ABSTRACT
Background
The first 35 years of pediatric heart transplantation (pHTx) in Sweden were investigated to determine outcomes following listing and transplantation, investigate sub‐populations of recipients, and describe the presence of donor‐specific antibodies (DSA) in a contemporary cohort.
Methods
Swedish children < 18 years, listed from 1/1/1989 to 31/12/2023, were included. The cohort was split based on the era of transplantation (ERA I: 1989–2008, ERA II: 2009–2023).
Results
A total of 254 children were listed and 185 (72.8%) reached pHTx, with no loss to follow‐up. Waiting list duration was 62 days and increased over time, while mortality on the waiting list decreased (30.5% in ERA I, 8.8% in ERA II). Congenital heart disease was the etiology of heart failure in 36.2% of recipients, including 24.9% with univentricular physiology. The frequency of ABO‐incompatible transplantations was 9.3% and 8.0% were considered to be at high immunological risk pre‐pHTx due to pre‐formed HLA‐antibodies with mean fluorescence intensity ≥ 5000. Ventricular assist device (VAD) was used in 26.9% of recipients. Long‐term survival was not affected by age, heart failure etiology, the use of pre‐transplant VAD, or elevated baseline indexed pulmonary vascular resistance. Era of transplantation was a determinant of listing, but not post‐pHTx outcome. Survival at 1‐, 10‐, and 30‐year follow‐up was 94.5%, 79.4%, and 57.1%, respectively. Of the total de novo DSA burden, 45.9% were HLA‐DQ‐type specific. Re‐transplantation was performed in 5.9% of recipients.
Conclusions
A high quality of care has been achieved in Sweden, despite modest pHTx numbers, in cooperation with the Scandiatransplant organization.
Keywords: donor‐specific antibodies, end‐stage heart failure, era‐dependent analysis, pediatric heart transplantation, pulmonary vascular resistance
The first 35 years of pediatric heart transplantation (pHTx) in Sweden were investigated to determine outcomes following listing and transplantation, investigate sub‐populations of recipients, and describe the presence of donor‐specific antibodies (DSA) in a contemporary cohort. A high quality of care has been achieved in Sweden, despite modest pHTx numbers, in cooperation with the Scandiatransplant organization.
Abbreviations
- ABOi
ABO‐incompatible transplantation
- AVT
acute vasoreactivity testing
- C1q
Complement component 1q
- CAV
coronary allograft vasculopathy
- CDC
complement‐dependent cytotoxicity
- CHD
congenital heart disease
- CM
cardiomyopathy
- DBD
donation after brain death
- DCD
donation after circulatory death
- DCM
dilated cardiomyopathy
- dnDSA
de novo donor‐specific antibodies
- ECMO
extracorporeal membrane oxygenation
- HCM
hypertrophic cardiomyopathy
- HLA
human leukocyte antigen
- MCS
mechanical circulatory support
- PH
pulmonary hypertension
- pHTx
pediatric heart transplantation
- PRA
panel‐reactive antibodies
- PVRi
indexed pulmonary vascular resistance
- RCM
restrictive cardiomyopathy
- Re‐pHTx
pediatric heart re‐transplantation
- UVH
univentricular heart
- VAD
ventricular assist device
1. Introduction
Since its inception in 1989, pediatric heart transplantation (pHTx) in Sweden has been centralized to Queen Silvia Children's Hospital in Gothenburg and Skåne University Hospital in Lund, which together provide for ~10.5 million inhabitants. Following the first two decades (1989–2009), a waiting list mortality of 31% and a post‐transplant 10‐year survival of 76% were reported by Gilljam et al. [1]. Sweden is part of Scandiatransplant, an organ allocation organization that coordinates all listing and solid organ exchange in the Nordic countries (Norway, Denmark, Finland, Iceland and Sweden) and Estonia [2], covering the needs of ~30 million inhabitants. The impact of short‐ and long‐term mechanical circulatory support (MCS), pulmonary hypertension (PH) and the presence of donor‐specific antibodies (DSA) on pHTx outcomes in Sweden remains to be determined.
The use of MCS as a bridge to pHTx, especially in small children and infants, has increased worldwide in the contemporary era [3, 4, 5]. Data on post‐transplant outcomes remain scarce [6, 7]. Pediatric ventricular assist device (VAD) and ECMO programs involve great risks [8, 9] and should be closely monitored [9, 10, 11].
PH in children with end‐stage heart failure is most often caused by pulmonary venous congestion secondary to high left atrial pressure; thus, post‐capillary PH [12]. If left untreated, pulmonary arterial vasoconstriction and remodeling ensue [13, 14], causing combined pre‐ and post‐capillary PH with elevated pulmonary vascular resistance index (PVRi). This may preclude children from pHTx, although data on the importance of elevated PVRi for long‐term outcomes following pHTx is ambiguous [12]. Practice for hemodynamic evaluation and acute vasoreactivity testing (AVT), as well as clinical management of elevated PVRi, displays significant heterogeneity in this patient group [15].
De novo DSA (dnDSA) has been detected in as many as 33%–43% of children following pHTx [16] and are increasingly being linked to antibody‐mediated rejection (AMR), coronary allograft vasculopathy (CAV) and poor long‐term outcomes in both the pediatric and adult populations [16, 17]. Interestingly, Dipchand et al. postulated a B‐cell memory response to be the cause of a significant amount of early detected dnDSA in children [17], suggesting pre‐pHTx sensitizing events despite the absence of significant levels of HLA‐antibodies pre‐pHTx.
The current study aimed to describe donor and recipient populations, and outcomes, as well as pHTx management during the first 35 years of pHTx practice, 1989–2023, in Sweden. Specific emphasis was put on exploring the impact of elevated PVRi, pre‐transplant short‐ and long‐term MCS, elevated immunological risk, univentricular physiology, and ABO‐incompatible (ABOi) transplant on post‐transplant survival. Additionally, we sought to describe the development of high‐risk dnDSA in a contemporary Swedish cohort of pHTx recipients.
2. Materials and Methods
2.1. Study Design
This was a nationwide, multicenter, retrospective observational cohort study approved by the Swedish Ethical Review Authority (Dnr 2020‐02140, Dnr 2023‐05036‐02 and Dnr 2024‐05951‐02) and conducted in accordance with the International Society for Heart and Lung Transplantation ethical statement. The requirement of informed consent was waived due to the retrospective nature of the study. Reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) [18] and Reporting of Studies Conducted Using Observational Routinely Collected Health Data (RECORD) guidelines [19].
2.2. Study Population and Data Sources
All children < 18 years residing in Sweden that were listed for pHTx, including simultaneous listing for other organs, between January 1st 1989 and December 31st 2023 were identified through the Scandiatransplant registry (Aarhus University Hospital, Aarhus, Denmark). Children listed for combined heart‐lung transplantation were excluded from this study, as they were considered to have significantly different clinical courses. Data at waiting list entry, as well as recipient and donor characteristics at both listing and transplantation, were acquired from the Scandiatransplant registry. When registry records were incomplete, complementary data were collected from the patients' medical records. Immunological data was obtained from HLA Fusion (One Lambda/Thermo Fisher Scientific), ProSang (Omda AS) and Scandiatransplant. All follow‐up data until December 31st of 2023 was included. Immunological data on HLA type and typing on the allelic level of resolution, dnDSA mean fluorescence intensity (MFI) values, complement‐dependent cytotoxicity (CDC) screening against HLA well‐types, healthy donors, and crossmatch tests (CDC and flow cytometry) were available and collected for patients transplanted from January 1st 2009 until December 31st 2023. Population data were collected from Statistics Sweden.
2.3. Definitions
Patients were grouped by era of transplantation, based on the availability of detailed immunological data, generating two eras: January 1st 1989–December 31st 2008 (ERA I, 20 years, no detailed immunological data available), January 1st 2009–December 31st 2023 (ERA II, 15 years, detailed immunological data available). A minimum duration of 1 day was considered for both waiting list time and post‐transplant survival. The total number of days on the waiting list was accounted for. Patients that deteriorated and died shortly after being taken off the list were counted as deceased in the waiting list mortality analysis. The end of follow‐up, and thus the censor date, was December 31st 2023. Patients withdrawn from the list due to improvement or deterioration, but later re‐listed and transplanted, were included only once in the waiting list and post‐transplant analyses. The second listing occasion, preceding pHTx or death, was used for defining the start of waiting list duration and for analysis of patient listing characteristics and outcomes in these patients. Patients listed for pediatric heart re‐transplantation (re‐pHTx) and re‐pHTx‐recipients were excluded from the main analyses and are commented on separately.
2.4. Immunological Methods and Analyses
Details on HLA‐typing and matching, HLA‐detection and identification, virtual panel reactive antibody (vPRA) calculations, and crossmatch test are all available in Data S1. Antibody levels were expressed as MFI and were rounded to two value figures due to the semi‐quantitative nature of the analysis. MFI > 1000 was considered positive for any given HLA antibody [20]. An HLA antibody was considered persistent if it displayed > 1 positive test with at least 6 months in between individual tests, whereas isolated positive findings were considered transient. Patients displaying any preformed HLA‐antibodies with MFI > 1000 were considered to have an elevated immunological risk, while patients displaying any preformed HLA‐antibodies with MFI > 5000 were considered to have a high immunological risk. Patients with only HLA‐antibodies with MFI < 1000 were considered to have normal immunological risk. One vPRA value was calculated for all preformed HLA‐antibodies with MFI > 1000. In addition, for all preformed HLA‐antibodies with MFI > 5000, the suggested MFI limit for high immunological risk with an increased risk for acute AMR [21], a second vPRA percentage was calculated. For HLA‐A, HLA‐B, HLA‐DR, and HLA‐DQ DSA, both de novo and preformed, antibodies were considered clinically significant if MFI > 1000. For HLA‐Cw, HLA‐DP, and HLA‐DR3/4/5, both de novo and preformed, antibodies were considered clinically significant if MFI > 2000. For HLA mismatch analysis, HLA‐A, B, C, DR, and DQ were included. Whilst on the waiting list, antibody screening was performed every 3 months. Follow‐up after detection of DSA‐antibodies was guided individually, depending on clinical suspicion of rejection.
2.5. Statistical Analysis
Normality distribution was assessed with histograms, normality tests (Kolmogorov–Smirnov and Shapiro–Wilk), Q–Q plots, and skewness. Variables are described using frequencies and percentages (categorical variables) and mean and standard deviation (SD) or medians and interquartile range (IQR) or range (continuous variables). Continuous variables were compared by independent samples t test or Mann–Whitney test (2 groups) and one‐way ANOVA or Kruskal‐Wallis test (> 2 groups). Categorical variables were compared by Fisher's exact test (2 groups) or Pearson χ 2 test (> 2 groups). All patients contributed time (in days) on the waiting list until removal from the list (death, pHTx or withdrawal) or the censor date, whichever came first. Additionally, transplanted patients contributed follow‐up time from the date of pHTx until the date of death or the censor date. Survival rates for all‐cause mortality were estimated using the Kaplan–Meier method, and the log‐rank test was used to compare differences between the curves. Cox proportional hazard regression was used with and without stepwise multivariable adjustment to estimate hazard ratio (HR) and 95% confidence interval (CI) for the association between era of transplantation and waiting list mortality or post‐transplant survival, with ERA I as the reference time period. Variables were selected a priori based on clinical experience. Mortality on the waiting list was adjusted for age, sex, blood group, and diagnosis. Post‐transplant survival was adjusted for recipient and donor sex, recipient and donor age groups, recipient and donor weight mismatch ratio at transplant, diagnosis, urgency, pre‐transplant kidney function (estimated or measured glomerular filtration rate [GFR]), cold ischemia time, cytomegalovirus (CMV) mismatch, ABOi transplant, and pretransplant treatment with a VAD or extracorporeal membrane oxygenation (ECMO). Simple linear regression analysis was performed to investigate the relationship between pre‐ or post‐AVT PVRi and post‐pHTx survival time. A 2‐sided p value of < 0.05 was considered statistically significant. Analyses were performed using Stata software (StataSE, version 17.0, StataCorp).
3. Results
3.1. Study Population, Recipient and Donor Characteristics
A total of 272 children were listed for pHTx during the study period. Eighteen children listed for combined heart‐lung transplantation were excluded, leaving 254 children to be included in the study. Three patients on the waiting list were transplanted abroad but had all follow‐up visits in Sweden. Characteristics at listing and transplantation are outlined in Tables 1 and 2, respectively. Figure 1 contains a flow chart of the included study population.
TABLE 1.
Patient characteristics at wait list entry.
| Listing characteristics | ||||
|---|---|---|---|---|
| ERA I (n = 118) | ERA II (n = 136) | Total (n = 254) | p | |
| Age at listing, years | 7.7 (IQR 0.8–14.0, range 2 days–17.6 years) | 9.4 (IQR 2.1–14.3, range 18 days–17.9 years) | 8.3 (IQR 1.6–14.2, range 2 days–17.9 years) | 0.422 |
| < 1 year | 31 (26.3%) | 22 (16.2%) | 53 (20.9%) | 0.079 |
| 1–10 years | 39 (33.1%) | 60 (44.1%) | 99 (39.0%) | |
| 11–17 years | 48 (40.7%) | 54 (39.7%) | 102 (40.2%) | |
| Sex, female | 52 (44.1%) | 62 (45.6%) | 114 (44.9%) | 0.808 |
| ABO | (n = 115) | (n = 136) | (n = 251) | |
| A | 56 (48.7%) | 58 (42.7%) | 114 (45.4%) | 0.177 |
| B | 7 (6.1%) | 18 (13.2%) | 25 (10.0%) | |
| AB | 11 (9.6%) | 8 (5.9%) | 19 (7.6%) | |
| O | 41 (35.7%) | 52 (38.2%) | 93 (37.1%) | |
| Diagnosis | (n = 118) | (n = 136) | (n = 254) | |
| CHD | 53 (44.9%) | 48 (35.3%) | 101 (39.8%) | 0.118 a |
| UVH | 33 (28.0%) | 38 (27.9%) | 71 (28.0%) | 0.996 |
| HLHS | 14 (11.9%) | 21 (15.4%) | 35 (13.8%) | 0.409 |
| CM | 65 (55.1%) | 88 (64.7%) | 153 (60.2%) | 0.118 a |
| DCM | 56 (47.5%) | 64 (47.1%) | 120 (47.2%) | 0.075 |
| HCM | 4 (3.4%) | 7 (5.2%) | 11 (4.3%) | |
| RCM | 5 (4.2%) | 17 (12.5%) | 22 (8.7%) | |
| Listing outcome | (n = 118) | (n = 136) | (n = 254) | |
| Death | 36 (30.5%) | 12 (8.8%) | 48 (18.9%) | < 0.001 |
| pHTx | 72 (61.0%) | 113 (83.1%) | 185 (72.8%) | |
| Withdrawal b | 10 (8.5%) | 8 (5.9%) | 18 (7.1%) | |
| On list | 0 (0.0%) | 3 (2.2%) | 3 (1.2%) | |
| Wait list duration, days c | 45 (IQR 12–115, range 1–527) | 79 (IQR 20–188, range 1–816) | 62 (IQR 13–154, range 1–816) | 0.005 |
Note: ERA I: 1989–2008, ERA II: 2009–2023. Variables are described using frequencies and percentages for categorical variables and median with interquartile range for continuous variables. Continuous variables were compared by Kruskal‐Wallis test. Binary and categorical variables were compared by Pearson χ 2 test.
Abbreviations: CHD, congenital heart disease; CM, cardiomyopathy; DCM, dilated cardiomyopathy; HCM, hypertrophic cardiomyopathy; HLHS, hypoplastic left heart syndrome; pHTx, pediatric heart transplantation; RCM, restrictive cardiomyopathy; UVH, univentricular heart.
p‐value for the comparison between CHD and CM.
Three patients in ERA II were withdrawn from the list, but later relisted and transplanted. Aiming to study patients, not listing occasion, outcomes and survival, we opted to present the listing occasion terminated with HTx in this table. Adjusting the frequencies to account for the primary listing occasion instead, the withdrawal frequency in ERA II and the total time period would have been 11 (8.1%) and 21 (8.3%) respectively, with p < 0.001 in the comparison between ERA I and ERA II.
Three patients in ERA II were still on the waiting list at the censor date (2023‐12‐31) and thus only contributed wait list duration until this date.
TABLE 2.
Recipient and donor characteristics at pediatric heart transplant.
| Recipient and donor characteristics | ||||
|---|---|---|---|---|
| ERA I (n = 72) | ERA II (n = 113) | Total (n = 185) | p | |
| Recipient age at transplant, years | 12.7 (IQR 5.5–15.7, range 0.1–17.7) | 11.4 (IQR 4.2–14.9, range 25 days‐18.7 years) | 11.6 (IQR 4.6–15.0, range 25 days‐18.7 years) | 0.469 |
| < 1 year | 10 (13.9%) | 8 (7.1%) | 18 (9.7%) | 0.164 |
| 1–10 years | 21 (29.2%) | 45 (39.8%) | 66 (35.7%) | 0.164 |
| 11–17 years | 41 (56.9%) | 60 (53.1%) | 101 (54.6%) | 0.164 |
| Donor age, years | 20 (8–37), n = 48 | 18 (6–37) | 20 (7–37), n = 161 | 0.816 |
| Recipient sex, female | 31 (43.1%) | 50 (44.3%) | 81 (43.8%) | 0.881 |
| Donor sex, female | 36 (50.7%), n = 71 | 61 (54.5%), n = 112 | 97 (53.0%), n = 183 | 0.651 |
| Recipient ABO | n = 72 | n = 113 | n = 185 | 0.123 |
| A | 40 (55.6%) | 48 (42.5%) | 88 (47.6%) | |
| B | 4 (5.6%) | 17 (15.0%) | 21 (11.4%) | |
| AB | 6 (8.3%) | 7 (6.2%) | 13 (7.0%) | |
| O | 22 (30.6%) | 41 (36.3%) | 63 (34.1%) | |
| Donor ABO | n = 70 | n = 113 | n = 183 | 0.487 |
| A | 31 (44.3%) | 44 (38.9%) | 75 (41.0%) | |
| B | 5 (7.1%) | 15 (13.3%) | 20 (10.9%) | |
| AB | 0 (0.0%) | 1 (0.9%) | 1 (0.6%) | |
| O | 34 (48.6%) | 53 (46.9%) | 87 (47.5%) | |
| ABOi | n = 70 | n = 113 | n = 183 | |
| Major a | 2 (2.9%) | 15 (13.3%) | 17 (9.3%) | 0.060 |
| Minor b | 17 (24.3%) | 23 (20.4%) | 40 (21.9%) | 0.060 |
| Positive crossmatch | 0 (0.0%), n = 63 | 5 (4.5%), n = 112 | 5 c (2.9%), n = 175 | 0.089 |
| Donor/recipient weight ratio d | 1.4 (IQR 1.1–1.7, range 0.9–3.3), n = 67 | 1.4 (IQR 1.1–1.8, range 0.7–3.6) | 1.4 (IQR 1.1–1.8, range 0.7–3.6), n = 180 | 0.984 |
| Diagnosis | n = 72 | n = 113 | n = 185 | |
| CHD | 27 (37.5%) | 40 (35.4%) | 67 (36.2%) | 0.772 |
| UVH | 14 (19.4%) | 32 (28.3%) | 46 (24.9%) | 0.173 |
| HLHS | 7 (9.7%) | 16 (14.2%) | 23 (12.4%) | 0.373 |
| CM | 45 (62.5%) | 73 (64.6%) | 118 (63.8%) | 0.772 |
| DCM | 41 (56.9%) | 52 (46.0%) | 93 (50.3%) | 0.085 |
| HCM | 1 (1.4%) | 6 (5.3%) | 7 (3.8%) | |
| RCM | 3 (4.2%) | 15 (13.3%) | 18 (9.7%) | |
| Wait list duration, days | 58 (IQR 12.5–120, range 1–527) | 78 (IQR 16–180, range 1–816) | 67 (IQR 13–163, range 1–816) | 0.161 |
| Highly urgent at transplant | 26 (37.1%), n = 70 | 34 (30.1%) | 60 (32.8%), n = 183 | 0.336 |
| CMV+ | ||||
| Recipient | 41 (57.8%), n = 71 | 52 (46.0%) | 93 (50.5%), n = 184 | 0.132 |
| Donor | 35 (60.3%), n = 58 | 63 (57.3%), n = 110 | 98 (58.3%), n = 168 | 0.744 |
| EBV+ | ||||
| Recipient | 40 (59.7%), n = 67 | 59 (52.2%) | 99 (55.0%), n = 180 | 0.355 |
| Donor | 19 (73.1%), n = 26 | 81 (77.1%), n = 105 | 100 (76.3%), n = 131 | 0.797 |
| Recipient preoperative status | ||||
| Pacemaker | 19 (27.5%), n = 69 | 21 (18.6%) | 40 (22.0%), n = 182 | 0.197 |
| VAD | 12 (17.4%), n = 69 | 37 (32.7%) | 49 (26.9%), n = 182 | 0.026 |
| ECMO | 1 (1.5%), n = 67 | 9 (7.7%) | 10 (5.6%), n = 180 | 0.093 |
| eGFR (mL/min/1.73m2) | 79 (IQR 70–90.5, range 51–137), n = 56 | 86 (IQR 73–106, range 29–191), n = 97 | 83 (IQR 71–97, range 29–191), n = 153 | 0.069 |
| PVRi, baseline (WU·m2) | 3.7 (IQR 2.2–6.3, range 1.4–13.0), n = 36 | 3 (IQR 1.7–4.8, range 0.4–19.0), n = 63 | 3.3 (IQR 2–5.40, range 0.4–19.0), n = 99 | 0.067 |
| PVRi, post‐AVT (WU·m2) | 3.6 (IQR 2.6–4.3, range 1.5–5.1), n = 13 | 3.0 (IQR 2.0–4.5, range 0.6–7.9), n = 17 | 3.1 (IQR 2.3–4.5, range 0.6–7.9), n = 30 | 0.901 |
| Cold ischemia time (minutes) | 199 ± 60, n = 70 | 199 ± 65, n = 110 | 199 ± 63, n = 180 | 0.962 |
| Recipient postoperative status | ||||
| Pacemaker | 3 (4.7%), n = 64 | 3 (2.7%) | 6 (3.4%), n = 177 | 0.669 |
| VAD | 3 (4.6%), n = 66 | 0 (0.0%) | 3 (1.7%), n = 179 | 0.049 |
| ECMO | 3 (4.8%), n = 63 | 13 (11.5%) | 16 (9.1%), n = 176 | 0.176 |
| Dialysis | 3 (4.7%), n = 64 | 7 (6.3%), n = 112 | 10 (5.7%), n = 176 | 0.749 |
| Immunosuppression, induction | n = 69 | n = 112 | n = 181 | |
| ATG | 66 (95.7%) | 76 (67.9%) | 142 (78.5%) | < 0.001 |
| Basiliximab | 1 (1.5%) | 33 (29.5%) | 34 (18.8%) | < 0.001 |
| Daclizumab | 1 (1.5%) | 3 (2.7%) | 4 (2.2%) | < 0.001 |
| Steroids | 64 (92.8%) | 112 (100%) | 176 (97.2%) | 0.007 |
| Other e | 0 (0.0%) | 1 (0.9%) | 1 (0.6%) | 1.000 |
| Immunosuppression, discharge | n = 69 | n = 112 | n = 181 | |
| Azathioprin | 53 (76.8%) | 8 (7.1%) | 61 (33.7%) | < 0.001 |
| MMF | 18 (26.1%) | 111 (99.1%) | 129 (71.3%) | < 0.001 |
| Cyclosporin | 44 (63.8%) | 0 (0.0%) | 44 (24.3%) | < 0.001 |
| Tacrolimus | 19 (27.5%) | 112 (100%) | 131 (72.4%) | < 0.001 |
| Immunosuppression, last follow‐up | n = 67 | n = 112 | n = 179 | |
| Prednisone | 18 (26.9%) | 16 (14.3%) | 34 (19.0%) | 0.049 |
| Azathioprin | 11 (16.4%) | 2 (1.8%) | 13 (7.3%) | < 0.001 |
| MMF | 41 (61.2%) | 92 (82.1%) | 133 (74.3%) | 0.003 |
| Cyclosporin | 21 (31.3%) | 9 (8.0%) | 30 (16.7%) | < 0.001 |
| Tacrolimus | 39 (58.2%) | 102 (91.1%) | 141 (78.8%) | < 0.001 |
| Everolimus | 11 (16.4%) | 35 (31.3%) | 46 (25.7%) | 0.034 |
| Sirolimus | 4 (6.0%) | 2 (1.8%) | 6 (3.4%) | 0.199 |
Note: ERA I: 1989–2008, ERA II: 2009–2023. Variables are described using frequencies and percentages for categorical variables and mean and standard deviation or medians and interquartile range for continuous variables. Continuous variables were compared by one‐way ANOVA or Kruskal‐Wallis test. Binary and categorical variables were compared by Pearson χ 2 test.
Abbreviations: ABOi, ABO‐incompatible transplantation; ATG, anti‐thymocyte globulin; CHD, congenital heart disease; CMV+, cytomegalovirus positive; DCM, dilated cardiomyopathy; EBV+, Ebstein‐Barr virus positive; ECMO, extracorporeal membrane oxygenation; eGFR, estimated glomerular filtration rate; HCM, hypertrophic cardiomyopathy; MMF, mycophenolate mofetil; PVRi, pulmonary vascular resistance index; RCM, restrictive cardiomyopathy; VAD, ventricular assist device; WU·m2, Woods units.
Major ABOi: A/B/AB → O or AB → A/B or A → B or B → A.
Minor ABOi: O → A/B/AB or A/B → AB.
Out of 5 positive crossmatches, only 2 were CDC positive following DTT. The remainder were negative on CDC‐test, but positive when performing flow cytometry crossmatch testing.
Ratio calculated as [donor weight]/[recipient weight].
One patient received rituximab, bortezomib and plasmapheresis as additional induction treatment at the time of transplantation.
FIGURE 1.
Study flow chart. pHTx, pediatric heart transplantation; re‐pHTx, pediatric heart re‐transplantation. *One child was taken of the waiting list due to clinical deterioration and died shortly after de‐listing (within 30 days) and is thus counted towards waiting list mortality in all analyses.
3.2. Waiting List Course of Events and Mortality
The cumulative incidence of listing for pHTx increased from 3.05 per million in ERA I to 4.40 per million in children < 18 years of age in Sweden (44% increase), and the incidence over the study period was 3.65 per million. A median number of 7 (IQR 4.5–11, range 1–15) children were listed annually during the study period. The median annual number of children listed during ERA I was 5 (IQR 3.8–11, range 1–11), and 9 (IQR 7.5–12, range 3–15) during ERA II. Kaplan–Meier survival estimates describing survival probability within the first 500 days following listing are shown in Figure 2. Waiting list mortality rates for the entire cohort at 30, 100, and 365 days were 9.7% (95% CI 6.4–14.5), 18.6% (95% CI 13.6–25.2) and 36.4% (95% CI 27.1–47.8) respectively (Figure 2A). The 30‐day waiting list mortality rate was 16.0% (95% CI 10.1–24.9) for ERA I and 4.2% (95% CI 1.8–9.8) for ERA II, while the 1‐year waiting list mortality was 56.1% (95% CI 41.2–72.0) and 20.2% (95% CI 10.7–36.4) for the respective eras (Figure 2B). Stepwise adjustment for covariates using Cox proportional hazard modeling displayed a significantly lower risk for waiting list mortality during ERA II compared to ERA I (HR 0.21, Table 3). The full model is available as Table S1.
FIGURE 2.
Survival probability during the first 500 days of listing in Sweden, 1989–2023. ERA I: 1989–2008, ERA II: 2009–2023. Kaplan–Meier estimated survival for the first 500 days on the waiting list for the full cohort (A) and stratified by era of transplantation (B), age group at listing (C) and diagnosis at listing (D). CHD, congenital heart disease; CM, cardiomyopathy.
TABLE 3.
Mortality on waiting list and post‐transplant survival by era of transplantation.
| HR | 95% CI | p | ||
|---|---|---|---|---|
| Wait list mortality | ||||
| ERA II (1989–2008) vs. ERA I (2009–2023) | Era unadjusted | 0.21 | 0.11–0.41 | < 0.001 |
| Era‐adjusted for sex and age group | 0.20 | 0.10–0.38 | < 0.001 | |
| Era adjusted for 4 covariates a | 0.21 | 0.10–0.42 | < 0.001 | |
| Post‐transplant survival | ||||
| ERA II (1989–2008) vs. ERA I (2009–2023) | Era unadjusted | 0.52 | 0.25–1.05 | 0.068 |
| Era‐adjusted for recipient and donor sex and age group | 0.44 | 0.20–0.95 | 0.037 | |
| Era adjusted for 7 covariates b | 0.46 | 0.21–1.04 | 0.063 | |
| Era adjusted for 13 covariates c | 0.46 | 0.15–1.43 | 0.182 | |
Note: Multivariable Cox proportional hazard regression for wait list and post‐transplant survival analyzed by ERA II (2009–2023) versus ERA I (1989–2008) of transplantation.
Abbreviations: ABOi, ABO‐incompatible transplant; CI, confidence interval; ECMO, extracorporeal membrane oxygenation; eGFR, estimated glomerular filtration rate; HR, hazard ratio; VAD, ventricular assist device.
Sex, age, group of diagnosis (CHD vs. CM) and blood group (ABO).
Recipient and donor sex, recipient and donor age groups, group of diagnosis, donor‐recipient weight mismatch ratio, and urgency at transplant. Variables were selected a priori based on clinical experience.
Recipient and donor sex, recipient and donor age groups, group of diagnosis, donor‐recipient weight mismatch ratio and urgency at transplant, eGFR, cold ischemic time, ABOi, CMV mismatch, preoperative VAD and preoperative ECMO. Variables were selected a priori based on clinical experience.
3.3. Transplantation and Donors
The cumulative incidence of pHTx increased from 1.86 per million in ERA I to 3.65 per million in ERA II (96% increase) in children < 18 years of age. The incidence over the entire study period was 2.66 per million. A median of 5 (IQR 3–7, range 0–13) children were transplanted annually: 3.5 (IQR 1.8–5, range 0–9) during ERA I and 7.0 (IQR 6–9, range 4–13) during ERA II. Induction and maintenance immunosuppressive regimens displayed significant variations over time (Table 2). Two concomitant liver transplantations were performed, both during ERA II. No concomitant kidney transplantations were performed. Donation after circulatory death (DCD) was legalized in Sweden in 2020; but so far, no children have undergone pHTx with DCD donors.
3.4. Post‐Transplant Survival
No patient was lost to follow‐up. Median follow‐up time was 8.6 years (IQR 3.2–15.0 years, range 1 day—34.1 years); 16.6 years for ERA I (IQR 9.8–24.1, range 1 day—34.1 years) and 5.6 years for ERA II (IQR 2.2–9.4 years, range 4 days—15.0 years). The 30‐day mortality was 2.2% (n = 4) over the entire study period, with rates of 2.8% (n = 2) during ERA I and 1.8% (n = 2) during ERA II. The 1‐ and 5‐year post‐transplant survival during the study period was 94.5% (95% CI 90.0–97.0) and 87.0% (95% CI 80.7–91.4), respectively. Kaplan–Meier estimates for post‐transplant survival are shown in Figure 3. Post‐transplant 10‐, 20‐, and 30‐year survival for the entire cohort was 79.4% (95% CI 71.6–85.2), 63.0% (95% CI 52.1–72.2) and 57.1% (95% CI 44.3–68.1), respectively (Figure 3A). The 10‐year survival rate was 73.6% (95% CI 61.8–82.3) in ERA I and 85.3% (95% CI 74.3–91.8) in ERA II (Figure 3B). In addition, survival post‐pHTx did not differ between age groups (Figure 3C, log‐rank p = 0.360) and was not associated with the etiology of heart failure (Figure 3D, log‐rank p = 0.816). After stepwise adjustment for covariates using Cox proportional hazard modeling, ERA II did not entail significantly lower risk for post‐transplant mortality compared to ERA I (HR 0.46, Table 3). The full model is available in Table S2. Causes of post‐transplant mortality (n = 44) are summarized in Table S3.
FIGURE 3.
Post‐transplant survival for pediatric heart transplant recipients in Sweden, 1989–2023. ERA I: 1989–2008, ERA II: 2009–2023. Kaplan–Meier estimated post‐transplant survival for the full cohort (A) and stratified by era of transplantation (B) age group at listing (C) and diagnosis at transplant (D). CHD, congenital heart disease; CM, cardiomyopathy.
3.5. Re‐Listing and Re‐Transplantation
Seven patients aged < 18 years (3.8% of all pHTx recipients) were listed for re‐pHTx (two from ERA I and five from ERA II). The median age at re‐listing was 12.9 years (IQR 12.5–13.2, range 0.80–13.4). Six of the re‐listed patients reached re‐pHTx (two from ERA I and four from ERA II), with a median time of 62 days (IQR 36–106, range 1–166) on the waiting list. Re‐pHTx took place after a median of 3.4 years (IQR 0.6–10.0, range 1 day–10.9 years) following primary pHTx. When including re‐transplants in adult age, a total number of 11 patients (5.9% of all pHTx recipients) underwent heart re‐transplantation during the study period. One patient received a third graft in adulthood.
3.6. Specific Populations of Listed and Transplanted Children in Sweden
3.6.1. Univentricular Physiology and Sub‐Types of CMs
The fractions of listed children with UVH and HLHS remained constant over time; however, there was an increase in the percentage of children with UVH and HLHS that reached pHTx over time (Tables 1 and 2). Listing and transplantation of patients with restrictive CM (RCM) increased over time (Tables 1 and 2). Mortality on the waiting list was not higher for UVH (Figure S1A, log rank p = 0.828) compared to non‐UVH CHD. Children with HLHS did not show worse waitlist survival (Figure S1B, log rank p = 0.711) than those with non‐HLHS UVH. Similar stratification revealed no significant differences in post‐transplant survival (Figure S1C, log rank p = 0.939 and Figure S1D, log rank p = 0.812). Waiting list mortality and post‐transplant survival did not differ significantly depending on CM phenotype (Figure S1E, log‐rank p = 0.472 and Figure S1F, log rank p = 0.843).
3.6.2. Short‐ and Long‐Term MCS in pHTx Recipients
Transplantations from VAD and ECMO were more frequent in the contemporary era (Table 2). Pre‐pHTx VAD did not affect post‐transplant survival (Figure 4A, log‐rank p = 0.733). Children with VAD after pHTx (Figure 4B, log‐rank p < 0.001) as well as pHTx from ECMO (Figure 4C, log‐rank test p = 0.022) and ECMO after pHTx (Figure 4D, log‐rank test p = 0.010) showed significantly worse survival.
FIGURE 4.
MCS and PVRi as determinants of post‐transplant outcomes. Kaplan–Meier estimated post‐transplant survival for pHTx recipients with pre‐transplant (A) and post‐transplant (B) VAD, as well as pre‐transplant (C) and post‐transplant (D) ECMO. Data on pre‐ and post‐transplant VAD was available for 182 and 179 patients, respectively, and data on pre‐ and post‐transplant ECMO was available for 180 and 176 patients, respectively. PVRi at baseline (n = 99) and post‐AVT (n = 30) are shown in (E), and paired measurements (n = 29) displayed in (F). Paired measurements revealed that only two patients were transplanted with PVRi over 6, post‐AVT. Kaplan–Meier estimated post‐transplant survival for pHTx recipients with baseline PVRi < 6 WU·m2 or ≥ 6 WU·m2 (G), according to data availability outlined above. Linear regression analysis for the correlation between baseline or post‐AVT PVRi and years of post‐transplant survival (H). AVT, acute vasoreactivity test; ECMO, extra‐corporeal mechanical oxygenation; PVRi, pulmonary vascular resistance index; VAD, ventricular assist device; WU·m2, Woods units.
3.6.3. Pulmonary Hypertension With Elevated Pre‐Transplant PVRi
In 99 pHTx recipients with available baseline pre‐transplant PVRi data, median PVRi was 3.3 indexed Woods Units (WU·m2) (Table 2, Figure 4E). Post‐AVT PVRi measurements were available in 30 patients with a median of 3.4 WU·m2 (Table 2, Figure 4E). Paired pre‐ and post‐AVT measurements were available in 29 patients, revealing post‐AVT PVRi < 6 WU·m2 in all but two patients (Figure 4F). When stratifying post‐transplant survival by baseline PVRi < 6 WU·m2 or ≥ 6 WU·m2 there was no statistically significant difference in survival (Figure 4G, log‐rank test p = 0.107). There was no significant correlation between pre‐transplant PVRi at baseline (r2 = 0.006, p = 0.419) or post‐AVT (r 2 = 0.033, p = 0.334) and post‐pHTx survival time (Figure 4H).
3.6.4. ABO‐Incompatible pHTx and Transplantation Against Positive Crossmatch
The percentage of major ABOi increased over time, from 2.9% in ERA I to 13.3% of all pHTx in ERA II (Table 2, p = 0.060). Twelve of the 17 ABOi recipients (70.6%) were younger than 24 months of age at the time of transplant, with the oldest recipient being 5 years and 2 months at ABOi transplant. Survival after pHTx with major ABOi was not significantly worse when compared to ABO‐compatible (ABOc) transplantation (Figure 5A, log rank p = 0.049). Five patients (2.9%) underwent transplantation against a positive crossmatch, deemed positive either on CDC following treatment with DTT or on flow cytometry crossmatch test, all in ERA II (Table 2, p = 0.089).
FIGURE 5.
ABO‐incompatible transplant, immunological risk status and DSA. Kaplan–Meier estimated post‐transplant survival for pHTx recipients undergoing major ABOi transplantation or ABOc/minor ABOi transplantation, 1989–2023 (A). Data on blood group compatibility was lacking for two patients, why n = 183 in A. Kaplan–Meier estimated post‐transplant survival for pHTx recipients by immunological risk status, 2009–2023 (B). Frequencies of the different types of DSA (C) and whether they were preformed or discovered de novo (D) are shown. ABOc, ABO‐compatible transplantation; ABOi, ABO‐incompatible transplantation; DSA, donor‐specific antibodies.
3.6.5. Immunological Risk and HLA‐DSA in Graft Recipients During the Contemporary Era, 2009–2023
Immunology data at pHTx for recipients between 2009 and 2023, including selected baseline characteristics for recipients and donors, stratified by immunological risk, are shown in Table 4.
TABLE 4.
Immunological risk, immunological data, and characteristics, 2009–2023.
| Normal immunological risk (n = 62) | Elevated immunological risk a (n = 42) | High immunological risk b (n = 9) | p | |
|---|---|---|---|---|
| Recipient | ||||
| Age (years) | 10.2 (IQR 2.6–14.4, range 25 days‐18.7 years) | 11.5 (IQR 7.2–15.0, range 1.1–17.7) | 12.5 (IQR 9.4–14.9, range 4.2–16.6) | 0.291 |
| < 1 year | 8 (12.9%) | 0 (0.0%) | 0 (0.0%) | 0.029 |
| Gender (female) | 28 (45.2%) | 19 (45.2%) | 3 (33.3%) | 0.790 |
| Diagnosis | ||||
| CHD | 23 (37.1%) | 12 (28.6%) | 5 (55.6%) | 0.282 |
| CM | 39 (62.9%) | 30 (71.4%) | 4 (44.4%) | |
| Preoperative VAD | 20 (32.3%) | 13 (31.0%) | 4 (44.4%) | 0.731 |
| Preoperative ECMO | 6 (9.7%) | 2 (4.8%) | 1 (11.1%) | 0.620 |
| Donor | ||||
| Age | 12.5 (IQR 4–25, range 0–59) | 28 (IQR 11–47, range 0–53) | 18 (IQR 13–22, range 5–58) | 0.007 |
| Gender | 32 (52.5%), n = 61 | 24 (57.1%) | 5 (55.6%) | 0.894 |
| Wait list duration | 83 (23–178) | 64.5 (9–187) | 72 (50–115) | 0.752 |
| Cold ischemic time | 198 + −69 | 200 + −62 | 205 + −56 | 0.906 |
| ABOi | 10 (16.1%) | 5 (11.9%) | 0 (0.0%) | 0.443 |
| Immunology | ||||
| vPRA | ||||
| HLA‐ab with MFI > 1000 (%) | N/A | 28.4 (10.4–53.1), n = 38 | 95.6 (70.3–98.5), n = 9 | < 0.001 |
| HLA‐ab with MFI > 5000 (%) | N/A | N/A | 52.8 (28.4–70.2), n = 9 | N/A |
| Cumulative MFI | ||||
| HLA with MFI > 1000 | N/A | 4600 (2100–7200) | 62 000 (17000–180 000), n = 8 | < 0.001 |
| HLA‐A, ‐B and ‐DR mismatches | ||||
| 0 | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0.631 |
| 1 | 1 (1.6%) | 1 (2.4%) | 0 (0.0%) | |
| 2 | 3 (4.8%) | 3 (7.1%) | 1 (11.1%) | |
| 3 | 11 (17.7%) | 4 (9.5%) | 2 (22.2%) | |
| 4 | 17 (27.4%) | 6 (14.3%) | 1 (11.1%) | |
| 5 | 15 (24.2%) | 14 (33.3%) | 4 (44.4%) | |
| 6 | 15 (24.2%) | 14 (33.3%) | 1 (11.1%) | |
| HLA‐DQ mismatches | n = 51 | n = 41 | n = 9 | |
| 0 | 7 (13.7%) | 8 (19.5%) | 0 (0.0%) | 0.647 |
| 1 | 22 (43.1%) | 16 (39.0%) | 5 (55.6%) | |
| 2 | 22 (43.1%) | 17 (41.5%) | 4 (44.4%) | |
| Clinically significant DSA c | ||||
| De novo and/or preformed | 4 (6.5%) | 12 (28.6%) | 9 (100.0%) | < 0.001 |
Note: Variables are described using frequencies and percentages for categorical variables and median with interquartile range for continuous variables. Continuous variables were compared by Kruskal‐Wallis test. Binary and categorical variables were compared by Pearson χ 2 test.
Abbreviations: ABOi, ABO‐incompatible; CHD, congenital heart disease; CM, cardiomyopathy; DSA, donor specific antibody; ECMO, extra‐corporeal membrane oxygenation; HLA, human leukocyte antigen; MFI, mean fluorescence intensity; PRA, panel reactive antibody; VAD, ventricular assist device.
If a patient displayed any HLA‐antibodies with an MFI > 1000, they were considered immunized.
If a patient displayed any HLA‐antibodies with an MFI > 5000, they were considered highly immunized.
For HLA‐A, HLA‐B, HLA‐DR and HLA‐DQ DSA, both de novo and preformed, antibodies were considered clinically significant if MFI > 1000. For HLA‐Cw, HLA‐DP and HLA‐DR3/4/5, both de novo and preformed, antibodies were considered clinically significant if MFI > 2000.
3.6.5.1. Pre‐Transplant Immunological Risk and Degree of Matching
In children with elevated immunological risk, the median vPRA value for preformed HLA with MFI > 1000 was 28.4%. The median cumulative MFI was 4700 for these patients. In recipients with high immunological risk, the median vPRA for preformed HLA with MFI > 1000 was 95.6%. The median cumulative MFI was 62 000 in this group (Table 4, both with p < 0.001). To investigate the impact of singular strong HLA‐antibodies on recipient pre‐pHTx immunization status, an additional vPRA value was calculated in patients at high immunological risk that only included preformed HLA with MFI > 5000. The median vPRA for these HLA was 52.8%, thus comprising > 50% of the total vPRA for all significant HLA (Table 4). In children at elevated and/or high immunological risk, DSA were present or developed to a greater extent than in non‐immunized children (Table 4, p < 0.001). Post‐transplant survival was not statistically different for recipients with high immunological risk (Figure 5B, log rank p = 0.101).
3.6.5.2. Clinically Significant DSA; HLA Specificity and Post‐Transplant Development
Clinically significant DSAs, pre‐formed and de novo, in pHTx recipients from 2009 to 2023 are listed in Table S4. DSAs were present, or developed, in 22.1% (n = 25) of all recipients during the contemporary era. The frequency of DSAs specific for different HLAs, pre‐formed and de novo, is presented in Figure 5C,D. HLA‐DQ DSAs, constituting 45.9% of all clinically significant dnDSAs detected in recipients (Figure 5C), were observed in 86.7% of all patients who developed dnDSAs. HLA‐DQ DSAs could be detected within the first month post‐pHTx (Table S4). MFI values for HLA‐DQ DSAs were plotted over time for all individual patients with lab screen analyses at > 4 separate time points (n = 8) and correlated with key clinical events and alterations in immunosuppression (Figure 6A–H).
FIGURE 6.
Management and monitoring of post‐transplant DQ‐type antibodies. MFI values of DQ‐type DSA were plotted over time, to visualize DSA development over time. Patient medical records were investigated to highlight what clinical measures were taken to combat and/or monitor a rise in MFI. A short description of the medical background is provided in the yellow box belonging to each graph. MFI > 25 000 is generally not reported as the analysis is saturated around this value, although specified here to provide an accurate description of the investigated values. ACR, acute cellular rejection; ATG, anti‐thymocyte globulin; CHD, congenital heart disease; CM, cardiomyopathy; DSA, donor‐specific antibodies; ECMO, extra‐corporeal membrane oxygenation; HTAD, hereditary thoracic aorta disease; IVIG, intravenous immunoglobulin; MFI, mean fluorescence intensity; MMF, mycophenolate mofetil; pAMR, pathological grade of antibody‐mediated rejection; pHTx, pediatric heart transplantation.
4. Discussion
The current study complements previous data reported by Gilljam et al., describing patient characteristics and survival until 2009 among the first 135 children listed for pHTx in Sweden [1]. Additionally, actual 1‐year and 10‐year survival of 93% and 72%, respectively, was reported for children from the Gothenburg cohort between 1990 and 2014 [22]. The present study investigates variables that have not been previously reported for the Swedish pHTx cohort, including data on MCS, elevated PVRi, ABO‐incompatible transplantation, immunological risk, and DSA‐status.
Since 2009, Sweden has experienced a significant decrease in waiting list mortality; although our data implicates that high infant waiting list mortality remains a challenge in Sweden, as it does globally [23]. Overall waiting list mortality 1989–2023 (18.9%) is comparable to the latest report from the UNOS database denoting a 16% waiting list mortality rate during the period 1999–2023 in the United States [24]. More importantly, waiting list mortality in Sweden had decreased to 8.8% between 2009 and 2023.
Increased waiting list duration was observed in ERA II, most likely due to improved waiting list survival. The increased utilization of durable long‐term MCS support, especially in small children and infants, may be an important contributor to both the increased waiting list duration and waiting list survival. Recently implemented Scandiatransplant waiting list regulations state that all hospital‐bound patients, with a body weight < 25 kg and supported by long‐term VAD, are to be considered candidates for urgent call listing after 3 months [25]. The change in policy will hopefully, without altering the number of available organs, level out differences in offers between patients on VAD and patients on inotropes in the intensive care unit (currently considered urgent at waiting list entry). Collaborative efforts within Scandiatransplant aim to optimize organ allocation [2]. In recent years, Scandiatransplant has had a net export of organs in relation to other European organ allocation organizations [26], highlighting the efficient utility of available donors in the region. Utilization of DCD donors in carefully selected recipients within Scandiatransplant could contribute to a greater donor pool, although the true long‐term outcomes in pediatric recipients of DCD organs are yet unknown [27].
An increase in listing, but more prominently in pHTx, incidence was seen in ERA II. The combination of an increased listing incidence among children, a longer time on the waiting list, and lower waiting list mortality suggests that children listed during ERA II likely were in a better overall clinical condition at waiting list entry than patients in ERA I, and that those in more critical heart failure benefited from the increased availability of VAD and improved medical and surgical/interventional treatment during the contemporary era.
Transplantation during the contemporary era was not a significant determinant of post‐transplant outcome in our analysis, most likely attributable to the low early mortality observed in Swedish pHTx practice. Improvements in early mortality in transplant practice have been postulated to be greater over time than improvements in late mortality [28], which may explain why the improvement in survival during ERA II in Sweden did not reach significance. Short‐ and long‐term post‐transplant outcomes in the Swedish cohort are comparable to recent reports from corresponding nationwide cohorts, both regionally and internationally [29, 30, 31]. The contribution of free accessible health care, the availability of high‐quality national heart failure follow‐up programs, as well as a high public awareness of available treatments all likely contribute to the good outcomes presented here. We observed similar outcomes for children with CHD and with CMs post‐transplant, contrary to recently published ISHLT reports [32]. This finding is also in contrast with recently published data from Scandiatransplant [33] where CHD patients had worse survival after pediatric heart transplant compared to non‐CHD patients. The challenge of pHTx in CHD patients is mainly comprised of two issues: surgical challenges in the reoperation setting and significant immunization due to prior surgery, transfusion, and circulatory support. Small surgical teams that maintain adequate volume and training, as well as close‐knit collaboration with immunologists at both Swedish pHTx centers may provide explanations for these outcomes. However, when comparing the Swedish cohort with largely North American registries like the ISHLT, Pediatric Heart Transplantation Society (PHTS) [34] and UNOS [29], the proportion of CHD and infants is lower in our cohort. Therefore, our results on CHD transplant may also be due to differences between our cohort and those of the ISHLT and PHTS.
The most common causes of post‐transplant death; graft failure, cardiovascular events, and malignancy correspond to the most common causes of death in the ISHLT registry [35]; although we did not investigate changes over time in our cohort due to the low number of events.
Our data supports previous reports suggesting that post‐transplant survival is independent of the use of pre‐transplant VAD in children [6, 11, 36, 37]. In Kaplan–Meier analysis, impeded survival was observed in children with post‐transplant VAD as well as pre‐ and post‐transplant ECMO in the Swedish cohort, like previously published data from the PHTS [38]. Pre‐transplant ECMO was, however, not deemed a risk factor for worse post‐transplant survival in our stepwise Cox proportional hazard regression analysis; although we acknowledge that this is most likely a type II statistical error. Post‐transplant ECMO and VAD could be surrogate markers for poor recipient status, donor and/or graft status; however, even in patients on post‐transplant MCS with good prognosis, associated events such as pump thrombosis, stroke, and infection could cause clinical deterioration [9, 39].
Pre‐ and post‐transplant kidney function has a significant impact on pHTx outcomes [40], but our analysis could not highlight elevated pretransplant GFR as a determinant of post‐pHTx outcome. A decrease in kidney function post‐transplant can be the result of chronic exposure to calcineurin inhibitors (tacrolimus). In recent years, the addition of Everolimus in a CNI‐sparing effort has been associated with improved kidney function in pediatric patients [41, 42]. We observed a trend of increased Everolimus use over time; however, we had no follow‐up data to investigate its possible impact on kidney function.
The dogma of elevated PVRi > 6 WU·m2 as an absolute contraindication for pHTx has been challenged in recent years [43], and there is a lack of international consensus regarding important management aspects in patients with elevated PVRi [12, 15]. Long‐term outcomes were somewhat impeded for patients with PVRi ≥ 6, although not statistically significant, as compared to those with PVRi < 6. Hemodynamics following AVT was available for 30 patients, with 29 having paired measurements pre‐ and post‐AVT. Only two patients were transplanted with a post‐AVT PVRi of > 6, indicating that PVRi > 6 was reversible with pulmonary vasodilators upon AVT in most children. Post‐AVT PVRi > 6 seems to have been a, possibly unspoken, cut‐off for pHTx eligibility in Sweden. As recently highlighted in a Scientific Statement from the American Heart Association [12], PVRi should be interpreted with caution, as many factors (sedation, shunts, and poor cardiac output) can influence the calculation of PVRi in children. Nonetheless, this data supports that children with baseline PVRi of > 6 should undergo AVT and not be automatically precluded from pHTx, and that most children can reach below this cut‐off with adequate treatment. The possibility of implantable hemodynamic monitors in children may increase the accuracy of pulmonary hypertension monitoring in the setting of heart failure and thus optimize pre‐pHTx management and timing of pHTx [44, 45]. One limitation of our study was the lack of baseline hemodynamic data for all listed patients, why our analysis could have been subject to selection bias.
ABOi transplantation and transplant against positive crossmatch was performed to a greater extent during ERA II, as new methods became available during this era. Our data support that ABOi recipients do not have worse short‐ and medium‐term outcomes after pHTx, in line with recent reports from the United States [46]. Implementation of perioperative immunoadsorption column protocols in Sweden and globally [47] has allowed for efficient management of anti‐A and anti‐B antibody titers in recipients.
The clinical utility of repeated post‐transplant measurements of anti‐HLA DSA in solid organ transplant recipients was highlighted as an area of interest for future study by the Sensitization in transplantation: assessment of risk (STAR) working group [48] and has been suggested to be an important tool for risk stratification of heart transplant recipients as well as a prognostic factor predicting CAV [49]. Descriptions of pediatric cohorts are sparse [48]. We provide a detailed description of immunological data in contemporary pHTx recipients, 2009–2023, and opted to define immunological risk status based on the presence of prominent anti‐HLA antibodies (MFI), rather than the width of the sensitization (vPRA). Additionally, we calculated two separate vPRA values to distinguish the contribution of strong individual anti‐HLA to the vPRA value. Children at high immunological risk, with any anti‐HLA antibody with MFI > 5000, displayed a high vPRA (median 96.3%) for all HLA‐antibodies with MFI > 1000. In the same patients, vPRA for anti‐HLA with MFI > 5000 was a median 50.1%, suggesting that a large part of the vPRA might be attributable to a limited number of strong anti‐HLA. Only a small subset of patients proved at high immunological risk in our cohort but demonstrated a trend towards worse survival in Kaplan–Meier analysis.
5. Conclusions
Early and long‐term outcomes of heart listing and pHTx in Sweden have improved continuously over the last 35 years. Data presented here support the notion that elevated pre‐transplant baseline PVRi > 6 WU·m2 is not an absolute contraindication to pHTx, and that AVT is an important tool for additional evaluation of PVRi. Reported outcomes in the Swedish cohort are comparable with outcomes reported by larger databases worldwide.
Author Contributions
All authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data. O.v.d.H., K.T.‐L., and M.O. drafted the manuscript. All authors edited, revised, and approved the final version of the manuscript.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Data S1: Supporting Information.
Acknowledgments
We acknowledge Scandiatransplant for provision of its registry data. We are grateful towards Dr. Joanna‐Maria Papageorgiou for the contribution of patient data. Flow charts were created with lucid.com.
van der Have O., Wåhlander H., Hofbard T., et al., “Comprehensive Analysis of the First 35 Years of Pediatric Heart Transplantation in Sweden,” Pediatric Transplantation 29, no. 6 (2025): e70154, 10.1111/petr.70154.
Funding: This work was supported by the Swedish Heart‐Lung Foundation (20230545 [K.T.‐L.] and 20220591 [K.T.‐L.]), Skane County Council (2022‐Projekt0168 [K.T.‐L.]), Knut and Alice Wallenberg Foundation [K.T.‐L.], the Swedish Research Council (2022‐00683 [K.T.‐L] and 2023‐03184 [J.N.]), by grants ALFGBG‐1006863 from the Swedish state under the agreement between the Swedish government and the county councils (J.H.) and the Fund for Transplant Research at the Centre for Transplantation, Skane University Hospital in Lund, Sweden (O.v.d.H.).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Gilljam T., Higgins T., Bennhagen R., and Wahlander H., “First Two Decades of Paediatric Heart Transplantation in Sweden—Outcome of Listing and Post‐Transplant Results,” Acta Paediatrica 100, no. 11 (2011): 1442–1447. [DOI] [PubMed] [Google Scholar]
- 2. Madsen M., Asmundsson P., Brekke I. B., et al., “Scandiatransplant: Thirty Years of Cooperation in Organ Transplantation in the Nordic Countries,” Clinical Transplants 121 (1998): 31. [PubMed] [Google Scholar]
- 3. Sutcliffe D. L., Pruitt E., Cantor R. S., et al., “Post‐Transplant Outcomes in Pediatric Ventricular Assist Device Patients: A PediMACS‐Pediatric Heart Transplant Study Linkage Analysis,” Journal of Heart and Lung Transplantation 37, no. 6 (2018): 715–722. [DOI] [PubMed] [Google Scholar]
- 4. Dipchand A. I. and Webber S. A., “Pediatric Heart Transplantation: Looking Forward After Five Decades of Learning,” Pediatric Transplantation 28, no. 1 (2024): e14675. [DOI] [PubMed] [Google Scholar]
- 5. Rohde S., van Puyvelde J., Veen K. M., et al., “The European Registry for Patients With Mechanical Circulatory Support (EUROMACS): Fourth Paediatric EUROMACS (Paedi‐EUROMACS) Report,” European Journal of Cardio‐Thoracic Surgery 66 (2024): ezae276. [DOI] [PubMed] [Google Scholar]
- 6. Kindel S. J. and Everitt M. D., “A Contemporary Review of Paediatric Heart Transplantation and Mechanical Circulatory Support,” Cardiology in the Young 26, no. 5 (2016): 851–859. [DOI] [PubMed] [Google Scholar]
- 7. Morgan C. T., Manlhiot C., McCrindle B. W., and Dipchand A. I., “Outcome, Incidence and Risk Factors for Stroke After Pediatric Heart Transplantation: An Analysis of the International Society for Heart and Lung Transplantation Registry,” Journal of Heart and Lung Transplantation 35, no. 5 (2016): 597–602. [DOI] [PubMed] [Google Scholar]
- 8. Power A., Navaratnam M., Murray J. M., et al., “Adverse Events Associated With Cardiac Catheterization in Children Supported With Ventricular Assist Devices,” ASAIO Journal 68, no. 9 (2022): 1174–1181. [DOI] [PubMed] [Google Scholar]
- 9. George A. N., Hsia T. Y., Schievano S., and Bozkurt S., “Complications in Children With Ventricular Assist Devices: Systematic Review and Meta‐Analyses,” Heart Failure Reviews 27, no. 3 (2022): 903–913. [DOI] [PubMed] [Google Scholar]
- 10. Niebler R. A., Amdani S., Blume B., et al., “Stroke in Pediatric Ventricular Assist Device Patients—A Pedimacs Registry Analysis,” Journal of Heart and Lung Transplantation 40, no. 7 (2021): 662–670. [DOI] [PubMed] [Google Scholar]
- 11. Marcos‐Alonso S., Gil N., García‐Guereta L., et al., “Impact of Mechanical Circulatory Support on Survival in Pediatric Heart Transplantation,” Pediatric Transplantation 24, no. 4 (2020): e13707. [DOI] [PubMed] [Google Scholar]
- 12. Hopper R. K., Hansmann G., Hollander S. A., et al., “Clinical Management and Transplant Considerations in Pediatric Pulmonary Hypertension due to Left Heart Disease: A Scientific Statement From the American Heart Association,” Circulation. Heart Failure 29 (2024): e000086. [DOI] [PubMed] [Google Scholar]
- 13. Brittain E. L., Thenappan T., Huston J. H., et al., “Elucidating the Clinical Implications and Pathophysiology of Pulmonary Hypertension in Heart Failure With Preserved Ejection Fraction: A Call to Action: A Science Advisory From the American Heart Association,” Circulation 146, no. 7 (2022): e73–e88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kulik T. J., “Pulmonary Hypertension Caused by Pulmonary Venous Hypertension,” Pulmonary Circulation 4, no. 4 (2014): 581–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Hopper R. K., van der Have O., Hollander S. A., et al., “International Practice Heterogeneity in Pre‐Transplant Management of Pulmonary Hypertension Related to Pediatric Left Heart Disease,” Pediatric Transplantation 27, no. 2 (2023): e14461. [DOI] [PubMed] [Google Scholar]
- 16. Barten M. J., Schulz U., Beiras‐Fernandez A., et al., “The Clinical Impact of Donor‐Specific Antibodies in Heart Transplantation,” Transplantation Reviews (Orlando, Fla.) 32, no. 4 (2018): 207–217. [DOI] [PubMed] [Google Scholar]
- 17. Dipchand A. I., Webber S., Mason K., et al., “Incidence, Characterization, and Impact of Newly Detected Donor‐Specific Anti‐HLA Antibody in the First Year After Pediatric Heart Transplantation: A Report From the CTOTC‐04 Study,” American Journal of Transplantation 18, no. 9 (2018): 2163–2174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. von Elm E., Altman D. G., Egger M., et al., “The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies,” Annals of Internal Medicine 147, no. 8 (2007): 573–577. [DOI] [PubMed] [Google Scholar]
- 19. Benchimol E. I., Smeeth L., Guttmann A., et al., “The REporting of Studies Conducted Using Observational Routinely‐Collected Health Data (RECORD) Statement,” PLoS Medicine 12, no. 10 (2015): e1001885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Tambur A. R., Campbell P., Claas F. H., et al., “Sensitization in Transplantation: Assessment of Risk (STAR) 2017 Working Group Meeting Report,” American Journal of Transplantation 18, no. 7 (2018): 1604–1614. [DOI] [PubMed] [Google Scholar]
- 21. Peacock S., Briggs D., Barnardo M., et al., “BSHI/BTS Guidance on Crossmatching Before Deceased Donor Kidney Transplantation,” International Journal of Immunogenetics 49, no. 1 (2022): 22–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Dellgren G., Westerlind A., Liden H., et al., “Continuous Improvement in Outcome After Heart Transplantation ‐ Long‐Term Follow‐Up After Three Decades of Experience,” International Journal of Cardiology 231 (2017): 188–194. [DOI] [PubMed] [Google Scholar]
- 23. Denfield S. W., Azeka E., Das B., et al., “Pediatric Cardiac Waitlist Mortality‐Still Too High,” Pediatric Transplantation 24, no. 3 (2020): e13671. [DOI] [PubMed] [Google Scholar]
- 24. Power A., Sweat K. R., Roth A., et al., “Contemporary Pediatric Heart Transplant Waitlist Mortality,” Journal of the American College of Cardiology 84, no. 7 (2024): 620–632. [DOI] [PubMed] [Google Scholar]
- 25. Guidelines for Thoracic Organ Exchange in the Scandiatransplant Area (April 2023) , “Guidelines for Organ Exchange in the Scandiatransplant Area,” (2023).
- 26. Scandiatransplant , “Scandiatransplant Annual Data Report 2024,” (2024), https://www.scandiatransplant.org/resources/annual‐report.
- 27. Kleinmahon J. A., Patel S. S., Auerbach S. R., Rossano J., and Everitt M. D., “Hearts Transplanted After Circulatory Death in Children: Analysis of the International Society for Heart and Lung Transplantation Registry,” Pediatric Transplantation 21, no. 8 (2017): 13064. [DOI] [PubMed] [Google Scholar]
- 28. Dharnidharka V. R., Lamb K. E., Zheng J., Schechtman K. B., and Meier‐Kriesche H. U., “Lack of Significant Improvements in Long‐Term Allograft Survival in Pediatric Solid Organ Transplantation: A US National Registry Analysis,” Pediatric Transplantation 19, no. 5 (2015): 477–483. [DOI] [PubMed] [Google Scholar]
- 29. Sharaf O. M., Bilgili A., Brennan Z., et al., “Pediatric Heart Transplantation Over 36 Years and Contemporary Volume‐Outcome Analysis of UNOS,” Annals of Thoracic Surgery 118, no. 6 (2024): 1288–1298. [DOI] [PubMed] [Google Scholar]
- 30. Salonen R., Jahnukainen T., Nikkilä A., and Endén K., “Long‐Term Mortality in Pediatric Solid Organ Recipients‐A Nationwide Study,” Pediatric Transplantation 27, no. 2 (2023): e14463. [DOI] [PubMed] [Google Scholar]
- 31. Rosenthal L. M., Krauss A., Miera O., et al., “Changes in Waiting Time, Need for Mechanical Circulatory Support and Outcomes in Paediatric Heart Transplant Recipients,” ESC Heart Fail 11, no. 6 (2024): 3626–3635. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Singh T. P., Hsich E., Cherikh W. S., et al., “The International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation: 2025 Annual Report of Heart and Lung Transplantation,” Journal of Heart and Lung Transplantation 25 (2025): S1053–2498. [DOI] [PubMed] [Google Scholar]
- 33. van der Have O., Tran‐Lundmark K., Wåhlander H., et al., “Pediatric Heart Transplantation Within the Scandiatransplant Region ‐ A Multinational Observational Study Spanning 38 Years,” Journal of Heart and Lung Transplantation (2025). [DOI] [PubMed] [Google Scholar]
- 34. Richmond M. E., Conway J., Kirklin J. K., et al., “Three Decades of Collaboration Through the Pediatric Heart Transplant Society Registry: A Journey Through Registry Data With a Highlight on Children With Single Ventricle Anatomy,” Pediatric Transplantation 28 (2024): e14615. [DOI] [PubMed] [Google Scholar]
- 35. Dipchand A. I., Kirk R., Edwards L. B., et al., “The Registry of the International Society for Heart and Lung Transplantation: Sixteenth Official Pediatric Heart Transplantation Report—2013; Focus Theme: Age,” Journal of Heart and Lung Transplantation 32, no. 10 (2013): 979–988. [DOI] [PubMed] [Google Scholar]
- 36. Blume E. D., Naftel D. C., Bastardi H. J., Duncan B. W., Kirklin J. K., and Webber S. A., “Outcomes of Children Bridged to Heart Transplantation With Ventricular Assist Devices ‐ A Multi‐Institutional Study,” Circulation 113, no. 19 (2006): 2313–2319. [DOI] [PubMed] [Google Scholar]
- 37. Davies R. R., Haldeman S., McCulloch M. A., and Pizarro C., “Ventricular Assist Devices as a Bridge‐To‐Transplant Improve Early Post‐Transplant Outcomes in Children,” Journal of Heart and Lung Transplantation 33, no. 7 (2014): 704–712. [DOI] [PubMed] [Google Scholar]
- 38. Dipchand A. I., Mahle W. T., Tresler M., et al., “Extracorporeal Membrane Oxygenation as a Bridge to Pediatric Heart Transplantation: Effect on Post‐Listing and Post‐Transplantation Outcomes,” Circulation. Heart Failure 8 (2015): 960–969. [DOI] [PubMed] [Google Scholar]
- 39. Simmonds J., Zangwill S. D., Wisotzkey B., et al., “Mechanical Circulatory Support Early After Pediatric Heart Transplantation‐An Analysis From the Pediatric Heart Transplant Society,” Journal of Heart and Lung Transplantation 44, no. 2 (2025): 227–233. [DOI] [PubMed] [Google Scholar]
- 40. Alsoufi B., Kozik D., Lambert A. N., et al., “Associated Factors and Impact of Persistent Renal Dysfunction in Pediatric Heart Transplantation,” Annals of Thoracic Surgery 117, no. 1 (2024): 136–142. [DOI] [PubMed] [Google Scholar]
- 41. Grimm K., Lehner A., Fernandez Rodriguez S., et al., “Conversion to Everolimus in Pediatric Heart Transplant Recipients Is a Safe Treatment Option With an Impact on Cardiac Allograft Vasculopathy and Renal Function,” Clinical Transplantation 35, no. 3 (2021): e14191. [DOI] [PubMed] [Google Scholar]
- 42. Rosenthal L. M., Nordmeyer J., Kramer P., et al., “Long‐Term Experience Using CNI‐Free Immunosuppression in Selected Paediatric Heart Transplant Recipients,” Pediatric Transplantation 25, no. 8 (2021): e14111. [DOI] [PubMed] [Google Scholar]
- 43. Richmond M. E., Law Y. M., Das B. B., et al., “Elevated Pre‐Transplant Pulmonary Vascular Resistance Is Not Associated With Mortality in Children Without Congenital Heart Disease: A Multicenter Study,” Journal of Heart and Lung Transplantation 34, no. 3 (2015): 448–456. [DOI] [PubMed] [Google Scholar]
- 44. Piccinelli E., Grutter G., Pilati M., et al., “Use of the CardioMEMS Device in Children and Patients With Congenital Heart Disease: A Literature Review,” Journal of Clinical Medicine 13, no. 14 (2024): 4234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Long Z., Yokota R., Peng L., Kaufman B., Chen S., and Kwiatkowski D., “PEDS11: Use of the CardioMEMSTM HF System in Pediatric Patients for Heart Failure Management,” ASAIO Journal 69, no. Supplement 2 (2023): 83. [Google Scholar]
- 46. Amdani S., Deshpande S. R., Liu W., and Urschel S., “Impact of the Pediatric ABO Policy Change on Listings, Transplants, and Outcomes for Children Younger Than 2 Years Listed for Heart Transplantation in the United States,” Journal of Cardiac Failure 30, no. 3 (2024): 476–485. [DOI] [PubMed] [Google Scholar]
- 47. Bansal N., West L. J., Simmonds J., and Urschel S., “ABO‐Incompatible Heart Transplantation‐Evolution of a Revolution,” Journal of Heart and Lung Transplantation 43, no. 9 (2024): 1514–1520. [DOI] [PubMed] [Google Scholar]
- 48. Lefaucheur C., Louis K., Morris A. B., et al., “Clinical Recommendations for Posttransplant Assessment of Anti‐HLA (Human Leukocyte Antigen) Donor‐Specific Antibodies: A Sensitization in Transplantation: Assessment of Risk Consensus Document,” American Journal of Transplantation 23, no. 1 (2023): 115–132. [DOI] [PubMed] [Google Scholar]
- 49. Loupy A., Coutance G., Bonnet G., et al., “Identification and Characterization of Trajectories of Cardiac Allograft Vasculopathy After Heart Transplantation: A Population‐Based Study,” Circulation 141 (2020): 1954–1967. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: Supporting Information.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.