The Authors’ Response

    loading  Checking for direct PDF access through Ovid

Excerpt

We would like to thank Stine and coauthors1 for their comments on our study examining the impact of portal vein thrombosis (PVT) on waitlist and postliver transplant outcomes based on the United Network for Organ Sharing (UNOS) data.2 The authors highlighted our analysis of risk factors for PVT1 which was based on a subset of patients who are listed for transplant without PVT but were documented to have PVT at the time of liver transplantation.2 The authors correctly point out that important independent predictors of PVT 0.001 included both diabetes mellitus (odds ratio, 1.23; 95% confidence interval, 1.11-1.36; P< 0.001) and obesity (body mass index, >30 kg/m2) (odds ratio, 1.13; 95% confidence interval, 1.03-1.23; P<0.001).1 However, we would like to clarify that we identified both nonalcoholic fatty liver disease (NAFLD) and cryptogenic cirrhosis (grouped together in our analysis) rather than cryptogenic cirrhosis alone as the most important predictor of PVT (odds ratio, 1.5; 95% confidence interval, 1.33-1.71; P<0.001).2 This would not be surprising because most patients with cryptogenic cirrhosis most likely had underlying NAFLD or nonalcoholic steatohepatitis (NASH).
The authors had also analyzed UNOS data with the aim of examining independent associations between NASH cirrhosis and PVT at the time of liver transplantation.3 In their analysis, Stine et al included all patients undergoing liver transplantation with PVT. Importantly, body mass index was analyzed as a continuous variable rather than using a specific cutoff value. Patients with cryptogenic cirrhosis were excluded, and as stated, diabetes was accounted for using interaction terms with NASH (interchangeable with fatty liver disease etiology in UNOS data) and obesity. Despite some differences in design and the analytic approach, they report nearly identical association of fatty liver disease etiology with PVT at transplantation (odds ratio, 1.55; 95% confidence interval, 1.33-1.81; P<0.001), although body mass index as a continuous variable was not a predictor of PVT in the final model. Despite some limitations of the UNOS data, including a reporting bias of higher grade PVT,2 taken together, our studies add further epidemiologic evidence for the prothrombotic risk associated with fatty liver disease and should at least maintain a curiosity in the impact of obesity on thrombotic risk. A recently published study described remarkably reduced hepatic venous pressure gradients in cirrhotics over a 16-week period with weight loss of at least 5% in regardless of etiology of cirrhosis, although Childs-Pugh score did not change.4 It is possible that the low portal venous flow state of portal hypertension in cirrhosis may accentuate the risk for PVT in patients with obesity and represent an independent risk factor.
Importantly, the authors highlight conflicting reports on the prohemostatic contributions of NAFLD versus obesity, and the complex interplay of these factors.5,6 We support their forward-looking call for further mechanistic studies in well-phenotyped patients to better define thrombotic risk in NAFLD and metabolic syndrome, with potential clinical impact beyond mitigating the risk of PVT in patients with cirrhosis.
    loading  Loading Related Articles