JAMIA: E-system IDs lung injury in trauma patients
An automated electronic system that screens intubated ICU trauma patients for acute lung injury (ALI) based on chest x-ray reports and arterial blood gas results is sufficiently accurate to identify many early cases of ALI, according to a single-center validation study in the July/August issue of the Journal of the American Medical Informatics Association.
Satjeet S. Khalsa, MS, from the department of radiology, medical informatics groups at the University of Pennsylvania School of Medicine in Philadelphia, and colleagues designed an automated electronic system that incorporates data from multiple hospital information systems to screen for ALI in mechanically ventilated patients. The authors evaluated the accuracy of the system in diagnosing ALI in a cohort of patients with major trauma, but excluding patients with congestive heart failure.
Researchers screened prospectively arterial blood gas data and chest x-ray reports for a cohort of ICU patients with major trauma for ALI requiring intubation by the system. The system was compared with a reference standard established through consensus of two blinded physician reviewers who independently screened the same population for ALI using all available arterial blood gas data and chest x-ray images. They evaluated the system's performance by measuring the sensitivity and overall accuracy, and the concordance with respect to the date of ALI identification (vs. the reference standard).
Khalsa and colleagues evaluated 199 trauma patients admitted to their level 1 trauma center with an initial injury severity score greater than 16 for development of ALI in the first five days in an ICU after trauma.
The system demonstrated 87 percent sensitivity (95 percent confidence interval) and 89 percent specificity (95 percent confidence interval), according to the authors. Two reviewers identified ALI before or within the 24-hour period during which ALI was identified in 87 percent of cases.
Based on their findings, the researchers hypothesized that, by linking the computerized ALI recognition system with an automated prompt to clinicians, the system may increase the use of lung protective ventilation and thereby improve the outcome of patients with ALI.
“Furthermore, our system can prompt clinicians to consider using lung protective ventilation in all cases of ALI that it identifies, but especially in early cases of ALI, which may not have otherwise been recognized,” they wrote.
Satjeet S. Khalsa, MS, from the department of radiology, medical informatics groups at the University of Pennsylvania School of Medicine in Philadelphia, and colleagues designed an automated electronic system that incorporates data from multiple hospital information systems to screen for ALI in mechanically ventilated patients. The authors evaluated the accuracy of the system in diagnosing ALI in a cohort of patients with major trauma, but excluding patients with congestive heart failure.
Researchers screened prospectively arterial blood gas data and chest x-ray reports for a cohort of ICU patients with major trauma for ALI requiring intubation by the system. The system was compared with a reference standard established through consensus of two blinded physician reviewers who independently screened the same population for ALI using all available arterial blood gas data and chest x-ray images. They evaluated the system's performance by measuring the sensitivity and overall accuracy, and the concordance with respect to the date of ALI identification (vs. the reference standard).
Khalsa and colleagues evaluated 199 trauma patients admitted to their level 1 trauma center with an initial injury severity score greater than 16 for development of ALI in the first five days in an ICU after trauma.
The system demonstrated 87 percent sensitivity (95 percent confidence interval) and 89 percent specificity (95 percent confidence interval), according to the authors. Two reviewers identified ALI before or within the 24-hour period during which ALI was identified in 87 percent of cases.
Based on their findings, the researchers hypothesized that, by linking the computerized ALI recognition system with an automated prompt to clinicians, the system may increase the use of lung protective ventilation and thereby improve the outcome of patients with ALI.
“Furthermore, our system can prompt clinicians to consider using lung protective ventilation in all cases of ALI that it identifies, but especially in early cases of ALI, which may not have otherwise been recognized,” they wrote.