Back

Acute Respiratory Distress Syndrome in Adults: A Retrospective Analysis of Temperature Trends, Demographic Factors, and Clinical Outcomes from the eICU Collaborative Research Database

Al Mahrizi, A. D.; Ezenwanne, C.; Mossolem, F.; Valladares, C.; Gill, H.; Boyle, M.

2025-01-17 respiratory medicine
10.1101/2025.01.16.25320573 medRxiv
Show abstract

BackgroundAcute Respiratory Distress Syndrome is characterized by the sudden onset of hypoxemia, reduced lung compliance, and bilateral pulmonary infiltrates. We investigated patient demographics and clinical predictors of outcomes in adult patients with ARDS. MethodsWe conducted a retrospective multi-center intensive care unit database study using the eICU Collaborative Research Database (eICU-CRD). The Analysis included 304 patients with ARDS who met all inclusion criteria, including key demographic variables. We Assessed associations between these variables and outcomes, including mortality and length of stay (LoS). ResultsFor each year of increased age, odds of mortality decreased by 4.7% (OR = 0.953, p < 0.001). Higher temperatures were associated with reduced mortality (OR=0.470, p=0.003). Gender (p = 0.596) and BMI (p = 0.964) were not significant predictors of mortality. A multiple linear regression model for predicting ICU length of stay was statistically significant at explaining the variance compared to the null model. High temperatures ({beta} = 0.981, p = 0.015) and low temperatures ({beta} = -1.038, p = 0.006) were significant predictors of LoS, while gender and BMI were not. There was significant association between ethnicity and hospital discharge status ({chi}2(5) = 13.123, p = 0.022), which suggests disparities in outcomes across ethnic groups. ConclusionOur study suggests that increasing age and higher temperatures may serve as protective factors, contrary to the popular belief of age-increasing fragility. Analysis also found that gender and BMI do not affect patient LoS nor mortality. These findings suggest clinicians should consider a patients age and core body temperature when assessing risk of mortality and prognosis, placing less emphasis on BMI and gender.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.