Metrics for size-matching in paediatric cardiac transplantation: A narrative review

Abstract

Several metrics exist for matching the size of donor to recipient in paediatric cardiac transplantation. Different centres employ different metrics for size-matching to determine the viability of donor hearts. Inconsistent evidence exists, with limited consensus as to the metric for size-matching that is most predictive of outcomes after cardiac transplantation. Furthermore, any metric must function within the tight timeline available for the assessment of the suitability of the donor. At the time of the writing of this paper, the most commonly used metric for size-matching in paediatric cardiac transplantation is the donor-to-recipient body weight ratio. In this article, we review published literature evaluating commonly used metrics for size-matching in paediatric cardiac transplantation, including weight, height, body surface area, and imaging parameters.

Improving the utilisation of limited donor hearts in paediatric cardiac transplantation is dependent upon the effective identification of suitable donors. Efficient matching of donor to recipient is imperative in order to mitigate mortality on the waitlist, increase the longevity of transplanted hearts, and improve the survival of recipients. ^1–3 While several factors are used to assess potential donors, the determination of size concordance is one of considerable variability. ⁴ Body weight has been the longest-standing and most widely studied metric for the evaluation of size in paediatric cardiac transplantation. ⁵ Other, less established, anthropomorphic metrics include body surface area and height. ^5–7 Radiographic evaluation also plays a role in size-matching through several modalities ^8–10 :

• direct analysis of cardiac dimensions with echocardiography,

• the calculation of total cardiac volume through computed tomography, and

• the determination of predicted heart mass through equations derived from cardiovascular magnetic resonance imaging.

Nevertheless, literature evaluating the utility of metrics for size-matching beyond weight is sparse, and the results are variable. Even in the context of weight alone, few centres use the same combination or hierarchy of metrics for size-matching in paediatric heart transplantation, leading to significant variability in practice. ⁴ This variety in the measurements used for size-matching in paediatric cardiac transplantation indicates a need for overarching review and analysis.

The purpose of this Narrative Review is to provide a comprehensive summary and comparison of outcomes and recommendations from studies investigating metrics for size-matching in paediatric cardiac transplantation, including the following metrics:

• body weight,

• height,

• body surface area, and

• measurements obtained through different imaging modalities.

By summarising and comparing the available literature, we aim to offer healthcare teams a more informed perspective on the current state of metrics for size-matching in paediatric cardiac transplantation.

Methods

While the studies included in this Narrative Review were not compiled using a formalised systematic and reproducible methodology, the general review process of our Literature Review is as follows. A comprehensive manual search of academic literature databases was conducted, including the following databases:

• PubMed,

• Google Scholar, and

• Scopus.

Keywords were used, including:

• “paediatric,”

• “heart transplant,”

• “size-matching,”

• “donor-to-recipient matching,”

• “matching,” and

• various size metrics (height, weight, etc.).

Studies were limited to those with full texts available, those published after the year 2000, and those published in English. Further, only studies relating to heart transplant size-matching in paediatrics were included, with studies about adults and studies focusing on the transplantation of other organs excluded from this analysis. Included studies were initially screened for titles and topics relevant to this review before more thorough screening for relevant content was performed by the first two authors (GPS and OMS). Size-matching data and recommendations were extracted using a standardised form, including information about:

• year,

• study type,

• participants,

• interventions, and

• outcomes.

Articles were then separated into categories representative of their metric of focus for thematically partitioned manuscript composition.

Results

Table 1 provides a summary of the findings about sizing and the recommendations from the included studies. Table 2 provides a summary of the pros and cons of the various size-matching metrics. These findings that are summarised in these two tables will be discussed in detail below, organised by the following topics:

Table 1. Summary of the findings about sizing and the recommendations from the included studies

DRWR = donor-to-recipient body weight ratio; ECMO = extracorporeal membrane oxygenation.

Table 2. Summary of pros and cons of various size-matching metrics

• Weight

• Height

• Body surface area

• Imaging criteria

Weight

Weight as a metric for size-matching in paediatric heart transplantation is the most studied and clinically validated metric within the literature, with a strong link between weight and cardiac size in neonates, infants, and children. ¹¹ Evidence already exists for the use of weight in size-matching for adult cardiac transplantation. However, in paediatric heart transplantation, wider variability in the anatomy and physiology of donors has obscured the link between weight matching and outcomes and has led to support for a less formulaic approach than in adults. Most studies evaluating weight utilise donor-to-recipient body weight ratio to assess outcomes and survival.

Numerous studies have found that elevations in donor-to-recipient body weight ratio up to 2.5−3.0 were not associated with poorer survival or outcomes compared with lower donor-to-recipient body weight ratio and were instead advantageous for patients with dilated cardiomyopathy and for expansion of the eligible donor pool. ^11–13 As the majority of patients eligible for organ donation after brain death are adults, children undergoing cardiac transplantation often receive organs from an inherently larger donor to facilitate timely transplantation. The paucity of small donor hearts can be seen in associations found between low donor-to-recipient body weight ratio and prolonged waitlist times in one study, while other studies report the association between younger, smaller patients and low donor-to-recipient body weight ratio, indicating the limited flexibility of the scarce donor pool. ^12,14 While recipients receiving hearts from smaller donors had longer waitlist times, an analysis of infant heart transplant recipients yielded that donor-to-recipient body weight ratios above 2 were associated with significantly shorter waitlist times. ¹⁵ Intentional oversizing as a strategy to minimise mortality on the waitlist in infants was studied by Lee and colleagues via a 25-year analysis of the United Network for Organ Sharing database. ¹⁶ Lee and colleagues recommended safe, “modest oversizing” with donor-to-recipient body weight ratio of 1.29–1.97, to both mitigate the potential adverse effects of oversized organs and minimise waitlist time. ¹⁶ This recommendation speaks to the utility and “protective effect” of donor pool expansion for such young and small recipients. ¹⁶ It should be noted that recipients in the third and fourth quintiles in this study (donor-to-recipient body weight ratios of 1.29–1.57 and 1.58–1.97, respectively) had significantly improved survival compared to recipients in the highest quintile (donor-to-recipient body weight ratio of 1.97–5.00), indicating that intentional oversizing may have diminishing returns at the extremes of weight mismatch. ¹⁶ In other analyses, incongruencies between donor and recipient size were not associated with poor outcomes, though the range of studied donor-to-recipient body weight ratios in these analyses did not reach 5.00. ^11–13 While with limited benefit, or potential detriment, to waitlist times, it has also been reported that the use of smaller donor hearts (donor-to-recipient body weight ratio of 0.6–0.8) had no negative impact on the short- and long-term survival of children who received a heart transplant. ^14,15

While the advantages of utilising donors with a wider range of donor-to-recipient body weight ratio are clear in the context of wait times and donor availability, postoperative outcomes must be considered. Multiple investigations note that elevated donor-to-recipient body weight ratios were associated with an increased rate of delayed sternal closure, yet Razzouk and colleagues found that this delay was not associated with worse outcomes or surgical site infection. ^11,13 Kanani and colleagues reported that donor-to-recipient body weight ratios greater than 3, termed “extreme mismatches,” were associated with an increased need for postoperative support with extracorporeal membrane oxygenation, although the sample size (n = 9) was small. ¹³ Elevated donor-to-recipient body weight ratio can lead to supra-optimal cardiac output with vascular and intracranial hypertension, although investigations have not found that these features impact short- or long-term outcomes. ^11–13,17

Donor-to-recipient body weight ratio is also not an isolated measurement and can be considered in the context of history, sex, and age. Regarding history, patients with a history of previous cardiac surgery, dextrocardia, heterotaxy, or situs inversus may better tolerate donor-to-recipient body weight ratios closer to 1 to accommodate acquired adhesions and altered mediastinal geometry. ¹¹ Thus, patients with this type of history may be limited to receiving hearts from matched or undersized donors due to their anatomical constraints. In the context of sex, one study reported the intersecting influence of sex and donor-to-recipient body weight ratio on survival, noting that sex and donor-to-recipient body weight ratio pairings impacted survival only in male patients. ¹⁸ Among male recipients, irrespective of the sex of the donor, survival after transplantation was greatest for higher donor-to-recipient body weight ratio and lowest for undersized donor-to-recipient body weight ratio; however, ideal (0.8–1.5) donor-to-recipient body weight ratio matches yielded the best survival when males received female hearts. ¹⁸ No differences in survival after transplantation existed for female recipients on the basis of gender-match, donor-to-recipient body weight ratio and gender and donor-to-recipient body weight ratio pairings. ¹⁸ Finally, age was considered by Tang and colleagues, who reported that the use of smaller donor hearts (donor-to-recipient body weight ratio of 0.6–0.8) had no negative impact on the short- and long-term survival of children who received a heart transplant, but infants with donor-to-recipient body weight ratio of 0.5–0.59 had lower 30-day survival (p = 0.045), albeit in a small cohort (n = 5). ¹⁴

While the literature consistently supports the use of weight as an indication of cardiac size, variations in outcomes still exist across donor-to-recipient body weight ratio scenarios. The most effective interpretation of these findings may be in a context-dependent manner. Higher donor-to-recipient body weight ratio may facilitate shorter waitlist time and be better tolerated in patients with dilated cardiomyopathy; however, extreme donor-to-recipient body weight ratios may be accompanied by increased need for delayed sternal closure and increased utilisation of extracorporeal membrane oxygenation after cardiac transplantation in the immediate postoperative period. Though not found in this review of the literature, hearts from larger donors may better suit the specific physiological needs of the recipient, such as increased peripheral vascular resistance. Lower donor-to-recipient body weight ratios are less common secondary to the scarcity of smaller donors and show inferior survival in infants, although lower donor-to-recipient body weight ratios may be better tolerated in some patients with a history of prior cardiac surgery. In sum, the present studies indicate that body weight has been thoroughly validated as a size-matching criterion in paediatric cardiac transplantation that can predict, although potentially case-dependent, outcomes. Regardless, transplant teams must consider relevant anatomical and physiological parameters in addition to weight to make a thorough judgement as to whether a donor is suitable for a recipient, weighing the benefits of organ availability against potential complications.

Height

While body weight is undoubtedly the most widely accepted metric for size-matching in transplantation, instances exist where height could offer distinct advantages over body weight such as in the setting of underweight or oedematous recipients. Using additional or alternative metrics such as height could better characterise the true body size of these patients.

Donor-to-recipient height ratio matched pairs in children with dilated cardiomyopathy were the focus of an investigation by Patel and colleagues. They noted that donor-to-recipient body weight ratio may incorrectly match donor and recipient cardiac size if the donor has a history of failure to thrive—as close to one-fifth of their patients were below the fifth percentile for weight. ¹⁹ Univariate analysis of donor-to-recipient height ratio showed a strong prediction of inferior 1-year survival for donor-to-recipient height ratio less than 0.87 compared to donor-to-recipient height ratio greater than 0.87, although differences in donor-to-recipient height ratio were no longer predictive of survival upon multivariate analysis. ¹⁹ Because of this lack of significance in the multivariate analysis, the authors asserted (1) that height had no superiority over the use of weight in the inherently multifactorial prediction of survival after paediatric cardiac transplantation and (2) that the potential pitfalls with the use of body weight are overshadowed by its superior outcome prediction.

Height was also a metric considered in a broad analysis of multiple size-matching criteria from within The International Society for Heart and Lung Transplantation (ISHLT) Registry. Outcome analysis revealed that height mismatch was not associated with decreased 1- or 5-year survival in both the overall cohort and stratified age groups. ⁵ However, multivariate analysis revealed that donor-to-recipient height mismatch was a significant risk factor for increased 5-year mortality, with no impact on 1-year mortality. ⁵ This 2019 analysis of the International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation (ISHLT) concluded “that donor-recipient weight match was independently associated with 1-year survival with undersized donors associated with lower survival, and donor-recipient height match was independently associated with 5-year survival, with oversized donors associated with lower survival.” ⁵ No comment was made on immediate postoperative considerations for taller donors, as was done in the analyses previously discussed in this current manuscript for heavier donors, leaving these topics still to be elucidated, including the need for delayed sternal closure in taller donors or the need for support with extracorporeal membrane oxygenation in taller donors.

The literature is divided on the utility of height in paediatric heart transplantation size-matching with support for the univariate prediction of short-term outcomes and the multivariate prediction of long-term outcomes. Although height remains a readily accessible metric to potentially consider when weight may not be reliable, transplant teams may choose to overlook this metric in favour of one with greater clinical validation. Furthermore, the relative scarcity of research on the use of height alone in size-matching for paediatric heart transplantation compared with the use of weight indicates a need for further clinical evaluation. In the meantime, these findings support the continued use of weight as the primary metric in size-matching. ¹⁹

Body surface area

Body surface area is a well-defined metric in the context of cardiac surgery. It is calculated based on the height and weight of the patient and is expressed in square metres. Traditionally, body surface area provides a more accurate assessment of the overall body size of a patient compared to weight or height alone. Moreover, body surface area has been a reliable metric for other physiological processes that are relevant to outcomes after transplantation, including cardiac index, blood volume, and aortic valvar size. ^20–24 Thus, body surface area can potentially provide a distinct cumulative metric for size-matching and evaluation of outcomes. While several formulas can calculate body surface area, the primary components, namely, height and weight, are readily available metrics in transplant databases. ²⁵ While body surface area has been recognised as a metric for size-matching in adult heart transplantation, the literature on its use in paediatric heart transplantation is, again, relatively scarce. Although body surface area is frequently reported as a descriptive statistic, only a handful of studies focused on assessing its prognostic value as a metric for size-matching.

One such single institutional study conducted by Tjang and colleagues evaluated the relationship between recipient mortality and recipient body surface area, donor body surface area, and the ratio of donor-to-recipient body surface area. Although limited by a small number of patients, they found that recipient body surface area and donor body surface area were associated with 30-day mortality and recipient body surface area was associated with 1-year mortality. Furthermore, they reported that a donor-to-recipient body surface area ratio mismatch of less than 0.9, indicative of a smaller donor heart, was an independent risk factor for 10-year mortality. ²⁶

Singh and colleagues evaluated predicted left ventricular mass, calculated directly from the body surface area of a patient. They postulated that because findings support left ventricular mass as the dominant contributor to heart mass after the first 4 to 6 weeks of life, predicted left ventricular mass would be a better metric for assessing donor-to-recipient size match in paediatric heart transplantation compared to weight. ⁶ They concluded that paediatric heart transplantation recipients with the lowest donor-to-recipient predicted left ventricular mass ratio faced a significantly higher risk of in-hospital mortality. Moreover, when size match was assessed with predicted left ventricular mass as a continuous variable, undersized donors were found to carry a higher adjusted risk of in-hospital mortality. ⁶ Despite this, they reported that predicted left ventricular mass was not a superior outcome predictor compared to directly using a body surface area ratio or a donor-to-recipient body weight ratio, pointing back to the more well-established metrics for size-matching. ⁶

Additional studies sought to evaluate the influence of donor-to-recipient body surface area on the growth of transplanted hearts. These studies compared the donor body surface area at the time of transplantation with the corresponding parameters of the recipient over time. Interestingly, it was found that regardless of donor-to-recipient size mismatch, cardiac volumes and dimensions increased in size in correlation with the body surface area of the recipient. ²⁷ Walter and colleagues ultimately concluded that using an oversized or undersized donor heart within a certain range appears to be well tolerated in paediatric heart transplantation. They additionally commented on their practice guidelines, noting that measurements of body surface area in the donor up to four times greater than the recipient have been well tolerated, although they remain suspicious of smaller donors, with concern for the maintenance of postoperative perfusion in a larger recipient. ²⁷

The utility of matching of body surface area in the prediction of outcomes after paediatric heart transplantation was challenged by findings from Lowrey and colleagues. They initially hypothesised that discordance in body mass index or body surface area was more closely associated with outcomes of transplantation than discordance in weight. ²⁸ However, in an analysis of the United Network for Organ Sharing database, they found that low body mass index ratios predicted short-term mortality, while body surface area was not predictive of any outcomes. ²⁸

While body surface area has been identified as a potentially useful metric for assessing donor-to-recipient size match in paediatric heart transplantation, current evidence suggests that it may not be the best option. Studies have reported mixed results, with some finding an association between body surface area and mortality, while others reported that other metrics are more accurate and reliable. However, the lack of dependability of body surface area as a metric for size-matching could very well be a product of the non-linear correlation and complex relationship between body surface area and echocardiographic measurements in comparison to the linear relationship between body length and echocardiographic measurements. ²⁹ As such, further clinical research is required to determine the reliability of body surface area (1) as a metric for size-matching and (2) as a metric to provide valuable insights into improving outcomes associated with paediatric heart transplantation.

Imaging criteria

Although the previously discussed anthropometric variables (weight, height, and body surface area) are often used to predict heart size and adequate mediastinal space, dissent exists in the literature regarding the association of these metrics with the matching of heart size and survival of recipients. ^7,29 Cardiac imaging has been proposed as a method that can facilitate a more efficient and accurate process of size-matching. However, it should be noted that studies exploring imaging criteria for paediatric cardiac transplantation size-matching only explore the feasibility of these novel metrics and their association with established measurements, without clinical validation or measurement of commonly reported outcomes after cardiac transplantation. Nevertheless, these methods represent significant advances within the field.

Echocardiographic imaging has been found to be important in the determination of donor or recipient size and is the third most widely used size metric overall, behind traditional weight and height. ⁴ Total cardiac volume, derived from echocardiographic measurements, is one proposed method to optimise size-matching. Although there are several ways to measure total cardiac volume, one reported technique uses a modified Simpson’s method applied to a standard four-chamber apical view of an echocardiogram. ⁹ Utilising this technique, calculated echocardiographic total cardiac volumes were compared with volumes from cardiac magnetic resonance imaging, with excellent correlation. This modified Simpson’s method was determined to be a reliable indicator of total cardiac volume (concordance correlation coefficient for modified echocardiographic total cardiac volume via Simpson’s method versus cardiac magnetic resonance imaging of 0.98; 95% confidence interval, 0.97−0.99) by Camarda and colleagues. ⁹ These authors also found that in patients with structurally normal cardiac anatomy, assumed to be donors in this case, there was a linear relationship between left ventricular end-diastolic volume and total cardiac volume. Thus, this standard echocardiographic measurement can be used to calculate total cardiac volume in donors, to be compared to the measured total cardiac volume in recipients, for optimal size-matching. It was additionally reported that these techniques are favourable in paediatrics because the studied model was increasingly successful at predicting total cardiac volume in smaller hearts. ⁹ This analysis by Camarda and colleagues concluded that “echocardiographic assessment of total cardiac volume for recipients and their potential donors is a simple process and can be prospectively applied as part of donor evaluation.” ⁹

Alternative techniques for measuring donor total cardiac volume have also been described. For example, a method for estimating total cardiac volume in heart donors using clinical data and a mathematical multiple linear regression model was explored by Szugye and colleagues. ³⁰ This model was compared to total cardiac volume directly calculated using segmented computerised tomographic images from a three-dimensional reconstruction software. Though segmented computerised tomographic determination of total cardiac volume directly is ideal, it is important to note that these scans are typically unavailable for donors. Three predictive models were explored, and the model with the most accurate prediction of total cardiac volume incorporated anthropometric variables (weight and height), sex, age, and a one-view anteroposterior projection on chest radiography to visualise and measure cardiac diameter. ³⁰ It should be noted that this information, along with chest radiography, is readily available during the evaluation of a donor. Therefore, by combining cardiac imaging via chest radiography and anthropometric measures, this technique proved to be a versatile and more complete method of assessing total cardiac volume, allowing for consideration of multiple metrics to predict something that is undoubtedly multifactorial.

While estimates of total cardiac volume based on chest radiography and anthropometric measures in the study by Szugye and colleagues were accurate, their study only included patients with normal cardiac anatomy. Patients with complex congenital heart disease or those who are supported with a ventricular assist device have unique anatomy that may not be easily predicted with basic imaging and anthropometric measures. In these challenging patients, three-dimensional reconstruction using computerised tomographic images may help accurately size match patients.

Besides total cardiac volume, cardiac imaging has allowed for the development of other metrics for size-matching. Zuckerman and colleagues suggested that an echocardiographic surrogate for height might be a strong predictor of heart size. Left ventricular end-diastolic diameter, an estimator of heart size, was compared to “the distance between the junction of the right atrium with superior caval vein and the junction of the right atrium with inferior caval vein,” or “superior caval vein to inferior caval vein distance.” ⁸ Superior caval vein to inferior caval vein distance was also compared to anthropometric measurements and age. Consequently, it was found that body height was directly correlated to left ventricular end-diastolic diameter and the “superior caval vein to inferior caval vein” distance, suggesting that these imaging metrics can be effective in supporting traditional body measurements. Much like previous research related to paediatric heart transplantation, this study suggested that to increase the donor pool and facilitate more accurate size-matching, paediatric cardiac transplant recipients could be listed by height, and consideration could be given to superior caval vein to inferior caval vein distance. ⁸ This analysis by Zuckerman and colleagues concluded that “the use of height and a novel superior caval vein to inferior caval vein distance measurement to evaluate heart size in potential paediatric heart transplant recipients and donors may allow for broadening of the donor pool and creation of a more efficient and accurate size-matching process. The prospective evaluation of these novel methods with respect to clinical outcomes is necessary.” ⁸

Although imaging has been used to assess and develop new ways to optimise size-matching of donor to recipient in children, no agreed-upon method exists that has been ubiquitously adopted across institutions. Perhaps this observation emphasises the need for a more holistic process that considers multiple variables to maximise the accuracy of models of size-matching and increase the survival of recipients. Further clinical implementation of these metrics is also needed to truly compare their utility to commonly accepted metrics like weight and height. Finally, it remains critical that any metric for size-matching to determine the viability of a donor’s heart must function within the tight timeline available for the assessment of the suitability of the donor.

Conclusions

Great variation exists in the evidence for the use of different metrics for size-matching in paediatric cardiac transplantation. Even within the analysis of a single metric, such as weight, disagreement exists in the efficacy of the metric as well as the impact of oversized or undersized donor hearts. Alongside anthropomorphic values, new imaging techniques and analytic processes have yielded promising metrics to assess donors and recipients. It is a fact that any new metric must function within the tight timeline available for the assessment of the suitability of the donor. At the time of the writing of this paper, the most commonly used metric for size-matching in paediatric cardiac transplantation is the donor-to-recipient body weight ratio. Directions for future research about this topic include additional exploration of alternative metrics for size-matching, as well as analysing the potential differential utility of the various metrics for different clinical conditions such as cardiomyopathy versus congenital heart disease and for different clinical scenarios such as patients with and without ventricular assist devices. In sum, this heterogeneity in the evidence for the use of different metrics for size-matching in paediatric cardiac transplantation points to the need for additional clinically validated studies to compare the utility of a variety of metrics, in an attempt to identify those metrics that are the strongest predictors of outcomes after paediatric cardiac transplantation and of the greatest use for paediatric heart transplantation teams.