Analysis Of The Natural Course Of Pediatric Portal Hypertension

Abstract

Presenter: Al-Faraaz Kassam MD | Cincinnati Children's Hospital Medical Center Background: The etiology of portal hypertension (pHTN) in children varies from that of adults and requires understanding of the differences in management strategies. We set out to assess the epidemiology and natural history of pediatric pHTN. Methods: 154 patients with pHTN from 2008-2018 were identified at a free-standing children’s hospital. Etiology, disease course, and interventions were assessed. Patients were then stratified by etiology of pHTN. The primary outcome was overall survival, with a secondary outcome of understanding the interventions and management required. Survival was calculated using Kaplan-Meier survival analysis. Statistical significance was set at p<0.05. Results: The distribution of etiology included 105 (68%) patients with cholestatic disease (CD), 16 (10%) with fibrotic or hepatocellular disease (HFD), four (3%) with hepatic vein obstruction (HVO), and 29 (19%) with pre-hepatic pHTN (PRE) caused by portal vein obstruction or portal vein sclerosis. Median follow-up for the cohort was 4.63 (IQR 2.02-7.25) years. Median age at diagnosis (years) varied between etiologies (0.34, IQR 0.06-0.9 CD vs 0.51, 0.06-3.33 HFD vs 5.85, 1.26-9.52 HVO vs 2.51, 0.85-5.09 PRE; p<0.01). There were no differences in sex, comorbidities, or years of follow up. Patients with PRE were more likely to require an EGD (79.3% vs 40% CD, 50% HFD, 50% HVO, p<0.01) and patients with PRE (2.0±0.4) and HVO (1.8±1.0) required more endoscopic interventions than CD (1.1±0.2) or HFD (1.4±0.5) (p<0.01). However, there were no differences regarding gastrointestinal bleeding or need for banding or sclerotherapy (p=0.31 and p=0.94 respectively). Interventions differed based on etiology (p<0.01) with CD more likely undergoing a transplant only (71.4%) and PRE more likely to undergo a shunt only (44.8%). 24.8% of CD, 62.5% of HFD, 100% of HVO, and 41.4% of PRE patients did not undergo an intervention. Kaplan-Meier analysis revealed no difference in survival based on etiology, however, analysis by type of intervention revealed a significant increase in mortality in the group who received no intervention compared to shunt, transplant, or both (p<0.01) (Figure 1). Of the 18 patients who died without intervention, 7 patients died while listed for transplant, five patients were not candidates due to progression of disease, and six patients had clinical deterioration before workup was completed. Of the 36 patients who are alive and have not undergone intervention, 27 have stable pHTN, six are waitlisted for organs, and three are not candidates for intervention due to severity of disease. Conclusion: The sequelae of pHTN in children is distinct from the adult process and a better understanding of this disease is necessary. In this single institution study, we found that most patients presented with CD as the cause of pHTN followed by PVO. While no difference in overall survival was noted between groups, patients who received no intervention had higher mortality than those who received an intervention. Early referral to specialized centers with experience managing these complex disease processes may allow for improved risk stratification and early intervention which may improve outcomes. Further work is needed to understand the predictors of mortality in patients with pHTN.