Reference intervals for the echocardiographic measurements of the right heart in children and adolescents: a systematic review

Background Transthoracic echocardiography is the primary imaging modality for the diagnosis of right ventricular (RV) involvement in congenital and acquired heart diseases. There is increasing recognition of the contribution of RV dysfunction in heart diseases affecting children and adolescents, but there is insufficient information on reference intervals for the echocardiographic measurements of the right heart in children and adolescents that represent all the continental populations of the world. Objective The aim of this systematic review was to collate, from published studies, normative data for echocardiographic evaluation of the right heart in children and adolescents, and to identify gaps in knowledge in this field especially with respect to sub-Saharan Africans. Methods We performed a systematic literature search to identify studies of reference intervals for right heart measurements as determined by transthoracic echocardiography in healthy children and adolescents of school-going age. Articles were retrieved from electronic databases with a combination of search terms from the earliest date available until May 2013. Results Reference data were available for a broad range of variables. Fifty one studies out of 3096 publications were included. The sample sizes of the reference populations ranged from 13 to 2036 with ages varying from 5 to 21 years. We identified areas lacking sufficient reference data. These included reference data for determining right atrial size, tricuspid valve area, RV dimensions and areas, the RV % fractional area change, pulmonary artery pressure gradients and the right-sided haemodynamics, including the inferior vena cava dimensions and collapsibility. There were no data for sub-Saharan African children and adolescents. Conclusion Reliable reference data are lacking for important echocardiographic measurements of the RV in children and adolescents, especially for sub-Saharan Africans.


Introduction
Transthoracic echocardiography is the primary imaging modality for the diagnosis of right ventricular (RV) failure [1][2][3][4]. Besides aiding in the diagnosis of conditions such as arrhythmogenic right ventricular cardiomyopathy, pulmonary embolism, and RV infarction, echocardiography plays a critical role in the diagnosis of congenital heart diseases where the RV often serves as the main pumping chamber [5][6][7][8].
The absence of reference intervals for cardiac structures in children and adolescents is an important problem [6]. Publications dedicated to the echocardiographic study of the right heart, especially on reference values of the structure and function of the right heart in children and adolescents are scarce. Many of the previous studies had limited sample sizes, and were conducted predominantly in North American and European populations. Thus there is potential for interpretation errors when assessing African children, given that environmental, social, economic and other factors may influence the anthropometric standards of a population [9]. Systematic reviews provide rigorous, objective evidence to assess the literature through the use of a prespecified protocol and access a variety of database search engines. The use of explicit, systematic methods in reviews limits bias and reduces the effects of chance, providing more reliable results [10]. The aim of this systematic review was to collate, from published studies, normative data for echocardiographic evaluation of the right heart in children and adolescents, in order to identify gaps in knowledge in this field, especially with respect to the available of information on sub-Saharan African children.

Types of studies
This review considered all publications reporting reference values for the right heart in healthy children and adolescents determined by echocardiography.

Types of reference individuals
Participants included school children and adolescents with no history or echocardiographic evidence of heart disease. No exclusion was set on sample sizes of the studies.

Inclusion and exclusion criteria
Inclusion criteria: Age range from 5 to 21 years. Documentation of age and/or an indicator body size for reference individual(s). Indicators of body size included height and/or weight and/or body mass index (BMI) and/or body surface area (BSA). Presentation of results as one reference value and/or -interval for the whole sample or, in relation to age and/or body size. Availability of English version of paper in the case of foreign-language articles.

Exclusion criteria:
Preterm infants. Deceased participants (autopsy studies). Participants with known cardiovascular disease. Measurements taken at high altitude ≥2400 metres above sea level. Age groups including adults (without subgroups for participants younger than 22 years). Presentation of reference values that had already been included in a previously published article.  We combined the outputs from the databases PubMed and ISI Web of Knowledge using a referencing program, Endnote® (Version X5; Thomson Reuters). After duplicate entries were removed, titles of citations were screened for possible inclusion. The titles of potentially relevant studies were reviewed, after which the abstracts and full text articles were examined for possible inclusion. We attempted to find an English copy of the full text article for all of the selected abstracts. Articles were graded as eligible, potentially eligible, or not eligible based on the inclusion and exclusion criteria.

Data extraction and analysis
We extracted the following data: the year of study, sample size, age range, study setting, sample selection methods, measurements and methods of measurements onto a data extraction form. For each age-group, the sample size with reported summary statistics (i.e., mean, median, centiles, standard deviation, confidence intervals, or standard error) for measurements were also documented.

Quality assurance
The systematic review was conducted according to the methods of the Cochrane Collaboration [11]. We included all available published reference values of echocardiographic evaluation of the right heart in children and young adults in an attempt to minimise publication bias.

Results
From the 3096 publications retrieved from the databases, we identified 51 studies for inclusion in this report. 2 846 articles were excluded on the basis of title alone, while a further 107 abstracts and 72 articles failed to meet the inclusion criteria. 18 of the 19 foreign language articles were excluded because no English translation was available. We were able to include data from the remaining foreign language article as the sub-headings of the tables containing reference data, were in English. Two articles were unobtainable.

Description of studies
The majority of included studies were conducted in clinical settings (such as hospitals, clinics, or medical centres), and research laboratories. 17 studies did not specify the setting. 15 of the included 51 studies were published before the year 2000, while 36 were published from 2000 onwards. The oldest publication dates back to 1977 [12]. Twenty two studies were from reference populations in North or South America, 17 from European countries, 8 from Asia, 1 from the Middle East, 1 from Australia and 1 from north Africa (Egypt).
The sample size of the reference populations ranged from 13 [13] to 2 036 [14]. The ages of the reference populations ranged from 5 to 21 years.
There were insufficient published data for the following measurements: RA size (width, length and area) for children younger than ten years, tricuspid valve (TV) area, RV dimensions (base, mid and length) and -areas  (diastolic and systolic), the RV fractional area change and the RV-to-RA peak pressure gradient. There were no published reference data for the RV mid-cavity dimensions presented as a sub-group for children younger than ten years; PA peak pressure gradient for children older than 12 years; inferior vena cava (IVC) diameter and percentage collapse for children younger than ten years. None of the studies estimated PA systolic pressure using an estimate of RA pressure that was based on the dimension and percentage collapse of the IVC. Lastly, there were no published data for RV volumes and ejection fraction using the area-length method.

Discussion
While performing this review, we recognized several limitations of available reference data in paediatric echocardiography, including a lack of technical standardization of measurements, inappropriate "normal" subjects and choice of population [6,8,9,, small sample sizes [12,13,21,22,33,37,49,51,56,60] and heterogeneous methods of reporting reference values. A valid metaanalysis could not be performed because of the variability in the populations studied, in the methodology for performing and normalizing measurements, and in ways to express normalized data. These limitations were also reported in a previous review [61]. We consider this review as the most complete report on the availability of reference data in children and young adults, highlighting areas lacking sufficient data with respect to measurements of the RA, the TV, RV size and -function, the pulmonary valve (PV), the PA, and right heart haemodynamics.

Right atrium
We identified only one publication presenting reference data for RA size for children younger than 10 years of age. In 1988, Hanseus et al. published a set of reference values of the RA on a sample of 120 healthy infants and children, aged three days to 15.5 years. His RA measurements included the width, length and area for the RA. He presented the data for the entire sample according to body surface area(BSA) using regression equations and nomograms [46].
Innelli et al. published a set of reference values for RA width on a sub-group of 40 healthy children and adolescents, aged 10-19 years. In this recent publication, reference values for the entire sub-group, aged 10-19 years were presented as one value [22].

Tricuspid valve area
Sufficient data exist on the diameter of the TV annulus, but we discovered only one publication presenting reference data on the TV area. Singh et al. published reference values on the TV area based on a reference population aged two months to 50 years. The data for participants less than 16 years were presented in 3 different age-categories [41].

Right ventricle size and/or volume
We identified one publication presenting reference data for RV basal diameter for children younger than 10 years of age [46]. This two decades old study obtained a reference population of 120 healthy infants and children, aged 3 days to 15.5 years. The data were adjusted for BSA, using regression equations and nomograms [46]. Innelli et al. recently published reference values on a sub-group of 40 healthy children and adolescents, aged 10-19 years as one value [22].
Two publications exist on reference values of the RV mid diameter measured in the apical four-chamber view [22,51]. Norgard et al. presented reference values for RV mid diameter on only 15 individuals, aged 6-16 years as one value for the entire sample [51]. More recently, Innelli et al. published reference values of a sub-group of 40 healthy children and adolescents, aged 10-19 years as one value [22]. We found no publications presenting reference values for RV mid diameter measured in the apical four-chamber view for children younger than 10 years.   Two publications were found containing reference values of the RV end-systolic area measured in the apical fourchamber view [45,51]. The usefulness of both of these publications however, is limited as they both present the data for the entire sample spanning more than 10 years as one value. One study had only 15 individuals, aged 6-16 years [51]. Boettler et al. published reference values of a larger series containing 129 reference individuals, aged 1 day to 16.9 years [45].

Right ventricular function: % fractional area change and isovolumic acceleration
Only one publication was identified for RV % fractional area change in a sample of healthy children [13]. The sample of healthy children was a control-group consisting of 13 healthy individuals. One of the objectives of the study was to compare quantitative measurements of the RV in single RV's to normal RV's. The % fractional area change of the control-group consisting of 13 healthy individuals was presented. The mean age of the healthy control-group was 5.7 ± 1.8 years [13]. A study aiming to generate reference values for RV fractional area change, containing a larger sample with greater age-span and categorization may be more useful in clinical practice.
We did not find any existing reference data on isovolumic acceleration.

Pulmonary artery peak pressure gradient
Zhendong et al. published reference values of the PA peak pressure gradient in 1998 in a reference population consisting of 88 healthy individuals, aged 3-12 years. The sample was divided into two age-categories [58]. No publications were found containing data for children older than 12 years.

Right-sided haemodynamics
McQuillan et al. have published the only reference data on right-sided haemodynamics for children and adolescents that we could find [43]. This study of 856 participants younger than 20 years published reference data for the TV RV-RA gradient. They presented estimated PA systolic pressure values assuming that the RA pressure for all reference individuals was 10 mmHg. The reference data were presented graphically, normalised to BSA [43]. Pena et al. published reference data for the tricuspid regurgitation (TR) peak velocity, but of 55 new-born infants recently [52]. The authors also presented estimated PA systolic pressure values assuming that the RA pressure for all reference individuals was 5 mmHg.
No reference values for the diameter and percentage collapse of the IVC for children younger than 10 years were found. Innelli et al. published reference values for the IVC diameter and percent IVC collapse in a subgroup consisting of 40 healthy children and adolescents but presented the data for the entire sample, aged 10-19 years as one value [22].
No reports were found on reference values for estimated PA systolic pressure with an estimate of RA pressure on the basis of IVC size and collapse.
We found sufficient reference data for the following variables: TV annulus, RV length, RV end-diastolic area, RV outflow-tract dimension, PV diameter, PA diameter, tricuspid annulus peak systolic excursion (TAPSE) and tissue myocardial velocities at the lateral tricuspid annulus.

Tricuspid valve annular size
Four authors have published reference data of the annulus of the TV based on reference populations ranging from infants to young adults, normalised to BSA [17,24,29,35].  N = 15, 6-16years) presented data for the entire sample, as one value [17,22,46,51].

Right ventricular size: outflow tract dimension
There are six existing publications reporting the dimension of the RV outflow tract, measured anterior to the aortic valve in the parasternal short-axis view [9,20,29,41,46,51]. The oldest publication dates back to 1988 [46] and the most recent publication was published in 2008 [29]. The sample sizes range from 15 [51] to 782 [29]. The ages of the samples range from birth to adolescent ages. Four authors presented data according to BSA [9,20,29,46]. Singh [28]. The data were presented according to BSA and age-categories.

Conclusions
In the clinical diagnosis and management of patients with cardiopulmonary disorders, arrhythmogenic right ventricular dysplasia and congenital heart disease, the evaluation of the RV is of major importance. Echocardiography is considered a cornerstone in the diagnosis and management of RV failure. Moreover, it is a leading technology, is less costly, is non-invasive and offers the advantages of wide applicability and availability when compared with other imaging modalities.
Clinically reliable reference data are lacking for important echocardiographic measurements of the RV in children and adolescents. There were insufficient published data of the following variables: RA size (width, length and area) for children younger than 10 years, TV area, RV dimensions (base, mid and length) and areas (diastolic and systolic), the RV percentage fractional area change and the RV-to-RA peak pressure gradient. There were no published reference data for the RV mid-cavity dimensions presented as a subgroup for children younger than 10 years; RV volumes and ejection fraction using the area-length method; PA peak pressure gradient for children older than 12 years; and IVC diameter and -percentage collapse for children younger than 10 years. None of the studies estimated systolic PA pressure using an estimate of RA pressure that was based on the dimension and percentage collapse of the IVC.