Impact of antiviral treatment on long-term prognosis in non-immunocompromised patients with CMV reactivation

This is a retrospective cohort study evaluating patients with CMV reactivation at Konkuk University Hospital, an 850-bed, community-based tertiary medical center in Seoul, Republic of Korea, during the study period between January 2010 and February 2018. Patients were tested for CMV DNAemia using real time polymerase chain reaction (RT-PCR), based on the clinical judgement of the attending physician suspecting CMV reactivation. CMV reactivation was defined as detection of CMV DNAemia (> 270 copies/ml) in whole blood by RT-PCR. Patients with clinical presentation suggesting primary CMV infection, such as infectious mononucleosis-like illness or fever of unknown origins were not included. Patients with hematologic or oncological disorders, with HIV infection, with previous history of pathologically confirmed CMV disease, those who had received SOT, or those who discontinued ganciclovir within 3 days or received other antiviral agents were excluded. Patients who died within 30 days were also excluded from the analysis (Fig. 1). Patients were divided into ganciclovir treatment group and non-treated group. Ganciclovir was administered at 5 mg per kilogram every 12 h, and the dose was adjusted according to patients’ renal function. The dose was not adjusted according to the amount of CMV PCR. We additionally included a third group of non-immunocompromised patients who tested negative for CMV infection, matching age, sex, and length of hospital stay after CMV test. The present study was approved by the Institutional Review Board (IRB) of Konkuk university of medical center (#2020–04-040). Informed consent was waived by the IRB of Konkuk university of medical center since the electronic medical record (EMR) was reviewed retrospectively with de-personalized identification number.

Data were collected from hospital database including the administrative, pharmaceutical, and laboratory information at Konkuk university Hospital. EMRs were reviewed, including treatment with ganciclovir, age, sex, underlying disease, intensive care unit (ICU) stay, length of hospital stay after CMV reactivation, mechanical ventilation, presence of pneumonia, and maximum titer of CMV PCR. CMV pneumonitis was defined as a case of atypical infiltration on Chest X-rays, progression of X-ray lung lesions despite broad spectrum antibiotic use, and bronchoaveolar lavage (BAL) or sputum CMV PCR of 10,000(copies/ml) or more [15]. Diagnosis of CMV pneumonitis was made regardless of concomitant bacterial infection. The severity of underlying disease at the time of CMV reactivation was estimated using Charlson’s weighted index of comorbidity (CWIs). The severity of illness at the time of CMV reactivation was assessed using quick sequential organ failure assessment (qSOFA) score. The primary endpoint was the 30-months survival. The secondary endpoints were 90-days and one-year survival.

The Real-Q CMV DNA quantification kit (Real-Q assay; BioSewoom, Seoul, Korea) was used for detection of CMV DNA in whole blood. The kit is designed to detect CMV genome in purified DNA samples via the gene coding for the glycoprotein B (gB). DNA was extracted on a MagNA Pure 96 instrument (Roche Diagnostics, Mannheim, Germany) with the “Pathogen Universal Protocol” (elution volume, 100 μL), according to the manufacturer’s recommendation. Detection and quantification of CMV DNA was performed by using the Real-Q assay. The PCR reaction was performed in a total volume of 25 μL (15 μL of PCR reaction mixture including probe and primers, plus 10 μL of template DNA). Results are indicated in copies per milliliter of whole blood (copies/ml) [16]. Additional tests for ganciclovir-resistant mutations were not conducted.

A total of 737 patients showed CMV DNAemia during the study period. After excluding immunocompromised hosts, patients died within 30 days, and patients with other exclusion criteria, 136 adult patients with CMV DNAemia were finally included in the study cohort for the evaluation of long-term prognosis (Fig. 1). Demographic characteristics of the patients between ganciclovir-treated (66 patients, 48.5%) and non-treated group (70 patients, 51.5%) are shown in Table 1. There were no significant differences between the ganciclovir-treated group and non- treated group with regard to age, sex, mechanical ventilation, qSOFA and CWIs. The proportion of patients who stayed in ICU (43.9% vs. 24.3%, respectively, P = 0.015) and who had > 30 days of hospitalization (63.6% vs 44.3%, respectively, P = 0.024) were higher in the ganciclovir-treated group compared to the non-treated group. The number of patient with CMV PCR titer over 5000 copies/ml is higher in ganciclovir-treated group than non-treated group (48.5% vs 22.9%, respectively, P = 0.002).

In the univariate analysis, no variables were positively associated with 30-months survival (Table 2). In the multivariable analysis, qSOFA score (HR 1.472, 95% CI 1.001–2.163, P = 0.049) were significantly associated with 30-months survival. Ganciclovir treatment was not associated with 30-months survival (HR 1.307, 95% CI 0.759–2.251, P = 0.334). The one-year survival also showed statistically similar results. In the multivariable analysis, qSOFA score was significantly associated with one-year survival (HR 1.595, 95% CI 1.049–2.426, P = 0.029). Ganciclovir treatment was not associated with 90-days and one-year survivals (Table S1 and S2). Ten out of 66 patients (15.2%) who treated with ganciclovir probably progressed to CMV pneumonitis, which showed BAL or sputum CMV PCR over 10,000 copies/ml. On the other hand, no patient out of 70 patients who were not received ganciclovir progressed to CMV pneumonitis. Four patients out of 136 patients developed cardiovascular event, but there were no statistically differences between ganciclovir-treated group and non- treated group. (6.1% vs 2.9%, respectively, P = 0.363). We further analyzed the 30-months survival between patients who undergone CMV test but were negative and patients with CMV reactivation. One hundred thirty-one patients were further analyzed by matching the CMV activation group (Table S3). Regardless of whether ganciclovir was treated or not, CMV reactivation did not affect the 30-month survival rate in the multivariate analysis (Table S4).