The study was reviewed and approved according to the guidelines of the Declaration of Helsinki by the responsible Ethics Committee of the Ludwig-Maximillian’s-University Munich (Munich, Germany) (Project no: 21-0165, date of approval: 05.07.2021) The data protection was observed in accordance with the guidelines of the Ethics Commission.

This study is a clinical trial and diagnostic study. The group selection was not randomized and not blinded. The results were collected prospectively, and all study methods were standardized and performed by two investigators. Fontan Patients were recruited and invited to participate in the study from the registry of the Department of Paediatric Cardiology and Paediatric Intensive Care of the LMU. A total of 25 patients with Fontan circulation were recruited for the study (group 1). To show the effect of food intake and respiration at normal physiological conditions, the healthy comparison group consisted of 50 volunteers (group 2). To assess the effect of age on liver stiffness, both groups were divided into two subgroups of different ages (group 1a: 10–19 years; group 1b: 20–29 years; group 2a: 15–19 years; group 2b: 20–25 years) (Figure 1). All participants were informed in detail about the procedure and background of the study in advance and gave written informed consent.

Subjects aged between 10 and 30 years, with Fontan completion were recruited for group 1. There were no specific exclusion criteria for group 1. For group 2, healthy subjects aged between 10 and 30 years were recruited. Here for exclusion criteria were liver disease (e.g., liver fibrosis, liver cirrhosis, infection, and liver tumor), other vascular diseases, diseases of the cardiovascular system (e.g., cardiac arrhythmias, hypertension, and heart failure), metabolic diseases (e.g., hyperthyroidism and disorders of sugar metabolism), diseases of the gastrointestinal tract (e.g., gallbladder diseases, gallstones, and bile duct obstruction), diseases of the kidneys, diseases of the nervous system, mental diseases, regular drug use including nicotine and alcohol, obesity, subjects with regular medication use (exception: contraceptives), and pregnant women.

After consent was obtained, vital signs and biometric data were recorded. In addition, blood flow in the hepatic vessels and the diameter of the inferior vena cava were determined in all subjects using a PW Doppler (Philips).

Subsequently, liver stiffness was measured once before food intake (=T0, fasting values) using elastography (FibroScan^® Echosens Paris, Philips). This was followed by ingestion of the liquid food, whose nutritional values correspond to a standardized meal (14, 15, 17). Liver stiffness was also measured 15, 30, 45, 60, 90, 120, 150, and 180 min after food intake. Liver stiffness was measured ten times at each time point in inspiratory and expiratory hold (2–4 min of measuring per time point). For the examination, the subjects were in supine position during the examinations with the transducer being placed in the right intercostal spaces.

For the simulation of food intake, a 500 mL cocoa drink was used, based on comparable studies. The nutrient distribution is as follows: 600 kcal, 54% carbohydrates, 17% proteins and 29% fats. This distribution corresponds approximately to an average meal. This composition was based on previously published nutritional standards (14, 15, 17). For Fontan patients with bronchitis plastica, an alternative variant with low-fat milk and MTC oil was offered. The exact nutritional values can be found in Table 1.

General medical history (previous illness/diagnosis, age).

Statistical analysis was performed using the Statistical Package for Social Sciences software (IBM SPSS version 26.0.0.1). In the context of descriptive statistics, mean and standard deviation were used to describe quantitative characteristics. Where values were not based on a normal distribution, median and interquartile range were used. In a T-test for unrelated target variables and a repeated measures ANOVA, values of inspiration and expiration were compared. For comparison of fasting and postprandial liver stiffness values a T-test for linked target values was applied. Statistically significance was set to a two-sided value of p <0.05. To all results the Bonferroni correction was applied.

A total of 25 Fontan patients and 50 healthy subjects were included in the data analysis. The distribution of patient characteristics and vital signs are shown in Table 2. There were significant differences in the categories weight (p = 0.008), heart rate (p = 0.046) and oxygen saturation (p < 0.001). Liver elastography was successfully performed in each participant. Liver stiffness is expressed in units of kPa.

In general, values for liver stiffness were significantly different in group 1 and 2. At T0 in inspiration there was a difference of 9.01 kPa [T (24.7) = 9.36, p = <0.002] while at T0 in expiration there was a difference of 10.54 kPa [T (24.9) = 10.69, p = <0.002] (Table 3; Figure 2). In group 1 our data showed a higher standard deviation than in group 2.

In group 2 there was a significant difference between the values measured in inspiration and expiration at T0 (fasting value). Liver stiffness at inspiration was measured at 3.97 kPa, while at expiration it was measured at 4.48 kPa [difference 0.51 kPa = 11.3%, T (98) = −2.9, p = 0.045]. For the other times of measurement, the T-Test showed no significance between inspiration and expiration (Tables 6, 7; Figure 2).

Overtime, the difference between inspiration and expiration was analyzed in a repeated measures ANOVA. In group 2 there was a significant change of the difference between inspiration and expiration between T0 and T30 (p = 0.003) and T0 and T120 (p = 0.045) [F (5.3, 259.32) = 3.55, p = 0.003].

Meanwhile, in group 1 there was no significant difference between the values for liver stiffness in inspiration and expiration at any time point of the examination [e.g., T0 insp = 12.98 kPa; T0 exp. = 15.01 kPa; difference 2.03 kPa = 13.5%, T (−1.5) = 48, p = 0.145] (Tables 6, 8; Figure 2).

The repeated measures ANOVA in group 1 also showed no significant changes in the difference between inspiration and expiration overtime.

The main groups were divided into two subgroups of different ages. Thus, the effect of age on liver stiffness (T0) could be determined in Fontan patients and in healthy subjects. The comparison of liver stiffnesses during fasting in group 1a (14.2 kPa) and in group 1b (13.7 kPa) showed no statistically significant difference (Figure 7). The liver stiffnesses of group 2a (4.4 kPa) and group 2b (4.1 kPa) was also not significantly different. In addition, the regression analyses did not reveal a statistically significant correlation between liver stiffness and age in any group (group 1: R2 = 0.011; group 2: R2 = 0.057).

Due to the lack of a subpulmonary ventricle, Fontan patients routinely present with increased central venous pressure and blood backing up into the liver. These factors can lead to complications such as Fontan-associated liver disease (FALD), bronchitis plastica, protein-losing enteropathy, and even failing Fontan during its course (18).

Beside the gold standard, i.e. liver biopsy, liver ultrasonography is used in the follow-up of Fontan patients as a diagnostic procedure for liver fibrosis and cirrhosis. It is a non-invasive and safe procedure that measures liver stiffness sonographically (10). In patients with a healthy heart and circulation, this procedure has been adequately evaluated for accuracy, confounding variables, and utility (19). The results of liver stiffness and biopsy were compared in 100 subjects by Jain. et al. Here, liver elastography was found to have a sensitivity of 90.3% and a specificity of 89.5% in the presence of significant fibrosis (F > 2) (20). Also, Castera et al. and and Foucher et al. defined cutoff values for the stages of fibrosis based on the METAVIR Fibrose-Score. The described values were for F0/F1 = 2.5–7.2 kPa, F2 = 7.2–12.5 kPa, F3 = 12.5–17.6 kPa, F4 = > 17.6 kPa (21, 22). Among the confounding variables identified are food intake, respiration, and age. For example, the research by Berzigotti et al. showed a postprandial increase in liver stiffness from 40.7 kPa to 51.2 kPa 30 min after food intake (15). For this reason, it is currently recommended that liver elastography is performed in a fasting state (23, 24). In addition, Yun et al. found a significant difference in liver stiffness between expiratory and inspiratory phases (exp: 8.7 kPa vs. insp: 7.9 kPa) in their study of 123 subjects (16). The effect of age on liver stiffness is weighted differently in current studies and remains incompletely elucidated (25, 26). In Fontan patients, neither the influence of age nor the influence of respiration and food intake on liver stiffness has been validated so far.

To address these questions, our study examined 25 patients after Fontan completion by liver elastography in the fasting state and after ingestion of a standardized meal and in the inspiratory and expiratory positions. For comparison, a group of 50 healthy subjects was examined under the same conditions.

The healthy subjects showed a typical postprandial increase in liver elastography values that was significant for the first 45 min during inspiration measurements and for the first 30 min during expiration measurements. The maximum was reached after 30 min (T30 = 4.9 kPa) in inspiration and after 15 min (T15 = 5.1 kPa) in expiration position. This corresponds to relative changes of 22.5% and 14.7% and indicates a clinically relevant difference. In contrast, the decreases in liver stiffness during expiration measurements at the time points T90 (6.5%), T120 (7.8%), T150 (6.9%), and T180 (8.0%) were clinically negligible. The increase in liver stiffness after meal ingestion has been described in other studies and can be explained by increased postprandial blood flow to the liver (14, 15, 23). Thus, also Alvarez et al. showed an increase from 7.8 kPa to 10.3 kPa 30 min postprandially in adult patients (23).

In contrast, the group of Fontan patients did not show a significant change in liver stiffness at any time point after food intake, neither in inspiration nor in expiration measurements. These results can be explained by the fact that in Fontan patients, due to chronic blood stasis in the liver, postprandial hyperemia cannot contribute to increased liver stiffness. The study presents the first scientific documentation of the lack of increased liver stiffness following food intake in Fontan patients.

For each of the nine measurement timepoints, liver stiffness was evaluated in inspiration and expiration. Physiologically, during inspiration the thoracic and intrapleural pressure lowers by expansion of the thorax through the breathing muscles, which creates an increase of venous return to the right heart (27). By Yun et al. it was reported that this increase in venous return leads to a significantly higher flow velocity in the hepatic veins than during expiration, where thoracic pressure rises leading to a decrease of blood flow to the heart and thereby leading to an elevation in intrahepatic blood volume which ultimately results in an increase of liver stiffness, that can be measured (12, 16).

As seen in our study, values for liver stiffness in expiration were generally higher than values measured in inspiration, which is consistent with results of other studies (12, 16). Values measured in group 2 showed a significant difference between inspiration and expiration at T0 (T0 insp = 3.97 kPa, T0 exp. = 4.48 kPa, difference 0.51 kPa). Except for the fasting value, no other measurement time point showed any significant difference. This could be influenced by the additional effect of meal ingestion on liver stiffness.

The Fontan patients in group 1 showed no significant difference between inspiration and expiration at any point during the study. Since the Fontan’s physiology leads to a chronically increased Central Venous Pressure it is possible that the physiological pressure gradients during inspiration and expiration do not affect the liver stiffness in Fontan patients. However, the lack of a significant difference between the values for inspiration and expiration in group 1 could also be due to the higher standard deviation. For a more detailed evaluation of this, further studies with a bigger sample size need to be performed.

These results could mean that liver elastography does not have to be performed in a specific phase of respiration in Fontan patients and is therefore also applicable to smaller children and patients who cannot follow breathing commands, with the results still being reliable.

No significant difference in liver stiffness was shown between the subgroups. In our patient collective, there was no correlation between age and liver stiffness. This correlation is evaluated differently on healthy subjects in the current data (25, 26, 28). Tokuhara et al. studied preschoolers, elementary school students, and adolescents and described a statistically significant correlation between age and liver stiffness. The values increased from 3.4 kPa to 3.8 kPa and 4.1 kPa (25). In contrast, Sirli et al. described no age differences with respect to liver elastography results (28). However, no Fontan patients were studied here. In subjects with Fontan circulation, one suspects an increasing liver stiffness during life due to the chronic progressive hemodynamic influences. Here for further studies with a larger patient population of Fontan patients are necessary.

There are several limitations to this study. A total of 25 Fontan patients were recruited from the registry of the Department of Pediatric Cardiology and Pediatric Intensive Care of the LMU. This number can be considered relatively small. Therefore, no statements could be made about the differences in the subtypes of the univentricular heart disease. The sample size calculation was based in values from the available literature. Since there are no studies on changes in liver stiffness during food intake and respiration in Fontan patients, the standard deviation was chosen smaller than what was later found in our data. Because of the small prevalence of Fontan patients, multicenter studies are needed for this. In addition, the amount of the drink was not adapted to the characteristics of the subjects. Younger patients with a lower BMI may respond differently to food intake than older patients with a higher BMI. Furthermore, the breathing maneuvers were performed by the subjects themselves and could therefore not be accurately compared.