Background. Human herpes viruses establish lifelong latency after primary infection and may reactivate in immunosuppressed patients causing significant morbidity and mortality. In immunocompetent patients, although reactivation may occur disease development is deterred by the competent host immune response. Recent studies indicate that approximately one third of CMV seropositive immunocompetent ICU patients present with CMV reactivation associated with poor outcome, potentially secondary to the stress incurred. CMV reactivation among immunocompetent critically ill children has not been assessed.

Study Hypothesis: Identifiable risk factors associated with CMV reactivation exist and may be used for future assessment of antiviral prophylaxis administration.

Aim: Primary aim is to identify risk factors associated with CMV reactivation and poor outcome in immunocompetent children and adults under severe stress. Whether CMV reactivation occurs in critically ill children and its clinical implications remains to be determined. Secondary aim is to study the role of cellular signaling pathways of inflammation and specific adaptive immunity during this process.

Work packages: A multicenter observational prospective study will be conducted among CMV seropositive pediatric and adult ICU patients. Patient clinical progress, laboratory findings, management, and complications will be recorded during the 28 days following ICU admission. Salivary free cortisol levels, plasma catecholamines, and serum cytokines levels will be measured to assess stress. CMV reactivation will be evaluated weekly by detecting CMV-DNA in peripheral blood and bronchial wash samples with real-time PCR. In a patient subsample, the nuclear factor κB and intracellular GC receptor will be measured in peripheral blood monocytes to study cellular signaling pathways of inflammation. The adaptive immune response to CMV infection following in vitro viral polypeptide stimulation will be prospectively examined in a subset of patients.

Expected Results: The study will provide original data on critically ill children. Further knowledge regarding risk factors associated with CMV reactivation and poor outcome will be accumulated. Novel information regarding the role of cellular inflammation and specific adaptive immune responses during CMV reactivation will be gathered.

This is a multicenter prospective observational study of CMV seropositive patients (children and adults) admitted in ICU. Two pediatric and four adult ICU units in tertiary teaching hospitals will be recruiting for 48 and 36 months respectively.

Research plan Upon ICU admission, patients (group A and B) will be screened for inclusion/exclusion criteria and after written informed consent is obtained the patient will be enrolled. Clinical severity will be estimated using standardized score systems (PRISM III and APACHI II respectively). All pertinent demographic, medical history, clinical presentation and medical interventions will be entered in a database. Prior to initiating any supportive treatment (inotropes or corticosteroids) blood and saliva will be obtained for the determination of biological markers of stress. Serum and whole blood will be collected for the determination of CMV-IgG antibodies and CMV-DNA respectively. CMV seronegative patients will be excluded from further evaluation. Seropositive patients will be followed prospectively for 28 days. Weekly follow up (7th, 14th, 21st and 28th day of hospitalization) will include: clinical and laboratory evaluation, documentation of therapeutic intervention and CMV reactivation assessment (CMV-DNA detection). Stress markers will be re-examined only on day 7. On day 28 (end of follow up) the clinical outcome will be recorded.

In a random subset of patients (20 children and 20 adults) the intracellular signaling pathway of inflammation and specific immune responses will be studied.

Laboratory Methods

A) Serology for CMV: Serum CMV-IgG antibodies will be measured by Enzyme Immunoassay Method (Elisa, Abbott Laboratories) in hospital serology lab.

B) Estimation of biological stress indicators

B.1. Measurement of free cortisol. Saliva samples will be collected three times a day (8.00 am, 2.00pm and 8.00 pm) to determine the circadian pattern cortisol secretion using special synthetic swabs (Salivetta-Salivette, Sarsted ®) and stored at -70oC. Cortisol levels will be determined with electrochemiluminescence immunoassay (Elecsys Cortisol reagent kit, ROCHE).

B.2. Plasma catecholamines. Catecholamine levels (noradrenaline, adrenaline and dopamine) will be measured before the initiation therapy with any inotropic agent. Samples will be taken from a venous catheter with the patient supine only in patients not receiving inotropes. Plasma will be stored at -85oC. The Liquid HPLC reverse phase (High Performance Liquid Chromatography - reverse phase) will be used.

B.3. Serum TNF-a will be measured by quantitative-sandwich-Enzyme Immunoassay method (Elecsys, ROCHE).

C) Detection and quantification of CMV-DNA from whole blood and bronchoalveolar lavage (BAL, in ventilated patients). If a patient is discharged before day 28, salivettes will be given at home to collect saliva for CMV-DNA determination (stored in home fridge) and returned on day 28. Samples will be stored at -70oC and all samples from each patient will be examined at the same time. DNA extraction and real-time PCR will be performed by commercial kits (Nanogen Advanced Diagnostics). A gene region encoding the Major Immediate Early Antigen (MIEA) of CMV will be detected. All PCR testing will be performed at the Cytology Laboratory, Attikon University Hospital, Athens Greece (Associate Prof P. Karakitsos).

D) Glucocorticoid receptor (GR) and nuclear factor κB (NFκB) measurements in whole cell and nuclear extracts from peripheral blood lymphocytes will be performed by immunoblotting at the Department of Biological Chemistry, Athens University (Ass. Prof P. Moutsatsou)8

E) Specific immune response against CMV. Specific immune responses of CD4+ and CD8+ T lymphocytes from peripheral blood after in-vitro stimulation with CMV polypeptides (including pp65 antigen) will be studied with intracellular staining (Intracellular staining, ICS). The proliferation of CD4+ (CFSE) and the cytotoxicity of CD8+ (perforin and granzyme B) will be assessed.

Statistical analysis For the data analysis adults and children will be examined separately. Seropositive patients without CMV reactivation will be used as the control group. A p-value (p) of ≤0.05 is the criterion for statistical significance. T-test and chi-square test will be used for quantitative and qualitative analysis, while the Mantel-Haenzel method will be used to calculate relative risk of different outcomes (mortality, intubation, etc) between the two groups. To identify potential risk factors associated with CMV reactivation and clinical outcome, logistic regression analysis using the statistical program SAS v.9 will be performed.

Sample calculation To calculate study sample size the statistical program EpiInfo version 6 was used. Based on recent data from adults, 30% may demonstrate CMV reactivation while in the ICU. The mortality rate in the pediatric and adult ICU has been estimated to reach approximately 8% and 63% respectively. Recent data suggest that CMV reactivation is associated with increased mortality (RR=1,93) and morbidity (RR=5,70). The sample size was calculated to meet the criteria of 99% confidence interval and 90% statistical power. We estimated that our CMV(+) population should include 165 children and 109 adults.

Work packages WP1 and WP2 involve clinical research; enrollment and clinical follow up of children and adults respectively. All data collection will be entered in an anonymous database which will include demographic data, past history, clinical and laboratory evaluation, clinical management and interventions upon admission and during follow up period. Adults and children will be evaluated separately. Once CMV results are provided (WP4), risk factors associated with CMV reactivation as well as adverse clinical outcomes, including both those directly (i.e. pneumonitis, febrile episodes, and hepatitis) and indirectly associated (i.e. extended ICU stay or mechanical ventilation, and increased rate of bacterial infections) will be examined.

WP3 involves the evaluation of stress from a clinical (WP3.1 and WP 3.2) and basic research (WP3.3) approach. Biochemical stress markers will be evaluated upon admission and on day 7 to examine duration of stress. Concomitant drug administration will be taken into account. Stress markers will be correlated with CMV reactivation and clinical outcome to examine whether they can be used as identifiable risk factors for future research. In WP3.3 cellular signaling pathways of inflammation in a subset of our cohort will be studied. Specifically the interaction of NFκΒ activation and glucocorticoid receptor sensitivity in patients' mononuclear cells will be evaluated and correlated with CMV reactivation and outcome.

WP4 involves the prospective detection and quantification of CMV-DNA in different biological samples. All data will be included in the main database and more importantly a DNA bank will be formed for possible detection of other herpesviruses in the future.

WP5 will evaluate specific immune responses to CMV in a subset of patients. These will be correlated with the underlying disease, reactivation and stress. Moreover, a bank of supernatants from those experiments will be formed to measure cytokines (Luminex) in the future.

Inclusion Criteria:

• previously healthy children 5-16 years old (group A) and adults (group B)
• no known immunosuppression (secondary to underlying disease or medications),
• residence near the ICU (ability to return for follow up on day 28 post admission)
• availability of patient guardian or first degree relative willing to provide written informed consent

Exclusion Criteria:

• imminent death
• expected ICU stay <48 hours
• intubation prior to admission (in a different center) for >48 hours

ICU patients