Patient radiation exposure during invasive graft surveillance after heart transplantation

Abstract

Background

Invasive procedures, including endomyocardial biopsy (EMB), right heart catheterisation (RHC) and coronary angiography (CA), are essential components of graft surveillance after heart transplantation (HTx). Fluoroscopy remains the primary imaging modality. Data on patient radiation exposure in HTx recipients are limited. The aim of this study was to investigate the kerma-area product (KAP, Gy.cm²), peak skin dose (PSD, Gy) and fluoroscopy time (s) in these procedures, assess trends in patient radiation exposure over time and identify the procedures contributing most to cumulative radiation during post-transplant surveillance.

Methods

Retrospective cohort study at a tertiary care centre including patients who underwent HTx between January 2018 and December 2019. All patients were followed through their invasive organ surveillance period with a median follow-up time of 4.9 years. For the primary end point, KAP (Gy.cm²), PSD (Gy) and fluoroscopy time (s) were assessed. Secondary end points were procedural success rate and procedure-related complications.

Results

A total of 382 procedures were performed: 330 EMB (86.4%), 22 EMB+RHC (5.8%), 30 EMB+CA with or without RHC (7.8%). The mean annual KAP from fluoroscopic procedures per patient was 50.9±3.95 Gy.cm² in the first post-transplant year, dropping to 11.1±6.1 Gy.cm² by year 5 (p<0.001). By day 387 after HTx, 50% of the total dose was reached. CA was the primary contributor to radiation peaks beyond year 1. Radiation per EMB procedure decreased significantly over time (p=0.002). Procedural success rate was 98.2%, with a 1.0% complication rate.

Conclusion

Radiation exposure is highest in the first-year post-HTx, with CA contributing most to later peaks. Time-specific and procedure-specific strategies are warranted to minimise exposure and optimise long-term care.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Heart transplant recipients are repeatedly exposed to ionising radiation due to routine graft surveillance, including endomyocardial biopsies, right heart catheterisation and coronary angiography (CA), all of which rely on fluoroscopy for imaging.

• Data on patient radiation exposure in adult heart transplant recipients in the modern fluoroscopy era are limited.

WHAT THIS STUDY ADDS

• This study provides contemporary real-world data on radiation exposure during invasive graft surveillance in adult heart transplant recipients.

• It demonstrates that institutional radiation doses can be kept relatively low using optimised imaging protocols, and that cumulative radiation exposure declines significantly over time, with approximately half occurring within the first-year post-transplant.

• Endomyocardial biopsy was identified as the main contributor to overall radiation burden.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• By documenting contemporary patient radiation values and trends, this study offers reference points for other heart transplant centres supporting the optimisation of procedural protocols and regular staff training to enhance patient safety.

Introduction

Heart transplantation (HTx) remains the gold standard therapy for advanced heart failure. Progress in surgical techniques, perioperative care and medical management, including advances in immunosuppressive therapies, has significantly improved outcomes over recent decades, with a current median survival rate of approximately 12.5 years after HTx.1,3

Despite these improvements, graft failure remains the leading cause of mortality within 5 years after HTx.¹ Therefore, rejection surveillance through endomyocardial biopsy (EMB) and right heart catheterisation (RHC), as well as screening for cardiac allograft vasculopathy (CAV) with coronary angiography (CA), are essential components of post-transplant follow-up.⁴ These procedures are typically performed under fluoroscopic guidance to ensure procedural safety and accuracy, although this inevitably exposes patients to ionising radiation. The risks associated with ionising radiation include stochastic effects, such as cancer, and deterministic effects, such as skin injuries and cataracts.5,8 These concerns are particularly relevant in HTx recipients, who undergo repeated imaging over many years and are a vulnerable patient population due to immunosuppressive therapy and comorbidities.⁹ Therefore, it is imperative to minimise radiation exposure to the lowest achievable level according to the As Low As Reasonably Achievable principle while maintaining procedure safety and efficacy.¹⁰

While considerable attention has been given to radiation optimisation strategies in the paediatric transplant population,^{11 12} data among adult HTx recipients in the modern fluoroscopy era remain limited.^{9 13} The aims of this study were to (1) define institutional typical radiation values for invasive procedures used for graft surveillance, (2) assess trends in radiation exposure over time in adult HTx recipients and (3) identify the procedures contributing most to cumulative radiation during post-transplant surveillance.

Methods

We conducted a single-centre retrospective cohort study including all patients who underwent an HTx between January 2018 and December 2019 at the University Hospital Zürich. This study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines (see online supplemental file 2).

Procedures for invasive organ surveillance included EMB, RHC and CA. The standard invasive surveillance protocol used at our centre is shown in the online supplemental table 1. All patients were followed throughout their invasive organ surveillance period. Patient radiation exposure was assessed using the kerma-area product (KAP, in Gy.cm²), peak skin dose (PSD, in Gy) and fluoroscopy time (t, in s). KAP is the product of the dose at the centre of a certain plane of the X-ray beam multiplied by the area of the X-ray field at that plane and is useful for estimating overall radiation risk. PSD, defined as the maximum dose received by any point on the patient’s skin, was chosen over cumulative dose because it directly reflects the maximum localised radiation burden, accounting for beam geometry, overlap, backscatter and patient positioning, and thus provides a more accurate indicator of deterministic skin injury risk.¹⁴ It corresponds to the fluoroscopy time recorded during the procedure. All procedures were performed with a biplane cardiology system Artis Q.zen biplane (Siemens Healthcare) equipped with two flat-panel detectors (20 cm×20 cm).

As secondary end points, procedural success rate and procedure-related complications were assessed. Procedure-related complications were defined as cardiovascular death, myocardial infarction, stroke, haemodynamically relevant pericardial effusion, iatrogenic coronary artery dissection, access site complications, arrhythmias, conduction abnormalities and severe tricuspid regurgitation.^{15 16}

Statistical analysis

Data were collected from a dose management system (DOSE, Qaelum, Belgium) and was cross-checked for accuracy with the patient information system. The prespecified primary outcome was patient radiation dose, assessed using KAP and PSD, with fluoroscopy time recorded as a procedural parameter. Institutional typical values for different procedures were calculated using the median and the first and third quartiles of overall KAP, PSD and fluoroscopy time. For comparison of the annual population-level KAP, the mean KAP and SEM were plotted by year. Comparison between years was made with repeated-measures analysis of variance. To better understand temporal trends, KAP values were first aggregated using a 90-day rolling window and subsequently smoothed using locally estimated scatterplot smoothing (LOESS).¹⁷ LOESS smoothing was applied separately for each procedure group. A linear-mixed effects model was used to examine changes in KAP for EMB procedures over time. Outliers were identified using standardised residuals from the linear mixed-effects model. Observations with absolute standardised residuals >2.5 were excluded. All hypotheses testing was two-tailed and p values <0.05 were considered statistically significant. Analyses were carried out with R V.4.2.3 (R Foundation for Statistical Computing, Vienna, Austria; https://www.R-project.org/).

Results

In total, 20 HTx patients were included in the study. The median age was 49 years, the median body mass index was 25.4 kg/m² and 70% were male. Reasons for HTx were ischaemic heart disease (45%), dilated cardiomyopathy (10%), hypertrophic cardiomyopathy (5%), amyloidosis (10%), congenital heart disease (10%) and other heart diseases (20%). The bicaval technique was used for all HTx procedures. Over a median follow-up time of 4.9 years, patients underwent a total of 382 procedures for invasive organ surveillance, of which 330 (86.4%) were EMB, 22 (5.8%) were EMB plus RHC and 30 (7.8%) were EMB plus CA with or without RHC. Per patient, the mean/median number of EMB was 18.3/20, EMB plus RHC was 2.2/1.5 and EMB plus CA with or without RHC was 2/2. Twenty-nine CAs were diagnostic; one was a percutaneous coronary intervention. Right jugular access was used in 344 procedures (90.1%), femoral access in 26 (6.8%), right subclavian access in two (0.5%) and combined access in 10 procedures (2.6%).

Institutional typical values for KAP, PSD and fluoroscopy time by procedure type are shown in table 1. The median KAP values per procedure were 2.7 Gy.cm² for EMB, 5.1 Gy.cm² for EMB plus RHC and 18.4 Gy.cm² for EMB plus CA with or without RHC. The mean/median KAP per patient for EMB was 66.1/58.9 Gy.cm², for EMB plus RHC was 12.3/7.7 Gy.cm² and for EMB plus CA with or without RHC was 56.9/43.5 Gy.cm². Among procedures, there was an interoperator variability in radiation exposure to the patient (see online supplemental tables 2 and 3). Overall, patient radiation exposure significantly decreased with increasing time since HTx during the post-transplant follow-up period (median follow-up 4.9 years; p<0.001). In the first year following HTx, the mean cumulative radiation exposure per patient was 50.9 (±3.95) Gy.cm² and by 5 years post-transplantation, it had decreased to 11.1 (±6.1) Gy.cm² (figure 1a). EMBs accounted for the largest share of total KAP (figure 1b). By day 387 after HTx, 50% of the cumulative radiation exposure over the follow-up period was reached (figure 1c, upper panel). Radiation exposure showed annual peaks during follow-up, initially driven by EMBs (up to day 330) and later by EMBs plus CA with or without RHC (figure 1c, lower panel). The first peak in radiation exposure was attributable both to a greater number of EMBs and to higher per-EMB procedural doses, which declined significantly over time (p=0.002) (figure 2).

Table 1. Institutional typical values shown for KAP, PSD and fluoroscopy time for EMB, EMB plus RHC and EMB plus CA with or without RHC.

Type of procedure	N	KAP (Gy.cm2)	PSD (Gy)	Fluoroscopy time (s)
EMB	330	2.7 (1.6–4.9)	0.02 (0.01–0.04)	156 (99.2–243)
EMB+RHC	22	5.1 (2.9–8.3)	0.04 (0.02–0.08)	231 (162–326)
EMB+CA±RHC	30	18.4 (14.1–34.4)	0.13 (0.08–0.21)	651 (535–872)

Values are presented as medians, with the first and third quartiles in parentheses.

CA, coronary angiography; EMB, endomyocardial biopsy; KAP, kerma-area product; PSD, peak skin dose; RHC, right heart catheterisation.

Figure 1. (a) Annual population-level KAP. Mean KAP and SEM plotted by year, shown as blue line with whiskers. Individual values for each year are reported as grey dots. (b) Total amount of KAP from each procedure type, shown as pie chart. (c) Population-level rolling average of cumulative KAP exposure. Upper panel: blue line with a shaded SEM represents a smoothed 90-day rolling average of KAP exposure across all patients over time. Light blue ribbon under the curve representing non-smoothed data. Dashed vertical red line indicates the time point at which 50% of the total cumulative population KAP exposure has occurred. Lower panel: population-level 90-day rolling average of KAP over time, decomposed by procedure type. Shaded ribbons indicate the SEM. ANOVA, analysis of variance; CA, coronary angiography; EMB, endomyocardial biopsy; HTx, heart transplantation; KAP, kerma-area product; RHC, right heart catheterisation.

Figure 2. (a) KAP per EMB procedure over time. Scatter plot indicating the relationship between KAP and time from HTx. Each dot point represents an individual EMB, with different colours corresponding to patients. The black trend line represents the fixed effect from a linear mixed-effects model. There was a significant fixed effect of time on KAP per EMB procedure. (b) Mean KAP per patient for EMB procedures. Mean KAP and SEM plotted by year, shown as blue line with whiskers. Grey dots representing individual patients. Linear trend shown as dashed line. EMB, endomyocardial biopsy; HTx, heart transplantation; KAP, kerma-area product.

The procedural success rate was 98.2% (375/382), as seven EMBs could not be histologically evaluated due to non-representative tissue. Procedure-related complications occurred in four (1.0%) cases, of which three were atrial fibrillation/flutter and one was rupture of a second-degree chorda tendinea of the tricuspid valve.

Discussion

This retrospective study assessed patient radiation exposure during procedures used for invasive organ surveillance in adult HTx recipients and investigated radiation trends over time. We report three findings:

1. First, our institution’s typical radiation values were lower than those reported in a previous single-centre study,¹³ with a KAP of 5.1 Gy.cm² for combined EMB and RHC procedures (vs 8.7 Gy.cm²) and 18.4 Gy.cm² for EMB combined with CA (vs 53.9 Gy.cm²). For isolated EMB procedures in the adult population, published data are limited; however, our findings are consistent with those of another single-centre study¹⁸ involving a small cohort of 63 cardiac biopsies, which reported a KAP of 2.7 Gy.cm². When compared with other interventional procedures performed at our institution,⁷ the radiation dose for EMB was comparable to that of pacemaker or implantable cardioverter defibrillator lead extractions, with a median KAP of 3.5 Gy.cm². These relatively low values may be explained by our institutional imaging protocol, which includes the use of low frame rates (five frames per second), consistent collimation and reduced radiation dose per pulse. The procedural success rate was 98.2% and procedure-related complications occurred in 1.0% of procedures. Previous retrospective data in 175 HTx recipients reported an overall EMB complication rate of 0.71%.¹⁶ These findings support the effectiveness of our radiation-reduction measures, including regular staff training to increase awareness of radiation safety in clinical practice.

2. Second, we observed a significant decline in patient radiation exposure over time (p<0.001), with a fivefold reduction from the first to the fifth year after HTx. The first year alone accounted for approximately 50% of the total cumulative radiation burden during invasive organ surveillance. This is in line with the results from a study with 202 HTx recipients over a 10-year follow-up period.⁹

3. Third, EMB procedures accounted for the largest share of cumulative KAP. Radiation exposure peaked in the first year after HTx due to a higher frequency of EMBs and increased radiation doses per EMB procedure, which showed a significant reduction over time (p=0.002). The increased radiation exposure during early post-transplant EMBs is likely attributable to technical challenges specific to this period. These include difficult navigation across the superior vena cava anastomosis, challenging bioptome navigation due to the presence of other catheters, such as dialysis catheter and a more cautious approach due to healing tissue and increased myocardial perforation risk. Additionally, increased awareness of radiation protection measures led to a reduction from 15 to 7.5 images per second for EMB procedures.

These findings highlight the importance of enhanced radiation safety measures particularly within the first year after HTx. In the following years, recurring annual peaks in radiation exposure were mainly driven by CA procedures, which represent additional opportunities for targeted dose-reduction strategies. Technological advances in coronary computed tomographic angiography (CCTA) have led to its adoption as a non-invasive alternative to CA,^{4 19} although radiation doses may vary substantially between and within centres due to patient-related, technology-related and operator-related factors.²⁰ In our patient cohort, CCTA was performed every 2 years for CAV screening in six patients. Beyond procedure-specific optimisation, emphasis should be placed on systemically identifying patient pathways that involve multiple repeated imaging and interventional procedures for a more comprehensive and longitudinal approach to radiation dose optimisation.

During follow-up, 15% of patients developed cancer (two patients developed skin cancer and one patient developed renal cell carcinoma). However, epidemiological data suggest that despite the increased cumulative radiation exposure in HTx recipients, the associated lifetime cancer risk remains relatively low, with an estimated additional radiation-associated cancer risk of 0.34% from a 10-year cumulative radiation exposure dose.^{9 21} In any case, cumulative radiation exposure should not preclude performing a radiological procedure if it provides clinically relevant information for patient management.

The limitations of the study are the small sample size and rather short follow-up period, as well as the single-centre design, which might limit the generalisability of the findings. A multicentre approach would be valuable to more comprehensively assess radiation exposure in HTx recipients.

Conclusion

Our findings offer guidance on key time periods during heart transplant follow-up where increased attention to radiation protection is advisable. The use of imaging protocols with low frame rates and consistent collimation may represent practical approaches to reducing radiation exposure. Further research is necessary to explore dose-reduction techniques and assess their impact on patient outcomes.

Data availability statement

Data are available on reasonable request.