#1784 The potential impact of chronic subclinical hypervolemia on left ventricular hypertrophy in children on dialysis: a prospective longitudinal study

Abstract Background and Aims Hypervolemia poses a significant challenge in pediatric patients undergoing dialysis, potentially leading to severe complications such as left ventricular hypertrophy (LVH) and cardiac dysfunction. Accurately assessing fluid status in growing children on dialysis could be particularly difficult, and although several non-invasive methods are emerging, each individual measurement has its own strengths and diagnostic limitations in terms of sensitivity and specificity. To date, only a few studies in children have compared the use of different methods simultaneously to detect fluid status, and none of these have evaluated the potential cardiovascular impact of a prolonged subclinical hypervolemia. The latter can be defined as a condition of euvolemia at clinical assessment but hypervolemia at technical evaluation. We evaluated the role of three methods—lung ultrasound (LUS), bioimpedance spectroscopy (BIS), and ultrasound assessment of inferior vena cava collapsibility index (IVC-CI)—in detecting subclinical hypervolemia and examined the cardiac implications. Method In this prospective observational longitudinal study, we included 23 children (<18 years) who were receiving chronic hemodialysis or peritoneal dialysis at the regional pediatric nephrology center of Meyer University Children's Hospital IRCSS, Florence, Italy, between 01/06/2018 and 01/11/2020. We excluded patients with clinical signs of fluid overload. Each patient underwent a multiparametric evaluation of their fluid status (physical examination, LUS, IVC-CI, BIS) every 2 months, and an echocardiography every six months starting from a baseline evaluation. The LVH was defined as a left ventricular mass index (LVMI) of ≥ 95th percentile or a left ventricular wall-thickness z-score > 2 z-score. We divided the patients into two groups and arbitrarily assigned them to the subclinical hypervolemia (SH) group if they had at least two out of three parameters of fluid overload at different methods (≥5 B-lines at LUS, overhydration normalized to extracellular water “OH/ECW” > 7% at BIS, and/or IVC-CI < 40%) in the absence of clinical signs of fluid overload. The others were included in the euvolemic (EU) group. The overload rate by weight (clinical OH), prescribed by the physician clinically, was calculated as the proportional increase from the dry weight (kg of fluid overload / dry weight in kg × 100). Results The two analyzed groups did not differ regarding gender, age and BSA (Table). The clinical OH was significantly higher in the SH population than in the EU group (2.2 ± 1.84 vs 1.30 ± 2.16%, p < 0.01). In the correlation analysis, we found that clinical OH positively correlated with the number of B-lines by LUS (R = 0.2923, p < 0.001) and OH/ECW by BIS (R = 0.4144, p < 0.001), while there was a negative correlation with IVC-CI (R = −0.2597, p = 0.019). However, each method taken individually (LUS, BIS, and IVC-CI) appeared to be inaccurate in predicting fluid status in several individual cases. Although the frequency of hypertension did not differ between the SH and EU groups, uncontrolled hypertension, treatment-resistant hypertension, and hospital admission for hypertensive crises were more frequent in the SH group, with the latter being statistically significant (67% vs 0%, p < 0.01). Furthermore, in the entire cohort, we observed a significant linear correlation between LVMI values and the average values of pre-HD systolic blood pressure recorded in the prior six months (R2 = 0.16, p = 0.002). It is worth noting that the presence of LVH at the last follow-up was significantly more prevalent in the SH group (75% vs 27%, p = 0.04). Conclusion Chronic subclinical hypervolemia, detected with a multi-method evaluation (including LUS, BIS, and IVC-CI), was significantly correlated with clinical OH and associated with a higher risk of LVH. To mitigate the intrinsic limitations of each technique, adopting two or more methods simultaneously can bypass and minimize the risk of evaluative error in assessing fluid status, and help reduce cardiovascular risk.