Laboratory Critical Values Should Support Effective Clinical Decision Making.

In their article on critical values (CVs), Doering et al1 made a valuable report. Their retrospective study on a large cohort of inpatients investigated the mortality rate in relation to the local (North Shore–LIJ Heath System) CV list. However, we believe that some of the points made by the authors and the method used for establishing the CV list and cutoffs deserve further consideration. As the authors stated, the study was performed using a set of CVs and cutoffs obtained in the literature, but at least one CV or cutoff used differed from those appearing in the references. For example, most of the published studies propose higher CVs for both the international normalized ratio (INR) and glucose than those advocated by the authors. Moreover, according to the current literature, lactate is defined as a critical test, while the high (critical) values cited for hematocrit and hemoglobin should be considered significantly abnormal laboratory results, according to the Joint Commission recommendation.2 Since the authors’ use of inappropriate CV limits resulted in an unnecessarily large volume of notifications, the CVs should be based on the answers to the following key questions: (1) Are the values potentially life-threatening? (2) Does the patient need prompt therapeutic action? (3) How can the appropriateness of the CVs be determined? In their conclusion, Doering et al1 maintained that, as a paradigm of appropriateness, high glucose results should not be considered in the CV list. This view contradicts the conclusions made by Pasquel and Umpierrez3 in their recent review: these authors highlight the importance of the hyperosmolar hyperglycemic state (HHS), which represents the most serious possible acute hyperglycemic emergency in patients with type 2 diabetes. Current HHS diagnostic criteria recommended by the American Diabetes Association and in international guidelines include a plasma glucose level more …