Case report
Coronary artery fistulas have been described in patients after heart transplantation more often than in the normal population. Reference Allen, Goldstein and Pahl1 Our aim was to review the incidence, anatomy, and evolution of coronary fistulas in our paediatric heart transplant patients.
Our centre’s database of paediatric heart transplant patients was reviewed retrospectively. All the patients transplanted between 2008 and 2020 with at least one coronarography during their follow-up were included and every coronary angiography was reviewed focusing on the presence of coronary artery fistulas, number, anatomic distribution, ending structure, size, and flow amount. Categorical data are reported as frequencies and percentages and have been analysed using the Fisher’s exact test. The continuous variables are reported as median and interquartile range.
During this period, a total of 50 patients were transplanted and 30 of them had at least 1coronary angiography. The median age at the transplant was 8.2 years (SIQR 5.3 years). The reason for the transplant was 73.3% cardiomyopathy, 23.3% congenital, and 3.3% retransplantation. All of them had at least one coronary angiography in the first 3years after heart transplantation, with median time until the first procedure of 1 year (SIQR 0.6) and median of two coronary angiographies during follow-up. Median follow-up of 5.3 years (SIQR 3.2). Table 1 describes the characteristics of the patients with coronary artery fistulas and Table 2 summarises the characteristics of the fistulas. A total of 13 (43.3%) patients had at least 1 coronary artery fistula on the first coronarography: 8 (61.5%) presented just one fistula, 2 (16.6%) two fistulas, 1 (12%) three fistulas, and 1 more than three. According to the importance of their flow, fistulas were classified as small in 7 patients (58.3%), medium in 4 (30.7%), and large in 2 (16.6%), with one of the large ones causing dilatation of the coronary arteries as a collateral effect. Their origin was the anterior descending coronary artery in 9, right coronary artery in 5, and the circumflex artery in 3, and 78.9% of them were terminated at the ipsilateral lung field. Two patients presented a second fistula draining to the pulmonary artery. Two fistulas to systemic thoracic arteries were identified. None of them presented any symptoms of cardiac dysfunction, nor electrocardiogram or echocardiography ischaemic signs. A second coronarography was performed in 7 patients, a median of 22 months after the first one. In 2 patients the amount of fistulas flow significantly decreased, and in 1 case, even disappeared.
Table 1. Characteristics of the patients.
ATG = Thymoglobulin; BAS = Basiliximab; Circ=Circumflex; CMOP = Cardiomyopathy; CONG = Congenital; COR = Corticosteroids; EV = Everolimus; LAD = Left anterior descending; MM = Mycophenolate; MPA = Main Pulmonary Artery; RCA = Right Coronary Artery; RPA = Right Pulmonary Artery; TC = Tacrolimus; TX = Transplant; U = Unknown
Table 2. Characteristics of the fistulas.
LAD = Left anterior descending; MPA = Main Pulmonary Artery; RCA = Right Coronary Artery; RPA = Right Pulmonary Artery
Compared to the previous descriptions of this finding, we didn’t find any statistically significant difference in the sex, age at heart transplant, indication for heart transplant, induction agent, immunosuppressant treatment nor median time of follow-up between the group with fistulas and the one without. In contrast to what is suggested in the literature, we have not found any statistically significant difference in the ischaemia time. No relationship has been found between the characteristics of the fistulas and mortality or retransplantation either.
Discussion
Coronary artery fistulas are a rare coronary anomaly, Reference Zaban, Elshershari and Hoyer4 and is found in 0.2% of patients undergoing coronary angiography. Although heart transplantation has been associated with a higher frequency of this finding. Reference Wei, Azarbal and Singh2
In the literature, coronary-to-right ventricle fistulas in heart transplant patients is suggested to be a secondary effect of myocardial biopsies performed on them. Reference Wei, Azarbal and Singh2 More recent reports show an important number of patients with fistulas connecting to other places, and some new potential mechanisms have been suggested, including angiogenesis secondary to the post-cardiac transplant inflammatory state, surgical trauma, and hypoxia. Reference Wei, Azarbal and Singh2
We report a series of paediatric heart recipients with coronary artery fistulas connecting the donor’s heart with structures from recipients. We describe two cases of coronary artery fistulas connecting the graft to systemic arteries of the receptor, a very unfrequent finding in pediatrics, as previously reported. Reference Camprubí-Tubella, Albert-Brotons and Betrián-Blasco3
We found a similar incidence of coronary artery fistulas (43.3%) than previous descriptions, Reference Allen, Goldstein and Pahl1 as well as a similar distribution of the fistula, but with a higher percentage of medium/large fistulas (47.3%). Also, as shown in the multicentre series of Allen et al., Reference Allen, Goldstein and Pahl1 all the non-cameral coronary artery fistulas were discovered on the first angiography and tended to disappear or decrease its size during the follow-up. We have also not found any correlation between non-cameral coronary artery fistulas and any adverse outcome or requirement of intervention.
Although the aim of our study was not to find out the possible mechanisms of the fistula’s formation, we can’t avoid thinking over the causes of that neoangiogenesis. Given that the fistulas are already present in the first coronarography, its appearance should be closely related to the changes that happen during the transplant procedure. A suitable hypothesis would be that the great post-inflammatory state that occurs after a transplant surgery, stimulates vascular growth factors, and has a determinant role in stimulating angiogenesis, and so, the formation of fistulas. This hypothesis would also agree with the downregulation of the fistula’s flow or its disappearance on the control coronary angiographies, because as we distance from the transplant, the organism tends to return to its basal state and the pro-inflammatory state decreases.
It is worth mentioning the main limitation of our study is a low size sample, making it very difficult to reach any statistical significance of the relation between fistula formation and the studied factors. Another limitation would also be the study has been done retrospectively and some patients have a short time of follow-up, so in many cases, we have just one coronary angiography.
In conclusion, non-cameral coronary artery fistulas after heart transplant in the paediatric population are not an uncommon finding, but those draining to systemic arteries seem to be less common. It would be necessary to carry out larger studies to better establish how frequent this finding is and if there are any involved factors. All fistulas showed involutional tendency. The main pathophysiological hypothesis for fistula formation is a great post-transplant pro-inflammatory state. No statistically significant relationship between non-cameral coronary artery fistulas and any of the studied factors were found, neither any relationship with mortality nor loss of the graft.
Acknowledgements
We would also like to acknowledge the invaluable contributions made by Dr. Paola Dolader (Pediatric Cardiac transplant unit) and Dr. Raul Abella (Pediatric Cardiac Surgery) and his surgical team.
Financial support
This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.
Conflict of interest
None.
Ethical standards
The authors assert that all procedures contributing to this work were in accordance with the ethical standards of the institutional research committee and with the 1975 Helsinki Declaration and its later amendments or comparable ethical standards.
