Trauma remains one of the leading causes of preventable deaths in the world (1). Most deaths that occur early after traumatic injury are due to hemorrhage or severe neurological injury (2). After this initial period, morbidity and mortality are largely due to the development of infections, thromboembolic complications, acute respiratory distress syndrome (ARDS), and multiple organ dysfunction syndrome (MODS).
In recent years, there has been increasing interest in the influence that sex may have on clinical outcomes after traumatic injury. Several epidemiological and prospective observational studies have examined this issue and have yielded conflicting results (2-8). In the studies that do find a sex-based difference in trauma outcomes, there seems to be a survival advantage conferred upon premenopausal women who sustain blunt trauma when compared against men or postmenopausal women (4, 5). In addition to a higher mortality rate, Mostafa et al. (6) have documented a higher incidence of MODS and longer intensive care unit (ICU) and hospital stays in male trauma patients. Similar observations were also made by Frink et al. (3), who observed sex-based differences in plasma cytokine levels.
Human clinical studies have also shown a sex dimorphism in coagulability. From the wealth of literature on postmenopausal hormone replacement therapy and hormone-based oral contraceptives, estrogens and progestins have been implicated in promoting coagulation (9-11). Additionally, partial estrogen antagonists have been shown to decrease fibrinogen levels and, thus, promote a less coagulable state (12). Using thromboelastographic (TEG) analysis, female trauma patients were found to be more hypercoagulable than male trauma patients on the first day after a traumatic injury (13).
The difference in clinical outcomes between men and women is not limited to trauma patients. An improved prognosis has also been noted in female sepsis patients (14). Although Schroder et al. noted a similar clinical sepsis course, they observed a mortality rate of 70% in men and 26% in women. In their study, they also noted that there seemed to be an intrinsic difference in the release of TNF-α between men and women, although they were not able to attribute this difference to sex hormones. Women have also been noted to have better outcomes than men after traumatic brain injury. This differential outcome has been hypothesized to be due to the differences in sex hormones that lead to a difference in the amount of secondary inflammation after brain injury (15). Similar outcome differences have also been noted in nontraumatic brain injury patients as well.
These sex-specific differences in morbidity and mortality from shock, trauma, and sepsis are highly suggestive that sex hormones play a key role in mediating these differences. This hypothesis has led to numerous experimental animal studies to examine the role that sex hormones play in the pathophysiology after traumatic injury (16-19). Data from these studies have suggested that male sex is a risk factor for death after hemorrhage and sepsis. Specifically, these studies have found that lower estrogen levels and/or higher testosterone levels lead to immunologic impairment. This impairment seems to be directly due to alteration of cell-mediated immunity and cytokine expression and is believed to lead to the development of ARDS and MOD (16).
Despite the wealth of information in animal studies on the roles of sex hormones in posttrauma pathogenesis, there remains no hormonal information in clinical trauma studies. We hypothesized that the sex dimorphism seen in previous clinical studies is due to sex hormones. The aim of the present study is to evaluate how the hormonal milieu in the immediate posttraumatic period influences coagulation, inflammation, and clinical outcomes.
MATERIALS AND METHODS
This was a prospective cohort study at one of two level-1 trauma centers in the state of Oregon. The Oregon Health & Science University Institutional Review Board approved the protocol. Trauma patients admitted to the ICU with a minimum injury severity score (ISS) of 4 were eligible for inclusion in this study. Patients whose initial blood collection could not be obtained within 24 h of injury and those with isolated head injury were excluded from this study. Other specific inclusion and exclusion criteria are outlined in Table 1. Informed consent was obtained from the patient or their surrogate medical decision maker.
The collection of clinical data and assay of standard laboratory studies, diagnosis and treatment of thromboembolic complications, and TEG analysis were described previously (13). Briefly, all clinical data, including age, sex, ISS, time to initial blood draw relative to time of injury, and complications, were collected prospectively for the duration of a patient's hospitalization. The initial blood sample was drawn within 24 h of the time of injury. As part of their routine care, all patients received daily physical examinations and once-daily laboratory assessments that included platelet counts, international normalized rations, partial thromboplastin times (PTTs), fibrinogen levels, and complete blood counts. Additionally, a 15-mL blood sample was obtained once-daily for the initial 4 days of the hospitalization for TEG, cytokine, and sex hormone analysis. These daily blood samples were obtained concurrently with the patient's routine morning blood draws for total blood sampling volume of approximately 25 mL/d.
The daily TEG assessments were performed using a 5000 series thromboelastograph hemostasis analyzer (Hemoscope Corporation, Niles, Ill). The four measured TEG parameters were r time (r), k time (k), angle (α), and maximum amplitude (MA). From these four variables, an overall coagulation index (CI) was calculated using the equation CI = −0.2454r + 0.0184k + 0.1655MA - 0.024α - 5.0220 (20). The r represents the time at which clot formation is first detected. The k represents the time at which the clot has achieved a defined level of firmness. α represents the rate of clot formation. k is shortened, and α is increased by increased fibrinogen level and increased platelet function. The MA is a measure of the maximum strength of the clot and is primarily determined by platelet function and fibrin cross-linking. The fifth measured TEG parameter is the LY30, which is defined as the reduction of the TEG amplitude 30 min after clot formation. It is considered to be a reflection of the rate of fibrinolysis.
In addition to the collection of standard patient demographics, the development of deep vein thrombosis (DVT), pulmonary embolism (PE), ARDS, and MODS was recorded for the duration that the patient was in hospital. The diagnosis of ARDS was determined clinically by a Pao2/Fio2 ≤ 200 and the presence of acute bilateral infiltrates without evidence of left atrial hypertension (21). Multiple organ dysfunction syndrome was defined by the modified Denver MODS score (22). Infections were defined by a positive bacterial culture from blood, sputum, or urine requiring appropriate antibiotic treatment. Cultures were routinely collected on febrile patients (temperature ≥ 38.5°C). It is our practice to routinely obtain sputum for culture by endotracheal tube aspirate, rather than bronchoalveolar lavage or protected brush. In persistently febrile patients without an obvious source of infection, cultures were obtained for 3 consecutive days.
Deep vein thrombosis prophylaxis consisted of sequential compression devices and subcutaneous prophylactic-dose Lovenox when no significant bleeding risk existed. Duplex ultrasound studies of the lower extremities were routinely performed weekly on all nonambulatory patients to screen for the development of DVTs. All patients clinically suspected of having a PE received a computed tomographic angiogram. All patients who were diagnosed with a PE or a clinically significant DVT were treated with therapeutic-dose intravenous heparin or therapeutic-dose Lovenox if not otherwise clinically contraindicated.
Plasma cytokine and hormone measurements
Platelet-poor plasma was derived from the collected whole blood samples by centrifugation and was stored at −80°C until they could be batch-analyzed, in duplicate, for cytokine and sex hormone measurement. The plasma cytokines were assayed using commercially available enzyme-linked immunosorbent assay kits (R & D Systems, Minneapolis, Minn): TNF-α, Quantikine HS; IL-6, Quantikine HS; IL-8, QuantiGlo. The plasma was assayed for estradiol, progesterone, and testosterone separately using commercially available enzyme immunoassay (EIA) kits (BioCheck, Inc., Foster City, Calif). These assays quantify free hormone levels in plasma; globulin-bound hormone is not detected with this methodology. The reported minimal detection limits of the EIA kits for these three steroids were 10 pg/mL, 0.3 ng/mL, and 0.05 ng/mL, respectively. Per the EIA kit manufacturer's literature, each assay has 100% reactivity with its targeted sex steroid and less than 3% cross-reactivity with all other steroids (including male and female sex steroids and their metabolites).
The TNF-α, IL-6, and sex hormone assays were read on a Dynatech Laboratory MRX microplate reader (Dynex Technologies, Inc., Chantilly, Va), whereas the IL-8 assay was done using a Fluoroskan Ascent FL microplate fluorimeter/luminometer (ThermoLab Systems, Helsinki, Finland).
Statistical analysis was performed with SPSS version 14 (SPSS Inc., Chicago, Ill) and Microsoft Excel 2003 (Microsoft Inc., Redmond, Wash). Categorical data were analyzed with either a chi-square test or Fisher exact test when the value in any cell was less than 5. Using Q-Q plot analysis, we determined that the cytokine and hormone data did not follow a Gaussian distribution; therefore, nonparametric tests were used. Independent continuous data were analyzed using the Mann-Whitney U test. The Wilcoxon matched-pair signed-rank test and the Friedman two-way ANOVA by ranks test were used for the comparison of multiple repeat measures. A two-tailed Spearman correlation coefficient was calculated to determine correlation between independent continuous variables. A two-tailed P value less than 0.05 was considered to be significant.
Sixty-two patients met entry criteria and were enrolled in the study. Eighteen of these patients were women (29%). The trauma population evaluated at our institution is historically 70% men and 30% women. Only one patient sustained a penetrating injury. There was no sex-based statistical difference in age, ISS, time to first blood draw, volume of fluid received, or volume of packed red blood cell transfusion. Forty-five patients had an ISS greater than 15. There was a proportionate distribution of men and women in both ISS strata. The patient demographics are summarized in Table 2.
Nineteen patients received chemical DVT prophylaxis during their entire hospitalization, but only 4 patients were started on chemical DVT prophylaxis during the 4-day study blood sampling period. Four patients were diagnosed with a thromboembolic event during their hospitalization. Three of these patients were diagnosed with a DVT, and the fourth had developed a PE. All four patients were men, and only one of the four patients who developed a DVT did not receive chemical DVT prophylaxis. None of the patients died as a result of their thromboembolic complication.
Of the 62 patients, 21 were intubated, and the mean duration of mechanical ventilation was 5.3 ± 4.9 days. Overall, 15 patients developed a bacterial infection. Ten of these patients were men (22.7% of men), and five were women (27.8% of women). Most of the infections were pneumonias (67%) but also included bacteremia, pelvic abscesses, and urinary tract infections (each 11%). All of the patients who developed pneumonia were intubated, with a mean duration of ventilation of 8 ± 6 days.
No patient met the criteria for MODS. Six patients (one woman) died during the course of this study. Three of the patients who died developed ARDS before death. Twelve patients developed ARDS (2 of whom were women). Eleven of the ARDS patients were intubated with a mean of 10.4 ± 9.9 days of mechanical ventilation. The 12th ARDS patient was successfully managed without the need for mechanical ventilation. Six patients who developed ARDS sustained injuries known to be risk factors for ARDS. Within these six patients, there were 5 long bone fractures and 4 patients requiring a massive transfusion (≥10 U packed red blood cells transfused with mean of 18.2 ± 4.3 U packed red blood cells).
Sex hormone secretion and injury severity
Plasma sex hormone levels are shown in Figure 1. All sex hormone levels are presented as median with the interquartile range (IQR) unless otherwise stated. Testosterone was found to be higher in men than in women throughout the 4-day period of our study. Testosterone was also noted to negatively correlate with age on all days (r = −0.384 to −0.467; P < 0.01). Estradiol levels were greater in women than men on all 4 days, but this difference reached statistical significance only on days 2 and 4. There was no statistical difference in progesterone levels during the study.
Estradiol and progesterone were noted to correlate with ISS on day 1 only (r = 0.272, P = 0.04; and r = 0.374, P = 0.003, respectively). As shown in Figure 2, patients with an ISS of 15 or less had persistently higher estradiol-to-progesterone (E2-Pr) and estrogen-to-testosterone (E2-T) ratios. Sex-based subgroup analysis of the hormone levels yielded similar results.
Correlation between sex hormones and coagulation
Estradiol levels on day 1 positively correlated with fibrinogen level (r = 0.460; P < 0.001), but this correlation was not present on subsequent days. Progesterone and testosterone, individually, did not have any correlations with any of the TEG parameters or with any of the standard coagulation laboratory studies. However, E2-Pr ratio correlated with several of the TEG parameters. As summarized in Table 3, the E2-Pr ratio, on day 1, was positively correlated with α, MA, and CI in both men and women together, and in men, it was also negatively correlated with r and k. There were no statistically significant correlations between E2-Pr and any of the TEG parameters on other study days. The correlations between E2-Pr and PTT are shown in Table 4. The ratio E2-Pr had no correlations with PTT on day 1, but was negatively correlated with PTT on study days 2 to 4. There were no significant correlations between hormone concentrations or hormone ratios with either international normalized ration or platelet counts.
Inflammatory serum cytokine response
Figure 3 shows TNF-α, IL-6, and IL-8 levels during the 4-day study period. There was no statistically significant change in TNF-α or IL-6 levels during the study. There was no correlation between ISS and TNF-α or IL-6 levels, but ISS was positively correlated with IL-8 on days 1 (r = 0.374; P = 0.007) and 2 (r = 0.400; P = 0.009). However, IL-8 was initially elevated and decreased rapidly after day 1 (P < 0.001) and remained at a constant level for the remaining 3 days. IL-8 levels were noted to be significantly elevated on all 4 days in patients who developed ARDS (P = 0.006 - 0.045). A similar IL-8 elevation was noted in patients who developed an infection, but this difference only persisted on days 1 (P = 0.006) and 2 (P = 0.014). However, in a subgroup analysis in which our cohort was divided by ventilator use, IL-8 was noted to be elevated only in those who required ventilator use and also had developed an infection (postinjury day 1; P = 0.035). There was no difference in TNF-α or IL-6 levels in those patients who developed ARDS or an infection.
Estradiol levels on days 1 and 2 negatively correlated with day 2 TNF-α levels (r = −0.464, P = 0.004; r = −0.356, P = 0.042, respectively). Subanalysis by sex showed no such correlation in men. In women, an even stronger correlation was seen between estradiol levels on days 1 and 2 and day 2 TNF-α levels (r = −0.929, P = 0.003; r = −0.952, P < 0.001, respectively). No similar correlations were found in men. No correlation was identified between estradiol, progesterone, testosterone, or their ratios with IL-6 or IL-8.
Despite the correlations observed between estradiol and TNF-α, there were no significant correlations between the sex hormone levels and leukocyte counts.
Sex hormones and clinical outcomes
As shown in Figure 4, testosterone was noted to be significantly elevated on day 1 in patients who developed ARDS (P = 0.045). A similar trend was noted when men and women were evaluated separately, but the differences did not reach statistical significance. As demonstrated in Figure 5, estradiol levels were noted to be higher on days 3 (P = 0.032) and 4 (P = 0.016) in those who developed either a DVT or a PE. No other statistically significant differences were noted.
Estradiol levels were noted to be higher in patients who developed an infection (Fig. 6); however, this difference reached statistical significance only on day 1 (P = 0.019), but no other significant differences were seen. However, when the analysis is repeated by excluding patients who developed a pneumonia (in this study, pneumonias only occurred in ventilated patients), the higher estradiol levels reached statistical significance on postinjury days 1 and 2 (P = 0.008 and P = 0.026, respectively) in those who developed an infection. There was no difference in the numbers of men and women in this subgroup analysis (P = 0.64).
There was no difference in estradiol levels when comparing survivors with nonsurvivors. However, both testosterone and progesterone (Fig. 7) were noted to be elevated on days 1 and 2 in those patients who did not survive.
The main finding of this study was the positive correlation of the E2-Pr ratio with TEG parameters and PTTs that favor hypercoagulability. Previous studies examining hormonal effects on coagulation have focused on the role of estradiol and have not examined the role of progesterone despite significant cross talk between the two receptor systems. Liganded progesterone receptors can quench, to differing degrees, the effects of estradiol-liganded estrogen receptors (23). Accordingly, this study notes that the balance between estradiol and progesterone seems to determine the degree hypercoagulability more so than each hormone independently.
The positive correlation between estradiol and plasma fibrinogen was also reflected by negative correlations with E2-Pr and the k and α TEG values, which is indicative of the fibrinogen concentration-dependent increase in clot propagation rate. The increase in fibrinogen in response to exogenous estrogen has been well described in the hormonal contraceptive literature (24). Additionally, use of raloxifene, a selective estrogen receptor modulator, has been shown to decrease fibrinogen levels (12).
The correlations with TEG parameters were significant only on postinjury day 1, and no similar correlations with TEG parameters were noted on postinjury days 2 to 4. There were, however, weak correlations between E2-Pr and PTT observed on subsequent days. By postinjury day 2, the coagulation and E2-Pr association is largely lost due to the normalization of the hypercoagulable state as previously described by Schreiber et al. (13).
Various studies have noted an increased rate of fibrinolysis in women who are using oral contraceptive pills and postmenopausal hormone replacement therapy (11, 24). This has been observed in formulations that contain only estrogens and with combination therapies. However, in our study, we did not observe any relationship between hormone levels and fibrinolysis (as reflected in the LY30). Sharma et al. (25) also failed to observe any changes in fibrinolysis in pregnant women who were found to be hypercoagulable. The increased fibrinolysis has been theorized to be due to a decrease in plasminogen-activator inhibitor type 1 (PAI-1) in response to estrogen therapy. This phenomenon has only been observed in those postmenopausal women with high basal levels of PAI-1 (11). If there are alterations to the PAI-1 levels in the trauma patient, it is not apparent in the fibrinolysis rate because of the heterogenous nature of the study population.
Estradiol levels were also negatively correlated with TNF-α levels early in the postinjury period. This correlation was noted only in women and did not seem to exist in men. Higher estradiol levels seemed to lead to lower TNF-α levels initially, but this trend was no longer present by postinjury day 3. This suppression of TNF-α is consistent with the sex-specific difference in ex vivo production of TNF-α by cultured peripheral blood lymphocytes obtained from trauma patients (26). A similar increase in TNF-α is also seen in the postmenopausal state (27). This estradiol-based attenuation of TNF-α production in the acute setting seems to be quite different from what is seen in chronic inflammatory diseases. In this setting, estradiol seems to increase the production of TNF-α (28).
Estradiol is classically thought to effect its actions via the estrogen receptor pathway and, subsequently, up-regulation of specific target genes with regulatory estrogen response elements (29, 30). However, estradiol has been shown to use nonclassical pathways to decrease TNF-α. The activated estrogen receptor homodimer antagonizes the binding of fos/jun heterodimers to the activating protein 1 (AP-1) response element by interacting with the heterodimer and by direct interaction with the AP-1 site to provide steric inhibition (31, 32). Our clinical findings, along with the other previously described data, suggest that this AP-1 site interaction predominates in the acute setting, but in the setting of chronic inflammation, this pathway may no longer be able to fully squelch TNF-α production.
Estrogens have been noted to have differing activity levels in different tissue types. This is believed to be due to various tissue types expressing different concentration of estrogen receptor subtypes and differing concentrations and types of coregulatory proteins. Complicating the picture is that coregulatory protein recruitment has been shown to be highly dependant upon the nature of the receptor ligand (33). By extension, we would expect that this phenomenon holds true for all nuclear receptors and their coregulator protein interactions. This would predict that gene products, under nuclear receptor regulation, would have differing degrees of expression depending on the tissue. Therefore, sex hormones, although active in numerous differing tissues, may have differing degrees of activity in different tissues.
We did not note any change in the IL-6 or IL-8 levels related to hormone levels, although previous studies have reported alterations in IL-6 due to estradiol exposure. This has been observed in both animal studies and cell culture studies. Elevated IL-6 levels have also been noted in men as compared with women (26). This difference is believed to be due to direct antagonism by the estradiol-estrogen receptor complex with nuclear factor κB at the IL-6 gene promoter site (27). Although it has not been as clearly studied, attenuation of IL-8 expression by estradiol likely occurs in a similar fashion with the antagonism of nuclear factor κB. It is not clear why this relationship between estradiol and IL-6 or IL-8 was not observed in our study. In studies evaluating the role of sex in the development of sepsis, researchers have noted that women had a lower rate of sepsis development than men. These same women were also noted to have a lower level of proinflammatory cytokines (especially IL-6) in the early posttraumatic period as compared with men (26).
The decreased proinflammatory cytokine load, due to estrogens, may lead to a lower risk of developing sepsis. However, this may predispose those with higher estrogen levels to the development of an infection, as was the case in our study. Patients who subsequently developed an infection during their hospitalization were also noted to have higher estradiol and IL-8 levels in the early posttraumatic period. However, we did not find any statistically different levels of TNF-α or IL-6 when comparing those who did develop and infection and those who did not. Christeff et al. noted that men who had developed sepsis had higher estrogen and lower testosterone levels as compared with similarly matched men who did not develop sepsis (34). Their study evaluated steroid levels after the development of sepsis and not the levels immediately after the primary insult. The hormonal milieu after the development of sepsis is likely to be a product of the compensatory state after sepsis develops, rather than a factor that predisposes or mitigates the development of sepsis.
In an analogous fashion, testosterone and IL-8 were noted to be higher in patients who developed ARDS. Although IL-8 levels remained higher throughout the study period, testosterone was higher only on postinjury day 1. This is again suggestive that it is the hormonal milieu at the time of injury that may predispose one to inflammatory complications such as ARDS and sepsis. The higher testosterone levels seen in those patients who developed ARDS are consistent with the observation that male patients with ARDS have a higher mortality rate when compared with their female counterparts (35). It is also consistent with the increased pulmonary dysfunction seen in males, but not females, after acute lung injury (36).
Patients with a thromboembolic complication were noted to have a higher estradiol level on days 1 and 2 as compared with patients who did not have any such complication. Although the E2-Pr ratio was the factor that correlated with coagulation, this hormone ratio was not correlated with the development of a thromboembolic event. It is not clearly evident as to why there is this discrepancy. It does, again, demonstrate that the sex hormone levels in the early posttrauma period are most associated with complications.
Although we are able to discern associations and differences in the sex hormone levels in the early posttrauma period, these levels have already been altered as part of the acute phase response to injury. Scheingraber et al. found that sex hormone levels are altered within 12 h after a major elective abdominal operation. Thus, one would expect a similar temporal relationship in nonelective injuries (37). This is a possible explanation as to why we do not see a difference in progesterone levels between men and women as well as why we see the correlation between ISS and estradiol levels.
Although our study is unable to define a causal relationship between sex hormone levels and coagulation and inflammation, it does provide further support for such a relationship. Despite being a relatively small study, we are able to clearly show a correlation between coagulation and early posttraumatic sex hormone levels. The women in this study were analyzed as a heterogeneous group as menopausal status and exogenous hormone use data were not collected, and we were not able to control for this. The small study size limited our ability to subanalyze our data by age which is a known modifier of sex hormone levels. However, the study size and small number of thromboembolic and inflammatory complications do limit our ability to conclusively define the exact relationship between sex hormones and outcomes after trauma.
Further research, with a larger cohort of trauma patients, will be able to better establish a clear relationship between sex hormones and outcomes. Additionally, further study is needed to delineate the molecular biology and biochemistry behind the physiological manifestations of the sex hormones. Understanding these pathways will facilitate the development of novel sex-based therapies for sepsis and ARDS.
1. Choudhry MA, Schwacha MG, Hubbard WJ, Kerby JD, Rue LW, Bland KI, Chaudry IH: Gender differences in acute response to trauma-hemorrhage. Shock
24(Suppl 1):101-106, 2005.
2. Peng R, Chang C, Gilmore D, Bongard F: Epidemiology of immediate and early trauma deaths at an urban Level I trauma center. Am Surg
3. Frink M, Pape HC, van Griensven M, Krettek C, Chaudry IH, Hildebrand F: Influence of sex and age on MODS and cytokines after multiple injuries. Shock
4. George RL, McGwin G Jr, Metzger J, Chaudry IH, Rue LW III: The association between gender and mortality among trauma patients as modified by age. J Trauma
5. George RL, McGwin G Jr, Windham ST, Melton SM, Metzger J, Chaudry IH, Rue LW III: Age-related gender differential in outcome after blunt or penetrating trauma. Shock
6. Mostafa G, Huynh T, Sing RF, Miles WS, Norton HJ, Thomason MH: Gender-related outcomes in trauma. J Trauma
53:430-434, 2002. discussion 434-435.
7. Napolitano LM, Greco ME, Rodriguez A, Kufera JA, West RS, Scalea TM: Gender differences in adverse outcomes after blunt trauma. J Trauma
8. Wohltmann CD, Franklin GA, Boaz PW, Luchette FA, Kearney PA, Richardson JD, Spain DA: A multicenter evaluation of whether gender dimorphism affects survival after trauma. Am J Surg
9. Gorton HJ, Warren ER, Simpson NA, Lyons GR, Columb MO: Thromboelastography identifies sex-related differences in coagulation. Anesth Analg
10. Harnett MJ, Bhavani-Shankar K, Datta S, Tsen LC: In vitro fertilization-induced alterations in coagulation and fibrinolysis as measured by thromboelastography. Anesth Analg
95:1063-1066, 2002, table of contents.
11. Koh KK, Mincemoyer R, Bui MN, Csako G, Pucino F, Guetta V, Waclawiw M, Cannon RO III: Effects of hormone-replacement therapy on fibrinolysis in postmenopausal women. N Engl J Med
12. Walsh BW, Kuller LH, Wild RA, Paul S, Farmer M, Lawrence JB, Shah AS, Anderson PW: Effects of raloxifene on serum lipids and coagulation factors in healthy postmenopausal women. JAMA
13. Schreiber MA, Differding J, Thorborg P, Mayberry JC, Mullins RJ: Hypercoagulability is most prevalent early after injury and in female patients. J Trauma
58:475-480, 2005. discussion 480-481.
14. Schroder J, Kahlke V, Staubach KH, Zabel P, Stuber F: Gender differences in human sepsis. Arch Surg
15. Bayir H, Marion DW, Puccio AM, Wisniewski SR, Janesko KL, Clark RS, Kochanek PM: Marked gender effect on lipid peroxidation after severe traumatic brain injury in adult patients. J Neurotrauma
16. Ananthakrishnan P, Cohen DB, Xu DZ, Lu Q, Feketeova E, Deitch EA: Sex hormones modulate distant organ injury in both a trauma/hemorrhagic shock model and a burn model. Surgery
17. Jarrar D, Wang P, Cioffi WG, Bland KI, Chaudry IH: The female reproductive cycle is an important variable in the response to trauma-hemorrhage. Am J Physiol Heart Circ Physiol
18. Kovacs EJ, Faunce DE, Messingham KA: Ethanol and burn injury: estrogen modulation of immunity. Alcohol
19. Krausz MM, Bashenko Y, Hirsh M: Improved survival in uncontrolled hemorrhagic shock induced by massive splenic injury in the proestrus phase of the reproductive cycle in the female rat. Shock
20. Salooja N, Perry DJ: Thrombelastography. Blood Coagul Fibrinolysis
21. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris A, Spragg R: The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med
22. Ciesla DJ, Moore EE, Johnson JL, Sauaia A, Cothren CC, Moore JB, Burch JM: Multiple organ dysfunction during resuscitation is not postinjury multiple organ failure. Arch Surg
139:590-594, 2004. discussion 594-595.
23. Katzenellenbogen BS: Mechanisms of action and cross-talk between estrogen receptor and progesterone receptor pathways. J Soc Gynecol Investig
24. Prasad RN, Koh S, Ratnam SS: Effects of three types of combined O.C. pills on blood coagulation, fibrinolysis and platelet function. Contraception
25. Sharma SK, Philip J, Wiley J: Thromboelastographic changes in healthy parturients and postpartum women. Anesth Analg
26. Majetschak M, Christensen B, Obertacke U, Waydhas C, Schindler AE, Nast-Kolb D, Schade FU: Sex differences in posttraumatic cytokine release of endotoxin-stimulated whole blood: relationship to the development of severe sepsis. J Trauma
48:832-839, 2000. discussion 839-840.
27. Pfeilschifter J, Koditz R, Pfohl M, Schatz H: Changes in proinflammatory cytokine activity after menopause. Endocr Rev
28. Janele D, Lang T, Capellino S, Cutolo M, Da Silva JA, Straub RH: Effects of testosterone, 17beta-estradiol, and downstream estrogens on cytokine secretion from human leukocytes in the presence and absence of cortisol. Ann N Y Acad Sci
29. Katzenellenbogen BS, Katzenellenbogen JA: Estrogen receptor transcription and transactivation: estrogen receptor alpha and estrogen receptor beta: regulation by selective estrogen receptor modulators and importance in breast cancer. Breast Cancer Res
30. Katzenellenbogen BS, Montano MM, Ediger TR, Sun J, Ekena K, Lazennec G, Martini PG, McInerney EM, Delage-Mourroux R, Weis K, et al.: Estrogen receptors: selective ligands, partners, and distinctive pharmacology. Recent Prog Horm Res
55:163-193, 2000. discussion 194-195.
31. Jakacka M, Ito M, Weiss J, Chien PY, Gehm BD, Jameson JL: Estrogen receptor binding to DNA is not required for its activity through the nonclassical AP1 pathway. J Biol Chem
32. Srivastava S, Weitzmann MN, Cenci S, Ross FP, Adler S, Pacifici R: Estrogen decreases TNF gene expression by blocking JNK activity and the resulting production of c-Jun and JunD. J Clin Invest
33. Gee AC, Carlson KE, Martini PG, Katzenellenbogen BS, Katzenellenbogen JA: Coactivator peptides have a differential stabilizing effect on the binding of estrogens and antiestrogens with the estrogen receptor. Mol Endocrinol
34. Christeff N, Benassayag C, Carli-Vielle C, Carli A, Nunez EA: Elevated oestrogen and reduced testosterone levels in the serum of male septic shock patients. J Steroid Biochem
35. Moss M, Mannino DM: Race and gender differences in acute respiratory distress syndrome deaths in the United States: an analysis of multiple-cause mortality data (1979-1996). Crit Care Med
36. Tsai BM, Wang M, Pitcher JM, Kher A, Brown JW, Meldrum DR: Endothelium-dependent pulmonary artery vasorelaxation is dysfunctional in males but not females after acute lung injury. Surgery
37. Scheingraber S, Dobbert D, Schmiedel P, Seliger E, Dralle H: Gender-specific differences in sex hormones and cytokines in patients undergoing major abdominal surgery. Surg Today