Search

Azygous venous system injuries are rare. Although not commonly classified as a thoracic great vessel, the azygous system manifests morbidity and mortality that is similar to that of other great vessel injuries.Over a 40 year period, data were retrospectively collected and charts reviewed from a vascular injury database.A total of 22 injuries to the azygous venous system were identified with 21 to the azygous vein and 1 to the hemiazygous vein. All were secondary to penetrating trauma, with 19 from gunshot wounds and 3 from stab wounds. Eight of 22 patients died for an overall mortality of 36%. All patients had associated injuries, with concomitant injury to the lung noted in all cases.Injuries to the azygous venous system have a significant associated lethality. As an azygous venous injury is not typically the indication for operation, anterior incisions are often employed, making exposure difficult. In the majority of injured patients, the azygous system injury was found after excluding injuries to the heart, lung, and great vessels. A key finding in detecting these injuries through an anterior incision is continued hemorrhage of dark blood from a posterior location in the thoracic cavity. The azygous venous system should be considered early as a source of significant hemorrhage from the posterior mediastinum.

The aim of this series is to describe a new and aggressive approach to definitive closure of the open abdomen.A retrospective review of 37 patients who underwent definitive abdominal closure using a combination of vacuum pack, vacuum-assisted wound management and human acellular dermal matrix (HADM).All patients' open abdomens were maintained with vacuum assisted wound management in attempts for primary closure. Once it was determined that the abdomen would not close primarily; it was closed with HADM and skin advancement. The mean duration of the open abdomen was 21.7 days (range 6-45), with an average of 127.78 cm of HADM, the largest number being 800 cm, with decreasing use of product later in the series. No major complications were seen with the repair. Superficial wound infection occurred with two patients that were easily treated with wet to dry dressing changes. No intraabdominal complications such as fistula or graft loss were seen. All patients left the hospital with an intact abdominal wall and skin. All 37 patients survived to discharge and were seen in follow-up within one month. No early hernia formation was seen at the one month follow up with the longest at three years. No abdominal wall complications were seen in subsequent follow up patients.Early aggressive closure of the open abdomen is possible with a combination of vacuum pack, vacuum-assisted wound management and HADM. Short term results are promising and warrant further study.

Background Although studies have ascertained that ten percent of soldiers killed in battle bleed to death from extremity wounds, little data exists on exsanguination and mortality from extremity injuries in civilian trauma. This study examined the treatment course and outcomes of civilian patients who appear to have exsanguinated from isolated penetrating extremity injuries. Methods Five and 1/2 years' data (Aug 1994 to Dec 1999) were reviewed from two Level I trauma centers that receive 95% of trauma patients in metropolitan Houston, TX. Records (hospital trauma registries, emergency medical system (EMS) and medical examiner data) were reviewed on all patients with isolated extremity injuries who arrived dead at the trauma center or underwent cardiopulmonary resuscitation (CPR) or emergency center thoracotomy (ECT). Results Fourteen patients meeting inclusion criteria were identified from over 75,000 trauma emergency center (EC) visits. Average age was 31 years and 93% were males. Gunshot wounds accounted for 50% of the injuries. The exsanguinating wound was in the lower extremity in 10/14 (71%) patients and proximal to the elbow or knee in 12/14 (86%). Ten (71%) had both a major artery and vein injured; one had only a venous injury. Prehospital hemorrhage control was primarily by gauze dressings. Twelve (86%) had "signs of life" in the field, but none had a discernable blood pressure or pulse upon arrival at the EC. Prehospital intravenous access was not obtained in 10 patients (71%). Nine patients underwent ECT, and nine were initially resuscitated (eight with ECT and one with CPR). Those undergoing operative repair received an average of 26 +/- 14 units of packed red blood cells. All patients died, 93% succumbing within 12 hours. Conclusion Although rare, death from isolated extremity injuries does occur in the civilian population. The majority of injuries that lead to immediate death are proximal injuries of the lower extremities. The cause of death in this series appears to have been exsanguination, although definitive etiology cannot be discerned. Intravenous access was not obtainable in the majority of patients. Eight patients (57%) had bleeding from a site that anatomically might have been amenable to tourniquet control. Patients presenting to the EC without any detectable blood pressure and who received either CPR or EC thoracotomy all died.

Background The aim of this modeling study was to examine how casualty load affects the level of trauma care in multiple casualty incidents and to define the surge capacity of the hospital trauma assets. Methods The disaster plan of a U.S. Level I trauma center was translated into a computer model and challenged with simulated casualties based on 223 patients from 22 bombing incidents treated at an Israeli hospital. The model assigns providers and facilities to casualties and computes the level of care for each critical casualty from six variables that reflect the composition of the trauma team and access to facilities. Results The model predicts a sigmoid-shaped relationship between casualty load and the level of care, with the upper flat portion of the curve corresponding to the surge capacity of the trauma assets of the hospital. This capacity is 4.6 critical patients per hour using immediately available assets. A fully deployed disaster plan shifts the curve to the right, increasing the surge capacity to 7.1. Overtriage rates of 50% and 75% shift the curve to the left, decreasing the surge capacity to 3.8 and 2.7, respectively. Conclusion This model defines the quantitative relationship between an increasing casualty load and gradual degradation of the level of trauma care in multiple casualty incidents, and defines the surge capacity of the hospital trauma assets as a rate of casualty arrival rather than a number of beds. The study demonstrates the value of dynamic computer modeling as an important tool in disaster planning.

Current massive transfusion guidelines are derived from washout equations that may not apply to bleeding trauma patients. Our aim was to analyze these guidelines using a computer simulation.A combined hemodilution and hemodynamic model of an exsanguinating patient was developed to calculate the changes in prothrombin time (PT), fibrinogen, and platelets with bleeding. The model was calibrated to data from 44 patients. Time intervals to subhemostatic values of each coagulation test were calculated for a range of replacement options.Prolongation of PT is the sentinel event of dilutional coagulopathy and occurs early in the operation. The key to preventing coagulopathy is plasma infusion before PT becomes subhemostatic. The optimal replacement ratios were 2:3 for plasma and 8:10 for platelets. Concurrent transfusion of plasma with blood is another effective strategy for minimizing coagulopathy.Existing protocols underestimate the dilution of clotting factors in severely bleeding patients. The model presents an innovative approach to optimizing component replacement in exsanguinating hemorrhage.

Patients with blunt aortic injury (BAI) often have concomitant liver or spleen (L/S) injuries. With increasing use of cardiopulmonary bypass with heparinization in repair of BAI, many advocate operative management of the L/S injury before aortic repair to eliminate risk of hemorrhage. We evaluated the safety of nonoperative management (NOM) of blunt L/S injuries in patients undergoing acute BAI repair with bypass.All patients admitted over a 6-year period with BAI were identified from the registry of our Level I trauma center. Patients with isolated L/S injuries without BAI admitted over the same period served as controls. Groups were compared with regard to demographics, injury characteristics, hospital course, and mortality.Eighty-four patients were diagnosed with BAI from 1994 to 2000; 28 (33%) also had blunt abdominal trauma. Three patients with severe brain injury did not undergo BAI repair, and five required laparotomy before BAI repair for other intra-abdominal injuries (two for hemodynamic instability with splenic injury, and three for concomitant bowel injury). Therefore, 20 of 28 (71.4%) BAI patients with grade I or II L/S injury (Aorta L/S group) underwent planned NOM. All BAIs were repaired using partial bypass with full heparinization. These 20 patients are compared with 894 patients with grade I or II L/S injuries with no BAI (L/S group) over the same time period. There was no difference in the nonoperative failure rate of the Aorta L/S group versus the L/S group (0% vs. 1.7%). Both groups had similar complication rates. The Aorta L/S group was also compared with 56 BAIs without solid organ injury (Aorta group). Although the Aorta L/S group was more severely injured than the Aorta group (Injury Severity Score of 35.3 vs. 26.8,0.0001), transfusion rates (5.7 U of packed red blood cells vs. 8.0 U of packed red blood cells, p = NS), hospital days (17.9 vs. 19.1, p = NS) and mortality (10% vs. 9%, p = NS) were similar.NOM of patients with grade I or II L/S injury who undergo systemic anticoagulation with heparin for repair of BAI is safe and associated with transfusion rates similar to BAI alone. Patients with low-grade liver or spleen injuries do not require laparotomy before BAI repair using partial bypass.

The critical decisions in patients with thoracoabdominal trauma are establishing the need to explore either or both cavities and determining appropriate sequencing. The causes and patterns of management difficulties were analyzed in 82 consecutive patients with penetrating thoracoabdominal injuries. Nine thoracotomies (11%) and 16 laparotomies (22%) were negative, with the major causes being misleading chest tube outputs, bullet trajectories, and abdominal tenderness. Inappropriate sequencing occurred in 19 patients (23%), and 15% required reoperation within 24 hours. Pitfalls and misjudgements in management of penetrating thoracoabdominal injuries occur in distinct patterns. The unreliability of chest tube output and abdominal examination must be kept in mind, and intraoperative clues of ongoing hemorrhage outside the operative field must be sought.

Emergency center thoracotomy was performed at our facility on 389 patients from 1984 through 1989. There were no patients excluded from the study, and survival for all patients was 8.3% with survival rates of 15.2% and 7.3% for stab and gunshot wounds, respectively. Emergency center thoracotomy was performed on 42 patients suffering from isolated extrathoracic injuries with 7% survival. There were no survivors of blunt trauma in this study. Fifty-three percent of the patients arrived with cardiopulmonary resuscitation (CPR) in progress. The average time of prehospital CPR for survivors was 5.1 minutes compared with 9.1 minutes for nonsurvivors. Of the survivors, prehospital endotracheal intubation prolonged successful toleration of CPR to 9.4 minutes compared with 4.2 minutes for nonintubated surviving patients (p less than 0.001). Emergency center thoracotomy is useful in the resuscitation of victims dying of penetrating truncal trauma. Prehospital endotracheal intubation significantly lengthened the time of successful CPR.

It has been stated that delay in operative repair of penetrating injuries to the gastrointestinal tract will result in a high rate of complications related to infection. To test this assertion, a group of patients with penetrating injuries to the colon were analyzed who had operative repair delayed (usually because of triage considerations) more than 6 hours after admission to the hospital. Nine hundred six patients who survived at least 48 hours after injury were divided into two groups. The immediate group of 769 patients was treated within 6 hours of admission and the delayed group of 137 patients was treated more than 6 hours after admission. The mortality for the immediate group was 4.0% vs. 1.5% for the delayed group. Colon-related infectious complications, defined as abscess or colon suture-line failure, occurred in 10% of the immediate group and 4.4% of the delayed group. To eliminate the effect of associated injuries, the group of patients with colon injuries only was analyzed separately. There was no mortality for 128 patients with colon injuries only operated on within 12 hours of injury, and the colon-related infectious morbidity rate was 3%. Eleven patients with colon injuries only were treated after 12 hours with a mortality of 9% and colon-related infectious morbidity of 18%. These data demonstrate that even patients with fecal contamination can have operative repair delayed for up to 12 hours without undue morbidity related to infection.

During a recent 11-year period, 233 consecutive patients with 358 penetrating iliac vascular injuries were treated at our institution. Injuries of the common and external iliac arteries were most often repaired with lateral suture (31%) although several other techniques were also employed. Lateral suture and ligation were used with nearly equal frequency in the management of venous injuries. The hospital mortality rate for the series was 28%, and 56/66 deaths (85%) were due to exsanguination or shock. One patient, initially treated with an end-to-end anastomosis of the iliac artery, died a year after discharge from a ruptured false aneurysm. Two patients treated with lateral suture of venous injuries died of pulmonary embolism. Arterial complications occurred in 15% of patients with arterial injuries and three patients required amputation. No graft infections occurred in 16 patients treated with PTFE interpositions, including four with associated colorectal injuries. Venous complications occurred in 12% of patients with venous injuries, and most were noted in those treated with ligation. Four patients treated by venous ligation developed chronic venous insufficiency. The prevention of death from exsanguination is the greatest problem in the management of patients with iliac vascular injuries. Although some late deaths and many complications may be related to the technique of vascular repair, circumstances often prohibit alternative methods. Despite two deaths from pulmonary embolism, insufficient data exist to condemn lateral suture of venous injuries.

The purpose of this study was to investigate the distribution of various mechanisms of injury and the relative severity of such injury cases throughout the different geographic zones of a large urban area using a computerized emergency medical services (EMS) dispatch/patient record database. The study city (population, 2 million residents) was divided into 156 geographic grids (each 4.5 by 3 miles) and the incidence and relative severity of various injury mechanisms were determined for each zone. Results: In one year (1988), there were more than 115,000 separate EMS incidents involving more than 150,000 patients, 26,000 of whom were transported for injuries incurred in 10,064 motor vehicle accidents, 4,587 falls, 4,015 lacerations/stabwounds, 1,796 beatings, 1,270 gunshots, and 952 auto-pedestrian accidents. Analysis of the 156 zones showed a disproportionate number of EMS responses in the city center with two centralmost grids accounting for about 25% of all responses. Call volume then progressively diminished toward the periphery of the city. However, with some very minor exceptions, the relative incidence and severity of the various injury mechanisms remained proportionally uniform within each zone, regardless of geographic location. Therefore, contrary to popular notoriety, the incidence and associated severity of any given injury type generally was not necessarily predicted by any particular neighborhood predilection for it, but rather by the overall demand for EMS in that zone of the city.

Houston is served by only two trauma centers: Hermann Hospital (University of Texas Medical School at Houston [UTMSH]), and Ben Taub General Hospital (Baylor College of Medicine). In mid-1988, Hermann Hospital, prompted by a shortage of ICU nurses and +8.0 million/yr of uncollectible trauma charges, began to divert critically ill and injured patients to the already overburdened Ben Taub General Hospital. The academic consequences to UTMSH included a severe loss of clinical experience by the surgical residents and medical students and a severe reduction in faculty-generated billing. The lost billing from the trauma service and the other clinical services approximated +8.0 million/yr. This equated to an +13.5 million decrease in the anticipated billings for the year. Alternative revenue sources were not apparent. Other centers with a heavy trauma system commitment are at risk to suffer similar unsettling academic sequelae as our trauma systems fail.

The term “cardiac contusion” has been used to describe an injury to the heart after blunt chest trauma. Histologically, it is characterized by a contused myocardium with hemorrhagic infiltrate, localized necrosis, and edema. Findings are best confirmed during surgery or at autopsy. Clinically, blunt cardiac injury (BCI) is the preferred diagnosis term, as blunt chest trauma can result in a broad range of cardiac injury. BCIs can be further described by specific injuries or observed dysfunction. Significant BCI usually occurs from high-impact trauma from motor vehicle accidents (50%), pedestrians struck by motor vehicles (35%), motorcycle crashes (9%), and falls from significant heights (6%). Diagnosing BCI can be challenging as there is no accepted gold standard diagnostic test. The diagnosis is fraught with further difficulty in the polytrauma patient. The reported incidence of cardiac injury following blunt chest trauma ranges from 8% to 76%, in large part due to a lack of standardized diagnostic criteria. In the absence of severe arrhythmia and hemodynamic instability, the significance of BCI is sometimes called into question. In the setting of blunt trauma, a high clinical suspicion for BCI is required, and certain patients should be monitored for adverse sequelae since there are no pathognomonic clinical signs or symptoms that correlate with the risk of cardiac complications. Indeed, the sequelae of BCI are more important than its label.

Survivors of innominate and other major cardiovascular injuries are being seen with increasing frequency. Penetrating injuries more frequently involve the distal innominate artery and innominate veins. Associated subclavian and carotid artery injuries are more frequent following penetrating trauma. Blunt trauma typically involves the proximal innominate artery. A variety of operative exposures is useful but the selection of incision frequently depends upon the presence or absence of associated mediastinal injuries. Partial or complete median sternotomy in combination with various cervical and thoracic extensions is advised. Successful management of innominate artery injury can be performed without the aid of cardiopulmonary bypass or arterial shunts.

Both penetrating and blunt cardiac injuries require urgent management. Delayed sequelae and complications have been reported to occur in 4 to 56% of survivors and frequently required secondary surgical corrections. Between January 1980 and June 1984, 204 patients presented with heart injuries. One hundred twenty-eight of these patients survived. Of the 90 undergoing cardiorrhaphy in the operating room, 78 survived. Two-dimensional (2-D) and pulsed-Doppler echocardiography (echo) and/or cardiac catheterization were performed in 40 patients for suspected significant residual injury. Eight of the 40 required secondary cardiac operations. Two-dimensional echo demonstrated pericardial effusion, abnormal chamber enlargement, abnormal cardiac wall motion, intracardiac missile, and intracardiac and pleural thrombus. Pulsed-Doppler findings included ventricular septal defect, tricuspid insufficiency and right ventricular turbulence secondary to arteriovenous fistulae. Cardiac catheterization detected lesions undetected by 2-D echo in one patient with a gunshot wound. It is concluded that in the absence of cardiac bullets seen on routine chest X-ray, 2-D with pulsed-Doppler echo is an excellent screening technique. With a positive echo study or a residual bullet overlying the cardiac silhouette, cardiac catheterization is indicated in selected patients.

Greater civilian use of firearms and improved transportation and resuscitation of the injured have provided our institutions with an increasing experience with subclavian vascular injuries. Ninety-three patients with subclavian vascular injuries are presented and two time periods are compared. Principles of management gained from the earlier experience have been utilized with a decline in mortality to 4.7% among the patients admitted with a palpable pulse or blood pressure. Successful treatment as before still lies in the recognition of the severity of the injury, rapidity of preparation for operation, and adequacy of surgical exposure. Recent trends have included an increased reliance on selective arteriography when the patient is stable, extensive use of the 'book' thoracotomy as a primary incision, preoperative and intraoperative autotransfusion, and a more frequent use of interposition grafting for vascular repair. Primary arterial repair was seldom accomplished; most patients required segmental resection with end-to-end anstomosis or interposition grafts.

A 20-year retrospective evaluation of vascular trauma in infants and children was undertaken. The study included 53 cases of blunt and penetrating vascular injuries in pediatric patients. There were 36 males and 17 females ranging in age from 24 days to 14 years (average, 10 years). The most frequently encountered sites of arterial trauma were the brachial or superficial femoral artery, and of venous trauma the inferior vena cava. Any patient presenting to the Emergency Center with an injury in proximity to a major vessel, hematoma formation, audible bruit, or palpable thrill underwent prompt arteriography or immediate operative exploration of the injury sit. All patients in the series were managed operatively. There were 41 major arterial and 32 major venous injuries. No patient required a major amputation. Most injuries were repaired by primary closure or segmental resection and end-to-end anastomosis; interposition vein grafts and substitute conduits were used in four patients with more extensive injuries. A 13% operative mortality was encountered: the most frequent cause of death was intraoperative exsanguinating hemorrhage. The triad for successful management of vascular trauma in pediatric patients is: 1) a high index of suspicion, 2) performance of aggressive diagnostic studies when indicated, and 3) prompt surgical intervention.

Recovery of intrathoracic and intraperitoneal blood and reinfusion by autotransfusion has been demonstrated to be safe and practical in selected trauma patients. Autotransfusion is ideally applicable to the trauma patient in whom replacement of six or fewer units of blood is required. In addition, autotransfusion provides readily available blood for patients with unusual blood types and for those in whom multiple transfusions may rapidly deplete available stores. The properties of an ideal autotransfusion device include rapid assembly, relatively low cost, ease of operation, in-line filtration, minimized air blood interface, simplified anticoagulation, and safety from air embolism and coagulopathies.

Records of 373 patients with penetrating wounds of the lung seen at the Ben Taub General Hospital over a 1-year period were reviewed. Intercostal tube thoracostomy was the only therapy required in 282 patients. Thoracotomy was performed in 91 patients with repair of a pulmonary lesion in only 45 patients. Pneumonorrhaphy was performed in 33 patients, segmentectomy in six, and lobectomy in two. Four patients required repair of tracheal injuries. Fourteen patients initially treated with intercostal tube drainage required thoracotomy for complications of clotted hemothorax in eight and empyema in six. There were 29 deaths. Penetrating lung trauma in the majority of patients may be treated conservatively with a low incidence of infection or complication. Of the patients who require thoracotomy, associated injuries will frequently represent the major operative indication. Early thoracotomy for complication of clotted hemothorax or empyema is encouraged.

During a 9-year period, 16 patients with a delay in diagnosis of an injury to the diaphragm after a penetrating wound were treated. The left hemidiaphragm was involved in 15 of 16 patients, and the delay in diagnosis from the time of arrival in the emergency center immediately after injury ranged from 16 hours to 14 years. In the patients in the Acute Group (delay of hours to days), three patients had diaphragmatic defects missed at the time of laparotomy, three patients had chest X-rays not immediately suggestive of diaphragmatic defects, two patients had false-negative lavages, and one patient treated elsewhere did not have a chest X-ray in the emergency room. In the patients in the Chronic Group (hernias presenting months to years after injury), four of seven patients had misreading of a recent chest X-ray or failure to have a chest X-ray performed during numerous return visits to the emergency center. Despite a variety of diagnostic maneuvers, these defects and hernias continue to be diagnosed after a delay. Careful review of early and late followup chest X-rays appears to be the easiest mechanism to avoid significant delays in diagnosis.

During an 18-month period, among 35,000 injured patients presenting to an urban trauma center, more than 3,500 were admitted, and 1,500 underwent surgery on the General Surgical Service. Three hundred fifty-two patients with prehospital systolic blood pressures of less than 90 mm Hg were randomized on an alternate day basis to receive treatment with 'MAST' (163 patients) or 'No-MAST' (189 patients). Age, mechanism of injury, prehospital management times, prehospital trauma scores, prehospital fluids administered, Injury Severity Scores, emergency center treatment, operative protocol, and calculated probability of survival were virtually identical for both groups. Eighty-eight per cent of the injuries were produced by penetrating wounds. The predominant injury area was the abdomen in 40%, the thorax in 37%, and the extremities in 13%. There was no statistically significant difference in evaluation and outcome data between the groups. Within a controlled catchment, only 2.5% of injured patients and 22% of those undergoing operation qualified for prehospital use of MAST. We conclude that for penetrating trauma with prehospital times of 30 minutes or less, MAST provide no advantage with regard to survival, length of hospital stay, or reduced hospital costs.

Traumatic injuries of the portal venous circulation occur infrequently but often pose management problems. During the past 13 years, 94 patients have been treated at the Ben Taub General Hospital for traumatic injury to the portal venous system, including 37 portal, 45 superior mesenteric, seven splenic, and nine inferior mesenteric venous injuries. Injury resulted from penetrating wounds in all but 17 patients. Associated vascular injuries were present in 80 patients. Twenty-six patients died from failure to control hemorrhage. Six deaths resulted from postoperative complications. Operative approach necessitated lateral venorrhaphy in 66 patients, ligation in 23, end-to-end anastomosis in one, saphenous vein mesocaval shunt in two, end-to-side portacaval shunt in one, and clamping and packing in five. In spite of numerous associated vascular and visceral injuries, portal venous injuries can be successfully managed utilizing generally available vascular reconstructive techniques. Language: en

In the last ten years, 22 patients with transection of the descending thoracic aorta were seen at the Ben Taub General Hospital. Five patients with massive left hemothorax died shortly after admission. Among 17 patients in whom proximal and distal control could be achieved, three died in the operating room; three died within 30 days of operation, and one patient died of sepsis more than 1 year postinjury: ten patients are longterm survivors. Partial pump bypass was utilized in six patients, and in one a heparinized shunt was used intraoperatively. Dacron tube grafts were used in 12 cases, and primary repair was accomplished in three. One patient was paraplegic on admission, and two others in whom this was not established before operation were found to have this complication postoperatively. One patient with an infected false aneurysm 1 year after operation underwent resection of the distal aortic arch and Dacron graft with extensive extravascular bypass procedures. Patients with transection of the descending thoracic aorta die if proximal control is not achieved before rupture of the hematoma. Successful repair of this injury requires aggressive diagnostic and surgical management.

Review of a case involving simultaneous management of injuries to the brachial, axillary, and subclavian arteries stresses the need for arteriography in suspected subclavian injuries and the need for knowledge of the operative exposures available to the operating surgeon.

External cardiac compression (ECC) was originally developed for patients with nontraumatic cardiac conditions, but it is now used for a wide variety of emergency conditions. As an integral part of cardiopulmonary resuscitation (CPR), ECC coupled with forced pulmonary ventilation may NOT be applicable to cases of cardiac arrest following penetrating and blunt thoracic and abdominal trauma. Review of 100 patients with penetrating or blunt truncal trauma who received CPR and ECC more than 3 minutes prehospital revealed NO survivors despite continued aggressive resuscitative efforts in 49 of the patients upon arrival at a trauma center. Major cardiovascular disruption was found at thoracotomy or autopsy in all patients. In another 12 patients receiving forced ventilation and prehospital ECC, air embolism to the coronary arteries was the cause of death. CPR by paramedics, physicians, nurses, or lay persons does not appear to be of value in patients who have sustained cardiac arrest from truncal trauma.

In a group of seven patients with percutaneous subclavian catheters in the pleural space, three were found to have continued blood loss from the catheter tip, and in four patients thoracotomy was prevented by discontinuing blood administration through the subclavian catheter during a re-evaluation phase.

Traumatic injuries to the upper abdominal vasculature pose difficult management problems related to both exposure and associated injuries. Among those injuries that are more difficult to manage are those involving the portal vein. While occurring rarely, portal vein injuries require specific therapeutic considerations. Between January, 1968, and July, 1974, over 2000 patients were treated operatively for abdominal trauma at the Ben Taub General Hospital. Among these patients, 22 had injury to the portal vein. Seventeen portal vein injuries were secondary to gunshot wounds, 3 to stab wounds, and 2 to blunt trauma. Associated injuries to the inferior vena cava, pancreas, liver and bile ducts were common. Three patients had associated abdominal aortic injuries, two with acute aorto-caval fistulae. Nine patients died from from failure to control hemorrhage. Eleven were long-term survivors, including two who required pancreataico-duodenectomy as well as portal venorrhaphy. Late complications were rare. The operative approach to patients with traumatic injuries to multiple organs in the upper abdomen, including the portal vein, requires aggressive management and predetermined sequential methods of repair. In spite of innumerable associated injuries, portal vein injuries can be successfully managed in a significant number of patients using generally available surgical techniques and several adjunctive maneuvers.

From 1978 through 1983, 206 patients had 236 polytetrafluoroethylene (PTFE) grafts inserted in vascular wounds. More than 85% of injuries were due to gunshot wounds, shotgun wounds, or stab wounds. Arterial grafts were inserted into vessels of the upper extremity (38.8%), lower extremity (46.1%), neck and chest (8.8%), and abdomen (6.3%). Grafts were most commonly placed in the brachial or superficial femoral arteries. Venous grafts were more commonly inserted into vessels of the extremities (96.7%), with the majority located in the superficial femoral vein. PTFE was found to be an acceptable prosthesis for interposition grafting in arterial wounds, but long-term patency was less than that seen when interposed saphenous vein grafts are used. Early and late occlusions were a significant problem with 4-mm PTFE grafts in the brachial artery, and this size is not recommended in this location. Peripheral PTFE graft infection did not occur in the absence of exposure of the graft or of osteomyelitis. Exposed grafts did not fare well and early coverage is recommended, even with extensive soft-tissue wounds around the graft. PTFE grafts inserted in proximal extremity veins are excellent temporary conduits which decrease hemorrhage in blast cavities and fasciotomy sites, but all grafts studied by venography at 7 to 14 days were either narrowed or occluded.