Echocardiography parameters used in identifying right ventricle dysfunction in preterm infants with early bronchopulmonary dysplasia: A scoping review

Background Bronchopulmonary Dysplasia (BPD) is a chronic condition that affects preterm infants and is associated with long-term complications. Haemodynamic effects of BPD can lead to right ventricular (RV) dysfunction. Objective To synthesise and map the evidence of echo parameters used in identifying RV dysfunction in the first two weeks-after-birth (WAB) of preterm infants with early BPD. Information Sources This scoping review included the databases: Medline, CINAHL, PubMed, EMBASE, Scopus, ProQuest, Web of Science, Cochrane Library, JBI Evidence-Based Practise and Gray Literature. Search Strategy The search utilised Boolean operators and descriptors registered in Medical Subject Headings. Inclusion and exclusion criteria Included were studies utilising echo parameters to examine RV function in preterm infants with early BPD in the first two WAB. Synthesis of results The results are presented as a map of the extracted findings in a tabular format with a narrative summary. Results Eight studies were included. Differences were observed in the number and timing of echo scans performed in the first two WAB and the variations in the echo parameters used to compare preterm infants with and without early BPD. Only echo scans performed at the end of the first WAB, demonstrated significant differences in the echo parameters measurements between preterm infants with and without BPD. Studies using RV Myocardial Performance Index (MPI) to identify RV-dysfunction associated with early BPD demonstrated similar findings. The Pulsed-Wave Doppler technique identified differences in RV-MPI between preterm infants with and without BPD, while Tissue-Doppler-Imaging did not demonstrate similar results. Speckle tracking can measure strain (S) and strain rate (SR) and diagnose RV-dysfunction. However, the findings of studies that utilised speckle tracking varied. Finally, two of the included studies added blood tests to their diagnostic model of early BPD, which was able to demonstrate significant differences in blood test results between BPD-affected and control preterm infants. Conclusion BPD could adversely affect the myocardium function of the RV; these negative influences can be captured in the first two WAB. However, there are still knowledge gaps regarding the appropriate number, timing and the most suitable echo parameters to assess RV function.


Introduction
Bronchopulmonary Dysplasia (BPD) was first described in 1967 (1). It affects the immature lungs of preterm infants. Preterm infants born before 32 weeks of gestation are at a greater risk of developing BPD, especially those requiring respiratory support with higher oxygen requirements (2). The incidence rate of this disease remains high; a moderate and severe form of BPD affects almost a third of the preterm infants born before 32 weeks of gestation (3). The incidence rate of BPD is expected to continue to be elevated, especially when the viability threshold is reduced to 22 weeks of gestation, while more at-risk extreme preterm infants are expected to survive (4).
Bronchopulmonary Dysplasia negatively affects the normal growth and development of the immature lungs' alveoli and vascular bed through complex processes, which can reduce gas exchange surfaces and subsequently result in a decline in pulmonary function (5). The vascular pathogenesis theory hypothesises that as the premature heart and the premature lungs' vascular bed are closely interlinked, a negative effect on one of them will also be reflected on the other. Abnormally affected lungs by early BPD will also impair angiogenesis of the vascular bed of the lungs and the formation of pulmonary vascular disease (PVD), which can adversely affect the function of the right ventricle (RV) of the heart.
The pulmonary vascular remodelling results in a rise in the vascular tone, altered reactivity, vasoconstriction, and increased pulmonary vascular resistance (6). These histologic changes can increase the RV afterload pressure. Subsequently, the chronic increase in RV afterload pressure and hypoxic episodes can result in RV dysfunction, hypertrophy and, in severe cases, failure (7). Furthermore, severe forms of BPD are associated with a higher incidence of pulmonary arterial hypertension (8), which results from an increase in pulmonary arterial pressure and is associated with significant co-morbidities and high mortality rates (9).
The literature showed that the pathological effects of early BPD and respiratory insufficiency could manifest as early as day 7 of postnatal life (10). Data from 1,735 infants born between 23 and 30 weeks of gestational age showed that the proportion of these preterm infants needing oxygen decreases from birth to day 7 of postnatal life, followed by a steep rise in the number of infants requiring oxygen in the same cohort at the start of the second week postnatal (10). Additionally, studies, albeit limited, such as by Czernik et al., 2012 (11), demonstrated that the negative effect of PVD associated with early BPD on the function of the RV is present as early as the first two weeks after birth.
To address the gap in knowledge and provide state-of-the-art current knowledge, the aim of this scoping review is to identify, synthesise and map the evidence of echo parameters used in identifying dysfunction of the RV in the first two weeks of postnatal life of preterm infants with early BPD. This review will examine the echo parameters in preterm infants born before 32 weeks of gestation since these infants are at higher risk of developing BPD. In addition, the review will include studies which performed echo scans in the first two weeks of postnatal life as it is the period when early BPD starts to manifest clinically. Specifically, the scoping review questions were: (i) What techniques are used to capture echo images in preterm infants? (ii) When are the echo scans performed in the first two weeks after birth? (iii) What echo parameters are used to assess the haemodynamic effects of early BPD on the function of the right ventricle? and (iv) How are the echo parameters measured and analysed?

Methods
The preliminary literature search on 01 April 2022 in Medline (OVID), Cochrane Library, PROSPERO and the JBI Evidence Synthesis Database revealed no scoping or systematic review available or currently being developed about this subject.
The proposed scoping review is conducted according to Joanna Briggs Institute (JBI) methodology (Peters M. D. J., 2020). Also, this scoping review used the reporting guideline "Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) extension for scoping reviews (PRISMA-ScR): checklist and explanation" (12).

Participants, concept and context
Participants: The scoping review included studies that examined the first two weeks of postnatal life of preterm infants born before 32 weeks of gestation.
Concept: The review included studies that examined the haemodynamic effects of PVD associated with early BPD on the RV function through the analyses of the echo parameters. The included studies should have at least one echo scan performed in the first two weeks after the birth of the participating preterm infants. In addition, eligible studies should have compared the data collected from echo parameters analyses in preterm infants with and without BPD.

Context:
The review included studies that recruited preterm infants admitted to a neonatal intensive care unit.

Types of evidence and sources
This scoping review considered all study designs, including experimental and quasi-experimental studies. Randomised controlled trials, non-randomised controlled trials, before and after studies and interrupted time-series studies were assessed for eligibility. Any systematic reviews that meet the inclusion criteria were considered, but none were retrieved. The review also considered analytical observational studies, such as prospective and retrospective cohort studies, case-control studies and analytical cross-sectional studies. There were no limitations regarding the date of the publication.
Papers were excluded if they did not fit in the conceptual framework of this review. In addition, studies were excluded if they used different imaging modalities, such as Magnetic Resonance Imaging or Computed Tomography or were not available in full text. Studies that did not assess the RV function or echo scans performed after the first two weeks after birth were also excluded. Studies were excluded if they were unavailable in English or were animal studies. Web of Science (((ALL = ((preterm infants) OR (premature infants) OR (preterm baby) OR (premature baby))) AND ALL = ((bronchopulmonary dysplasia))) AND ALL = ((right ventricular function) OR (right ventricular dysfunction))) AND ALL = (echocardiography OR echo OR echocardiogram)

Search terms
The selected text words and index terms, such as preterm infants, right ventricular function, bronchopulmonary dysplasia, and echocardiography, were utilised to formulate a full search strategy ( Table 1).

Search strategy
A comprehensive search strategy was followed and consisted of three steps.
First step: An initial limited search of two online databases, Medline (OVID) and CINHAL Plus Full text (EBSCOhost), was conducted to allocate related articles. The examination of the relevant articles resulted in the identification of the text words included in the titles and the abstracts and the index terms related to these articles to guide the subsequent detailed search as per the advice of a research librarian.
Second step: The full search strategy was applied to the included databases with the identified keywords and index terms. The descriptors registered in the Medical Subject Headings (MeSH) were used wherever possible. The synonyms were combined with the Boolean operator "OR" while the groups of words were combined with the operator "AND" ( Table 1). The following databases were searched: Medline (OVID), CINAHL Plus with Full Text (EBSCOhost), PubMed, EMBASE, Scopus, ProQuest, Web of Science, Cochrane Library, JBI Evidence Based Practise (EBP) and Gray Literature (British Library, google scholar & EThOS).
Third step: A reference list containing all the identified studies was formed. There were no scoping nor systematic reviews identified.
A full search was performed by two reviewers (WM and JML). Any unresolved dispute was discussed with a third reviewer (EN).

Studies selection
The studies were selected from the included databases. All the identified citations were uploaded into EndNote 20 (Clarivate Analytics, PA, USA), followed by removing duplicates. Then, studies were selected by two reviewers (WM and JML) in two stages; titles and abstracts review; and full-text review, against the inclusion criteria. A detailed assessment of the full text of the selected articles against the inclusion criteria was performed by two independent reviewers (WM and JML). Any disagreements were discussed and resolved by both reviewers.

Data extraction
Two independent reviewers (WM and JL) performed the data extraction process using the tool developed by the reviewers for this purpose. The data extraction tool contained details of the population, concept, context, methodology and relevant findings. No modifications to the data extraction tool were performed.

Results
The final scoping review full report of the search results is presented in the PRISMA flow diagram ( Figure 1). There were 501 records identified from 12 databases. Sixty-two duplicates were removed, leaving 439 records eligible for the first screening stage, i.e., titles and abstracts were screened. After the exclusion of 415 records, only 24 papers were eligible for the second stage of full-text screening. Sixteen articles were excluded (Figure 1), and eight studies were included in the final critical appraisal and analysis.

Data charting
Data were extracted from the full texts of the eight included studies and detailed into two tables (Tables 2, 3). Table one described the studies' methodologies, while table two included details about the echo scans' timing, echo parameters analyses and the key findings. The results are presented in narrative descriptions and mapping of the data.

Narrative description of studies
The eight included studies were conducted in the USA and European countries. All studies were conducted in tertiary perinatal centres. Only four studies examined exclusively preterm infants born before 32 weeks of gestation (11, 13-15) ( Table 1). The study by Helfer et al., 2014, included preterm infants born at 32 weeks gestation as the recruitment was per birth weight rather than gestational age (16). Another study included preterm infants with a higher gestation age, and the birth weight needed to be within a specific range (≤34 weeks gestation and birth weight between 500 and 1,250 grams) (17). The other two studies by Levy and colleagues focused on recruiting preterm infants with lower gestational age (23- The studies' timing and frequency of echo scans varied ( Table 2). Three studies performed one echo scan within the first two weeks after birth (13, 15, 17); one study had the echo scans performed on the first day after birth (DAB) (17), and the other two studies performed the echo scans at the 7th DAB, respectively (13, 15). The remaining five studies had several echo  A total of 89 preterm infants (<32 weeks gestation) were included: Echo parameters were obtained via one of the following methods: -Pulsed wave (PW) Doppler.
-Tissue Doppler Imaging (TDI). Appropriate statistical tests were used for analyses.

Helfer et al., 2014
Germany The study aims to examine the development of cardiac function in preterm infants by measuring tissue Doppler-derived peak systolic strain (PSS) and strain rate (PSSR) in the first 28 days after birth. It is also to assess the impact of BPD on PSS & PSSR.
A total of 119 preterm infants (birth weight <1500 g) were included: Images from 110 preterm infants were suitable for analysis: -Preterm infants without BPD n = 79.
-Preterm infants with BPD n = 31. A prospective, single-centre trial was conducted at a tertiary perinatal centre from September 2008 to January 2010. The study was approved by the ethics committee at the research team institution.
Appropriate statistical tests were used for analyses. 3.2.3. What echo parameters are used to assess the haemodynamic effects of early BPD on the function of the right ventricle?
The critically appraised studies utilised different echo parameters to assess the function of the RV in healthy preterm infants (controls) and the ones affected by BPD (cases); RVmyocardial performance index (MPI), RV strain (S) and strain rate (SR), RV fractional area of change (RV-FAC), TDI systolicand diastolic-velocities and tricuspid annular plane systolic excursion (TAPSE).

How were the echo parameters measured and analysed?
Studies utilising RV-MPI to assess RV function, the calculation of RV-MPI was performed by using one of the following two equations: Four studies showed a delay in the normal maturation of the RV-MPI, i.e., persistently raised RV-MPI in BPD-affected preterm infants in comparison to the controls. The differences in the RV-MPI between the two groups (preterm infants with and without early BPD) were detected in the echo scans performed after the first three days after birth (11,(13)(14)(15)). In addition, Neumann et al., 2021, demonstrated that TAPSE is lower in BPD-affected preterm infants (15).
Other findings were identified when examining all the included studies. Neumann et al., 2021, demonstrated that TAPSE is lower in BPD-affected preterm infants (15). Levy et al., 2015, showed differences in the RV-FAC measurements between controls and BPD-affected preterm infants when assessed at 32 weeks of PMA echo scans (18). Two studies utilising S and SR also demonstrated statistically significant differences in these measurements between their controls and the cases of preterm infants (16, 19). However, the study by Helfer et al., 2014, demonstrated that the differences in the S & SR measurements between controls and BPD-affected preterm infants would start to manifest in the echo scans at day 14 after birth while the The study aims to assess the influence of BPD on the maturation of the myocardium in very low birth weight infants (VLBWIs).
Preterm infants born at ≤32 weeks gestation and/or ≤1500 -Preterm infants with BPD n = 15 A prospective, observational, single-centre study was conducted at a tertiary perinatal centre from January 2015 to 2017. The study was approved by the ethics committee at the study institution.
Echo parameters were obtained via: -Tissue Doppler Imaging (TDI). The study was approved by the ethics committees at the two study institutions.
Echo parameters were obtained via one of the following methods: -Pulsed wave (PW) Doppler.
-M-Mode. Appropriate statistical tests were used for analyses.  Echo scans were performed at three-time points: -First day after birth.
-B was calculated from pulsed Doppler waveforms of TV inflow. Increased RV-MPI via PW-Doppler at 28 days after birth in preterm infants with severe-BPD group compared with both the preterm infants without BPD (p = .014) and the preterm infants with mild-BPD groups (p = .031).
No difference in RV-MPI between preterm infants with no-BPD and preterm infants with mild-BPD groups was detected (p = .919).
No difference in E/E' RV ratio between the groups at any time point.

Czernik et al., 2012
Echo scans were performed at four-time points: -At the second day after birth.
-At the 7th day after birth.
-At two weeks after birth.
-At the 28th day after birth.
-B was calculated from pulsed Doppler waveforms of TV inflow.
When grouping the study participants by BPD or PH status, there was no significant association between TDI measures and the presence or absence of PH or BPD.

Helfer et al., 2014
Echo scans were performed at four-time points: -At the first day after birth (0-3).
-At the 28th day after birth (22-31).  In addition to using echo parameters, two of the eight studies performed blood tests. Mendez-Abad et al., 2020, analysed Nterminal pro-B type natriuretic peptide (NTproBNP) plasma level on day 14 after birth (14). Neumann et al., 2021, analysed the plasma concentrations of mid-regional pro-atrial natriuretic peptide (MR-proANP) and C-terminal pro-endothelin-1 (CT-proET1) on day 7 after birth (15). Both studies demonstrated a statistically significant difference in the plasma levels of the selected tests between the controls and BPD-affected preterm infants.

Discussion
There were methodological variations between the selected studies. There are differences in the number and the timing of echo scans performed in the first two weeks of postnatal life, together with the variations in the echo parameters used to compare the preterm infants as controls and the preterm infants affected by BPD.
Echo scans performed in the first three days after birth did not demonstrate a difference in the echo parameters measurements between the control group and preterm infants with early BPD. In comparison, significant differences in the echo parameters measurements between the controls and preterm infants with BPD were seen in the echo scans performed at the end of the first week after birth and the subsequent weeks (11,13,15,16). Similarly, the study by Levy et al., 2015, showed that RV-FAC measurements in the first three days after birth did not differ between preterm infants without BPD and the ones who have BPD (18).
Studies using RV-MPI to identify RV dysfunction associated with early BPD have similar findings. Bokiniec et al., 2017, showed that PW-Doppler could identify the differences in RV-MPI between controls and preterm infants affected by early BPD, while TDI could not demonstrate differences in RV-MPI between preterm infants with and without BPD (13 Despite the limitations of PW-Doppler in calculating RV-MPI, e.g., it utilises the measurements from different cardiac cycles, three of the eight selected studies used PW-Doppler and were able to diagnose RV dysfunction in preterm infants with early BPD. On the other hand, despite TDI being a valuable and practical tool in assessing RV's systolic and diastolic function, TDI was unable to identify the RV dysfunction (13, 17). This could be explained by the inherent limitations of TDI, which is angle-dependent and influenced by the global cardiac motion and the inability to differentiate between passive and active motion (24). In addition, velocities and displacements depend on the cardiac sizes; hence, larger hearts have higher velocities and displacements. Therefore, it is crucial to normalise the TDI measurements according to the cardiac size (25). None of the studies that utilised TDI technique normalised the cardiac TDI measurements, which might negatively affect the ability of these studies to demonstrate a significant difference in the TDI measurements between controls and cases.
Speckle tracking can be used to measure S and SR and diagnose RV dysfunction. However, the findings of the studies that utilised speckle tracking in this review varied. A study by Levy et al., 2017, could not diagnose the RV dysfunction in the echo scans performed at day 1, 2, and 5 to 7 after birth (19). In contrast, a study by Helfer et al., 2014, demonstrated a significant difference in the S and SR between preterm infants with and without BPD when echo scans were performed at day 14 after birth (16). The findings of these two studies demonstrate that results can vary even when a reliable technique such as speckle tracking was used. This might be well related to the timing of the echo scans, i.e., if the echo scan were performed too soon, it might not Echo scan and blood tests were performed on the 7th day after birth. Blood tests to measure the Plasma concentrations of mid-regional pro-atrial natriuretic peptide (MR-proANP) and Cterminal pro-endothelin-1 (CT-proET1), were performed at the same time of the echo scans. -TAPSE.
-B was calculated from pulsed Doppler waveforms of TV inflow. -A was calculated from pulsed Doppler waveforms of RV outflow.
capture the hemodynamic changes related to RV dysfunction that is associated with early BPD. Furthermore, two of the selected eight studies added blood tests to their diagnostic model of early BPD (14, 15). Both studies demonstrated that adding specific blood tests would enhance the diagnostic ability of RV dysfunction associated with early BPD (14, 15).
Although, adding a specific blood test to a diagnostic model might enhance its diagnostic ability, it might also have a negative impact on its applicability in many perinatal centres worldwide, especially in developing countries, due to, for example, the cost and unavailability of these blood tests. Likewise, it will negate the idea of having a diagnostic model based on a practical, applicable and non-invasive test such as an echo scan.
The primary strength of this review was its broad literature search. The process of synthesising the scientific evidence was transparent and systematic in identifying and mapping the evidence related the use of echocardiography to diagnose RV dysfunction associated with early BPD. Nevertheless, as this review designed as a systematic scoping review, it did not assess the quality of the evidence, so it does not provide the scientific basis to inform clinical decision making. Finally, language restrictions may have led to the exclusion of a few studies.

Conclusion
This scoping review showed the need to develop a diagnostic model based on a non-invasive echo scan. The review demonstrated the emergence of unanswered questions regarding the practicalities of when and how many echo scans are needed and which cardiac parameter is required to diagnose early BPD in preterm infants.
BPD will adversely affect the myocardium function of the RV; these negative influences can be captured in the first two weeks of postnatal life (15, 17). However, there are still knowledge gaps regarding the appropriate number, timing of the echo scans and the most suitable cardiac parameters to assess RV function.

Author contribution
WM: contributed to the main design and detailed concept and discussion of this review article; he drafted the first manuscript. JML: as a senior author revised the manuscript and added details to it. EN and JH revised the manuscript and added details. All authors contributed to the article and approved the submitted version.