Do vented chest seals differ in efficacy? An experimental evaluation using a swine hemopneumothorax model.

OBJECTIVE: Airways compromise was the second leading cause of potentially preventable death among combat casualties. We investigated the ability of five Food and Drug Administration-approved nonocclusive chest seals (CSs) to seal a bleeding chest wound and prevent tension hemopneumothorax (HPTX) in a swine model. METHODS: Following instrumentation, an open chest wound was created in the left thorax of spontaneously air-breathing anesthetized pigs (n = 26; 43 kg). Autologous fresh blood (226 mL) was then infused into the pleural cavity to produce HPTX. The chest wounds were then sealed with CSs. The sealant strength and venting function of CSs were challenged by infusion of 50 mL more blood directly into the chest wound and incremental air injections into the pleural cavity. Tension HPTX was defined as intrapleural (IP) pressure equal to or more than +1 mm Hg and more than 20% deviation in physiologic measurements. RESULTS: An open chest wound with HPTX raised IP pressure (~ -0.7 mm Hg) and caused labored breathing and reductions in PaO2 and SvO2 (p < 0.01). Sealing the wounds with the CSs restored IP pressure, and improved breathing and oxygenation. Subsequent blood infusion into the wound and IP air injections produced CS-dependent responses. Chest seals with one-way valves (Bolin and SAM) did not evacuate the blood efficiently; pooled blood either detached the CSs from skin and leaked out (75%), or clotted and clogged the valve and led to tension HPTX (25%). Conversely, CSs with laminar venting channels allowed escape of blood and air from the pleural cavity and maintained IP pressure and oxygenation near normal levels. Success rates were 100% for Sentinel and Russell (6/6); 67% for HyFin (4/6); 25% for SAM (1/4); and 0% for Bolin (0/4) CSs (p = 0.002). CONCLUSION: The sealant and valve function of vented CS differed widely in the presence of bleeding chest wounds. Medics should be equipped with more effective CSs for treating HPTX in the field.

Does small-volume resuscitation with crystalloids or colloids influence hemostasis and survival of rabbits subjected to lethal uncontrolled hemorrhage?

BACKGROUND: Prehospital, small-volume resuscitation of combat casualties with a synthetic colloid (6% hydroxyethyl starch [HES] 670/0.75) has been recommended when blood or blood components are unavailable. We studied hemostatic effects of a newer synthetic colloid (6% HES, 130/0.4) compared with either a natural colloid (albumin) or to crystalloids in an uncontrolled hemorrhage model. METHODS: Spontaneously breathing New Zealand white rabbits (3.4 ± 0.1 kg) were anesthetized, instrumented, and subjected to a splenic injury with uncontrolled bleeding. Fifteen minutes after injury, rabbits were in shock (mean arterial pressure [MAP] = 26 ± 1.3 mm Hg, and received colloids (6% HES, 130/0.4 or 5% albumin at 15 mL/kg), or crystalloids (normal saline at 30 mL/kg or 5% hypertonic saline at 7.5 mL/kg) for resuscitation in two intravenous bolus injections (15 minutes apart) to raise their MAP to 65 mm Hg, n = 9/group. Animals were monitored for 2.5 hours or until death, and blood losses were measured. Blood samples were analyzed for arterial blood gas, complete blood count, and coagulation measures. RESULTS: There were no differences among groups in baseline measures and initial hemorrhage volume (11.9 ± 0.6 mL/kg) at 15 minutes postinjury. Twenty minutes after fluid resuscitation (1 hour postinjury), MAP was higher, shock indices were lower, and blood pH was higher in colloids versus. crystalloids groups (p < 0.05). Administration of 6% HES 130/0.4 colloid produced the largest hemodilution (54% decrease in hematocrit, p < 0.05 vs. hypertonic saline). Activated partial thromboplastin time increased approximately 35% above baseline in all groups except in 6% HES 130/0.4 group in which it doubled. Clot strength was reduced (15%) only in the 6% HES 130/0.4 group. 6% HES 130/0.4 resuscitation produced the largest blood loss and 33% survival rate that was not different than the crystalloid groups. Albumin produced the best hemostatic and survival outcomes (78%). CONCLUSION: Small-volume resuscitation with crystalloids appeared inadequate to treat hypovolemic shock and prevent death. 6% HES 130/0.4 was effective hemodynamically but detrimental to hemostasis. Albumin produced the best outcomes consistent with our previous observations. Further studies are needed to prove benefit of albumin solution as a possible resuscitation fluid for treating combat casualties at the point of injury.

Influences of limited resuscitation with plasma or plasma protein solutions on hemostasis and survival of rabbits with noncompressible hemorrhage.

BACKGROUND: Plasma infusion with or without red blood cells is the current military standard of care for prehospital resuscitation of combat casualties. We examined possible advantages of early and limited resuscitation with fresh plasma compared with a single plasma protein or crystalloid solutions in an uncontrolled hemorrhage model in rabbits. METHODS: Anesthetized spontaneously breathing rabbits (3.3 ± 0.1 kg) were instrumented and subjected to a splenic uncontrolled hemorrhage. Rabbits in shock were resuscitated at 15 minutes with Plasma-Lyte (PAL; 30 mL/kg), PAL + fibrinogen (PAL + F; 30 mL + 100 mg/kg), fresh rabbit plasma (15 mL/kg), or 25% albumin (ALB; 5 mL/kg) solution, all given in two bolus intravenous injections (15 minutes apart) to achieve a mean arterial pressure of 65 mm Hg, n = 8 to 9/group. Animals were monitored for 2 hours or until death, and blood loss was measured. Blood samples and tissues were collected and analyzed. RESULTS: There were no differences among groups in baseline measures and their initial bleeding volume at 15 minutes. At 60 minutes after injury, mean arterial pressure was higher with ALB than with crystalloids (PAL or PAL + F), but shock indices were not different despite the large differences in resuscitation volumes. Fibrinogen addition to PAL only increased clot strength. Plasma resuscitation increased survival rate (75%) without significant improvement in coagulation measures. Albumin administration replenished total plasma protein and increased survival rate to 100% (p < .05 vs. crystalloids). No histological adverse events were identified in the vital organs. CONCLUSIONS: Fibrinogen administration added to a compatible crystalloid did not improve hemostatic outcomes. Plasma resuscitation increased survival rate; however, its effects did not differ from those obtained with 25% ALB at one-third of the volume. The ALB advantage was consistent with our previous findings in which 5% ALB was used at a volume equal to plasma. The benefit of plasma for resuscitation may be mostly due to its ALB content rather than its coagulation proteins.

Clot-inducing minerals versus plasma protein dressing for topical treatment of external bleeding in the presence of coagulopathy.

BACKGROUND: Previous studies identified WoundStat (WS, smectite) and Combat Gauze (CG, kaolin-coated gauze) as the most effective available agents for controlling arterial bleeding with potential utility in casualty care. Tissue sealant properties of WS suggested its potential advantage over clot-promoting CG for treating coagulopathic bleeding. This study compared the efficacy of CG and WS with a fibrinogen-based (FAST) dressing to control bleeding in coagulopathic animals. METHODS: Coagulopathy was induced in pigs (n = 55, 35 kg) by ∼50% isovolemic hemodilution and hypothermia (core temperature, 33°C ± 0.5°C). A 6-mm arteriotomy was made in the femoral artery and free bleeding allowed for 30 seconds. A test agent (n = 13-15 per group) or control product (gauze, GZ, n = 12) was applied to the wounds and compressed with a Kerlix gauze for 2 minutes. Fluid resuscitation was given, titrated to a mean arterial pressure of 65 mm Hg. Animals were observed for 180 minutes or until death. Angiography using the computed tomography method was performed on survivors, and local tissues were collected for histology. RESULTS: No differences were seen in baseline measures. Coagulopathy, confirmed by a 31% increase in prothrombin time and a 28% reduction in clotting strength (maximum amplitude, thrombelastography assay), was similar in all groups before injury. The average pretreatment blood loss was 11.9 mL/kg ± 0.4 mL/kg with no difference among groups. Posttreatment blood loss, however, was significantly different (p = 0.015) ranging from 18.2 mL/kg ± 8.8 mL/kg (FAST) to 63.3 mL/kg ± 10.2 mL/kg (GZ controls). Stable hemostasis was achieved in 10 of 13 (FAST), 5 of 15 (CG), 2 of 15 (WS), and 1 of 12 (GZ) animals in each group, resulting in significantly different survival rates (8-77%; p = 0.001). The average survival times were 145 (FAST), 119 (CG), 75 (WS), and 74 (GZ) minutes for different groups (p < 0.002). The outcomes with the FAST dressing were significantly better than with WS or GZ in this coagulopathic bleeding model. Essentially, no difference was found between WS and GZ control. Computed tomography images showed limited blood flow only through the vessels treated with FAST dressings. Histologic observations of the vessels indicated minimal damage with FAST and CG and greater injury with WS with some residues present on the tissues. CONCLUSION: The tissue sealant property of WS is apparently mediated by clot formation in the wound; therefore, it was ineffective under coagulopathic conditions. CG was partially effective in maintaining blood pressure up to 1 hour after application. FAST dressing showed the highest efficacy because of the exogenous delivery of concentrated fibrinogen and thrombin to the wound, which bypasses coagulopathy and secures hemostasis.