A dynamic view of mHLA-DR expression in management of severe septic patients

Increasing evidence suggests that the secondary phase of sepsis (that is, after the first proinflammatory hours) is characterized by the occurrence of a systemic failure of the immune system. In the most immunodepressed patients, therapies could be used to restore normal immune functions. However, biomarkers need to be developed to beforehand specifically identify these patients. Of these biomarkers, diminished monocyte HLA-DR expression has rapidly become the most popular. Herein, novel perspectives regarding monocyte HLA-DR assessed as a dynamic parameter in septic patients will be discussed in the context of a recently published study investigating daily evolution of monocyte HLA-DR with regard to 28 day-mortality after severe sepsis.

Th e work of Wu and colleagues [1] is in accordance with the recent concept of sepsis-induced immunosuppression [2,3]. Th ere is now agreement that many severe septic patients survive the fi rst critical hours of the syndrome but eventually die later in a state of immunosuppression that is illustrated by patients' diffi culty to fi ght the primary bacterial infection, decreased resistance to secon dary nosocomial infections and reactivation of viral infections normally solely pathogenic in the immunocompromised host [4,5]. Consequently, immunostimulatory therapies might be used to restore immune functions in the most immunodepressed patients [6]. In the absence of any specifi c clinical signs of immune failure, however, it is beforehand critical to determine the best biological tools (markers of septic patients' immune failure) enabling patient stratifi cation [7].
Th e most frequently assessed biomarker in the fi eld to date is undeniably the measurement of HLA-DR expression on circulating monocytes (mHLA-DR). Th ere appears to be general consensus that diminished mHLA-DR is a reliable marker for the development of immuno suppression in critically ill patients. Indeed, decreased expression of this marker is regularly reported to be associated with higher mortality/risk for nosocomial infections in critically ill patients. More than 100 articles on this topic have been published in diff erent ICU conditions, including sepsis, trauma, burns, and stroke.
It is becoming increasingly clear that the critical point after injury is the recovery of normal mHLA-DR. Schema ti cally, mHLA-DR rapidly returns to normal values (generally in less than 1 week) in injured patients with uneventful recovery, whereas this parameter remains constantly decreased in patients with adverse outcome/ secondary septic complications. In line with this hypothesis, Wu and colleagues showed that low mHLA-DR was associated with increased mortality in severe sepsis [1]. Most importantly, the authors propose that, more than a single value at a given time point, the dynamic change of mHLA-DR over time would be a better predictor of mortality. Indeed, in their study, single measurements of mHLA-DR within the fi rst week after patient admission (either days 0, 3 or 7) had no predictive value regarding mortality. In contrast, results expressed as dynamic parameters (that is, Δ between two time points) provide excellent predictive values, especially Δ calculated between days 0 and 3 or between days 0 and 7 (areas under the curve of 0.92 and 0.94, respectively, in receiver operating characteristic analysis). Most importantly, after multivariate analysis, the authors show that these two parameters remain the sole independent predictors of mortality with an elevated signifi cant odds ratio.
Overall, the present results confi rm the concept that patients who do not start to restore normal immune functions are those who are going to die [1]. Th ese results are in agreement with two recent studies in which a weak slope of mHLA-DR recovery was associated with increased risk of secondary infections in a mixed ICU

Abstract
Increasing evidence suggests that the secondary phase of sepsis (that is, after the fi rst proinfl ammatory hours) is characterized by the occurrence of a systemic failure of the immune system. In the most immunodepressed patients, therapies could be used to restore normal immune functions. However, biomarkers need to be developed to beforehand specifi cally identify these patients. Of these biomarkers, diminished monocyte HLA-DR expression has rapidly become the most popular. Herein, novel perspectives regarding monocyte HLA-DR assessed as a dynamic parameter in septic patients will be discussed in the context of a recently published study investigating daily evolution of monocyte HLA-DR with regard to 28 day-mortality after severe sepsis. population [8] and in trauma [9]. Th is outcome could have important consequences in patient management, by potentially allowing for the adminis tration of tailored therapies aimed at restoring immune functions based on dynamic changes of immuno logical parameters.
Th ere are some limitations to Wu and colleagues' study that should be noted. First, the study is monocentric in a small cohort of surgical patients (that is, not necessarily representative of the whole septic population) that present with relatively elevated mHLA-DR values (>50%) in comparison with results from the literature (usually below 50% in severe septic patients). Th is moderate severity and the lack of statistical power due to the small size of the cohort may explain surprising results after multivariate analysis (Sequential Organ Failure Assess ment and Acute Physiology and Chronic Health Evaluation II scores were not signifi cantly diff erent between survivors and non survivors, odds ratio with very large confi dence intervals). Another limitation lies in the expression of mHLA-DR results. Indeed, the standard ized recommended method for expressing mHLA-DR results is as numbers of antibodies bound per cell and not as the percentage of positive cells [10,11]. As an example, the authors suggest that a diff erence of 4.8% in mHLA-DR between days 0 and 3 is of signifi cance. Although most probably correct from a statistical perspective, this threshold is hardly applicable in routine/technical practice because such a small percen tage diff erence could be due to measurement variability by fl ow cytometry.
Overall and beyond these limitations, appropriately acknowledged by the authors, this study confi rms that after injury (for example, severe sepsis) survivors tend to progressively normalize mHLA-DR, contrary to nonsurvivors. Th is biologic parameter could thus provide critical informa tion when assessed as a dynamic variable over time. Th is potential aspect now deserves to be validated in multi centric clinical studies using standardized fl ow cytometry protocols.