© 2005 American Thoracic Society
From the Authors:In their work, Branco and colleagues analyzed children with sepsis, severe sepsis, and septic shock (1), whereas we also analyzed children with SIRS (2). According to their data (1), in our population excluding children with SIRS, mortality rose from 0% in those with one organ dysfunction (OD) to 7.5% in those with two OD (p = 0.2), whereas it rose from 7.5% in those with two OD to 27.0% in those with three OD (p = 0.02) (2). Among our children with sepsis, severe sepsis, and septic shock, mortality was 0% in those with less than two OD and 27.6% in those with two or more OD (p = 0.0002), whereas mortality was 3.8% in those with two or less OD and 41.5% in those with more than two OD (p < 0.0001) (2). As observed by Branco and colleagues (1), mortality of our children with more than two OD was not different between those with and those without a septic state (p = 0.58) (2). However, these descriptive data do not indicate how septic states influence the outcome of children with MODS. In our study, the Cox model, using the OD number as a fixed covariate and the worst septic state (from SIRS to septic state) as a time dependent variable, showed that each OD multiplied the hazard ratio of death (HR) by 2.374 (p < 0.0001) whatever the septic state category (2). Each septic state category multiplied the HR by a value (all p < 0.05) independent of the OD number (2). The cumulative influence was obtained by multiplying the HR of the OD number by the HR of the septic state category (2). To better understand the interactions between sepsis and MODS, elucidating the pathophysiology of MODS is needed. An important contribution from Branco and colleagues (3) is in reporting the association in children with septic shock of a peak level glucose > 178 mg/L (indicating metabolic dysfunction) and an increased risk of death, which relates to a study in which anion gap acidosis (related to altered glucose metabolism) was associated with higher MODS scores and death in children in shock (p < 0.001) (4). Has "the definition of MODS become obsolete" and does "using a definition of MODS requiring the presence of three or more OD better reflect the effect of OD on outcome" (1)? MODS is traditionally defined by the dysfunction of two or more organs (2, 5, 6). Although Wilkinson and colleagues considered OD as present or absent (5), OD is an ordinal process and a score based on scaled and weighted variables is probably more informative (1, 6). Thus, we think that the PELOD score should be preferred to the OD number, as the OD number does not take into account the severity level and the weight of each OD (1, 6). As mortality in PICU is a function not only of the OD number but also of the relative risk and the degree and duration of each OD, there is no reason to use only the two states of normal function or dysfunction. Finally, detecting developing OD with scaled MODS scores would be beneficial so that early treatment can be initiated.
Université de Lille Lille, France and Université de Montréal Montréal, Quebec, Canada FOOTNOTES Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
