The role of lactic acid in carbon monoxide poisoning and the difficulty of defining the worse outcome.

Abstract

On behalf of all coauthors of our manuscript, I would thank to the authors for their interest and comments to the article. The main obstacle in designing studies involving carbon monoxide (CO) poisoning was defining an accurate primary outcome. In this population, immediate or delayed neurologic sequelae was an important concern in evaluating patients with CO exposure; however, mortality was quite low, particularly if the patients could reach to the emergency department with unintentional poisoning. The referral for hyperbaric oxygen (HBO) treatment was considered an important sign of severe intoxication in many centers, with the decision to referral depends not only on established criteria but also on overall clinical picture and subjective decision of clinicians. The severity of the symptoms, which was defined by the authors (e.g. altered mental status, hemodynamic instability, and ischemic symptoms) was already used for HBO treatment need and defining the severity of the disease. As we observed that setting an exact primary outcome for this population was somewhat difficult, we accepted the HBO treatment decision as a sign of severe poisoning. Also Cervellin et al. used the HBO treatment to compare the characteristics of the patients. However, we need more accurate end points to claim the severity of the disease. Obviously, a cohort study configured with a composite outcome (e.g. delayed neuropsychiatric deficits, behavioral disorders, and attention disorders) in the future may add more valuable information about the effects of CO and the role of lactate. As the authors mentioned, we stated that a lactate value greater than 1.85 mmol/L predicts the need for HBO treatment. However, this value has moderately low sensitivity and specificity (70.8% and 78.0%, respectively). We agree to the authors that this sensitivity and specificity measures should be considered carefully in treatment decisions. However, a more remarkable finding in our study was the high negative predictive value (96.3%) of lactate levels of <1 mmol/L. Lactic acid is the product of anaerobic glycolysis in association with poor oxygen supply, and it can be taken as a measure of an early warning sign in many poisonings, which affect cellular oxygen supply and demand. Although data about the role of initial lactate levels in CO poisoning revealed conflicting results, the scientific evidence strengthened by recently published studies. We disagree to the authors, as they stated the lactate levels in ‘‘normal ranges’’ could not be utilized as a marker of disease severity. As previously demonstrated in many studies, increase in lactate levels was a dynamic process in critically ill patients, and moreover, future studies involving lactate clearance may add more valuable prognostic information about CO poisoning.