A group of Brazilian researchers are among the creators of innovative technology that is modernizing one of the most widely used tools in intensive care units (ICUs) around the world. This tool updates a scale that has helped doctors since 1996 assess the severity of critically ill patients admitted to hospitals.
System for intensive care units
The updated system, titled Sequential Organ Failure Assessment-2 (SOFA-2), was published in a scientific journal. The development was carried out by an international consortium of 60 specialists in intensive medicine. The Brazilian researchers who participated are from the DUR Institute for Research and Education (IDOR).
Dr. Jorge Salouh was a member of the study coordination committee. “The participation of Brazil was very important. We had seven Brazilian specialists in this group, so the representation was great, and some of them played important prominent roles. Among them, I, who was the leader of the dataset and validation, and Professor Otavio Ranzani, who is the first author of the main development and validation study, and who is also Brazilian.”
Salouh explains that the system currently used in intensive care units has become outdated as it was implemented three decades ago. “When it was developed, there were no methods common today in intensive care, such as continuous hemodialysis, the use of ECMO, that is, extracorporeal oxygenation, which is widely used in more serious cases, as we saw during the pandemic, for example, and even other organ support devices. All this is included and included in the updated version,” explains the specialist.
“Changes in organ function have an impact on the outcome, on patient mortality, because as intensive care has changed so much over these 30 years, the things or factors, whether it’s examinations or changes in organ function that had a big impact on mortality to begin with, today, have a much less impact,” the researcher adds.
Dr. and researcher Jorge Salouh
Global database
An important point in the update is that the current study took into account that healthcare systems and patients are different around the world. Therefore, he combined a database of patients of different nationalities, with information from daily life in real intensive care units.
“Patient databases were used from both high- and middle-income and lower-middle-income countries, including countries in Asia, Latin America and, obviously, databases from Europe, North America, Australia and New Zealand,” explains George Saleh.
Updates
- The model in use actually evaluates six systems in the human body: brain, liver, kidney, coagulation, respiration and heart, assigning a score from 0 to 4 to each. The sum of these values indicates the degree of organic dysfunction to determine the patient’s severity.
- Now, SOFA-2 maintains the six-organ system assessment, but updates the scoring criteria using modern critical care medicine resources.
- Respiratory: Upgrading includes the use of non-invasive ventilation and ECMO (extracorporeal oxygen support).
- Cardiovascular system: Takes new medications and circulatory support devices.
- Kidney: now includes the use of dialysis, including chronic hemodialysis.
- Brain: Provides assessment of delirium, an important sign of neurological dysfunction in intensive care units.
- The result then becomes more accurate and applicable whether in highly complex hospitals or in units with limited resources.
- The main function of SOFA-2 is to provide a common and updated language for monitoring the development of critically ill patients. This makes it easy to compare results between teams, organizations and countries.
artificial intelligence
System modernization for intensive care units is made possible thanks to the use of artificial intelligence and big data. In this way, information was collected from more than 3.3 million hospital admissions in 1,300 intensive care units in nine countries between 2014 and 2023.
According to the creator, the implementation process is simple, as the collaborative research initiative is not commercial in nature. The tool is free and the full description is in the original articles of the publication.
The aim of the study was to provide an accessible tool that can be used without the need for structures with high processing power. “What we wanted to do was something cutting-edge, from the point of view of contemporary international modality and data provenance, and at the same time, that could be applied in any reality, whether high-resource ICUs or low-resource ICUs, anywhere in the world,” concludes Salouh.