OBJECTIVES:Cardiac dysfunction has been implicated in the genesis of hepatorenal syndrome (HRS). It is unclear whether a low cardiac output (CO) or attenuated contractile response to hemodynamic stress can predict its occurrence. We studied cardiovascular hemodynamics in cirrhosis and assessed whether a diminished cardiac reserve with stress testing predicted the development of HRS on follow-up.METHODS:Consecutive patients undergoing liver transplant workup with dobutamine stress echocardiography (DSE) were included. CO was measured at baseline and during low-dose dobutamine infusion at 10 g/kg/min. HRS was diagnosed using guideline-based criteria.RESULTS:A total of 560 patients underwent DSE, of whom 488 were included after preliminary assessment. There were 64 (13.1%) patients with established HRS. The HRS cohort had a higher baseline CO (8.0 ± 2 vs 6.9 ± 2 L/min; P < 0.001) and demonstrated a blunted response to low-dose dobutamine (?CO 29 ± 22% vs 44 ± 32%, P < 0.001) driven primarily by inotropic incompetence. Optimal cutpoint for ?CO in patients with HRS was determined to be <25% and was used to define a low cardiac reserve. Among the 424 patients without HRS initially, 94 (22.1%) developed HRS over a mean follow-up of 1.5 years. Higher proportion with a low cardiac reserve developed HRS (52 [55.0%] vs 56 [16.9%]; hazard ratio 4.5; 95% confidence interval 3.0-6.7; P < 0.001). In a Cox multivariable model, low cardiac reserve remained the strongest predictor for the development of HRS (hazard ratio 3.9; 95% confidence interval 2.2-7.0; P < 0.001).DISCUSSION:Patients with HRS demonstrated a higher resting CO and an attenuated cardiac reserve on stress testing. On longitudinal follow-up, low cardiac reserve was an independent predictor for the development of HRS. Assessment of cardiac reserve with DSE may provide a novel noninvasive risk marker for developing HRS in patients with advanced liver disease.HRS is a life-threatening complication of liver disease. We studied whether an inability to increase cardiac contraction in response to stress can assist in the prediction of HRS. We demonstrate that patients with liver disease who exhibit cardiac dysfunction during stress testing had a 4-fold increased risk of developing HRS. This may improve our ability for early diagnosis and treatment of patients at a higher risk of developing HRS.