Personalized treatment for ARDS: Ready for prime time?
For people in the realm of acute respiratory distress syndrome research, the LIVE trial, conducted by Constantin and colleagues and published in The Lancet Respiratory Medicine, is intriguing. The concept of standardizing care based on lung morphology in patients with ARDS has been studied extensively during the past 10 years or so, but the key question here is whether there is a benefit to treating patients with focal vs. nonfocal disease differently. This premise is valid and is of great interest to many in the field.
Before we can assess the study results and their clinical applications though, we must understand the study design and its execution. We know that looking at chest X-rays to determine ARDS is a tricky proposition: Clinicians who have a vested interest in and specialize in this area do well, but for the usual bedside provider, using chest X-rays to make a determination for ARDS is likely hit-or-miss. Therefore, when the authors designed the study, they aimed to use CT scans to assess who had focal or nonfocal disease. This is the right way to do this in terms of deciding lung morphology, but we also have to realize that it is a time-consuming and resource-intensive process that adds an extra layer before we can move forward. This too became an issue when the study was initiated, as approximately three of every four patients did not get CT scans most patients were determined to have focal or nonfocal disease based solely on chest X-ray results. Consequently, 21% of patients were misclassified, which was a significant stumbling block when interpreting the results. At the heart of the study is a fascinating question, but there was a disconnect between the way that the authors wanted to do the study and the way that the study was done.
Problems and pitfalls
In terms of the findings, the primary outcomes were negative, potentially because of the reasons discussed, but the results are still worth noting. After removing patients who were misclassified, the data showed that patients may benefit from receiving tailored therapy based on their underlying phenotype of ARDS, and, in fact, patients who received tailored therapy fared better than patients who were misclassified. Unfortunately, the authors were not asking this question; they were trying to show that we can predict the type of ARDS in these patients and then tailor our therapy accordingly.
For me, the LIVE study raises an important question. Although there certainly are different ARDS phenotypes, can we base our care plan on these specific phenotypes given the tools at our disposal and the typical bedside provider’s difficulty in determining lung morphology based on chest X-rays alone? The risk for misclassification is high, and, as we see in the study, there is a consequence of that misclassification. For patients who were misclassified vs. those who were classified correctly, the risk for 90-day mortality was nearly threefold higher. While we can explore different treatment options for different ARDS phenotypes, the wrong treatment has a significant cost.
One other point to note is that the authors did not use only one intervention in the personalized treatment group. They used three interventions, with tidal volumes, positive end-expiratory pressure strategies and use of prone position ventilation or recruitment maneuver differing according to phenotype. When multiple interventions are applied, it becomes difficult to identify the relationship between treatment and outcomes. Is one intervention causing the effect? Is it the use of all three interventions? And, ultimately, there is the question of whether everyone classified as having focal or nonfocal disease needs all three interventions or whether just one or two would suffice. This is another potential pitfall that we need to bear in mind before we can apply these findings at the bedside.
Overall, the LIVE study, which is the first to truly evaluate precision medicine in ARDS, shows that a one-size-fits-all strategy for patients with ARDS is not the best treatment strategy. Can we improve by tailoring therapy according to phenotype? We can, as long as we are very mindful of the importance of correctly classifying the patient. However, in a real-world setting, are we ready to determine a patient’s phenotype with the tools that we have? If everyone in the study had undergone a CT scan, how many misclassifications could have been prevented? These data point us in the right direction, but I do not know if we are ready to say that we can safely screen these patients into phenotypes based on the way it was done in this study.
For me, when thinking about the hypothesis of personalizing therapy according to ARDS phenotype in a clinical setting, it makes perfect sense. Is it ready for prime time? Not yet. My two main concerns are the high risk for misclassification and the fact that bedside providers are more prone to inaccurately diagnosing these patients. So, should tailored treatment based on phenotype be employed in every ICU right now? I do not believe we can say that at this point in time.
My final thought is that this study was intriguing, and the authors were transparent in identifying where they missed the mark, but it is difficult to make assertions about outcomes because patients with focal and nonfocal disease are very different and we do not yet have the power determine whether personalized treatment does or does not make a difference in these subgroups.
For more information:
Abhijit Duggal, MD, director of critical care research in the medical intensive care unit and associate program director of critical care fellowship at the Cleveland Clinic, and assistant professor of medicine at the Lerner School of Medicine of Case Western Reserve University, can be reached at email@example.com.
Disclosure: Duggal reports no relevant financial disclosures.