TY - GEN
T1 - Cytomegalovirus reactivation in critically ill patients with acute respiratory distress syndrome
AU - Ong, David S Y
AU - Spitoni, Cristian
AU - Klein Klouwenberg, Peter M C
AU - Verduyn Lunel, Frans M
AU - Schultz, Marcus J
AU - Bonten, Marc J M
AU - Cremer, Olaf L
PY - 2014/9/30
Y1 - 2014/9/30
N2 - INTRODUCTION. Cytomegalovirus (CMV) reactivation frequently occurs in intensive care unit (ICU) patients, even in those without known prior immune deficiency. Patients with acute respiratory distress syndrome (ARDS) are thought to be at particular risk. CMV reactivation is associated with a prolonged stay in the ICU and an increased mortality. However, it remains uncertain whether these associations are causal, because previous studies have not adequately adjusted for all possible sources of bias.
OBJECTIVES. To estimate the attributable risk of CMV reactivation on the duration of mechanical ventilation and mortality in immune competent critically ill patients with ARDS who were latent carriers of the virus and remained mechanically ventilated beyond day four of ICU admission.
METHODS. We prospectively included consecutive patients with ARDS who were admitted to the ICUs of two tertiary care hospitals in the Netherlands and who were mechanically ventilated for at least 4 days. Immunocompromised Patients with known immunocompromise and those receiving antiviral treatment prior to ICU admission were excluded. CMV serostatus was determined by an enzyme immuno assay. Subsequently, in seropositive patients only, a real-time Taqman CMV-DNA polymerase chain reaction was used to determine the viral load in plasma on a weekly basis. CMV reactivation was defined as a load > 100 IU/mL.
We used competing risk and time-varying survival analyses, fitting CMV reactivation status as a time-dependent variable, while adjusting for baseline markers of disease severity.
RESULTS. Between January 2011 and March 2013, 306 ARDS patients were included, of whom 209 were CMV seropositive. CMV reactivation occurred in 53 (26%) of these cases. Patients in whom reactivation occurred, had a longer duration of mechanical ventilation and higher mortality than non-reactivated patients: 15 (IQR 10-27) versus 8 (IQR 6-11) (p=< 0.001) and 30% versus 17% (p=0.03), respectively.
After adjustment for APACHE IV score, presence of sepsis, use of high dose corticosteroid therapy, and blood products transfusion, patients with CMV reactivation had lower successful weaning rates (adjusted cause-specific hazard ratio (CSHR) 0.48 95% C.I. 0.32-0.73) but not increased fatality rates (adjusted CSHR 0.74 95% C.I. 0.38-1.45). However, when taking competing risks into account, CMV reactivation was associated with increased mortality (subdistribution hazard ratio 2.2 95% C.I. 1.0-4.8). The increased length of mechanical ventilation caused by CMV reactivation was 4 (IQR 1-6) days.
CONCLUSIONS. CMV reactivation increases the risk of dying in the ICU by prolonging the need for mechanical ventilation rather than by a direct effect on the daily risk of death.
AB - INTRODUCTION. Cytomegalovirus (CMV) reactivation frequently occurs in intensive care unit (ICU) patients, even in those without known prior immune deficiency. Patients with acute respiratory distress syndrome (ARDS) are thought to be at particular risk. CMV reactivation is associated with a prolonged stay in the ICU and an increased mortality. However, it remains uncertain whether these associations are causal, because previous studies have not adequately adjusted for all possible sources of bias.
OBJECTIVES. To estimate the attributable risk of CMV reactivation on the duration of mechanical ventilation and mortality in immune competent critically ill patients with ARDS who were latent carriers of the virus and remained mechanically ventilated beyond day four of ICU admission.
METHODS. We prospectively included consecutive patients with ARDS who were admitted to the ICUs of two tertiary care hospitals in the Netherlands and who were mechanically ventilated for at least 4 days. Immunocompromised Patients with known immunocompromise and those receiving antiviral treatment prior to ICU admission were excluded. CMV serostatus was determined by an enzyme immuno assay. Subsequently, in seropositive patients only, a real-time Taqman CMV-DNA polymerase chain reaction was used to determine the viral load in plasma on a weekly basis. CMV reactivation was defined as a load > 100 IU/mL.
We used competing risk and time-varying survival analyses, fitting CMV reactivation status as a time-dependent variable, while adjusting for baseline markers of disease severity.
RESULTS. Between January 2011 and March 2013, 306 ARDS patients were included, of whom 209 were CMV seropositive. CMV reactivation occurred in 53 (26%) of these cases. Patients in whom reactivation occurred, had a longer duration of mechanical ventilation and higher mortality than non-reactivated patients: 15 (IQR 10-27) versus 8 (IQR 6-11) (p=< 0.001) and 30% versus 17% (p=0.03), respectively.
After adjustment for APACHE IV score, presence of sepsis, use of high dose corticosteroid therapy, and blood products transfusion, patients with CMV reactivation had lower successful weaning rates (adjusted cause-specific hazard ratio (CSHR) 0.48 95% C.I. 0.32-0.73) but not increased fatality rates (adjusted CSHR 0.74 95% C.I. 0.38-1.45). However, when taking competing risks into account, CMV reactivation was associated with increased mortality (subdistribution hazard ratio 2.2 95% C.I. 1.0-4.8). The increased length of mechanical ventilation caused by CMV reactivation was 4 (IQR 1-6) days.
CONCLUSIONS. CMV reactivation increases the risk of dying in the ICU by prolonging the need for mechanical ventilation rather than by a direct effect on the daily risk of death.
M3 - Conference contribution
BT - LIVES 2014 (ESICM annual meeting)
ER -