New Prediction Model Assesses the Risk for Acute Respiratory Distress Syndrome in Burn Patients

For the first time, investigators have devised a model to predict burn patients who are likely to develop life-threatening acute respiratory distress syndrome (ARDS).

For their study, investigators from Loyola Medicine and Loyola University Chicago Stritch School of Medicine set out to derive and validate a prediction model for the development of ARDS in burn-injured patients.

ARDS is a form of respiratory failure in patients who are receiving mechanical ventilation. The disease usually occurs in patients who already are critically ill from predisposing conditions such as sepsis, pneumonia, and burns.

Previously, patients with large burns had high mortality rates, so few survived to develop ARDS. Due to improvements in treatment, however, more patients are surviving burns, and now, over 25% of patients with large burns or major inhalation injuries who survive the first 24 hours go on to develop ARDS.

In the new prospective study, published in the journal Annals of Surgery, investigators screened 113 adult patients who had burns over at least 10% of their bodies and/or were suspected of having inhalation injuries. About one-third, (33.6%) developed ARDS a median of 2.2 days after their injuries.

Measuring plasma biomarkers that have been previously identified as playing a role in endothelial injury, epithelial injury, or inflammatory changes in ARDS, the investigators set out to develop and validate a prediction model for ARDS in burn-injured patients.

To do this, the investigators examined clinical characteristics including burn and inhalation injury, alcohol/tobacco use, and existing health issues such as diabetes, congestive heart failure, heart disease, and 5 protein biomarkers found in plasma.

The specific plasma protein biomarkers—von Willebrand Factor (vWF), soluble Intercellular Adhesion Molecule 1 (ICAM-1), surfactant protein D (SP-D), soluble tumor necrosis factor receptor 1 (sTNFR1), interleukin (IL)-8 and interleukin (IL)-6—have been examined in multiple ARDS studies with promising results for phenotyping and predictive modeling, according to study authors.

The investigators postulated that the combination of clinical risk factors and biomarkers compared with either alone would provide a superior prediction model to identify burn patients who are at risk for ARDS development.

The investigators found that among the biomarkers, log-transformed levels of the A2 domain of the von Willebrand factor in the first 24 hours had the greatest odds ratio for the development of ARDS based on an increase of 1 standard deviation (OR 7.72; 95% CI: 1.64-36.28 P= 0.03).

A “Burn patients with ARDS had lower mean 28-day ventilator free days (8.9 days vs 23.3 days, P <0.001) and a greater proportion of in-hospital deaths (42% vs 4.0%, P < 0.001) that non-ARDS counterparts,” the authors write.

Investigators also report that diabetes was the only clinical characteristic associated with ARDS development (OR 6.34, 95% CI 1.84-21.89, P=0.004).

Among the different models that were analyzed, a model consisting of inhalation injury, the von Willebrand factor biomarker, and the percent of body burned did the best job of predicting which patients were more likely to develop ARDS.

The 3-pronged model "could be used to better identify at-risk patients for both the study and prevention of ARDS in patients with burn injury," Majid Afshar, MD, MSCR, assistant professor, Loyola Medicine and colleagues, said in a recent statement.

If the model is validated by other studies, it could guide clinical trials designed to prevent ARDS and identify burn patients who are at risk for ARDS.