A literature search was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [21] using MEDLINE database. Search terms were defined to identify literature on PRA in healthy children as well as in children with heart failure. In November 2021, the search term “(plasma renin activity) AND (Paediatric OR newborn OR infant OR toddler OR child) AND (Heart failure OR dilated cardiomyopathy OR congenital heart defect OR congenital heart disease)” was utilized to ascertain literature on PRA in children with heart failure. In January 2022, the search term “(plasma renin activity) AND (Paediatric OR newborn OR infant OR toddler OR child) AND (healthy OR “control group”)” was used to detect literature on PRA in healthy children. The following filters were set for both searches: Humans, English, German, Child: birth – 18 years.

The inclusion criteria were set as follows. Studies were included if they provided PRA in healthy children or children with heart failure from birth to 18 years of age. In addition, for better comparability, PRA had to be reported in the study as arithmetic mean $\pm$ standard deviation (SD) or standard error (SE). Due to the smaller amount of data in children with heart failure, studies in which PRA was reported as arithmetic mean and range were also allowed. Age had to be reported in the study as arithmetic mean $\pm$ SD or SE or alternatively as range. SE was converted to SD for the graphs and tables. Healthy children were allowed to have no or only mild diseases without known influence on the RAAS. Heart failure was allowed to be due to congenital or acquired heart defects.

The exclusion criteria were set as follows. Studies were excluded if age, type of statistical parameters used, or health status were not accurately reported. In addition, studies in which PRA was measured in preterm infants, in fetal blood or in umbilical cord blood were excluded. Further exclusion criteria were if only stimulated PRA or only renin concentration was measured. As the influence of ACEi treatment was to be investigated, studies on children with heart failure were excluded if no information on therapeutic medication was provided. Moreover, studies on children with heart failure were excluded if only postoperative or intraoperative PRA was measured. Further details are provided in the PRISMA flow diagram (Fig. 1).

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of the conducted literature search.

In the first step of study selection, the titles and abstracts of all records identified with the above search terms were screened. Records that did not meet the inclusion criteria or fulfilled an exclusion criterion were excluded at this stage. Subsequently, the full text of the remaining records was sought. Reports for which neither a printed nor a digital version of the full text was available could not be considered further. In the next step, all available full texts were screened. Reports that did not meet the inclusion criteria or fulfilled an exclusion criterion were excluded. Finally, all studies that were not excluded and met the inclusion criteria were included in the review.

To obtain an overview of the RAAS in children, a non-systemic literature search for all RAAS parameters was conducted as a preliminary search before the systemic literature search. Additional publications found in the preliminary search that met all inclusion criteria were also included.

After inclusion of all suitable publications, it was evaluated how many publications contain information about PRA in young healthy children aged up to two years.

In addition to the literature review, data from the European LENA project (EudraCT 2015-002335-17; EudraCT 2015-002396-18) were analysed [22]. In the LENA project, enalapril in the form of orodispersible minitablets (ODMT) was tested in 102 children with heart failure due to congenital heart disease (CHD) or dilated cardiomyopathy (DCM). Of 102 subjects, 35 subjects were ACEi naïve and were included in our analysis. The dosing regimen of the LENA studies was established using a physiologically based pharmacokinetic (PBPK) simulation. It includes age- and weight-dependent doses which were predicted to result to similar enalapril and enalaprilat exposures as observed in adults for a start dose of 2.5 mg and a maintenance dose of 20 mg [23]. Daily doses $\le$0.25 mg were administered once daily in the morning and daily doses $>$0.25 mg were divided into two equal doses, each administered in the morning and evening.

In addition to heart failure therapy, other drugs were also administered during the observation period that are not expected to influence PRA. Other concomitant medications were antiplatelet drugs, ampicillin/sulbactam, cephalosporins, chloral hydrate, folic acid, heparins, ibuprofen, iron supplement, levothyroxine, meropenem, methylprednisolone, morphine, palivizumab, paracetamol, polyethylene glycol, potassium, prednisone, ranitidine, red cell concentrate and vitamin D3.

As part of the pharmacokinetic/pharmacodynamic (PK/PD) studies, blood samples had been collected and analysed for PRA levels before, 4 hours after and within the first 8 days of enalapril treatment. Blood was collected in a cooled ethylenediamine tetraacetic acid (EDTA) tube, carefully mixed and immediately centrifugated under cooled conditions (0–4 °C). After centrifugation, the supernatant was transferred into a cryo tube and was stored at –80 °C until analysis. The sample was taken in supine position and, if possible, when the children were quiet. It was advised to collect the blood sample before 10:00 AM. If this time could not be kept, the sample collection should always take place around the same time to minimise the influence of the circadian rhythm as much as possible. Resting time and behaviour during sampling (relaxed, moving or crying) were noted. PRA was determined by using an validated in-house customised enzyme-linked immunosorbent assay (ELISA) [24]. Further information on the study procedure can be found in the study protocol [22].

For the evaluation of the PRA level, the healthy children from the literature were divided in four groups after visual inspection of the data. In the visual inspection, age ranges in which the extent and variability of PRA was similar were defined as one age group. Based on mean age or centerpoint of age range of the study group, results reporting mean PRA were summarized into four groups of age ranges: Neonates up to 30 days of age, infants from 1–24 months, children from 2–10 years, and children and adolescents older than 10 years. For each age group, the weighted mean of the reported mean PRA was calculated as an overall approximation. The number of PRA measurements was used for weighing. The weighted mean of the four age groups was compared by calculating the percentage change between the groups.

As PRA and age were not normally distributed at all time points in the LENA studies, we reported the median and where appropriate the range for our analyses. For comparison with the literature data, we calculated mean and SD of age and PRA to have better comparability with the literature data, which were all available as mean $\pm$ SD or mean and range. To compare the 35 ACEi naïve subjects with the literature, they were divided into four age groups (1 month, 1–3 months, 3–6 months and $>$6 months).

Since the conditions regarding normal distribution for the application of the parametric tests for dependent and independent samples were not fulfilled, nonparametric tests were performed.

To analyse the effect of ACEi on PRA, PRA levels before, after 4 hours of enalapril treatment and within the first 8 days of enalapril treatment were compared. For this purpose, the Friedman test for more than two paired samples was conducted. After that, the Wilcoxon test for paired samples was conducted to compare the PRA before and after 4 hours of enalapril treatment as well as the PRA before and within the first 8 days of enalapril treatment. For both the Friedman test and the Wilcoxon test for paired samples, data from those children who had a complete data set of three PRA measurements were used (n = 29).

The PRA of the asymptomatic children and children with symptomatic heart failure was compared before and within the first 8 days of enalapril treatment using the Wilcoxon test for unpaired samples. Similarly, the age of the asymptomatic children and the children with symptomatic heart failure was compared using the Wilcoxon test for unpaired samples. p values 0.05 were considered as indicator for statistical significance. If not otherwise indicated, all data mentioned were expressed as mean $\pm$ SD. Statistical analyses and graphics were produced with R software version 4.0.5 (The R Foundation for Statistical Computing, Vienna, Austria) and OriginPro 2021b version 9.8.5.201 (OriginLab Corporation, Northampton, MA, USA).

The literature search yielded 29 publications on PRA in healthy children (Fig. 1). The plasma renin activities reported in these 29 publications were from a total of 1482 healthy children. A total of 14 of the 29 publications provided values of PRA from healthy children younger than or equal to two years of age. Overall, 344 of the 1482 healthy children were younger than or equal to two years old. Visual inspection of the literature data revealed roughly four groups: Neonates up to 30 days of age, infants from 1–24 months, children from 2–10 years, and children and adolescents older than 10 years. The results of PRA level evaluation in healthy subjects showed a decrease of PRA level over age (Fig. 2, Table 1, Ref. [5, 7, 8, 12, 13, 14, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]). The overall approximation of mean PRA levels for the four age ranges were 15.4, 11.8, 3.5, and 2.2 ng/mL/h. Compared to PRA levels of neonates up to 30 days of age, this means a drop of 23% in infants 1–24 months, 77% in children 2–10 years and 85% in children and adolescents older than 10 years. Comparing neonates with older children, PRA is up to 7 times higher in neonates than in older children. The percentage decrease in PRA between the different age groups is highest between infants aged 1 to 24 months and children aged 2 to 10 years. The highest mean PRA was determined by Vincent et al. [27] for 16 children between 6 and 30 days of age at 29.8 $\pm$ 28.6 ng/mL/h. PRA levels were lower in children and adolescents older than 10 years. Nevertheless, the PRA in children and adolescents older than 10 years was still slightly above the values reported in adults. For example, Van Acker et al. [28] found a PRA of 0.96 $\pm$ 0.6 ng/mL/h in 20 adults aged 22 to 45 years. In the first two years of life, the reported PRA levels partially varied strongly between the different studies.

Fig. 2.

Age-related change of plasma renin activity (PRA) in healthy children from birth to 18 years of age. PRA is expressed as mean $\pm$ standard deviation (SD). Age is expressed as $○$: Mean $\pm$ standard deviation (SD), ⚫: Mean and range or ▲: Centerpoint of range and range. A gray shaded box highlights PRA levels in healthy children from 1 to 18 years.

Of the three influencing factors analysed, age has the greatest influence on PRA in childhood according to our analyses. The comparison of the results of the different studies in the systematic review confirms previous separate observations on the decrease of PRA with age [12, 13, 14] and furthermore provides information on the extent of the age-related decrease of PRA in healthy children.

Considering the whole childhood, our analyses show a strong decrease of PRA by up to 85%. Such a strong decrease with age in childhood is not extraordinary and is also known for example for NT-proBNP [52].

The reasons for the increased PRA in children and especially in neonates and infants are not known with certainty. It is known that renin release is promoted by a decrease in renal perfusion pressure [53]. Consequently, the increase in blood pressure with age in children [54], could be a possible explanation for the age-related decrease of PRA in childhood.

Neonates and infants up to two years of age not only have the highest PRA levels, but also the greatest variability between the different studies. Possible reasons for the variability of PRA between the different studies could be differences in sampling or sodium intake of the subjects as well as crying during sampling. It is unlikely that the variability is due to the influence of a different position during sampling [7] or a different time of sampling [8, 9] in the studies, as in most of the included studies with children under two years of age, sampling was performed in the morning in the supine position. Different sodium intake could be one reason for the variability between the studies, as other investigations show an inverse correlation between PRA and sodium intake [31, 55]. As sodium intake was rarely assessed in the included studies, differences in sodium intake between the studies could possibly explain parts of the variability between the different studies. In addition, crying during sampling increases PRA [56]. Neonates and infants are more likely to cry during blood collection than older children. A different proportion of crying subjects in the different studies could be another reason why the variability between the different studies is high for children under two years of age.

The second factor influencing PRA is heart failure, which appears associated with 3–4 times higher PRA levels in neonates and young infants compared to healthy peers. Children with heart failure younger than 6 months from both the literature and the LENA studies had higher PRA levels than healthy children of the same age.

The activation of the RAAS in children with heart failure is part of the pathophysiology [4]. Consistent with our results in children with heart failure younger than 6 months, Nijst et al. [57] found about 5 times higher PRA in adults with chronic heart failure on optimal medical therapy compared with healthy controls. Anand et al. [58] reported a 9.5 times higher PRA in adults with severe clinical congestive cardiac failure. The greater difference between patients and healthy subjects in the latter study could be due to the fact that the subjects of Anand et al. [58] were completely untreated patients, while LENA subjects had been pretreated with various heart failure medications other than ACEi.

We assume that the PRA in children with heart failure is influenced by the severity of heart failure. The influence of the severity of heart failure on PRA is supported by the fact that our subjects with symptomatic heart failure (Ross score $\ge$3) had a significantly higher PRA than our subjects with asymptomatic heart failure (Ross score $\le$2). However, we could not find any association between the different severities of symptomatic heart failure classified by the Ross score and PRA. One reason for this could be that the Ross score includes various symptoms that are not all equally related to the RAAS. Of course, children with symptomatic heart failure receive more intensive therapy than children with asymptomatic heart failure. Accordingly, the dosage of furosemide was higher in the children with symptomatic heart failure. Since furosemide increases PRA [15], it is likely that current concomitant medication, in addition to the severity of heart failure, contributed to the fact that the children with symptomatic heart failure had a higher PRA. Another indication of the influence of the severity of heart failure on PRA is the fact that Buchhorn et al. [6] found a correlation of PRA with respiratory rate and an inverse correlation of PRA with weight gain in children with heart failure.

Data from the literature and the LENA studies indicate that PRA levels, similar to healthy children, tend to decrease with age also in patients with heart failure. However, the amount of data available and the age range of children with heart failure were too small to interpret the influence of age in children with heart failure more precisely.

A surprising finding was that the oldest subjects with heart failure from the literature and the LENA studies even had PRA levels in the range of PRA of healthy children. We suggest two reasons why PRA is not increased in older children with heart failure. The first reason could be that the older children have milder heart failure, as patients with more severe clinical symptoms due to CHD are usually operated earlier. The subjects in the oldest study group (aged 6 $\pm$ 2 months) from the literature had a respiratory rate 50 min${}^{-1}$, indicating milder heart failure considering their age [6]. In addition, the oldest group in the LENA studies had a Ross score that led to the classification as mild heart failure patients. However, the LENA patients with an age of 3 to 6 months and an increased PRA compared to healthy children were also classified as having mild heart failure according to the Ross score. Thus, there must be a second factor influencing PRA. The second reason could be that older children are more likely to have been stabilized on their medication. Furosemide [15], spironolactone [59] and milrinone [60] increase PRA whereas beta-blocker [17, 19] and digoxin [18] decrease PRA. Consequently, a lower proportion of furosemide, spironolactone or milrinone and/or a higher proportion of beta-blocker or digoxin in the current medication of the respective study group could also lead to a lower PRA. The oldest study group in the literature also had the lowest percentage of diuretics in their current medication compared to the other groups in this study, which had higher PRA levels [6]. The oldest group of LENA subjects received the lowest percentage of concomitant medication that increased PRA and the highest percentage of concomitant medication that decreased PRA. Therefore, PRA in children with heart failure is probably influenced by both the severity of heart failure and the current heart failure medication.

The third factor influencing PRA is the ACEi treatment, which increases PRA by 4.5-fold in children with heart failure. Thus, our results suggest that treatment with an ACEi may have an even greater influence on PRA than heart failure itself.

Previous studies on PRA after captopril administration in children with heart failure either did not find a significant increase in PRA [61, 62] or did not state whether the increase in PRA was significant [50, 51]. The studies that did not find a significant increase investigated only a small number of 8 respectively 12 subjects [61, 62]. In contrast, our analysis is based on a considerable higher number of 29 subjects.

In accordance with our results in children with heart failure, enalapril administration significantly increases PRA by approximately a factor of four in adults with heart failure [16]. Compared to adults with heart failure, the increase in PRA in children with heart failure was not yet significant 4 hours after administration, but only within the first 8 days of treatment. The delayed effect of enalapril in children with heart failure could have several reasons. One reason could be the higher starting dose used in the adult study compared to the starting dose in the LENA studies. In the study on enalapril in adults with heart failure a high dose between 10 and 40 mg was used as the first dose. The children in the LENA study received a dose comparable to a starting dose of 2.5 mg enalapril in adults. Another reason may be that the maximum enalaprilat concentration is reached later in children under one year of age. The pharmacokinetic analysis of the LENA studies revealed that the subjects younger than one year had a median time to reach maximum enalaprilat concentration of 6 hours [23]. Whereas in subjects older than one year, the median time to reach maximum enalaprilat concentration was 4 hours [23], comparable to that of adults [63]. Since 28 of the 29 subjects analysed were younger than one year, the time of PRA measurement in these children may have been too early to observe a significant increase in PRA. As a third reason, the PRA increase may be due not only to a direct negative feedback of angiotensin II on renin release but also to an increase in renin synthesis, as treatment with enalapril for several days causes an increase in renin mRNA in rats [64]. Finally, the number of subjects may have been too small to observe a significant increase in PRA after 4 hours in children with heart failure.

As concomitant medication could have an impact on PRA during the observation period, we investigated duration, changes and discontinuations of concomitant medication. Most of the concomitant medication was taken at least three days before the start of enalapril administration. It can therefore be assumed that the majority of subjects and their PRA had stabilized on their therapy before enalapril was administered. Since furosemide [15] and milrinone [60] increase PRA, reducing the dosage of furosemide and discontinuing milrinone therapy in one subject may at most have attenuated the observed increase in PRA due to enalapril. As the increase in furosemide dose in one subject only compensates for the reduced bioavailability due to the switch from intravenous to peroral administration, this change is not expected to have any effect. For the above reasons, we consider an impact of concomitant medication on PRA during the observation period as unlikely.

Of importance is the fact that the significant difference in PRA between asymptomatic children and children with symptomatic heart failure disappeared under enalapril therapy. The reason for this could be that, according to our results, the influence of ACEi treatment appears to be greater than the influence of heart failure itself. However, the comparison between asymptomatic children and children with symptomatic heart failure is based on a small number of eight asymptomatic patients, and the improvement in heart failure with therapy may have attenuated the increase in PRA differently in the two groups.

The existing studies on PRA after captopril administration in children with heart failure had to be excluded from the systematic review due to missing information on PRA values [61], contradictory information on the unit [62] or imprecise or missing information on age [50, 51]. Therefore, to our knowledge, this is the first report of PRA after ACEi administration in children with heart failure with exact age information and the first report of PRA after enalapril in children with heart failure ever.

The systematic review may not include all publications on PRA in healthy children and children with heart failure due to limitations of the method. Only publications listed in the MEDLINE database in English or German were considered. Publications in English and German for the literature search on PRA in healthy children and in children with heart failure, however, encompass the majority of all publications under the search terms used, with 92% and 90% respectively. Publications in which parts of the search terms are not mentioned in the title or abstract may have been overlooked. For healthy children, six additional publications from the preliminary search were included. In three of these cases were only the title and not the abstract available in the MEDLINE database. In the other three cases, other keywords were used for healthy children, for example, they were named normal children. Due to the nature of the literature data, it was not possible to analyse the literature data beyond descriptive statistics. Our relatively strict inclusion criteria led to the exclusion of some studies. Excluding studies with inaccurate age information reduced the data set, but as these analyses show, accurate age information is necessary to compare PRA values. Specifying the allowable statistical parameters in the inclusion criteria resulted in a smaller data set but improved the comparability of the data. Despite these limitations, the systematic review contains a large amount of data on PRA from almost 1500 healthy children and almost 60 children with heart failure.

Our age classification for the evaluation of the PRA level of the healthy children from the literature is based only on the visual inspection of the data. Both the age group of children from 2–10 years and the age group of children and adolescents over 10 years encompass a relatively wide age range. However, the age classification chosen is comparable to the age classification of the European Medicines Agency and the US Food and Drug Administration, where the age groups are for children from 2–11 years and for adolescents from 12–18 years.

The analysis of PRA in children with heart failure from the LENA studies is faced with limitations. One limitation is that the number of subjects that could be analysed for the effect of enalapril on PRA was limited because only a part of the subjects was without ACEi pretreatment. However, almost all ACEi naïve children were under one year old. Considering this very young age, a large number of very young subjects with heart failure were studied. Furthermore, PRA was analysed in children with heart failure of different aetiology. However, due to the small number of three ACE naïve children with DCM, an analysis regarding a possible influence of the aetiology of heart failure was not possible. Data from the LENA studies cannot provide information on the prognostic value of PRA in children with heart failure. The number of 35 ACE naïve subjects was unfortunately too small to perform a multiple regression analysis. Thus, an overlap of the influences of age, heart failure and ACEi treatment is possible. However, when comparing PRA between healthy children and children with heart failure, the LENA subjects were divided into four age groups to compare them with healthy peers and to keep the influence of age as low as possible. When comparing PRA between children with asymptomatic and symptomatic heart failure, there was no significant age difference between the two groups. The influence of enalapril on PRA was investigated in an observation period with a maximum of eight days. Therefore, we consider the influence of age to be negligible in the selected observation period. Since PRA decreases with age, age could at most have attenuated the observed effect of enalapril on PRA. As the Ross score improved in some of the subjects during the observation period, the improvement in heart failure may have attenuated the observed increase in PRA due to enalapril. However, as only two subjects had a change in Ross score large enough to change the heart failure classification, we assume that the effect of the improvement in heart failure on PRA will be rather small here. Moreover, we analysed the effect of enalapril on PRA in children with heart failure only in the first days of therapy. Further analysis of the studies data is planned to determine whether PRA remains elevated with prolonged ACEi therapy.