Arteriovenous fistula. A potential complication of venipuncture.

We describe a very unusual complication of venipuncture: an arteriovenous fistula between the brachial artery and median cubital vein. The clinical presentation and treatment of arteriovenous fistulas and methods of preventing this complication are reviewed.

Selective use of ventilator therapy in flail chest injury.

We have prospectively treated 36 patients with flail chest using a treatment protocol for limited use of mechanical ventilation. Age of the patients ranged from 6 months to 83 years. Patients were divided into three groups dependent upon their clinical presentation and need for respiratory support: Group I patients had severe pulmonary dysfunction-tachypnea, dyspnea, arterial PO2 less than or equal to 60 torr, arterial PCO2 greater than or equal to 50 torr or shunt fraction greater than or equal to 25%. Group II patients had no pulmonary dysfunction but did require temporary respirator support for an associated injury. Group III patients had no pulmonary dysfunction. Thirteen patients were assigned to Group I. They required respiratory support for an average of 10.5 days; 11 of the 13 had complications, and there were two deaths in this group resulting from a combination of respiratory failure and myocardial infarction. Seven patients were assigned to Group II. six patients were extubated immediately postoperatively; one patient with a head injury was hyperventilated for 48 hours to reduce intracranial pressure and then extubated. Sixteen patients were assigned to Group III. Fifteen required no ventilatory support. One 83-year-old man developed pneumonia and was mechanically ventilated for 31 days. Early effective pain control and chest physiotherapy were critical to success and were used in all patients. Increase in respiratory rate, fall in tidal volume or vital capacity, and increased pain were used as criteria for administration of analgesia. Nonventilatory therapy of flail chest reduces morbidity, mortality, and hospital cost.

The effect of lesion volume on cerebral vasomotor tone after focal brain injury and shock.

We have previously shown that the volume of a focal brain injury influences cerebral blood flow. We hypothesized that the cerebral vasomotor tone after traumatic brain injury and shock is related to lesion volume and that the size of the lesion would affect vasomotor reactivity. Swine were randomized to receive either a large or small cryogenic injury followed by shock, and were studied for 5 h postresuscitation. A small brain injury and shock produced a significant and sustained increase in bihemispheric pial arteriolar diameter when compared to the large lesion group and controls. A large brain injury and shock resulted in a significant decrease in the pial arteriolar diameter in the injured hemisphere. We also noted significant differences between and within groups in interhemispheric pial arteriolar diameter and pial arteriolar reactivity to acetylcholine and hypocarbia. These data suggest that the volume of injured tissue influences cerebral blood flow by a vascular mechanism, which may be due in part to an alteration in cerebral endothelial cell function.

Renal, cerebral, and pulmonary effects of hypertonic resuscitation in a porcine model of hemorrhagic shock.

To determine the safety and efficacy of a hypertonic solution for hypovolemic resuscitation, we compared the acute and delayed effects of hypertonic sodium lactate solution (514 mOsm) to Ringer's lactate solution (274 mOsm) in a porcine model of hemorrhagic shock. Cardiovascular, pulmonary, renal, and cerebral functions were examined in mature swine after their blood volume had been reduced by 40%. Hemorrhage produced significant decreases in blood pressure, cardiac output, and creatinine clearance, which were reversed with resuscitation. Resuscitation with Ringer's lactate solution required significantly more fluid and produced a significantly greater increase in intracranial pressure than did hypertonic sodium lactate solution. HSL produced significant increases in serum sodium and osmolality, which resolved within 48 hours. Hypernatremia and hyperosmolality were not associated with renal or cerebral dysfunction and were corrected through increased sodium excretion, free water intake, and a negative free water clearance.

The clinical spectrum of Takayasu's arteritis.

Takayasu's arteritis is a rare inflammatory disease of the aorta, its major branches, and the pulmonary artery in which the varying anatomic involvement, the time course, and the periodicity of exacerbations give rise to a wide variety of signs and symptoms. We have recently encountered five patients with Takayasu's arteritis whose symptoms and findings demonstrate the clinical spectrum of this disease. All five patients are women, with a current mean age of 41 years. Although findings at initial evaluations included systemic manifestations and elevation of the erythrocyte sedimentation rate in four patients, the fifth patient had a normal erythrocyte sedimentation rate and signs of abdominal aortic occlusion. Two patients had a history of hypertension, and four patients complained of upper- or lower-extremity claudication. Arteriographic examination revealed aortic arch branch vessel involvement (type I) in two patients who also had aortic valvular insufficiency; three patients had combined arch vessel and distal aortic disease (type III). All patients have been maintained on steroid medications, and one patient has undergone a trial of cytotoxic agents. Three patients underwent surgical procedures: aortic valve replacement in two patients, and aortorenal bypass in one patient. Takayasu's arteritis gives rise to a variety of symptoms and findings resulting from the distribution and severity of the inflammatory process. With adequate immunosuppression and selective application of surgical therapy, there is a good prognosis for survival and a return to functional status.

Hypertonic sodium lactate versus lactated ringer's solution for intravenous fluid therapy in operations on the abdominal aorta.

Fifty-eight patients who were to undergo aortic reconstruction were prospectively randomized into two groups to compare the effects of perioperative fluid replacement with isotonic and hypertonic crystalloid solutions. Blood loss was replaced with packed red blood cells, and additional fluid was given as either Ringer's lactate solution (RL, 130 mEq sodium/L, 274 mOsm/L) or a hypertonic balanced salt solution (HSL, 250 mEq sodium/L, 514 mOsm/L). Fluid was administered to maintain the cardiac filling pressure within 3 torr of the preoperative level and the cardiac output (CO) at or above the preoperative level. The groups were similar with respect to age, preexisting disease, duration of operation, and operative blood loss. During the operation, the RL group required 9.5 +/- 0.8 L of fluid, whereas the HSL group required 4.5 +/- 0.3 L (P less than 0.001). Pulmonary, cardiac, and renal functions were adequately maintained in both groups. There were no significant differences between the groups with regard to CO, urine output, or creatinine clearance during the operation and early postoperative period. Postoperatively, the intrapulmonary shunt was 20 +/- 1% in the RL group and it was 16 +/- 1% in the HSL group (P less than 0.05). The amount of sodium infused and the cumulative sodium balance at the completion of the study period were similar in both groups. Serum sodium and osmolarity were significantly greater in the HSL group (P less than 0.001), reaching a maximum of 151 +/- 1 mEq/L and 305 +/- 2 mOsm/L, respectively. Two patients in the HSL group had a persistent elevation in serum osmolarity (greater than 320 mOsm/L) during operation, for which they received RL for the balance of the resuscitation. There were no complications that could be attributed to the hypertonicity of the solution. HSL is effective for resuscitation of patients with extracellular fluid deficit and is safe provided that the serum sodium and osmolarity are monitored during periods of large volume administration.

The management of splenic trauma in a trauma system.

Large teaching centers have reported splenic salvage rates of 40% to 50% in adults after splenic trauma. It is unknown whether similar salvage rates can be achieved safely in community trauma centers with a lower volume of patients and less experience. Between August 1984 and August 1988, 117 patients with splenic injury were treated at a level I center and 311 were treated at four level II centers. Splenectomy was performed in 252 patients (59%), splenorrhaphy was performed in 160 patients (37%), and 16 patients (4%) were observed. While the splenic salvage rate was higher at the level I center (50% vs 38%), selective splenorrhaphy was successful in the level II centers where the volume of splenic injury was lower (15 to 25 cases per year).

A prospective, randomized comparison of computed tomography with conventional diagnostic methods in the evaluation of penetrating injuries to the back and flank.

We prospectively compared computed tomography with conventional diagnostic methods in the evaluation of penetrating injuries to the back and flank in 85 patients. Immediate laparotomy was performed in 24 patients because of physical findings, and these patients were not randomized. Nine unnecessary operations were performed in this group, and these nine patients had significantly higher hospital costs than patients in either randomized group. In the randomized patients, there were only three true-positive and three false-positive findings. Both computed tomographic evaluation (31 patients) and conventional evaluation (30 patients) were highly accurate and specific for injuries that required operation. Evaluation with computed tomography required a longer time to make a hospital disposition but required fewer diagnostic tests. Computed tomography can be useful in the assessment of penetrating injuries to the back and flank.

Selective nonoperative management of liver and spleen injuries in neurologically impaired adult patients.

BACKGROUND: Nonoperative management of blunt liver and spleen injuries in hemodynamically stable, neurologically intact patients has become an accepted treatment in recent years. OBJECTIVE: To determine the morbidity and mortality in neurologically impaired adult patients who had sustained blunt liver or spleen injuries and who had been managed nonoperatively in a monitored setting, owing to the preponderance of blunt trauma and associated head injuries in Vermont. DESIGN: Case-control study. SETTING: Regional level I trauma center in northern Vermont. PATIENTS: One hundred eighty-seven consecutive patients with documented blunt splenic or hepatic trauma who were admitted to a regional rural trauma referral center in Vermont during an 8-year period, beginning in January 1987, were studied. Hemodynamically stable patients underwent diagnostic imaging studies and were classified by mental status as either normal or altered. Patients who required operative intervention were excluded. MAIN OUTCOME MEASURES: Morbidity and mortality rates for each group were recorded and compared to determine if statistically significant differences between the two groups existed. RESULTS: The groups were similar in age, systolic blood pressure, and hematocrit at admission. The group of patients with an altered mental status were more severely injured and had a longer hospital stay. Intensive care unit stays were not significantly different. Transfusion requirements for both groups were minimal; however, the group of patients with an altered mental status received more blood transfusions compared with the group of patients with a normal mental status. There was no significant difference in morbidity and mortality between the two groups. There were no failures of nonoperative management, no complications, and no missed visceral injuries in the group of patients with an altered mental status. Patients older than 50 years had higher morbidity and mortality. CONCLUSIONS: Nonoperative management in patients with an altered mental status can be done safely in a monitored setting. This challenges the current criteria of excluding neurologically impaired patients with liver or spleen trauma from nonoperative management.

Cardiovascular and neurohumoral responses following burn injury.

Cardiovascular and endocrine responses were evaluated in 12 adult patients over a 5-day period following 30% to 66% burn injury. Heart rate, mean arterial pressure, central venous pressure, pulmonary capillary wedge pressure, cardiac output, systemic vascular resistance, and stroke volume were measured. Plasma concentrations of angiotensin II, atrial natriuretic peptide, vasopressin, neuropeptide Y, norepinephrine, and epinephrine were measured. On the day of burn injury, systemic vascular resistance was markedly elevated, and stroke volume and cardiac output were low, but all normalized by day 3, and cardiac output and stroke volume increased by day 5 without significant changes of central venous pressure or pulmonary capillary wedge pressure. Mean arterial pressure and heart rate did not change significantly over the 5-day period. Vasopressin, angiotensin II, neuropeptide Y, norepinephrine, and epinephrine concentrations in plasma were elevated on admission. Vasopressin concentrations were elevated 50 times normal on admission and returned to normal levels by days 4 to 5. Plasma atrial natriuretic peptide concentrations were normal on admission and increased significantly on days 3 to 5. The reciprocal relationship between systemic vascular resistance and cardiac output and between vasopressin and atrial natriuretic peptide correlate with each other and the observed physiologic events that occurred following burn injury and resuscitation. All of these changes in cardiac performance occurred without significant alterations in preload or afterload as measured by central venous pressure, pulmonary capillary wedge pressure, and mean arterial pressure. Increases in plasma levels of atrial natriuretic peptide correlated with the increased stroke volume and cardiac output observed in these patients. The results of this study are consistent with the conclusion that the extreme elevations of plasma vasopressin levels contribute to the vascular complications of increased systemic vascular resistance and decreased cardiac output and contractility seen following burn injury.

Intra-abdominal injury following blunt trauma. Identifying the high-risk patient using objective risk factors.

The initial physical examination is frequently unreliable in identifying patients with blunt trauma at high risk for having serious intra-abdominal injury. Intra-abdominal injury may be associated with specific injuries or risk factors, but the usefulness of such objective clinical criteria in predicting intra-abdominal injury has not previously been determined. The presence or absence of each of 11 clinical indicators and their association with serious intra-abdominal injury were analyzed in 3223 patients with blunt trauma. Linear and logistic regressions were used to determine which factors were significant predictors of an increased probability of intra-abdominal injury. Arterial base deficit less than -3 mEq/L, major chest injury, hypotension, and pelvic fractures were found to significantly increase the chance of intra-abdominal injury. Early diagnostic evaluation of the abdomen using diagnostic peritoneal lavage or computed tomography should be strongly considered in patients with blunt trauma who present with these associated factors.

The impact of in-house surgeons and operating room resuscitation on outcome of traumatic injuries.

As trauma systems develop, more patients can potentially benefit from immediate surgery. With in-house surgeons available, enthusiasm for direct transfer from the scene to the operating room (OR) has developed in many institutions. The purpose of this study was to define precisely which patients should be taken to the OR for resuscitation. Three hundred twenty-three patients were taken to the OR directly from the field during a 4-year period (6.9% of trauma activations). Indications included the following: (1) cardiac arrest--one vital sign present, (2) persistent hypotension despite field intravenous fluid, and (3) uncontrolled external hemorrhage. A board-certified surgeon and resuscitation team met the field transport team in the OR in all cases. Cardiopulmonary resuscitation for patients with blunt trauma was not accompanied by survival even with immediate surgery by a trained surgeon and it wastes valuable OR resources. Patients with prehospital hypotension unresponsive to fluid resuscitation indicate the need for rapid surgery. Patients with blunt injuries even with hypotension infrequently undergo operations in less than 20 minutes and can be resuscitated in traditional areas where better roentgenograms are obtained. Penetrating injuries to the chest and abdomen with hypotension are the primary indications for OR resuscitation. It can be anticipated with field communication and accompanied by enhanced survival.

High-risk behavior and the public burden for funding the costs of acute injury.

OBJECTIVE: To determine if high-risk behavior is associated with increased injury severity and cost and if public agencies bear a disproportionate burden of that cost. DESIGN: Case comparison study utilizing patient data collected over a 10-year period. SETTING: Five level 1 and 2 trauma centers in an urban-suburban community with a population of 2.4 million. PARTICIPANTS: Trauma registry data from 37,304 consecutive hospitalized adult patients with trauma. Financial data were reported and analyzed on 28,842 of these. MAIN OUTCOME MEASURES: Incidence of alcohol intoxication, other drug use, use of vehicular protective devices, and firearm violence injuries in patients with private vs public health care sponsorship. Length of hospital stay, injury severity, and hospital unit charges were assessed for high-risk behavior. RESULTS: High-risk behavior was more prevalent among trauma patients relying on public funding to cover the costs of their injuries (P < .001). Total hospital unit charges were 28% and 35% higher for motorists not wearing seat belts and motorcyclists not wearing helmets, respectively. Injury severity and length of stay were also higher (P < .001). CONCLUSIONS: High-risk behavior is associated with increased injury severity and cost. Trauma victims exhibiting high-risk behavior more often depend on public agencies to cover the cost of acute injury. Failure to establish and enforce laws and policies designed to reduce or prevent injury may generate enormous trauma care costs, borne to a large extent by public agencies. Further restriction of certain types of high-risk behavior and the institution of "users' fees," taxes, or penalties may be necessary to reduce the disproportionate public agency cost generated by this activity.

Delayed fluid resuscitation of head injury and uncontrolled hemorrhagic shock.

OBJECTIVE: To evaluate the effects of delayed vs early fluid resuscitation on cerebral hemodynamics after severe head injury and uncontrolled hemorrhagic shock. DESIGN: Prospective, randomized, controlled experimental trial. SETTING: Surgical research laboratory. PARTICIPANTS: Immature swine (N=16) weighing 40 to 50 kg. INTERVENTIONS: Twelve swine were subjected to cryogenic brain lesion and hemorrhage to maintain a mean arterial pressure (MAP) of 50 mm Hg. Animals were randomized to receive 1 L of Ringer lactate solution in 20 minutes, starting 20 minutes after injury and hemorrhage, followed by 1 L of Ringer lactate solution in 30 minutes (ER group) (n=6), or no fluid resuscitation (DR group) (n=6). The 4 control animals underwent instrumentation only. The study ended 70 minutes after head injury and hemorrhage. MAIN OUTCOME MEASUREMENTS: Measurements of MAP, bilateral regional cerebral blood flow, serum hemoglobin level, systemic and regional cerebral oxygen delivery, and intracranial pressure performed at baseline and 20 (phase 1), 50 (phase 2), and 70 minutes (phase 3) after head injury and hemorrhage. Lesion size (percentage of ipsilateral cortex) was measured post mortem. RESULTS: All animals survived the experimental period. Systemic cerebral oxygen delivery in the DR group was significantly lower at phase 3 compared with that of the ER group (31.5% vs 53.1% at baseline) (P=.03). However, bilateral regional cerebral oxygen delivery was significantly greater in the DR group at phase 3 compared with that of the ER group (71.5% vs 47.0% at baseline in the injured side; 72.9% vs 48.4% at baseline in the noninjured side) (P=.02). Bilateral cerebral blood flow was similar in all groups at all times. The ER group showed a trend toward a greater intracranial pressure elevation (6.8 vs -0.25) (P=.07) and lesion size (37.0% vs 28.6%) (P=.07). Hemoglobin level became significantly lower in the ER group at phase 2 (7.0 vs 10.7) (P=.03) and remained lower at phase 3 (6.9 vs 11.7) (P=.01). CONCLUSIONS: Early fluid resuscitation with Ringer lactate solution following head injury and uncontrolled hemorrhagic shock worsens cerebral hemodynamics. Cerebral pressure autoregulation is sufficiently intact following head injury to maintain regional cerebral oxygen delivery without asanguineous fluid resuscitation.

Impact of a trauma system on outcome of severely injured patients.

We examined the impact of a trauma system on the survival of patients with a Trauma Score of 8 or less. We compared the observed survival with that predicted using a method that calculates the probability of survival (Ps) based on age, physiologic score, and anatomic severity of injury. Of 3394 patients triaged to trauma centers in a 12-month period, 283 (8.3%) had a Trauma Score of 8 or less. Sufficient data were available in 189 patients with blunt trauma to make the survival comparison. The Ps was 18%; the observed survival was 29%. Of 60 patients with penetrating trauma and complete data, the Ps was 8%; the observed survival was 20%. We attribute the improved survival to the integration of prehospital and hospital care and expeditious surgery.

Trauma deaths in a mature urban vs rural trauma system. A comparison.

OBJECTIVE: To compare the timing, severity, and injury characteristics of patients dying from trauma in an urban vs a rural setting. DESIGN: Retrospective review of autopsy database (urban) and medical examiner database (rural), with selected medical chart review. SETTING: An organized urban trauma system with 6 trauma centers and a rural state with no formal trauma system and 1 trauma center. PATIENTS: All trauma fatalities occurring in an urban (n = 612) and a rural (n = 143) setting during a 1-year period. RESULTS: In the urban system, 248 patients (40.5%) died at the scene of injury compared with 103 (72%) patients in a rural environment. During the first 24 hours of hospitalization 243 (39.7%) urban patients died compared with 23 (16%) rural patients. Eighty-nine urban patients (14.5%) and 17 rural patients (11.8%) survived for more than 24 hours but later died in the hospital. The mean age of those who died was significantly greater in the rural trauma system than in the urban trauma system (P < .001), and the Injury Severity Score was significantly less in the rural trauma system than in the urban trauma system (P < .01). In the patients who died after being admitted to the hospital for more than 24 hours there was a significantly higher rate of preexisting comorbidity in the rural patients than in the urban patients (P < .05). The most frequent cause of death in the rural setting was multisystem organ failure; head injury was the most common cause of death in the urban setting. CONCLUSIONS: Patients who die in a rural area without a formal trauma system are more likely to die at the scene, are less severely injured, and are older. Rural trauma patients who are admitted to a hospital and who survived for at least 24 hours before dying are older, less severely injured, have significantly more comorbidities, and are more likely to die of multisystem organ dysfunction than their urban counterparts. These differences reflect the different patient populations and injury patterns that confront urban and rural trauma centers. The higher proportion of scene deaths in the rural environment may reflect the longer discovery and transport times that occur in a rural setting.

The epidemiology of traumatic death. A population-based analysis.

Despite the proliferation of trauma systems, there are no population-based data describing the epidemiology of traumatic death. To provide these data, we reviewed all trauma deaths occurring in San Diego (California) County during 1 year. There were 625 traumatic deaths during the study (27.3 deaths per 100,000 population per year). Motor vehicle trauma was the most common cause of injury leading to death (N = 344 [55.2%]; 15.0 annual deaths per 100,000 population). Central nervous system injuries were the most common cause of death (48.5%, or 13.2 deaths per 100,000 population per year). Sepsis was responsible for only 2.5% of the overall mortality. Based on life-table data, traumatic death resulted in an annual loss of 1091 years of life per 100,000 and an annual loss of 492 years of productivity per 100,000. Injury continues to account for an enormous loss of life despite improvements in survival wrought by trauma systems.

Five-year follow-up of prophylactic vena cava filters in high-risk trauma patients.

OBJECTIVE: To assess the short- and long-term outcomes of vena cava filter (VCF) placement for prophylaxis against pulmonary embolism in patients at high risk due to trauma. DESIGN AND SETTING: Case series at a level I trauma center. PATIENTS: Patients were considered for prophylactic VCF placement if they met 1 of the injury criteria--spinal cord injuries with neurologic deficit, severe fractures of the pelvis or long bone (or both), and severe head injury--and had a contraindication to anticoagulation. INTERVENTION: Vena cava filters were placed percutaneously by the interventional radiologists when the acute trauma condition was stabilized following admission. MAIN OUTCOME MEASURES: Filter tilt of 14 degrees or more, strut malposition, insertion-related deep vein thrombosis, pulmonary embolism, or inferior vena cava patency. RESULTS: There were 132 prophylactic VCFs placed. A 3.1% rate of insertion-related deep vein thrombosis occurred, all of which were asymptomatic. Filter tilt occurred in 5.5% of patients and strut malposition in 38%. Three cases of pulmonary embolism (1 fatal) occurred in a prophylactic VCF, and all patients had either filter tilt or strut malposition. The risk of pulmonary embolism developing was higher in those patients with filter tilt or strut malposition than in those who did not have these complications (6.3% vs 0%; P=.05; Fisher exact test). The 1-, 2-, and 3-year inferior vena cava patency rates (+/-SD) were 97%+/-3%. CONCLUSIONS: Prophylactic VCF can be placed safely with an acceptable rate of insertion-related deep vein thrombosis and long-term inferior vena cava patency. Patients with prophylactic VCF remain at risk for pulmonary embolism if the filter is tilted 14 degrees or more or has strut malposition. In such patients, consideration should be given to placing a second filter.

A multi-institutional study of factors associated with fetal death in injured pregnant patients.

HYPOTHESIS: Factors associated with fetal death in injured pregnant patients are related to increasing injury severity and abnormal maternal physiologic profile. DESIGN: A multi-institutional retrospective study of 13 level I and level II trauma centers from 1992 to 1996. MAIN OUTCOME MEASURE: Fetal survival. RESULTS: Of 27,715 female admissions, there were 372 injured pregnant patients (1.3%); 84% had blunt injuries and 16% had penetrating injuries. There were 14 maternal deaths (3.8%) and 35 fetal deaths (9.4%). The population suffering fetal death had higher injury severity scores (P<.001), lower Glascow Coma Scale scores (P<.001), and lower admitting maternal pH (P = .002). Most women who lost their fetus arrived in shock (P = .005) or had a fetal heart rate of less than 110 beats/min at some time during their hospitalization (P<.001). An Injury Severity Score greater than 25 was associated with a 50% incidence of fetal death. Placental abruption was the most frequent complication, occurring in 3.5% of patients and associated with 54% mortality. Cardiotrophic monitoring to detect potentially threatening fetal heart rates was performed on only 61% of pregnant women in their third trimester. Of these patients, 7 had abnormalities on cardiotrophic monitoring and underwent successful cesarean delivery. CONCLUSIONS: Fetal death was more likely with greater severity of injury. Cardiotrophic monitoring is underused in injured pregnant patients in their third trimester even after admission to major trauma centers. Increased use of cardiotrophic monitoring may decrease the mortality caused by placental abruption.

Routine prophylactic vena cava filter insertion in severely injured trauma patients decreases the incidence of pulmonary embolism.

BACKGROUND: Trauma patients are at increased risk for venous thromboembolism despite routine prophylaxis. A five-year retrospective review of pulmonary embolism at the Medical Center Hospital of Vermont showed that four types of injuries (head injuries, spinal cord injuries, complex pelvic fractures, and hip fractures) accounted for 92 percent of pulmonary emboli in patients on the trauma service. STUDY DESIGN: Beginning July 1991, all patients who met criteria for a high-risk injury (excluding hip fractures) had prophylactic vena cava filters inserted percutaneously in the radiology suite. Weekly impedance plethysmography was performed for the detection of deep vein thrombosis. Following discharge, abdominal duplex ultrasound was performed at one month, six months, and yearly to check filter position and patency. RESULTS: To date, a total of 63 patients, or 2 percent of the total trauma population, had prophylactic vena cava filters inserted. There were 19 (30 percent) deep vein thromboses in this population of patients and one patient had a pulmonary embolism (1.5 percent). Overall there was significant (chi-square, p < 0.00072) reduction of pulmonary embolism on the trauma service compared to the historical controls. Follow-up examination with abdominal duplex ultrasound showed a 30-day patency rate of 100 percent and a one- and two-year patency rate of 96.1 +/- 3.8 percent by life-table analysis. CONCLUSIONS: We concluded that prophylactic vena cava filters are efficacious in decreasing the risk of pulmonary embolism in high-risk trauma patients.