Related Experiment Video
Updated: Apr 20, 2026
A Model of Acute Lung Injury Following Visceral Ischemia-Reperfusion by Supra-Coeliac Aortic Cross Clamping in Rats
Published on: August 15, 2025
A ternary model of decompression sickness in rats
Peter Buzzacott1, Kate Lambrechts2, Aleksandra Mazur2
1Université de Bretagne Occidentale, Laboratoire Optimisation des Régulations Physiologiques (ORPhy), UFR Sciences et Techniques, 6 avenue Le Gorgeu, CS 93837, 29200 Brest Cedex 3, France; School of Sports Science, Exercise and Health, the University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia.
This study developed a new ternary model to predict the probability of decompression sickness (DCS) in rats, distinguishing between no-DCS, survivable-DCS, and death. The model shows reliability within specific parameter ranges for dive profiles and rat characteristics.
Area of Science:
- Physiology
- Toxicology
- Biostatistics
Background:
- Decompression sickness (DCS) in rats is typically modeled as a binary outcome (present or absent).
- A need exists for a more nuanced model that predicts the probability of different DCS outcomes.
- This study addresses the limitation of binary models by proposing a ternary prediction system.
Purpose of the Study:
- To develop and validate a ternary model for predicting DCS probability in rats.
- The model aims to differentiate between no-DCS, survivable-DCS, and fatal DCS outcomes.
- Predictive accuracy is based on compression/decompression profiles and rat physiological characteristics.
Main Methods:
- A comprehensive literature search identified relevant dive profiles and DCS outcomes (no-DCS, survivable-DCS, death).
- Data from 1602 rats across 15 studies, encompassing 22 dive profiles, were compiled, adhering to strict inclusion criteria.
- Ordinal logistic regression was employed for model optimization and validation using two independent datasets.
Main Results:
- The model demonstrated 75% accuracy in predicting DCS status within the calibration dataset (15/20 cases correctly classified).
- A high degree of confidence was observed, with 92.5% of predictions falling within 95% confidence intervals.
- Model predictability decreased significantly when applied to data outside the calibration parameter range, particularly concerning rat weight.
Conclusions:
- The developed ternary model is reliable for predicting DCS status in rats when dive profiles and physiological parameters fall within the model's optimization range.
- Further improvement in the model's applicability can be achieved by incorporating data with a broader range of parameters.
- This study advances DCS research by providing a more refined predictive tool for animal models.
