Purpose
To investigate whether adrenal gland radiodensities alone or set in relation to either the inferior vena cava (IVC) or the spleen can predict hospital mortality in intensive care unit patients.
Methods
One hundred thirty-three intensive care patients (90 males, age: 66.3 ± 14.5 years) with an acute clinical deterioration were included in this retrospective analysis. CT attenuation (Hounsfield units) of adrenal glands, IVC, and spleen was evaluated by 2 radiologists separately. Adrenal-to-IVC and adrenal-to-spleen ratios were calculated. Receiver operating characteristic (ROC) analysis, combined with the Matthews correlation coefficient (MCC) as a classifier, was used to assess which parameter is the most suitable for short-term, intermediate-term, and overall mortality prediction. Interrater agreement was assessed using intraclass correlation coefficient (ICC).
Results
The highest discriminative power to distinguish between deceased and survivors was found for the adrenal gland-to-spleen ratio for the 72-h mortality. A threshold of > 1.4 predicted 72-h mortality with a sensitivity of 79.31% and a specificity of 98.08% (area und the curve (AUC) = 0.94; p < 0.0001; MCCs = 0.81). The positive likelihood ratio was 41; the positive predictive value was 92.20%. Adrenal gland-to-spleen ratio was also best suited to predict the 24-h and overall mortality. ICCs of HU measurements in adrenal gland, IVC, and spleen indicated a high interrater agreement (ICC 0.95–0.99).
Conclusions
To conclude, the adrenal-to-spleen ratio in CT in portal venous phase may serve as an imaged-based predictor for short, intermediate, and overall mortality and as reproducible prognostic marker for patient outcome.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.