Abstract
NTRODUCTION. Lactate (LCT) and central venous oxygen saturation (ScvO2) may reflect adequacy of tissue perfusion and their normalisation are used as resuscitation endpoints.
OBJECTIVES. We hypothesise that LCT and ScvO2 trends (reflecting effectiveness of resuscitation) following ICU admission independently predict changes in the severity of organ dysfunction (as defined by SOFA) since adequate resuscitative efforts should theoretically improve organ function.
METHODS. Population: We did a retrospective single-centre cohort study using data from the comprehensive electronic clinical information system (CareVue™). All admissions between 04/2004-07/2009 who had LCT and ScvO2 measured within the first 24 hours after admission were included. Baseline, daily follow-up (physiological, SOFA), ICU length of stay (LOS) and mortality data were recorded.
ANALYSIS. Descriptive analysis used standard statistical techniques.We performed longitudinal linear regression modelling treating day 1-4 changes in LCT and ScvO2 as the main predictor variables, and the longitudinal day 2-5 change in total (t-) SOFA score as dependent variable. Other potential confounders/effect modifiers were also included, but variables included in the SOFA calculations were excluded. We did include renal replacement therapy (RRT) as it may confound the estimates for LCT. To account for the correlation due to repeated measurement, we used a mixed effects model with random intercept for each cluster (patient). Slopes (sl) are presented as coefficients (95% CI) and intercepts (int) in base units.
RESULTS. Demographics: 1,454 patients were included. Mean (SD) age, APACHE 2 and ScvO2 were 63 (16.1) years, 19.2 (6.3) and 68.6 (10.3), respectively. Median (IQR) LCT and t-SOFA was 3 (2.9) and 7 (5), respectively. 75.9% of patients had at least 1 organ failure at admission. Median (IQR) ICULOS was 7 (10) days and ICU mortality was 22.6%.On average LCT decreased (sl: -2.5; -2.2 to -2.8) while ScvO2 increased (sl: 0.59; 0.33 to 0.86) from day 1-4, both suggesting effective resuscitation. Organ function, reflected by t-SOFA trends, also improved from day 2-5 (sl: -1.27; -1.02 to -1.52).In multivariable analysis LCT and ScvO2 trends were both associated with t-SOFA trends: LCT - sl: 0.12 (0.08 to 0.15); ScvO2 - sl: -0.02 (-0.03 to -0.01). Interestingly RRT was also strongly associated with t-SOFA trends (sl: 2.16; 1.87- 2.45).
CONCLUSIONS. Improvements in LCT and ScvO2 are associated with improvements in aggregate organ function. This finding suggests that effective early and on-going resuscitation leads to improvements in subsequent organ function.
OBJECTIVES. We hypothesise that LCT and ScvO2 trends (reflecting effectiveness of resuscitation) following ICU admission independently predict changes in the severity of organ dysfunction (as defined by SOFA) since adequate resuscitative efforts should theoretically improve organ function.
METHODS. Population: We did a retrospective single-centre cohort study using data from the comprehensive electronic clinical information system (CareVue™). All admissions between 04/2004-07/2009 who had LCT and ScvO2 measured within the first 24 hours after admission were included. Baseline, daily follow-up (physiological, SOFA), ICU length of stay (LOS) and mortality data were recorded.
ANALYSIS. Descriptive analysis used standard statistical techniques.We performed longitudinal linear regression modelling treating day 1-4 changes in LCT and ScvO2 as the main predictor variables, and the longitudinal day 2-5 change in total (t-) SOFA score as dependent variable. Other potential confounders/effect modifiers were also included, but variables included in the SOFA calculations were excluded. We did include renal replacement therapy (RRT) as it may confound the estimates for LCT. To account for the correlation due to repeated measurement, we used a mixed effects model with random intercept for each cluster (patient). Slopes (sl) are presented as coefficients (95% CI) and intercepts (int) in base units.
RESULTS. Demographics: 1,454 patients were included. Mean (SD) age, APACHE 2 and ScvO2 were 63 (16.1) years, 19.2 (6.3) and 68.6 (10.3), respectively. Median (IQR) LCT and t-SOFA was 3 (2.9) and 7 (5), respectively. 75.9% of patients had at least 1 organ failure at admission. Median (IQR) ICULOS was 7 (10) days and ICU mortality was 22.6%.On average LCT decreased (sl: -2.5; -2.2 to -2.8) while ScvO2 increased (sl: 0.59; 0.33 to 0.86) from day 1-4, both suggesting effective resuscitation. Organ function, reflected by t-SOFA trends, also improved from day 2-5 (sl: -1.27; -1.02 to -1.52).In multivariable analysis LCT and ScvO2 trends were both associated with t-SOFA trends: LCT - sl: 0.12 (0.08 to 0.15); ScvO2 - sl: -0.02 (-0.03 to -0.01). Interestingly RRT was also strongly associated with t-SOFA trends (sl: 2.16; 1.87- 2.45).
CONCLUSIONS. Improvements in LCT and ScvO2 are associated with improvements in aggregate organ function. This finding suggests that effective early and on-going resuscitation leads to improvements in subsequent organ function.
| Original language | English |
|---|---|
| Pages (from-to) | S154-S154 |
| Number of pages | 1 |
| Journal | Intensive Care Medicine |
| Volume | 37 |
| Publication status | Published - Sept 2011 |