Impact of Injury Severity on the Inflammatory State and Severe Anemia

Abstract

Background:

Severe traumatic injury is a major cause of morbidity and mortality. Our goal was to analyze blunt traumatic injury by injury severity score and compare to elective hip repair, as a transient injury, and healthy control with the hypothesis that more severe injury would lead to an increase in neuroendocrine activation, systemic inflammation and worse anemia.

Materials and methods:

A prospective observational cohort study was performed at a level one trauma center, comparing blunt trauma patients (n=37), elective hip replacement patients (n=26), and healthy controls (n=8). Bone marrow and plasma were assessed for hyperadrenergic state, erythropoiesis and systemic inflammation. Trauma patient’s injury severity score (ISS) ranged from 4 to 41 and were broken down into quartiles for analysis. The ISS quartiles were: 4–13, 14–20, 21–26, and 27–41.

Results:

Plasma norepinephrine, interleukin-6, tumor necrosis factor-alpha and hepcidin increased progressively as ISS increased. Hemoglobin significantly decreased as ISS increased and packed red blood cell (pRBC) transfusion increased as ISS increased. Elective hip replacement patients had an appropriate increase in the bone marrow expression of erythropoietin and the erythropoietin receptor which was absent in all trauma patient groups.

Conclusions:

Increased neuroendocrine activation, systemic inflammation, and anemia correlated with worsening injury severity, lower age and increased pRBC transfusions. Elective hip replacement patients have only minimal systemic inflammation with an appropriate bone marrow response to anemia. This study demonstrates a link between injury severity, neuroendocrine activation, systemic inflammation and the bone marrow response to anemia.

Introduction

Based on anatomic injury severity, the injury severity score (ISS) assesses the multiply injured patient and is used to help predict outcomes.¹ Improvements in the management of patients with severe traumatic injuries have led to prolonged intensive care unit (ICU) lengths of stay as well as the development of new and chronic morbidities. Approximately ninety-five percent of all trauma patients who remain critically ill for three or more days are affected by chronic anemia, despite the cessation of bleeding and the treatment of major injuries.^{2, 3} Allogenic red blood cell transfusion is the typical mainstay for initial management of that severe anemia. Following the first week after injury, despite having gained control of bleeding, patients continue to be transfused on average, one to two units of packed red blood cells per week and this has been linked with immune suppression, infectious complications and increased mortality.^4–6

While an acute anemia is diagnosed in the majority of trauma patients admitted to the ICU, anemia has been found to persist for the duration of their critical illness, and can last for weeks to months following the initial injury.^{2, 7} Other than treatment with blood transfusions, persistent injury-associated anemia has largely been refractory to exogenous erythropoietin administration and iron supplementation or transfusion. The pathophysiology of this persistent injury-associated anemia is not completely understood; however, several factors are thought to contribute. Persistent injury-associated anemia is associated with an increase in neuroendocrine activation and systemic inflammation.^8–10 In preclinical rodent studies of blunt trauma, hemorrhagic shock and chronic stress, post-injury hypercatecholaminemia is associated with a suppression of bone marrow erythroid progenitor growth, reduced erythroid commitment, an abnormal erythropoietin response, prolonged mobilization of hematopoietic progenitor cells and dysregulation of iron homeostasis.^11–12 Norepinephrine, a key modulator of hematopoietic progenitor mobilization and an inhibitor of bone marrow erythroid cell growth, also directly stimulates IL-6 release, and several of our prior studies have implicated these cytokines directly in the development of persistent injury-associated anemia.^13–14 Tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) are both pro-inflammatory cytokines that influence hematopoiesis and have altered expression following severe trauma.^15–18 IL-6 has been found to upregulate hepcidin expression, thereby leading to the sequestration of iron within macrophages, which can play a role in the persistent anemia seen in severely injured trauma patients, and it also activates secretion of C-reactive protein (CRP), which contributes to a systemic inflammatory state.^19–20 Dysfunctional erythropoiesis in a rodent trauma model can be ameliorated with chemical sympathectomy using either propranolol or clonidine, suggesting that a reduction of hypercatecholaminemia improves erythropoiesis.^{11–13, 21–26}

Prior research has evaluated the post-injury inflammatory state in severely injured trauma patients and demonstrated their erythropoietic dysfunction and iron dysregulation.²⁷ To validate our preclinical findings and to expand upon our prior study of the post injury inflammatory state in humans, the purpose of this study was to evaluate additional severely injured trauma patients and determine if there is a relationship between the ISS in trauma patients, and the degree of neuroendocrine activation and systemic inflammation that is related with persistent injury-associated anemia as well as to further characterize the bone marrow response to injury. To compare a continuum of neuroendocrine activation, we evaluated erythropoietic function in healthy controls, elective hip replacement patients and blunt trauma patients. We hypothesize that in blunt trauma patients, erythropoietic dysfunction occurs in direct proportion to the severity of anatomic injury.

Material and methods

Study Population

A prospective observational cohort study was performed to compare three different populations: severely injured blunt trauma patients (n=37), elective hip replacement patients (n=26) and healthy controls (n=8). These three groups were included in order to compare a transient stress response (elective hip replacement) and persistent hypercatecholaminemia (severe blunt trauma) to healthy controls. As reported by Loftus et. al.²⁷, trauma patients from a Level 1 trauma center were screened on admission for inclusion in the study. This study was registered at ClinicalTrials.gov () and approved by the University of Florida Institutional Review Board (IRB201601386). Written consent was obtained from the patient or their designated health care proxy, either prior to their orthopedic surgery or within 96 hours after surgery using a delayed consent waiver. Inclusion criteria for the blunt trauma cohort were: age ≥ 18, long bone or pelvic fractures requiring open reduction and internal fixation or closed reduction and percutaneous pin fixation, in addition, these patients had either hemorrhagic shock (defined by systolic blood pressure ≤ 90 mm Hg, or mean arterial pressure ≤ 65 mm Hg, or base deficit ≥ 5 meq/L or lactate ≥ 4 mmol/L) or an ISS ≥ 15. Exclusion criteria were: survival < 48 hours, incarceration, pregnancy, patients receiving chronic corticosteroids or immunosuppressive therapies, previous bone marrow transplantation, or end stage renal disease. Blunt trauma patients were then broken down into quartiles for retrospective analysis. To ensure similar N for each ISS quartile, the groups were defined as an ISS 4–13, ISS 14–20, ISS 21–26, and ISS 27–41.

The elective hip replacement cohort were age > 18 years and underwent an elective hip replacement with the same exclusion criteria as the trauma cohort. Healthy control bone marrow samples were purchased from American Type Culture Collection (ATCC). Control samples were obtained by iliac crest biopsy of four female and four male subjects, age 21–37 years.

Sample collection

As previously described²⁷, the bone marrow samples were collected during the scheduled open reduction and internal fixation of the hip or long bone fracture for the trauma and hip replacement patients, using an 11-gauge bone marrow biopsy needle (Covidien, Minneapolis, MN) at the fracture site. An investigator (TJL, ESM, CGA, JAS) was present in the operating room for collection of all bone marrow samples, which were obtained from the sterile field by an orthopedic surgeon (JEH, HKP, KKS, MP) to ensure consistency of the sample collection.

Blood samples were obtained from the trauma and hip replacement patients during the index operation of the hip or long bone fracture and collected in a fully filled heparinized 6ml tube (Becton Dickinson, Franklin Lakes, NJ). Blood samples from the same healthy control bone marrow donors were not available, so blood samples were drawn from age-matched healthy volunteers to serve as the control group for plasma cytokine analyses. Plasma was collected and stored as previously described.²⁷

Bone marrow analysis

Real-time polymerase chain reaction was performed to analyze bone marrow expression of inflammatory and erythropoiesis markers, including erythropoietin (EPO), the erythropoietin receptor (EPO-R), ferroportin, the transferrin receptor-1 (TFR-1) and transferrin. RNA was isolated from human bone marrow and reverse transcribed to cDNA, per our standard protocol.²⁷ Real-time polymerase chain reaction was performed using Brilliant II SYBR Green QPCR Master Mix (Agilent Technologies, Santa Clara, CA) and the Mx3005P qPCR System (Agilent Technologies), reported as mRNA fold change relative to healthy controls. Primers were designed using Primer3 Web software.

Plasma analysis

Plasma IL-6 (R&D Systems, Minneapolis, MN), TNF-alpha (R&D Systems, Minneapolis, MN), CRP (R&D Systems, Minneapolis, MN), norepinephrine (Abnova, Taiwan) and hepcidin (R&D Systems, Minneapolis, MN) was measured by enzyme linked immunosorbent assay. All samples were run in duplicate following the manufacturer’s protocol.

Statistical Analysis

Statistical analysis was performed in GraphPad Prism version 7.05 (GraphPad Software, La Jolla, CA). D’Agostino & Pearson test for normality was used to test normality in all variables, and one-way analysis of variance using Tukey’s multiple comparisons test was used for variables whose distribution was normal. Kruskal-Wallis non parametric test was used to analyze groups that were found to be skewed. Significance was set at α = 0.05 and data were reported as both mean ± standard deviation (SD) and median (IQR) for clinical variables. *p < 0.05 vs healthy controls and **p < 0.05 between groups.

Results

Clinical Data

Thirty-seven trauma patients were included in this study; twenty-one of these patients were male, fourteen are female. Ages ranged from 18 to 90 years, with a mean of 43 years. Median age for the trauma patients by ISS quartile was 40, 49, 48 and 35 years respectively. Twenty-six elective hip replacement patients were included in this study; nine of these patients were male, seventeen were female. Ages ranged from 32 to 80 years, with a median of 65 years. 35% of the elective hip replacement patients were male, whereas 57% of the trauma patients were male (Table 1).

Table 1:

Clinical Data

	Healthy Control	Hip Replacement	ISS 4–13	ISS 14–20	ISS 21–26	ISS 27–41
N	8	26	8	9	10	10
Age (years)	28 (24–34)**	65 (54–71)	40 (30–4)**	49 (39–62)	48 (35–55)**	35 (22–50)**
Initial Lactate (mg/dL)	n/a	n/a	2.4 (1.8–3.2)	3 (1.5–3.5)	3.8 (3–5.2)	3.2 (2.2–3.8)
Day of Surgery Hemoglobin (g/dL)	n/a	13.7 (13–14.6)	12.4 (11–13.5)	10.2 (8.1–11.4)**	8.2 (7.4–9.5)**	8.8 (7.9–10.5)**
# pre-op pRBC transfusion	n/a	0.0 ± 0.0	0 (0–1.5)	0 (0–1.5)	1.5 (0–2.3)**	2 (0–3.3)**
# total pRBC transfusion	n/a	0.0 ± 0.0	0 (0–1.5)	3.5 (0.5–6)**	3.5 (1–6)**	3.5 (2–6)**
HLOS (days)	n/a	2 ± 1	5.5 (4–9)**	14 (10–17)**	16 (10–24)**	13 (7–16)**
ICU LOS (days)	n/a	0.0 ± 0.0	0 (0–1.5)	4 (0.5–14)**	4 (3–17)**	6 (5–7)**

pRBC: packed red blood cells, HLOS: hospital length of stay, ICU: intensive care unit, ISS: injury severity score;

^**p<0.05 between groups

Bone marrow and plasma sample collection was on hospital day 0 for the elective hip replacement patients. From the time of admission to operative repair for the trauma patients, the median time to bone marrow and plasma sample collection was 2 days with an interquartile range of 1 day. The median time from admission to operative repair by ISS quartile group was as follows: ISS 4–13 – 1.5 days, ISS 14–20 – 2 days, ISS 21–26 – 2 days and ISS 27–41 – 3 days. Those patients with higher ISS had a slightly higher median time from admission to operative repair but this was not statistically significant.

Median hospital length of stay was significantly increased in all trauma ISS groups when compared to the hip replacement patients. The 27–41 ISS group had a significantly longer hospital length of stay when compared to the 4–13 ISS group, with a 98% increase (**p=0.01) (Table 1). Median ICU length of stay for the trauma patients in the 14–20, 21–26 and 27–41 ISS groups was also significantly increased when compared to the hip replacement group (Table 1). The median ICU length of stay was significantly higher in the 21–26 ISS group compared to the 4–13 ISS group (**p<0.01) (Table 1).

Hemoglobin on the day of surgery was significantly decreased in 14–20, 21–26 and 27–41 ISS groups when compared to hip replacement patients, with a 27%, 41% and 37% decrease respectively (**p<0.01 for all three groups) (Table 1). The 14–20, 21–26 and 27–41 ISS groups also all experienced a decrease in day of surgery hemoglobin when compared to the 4–13 ISS group, with a 21%, 36% and 32% decrease respectively (**p<0.01 for all three groups). The median day of surgery hemoglobin for 21–26 ISS group was also significantly decreased when compared to the 14–20 ISS group, with a 20% decrease (**p=0.03) (Table 1).

Packed red blood cell (pRBC) transfusions were collected and the hip replacement group received no preoperative or postoperative pRBC transfusions. The trauma patients received a median of 0, 0, 1.5 and 2 pRBC transfusions preoperatively for each 14–20, 21–26 and 27–41 ISS group respectively. The increase in preoperative transfusions in the 21–26 and 27–41 ISS group was statistically significant when compared to the hip replacement group (**p=0.02, **p<0.01 respectively) (Table 1). Regarding total pRBC transfusions for the hospital stay, the median number of pRBC transfusions was 0, 3.5, 3.5 and 3.5 for each14–20, 21–26 and 27–41 ISS group respectively.

There was a significant correlation between the highest ISS group and the severity of anemia on the day of surgery (Pearson r = −0.546, p=0.001). However, pRBC transfusions preoperatively (Pearson r = 0.241, p=0.157) and ICU length of stay (Pearson r = 0.315, p= 0.069) did not significantly correlate.

Inflammatory Cytokines

As ISS increases, plasma concentration of IL-6 increases significantly when compared to the healthy control patients (Figure 1A). The 27–41 ISS group was significantly increased with respect to the healthy control group (*p=0.03). Similarly, plasma IL-6 concentration increased significantly when compared to the elective hip replacement patients in the 14–20, 21–26 and 27–41 ISS groups, (**p=0.01, **p=0.01, **p<0.01 respectively) (Figure 1A).

Figure 1. Neuroendocrine activation and systemic inflammation worsened with increasing injury severity score.

1A. Plasma IL-6 is significantly higher in the 14–20, 21–26 and 27–41 ISS groups when compared to hip replacement. 1B. Plasma TNF-alpha is significantly higher in 27–41 ISS group when compared to hip replacement. 1C. Plasma CRP is significantly higher in all trauma groups when compared to both healthy controls and hip replacement. 1D. Plasma NE increased as ISS increased. Healthy control subjects have mean value represented by the dashed line. *p<0.05 vs. healthy controls, **p<0.05 between groups.

Plasma TNF-alpha concentration increased significantly in the 14–20 and 27–41 ISS groups when compared to the healthy control groups (*p=0.05, *p=0.02 respectively) (Figure 1B). Plasma TNF-alpha concentration also increased 151% in the 27–41 ISS group when compared to the elective hip replacement patients (**p=0.02) (Figure 1B).

All four ISS quartile groups significantly increased plasma CRP concentration when compared to both healthy controls and elective replacement repairs (Figure 1C). With respect to healthy controls, the plasma CRP concentration in the trauma patients increased 50–60% (Figure 1C). When compared to the elective hip replacement patients, the plasma CRP concentration in all four ISS quartile groups increased more than 100-fold (**p<0.01 for all groups).

Plasma NE concentration increased as ISS worsened, and this was significant when compared to healthy controls in the 21–26 and 27–41 ISS groups (*p=0.0015, *p=0.0002 respectively) (Figure 1D). These two groups also had a 193% and 229% increase in plasma NE concentration when compared to elective hip replacement (**p=0.0012, **p<0.0001 respectively). NE concentration was most elevated in the 27–41 ISS quartile when compared to the other trauma groups and the elective hip replacement group, with a mean of 44.1 ng/ml when compared to the mean of 13.4 ng/ml in the elective hip replacement population (Figure 1D).

Markers of Erythropoiesis

Bone marrow expression of EPO was significantly increased in hip replacement patients when compared to healthy controls (*p<0.0001) (Figure 2A). In the trauma patients, bone marrow expression of EPO decreased as ISS increased, with the lowest expression in the 27–41 ISS group. EPO expression in the 27–41 ISS group decreased significantly by 70% when compared to elective hip replacement patients (**p=0.0443) (Figure 2A).

Figure 2. Erythropoietic dysfunction following severe trauma.

2A. Bone marrow EPO expression increased following hip replacement. 2B. Bone marrow EPO-R expression is significantly increased following hip replacement and is significantly decreased in all trauma ISS groups. *p<0.05 vs. healthy controls, **p<0.05 between groups.

Likewise, bone marrow expression of EPO-R was significantly increased in hip replacement patients when compared to healthy controls (*p=0.0046) (Figure 2B). In the trauma patients, bone marrow expression of EPO-R decreased as ISS increased, and all four trauma ISS groups, 4–13, 14–20, 21–26, and 27–41, were significantly decreased when compared to hip replacement patients (**p<0.0001, **p=0.0343, **p=0.0012, **p=0.0005 respectively) (Figure 2B).

Markers of Iron Dysfunction

Plasma hepcidin concentration was significantly increased in all four trauma ISS groups, when compared to healthy controls (*p=0.0002, *p<0.0001, *p<0.0001, *p<0.0001 respectively) (Figure 3A). Plasma hepcidin concentration increased 300-fold in all four trauma ISS groups when compared to elective hip replacement patients (**p<0.0001 for all groups) (Figure 3A).

Figure 3. Iron dysfunction following severe trauma.

3A. Plasma hepcidin is significantly increased in all trauma ISS groups when compared to hip replacement and healthy controls. 3B. Bone marrow transferrin expression is significantly increased in all trauma ISS groups when compared to hip replacement and healthy controls. 3C. Bone marrow ferroportin expression is significantly increased in 21–26 and 27–41 ISS groups. 3D. Bone marrow transferrin receptor-1 expression decreased is significantly decreased in the 21–26 and 27–41 ISS groups when compared to hip replacement. Healthy control subjects have mean value represented by the dashed line. *p<0.05 vs. healthy controls, **p<0.05 between groups.

Bone marrow expression of transferrin in all four trauma ISS groups, 4–13, 14–20, 21–26, and 27–41, was significantly increased when compared to healthy controls and elective hip replacement patients (Figure 3B).

Bone marrow expression of ferroportin was significantly increased, with a 282% increase in the hip replacement group (*p<0.01), 326% increase in the 21–26 ISS group (*p=0.02), and 308% increase in the 27–41 ISS group (*p=0.02) when compared to the healthy controls (Figure 3C). The 4–13 ISS group had a 63% decrease in expression of ferroportin when compared to the elective hip replacement group (**p=0.03) (Figure 3C).

Bone marrow expression of transferrin receptor-1 (TfR-1) was significantly increased in hip replacement patients when compared to healthy controls (*p=0.0392) (Figure 3D). In contrast, bone marrow expression of TfR-1 decreased in the trauma groups as ISS increased, and was significantly decreased in the 21–26 and 27–41 ISS groups when compared to elective hip replacement patients, with a 78% (**p=0.0401) and 84% (**p=0.0407) decrease respectively (Figure 3D).

Discussion

Critically injured trauma patients with higher ISS developed severe systemic inflammation and neuroendocrine activation that correlated with the degree of erythropoietin dysfunction, iron dysregulation and anemia. In comparison, hip replacement patients exhibited minimal systemic inflammation and neuroendocrine activation with an appropriate upregulation of erythropoietin as well as molecular pathways for iron mobilization that was absent in severely injured trauma patients. Based on these findings, the study uniquely demonstrates a link between injury severity, neuroendocrine activation, systemic inflammation and the bone marrow response to anemia.

This study demonstrated that trauma patient plasma IL-6 concentration significantly increased with each ISS group. This is consistent with previous data, which has shown that IL-6 is a pro-inflammatory cytokine and a reliable marker for systemic inflammation in patients with moderate and severe trauma.¹⁶ In a rodent model of trauma and hemorrhagic shock with resuscitation, IL-6 levels were found to be significantly elevated when compared to sham rodents at 24 hours.²⁸ In addition, Qiao et al.²⁹ reported in their meta-analyses, demonstrating that IL-6 is an early marker for the development of complications in trauma and that patients with higher IL-6 levels are more likely to experience complications or death.²⁹ Also, trauma plasma TNF-alpha concentrations were significantly increased in the 21–26 and 27–41 ISS groups. Brouckaert et al.³⁰ observed that IL-6 circulating levels were significantly elevated in healthy humans infused with recombinant human TNF, demonstrating that TNF is a potent inducer of IL-6. Both increased TNF-alpha and IL-6 play a role in the propagation of the systemic inflammation that correlates with worsening ISS.

CRP is a generalized marker for inflammation and is activated by IL-6.³¹ Plasma CRP concentration was significantly elevated in all four trauma groups. Golabek-Dropiewska et al.²⁰ found that there was a correlation between elevated CRP level and the occurrence of infectious complications in severely injured trauma patients. While, Alper et al.³² demonstrated that mean CRP on admission was significantly higher in trauma patients when compared to controls, CRP did not directly correlate with ISS and therefore was not helpful in predicting injury severity or outcomes. In contrast, our data does show an overall trend of increasing CRP concentration with increasing ISS. This difference may be due to the timing of the samples since our samples were not taken on admission, rather on the day of orthopedic surgery.

After trauma, hypothalamic activation of the sympathetic nervous system has been shown to trigger NE secretion from the presynaptic nerve terminals and catecholamine discharge from the adrenal medulla.³² In a rodent model of lung contusion, hemorrhagic shock and chronic restraint stress, prolonged hypercatecholaminemia, specifically related to NE, was shown to be a key mediator of prolonged bone marrow dysfunction.²⁴ In this study, norepinephrine concentrations increased following trauma and correlated with increasing ISS groups. The 21–26 and 27–41 ISS groups had plasma NE concentrations that were significantly increased. These results help us predict that patients with higher ISS will likely have prolonged bone marrow dysfunction and chronic anemia.

Plasma hepcidin concentrations were increased in all ISS groups, and hepcidin has been shown to contribute to dysregulated iron trafficking after severe trauma, and its elevation has been shown to lead to a relative iron deficiency and it has been associated with anemia of inflammation.³³ Exogenous hepcidin injection in a murine model was associated with decreased iron bioavailability and iron-restricted erythropoiesis.³⁴ Bone marrow expression of ferroportin, the receptor for hepcidin, was increased in all trauma ISS groups when compared to healthy controls. Stimulated by increased IL-6, an increase in systemic hepcidin causes ferroportin to be internalized, thereby decreasing mobilization of iron stores.³³ We also found that the bone marrow expression of transferrin was increased significantly in all trauma ISS groups, despite its role as a negative acute phase reactant. Yet, bone marrow expression of TfR-1 was decreased following severe trauma, representing the rate-limiting component of iron uptake by erythroid progenitors.^{27, 35} Loftus et al.²⁷ postulated that for iron homeostasis and normal erythropoiesis, iron bioavailability must be accompanied by iron uptake in erythroid progenitors, which is facilitated by transferrin and its receptor via receptor-mediated endocytosis. Our findings in this study are consistent with the iron homeostasis process following trauma being interrupted both by decreased iron bioavailability due to an increase in hepcidin concentration as well as a decrease in transferrin receptor expression preventing iron uptake by the erythroid progenitors. In contrast, elective hip replacement patients lack an increase in hepcidin and have increased bone marrow expression of the transferrin receptor.

Bone marrow expression of EPO and EPO-R was significantly increased in hip replacement patients which is an appropriate response to anemia. The bone marrow expression of EPO did not increase in any of trauma patient ISS groups in response to anemia. In addition, there was a significant decrease in the bone marrow expression EPO-R in each trauma ISS group. The trauma patients in the 27–41 ISS group had the lowest expression of both EPO and EPO-R when compared to the other trauma groups and had the most severe anemia. This decreased bone marrow expression of EPO and EPO-R in critically ill trauma patients is consistent with previous studies, as Loftus et al.²⁷ described the down-regulation of the EPO-R expression in the trauma bone marrow in the post-injury inflammatory state that correlated with decreased growth erythroid progenitor cells. The increased plasma concentrations of hepcidin, seen in all trauma ISS groups, is associated with a blunted response in EPO and EPO-R bone marrow expression.

As ISS increased, so did the hospital and ICU length of stay, which correlated with anemia and the need for pRBC transfusions. This is consistent with what Kuo et al.³⁶ reported in their cross-sectional study comparing the ISS to the New Injury Severity Score. The role of cytokines in the inflammatory response is well described as they activate, modulate and inhibit inflammation in response to tissue damage but their role in the bone marrow response to anemia is not as well understood.²⁰

This study was limited by small sample sizes, data from a single institution, and the lack of serial biochemical peripheral blood analyses. Although sample sizes were small, the results suggest that enrollment was adequate. This study was also limited by the fact that we included trauma patients who received corticosteroids, experienced a traumatic brain injury (TBI), and/or were treated with beta blockers. When reviewing our data, we found that four patients (11%) were treated with corticosteroids and one was in each of the quartile groups. We found that five patients (14%) had a TBI but one was in the ISS 4–14, one in the 21–26 and two in the 27–41 group. Thirteen patients (35%) were treated with beta blockers during their hospital stay which included patients who received one dose. We acknowledge that these are potential confounders, however, we chose to include these patients because they were fairly equally distributed across each ISS quartile. We believe that the differences seen in the inflammatory markers and bone marrow dysfunction correlate with the graded ISS, however another limitation of our study is that we cannot exclude the role of age, transfusion, or the incidence of shock. Preoperative pRBC transfusion requirement increased by ISS quartile, which could influence the inflammatory markers as the pattern correlates. The age of the patients increased from ISS 4–13 quartile toEPO the 14–20 group, but then decreased in the next two groups until the ISS 27–41 group which had the lowest median age of all the groups, and the inflammatory cytokine and degree of bone marrow dysfunction did not correlate with this age pattern. Future multivariate analysis would be helpful to delineate potential roles of any confounding variables, when we have a larger sample size in each group.

Conclusions

In summary, this study further characterizes the relationship between the neuroendocrine response, systemic inflammation, hepcidin and bone marrow response to anemia following severe trauma as it correlated to injury severity. Elective hip replacement patients have only minimal systemic inflammation with an appropriate compensatory upregulation of the erythropoietin response and molecular pathway for iron mobilization which was absent in the trauma patients. Injury severity, neuroendocrine activation, and systemic inflammation are pivotal in the bone marrow response to anemia.