Echocardiographic Systolic and Diastolic Function Parameters Are Not Significantly Altered in Isolated Pediatric Excess Weight

Obesity is a classic risk factor for heart disease and the relationship between the two has been extensively studied in adults¹. The rationale for assessing possible effects of excess weight on cardiac function in children stems from the unique opportunity to obtain data from a population that will not typically exhibit additional, confounding risk factors. The possibility that these individuals could potentially develop premature cardiovascular disease also warrants attention.

Echocardiography presents itself as a widely accessible, non-invasive tool that can easily provide the necessary information. Given the increasing burden of pediatric excess weight, we employed echocardiography to assess the presence and extent of cardiac dysfunction in a cohort of excess weight children compared to normal weight controls.

Patients under 18 years of age who presented to the pediatrics or pediatric cardiology departments of „Dr. V. Gomoiu” Clinical Hospital for Children in Bucharest between 2016 and 2018 were evaluated for enrollment in the study. Excess weight (overweight or obese), otherwise healthy children and normal weight, healthy controls were included, subject to parents or carers and adolescents signing a dedicated consent form. Weight status was assessed according to the Centers for Disease Control and Prevention (CDC) criteria² using an online calculator³ and defined as: healthy weight (5^th percentile up to the 85^th percentile, included in the control arm), overweight (85^th to less than the 95^th percentile) or obese (equal to or greater than the 95^th percentile), both enrolled in the excess weight arm. Body surface area (BSA) was calculated using the Du Bois method⁴. Each visit comprised clinical examination, routine blood tests, total serum cholesterol and triglyceride levels, 12-lead ECG, 24 hour ambulatory blood pressure monitoring (ABPM) and echocardiography. Excess weight children were then scheduled for clinical visits at 6-month intervals and were followed up for one year.

Standard transthoracic echocardiography (2D, M-mode, colour and spectral Doppler) and Tissue Doppler Imaging were performed. A minimum of 2 consecutive cardiac cycles (usually 3 or 4) were recorded. We determined cardiac and great vessel dimensions, systolic and diastolic function parameters and performed an assessment of heart valves.

2D-guided M-mode was employed to calculate fractional shortening (FS) and the ejection fraction using Teichholz's formula. Left atrial volume was calculated using the area-length method. Left ventricular (LV) end-diastolic and end-systolic volumes were determined using the biplane disc-summation technique. LV ejection fraction was also calculated using the modified Simpson method. We calculated stroke volume and cardiac output. Diastolic function assessment was carried out by documenting mitral inflow and early diastolic flow propagation velocity (Vp). We employed tissue Doppler imaging (TDI) to record both systolic and diastolic velocities of mitral annular motion.

Stroke volume and cardiac output were normalized to body surface area and height to its age-specific allometric power⁵. MAPSE was normalized using an online calculator^6,7, avaliable at https://www.pedz.de/en/heart.html (accessed May 26^th 2020). Ratios (ejection fraction, E/A, E/Vp) were not indexed. All other parameters were normalized to body surface area using the Z-score calculator provided by Boston Children's Hospital⁸.

46 excess weight (overweight or obese), otherwise healthy children (mean age 10.31 ± 3.1, range 2–16) and 28 normal weight, healthy controls (mean age 13.06 ± 3.58, range 4–17) were enrolled in the study. 91.3% of participants in the excess weight arm were obese according to the CDC criteria. Baseline participant characteristics are depicted in Table 1; further details regarding blood pressure data may be found elsewhere⁹.

Most previous studies have identified some degree of diastolic dysfunction in obese children^{15,16,17,18,19,20}. Some authors have demonstrated changes in standard mitral inflow parameters (E/A, DT)^16,18,20 or pulmonary venous flow¹⁵, but combined parameters using TDI (E/e’, E’/A’) appear to be more sensitive. Existing evidence is less convincing when it comes to the systolic function. While there are papers which have reported minor subclinical changes in some systolic function parameters (myocardial performance index, stroke volume)^21,22, many others report no difference from normal weight children^18,19,20.

In our study LV systolic function (assessed by various means) was mainly preserved across both groups. Although a significantly higher proportion of excess weight children had a slightly lower ejection fraction (below 55%, Simpson's biplane method), other systolic function parameters (MAPSE, lateral and septal peak systolic tissue velocities) were normal or supranormal. Interestingly, with respect to peak systolic tissue velocities, there was a higher number of supranormal values in the normal weight group, although this did not reach statistical significance. Arguably, this data may suggest minimal impairment of LV function in some excess weight children that could potentially be demonstrated by more sensitive tests. Supporting this is the fact that both indexed stroke volume and cardiac output were higher in normal weight children.

Diastolic function was also normal when assessed using standard echocardiographic parameters (E/A, DT). With respect to combined mitral inflow and TDI parameters, E/e’ was abnormal in only 2 participants, both having excess weight. Only the septal E/e’ was abnormal in one participant, who also exhibited a low Vp and increased E/Vp; as for the other, both septal and lateral E/e’ were abnormal, but no changes were noted using color M mode. It is thus difficult to appreciate genuine diastolic disfunction. LV ejection fraction, as well as other systolic function parameters, were normal in both cases.

Early diastolic flow propagation velocity (Vp) was reduced in some cases without a corresponding increase in E/Vp ratio (with the single exception described above). Calculating E/Vp may thus be a more accurate method of evaluating LV diastolic function.

Finally, a significant number of children from both groups exhibited a longer IVRT when using age-appropriate (age under 20) cutoff values. This finding is difficult to interpret, as other diastolic function parameters were mainly normal and there was no echocardiographic evidence of an increase in left atrial pressure. Previous work¹³ suggests limited applicability of IVRT in children because of close correlation with heart rate, which is both higher and overly variable in this population (with age, emotional context etc.).

In this study, both systolic and diastolic function parameters were normal in most cases regardless of weight status. We were unable to exclude a minimal degree of left ventricular systolic impairment that could potentially be demonstrated by means of myocardial strain imaging or cardiac magnetic resonance. The few cases which exhibited modified diastolic function parameters are difficult to interpret, as these changes were not consistent, thus causing us to question true diastolic dysfunction in these children. Vp and IVRT do not seem to be well suited for diastolic function assessment in children based on our data.

These findings do not align with some of the previously published data as discussed above. This could be since our excess weight group included overweight as well as obese children. Moreover, participants in both groups were quite young (in the excess weight arm the average age was 10, while the youngest participant was only 3 years old), thus limiting the time of exposure to excess weight. It is indeed possible that more definite changes may appear in older children with a more severe weight issue. Measurements performed in this study are however mostly standard, at least for adult studies. Incorporating calculating the ejection fraction with a volumetric method, systolic and diastolic mitral annular velocities, analysis of the mitral inflow pattern and combined TDI parameters into a dedicated assessment protocol is not difficult, may yield minor changes in some individuals and can be augmented with more complex parameters if needed.

Standard echocardiographic systolic and diastolic function parameters do not appear to be significantly altered in isolated pediatric excess weight. However, a thorough examination (ideally following a standardized protocol) is advised in these children, as subtle changes may be identified in some, signaling a need for closer follow-up in these cases.