In this analysis of children undergoing chronic HD for ESKD with the typical range of underlying congenital and acquired kidney conditions as reported from the United States Renal Data System on ESRD in children and adolescents in 2018, we observed intradialytic changes in both systolic and diastolic function using novel STE markers. Both mid-HD and post-HD GLS significantly impaired from baseline. Progressive impairment in diastolic function was detected in terms of LV Ds at all three time points. Mid-HD LAS and Sr were impaired from baseline and both LAS and Sr returned to baseline at the end of HD.
Systolic Function
Similar to other studies, LVEF remained normal (≥ 55%) throughout HD in all of our patients. Prior studies in children also showed preserved LVEF when compared to healthy control and nonsignificant LVEF change during HD.7,8
Baseline GLS in children with ESKD and HD have been noted to be mildly impaired in 2 separate previous studies (-19.7 ± 2.8%8 and − 20.31 ± 3.58%7), which reported GLS similar to the pre-HD GLS in our study (-19.9 ± 1.9%). Our results showed 53% had abnormal pre-HD GLS and LV GLS worsened during HD with 100% had abnormal GLS at mid-HD and post-HD. Studies in adults demonstrated LV GLS was associated with ventricular arrhythmia or sudden cardiac death (area under the curve = 0.79), uremic toxins, obesity, and arterial stiffness. 5,17 In our study, post-HD GLS and intradialytic GLS impairment was associated with the degree of decrease in mean BP during HD with greater decrease in BP associated with more significant post-HD GLS impairment. There was a trend of more post-HD GLS impairment with greater BV removal, but this association was not statistically significant.
Intradialytic LV MDI change was not significant in our study with a baseline MDI measured 36 ± 9 vs 37 ± 9 msec after HD. Predialysis and dialysis adult patients with ventricular arrhythmia or sudden cardiac death had larger MDI 66 (55–74) vs 52 (43–65) msec (p = 0.04) with nonsignificant vintage difference between the 2 groups.5 Our patients’ median HD vintage of 13 months (IQR 9, 25) was not different from that was reported in this study.5 We hypothesized the difference in MDI between pediatric patients and adults were related to underlying etiologies and fewer comorbidities in children.
Left ventricular GLS was the only marker of systolic function that demonstrated intradialytic impairment in our study. The impaired GLS would not have been detected by conventional echo imaging as LVEF remained normal, indicating that more sophisticated noninvasive imaging techniques might be paramount to detect adverse impact of HD in children which likely occurs repetitively with each HD session in children.
Diastolic strain, DSrE, and E/DSrE
Diastolic dysfunction diagnosed by spectral and tissue Doppler imaging has been reported in children with ESKD requiring long-term HD.8,18,19 Diastolic dysfunction developed early in the course of children with chronic kidney disease and was present up to 18 months after renal transplantation.8 Studies of global Ds and Sr indices demonstrated myocardial diastolic dysfunction in adults with preserved LVEF, who received peritoneal dialysis.20,21 The E/DSrE ratio improved the prediction of cardiovascular mortality and major adverse cardiovascular events in HD adult patients over clinical and conventional echocardiographic parameters alone.6 In our study, intradialytic E/DSrE change was nonsignificant. We observed progressive intradialytic Ds and DSrE impairment which was not associated with BV removal or BP change. Our study demonstrated progressive intradialytic ventricular diastolic changes that are likely repetitive and potentially contributory to diastolic dysfunction in patients on chronic HD for ESKD, which may persist after transplantation.8 Evolving data in adults showed asymptomatic diastolic dysfunction with preserved LVEF is associated with subsequent heart failure and mortality; addressing these preclinical changes and echocardiographic surveillance may delay symptomatic progression and improve outcome.22 Longitudinal studies are necessary to understand how diastolic dysfunction plays a role in cardiovascular outcome of children who are on chronic HD. It remains to be explored whether management strategies including monitoring using these novel STE markers can be developed to slow down or prevent its progression to improve patient outcomes.
Left Atrial Strain and Sr
Left atrial strain reservoir has been identified as a new emerging reliable marker of diastolic dysfunction in adults; it better reflects instantaneous left atrial pressure with greater prognostic value than left atrial volume index.21 Left atrial strain is a sensitive marker of early diastolic dysfunction, linearly worsens to the progression of diastolic dysfunction, and improves as LV filling pressure improves.23 While the use of LAS has been studied in HD adults and found to be predictive of outcome,24 we found no prior study assessing LAS in pediatric chronic HD. In our study, LAS and Sr during the atrial early emptying (conduit) and atrial filling (reservoir) were impaired from pre-HD to mid-HD but returned to baseline at the end of HD. When combined with mitral inflow E velocity, the E/Ɛres increased at mid-HD and remained increased at post-HD. The E/Ɛres ratio has been identified as a useful parameter in the prediction of total and cardiovascular mortality in adult HD patients.24 Similar to Ds, these LAS and Sr changes in our study were not associated with BV removal or BP change. Left atrial contractile function changes during HD assessed by tissue velocity and Sr during late diastole were noted to be preload-independent in a study in 41 adult HD patients,25 which was similar to our observation in children in this study. Further studies are needed to understand the role of these intradialytic changes in the progression of diastolic dysfunction in children with chronic HD.
Children with ESKD are at risk of cardiovascular health issue, possibly related to their disease process, complications of the disease, and treatment which can lead to microvascular dysfunction.15 Our study results demonstrated progressive intradialytic LV diastolic dysfunction, which was not associated with the degree of fluid removal or BP change. It may have occurred as a natural history of the disease process. Addressing vascular dysfunction is valuable in adults to reduce long-term major adverse cardiovascular complications.15 We do not believe our results are sufficient to postulate pathophysiology related to cardiovascular complications seen in these children. However, they may serve as a proof that these novel tools can be utilized in clinical practice to measure the magnitude of subclinical function impairments and future research as we strive to understand the disease process, minimize harm, and improve the care and outcome of children with ESKD requiring HD.
Our study was a secondary analysis of a prospective study and therefore has limitations. The sample size, as typical of most pediatric dialysis studies, was small. We were only able to assess patients during one HD treatment and were not able to follow them longitudinally post treatment to assess whether the observed abnormalities resolve during the recovery period after HD. We did not have access to real time hemodynamic indices such as continuous BP or cardiac output monitoring. However, none of our patients had overt cardiovascular risk factors, all had uncomplicated asymptomatic treatments, and relatively short vintage, underscoring the adverse hemodynamic burden of HD on organ function.