124 Chapter 6 Variables Baseline Season (n= 13,440) ILI Season (n= 12,994) P Value Duration of mechanical ventilation, hours 5.6 (3.5-9.5) 5.6 (3.6-9.5) .30 P/F ratio 1 day after ICU admission 271 (214-339) 269 (210-332) .03 P/F ratio 3 days after ICU admission 260 (180-333) 273 (180-348) .40 P/F ratio 7 days after ICU admission 217 (150-284) 236 (163-304) .14 Length of stay ICU, days 1.9 (1.1-2.8) 1.9 (1.1-2.9) .43 Length of stay hospital, days 7.6 (6-11) 8.0 (6-11.1) <.0001 In-hospital mortality 165 (1.22) 224 (1.72) .001 ICU mortality 116 (0.86) 125 (0.96) .43 Ward mortality 49 (0.36) 99 (0.76) <.0001 Table 2: Unadjusted analysis showing differences in outcomes between baseline season and Influenza-like Illness season. Values are median (interquartile range) or n (%). ICU, intensive care unit; ILI, influenza-like illness; P/F, to fraction of inspired oxygen. There is a distinct yearly pattern in the number of people with ILI, with differences in the duration of ILI season between the years (Figure 1). Visually, there is no relationship between the ILI prevalence (per 100,000 persons) and the duration of mechanical ventilation in hours. Primary Outcome Measure A Cox regression analysis was performed (Figure 2). For 2015-2016, the hazard ratio (HR) of stopping IMV was 1.22 (95% CI, 1.12-1.33), indicating that independent of other variables, time until IMV interruption was slightly shorter in the ILI season compared with baseline season. Similar results were found for 2014-2015 (HR 1.15; 95% CI, 1.04-1.26) and 2017-2018 (HR 1.25; 95% CI, 1.16-1.36). The HR of stopping IMV did not significantly differ in 2013-2014 (HR 1.14; 95% CI, 0.96-1.34), 2016-2017 (HR 0.99; 95% CI, 0.92-1.07), or 2018-2019 (HR 1.01; 95% CI, 0.931.10). Secondary Outcome Measure In a multivariable regression analysis, undergoing cardiac surgery during ILI season increased the odds ratio of in-hospital mortality (odds ratio 1.67; 95% CI, 1.14-2.46; P = .009). The interaction term of ILI season and year of surgery was not significant in a univariable regression analysis and was therefore not included in the multivariable regression analysis, but the individual variables (year of surgery and ILI season) were included.
RkJQdWJsaXNoZXIy MTk4NDMw