Active Cytomegalovirus Infection in Acute Respiratory Distress Syndrome Patients: Incidence, Risk Factors, and Clinical Outcomes


 Background: Acute respiratory distress syndrome (ARDS) patients have been reported to have a high seroprevalence of cytomegalovirus (CMV). However, the role of active CMV infection in ARDS patients has not been clearly established.Objective: This study aimed at determining the incidence, risk factors, and clinical outcomes of active cytomegalovirus (CMV) infection in acute respiratory distress syndrome (ARDS) patients.Methods: We retrospectively reviewed medical records for ARDS patients who had been admitted to the intensive care unit (ICU) from January 1st, 2018 to December 31st, 2020 at a national teaching hospital in China. Study participants were divided into active CMV infection and non-active infection groups based on CMV DNAemia within a 28-day hospitalization period in ICU. Clinical features, laboratory findings, treatment measures, and clinical outcomes were compared between the two groups.Results: Among 168 ARDS patients, 31 (18.5%) exhibited active CMV infection within the 28-day hospitalization period in ICU. In multivariate logistic regression analysis, monocyte counts, hemoglobin levels, blood transfusion, and septic shock were significantly independently associated with active CMV infection (p < 0.05). Oxygenation (PaO2/FiO2) of active CMV infection patients was worse than for non-active CMV infection (p < 0.05). Duration of invasive mechanical ventilation, 28-day ventilator-free days, length of ICU stay, and 28-day all-cause mortality rates in active CMV infection patients were significantly higher than in those without active CMV infection (p < 0.05).Conclusions: Active CMV infection is common among critically ill ARDS patients. Monocytes, hemoglobin, blood transfusion, and septic shock are risk factors for active CMV infection, which has a negative effect on oxygenation. Moreover, active CMV infection is associated with several adverse prognoses. Prospective studies should be performed to evaluate the impact of prophylactic antiviral therapy for prognoses among ARDS patients.

ARDS patients [6]. However, studies have not elucidated on the role of active CMV infection in ARDS patients. Moreover, there is a need to establish their clinical incidences, risk factors, and prognoses. Therefore, this study aimed at investigating the incidences, risk factors, and clinical outcomes of active CMV infection in ARDS patients.

Study participants
We retrospectively reviewed medical records for ARDS patients who had been admitted to the ICU between January 1st, 2018 and December 31st, 2020 at the First A liated Hospital of Guangzhou Medical University, China. Detection of active CMV infection was done using real-time PCR with blood plasma as the sample. Study participants were divided into active CMV infection (CMV DNAemia ≥ 500 copies/mL) and non-active infection groups (CMV DNAemia < 500 copies/mL) within 28-day hospitalization period in the ICU. The Ethics Committee of the First A liated Hospital of Guangzhou Medical University approved the study protocol (No. GY-2021-K04). The need for an informed consent was waived due to the retrospective nature of the study.

Data collection
Clinical data for 168 cases were extracted from the electronic records by two independent researchers who subsequently cross-checked the data for accuracy, including clinical features, laboratory ndings, treatments, complications, and clinical outcomes. Disagreements between the two researchers were further adjudicated by a third independent reviewer who was an expert in critical care medicine. Data were entered into an electronic database for statistical analysis.

Study de nitions
The inclusion criteria were: i. Patients that met the diagnostic criteria for ARDS (Berlin de nition 2012) [7] and ii. Patients aged > 18 years. The exclusion criteria were: i. Pregnant or lactating patients; ii. Survival time < 72 h; iii. Lack of CMV detection; and iv. Patients administered with antiviral therapy before ICU entry.
According to the World Health Organization Standard, active CMV infection was de ned as viral load greater than or equal to 500 copies/mL in plasma [8]. Screening for CMV viral load in plasma was part of routine clinical practice in the hospital.

Study outcomes
Assessed outcomes included incidences, risk factors, and prognoses of the study participants. Moreover, we assessed the impact of CMV infection on oxygenation (PaO 2 /FiO 2 ).

Statistical analysis
Continuous variables were expressed as Mean ± SD or Median (interquartile ranges, IQRs) and compared using the Wilcoxon rank-sum test. Categorical variables were expressed as counts and percentages, and compared using the Fisher's exact test. Risk factors for active CMV infection were screened in the univariate logistic regression model, variables with p≤ 0.05 were considered potential risk factors and were further imported into the multivariate logistic regression analysis. Active CMV infection risk model was established by calculating the regression coe cient (β), odds ratios (OR), and 95% con dence interval (CI). The receiver operating characteristic (ROC) curve was used to evaluate the predictive value of active CMV infection. The area under ROC curve (AUC), 95% CI, p-value, cut-off, sensitivity, and speci city were calculated. The signi cance threshold was set at a two-sided p≤ 0.05. Atatistical analyses or charting were performed using SPSS version 25.0 (SPSS Inc., USA) and GraphPad Prism 8.0 (Graphpad Software Inc., USA).

Active CMV infection
During the study period, a total of 4,261 patients were admitted to the ICU. Among them, 4,022 patients were initially excluded for reasons that included: i. Not meeting the ARDS diagnostic criteria (n=3,987); ii.
Survived for less than 72 h (n=23); iii. Pregnant or lactating women (n=7), and iv. Younger than 18 years (n=5). Preliminarily, 239 ARDS patients were screened, however, 71 were excluded for; i. Lacking CMV detection (n=69) and ii. Receiving antiviral therapy before entering the ICU (n=2). Finally, a total of 168 ARDS patients were enrolled (Figure 1) Among the 168 enrolled ARDS patients, there were 31 (18.5%) cases of active CMV infections within the 28-day ICU hospitalization period ( Figure 1). A total of 25 (80.7%) of the 31 cases with active CMV infection exhibited positive results at ICU admission, moreover, 4 cases at day 14 and 2 cases at day 21.

Clinical features
We recruited 168 patients with complete data, among whom 111 were male (66.1%). The mean age for all study participants was 58 ± 15 years. Based on the median scores for APACHE (20) and SOFA (9), the disease was very severe. Patients with mild, moderate, and severe ARDS were 32 (19.0%), 66 (39.3%), and 70 (41.7%), respectively. The main cause of ARDS was pneumonia (91.7%). Patients' heart and respiratory rates were higher. The main comorbidities were hypertension (n=47, 28.0%), cardiovascular diseases (n=47, 28.0%), and connective tissue diseases (n=31, 18.5%). Except for connective tissue disease [35.5% vs. 14.6% (n), p=0.011], differences in these clinical characteristics between the two groups were not signi cant (Table 1) 1.32, p=0.030). There were no signi cant differences for other laboratory ndings between the two groups ( .003] were found to have been highly administered in the group with active CMV infection before ICU admission. Other therapeutic measures were not signi cantly different between the two groups (Table 4).
Severe pneumonia with a prevalence of 91.7% was found to be a major complication in ARDS patients.
Compared to the non-active CMV infection group, the number of ARDS patients diagnosed with septic shock [90.3% vs. 71.5% (n), p=0.037] in the active CMV infection group was higher. Although statistical signi cance was not reached, the rate of AECOPD was low in cases without (11.7%) than in cases with (22.6%) active CMV infection ( Table 4) (Table 4).

Oxygenation in uence
Continuous observation of arterial oxygenation over 7 days of active infection with CMV revealed that the CMV group exhibited worse outcomes in oxygenation within 5-day hospitalization in ICU than the nonactive CMV infection group. On day 5, oxygenation was signi cantly worse in the active CMV group than in the non-active CMV infection group [median: 179 vs. 193 (P/F), p=0.046] ( Figure 2).

Discussion
We evaluated the incidence, risk factors, and clinical outcomes of active CMV infection in ARDS patients. Among ARDS patients, the incidence rate of active CMV infection was 18.5%. Clinical features, including connective tissue disease, monocytes, hemoglobin, platelet, Th/Ts, immunosuppressive drugs and blood transfusion, and septic shock were associated with active CMV infection. Monocytes, hemoglobin, blood transfusion, and septic shock were found to be independent risk factors for active CMV infection. Moreover, active CMV infection had a negative effect on oxygenation and was associated with adverse prognoses.
Active CMV infection is not a rare phenomenon among patients admitted to the ICU. It has been reported that the incidence of active CMV infection in critically ill patients is 31% [4]. However, the population included in most of the current studies did not differentiate between speci c disease types, especially ARDS. Two prospective studies revealed that the incidence of active CMV infection in ARDS patients was 18.6%-22.0% [9,10], which was comparable to our ndings. We found that most of the patients had active CMV infections at the time of ICU admission, implying that active CMV infection was likely to have been present before ICU admission. Therefore, detection of active CMV infection should not be restricted to ICU post-admission but should be extended to ICU pre-admission.
Active CMV infection has previously been associated with several conditions, including immunosuppressive drugs, blood transfusion, and septic shock [3,[10][11][12], consistent with our results. Immune dysfunction is closely correlated with the occurrence of active CMV infection [2,13]. Immunosuppressive drugs result in suppressed immune cell (especially T lymphocytes) levels, which inhibits viral clearance, making latent CMV infection more susceptible to active infection, as seen mostly in transplant patients [3,[13][14][15]. Active CMV infection through transfusion is a challenge in the treatment of critically ill patients. It has been reported that donations from new CMV-IgG-positive donors bear the highest risk for transmitting CMV infections because they contain elevated CMV-DNA levels, which is a risk factor for active CMV infection [16]. Leucocyte depletion of cellular blood products and selection of CMV-IgG-negative donations might reduce the occurrence of active CMV infection [16,17]. Sepsis induces active CMV infection through sepsis-related cytokine storm, which triggers transcriptional CMV replication, a mechanism that has been con rmed in animal models [18,19]. Furthermore, several studies have shown that active CMV infection impairs hematopoiesis and immune function [20,21], suppressing multiple blood cell levels, consistent with our results. The reason for this association is correlated with direct pathological damage caused by CMV infection and the indirect damage caused by in ammatory factors.
Active CMV infection has been strongly associated with sepsis, mechanical ventilation, as well as hypertension induced by glucocorticoids and catecholamines. Besides, there was no correlation for disease scores, such as the APACHE and SOFA scores. There is no evidence that CMV reactivation is agerelated, and whether it is gender-related or not has not been established [3]. Moreover, risk factors for active CMV infection in ARDS patients have not been clearly elucidated. We found that monocytes, hemoglobin, blood transfusion, and septic shock were independent risk factors for active CMV infection in ARDS patients. The mechanisms through which hemoglobin, blood transfusion, and septic shock cause active CMV infection have been described above. Monocytes are essential for effective control and clearance of viral infections, especially their direct involvement in non-speci c immunity and indirect regulation of speci c immunity [22]. Monocytes play a pivotal role in viral dissemination to organ tissues during primary infections and the following reactivation from latency [23]. Main targets of CMV are monocytes, where they induce their differentiation into macrophages [23,24]. Once CMV infected monocytes differentiate into macrophages, expression of immediate early viral genes are detectable, followed by viral replication and long term infectious viral particle release [23]. Furthermore, CMV have been shown to alter the expression of monocyte transcripts and are involved in in ammatory responses, which enhances CMV replication [23][24][25]. Therefore, suppressed monocyte levels directly affect CMV clearance and indirectly re ect elevated CMV replication levels.
Active CMV infection is associated with adverse prognoses for critically ill patients, consistent with our ndings, including prolonged duration of mechanical ventilation, increased length of hospitalization, and mortality [2-4, 9-12, 18]. These adverse prognoses are associated with various factors, including direct injury (such as CMV pneumonia) and indirect injury (such as immune disorders) [18,26]. We found that active CMV infection is associated with poor oxygenation in ARDS patients, as previously reported [12]. This mechanism may be involved in the initiation of pulmonary brosis by CMV and has been validated in in vivo experiments [27]. CMV may be a potential factor in the development of pulmonary brosis in ARDS patients.
This study has several limitations. First, as a retrospective study, the time points at which accurate viral detections were made were impossible to be determined. Second, since CMV positive cases were present at ICU admission, only active infection incidences and not reactivation incidences could be assessed. Third, due to the lack of CMV detection in the airways, we could not establish the clinical signi cance of CMV on oxygenation among ARDS patients. Therefore, prospective, multicenter studies are needed in future. Further analysis of the effect of CMV in the airways of ARDS patients is needed, and evaluation of CMV reactivation needs to be extended to the entire hospitalization period and not limited to the ICU stay period.

Conclusion
Active CMV infection is a common phenomenon in ARDS patients. Monocytes, hemoglobin, blood transfusion, and septic shock are risk factors for active CMV infection, which has a negative effect on oxygenation. Moreover, active CMV infection is associated with several adverse clinical outcomes. Prospective studies should aim at evaluating the impact of prophylactic antiviral therapy on the prognosis of ARDS patients.

Consent for publication
Not applicable.

Competing interests
None of the authors has any con ict of interest to report.

Statement
All methods were carried out in accordance with relevant guidelines and regulations in the manuscript.     Figure 1 Flowchart for patient enrollment. Oxygenation levels for the study participants *p< 0.05

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. 2OnlinesupplementResults.doc