Increased Levels of Serum S100B Protein in Critically Ill Patients Without Brain Injury

Abstract

Increased levels of serum S100B protein in critically ill patients without brain injury.Shock 26:20-24, 2006. To the Editor: The recent study by Routsi et al. in Shock has given further support to evidence that serum levels of S100B protein may be elevated in the absence of gross brain damage (e.g., traumatic brain injury or subarachnoid hemorrhage) (1). The authors measured S100B in 46 mechanically ventilated patients in the intensive care unit (ICU) who were not diagnosed with overt brain injury and found that this protein-previously shown as a marker of neurological injury (2-5)-was elevated in every patient at least once during their ICU stay. Two hypotheses are put forth by Routsi et al. to possibly explain these findings: (1) S100B elevation reflects brain injury that has resulted from brain tissue hypoxia, or (2) S100B is released from extracerebral tissues injured as a result of critical illness. Although both hypotheses have merit and warrant further investigation, the current study provides weaker support to the latter hypothesis (i.e., the study was not designed to thoroughly and systematically evaluate the enrolled patients in ICU for brain injury). No diagnoses of structural brain damage were made, but brain dysfunction, in the absence of stroke or trauma, is now known to be a frequent consequence of critical illness (6, 7). Delirium is the most common acute form of critical illness-related brain dysfunction and can be rapidly and reliably diagnosed in mechanically ventilated patients using a well-validated tool known as the Confusion Assessment Method for the ICU (8, 9). Although pathophysiological mechanisms remain poorly understood, the development of delirium in up to 80% of mechanically ventilated patients in the ICU (10-12) makes it unlikely that the patients in the study by Routsi et al. had no symptoms of brain dysfunction. An analysis of the association between S100B levels and the presence of delirium may have helped in determining whether the elevated S100B noted in their study was a reflection of neurological injury. We realize that they do not have those data, but their absence does not allow for the conclusion that S100B was elevated in patients "without brain injury." There are well-validated tools now available in many languages by which the diagnosis of delirium is tracked daily in thousands of ICUs all over the world (see www.icudelirium.org). In addition, critically ill patients are at significant risk for development of long-term brain dysfunction in the form of cognitive impairment. Long-term cognitive impairment is now known to occur in nearly half of patients who survive critical illness, suggesting that significant brain dysfunction occurs in the context of critical illness (13, 14). The possibility that S100B elevations during or after critical illness correlate with neurocognitive outcomes remains unstudied. As we, as researchers and clinicians, strive to improve the outcomes of patients after critical illness and injury, it is crucial that we broaden our understanding of critical illness-associated brain dysfunction and evaluate patients for both delirium and long-term cognitive impairment in studies of the neurological consequences of critical illness. We look forward to seeing future work from this group of authors as they continue tocontribute to this exciting-and still emerging-field of research. Bryan A. Cotton, MD Timothy D. Girard, MD E. Wesley Ely, MD, MPH Vanderbilt University School of Medicine Nashville, Tennessee