  • long-term management;
  • non-cirrhotic portal fibrosis (NCPF).


A prospective study was conducted to compare the results of long-term endoscopic variceal sclerotherapy in patients with different aetiologies of portal hypertension. A total of 404 consecutive patients were included. There were 234 patients with hepatic cirrhosis, 83 with non-cirrhotic portal fibrosis (NCPF) and 87 with extrahepatic portal venous obstruction (EHO). The mean follow-up for patients with cirrhosis, NCPF and EHO was 25, 37 and 28 months. A total of 73 (31%) patients with cirrhosis, 19 (23%) with NCPF and 10 (11.5%) with EHO rebled (P < 0.05) on follow-up, prior to eradication of varices. Irrespective of the aetiology, 40 (17%) patients of Child's A class, 42 (33%) of Child's B and 20 (50%) of Child's C class rebled (P<0.01). The median bleeding free period (BFP) was longer (P<0.05) in patients with EHO than in cirrhotics. Patients in Child's A class had significantly longer BFP than those in Child's B, and the latter had a longer BFP than those in Child's C class (P<0.01). The probability of 7-year survival was also better with EHO (97.5%) and NCPF (73.6%) than cirrhotics (41%). Survivals in patients with EHO and NCPF were comparable (P<0.1). Similarly 7-year survival irrespective of aetiology in Child's A patients (90.7%) was longer than in Child's B (28.8%), and longer in Child's B than Child's C patients (0%). Success of eradication was greater (P<0.05) in EHO (92%) and NCPF (87%) than cirrhotic patients (75%). This was also related to the Child's status of patients. Recurrence of varices, and complications related to the procedure, were not influenced by Child's status or by the aetiology of portal hypertension. For a given Child's class, the results of scierotherapy in the different aetiological groups were similar.

We conclude that the results of long-term sclerotherapy are affected primarily by the Child's scores and indirectly by the aetiology of portal hypertension.