Cirrhotic cardiomyopathy is characterized by a hyperdynamic circulatory state with high cardiac output at rest due to a low systemic vascular resistance (17). It remains a poorly described entity in pediatrics. It carries significant mortality and morbidity including increased incidence of perioperative complications and graft rejection (18, 19). Our study assesses conventional and strain echocardiographic parameters of ventricular function in pediatric patients with cirrhosis who are listed for liver transplantation.
We found that an overwhelming 74% of our cohort had abnormally low RV GLS values prior to liver transplantation when RV FAC was normal. However, RV GLS did not seem to correlate with post LT clinical course. The mechanism of RV dysfunction is likely due to increased venous return to the right heart caused by the development of portosystemic collaterals to counterbalance the increased intrahepatic vascular resistance to portal blood flow as well as elevated pulmonary arterial pressure that may further contribute to the dysfunction (1). Our study found also found that TAPSE, a parameter of RV function, correlates with post-transplant clinical course including post transplantation hospital length of stay, length of ICU stay and duration of mechanical ventilation thus, highlighting the importance of assessment of RV in patients with cirrhosis.
Overt systolic dysfunction has been observed in only 2% of adult liver transplant candidates (20). This is similar to our data that almost all patients had normal LV function as measured by EF and only one patient had abnormal LV GLS strain. Previous pediatric studies have shown that systolic strain was impaired in patients with cirrhosis (1). LV function by EF or strain did not correlate with post LT clinical course. However, none of the patients in our cohort required inotropes post LT, which might affect the results of our study.
Our study found a significant burden of diastolic dysfunction (DD) in pediatric patients with almost half of the patients having DD on pre-LT echocardiogram. Increasing numbers of recent studies have emphasized the importance of LVDF in predicting poor early and late outcomes after LT (17, 21, 22). Raevens et al. found incidence of DD in 43% in 173 adult liver transplant candidates (20). Our study showed almost 50% of patients had DD. This is in line with the existing evidence (23, 24). In our cohort the markers of DD in pediatric patients correlated with longer post LT hospital length of stay, length of ICU stay and duration of mechanical ventilation. In adults, combination of systolic and diastolic dysfunction has been shown to predict survival outcomes in patients undergoing LT (17). Thus, it becomes important to evaluate these parameters in patients. Our small sample size did not allow for assessment of mortality outcome, as no mortalities occurred in this cohort.
In our study, there was no correlation between LV systolic function and severity of disease. However, TAPSE and parameters of DD correlated with severity of liver disease as seen by natural PELD/MELD score, prothrombin time, INR and decompensated portal hypertension. Evidence of this is in contrast to a study of patients with cirrhosis where no significant association was found between echocardiographic changes and Child-Pugh score (25).
Pediatric liver wait-list morbidity and mortality is high overall but unpredictable in individual cases, resulting in poor stratification and general dissatisfaction with the current system of liver allocation in infants and children (26). In the absence of improved organ availability, new objective parameters for children are needed. The echocardiographic parameters and data from our study can be used to model further efforts in the stratification of cardiovascular risk and may change treatment decisions.
There are limitations to this study. It is a single-center retrospective analysis with a limited sample size and a significant heterogeneity in primary diagnosis. STE is dependent on image quality and strain values are subject to error, though they were performed in a uniform and standardized manner. However, our site has previously published data on good strain in patients with PVC with an inter-rater reliability value of 0.87 demonstrating that the values are reproducible (7). The measurement of LV and RV function by the area length method and fractional area shortening have their inherent assumptions and limitations. Only a single echocardiogram was reported for each patient, so the authors cannot extrapolate the data longitudinally. The retrospective nature of the study did not allow for apical 2-chamber and 3-chamber evaluations. We also used absolute values for TAPSE and not the TAPSE z scores as the z scores have not been validated in this cohort. In addition, clinically a lot of our patients had low PELD/MELD scores (range is from 6–40), indicating less severe liver disease, which may affect our findings and limit the application to all pediatric cirrhosis patients.