Recognizing the mobilization of neutrophils with banded nuclei early after trauma

Dear Editors, After major trauma neutrophil function is negatively affected (eg, diminished responsiveness to stimuli, chemotaxis, phagocytosis1), contributing to increased susceptibility of infection. The initial presence of banded, segmented, and hypersegmented neutrophils (based on CD16/CD62L expression/associated with nuclear segmentation) is linked to late infectious complications2 and the development of multi-organ failure3 during admission in major trauma patients. Therefore, it is of great importance to recognize neutrophil subsets to obtain all information on neutrophil function after trauma. Automated hematological analyzers made neutrophilic differentiation an easily accessible tool to automatically analyze the number of progenitors and banded cells based on predefined algorithms.4 White blood cell counts and, particularly, neutrophil left shifts (increased counts of immature or banded neutrophils) have proven their diagnostic value in bacterial infection.5 The value for diagnosis of (sterile) inflammatory diseases is far less obvious. In the case of trauma, the presence of banded neutrophils in the first day has been described.6 Despite the consensus regarding the occurrence of banded neutrophils during infection, the situation after traumatic injury is much less clear. Differentiation data obtained with classical microscopy showed fast numbers (10%-98%) of banded neutrophils recirculating immediately after major trauma.6 However, applying automated leukocyte differentiation by a routine hematological analyzer in earlier studies on trauma patients revealed minimal band counts (0%-5%).3 We aimed to investigate the accuracy of a fully automated hematological analyzer to identify the neutrophil left shift following major trauma and compare results to manual differentiation and immunophenotyping by flow cytometry. Eight samples from six adult polytrauma patients (<48 hours after trauma, injury severity score >16, all males, median age


Dear Editors,
After major trauma neutrophil function is negatively affected (eg, diminished responsiveness to stimuli, chemotaxis, phagocytosis 1 ), contributing to increased susceptibility of infection. The initial presence of banded, segmented, and hypersegmented neutrophils (based on CD16/CD62L expression/associated with nuclear segmentation) is linked to late infectious complications 2 and the development of multi-organ failure 3 during admission in major trauma patients.
Therefore, it is of great importance to recognize neutrophil subsets to obtain all information on neutrophil function after trauma.
Automated hematological analyzers made neutrophilic differentiation an easily accessible tool to automatically analyze the number of progenitors and banded cells based on predefined algorithms. 4 White blood cell counts and, particularly, neutrophil left shifts (increased counts of immature or banded neutrophils) have proven their diagnostic value in bacterial infection. 5 The value for diagnosis of (sterile) inflammatory diseases is far less obvious. In the case of trauma, the presence of banded neutrophils in the first day has been described. 6 Despite the consensus regarding the occurrence of banded neutrophils during infection, the situation after traumatic injury is much less clear. Differentiation data obtained with classical microscopy showed fast numbers (10%-98%) of banded neutrophils recirculating immediately after major trauma. 6 However, applying automated leukocyte differentiation by a routine hematological analyzer in earlier studies on trauma patients revealed minimal band counts (0%-5%). 3 We aimed to investigate the accuracy of a fully automated hematological analyzer to identify the neutrophil left shift following major trauma and compare results to manual differentiation and immunophenotyping by flow cytometry.  Database, a relational database infrastructure. 8 For the manual differentiation, either blood smears of whole blood or cytospins of 1 × 10 5 leukocytes were used (<4 hours), based F I G U R E 2 A and B, Gating strategy for flow cytometry identification of immature and banded neutrophils. In A an example of a healthy control, in B an example of a polytrauma patient upon presentation at the emergency department. Granulocytes are identified on their specific forward/side scatter, after which neutrophils are characterized on CD16/CD62L expression. For sorting, Q1 was identified as the banded and Q2 as mature neutrophil gate. See also Figure 2B. C. Sorting of CD16 dim and CD16 high neutrophils classifies banded and segmented neutrophils correctly. CD16 dim and CD16 high neutrophils were sorted and single-cell cytospins prepared and scored for the presence of banded cells. Results were regarded as significant when P < .05.
In none of the trauma patient samples, the automated hematology analyzer detected any immature or banded neutrophils (0% ± 0%). In surgical patients with acute infection, the automated differentiation did reveal banded neutrophils (22% ± 8%). In marked contrast to automated analysis, manual differentiation revealed large numbers of banded neutrophils in both trauma patients (37% ± 17%, P = .0078) and infection patients (22% ± 6%, P = .8125) ( Figure 1A,B). both groups without significant differences between groups. In the trauma group, no significant differences in cell size between CD16 dim and CD16 high neutrophils were found. An exemplary scatter plot is shown in Figure 3.
The Cell-Dyn algorithm uses neutrophil scatter parameters including increased 0-degree scatter (proxy for neutrophil size) and 90-degree scatter (proxy for neutrophil lobularity) to detect immature neutrophils (exact algorithm remains confidential). Thus, cell size is part of this algorithm and we found that neutrophil cell size was increased during infection, whereas it was not after trauma. An increase in neutrophil size has been found after in vitro activation with inflammatory mediators. 10 So, it is possible that mean neutrophil size was not increased because of the presence of larger banded neutrophils, but simply because of neutrophil enlargement due to in vivo activation of neutrophil during infections. This is supported by our finding that there was no difference in cell size between banded (CD16 dim /CD62L high ) and segmented (CD16 high /CD62L high ) neutrophils after trauma. This F I G U R E 3 Exemplary scatter plots showing forward scatter (size, ALL) and side scatter (lobularity, PSS) of infection and trauma patients might be (part of) the explanation for the inability of the hematological analyzers algorithm to detect these banded cells. For the purpose of this study, only one analyzer was tested. Although it is tempting to speculate that other analyzers would have the same problem, as algorithms are mostly based on light scatter patterns, caution should be taken in extrapolating these findings to other hematological analyzers.
Although we tested a limited number of patients, we provided proof for the principle that algorithm-based automated neutrophil differentiation using a Cell Dyn hematology analyzer is not suited to detect the presence of banded neutrophils in the circulation following major trauma.
On the contrary, it accurately detected banded neutrophils in patients suffering from infections. Phenotyping neutrophils based on their CD16/ CD62L expression was a more accurate method to automatically identify the presence of banded neutrophils in the circulation of trauma patients.
Thereafter, studies should focus on the differences in banded neutrophils and whether these cells only differ in phenotype or also in function.

K E Y WO R DS
banded, diagnosis, infection, neutrophil, trauma