| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Immunology Laboratory and Intensive Care Units, Lyon-Sud University Hospital, 69495 Pierre-Bénite, France
aaddress correspondence to this author at: Flow Cytometry Unit, Immunology Laboratory, Lyon-Sud University Hospital, 69495 Pierre-Bénite, France; fax 33-4-7886-3344, e-mail guillaume.monneret{at}chu-lyon.fr
The concept of immunoparalysis has recently been proposed for explaining the failure of 20 years of clinical trials using antiinflammatory drugs in sepsis (1)(2)(3). Immunoparalysis is characterized mainly by the paralysis of monocytic functions. In particular, because of decreased expression of HLA-DR, antigen-presenting capacity is severely depressed (4)(5). Recent clinical studies have confirmed that among a large panel of activation markers expressed on different leukocyte populations (i.e., neutrophils, lymphocytes, and monocytes), decreased HLA-DR expression on monocytes constitutes a reliable marker of immunoparalysis and seems to correlate with an increased risk for fatal outcome (6)(7)(8). Nevertheless, little work has been devoted to analytical aspects related to its measurement by flow cytometry. Indeed, flow cytometry cannot yet be considered a standardized tool, and many variables must be taken into account for ensuring the technical quality of results (9). This is especially required in clinical research when clinicians and immunologists are assessing the potential value of a new marker, as is presently the case for monitoring septic patients, and it is particularly true for the determination of HLA-DR, which is considered a rapidly up- or down-regulated marker (5)(10). The present study was designed to determine whether different protocol procedures could lead to discrepant results for HLA-DR measurements. After establishing a reliable protocol, the second objective was to demonstrate immunoparalysis in monitoring patients with septic shock.
Samples of peripheral blood were collected in EDTA anticoagulant tubes. Staining was performed on whole blood using PC5-labeled CD45, fluorescein isothiocyanate (FITC)-labeled CD14 (Immunotech), and phycoerythrin-labeled HLA-DR (clone L243; Becton Dickinson). Samples were lysed manually by use of FACS lysing solution (Becton Dickinson) or the automated Q-Prep system (Beckman-Coulter). Cells were analyzed on a Coulter EPICS XL flow cytometer (System II software; Beckman-Coulter). After debris was excluded by means of a leukogate (11), monocytes were gated out from other cells on the basis of labeling with FITC-CD14 according to the latest consensus conference (12). All results are expressed either as percentages of HLA-DR-positive monocytes in the total monocyte population (a threshold was defined with the isotype control) or as the mean of fluorescence intensities (MFI) related to the entire monocyte population. Results are presented as the mean ± SE. The Wilcoxon matched-pairs test and the Mann–Whitney test were used for the statistical study.
The patient group consisted of patients with septic shock. We used the diagnostic criteria for septic shock of the American College of Chest Physicians/Society of Critical Care Medicine consensus conference committee (13). Because we focused our interest on the secondary stage (immunoparalysis) of septic shock, we excluded patients who did not survive more than 48 h after being admitted to intensive care units. Severity was assessed by the Simplified Acute Physiologic Score II (SAPS II) (14). Mortality was defined as death occurring within 28 days after diagnosis. To provide a panel of reference values, we also included 58 apparently healthy individuals from our laboratory.
Regarding the expression of results as a percentage of HLA-DR-positive monocytes or as MFI, we rapidly noticed that these two values, although correlated, could provide different information. It remains to be demonstrated which of these two results will better correlate with clinical resolution or improved survival. Furthermore, MFI results provided the opportunity to convert data as numbers of HLA-DR sites per monocyte by use of calibrated beads, which constitutes a first approach to standardizing results. Because using MFI implies working with antibodies at saturating concentrations, we had to ensure this point. We tested serial dilutions of the antibody directed against HLA-DR. As hypothesized, dilutions led to artificially lower results. We statistically confirmed these data in eight healthy donors by comparing the two highest concentrations of antibodies (1/1 vs 1/2). Regarding the percentages of HLA-DR-positive monocytes, we found no difference. In contrast, significantly different MFI results (17 ± 4 vs 23 ± 1 MFI; P <0.01) supported the fact that antibodies must be used at saturating concentrations (20 µL/100 µL of whole blood) when expressing results as MFI.
Because HLA-DR is a marker that is rapidly regulated, the storage temperature before staining may be an important issue. We compared results from fresh whole blood and those obtained after storage for 24 h at 4 °C. In healthy donors, storage significantly increased MFI results (Table 1
). In septic patients, both the percentage values and MFI were increased after storage (Table 1
). We also stored blood from five healthy donors at 37 °C for 1 h and found a dramatic increase in the HLA-DR values measured for each sample, which supports a rapid nonspecific in vitro modulation of this expression (Table 1
). Finally, we investigated the stability of HLA-DR expression at room temperature in five healthy patients (Table 1
). Our results indicated that samples should be stained within 1 h after sampling. We also demonstrated that temperature variations during staining led to discrepant results (data not shown). Thus, we chose to work as recommended by the manufacturer (i.e., at room temperature).
|
To ensure that the lysis step did not induce an analytical bias, because different lysis systems are used in multicenter studies, we compared an automated lysis system to a manual reagent. We found discrepant results with a high variability independent of the lysis system (data not shown). Because we wanted to develop a protocol usable for a multicenter study, we used the manual lysis procedure, as automated lysis systems are not available worldwide. In many hospitals, flow cytometers are usually not available 24 h a day for rapid measurement. An alternative is to immediately stain and fix cells after sampling and analyze them afterward. We verified this procedure by analyzing 25 samples just after staining and after 24 h at 4 °C and found no discrepant results.
On the basis of our results, we defined the most reliable protocol for measuring monocytic HLA-DR as follows. Fresh EDTA whole blood is immediately stained at room temperature for 30 min. Samples are lysed with the FACS lysing solution (Becton Dickinson) and immediately analyzed. After debris is excluded by means of a leukogate, monocytes are gated out from other cells on the basis of labeling with FITC-CD14. Results are expressed as the percentage of monocytes expressing HLA-DR (a threshold is defined with the isotype control) and as MFI.
We also performed two sets of analyses to study intraassay precision: one sample stained once and analyzed 10 times or one sample stained 10 times and analyzed 10 times. The intraassay CVs ranged from 3.9% to 5.0% (percentage of HLA-DR-positive monocytes) and from 2.0% to 2.3% (MFI).
The present study confirmed that HLA-DR is a marker whose determination is extremely dependent on analytical procedures. Consequently, a common protocol should be defined to make data from different centers as comparable as possible. The elapsed time between blood collection and staining constitutes a critical step: we observed a rapid nonspecific increase in measured HLA-DR values after sampling. Our precision results also illustrate that despite critical steps in the measurement, accurate determination of HLA-DR on monocytes allows day-to-day comparisons. When MFI and calibrated beads for fluorescence quantification are used, results become comparable between different laboratories.
For our clinical findings, we studied 17 patients with septic shock (mean age, 58 years; 13 males and 4 females) with a mean SAPS II admission score of 48 (range, 32–89) and a global mortality of 47%. We also included 58 apparently healthy donors (age range, 25–65 years). At 48 h after admission, the expression of HLA-DR on monocytes was severely and significantly reduced in septic patients (25% ± 4%; MFI value, 6 ± 1) compared with healthy donors (89 ± 1%; MFI value, 35 ± 3) without overlap. When survivors and nonsurvivors were compared, there was no difference in HLA-DR monocytic expression at 48 h. However, HLA-DR concentrations 5 days after admission were significantly higher values in survivors, suggesting an ongoing recovery of their immunologic status (Fig. 1
).
|
In clinical reports, poor prognosis is associated with lower expression of HLA-DR on monocytes in the first days after admission, and it has been suggested that poor outcome might be attributable to this monocyte deactivation (6)(7)(8). The present study confirms and extends these data. We report that 5 days after admission, the persistence of a very low concentration of monocytic HLA-DR (<40%) is associated with a fatal outcome. Clinical and experimental results suggest that patients presenting with a low monocytic HLA-DR concentration should receive therapy, such as interferon-
, to stimulate immune function. However, despite a promising trial using this approach in 10 septic patients with <30% HLA-DR-positive monocytes (6), there is no definitive evidence that boosting the immune system can reduce mortality from sepsis in immunocompromised patients. A placebo-controlled multicenter trial based on monocytic HLA-DR concentrations is necessary. As we have shown in the present study, a critical issue in such a trial will be the capacity for each center to perform standardized measurements of HLA-DR.
Acknowledgments
We thank Justin Kingsley for greatly assisting in the preparation of this manuscript.
References
treatment. Nat Med 1997;3:678-681.[ISI][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  A. Pachot, M.-A. Cazalis, F. Venet, F. Turrel, C. Faudot, N. Voirin, J. Diasparra, N. Bourgoin, F. Poitevin, B. Mougin, et al. Decreased Expression of the Fractalkine Receptor CX3CR1 on Circulating Monocytes as New Feature of Sepsis-Induced Immunosuppression J. Immunol., May 1, 2008; 180(9): 6421 - 6429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.-D. Docke, C. Hoflich, K. A. Davis, K. Rottgers, C. Meisel, P. Kiefer, S. U. Weber, M. Hedwig-Geissing, E. Kreuzfelder, P. Tschentscher, et al. Monitoring Temporary Immunodepression by Flow Cytometric Measurement of Monocytic HLA-DR Expression: A Multicenter Standardized Study Clin. Chem., December 1, 2005; 51(12): 2341 - 2347. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Le Tulzo, C. Pangault, L. Amiot, V. Guilloux, O. Tribut, C. Arvieux, C. Camus, R. Fauchet, R. Thomas, and B. Drenou Monocyte Human Leukocyte Antigen-DR Transcriptional Downregulation by Cortisol during Septic Shock Am. J. Respir. Crit. Care Med., May 15, 2004; 169(10): 1144 - 1151. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
