Incidence and severity of cytomegalovirus infection in seropositive heart transplant recipients

Abstract Background The frequency and significance of cytomegalovirus (CMV) infection in seropositive (R+) heart transplant recipients (HTR) is unclear, with preventative recommendations mostly extrapolated from other groups. We evaluated the incidence and severity of CMV infection in R+ HTR, to identify risk factors and describe outcomes. Methods R+ HTR from 2010 to 2019 were included. Antiviral prophylaxis was not routinely used, with clinically guided monitoring the local standard of care. The primary outcome was CMV infection within one‐year post‐transplant; secondary outcomes included other herpesvirus infections and mortality. Results CMV infection occurred in 27/155 (17%) R+ HTR. Patients with CMV had a longer hospitalization (27 vs. 20 days, unadjusted HR 1.02, 95% CI 1.00–1.02, p = .01), higher rate of intensive care readmission (26% vs. 9%, unadjusted HR 3.46, 1.46–8.20, p = .005), and increased mortality (33% vs. 8%, unadjusted HR 10.60, 4.52–24.88, p < .001). The association between CMV and death persisted after adjusting for multiple confounders (HR 24.19, 95% CI 7.47–78.30, p < .001). Valganciclovir prophylaxis was used in 35/155 (23%) and was protective against CMV (infection rate 4% vs. 27%, adjusted HR .07, .01–.72, p = .025), even though those receiving it were more likely to have received thymoglobulin (adjusted OR 10.5, 95% CI 2.01–55.0, p = .005). Conclusions CMV infection is common in R+ HTR and is associated with a high burden of disease and increased mortality. Patients who received valganciclovir prophylaxis were less likely to develop CMV infection, despite being at higher risk. These findings support the routine use of antiviral prophylaxis following heart transplantation in all CMV R+ patients.

Investigator Grant, Grant/Award Number: GNT1174673 were less likely to develop CMV infection, despite being at higher risk. These findings support the routine use of antiviral prophylaxis following heart transplantation in all CMV R+ patients.

K E Y W O R D S
cytomegalovirus, heart transplant, recipient positive

BACKGROUND
Cytomegalovirus (CMV) remains a major contributor to morbidity and mortality in heart transplant recipients (HTR). 1,2 It causes direct clinical illness and indirectly contributes to poor outcomes including cardiac allograft vasculopathy. [3][4][5][6][7] Preventative approaches are effective and include antiviral prophylaxis or pre-emptive therapy. 8,9 Primary prophylaxis with valganciclovir has emerged as the dominant CMV preventative strategy and is associated with less CMV infection and improved outcomes. [10][11][12][13] Despite this, CMV infection remains a problem and is difficult to predict. [14][15][16] The pre-emptive approach can be challenging and resource-intensive to implement. Routine use of prophylaxis increases the risk of side effects such as neutropenia, drug costs, and antiviral resistance.
Patients who acquire CMV from their transplant donor (donor positive/recipient negative, D+/R-) are at the highest risk of infection and generally receive 6 months of valganciclovir prophylaxis. [17][18][19][20] Recipients who have had an infection with CMV prior to transplantation (R+ HTR) are considered intermediate risk unless they receive antilymphocyte agents such as thymoglobulin for induction or rejection. [21][22][23] Their risk of CMV is less clear, with estimates in prior studies ranging widely. [24][25][26][27][28][29] Practice varies in these patients with many receiving 3-6 months of prophylaxis but others undergoing virologic monitoring.
Few dedicated studies have been performed in this subgroup, with recommendations mostly extrapolated from other solid organ transplant groups. 30,31 Given the widespread use of universal valganciclovir prophylaxis in many transplant programs, defining the burden of CMV infection in a contemporary cohort of R+ HTR who have not received prophylaxis is difficult. At our center, antiviral prophylaxis has been implemented selectively among R+ HTR who are considered at higher risk. The aims of this retrospective cohort study were to describe the frequency and severity of CMV infection in our R+ HTR, to understand the relationship between CMV and clinical outcomes, and to explore risk factors for CMV infection.

Study design and population
Patients who received a heart transplant at our institution between 2010 and 2019 who had positive CMV serology prior to transplantation were eligible for inclusion. Patients were excluded if they died during or less than 72 h following the transplant surgery. Clinical data were obtained from hospital medical records. The study was approved by the Alfred Health Ethics Committee, with informed consent not required given its retrospective nature and minimal risk.

Outcomes
Our primary outcome, CMV infection, was defined as a positive test for CMV at any anatomical site (mostly serum viral load testing).
Asymptomatic patients with an isolated low-level positive test that spontaneously resolved without antiviral therapy who did not have any other clinical evidence for disease were not classified as CMV infection. Secondary outcomes included herpes simplex virus (HSV) and varicella zoster virus (VZV) infections, which were classified as localized if restricted to one anatomic site (e.g., single dermatomal shingles or oral HSV) or disseminated if beyond this. We also examined overall mortality.

CMV prophylaxis, diagnosis, and treatment
Antiviral prophylaxis was not routine in this cohort but rather targeted to patients considered high-risk by their treating clinicians, such as those receiving an anti-lymphocyte agent for induction or rejection. Duration of therapy was individualized and determined by clinical and virologic endpoints but was generally at least 2-3 weeks.

Statistical analysis
Categorical data were reported as counts with percentages, continu-

Valganciclovir prophylaxis use
Because of the targeted use of prophylaxis in our cohort and the known association between this and the timing and likelihood of CMV infection, we wanted to understand the differences between the 35 (23%) patients who did and the 120 (77%) who did not receive prophylaxis (
Several factors were identified as being associated with increased mortality (  Figure S1).

DISCUSSION
CMV remains an important opportunistic infection in HTR and contributes to poor outcomes, though the risk and significance in R+ patients has not been well defined. In this study we have described