Delay in ICU transfer is protective against ICU readmission in trauma patients: a naturally controlled experiment.

BACKGROUND: Unplanned intensive care unit (ICU) readmission-ICU bounce back (ICUbb)-is associated with worse outcomes. Patients not requiring organ system support or intensive nursing are deemed 'ICU discharge ready' and transfer orders are placed. However, actual transfer only occurs when an appropriate, non-ICU bed is available. This is dependent on inherent system inefficiencies resulting in a naturally controlled experiment between when patients actually transfer: Early (<24 hours) or Delayed (>24 hours) transfers, after order placement. This study leverages that natural experiment to determine if additional ICU time is protective against ICUbb. We hypothesize that Delayed transfer is protective against ICUbb. METHODS: Using a retrospective, cohort design, we queried a trauma research repository and electronic medical record during a 10-year period to capture traumatized patients admitted to the ICU. Patients were categorized into Early (<24 hours) or Unintended-Delayed (>24 hours) groups based on actual transfer time after order placement. Patient characteristics (age, Charlson Comorbidity Index (CCI)) and Injury Severity Score (ISS) were analyzed. Univariate and multivariate analyses were performed to compare ICUbb rates among Early and Unintended-Delayed groups. RESULTS: Of the 2004 patients who met the criteria, 1690 fell into the Early group, and 314 fell into the Delayed. The Early group was younger (mean age 52±23 vs. 55±22 years), had fewer comorbidities (median CCI score 1 (0, 3) vs. 2 (1, 3)), and was less injured (median ISS 17 (10-22) vs. 17 (13-25)), all p<0.05. Overall, 113 (5.6%) patients experienced ICUbb: Early 109 (6.5%) versus Unintended-Delay 4 (1.3%), p<0.05. By regression analysis, age, CCI, and ISS were independently associated with ICUbb while Delayed transfer was protective. DISCUSSION: Despite higher age, CCI score, and ISS, the Unintended-Delayed group experienced fewer ICUbb. After controlling for age, CCI and ISS, Delayed transfer reduced ICUbb risk by 78%. Specific care elements affording this protection remain to be elucidated. LEVEL OF EVIDENCE: Level III. STUDY TYPE: Therapeutic study.

Effect of ascorbic acid concentrations on hemodynamics and inflammation following lyophilized plasma transfusion.

BACKGROUND: Compared with lyophilized plasma (LP) buffered with other acids, LP with ascorbic acid (AA) attenuates systemic inflammation and DNA damage in a combat relevant polytrauma swine model. We hypothesize that increasing concentrations of AA in transfused LP will be safe, will be hemodynamically well tolerated, and will attenuate systemic inflammation following polytraumatic injury and hemorrhage in swine. METHODS: This prospective, randomized, blinded study involved 52 female swine. Forty animals were subjected to our validated polytrauma model and resuscitated with LP. Baseline control sham (n = 6), operative control sham (n = 6), low-AA (n = 10), medium-AA (n = 10), high-AA (n = 10) groups, and a hydrochloric acid control (HCL, n = 10) were randomized. Hemodynamics, thrombelastography, and blood chemistries were assessed. Inflammatory cytokines (tumor necrosis factor α, interleukin 6 [IL-6], C-reactive protein, and IL-10) and DNA damage were measured at baseline, 2 hours, and 4 hours after liver injury. Significance was set at p < 0.05, with a Bonferroni correction for multiple comparisons. RESULTS: Hemodynamics, shock, and blood loss were similar between groups. All animals had robust procoagulant activity 2 hours following liver injury. Inflammation was similar between groups at baseline, and AA groups remained similar to HCL following liver injury. IL-6 and tumor necrosis factor α were increased at 2 hours and 4 hours compared with baseline within all groups (p < 0.008). DNA damage increased at 2 hours compared with baseline in all groups (p < 0.017) and further increased at 4 hours compared with baseline in HCL, low-, and high-AA groups (p < 0.005). C-reactive protein was similar between and within groups. IL-10 increased at 2 hours compared with baseline in low- and high-AA groups and remained elevated at 4 hours compared with baseline in the low-AA group (all, p < 0.017). CONCLUSION: Concentrations of AA were well tolerated and did not diminish the procoagulant activity of LP. Within our tested range of concentrations, AA can safely be used to buffer LP.

Reconstitution fluid type does not affect pulmonary inflammation or DNA damage following infusion of lyophilized plasma.

BACKGROUND: Dysfunctional inflammation following traumatic hemorrhage can lead to multiple-organ failure and death. In our polytrauma swine model, lyophilized plasma (LP) reconstituted with sterile water and ascorbic acid suppressed systemic inflammation and attenuated DNA damage. However, it remains unknown whether the inflammatory response is affected by the type of fluid used to reconstitute LP. We hypothesized that common resuscitation fluids such as normal saline (LP-NS), lactated Ringer's solution (LP-LR), Hextend (LP-HX), or sterile water (LP-SW) would yield similar inflammation profiles and DNA damage following LP reconstitution and transfusion. METHODS: This was a randomized, prospective, blinded animal study. LP was reconstituted to 50% of original volume with NS, LR, HX, or SW buffered with 15-mM ascorbic acid. Forty swine were subjected to a validated model of polytrauma, hemorrhagic shock, and Grade V liver injury and resuscitated with LP. Serum interleukin 6 (IL-6), IL-10, plasma C-reactive protein, and 8-hydroxy-2-deoxyguanosine concentrations were assessed for systemic inflammation and DNA damage at baseline, 2 hours, and 4 hours following liver injury. Lung inflammation was evaluated by Real Time Polymerize Chain Reaction (RT-PCR). RESULTS: Reconstituted LP pH was similar between groups before resuscitation. IL-6 and IL-10 increased at 2 hours and 4 hours compared with baseline in all groups (p < 0.017). DNA damage increased at 2 hours and 4 hours compared with baseline and from 2 hours to 4 hours in the LP-NS, LP-LR, and LP-SW groups (all p < 0.017). Animals resuscitated with LP-HX not only demonstrated increased DNA damage at 4 hours versus baseline but also had the lowest C-reactive protein level at 2 hours and 4-hours (p < 0.017). Overall, differences between groups were similar for DNA damage and lung inflammation. CONCLUSION: Reconstitution fluid type does not affect inflammatory cytokine profiles or DNA damage following LP transfusion in this swine polytrauma model. Based on universal availability, these data suggest that sterile water is the most logical choice for LP reconstitution in humans. LEVEL OF EVIDENCE: Prognostic, level II.

Traumatic brain injury is not associated with coagulopathy out of proportion to injury in other body regions.

BACKGROUND: Coagulopathy following trauma is associated with poor outcomes. Traumatic brain injury has been associated with coagulopathy out of proportion to other body regions. We hypothesized that injury severity and shock determine coagulopathy independent of body region injured. METHODS: We performed a prospective, multicenter observational study at three Level 1 trauma centers. Conventional coagulation tests (CCTs) and rapid thrombelastography (r-TEG) were used. Admission vital signs, base deficit (BD), CCTs, and r-TEG data were collected. The Abbreviated Injury Scale (AIS) score and Injury Severity Score (ISS) were obtained. Severe injury was defined as AIS score greater than or equal to 3 for each body region. Patients were grouped according to their dominant AIS region of injury. Dominant region of injury was defined as the single region with the highest AIS score. Patients with two or more regions with the same greatest AIS score and patients without a region with an AIS score greater than or equal to 3 were excluded. Coagulation parameters were compared between the dominant AIS region. Significant hypoperfusion was defined as BD greater than or equal to 6. RESULTS: Of the 795 patients enrolled, 462 met criteria for grouping by dominant AIS region. Patients were predominantly white (59%), were male (75%), experienced blunt trauma (71%), and had a median ISS of 25 (interquartile range, 14-29). Patients with BD greater than or equal to 6 (n = 110) were hypocoagulable by CCT and r-TEG compared with patients with BD less than 6 (n = 223). Patients grouped by dominant AIS region showed no significant differences for any r-TEG or CCT parameter. Patients with BD greater than or equal to 6 demonstrated no difference in any r-TEG or CCT parameter between dominant AIS regions. CONCLUSION: Coagulopathy results from a combination of tissue injury and shock independent of the dominant region of injury. With the use of AIS as a measure of injury severity, traumatic brain injury was not independently associated with more profound coagulopathy. LEVEL OF EVIDENCE: Epidemiologic study, level III.

A predictive model of early mortality in trauma patients.

BACKGROUND: Rapid thrombelastography (rTEG) is a real-time whole-blood viscoelastic coagulation assay. We hypothesized that admission rTEG and clinical data are independent predictors of trauma-related mortality. METHODS: Prospective observational data (patient demographics, admission vital signs, laboratory studies, and injury characteristics) from trauma patients enrolled within 6 hours of injury were collected. Mann-Whitney U test and analysis of variance test assessed significance (P ≤ .05). Logistic regression analyses determined the association of the studied variables with 24-hour mortality. RESULTS: Seven hundred ninety-five trauma patients were enrolled, of which 55 died within 24 hours of admission. Admission variables which independently predicted 24-hour mortality were as follows: Glasgow Coma Scale ≤8, hemoglobin <11 g/dL, international normalized ratio >1.5, Ly30 >8%, and penetrating injury (P < .05). This 5-variable model's area under the receiver operator characteristic curve was .88. The Hosmer-Lemeshow goodness-of-fit test was .90. CONCLUSIONS: This 5-variable model provides a rapid prediction of 24-hour mortality. The inclusion of rTEG Ly30 demonstrates the association of fibrinolysis with outcome and may support the early use of antifibrinolytic therapies.

Comparison of the hemostatic efficacy of low-volume lyophilized plasma reconstituted using sterile water, lactated Ringer's, normal saline, and Hextend solutions.

BACKGROUND: Low-volume ascorbic acid-buffered reconstituted lyophilized plasma (LP) provides logistic advantages, reduces the risks for large-volume resuscitation, modulates inflammation, and is equally effective for hemostatic resuscitation as full-volume LP. We compared the physiologic effects of resuscitation using LP reconstituted with sterile water (LP-SW), lactated Ringer's solution (LP-LR), normal saline (LP-NS), and Hextend (LP-Hx). METHODS: Plasma was collected from swine, lyophilized, and then reconstituted into four test solutions: LP-SW, LP-LR, LP-NS, or LP-Hx. Forty swine were anesthetized and subjected to a validated model of polytrauma and hemorrhagic shock (including a Grade V liver injury), then randomized to receive one of the four test solutions. Physiologic parameters, blood loss, lactate, and hematocrit were followed up. Coagulation status was evaluated using thrombelastography. Inflammatory mediator expression was evaluated by multiplex serum assay. RESULTS: Forty animals were included in the study (10 animals per group). One animal died following LP-Hx resuscitation. There was less blood loss in the LP-SW and LP-LR groups compared with the LP-NS and LP-Hx groups (p < 0.05). The LP-SW group exhibited less early coagulopathic changes by thrombelastography, and the LP-Hx group had persistently elevated international normalized ratios at the end of the study period (p < 0.05). Serum interleukin 6 was lower after 4 hours in the LP-SW group compared with LP-NS (p < 0.05). CONCLUSION: Resuscitation using low-volume LP-SW and LP-LR buffered with ascorbic acid confers an anti-inflammatory benefit and results in less blood loss. Sterile water is a safe, cost-effective, and universally available fluid for creating a low-volume hemostatic LP resuscitation solution.

Pericardiocentesis in trauma: a systematic review.

BACKGROUND: Pericardiocentesis (PCC) had been taught as a mandatory skill in the Advanced Trauma Life Support (ATLS®) course as a bridge to definitive surgical therapy for traumatic pericardial tamponade since its inception in 1978. Immediate thoracotomy for penetrating trauma to the heart and chest has resulted in the decreased use of PCC in trauma. PCC is now offered as an optional skill in the ninth edition of the ATLS®. A review of the literature regarding the use and effectiveness of PCC in traumatic pericardial tamponade in the modern era is necessary to better define its current role in trauma care. METHODS: Scientific publications from 1970 to 2010 involving PCC after trauma were identified. The Preferred Reporting Items for Systematic reviews and Meta-Analyses was used. Human studies describing acute traumatic tamponade were included. Publications involving nontraumatic or chronic pericardial tamponade from effusions caused by inflammatory, infectious, or neoplastic etiology were excluded. Publications were categorized by level of evidence. RESULTS: Of the 135 publications identified, 27 were included, composing of 2,094 trauma patients with suspected cardiac tamponade. The reported use of PCC decreased from 45.9% of patients in the period 1970 to 1979 down to 6.4% of patients in the period between 2000 and 2010 (p < 0.05). Reported rates describing the use of PCC as the sole intervention decreased from 13.7% in the period 1970 to 1979 to 2.1% in the period 2000 to 2010 (p < 0.05). Survival analysis after PCC was possible for 380 patients. Overall survival following PCC was 83.4% (n = 317) and 91.8% (n = 145) when used as the sole intervention. In patients who received PCC then thoracotomy, survival rate was 79.5% (n = 178). CONCLUSION: Studies on the use of PCC for trauma are limited and biased toward survivors. The reported survival rate is high. There remains a limited role for PCC in nontrauma centers where definitive surgical management is not immediately available and transport time to a higher level of care facility supports the use of temporary decompression by PCC. LEVEL OF EVIDENCE: Systematic review, level III.

Hyperosmolar reconstituted lyophilized plasma is an effective low-volume hemostatic resuscitation fluid for trauma.

BACKGROUND: We performed this study to optimize reconstituted lyophilized plasma (LP) into a minimal volume fluid that provides effective hemostatic resuscitation for trauma while minimizing logistical limitations. METHODS: We performed a prospective, blinded animal study. Plasma was lyophilized following whole blood collection from anesthetized swine. The minimal volume needed for reconstitution was determined, and this solution was evaluated for safe infusion into the swine. Reconstituted LP was analyzed for electrolyte content, osmolarity, and coagulation factor activity. Twenty swine were anesthetized and subjected to a validated model of polytrauma and hemorrhagic shock (including a Grade V liver injury), then randomized to resuscitation with LP reconstituted to either 100% of the original plasma volume (100%LP) or the minimal volume LP fluid. Physiologic data were monitored, and blood loss and hematocrit were measured. Coagulation status was evaluated using thrombelastography. RESULTS: The minimal volume of reconstituted LP safe for infusion in swine was 50% of the original plasma volume (50%LP). The 50%LP had higher electrolyte concentrations, osmolarity, and increased coagulation factor activity levels by volume compared with 100%LP (p < 0.05). Blood loss, hematocrit, mean arterial pressure, and heart rate did not differ between animals receiving 100%LP (n = 10) or 50%LP (n = 10) at any time point (p > 0.05). International normalized ratio and thrombelastography parameters were not different between groups (R time, α angle, or maximal amplitude, p > 0.05). CONCLUSION: Resuscitation with 50%LP fluid was well tolerated and equally effective compared with 100%LP, with respect to physiologic and hemostatic properties. The smaller volume of fluid necessary to reconstitute hypertonic LP makes it logistically superior to 100%LP for first responders and may reduce adverse effects of large-volume resuscitation.

Correlation of conventional thrombelastography and rapid thrombelastography in trauma.

BACKGROUND: Conventional thrombelastography has been in use for over 6 decades and provides a functional assay of coagulation. Rapid thrombelastography was developed to provide more rapid comprehensive analysis of coagulation status in an emergency setting. The purpose of this study was to determine the correlation of rapid thrombelastographic values with conventional thrombelastographic values in trauma patients. METHODS: We performed a prospective study on trauma patients at a university level 1 trauma center. Conventional thrombelastography and rapid thrombelastography were performed on 190 consecutive major trauma patients upon admission between 2010 and 2012. Conventional thrombelastographic and rapid thrombelastographic parameters were analyzed using bivariate analysis with Pearson correlation. Group comparisons were performed using the Mann-Whitney U test. RESULTS: Patients were predominantly male (71.6%, P < .05) with a median Injury Severity Score of 17 (range 10 to 29) and a median age of 43 years (range 29 to 53 years). There were significantly more patients with blunt trauma than penetrating trauma (72% vs 28%, P < .05). There was a strong correlation between the rapid thrombelastographic and conventional thrombelastographic maximal amplitude value, which represents platelet function (r = .80). There was a moderate correlation between the G (overall clot strength, r = .70), k (speed of clot formation, r = .66), and α-angle (r = .38), which reflects the degree of fibrin cross-linking. Lysis at 30 minutes correlated poorly (r = .19). CONCLUSIONS: Overall, there is a strong correlation between rapid thrombelastography and conventional thrombelastography in terms of overall clot strength and platelet function. There is a moderate correlation in assessing the degree of fibrin cross-linking and a poor correlation in evaluating thrombolysis. These correlations should be considered when evaluating coagulation status using rapid thrombelastography.

The effects of location and low-molecular-weight heparin administration on deep vein thrombosis outcomes in trauma patients.

BACKGROUND: Deep vein thrombosis (DVT) is common after trauma. Pulmonary embolism (PE) is a feared complication of DVT. Standard doses of low-molecular-weight heparin (LMWH) are commonly used to prevent and treat DVT and PE. There is variable bioavailability of LMWH with standard therapy. The traditional concept that below-knee DVT is associated with a lower risk of subsequent PE may result in less aggressive therapy. The purposes of this study were to assess the rates of PE in above-knee versus below-knee DVT and longitudinally evaluate outcomes of DVT treated with LMWH therapy. METHODS: This was a retrospective review of patients at a university Level I trauma center during the years 2005 through 2010. Patients diagnosed with lower-extremity DVT were included in this study. Patients were classified by location of lower-extremity DVT and type of LMWH therapy received. All high-risk trauma patients were evaluated with weekly duplex Doppler ultrasonography. All duplex studies were reviewed for DVT resolution or improvement. Symptomatic patients were evaluated with computed tomographic angiography to rule out PE. Demographics, total length of hospital stay, length of intensive care unit stay, and Injury Severity Score (ISS) were collected. RESULTS: Three-hundred eight trauma patients with lower-extremity DVT were included. More patients developed below-knee DVT (65.6%) compared with above-knee DVT (34.4%). Increased length of hospital stay, intensive care unit stay, and higher ISS were noted in patients with above-knee DVT. More patients had below-knee DVT in the prophylactic dosing group. With LMWH therapy, three PEs occurred in patients in the prophylactic dosing group with below-knee DVT, and no PEs occurred in the therapeutic treatment group. The incidence of PE between patients with below-knee DVT compared with above-knee DVT overall was not different (3.3% and 4.7%, p = 0.59). To assess DVT outcomes, 157 of the 308 patients had serial duplex studies following diagnosis of lower-extremity DVT. The number of patients receiving either therapeutic or prophylactic LMWH was similar (51% and 49%). There was no difference in rates of resolution or improvement between LMWH dosing groups or location of DVT. CONCLUSION: In screened trauma patients, below-knee DVT is more common than above-knee DVT. There was no difference in the incidence of PE between groups. Standard prophylactic and therapeutic dosing of LMWH does not affect the rates of resolution or improvement of lower-extremity DVT. Rates of resolution and improvement of DVT is not dependent of location of lower-extremity DVT. LEVEL OF EVIDENCE: Therapeutic study, level IV; epidemiologic study, level III.

The use of lyophilized plasma in a severe multi-injury pig model.

BACKGROUND: Shock and severe tissue injury lead to an endogenous coagulopathy mediated by activation of Protein C and hyperfibrinolysis known as acute traumatic coagulopathy. Together, hemodilution, acidosis, inflammation, and hypothermia result in a global trauma-induced coagulopathy. Coagulopathy in trauma is associated with mortality. Early and effective hemostatic resuscitation is critical in restoring perfusion, correcting coagulopathy, and saving lives in exsanguinating trauma. Lyophilized plasma (LP) provides a logistically superior alternative to fresh frozen plasma (FFP). STUDY DESIGN AND METHODS: Plasma was lyophilized following whole blood collection from anesthetized swine. A series of studies were performed using anesthetized swine subjected to a validated model of polytrauma and hemorrhagic shock including a Grade V liver injury. Animals were randomized to resuscitation using reconstituted LP fluids. Physiologic data and blood loss were measured. Coagulation status and inflammatory mediators were evaluated. RESULTS: Full volume reconstituted LP (100%LP) retains on average 86% coagulation factor activity compared to fresh plasma and when used in 1:1 ratios with red blood cells demonstrated superior hemostatic efficacy compared to FFP. Hypertonic LP reconstituted using 50% of the original plasma volume (50%LP) had higher coagulation factor concentrations, was well tolerated in swine, and equally effective compared to 100%LP with respect to physiologic and hemostatic properties. Buffering with ascorbic acid resulted in significant reductions in serum levels of tumor necrosis factor alpha and interleukin-6. CONCLUSION: By minimizing the volume of reconstituted LP and optimizing its anti-inflammatory properties, an LP resuscitation fluid may be created to provide effective hemostatic resuscitation with superior logistical properties.

The effect of preoperative radiotherapy on complication rate after microsurgical head and neck reconstruction.

BACKGROUND: The introduction of radiotherapy (XRT) has resulted in increased survival of patients diagnosed with head and neck malignancies. However, the potentially deleterious impact of radiotherapy on reconstructive efforts continues to be the subject of intense debate. The present study was designed to evaluate the effects of preoperative XRT on complication rates in patients undergoing microsurgical reconstruction of head and neck defects after oncosurgical resection. METHODS: A retrospective cohort study was conducted of all patients who underwent immediate microsurgical reconstruction of post-ablative defects over a 3-year period. Study subjects were divided into two groups: (1) those who did not receive XRT and (2) those who received preoperative XRT. Clinical variables examined and analysed included age, gender, co-morbid conditions, tobacco history, the presence of recurrent disease and ischaemia time. Outcomes of interest included length of intensive care unit (ICU) and hospital stay and postoperative complications. Complications were further classified as flap-related as well as 'medical'. RESULTS: A total of 60 patients were included in this study (group 1: 26 patients; group 2: 34 patients). Results were similar between the study groups with the exception of a higher rate of flap-related complications in patients undergoing XRT. Overall, 19 patients (31.7%) experienced flap-related complications, with 12% of the patients being in group 1 (N=3) versus 47% of patients being in group 2 (N=16) (p=0.003). CONCLUSIONS: Our data suggest that preoperative radiotherapy is associated with a significant increase in postoperative flap-related complications. However, these did not result in a prolonged hospital stay, reflecting the fact that the majority of flap-related complications can be managed on an outpatient basis. Although microsurgical reconstruction is frequently successful, patients with a history of XRT should be informed preoperatively about their increased risk of complications.

Stromal cell-derived factor-1 stimulates vasculogenesis and enhances Ewing's sarcoma tumor growth in the absence of vascular endothelial growth factor.

Stromal cell-derived Factor-1alpha (SDF-1alpha) stimulates the migration of bone marrow (BM) cells, similar to vascular endothelial growth factor (VEGF). We previously demonstrated that inhibition of VEGF(165) by small interfering RNA inhibited Ewing's sarcoma tumor growth, tumor vessel formation and recruitment of BM cells to the tumor. To determine the importance of BM cells in tumor vessel development, we investigated the effects of SDF-1alpha on VEGF-inhibited TC/siVEGF(7-1) Ewing's tumor neovasculature formation and growth. The effect of SDF-1alpha on CD34(+) progenitor cell chemotaxis was determined in vivo. Using a BM transplantation model with GFP(+) transgenic mice as BM donors and nude mice as recipients, we evaluated the effect of SDF-1alpha on the recruitment of BM-derived cells to VEGF(165)-inhibited TC/siVEGF(7-1) tumors, as well as its effect on neovasculature development, vessel morphology and tumor growth. SDF-1alpha stimulated the migration of CD34(+) progenitor cells to Matrigel plugs in vivo and promoted the retainment of BM-derived pericytes in close association with perfused, functional tumor vessels. Intratumor inoculation of Ad-SDF-1alpha into TC/siVEGF(7-1) tumors resulted in increased SDF-1 and PDGF-BB expression, augmented tumor growth, an increase in the number of large, lumen-bearing vascular structures, and enhanced vessel pericyte coverage, with no change in VEGF(165). SDF-1alpha stimulates BM cell chemotaxis and the association of these cells with functional tumor vessels. Furthermore, SDF-1alpha enhances tumor neovascularization and growth with no alteration in VEGF(165). Our work suggests that SDF-1-mediated vasculogenesis may represent an alternate pathway that could potentially be utilized by tumors to sustain growth and neovasculature expansion after anti-VEGF therapy.

Production of VEGF165 by Ewing's sarcoma cells induces vasculogenesis and the incorporation of CD34+ stem cells into the expanding tumor vasculature.

The Ewing's sarcoma cell line TC71 overexpresses vascular endothelial growth factor isoform 165 (VEGF165), a potent proangiogenic molecule that induces endothelial cell proliferation, migration, and chemotaxis. CD34+ bone marrow stem cells can differentiate into endothelial and hematopoietic cells. We used a transplant model to determine whether CD34+ cells migrate from the bone marrow to Ewing's sarcoma tumors and participate in the neovascularization process that supports tumor growth. We also examined the role of VEGF165 in CD34+ cell migration. Human umbilical cord CD34+ cells were transplanted into sublethally irradiated severe combined immunodeficient mice. Seven days later, the mice were injected subcutaneously with TC71 tumor cells. Tumors were excised 2 weeks later and analyzed by immunohistochemistry. The tumor sections expressed both human VE-cadherin and mouse CD31, indicating involvement of donor-derived human cells in the tumor vessels. To determine the role of VEGF165 in the chemoattraction of CD34+ cells, we generated two VEGF165-deficient TC71 clones, a stable anti-sense VEGF165 cell line (Clone 17) and a VEGF165 siRNA-inhibited clone (TC/siVEGF(7-1)). The resulting VEGF165-deficient tumor cells had normal growth rates in vitro, but had delayed growth when implanted into mice. Immunohistochemical analysis revealed decreased infiltration of CD34+ cells into both VEGF165-deficient tumors. These data show that bone marrow stem cells contribute to the growing tumor vasculature in Ewing's sarcoma and that VEGF165 is critical for the migration of CD34+ cells from the bone marrow into the tumor.