Five databases were used to find 2818 articles related to COVID-19 infection in HT and heart-kidney transplant recipients. Thirty-three studies were then deemed eligible for inclusion in this review [7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39]. A PRISMA flow diagram detailing the process of identification, inclusion, and exclusion of studies is shown in Fig. 1.
Of the thirty-three studies, eleven were retrospective case series [7, 11, 14, 19, 20, 26, 29,30,31,32,33, 38, 39], while twenty-two were case reports [8,9,10, 12, 13, 15,16,17,18, 21,22,23,24,25, 27, 28, 34,35,36,37]. All were peer-reviewed. All articles were published in 2020 and 2021. The total number of patients included in the review is 74.
Risk of bias within the studies
In comparison of the case reports, all the articles were determined to have a very low risk of bias. All the retrospective studies were rated as having low risk of bias. One case report had an intermediate risk of bias . These results are included in Table 1.
Results of individual studies
Categorical variables were described as number (%) and continuous variables were described with mean ± standard deviation or median (IQR), as appropriate.
The search identified 2818 publications. After removing duplicates and screening for exclusion and inclusion criteria, 33 publications were included in the analysis (Fig. 1). A summary of findings from all studies is presented in Table 2.
Synthesis of results
We identified 74 cases of SARS-CoV-2 infection in heart transplant recipients with data available on individual patients. The mean age was 60.5 ± 15.8 years, and 82.4% were males.
The reports came from 9 countries, including 27 (36.4%) from North America, 27 (36.4%) from Europe, 18 (24.3%) from South America, and 2 (2.7%) from Asia. In total, 68 (91.8%) were HT recipients, while 6 (8.1%) were HKT recipients and 55 (74.3%) were > 1-year post-transplant. The youngest HT recipient infected with COVID-19 who deteriorated and underwent retransplantation was a 22-year-old woman in France .
Patients were identified based on testing positive for COVID-19, having a prior HT or HKT. Immunosuppression was assessed based on (1) background immunosuppression; (2) changes in immunosuppression while admitted and then selected clinical outcomes were evaluated.
The median time from symptom onset to presentation was 3.5 (2.75–7) days, and fever (59.4%) and cough (59.4%) were the most prevalent symptoms, followed by dyspnea (47.2%), diarrhea (33.8%), and myalgia (25.6%). Fatigue (12.2%), anosmia (12.2%), anorexia (9.4%), and rigors (8.1%) were less common. Bilateral pulmonary infiltrates were seen on chest X-ray in 44 (59.4%) of patients, absent or not, reported in the rest.
The mean temperature recorded was 38.2 ± 0.8 °C, and the mean blood pressure was 123/79 ± 15.6/11.1 mmHg. The distribution of presenting complaints and associated symptoms is found in Table 3.
Past medical history
Of the cohort with fully reported comorbidities, the distribution of comorbidities was as expected from a post-transplant population. Hypertension was reported in 50%, diabetes mellitus in 36.4%, and chronic kidney disease in 31.1% patients and there was significant overlap of these comorbidities in the same patients. The list of comorbidities is also found in Table 3.
A summary of laboratory tests is found in Table 4. Mean white blood cell (WBC) count was normal at 6027 ± 3383 cells/mm3, and C-reactive protein (CRP) levels were increased at 42.6 ± 48.2 mg/L. Oxygen saturation and arterial blood gas was unavailable for most studies. Some studies reported either increased troponin-T [7, 9, 14, 19, 24, 25, 27] and/or NT-pro brain natriuretic peptide (NTproBNP) [11, 25].
A list of the diagnostic tests and imaging techniques utilized in the studies is provided in Table 5. All patients were confirmed COVID-19 positive. In patients who underwent chest imaging with computed tomography (CT) scans or chest X-rays (CXR), bilateral infiltrates and ground-glass opacities were discovered in 59.4% of patients.
In five of the studies, 2-D echocardiography revealed decreased LV ejection fraction (EF) [11, 14, 24, 28, 32]. Of the five studies, one did not undergo repeat endomyocardial biopsies due to a recent negative biopsy and the other who had severe biventricular dysfunction that required retransplantation had an endomyocardial biopsy that ruled out acute humoral rejection and no mention was made of viral particles within the myocardium or inflammation . The last case  had no evidence of acute cellular rejection (grade 0R) or antibody-mediated rejection and ejection fraction and recovered after 4 days spontaneously.
General management strategies
In terms of anti-inflammatory, antiviral therapies, and immunomodulators, steroids were used most (39.2%), hydroxychloroquine was given to 37.8% and remdesivir to 9.4%, and other anti-viral agents such as lopinavir/ritonavir (5.4%), favipiravir (1.3%), and ganciclovir (2.7%) were rarely used. The use of interleukin 6 (IL-6 inhibitors) mainly tocilizumab and clazakizumab was infrequent occurring in only 9.4% of the cohort. The utilization of antibiotics was common in 25.6% of patients.
The use of vasopressors, human immunoglobulin, renal replacement therapy (excluding long term dialysis) (4%), and venoarterial extracorporeal membrane oxygenation (VA ECMO) (2.7%) was rare. An outline of the management and immunosuppressive therapies used is outlined in Table 6.
Immunosuppressive management strategies
In a majority of patients, 40 (54.1%) were on triple regimen of immunosuppression with either an anti-metabolite (mycophenolate mofetil (MMF), mycophenolic acid (MPA) or azathioprine), calcineurin inhibitor (CNI) (tacrolimus or cyclosporine), steroid, or mammalian target of rapamycin inhibitor (mTORi) (sirolimus or everolimus). Single agent for immunosuppression was rare. Of the patients on triple regimen, the commonest combination was an anti-metabolite, CNI, and steroid combination (48.6%) of the patient.
A 2-drug immunosuppression regimen of anti-metabolite and CNI was seen in 59.4% of patients.
The commonest strategy of managing immunosuppression while admitted was discontinuing or reducing the dose of the anti-metabolite in 36 out of the 63 patients (57.1%) on anti-metabolites. Anti-metabolite dose was reduced in 12 of the 63 patients (19%) and held while admitted in 24 of the 36 patients (38.1%). There was no change in immunosuppression in 20 of the 74 patients (27%) on immunosuppression. The CNI was decreased or stopped at some point in the management course in 11 out of the 64 patients (17.4%) on CNI.
Patient outcomes were included in most of the case reports and case series. Among cases with fully reported data, the median length of stay was 13 days.
Most patients (n = 68, 91.8%) were hospitalized and 5 (6.7%) were categorized as having mild COVID-19 and treated as outpatients. Three patients (4.1%) remained inpatient with uncertain outcome at the time of publication.
Fourteen deaths were reported—two patients died from progressive ARDS and vasoplegic shock that progressed to multiorgan system failure; the third patient had a significant history of multiple cellular and humoral rejection, presented with cardiac arrest, and died. The fourth patient had a hospital course complicated by ischemic cerebrovascular accident and died. Other listed causes of death were septic shock, cardiogenic shock and acute respiratory failure, cardiac arrest, and multi-system organ failure. Two patients were inpatient at the time of case publication , while 55 patients were discharged home. Twelve patients required ICU admission (16.2%) of the cohort. Survival in hospitalized patients was 80.2%, while that in patients admitted to the ICU was very poor at 25%.
Outcomes were known for 69 patients: 55 survived, 14 died (mortality was 20.2%).
Risk of bias across the studies
Due to the nature of the descriptive studies, the results presented are liable to investigator bias, selection procedure bias, and selection bias.
Limitation of the study
Most of the reports were observational in nature. Statistical analyses were not performed as there were no control/comparison groups in the included studies. All the desired datasets were not available in all the reports and case series. The therapies for COVID-19 were also changing rapidly during the first 9 months of the pandemic and could have impacted management patterns and outcomes.