Early alterations of circulating immune cell counts in severe trauma patients are related to later occurrence of nosocomial infection, sepsis and mortality

Background: Severe trauma leads to extensive disturbances of the innate and adaptive arms of the immune system, which in turn may affect the prognosis. The main objective of this study was to investigate the relationship between the alterations of circulating immune cell counts in the early stage of severe trauma and the later occurrence of nosocomial infection, sepsis and mortality. Methods: This was a retrospective study of 876 patients with an Injury Severity Score (ISS) ≥ 16. Demographic data, the absolute counts of neutrophil, lymphocyte and monocyte (ANC, ALC and AMC) on days 1, 3, and 7 (D1, D3, and D7) after trauma, and whether nosocomial infection, sepsis or death occurred within 60 days were recorded. Ratios were calculated between immune cell counts of each two time points, namely day 3/day 1 (D3/D1) and day 7/day 3 (D7/D3). Patients were grouped based on ISS and the occurrence of nosocomial infection, sepsis or death. Comparative studies were conducted between each two groups. Univariate and multivariate logistic regression analysis were used to identify variables related to the risk of nosocomial infection, sepsis, and mortality. Receiver operating characteristic (ROC) curve was plotted to assess the predictive value of various risk factors. Results: More severe trauma leads to more pronounced increase in ANC and more slowly recovery of ALC. In patients with subsequent nosocomial infection and sepsis, ANC was higher and ALC recovery was slower than those without nosocomial infection and sepsis within 7 days. In non-survivors, ALC had not recovered and AMC (D3) and AMC (D7) was lower than survivors within 7 days. ALC (D3) and ALC (D3/D1) are independent risk factors for nosocomial infection and sepsis. ALC (D3), obstructive pulmonary diseases.


Background
Trauma is a disease with high morbidity and mortality. More than five million people die of trauma every year. It is also the primary cause of death in individuals under 40 years old [1,2]. The main cause of early death at the scene or after admission is mainly severe brain damage or massive bleeding due to heart or large blood vessel injury. With the advancement of resuscitation strategies, damage control surgery and supportive care, many inpatients can survive the early stage. However, for these early survivors, delayed death due to sepsis and multiple organ dysfunction syndrome (MODS) is also frequent [3][4][5][6].
Extensive tissue injury and ischemia-induced release of damage-related molecular models (DAMP) after severe trauma leads to a tremendous inflammatory response (systemic inflammatory response syndrome, SIRS) and an extremely important antiinflammatory response (compensatory anti-inflammatory response syndrome, CARS) occurs almost simultaneously [4,[7][8][9][10]. These early events are generally believed to disrupt the homeostasis of the immune system and are important causes of nosocomial infection, sepsis and MODS in the later stage [11][12][13][14]. And this disruption of homeostasis is quite complex, affecting the innate and adaptive arms of the immune system, leading to significant changes in the count, function and phenotype of the immune cells [4,15,16]. Nevertheless, if we continue to systematically deepen the study of these immune cell changes, we may also have the opportunity to find out general rules in the complexity. This in turn may help us to predict that some patients may have adverse complications or poor prognosis in advance, for which we can conduct early intervention. Moreover, the changes of these immune cells may also be the cause of adverse complications or poor prognosis, which may also help us to expand the idea of immunotherapy.
Many studies have previously reported some changes in the post-traumatic immune system and the impact of these changes on prognosis. However, the following points may be worth further consideration: Simultaneously studying the response of innate and adaptive immune cells in the same group of patients to post-traumatic immune system status is more comprehensive than studying one type of immune cells alone; Considering that the immune system is dynamic, it is more persuasive to study the law of change in a period of time than to investigate the state of a time point;  Table S1). ANC and AMC rise and ALC decline were common in severe trauma patients. The proportions of patients with immune cell counts above the upper limit of the normal range of ANC and AMC or below the lower limit of the normal range of ALC are shown on the x-axis (Fig. 1). Patients with higher ISS had higher ANC within 7 days. Although ANC of both groups showed a decrease on the 3rd day, it rebounded in patients with higher ISS on the 7th day (Fig. 1A). On the 1st day, there was no significant difference in ALC between the two groups.
Then, ALC of the lower ISS group gradually increased from the 3rd day, while ALC of the higher ISS group began to rise on the 7th day. And ALC of the latter was lower than the former on the 3rd and 7th day (Fig. 1B). Besides, AMC increased in the higher ISS group on the 7th day, but there was no significant difference between the two groups (Fig. 1C). These results show that patients suffering from more severe trauma have worse alterations in immune cell counts, especially ANC and ALC.

Nosocomial infection
According to whether nosocomial infection occurred, patients were divided into two groups: non-nosocomial infection and nosocomial infection. A total of 476 patients had nosocomial infection, with the lungs being the most common site, and Gramnegative bacteria being the most common pathogen (data not shown). Nosocomial infection group had higher ISS and APACHE II score, lower GCS and longer ventilator use time (Additional file 1: Table S1). The trends of differences in ANC and ALC between the non-nosocomial and nosocomial infection groups were similar to those between the lower and the higher ISS groups (

Sepsis
According to whether sepsis occurred, we divided the patients into two groups: nonsepsis and sepsis. A total of 238 patients developed sepsis, with the lungs being the most common site of infection, and Gram-negative bacteria being the most common pathogen (data not shown). Sepsis patients were older, with higher ISS and APACHE II score, lower GCS and longer ventilator use time (Additional file 1: Table S1). The trends of differences in ANC and ALC between the non-sepsis and sepsis groups were similar to those between the lower and higher ISS groups or the non-

Mortality
According to whether death occurred within 60 days, patients were divided into two groups: alive and death. A total of 161 patients were included in the latter group, and they were older, with higher ISS and APACHE II score, lower GCS, and longer ventilator use time compared with the former group (Additional file 1: Table S1).
ANC of both groups reduced on the 3rd day and rebounded on the 7th day, but the latter group was different from the former group in that the ANC increased to the same level as the 1st day on day 7. And there was a difference in ANC between the two groups on day 7 (Fig. 4A). There was no difference in ALC between the two groups on day 1, it gradually increased in the alive group but no significant change in the death group. As a result, ALC differed between the two groups on the 3rd and 7th days (Fig. 4B). Besides, AMC increased on the 7th day in the alive group and was higher than the death group on the 3rd and 7th days (Fig. 4C). Moreover, ALC (D3/D1), AMC (D3/D1) and AMC (D7/D3) were different between the two groups ( Fig. 4D). These results suggest that worse alterations in immune cell populations may increase the risk of mortality.
We further divided the non-survivors into two groups according to the cause of death: sepsis and trauma. There were 96 (59.6%) patients died of sepsis and 65 (40.4%) patients died of trauma. There were no significant differences in age, gender, ISS, APACHE II score, and GCS between the two groups, as well as ANC, ALC, and AMC (Fig. 4E, 4F, 4G and Additional file 1: Table S1). These results imply that regardless of whether the patients eventually died of sepsis or trauma, they may suffer a similar degree and potentially fatal immune system insult in the early stage.

Risk factor analysis
To further determine the relationship between the immune cell counts and the occurrence of nosocomial infection, sepsis and mortality, we performed risk factor analysis. Since the median time of nosocomial infection was day 6 (data not shown) in our study, we did not analyze the effect of ANC (   The neutrophil-lymphocyte ratio (NLR) has been widely proved to be a simple and reliable marker for assessing systemic inflammation and prognosis in non-traumatic and traumatic patients [1, [18][19][20][21][22][23]. In our study, we validated the predicted value of NLR (D3) and NLR (D7) using ROC curve given their differences between each two groups (Fig. 5). Since the median time of nosocomial infection was day 6 (data not shown) in our study, which may be the cause of the rise of ANC (D7), we did not analyze the effect of NLR (D7) on nosocomial infection. The results show that NLR (D7) has a larger area under the ROC curve (AUC) in predicting the risk of sepsis and mortality than NLR (D3) ( Table 4). Then, we selected NLR which has a larger AUC and the independent risk factors of nosocomial infection, sepsis and mortality (Tables 2 and 3) for comparison. As depicted in Fig. 6A, 6B, 6C and Table 5   Considering that the independent assessment of immune cell counts at fixed time points or their slopes of two time points is one-sided, the combination of them was assessed using ROC curve. The results are shown in Fig. 6 and Table 6.  < 0.001 Table shows  were independent risk factors for sepsis, however, this may be only the manifestation of nosocomial infection. Therefore, it cannot be concluded that higher ANC will increase the susceptibility to nosocomial infection and sepsis. Moreover, ANC (D7) of the death group was higher than the alive group, but multivariate analysis showed that it was not an independent risk factor for death. Nevertheless, it should be noted that the elevation of ANC is only one aspect of neutrophil abnormalities (e.g. neutrophil function is also impaired) [4,[25][26][27]. The overall abnormal response of neutrophils may still lead to immunosuppression, which increases the risk of adverse complications and poor prognosis [5].
Lymphocytes play a central role in the anti-infective immune response due to their ability to interact with cells of the innate immune system and other cells of adaptive immunity. They are not only passive bystanders, but also play a key role in the proper regulation of the inflammatory response [28,29]. In contrast to the increase of ANC after trauma, ALC decreased, which has been widely reported [8,30,31].
Post-traumatic lymphopenia was also common in our study. More severe trauma did not initially lead to more severe lymphocyte depletion, but resulted in more slowly recovery of ALC. So that patients with higher ISS on days 3 and 7 had lower ALC.
This changing trend of ALC was also applicable to patients with subsequent nosocomial infection and sepsis. Moreover, for patients with subsequent nosocomial infections and sepsis, ALC showed a tendency to recover on the 7th day, but for patients with subsequent death, ALC had no tendency to recover at all within 7 days. According to the results of slope, ALC did not recover or even decline on the 3rd day in patients with subsequent nosocomial infection, sepsis or death. In addition, multivariate analysis showed that ALC (D3) and ALC (D3/D1) were independent risk factors for nosocomial infection, sepsis and mortality, and ALC (D7) was also an independent risk factor for mortality. These results are similar to previous findings in critically ill patients such as trauma and sepsis, the extent and delayed recovery of lymphopenia have been widely reported to be related to nosocomial infection, sepsis and death [28,30,[32][33][34][35]. Taking  This study has several limitations. Firstly, this is a single-center retrospective study.
The nature of this study may lead to unintentional bias in patient selection. Besides, blood transfusion can affect the function of immune system [44][45][46], but the interference of blood transfusion is not excluded in our study. Finally, although we excluded patients who underwent surgery from after blood collection on day 1 to before blood collection on day 7, emergency surgery performed at admission and surgery performed 7 days after admission may also have an impact on the results.

Declarations
Ethics approval and consent to participate: Approval was obtained from the Medical Ethics Committee of the hospital with a waiver of informed consent.

Consent for publication: Not applicable.
Availability of data and materials: The datasets used and/or analyzed during the present study are available from the corresponding author on reasonable request.
Competing interests: All authors declare that they have no competing interests. Authors' contributions: XJD and CTW were responsible for data acquisition, statistical assistance and data analysis. XHL were responsible for manuscript drafting and revision. XJD and ZFL were responsible for concept and design, data analysis, manuscript drafting and revision. XJD, CTW XHL, XJB and ZFL discussed the results. All authors read and approved the final manuscript. Early circulating immune cell counts show worse alterations in non-survivors, but there is no Differences in NLR (neutrophil-lymphocyte ratio) were compared between non-nosocomial inf Additional file 1.xlsx