Heart transplantation (HTX) confers excellent long-term survival in select patients with symptomatic end-stage heart failure. Cardiac surgery after heart transplantation (CASH) is a rarely used approach to improve allograft function and quality of life, and prolong survival (1), but has been described only in case reports and single-center experiences (2, 3). Although symptomatic tricuspid regurgitation is known as the most common cause for CASH (2), other valvular diseases, aortic pathology, and cardiac allograft vasculopathy (CAV) can occur in the HTX population (1–3). Coronary artery bypass grafting (CABG) is a safe surgical therapy for CAV, in selected cases, with acceptable long-term outcomes (1, 2). Furthermore, transcatheter and minimally invasive strategies to treat allograft pathologies have been reported with excellent short-term outcomes (4–9).

In this retrospective, multicenter, cohort study, we evaluated the safety of CASH and risk factors for subsequent morbidity and mortality.

The study cohort was comprised of heart transplant recipients who underwent CASH between January 1984 and December 2020. CASH includes cardiac and aortic transcatheter interventions but does not include retransplantation, pacemaker/defibrillator placement, or reoperation due to bleeding complications. In total, 60 high-volume cardiac transplant centers performing more than 15 HTXs per year were invited to participate in the study. Each participating center obtained ethics approval from its institutional review board (Vienna ethics committee reference number: 1894/2017) and a data use agreement was executed with each center.

Relevant data on demographics, medical history, medications, surgeries, cardiac testing, outcomes, and complications were collected from the patients’ medical records and were de-identified by each center’s study coordinator. A password-protected, validated Excel file was sent to the primary investigator for statistical analysis. The European System for Cardiac Operative Risk Evaluation (EuroSCORE II) (10, 11), a validated scoring system that predicts the risk of in-hospital mortality after major cardiac surgery, was calculated.

The inverse Kaplan–Meier method was used to quantify median follow-up (12). To evaluate the effect of selected clinical factors on in-hospital mortality, we used univariate logistic regression models accounting for the center as a random effect. Survival after reoperation was described using the Kaplan–Meier method. Two-sided p < 0.05 were considered statistically significant. Statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, United States).

Our multicenter study describes the largest cohort of patients with CASH worldwide. We demonstrate that CASH is an acceptable therapy for different cardiovascular pathologies early and late after HTX. Overall in-hospital and 1-year mortality after CASH were acceptable (9.1 and 13.8%, respectively). Urgent indication for CASH, such as endocarditis, infected pseudoaneurysm, aortic dissection, and iatrogenic complications, was strikingly, but not statistically significantly, associated with higher in-hospital mortality. These indications are also associated with high mortality in the general heart surgery population, and in previous reports on CASH (2, 3). Postoperative acute kidney injury requiring renal replacement therapy was the most common complication after CASH and was associated with higher in-hospital mortality. Age at time of CASH and higher EuroSCORE II were also associated with higher in-hospital mortality. However, the effect of age was not statistically significant.

The incidence of postoperative systemic infection and deep sternal wound infection was low (13, 14). Reduction or discontinuation of mammalian target of rapamycin (mTOR) inhibitors several weeks to months prior to elective CASH may be considered due to reports of impaired wound healing (15).

Severe tricuspid regurgitation is the most common reason for CASH (16). HTX-specific tricuspid valve pathologies are biopsy-induced injury to the chordae (leaflets), ischemic injury to the papillary muscle as a consequence of CAV or rejection, and distortion of the valvular apparatus [biatrial implantation technique (2, 17–24). Endocarditis is rare, but is more common than in the general population due to increased risk arising from immunosuppression and frequent central venous access, such as endomyocardial biopsies (25).

Indications for surgery must be carefully considered, especially in patients with ventricular dysfunction and/or pulmonary hypertension because they are at risk of right ventricular failure after CASH. Potential underlying disease processes, such as CAV or acute rejection must be ruled out prior to CASH.

Tricuspid valve repair with annuloplasty should only be performed in HTX patients with annular dilation; valve replacement is recommended in complex valvular pathologies or residual regurgitation after repair (21). Besides the risks associated with life-long anticoagulation, mechanical valve replacement rules out the right-sided endomyocardial biopsy. Interventional edge-to-edge repair of the tricuspid valve using the MitraClip system has been reported with perioperative success (26), but long-term data are lacking.

Mitral valve surgery after HTX has been described rarely. Pathology can be related to annular dilation or degeneration of the leaflets or papillary muscles, typically as a consequence of CAV or acute rejection, often accompanied by ventricular dysfunction (27–29). Iatrogenic injury after left-sided endomyocardial biopsy may lead to acute mitral regurgitation. Mitral stenosis has been described in dialysis-dependent HTX patients in association with hyperparathyroidism (30). Mitral valve replacement may be preferred over repair due to complex valvular pathologies in patients with HTX, and to achieve shorter cardiopulmonary bypass times in patients with ventricular dysfunction. In three of our patients, concomitant tricuspid valve repair or replacement was performed without perioperative complications, but postprocedural death occurred 0.5, 0.9, and 2.8 years after CASH.

Minimally invasive CASH via thoracotomy confers the advantage of avoiding resternotomy. Transcatheter interventions may be reasonable for patients at high surgical risk with appropriate valvular pathology, but the atrial and atrio-ventricular anatomy can be challenging due to distortions after HTX (9).

Cardiac surgery after heart transplantation due to symptomatic degenerative aortic stenosis in the allograft may occur more often with the acceptance of marginal donor hearts. Single-center case reports have reported acceptable perioperative outcomes (16, 31–33), and case reports on TAVR in HTX patients with high surgical risk have presented data on favorable short-term outcomes (4–8); however, long-term data, as well as data are data on the durability of biological and mechanical valves, are lacking. Aortic valve endocarditis after HTX is extremely rare (34). Our cohort included a patient with biological valve replacement due to endocarditis and 7.7 years of follow-up without valvular degeneration.

Aortic surgery was the most heterogeneous group in our series. Case reports have described successful CASH for ascending aortic aneurysm, dissection, or pseudoaneurysm of the aortic anastomosis (1–3, 35). Aortic pathologies after HTX typically arise at the site of aortic anastomosis due to flow turbulence (donor/recipient aortic size mismatch), infection, or hypertension (36). Due to urgency, the precise preoperative planning of surgery is limited in acute type A dissection and infected pseudoaneurysm. Our data are in line with previously published data (1). Two patients with urgent surgery due to infected pseudoaneurysm had early surgery-related deaths, highlighting the high risk of mortality associated with this rare disease (36).

Surgical revascularization for CAV is safe, with acceptable long-term results, in HTX patients with acceptable coronary anatomy (type A lesion, Stanford Classification), and elective indication for surgery (1–3, 37). This approach is generally limited, however, by the diffuse nature of CAV (37, 38) and inexorable disease progression in most patients (3, 39), which may necessitate additional interventions (3, 40). The patency of arterial grafts is superior to vein grafts in CASH, with patency of the internal thoracic artery reported up to 20 years (3, 41). Though the left internal thoracic artery was used in most of our patients, the radial artery had good mid-term results in a published case series and may be an adjunct graft in patients with CABG prior to HTX (3).

Constrictive pericarditis after HTX is rare and is typically associated with recurrent pericardial effusions, allograft rejection, or biopsy-related complications (42–44). As most of the participating centers did not report any cases of constrictive pericarditis requiring pericardiectomy, we cannot draw conclusions as to whether patients were undiagnosed at some centers and whether surgical and/or treatment strategies differed at the three centers that reported cases of constrictive pericarditis.

This study is limited by its retrospective nature, the small numbers for each procedure type, and the fact that participation was by invitation only—not all eligible heart transplant centers reported data. It may be of interest, however, for multinational heart transplant registries to begin collecting data on CASH, given the growing population of heart transplant recipients worldwide with improved long-term survival and the increasing use of minimally invasive surgical and transcatheter approaches.

We conclude that CASH is generally safe, with low in-hospital mortality and postoperative complications in carefully selected patients. Nevertheless, it is rarely performed, with differences in practice between heart transplant centers worldwide. Higher EuroSCORE II and postoperative acute kidney injury are associated with higher in-hospital mortality.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

JG and AA-Z designed the study and wrote the manuscript. JG organized the database. AK performed the statistical analysis. All authors contributed to manuscript revision and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.