He viral load in positive BAL fluids. A CMV and HSV negative specimen was used as a negative control.G. DefinitionsThe CMV infection group was defined as patients suspected of having pneumonia with positive CMV DNA detection in BAL fluid and/or positive antigenemia and/or the presence of IgM for CMV. We did not differentiate between endogenous reactivation or exogenous infection as the cause of the active infection. The HSV infection group was defined as patients suspected ofhaving pneumonia associated with positive HSV DNA detection in BAL fluid, or the presence of IgM for HSV in patients without CMV identification by antigenemia, real-time PCR or specific IgM. 1516647 one third of all patients were ventilated for ARDS. Ten percent of all patients were previously immunosuppressed (Table 1). 2. Virological status. Twenty-six patients (28 ) were included in the HSV group, 22 (24 ) in the CMV group, andI. Statistical AnalysisData are expressed as the median with an interquartile range (IQR) or as number of events (percentage). Continuous variables were compared using a Kruskall-Wallis one-way analysis of variance on ranks, with a pairwise multiple comparisons procedure using Dunn’s method. The chi-squared test was used to compare categorical variables. All reported P values are two-sided. For all Table 4. Outcomes.All (n = 93) Mortality at day 60, n ( ) ICU Mortality, n ( ) 32 (34) 32 (34)HSV infection group (n = 26) 11 (42) 11 (42) 14.5 [10?6] 5.5 [0?3] 36.5 [0?5] 18 [11?0] 10 (39) 6 (23) 3 (12) 5 (19) 2 (8)CMV infection group (n = 22) 12 (55)* 12 (55)* 19.5 [13?4]{ 0 [0?]{ 0 [0?5]{ 25.5 [15?3]{ 17 (77)*{ 11 (50)* 10 (46)* 6 (27) 4 (18)Control group p (n = 45) 9 (20) 9 (20) 10 [3?5] 18.He viral load in positive BAL fluids. A CMV and HSV negative specimen was used as a negative control.G. DefinitionsThe CMV infection group was defined as patients suspected of having pneumonia with positive CMV DNA detection in BAL fluid and/or positive antigenemia and/or the presence of IgM for CMV. We did not differentiate between endogenous reactivation or exogenous infection as the cause of the active infection. The HSV infection group was defined as patients suspected ofhaving pneumonia associated with positive HSV DNA detection in BAL fluid, or the presence of IgM for HSV in patients without CMV identification by antigenemia, real-time PCR or specific IgM. 12926553 The “control group” was defined by the absence of a CMV or HSV infection in patients suspected of having pneumonia. A bacterial coinfection required the presence of at least one bacteria at a concentration exceeding 104 cfu/mL in the BAL fluid.Impact of CMV and HSV on Ventilated PatientsTable 3. Virological results.All (n = 93) HSV status, n( )HSV infection group (n = 26)CMV infection group (n = 22)Control group (n = 45)IgM HSV IgG HSV BAL RT-PCRCMV status, n( )3 (3) 81 (87) 31 (33)2 (8) 24 (92) 25 (96)1 (5) 19 (86) 6 (27)0 (0) 38 (84) 0 (0)IgM CMV IgG CMV BAL RT-PCR Antigenemia8 (9) 72 (77) 16 (17) 10 (11)0 (0) 18 (69) 0 (0) 0 (0)8 (36) 20 (91) 16 (73) 10 (46)0 (0) 34 (76) 0 (0) 0 (0)BAL, bronchoalveolar lavage; RT-PCR, real time polymerase chain reaction. doi:10.1371/journal.pone.0051340.tH. Study OutcomesThe primary outcome was mortality for both viruses, evaluated at day 60. 2. Secondary outcomes. Secondary outcomes were the ICU mortality, the day-28 mortality, the number of days with mechanical ventilation, the number of ventilator-free days (days alive and with a successful weaning from mechanical ventilation for at least 48 hrs) between day 1 and day 28, and between day 1 and day 60 [32].1. Primary outcome.statistical tests used, a p value of ,0.05 was considered significant. A Bonferroni method was applied for multiple comparisons when necessary (leading to a significant p value of ,0.016 when applied). Multiple logistic regressions were used to adjust the day 60 mortality regarding 2 pre-defined variables (SAPS II score on admission and SOFA score the day of BAL).Results A. Patients Characteristics1. All patients. During the study period, ninety-three consecutive patients met the inclusion criteria and were prospectively included in the study. Less than 1516647 one third of all patients were ventilated for ARDS. Ten percent of all patients were previously immunosuppressed (Table 1). 2. Virological status. Twenty-six patients (28 ) were included in the HSV group, 22 (24 ) in the CMV group, andI. Statistical AnalysisData are expressed as the median with an interquartile range (IQR) or as number of events (percentage). Continuous variables were compared using a Kruskall-Wallis one-way analysis of variance on ranks, with a pairwise multiple comparisons procedure using Dunn’s method. The chi-squared test was used to compare categorical variables. All reported P values are two-sided. For all Table 4. Outcomes.All (n = 93) Mortality at day 60, n ( ) ICU Mortality, n ( ) 32 (34) 32 (34)HSV infection group (n = 26) 11 (42) 11 (42) 14.5 [10?6] 5.5 [0?3] 36.5 [0?5] 18 [11?0] 10 (39) 6 (23) 3 (12) 5 (19) 2 (8)CMV infection group (n = 22) 12 (55)* 12 (55)* 19.5 [13?4]{ 0 [0?]{ 0 [0?5]{ 25.5 [15?3]{ 17 (77)*{ 11 (50)* 10 (46)* 6 (27) 4 (18)Control group p (n = 45) 9 (20) 9 (20) 10 [3?5] 18.