Abstract
We read with interest the study by Doğan et al. 1 and question the authors’ interpretation.
The study’s primary outcome is to determine whether serum lactate correlates with hyperbaric oxygen (HBO) therapy. The main rationale for HBO in carbon monoxide (CO) poisoning is to reduce the risk of delayed neurologic sequelae. 2,3 However, HBO is a treatment modality and not a measurable clinical outcome. The decision to initiate HBO is dependent on multiple factors, and administration of HBO does not necessarily indicate more severe poisoning.
Furthermore, the authors state, “a lactate value greater than 1.85 mmol/L predicts the need for HBO.” We find this problematic when the normal range for serum lactate is below 2.1 mmol/L. It is concerning to state that a normal laboratory result should be an indication of severity of disease as well as a trigger to initiate HBO.
In this study, a lactate value of 1 mmol/L has a negative predictive value of 96.3%. This negative predictive value does not predict disease severity but rather whether HBO was initiated. While patients with low lactate values are less likely to have significant CO-related symptoms (e.g. neurologic impairment, altered mental status, ischemic symptoms, and hemodynamic instability 4,5 ), a patient who has these clinical findings would be considered severely CO poisoned despite the normal serum lactate.
We commend the authors’ efforts in this study. However, we believe it is premature to utilize the lactate level as a marker of CO poisoning based on this study.