Faculty Bibliography

The most significant contribution to the management of hepatic injuries over the past 5 years has been the nonoperative management of blunt injuries in the adult patient. Recent data suggest that as many as 80% of all blunt hepatic injuries may be treated in this fashion, with a success rate exceeding 95%. The fear of missing hollow viscus injuries, as well as the risk of sudden hemorrhage in the observational period, leading to an increase in hepatic-related deaths, seems exaggerated. The intraoperative management of complex hepatic injuries revolves around strict adherence to resuscitation prior to addressing the lesion itself. At times, 'damage control' with termination of surgery and 'packing' the patient with planned re-exploration are critical, as these maneuvers are often lifesaving. The Pringle maneuver and intrahepatic hemostasis for grades III to IV injuries have resulted in a mortality rate under 10%. Juxtahepatic venous injuries continue to carry an inordinately high mortality rate. Intracaval shunts, when used, should be inserted early in the course of the operation before excess transfusions are given and acidosis and hypothermia develop

INTRODUCTION: While nonoperative management of blunt splenic injury in the stable patient has become the standard of care, splenectomy is still advocated as the safest management for rupture of the diseased spleen. The combination of splenectomy and underlying immunosuppression may render these patients particularly susceptible to postsplenectomy infection, and thus we undertook a prospective trial of nonoperative management of the ruptured pathologic spleen. METHODS: Hemodynamically stable patients with preexisting pathologic splenomegaly and isolated splenic disruptions diagnosed by computed tomographic (CT) scan (American Association for the Surgery of Trauma (AAST) grades 1-4) requiring 2 or less units blood transfusion were prospectively studied. Patients were monitored in a critical care setting, and resolution of splenic disruption was followed by serial CT examinations. RESULTS: Nonoperative management was successful in all 11 patients (eight, HIV/AIDS; one each, acute leukemia, infectious mononucleosis, sickle cell anemia). The mean transfusion requirement was 0.7 units; the mean length of stay was 16 days. CONCLUSIONS: The pathologic spleen can heal after parenchymal disruption. While not appropriate for all patients, a subset of hemodynamically stable patients can be successfully managed nonoperatively using CT diagnosis, close clinical monitoring, and minimal transfusions

INTRODUCTION: Nonoperative management is presently considered the treatment modality of choice in over 50% of adult patients sustaining blunt hepatic trauma who meet inclusion criteria. A multicenter study was retrospectively undertaken to assess whether the combined experiences at level I trauma centers could validate the currently reported high success rate, low morbidity, and virtually nonexistent mortality associated with this approach. Thirteen level I trauma centers accrued 404 adult patients sustaining blunt hepatic injuries managed nonoperatively over the last 5 years. Seventy-two percent of the injuries resulted from motor vehicle crashes. The mean injury severity score for the entire group was 20.2 (range, 4-75), and the American Association for the Surgery of Trauma-computerized axial tomography scan grading was as follows: grade I, 19% (n = 76); grade II, 31% (n = 124); grade III, 36% (n = 146); grade IV, 10% (n = 42); and grade V, 4% (n = 16). There were 27 deaths (7%) in the series, with 59% directly related to head trauma. Only two deaths (0.4%) could be attributed to hepatic injury. Twenty-one (5%) complications were documented, with the most common being hemorrhage, occurring in 14 (3.5%). Only 3 (0.7%) of these 14 patients required surgical intervention, 6 were treated by transfusions alone (0.5 to 5 U), 4 underwent angio-embolization, and 1 was further observed. Other complications included 2 bilomas and 3 perihepatic abscesses (all drained percutaneously). Two small bowel injuries were initially missed (0.5%), and diagnosed 2 and 3 days after admission. Overall, 6 patients required operative intervention: 3 for hemorrhage, 2 for missed enteric injuries, and 1 for persistent sepsis after unsuccessful percutaneous drainage. Average length of stay was 13 days. Nonoperative management of blunt hepatic injuries is clearly the treatment modality of choice in hemodynamically stable patients, irrespective of grade of injury or degree of hemoperitoneum. Current data would suggest that 50 to 80% (47% in this series) of all adult patients with blunt hepatic injuries are candidates for this form of therapy. Exactly 98.5% of patients analyzed in this study successfully avoided operative intervention. Bleeding complications are infrequently encountered (3.5%) and can often be managed nonoperatively. Although grades IV and V injuries composed 14% of the series, they represented 66.6% of the patients requiring operative intervention and thus merit constant re-evaluation and close observation in critical care units. The optimal time for follow-up computerized axial tomography scanning seems to be within 7 to 10 days after injury

OBJECTIVE: Injuries to the portal triad are a rare and complex challenge in trauma surgery. The purpose of this review is to better characterize the incidence, lethality, and successful management schemes used to treat these injuries. DESIGN: A retrospective review of the experience of eight academic level I trauma centers over a combined 62 years. RESULTS: A retrospective review of the experience of eight anatomical structures of the portal hepatis: 118 injuries to the anatomical structures of the portal hepatis: 55 extrahepatic portal vein injuries, 28 extrahepatic arterial injuries, and 35 injuries to the extrahepatic biliary tree. Sixty-nine percent of the injuries were by penetrating mechanism and 31% were by blunt mechanism. All patients had associated injuries with a mean Injury Severity Score of 34 in blunt trauma patients. Overall mortality was 51%, rising to 80% in patients with combination injuries. Sixty-six percent of deaths occurred in the operating room, primarily from exsanguination; 18% of deaths occurred within 48 hours of injury from refractory shock, coagulopathy, or cardiac arrest; 16% occurred late. Ten percent of patients undergoing portal vein ligation survived, compared to 58% managed by primary repair. Survival after hepatic artery ligation was 42%, compared to 14% after primary repair. Survival after biliary-enteric anastomosis as treatment of extrahepatic bile duct injury was 89%, compared to 50% after primary repair and 100% after ligation of lobar bile duct injuries. Missed bile duct injuries had a high (75%) severe complication rate. CONCLUSIONS: Injuries to the anatomical structures of the portal triad are rare and often lethal. Intraoperative exsanguination is the primary cause of death, and hemorrhage control should be the first priority. Bile duct injuries should be identified by intraoperative cholangiography and repaired primarily or by enteric anastomosis; lobar bile ducts can be managed by ligation

BACKGROUND: Severe blunt trauma to the torso can result in diaphragmatic disruption. Prompt recognition of this potentially life-threatening injury is difficult when the initial chest roentgenogram is unrevealing and immediate thoracotomy or celiotomy is not performed. This retrospective study was undertaken to: (1) determine the incidence of missed diaphragmatic injuries on initial evaluation; (2) identify factors contributing to diagnostic delays; and (3) formulate a diagnostic approach that reliably detects diaphragmatic rupture following blunt trauma. METHODS: Retrospective review of hospital records and radiographs from our 18-year experience with blunt diaphragmatic injuries. RESULTS: Seven of 57 (12%) blunt diaphragmatic injuries were missed on initial evaluation. Recognition followed 2 days to 3 months later. Two (4%) isolated left-sided injuries initially presented with normal chest roentgenograms. Five patients (9%) (4 with right-sided ruptures) had abnormalities on chest roentgenogram or computed tomography (CT) initially attributed to chest trauma. They were diagnosed by radionuclide, ultrasound, or CT investigations of hemothorax, pulmonary sepsis, and right upper quadrant pain; and, in 1 case, at thoracotomy for a persistent right hemothorax. In the remaining 50 patients (88%), the diagnosis was established within 24 hours. In 21 (42%) of these, the problem was initially recognized at the time of celiotomy for accompanying injuries. CONCLUSIONS: Blunt diaphragmatic injuries are easily missed in the absence of other indications for immediate surgery, since radiologic abnormalities of the diaphragm--particularly those involving the right hemidiaphragm--are often interpreted as thoracic trauma. In this setting, a high index of suspicion coupled with selective use of radionuclide scanning, ultrasound, and CT or magnetic resonance imaging is necessary for early detection of this uncommon injury

This review of 14 recent publications encompassing 495 patients highlights the current role of the nonoperative management of adult blunt hepatic injuries. When careful inclusion criteria were met, the most important of which is hemodynamic stability, a 94% success rate was achieved, clearly attesting to the safety and efficacy of this approach. A 0% liver-related mortality in these 495 patients was achieved, and there were no documented missed enteric injuries. Delayed hemorrhage that led to laparotomy occurred in 2.8% of patients. The mean length of hospital stay was 13 days, and the mean transfusion requirement was 1.9 units of blood per patient. Computed axial tomography scanning was essential and played an integral role in delineating the extent of the injury, identifying other intra-abdominal injuries that would mandate immediate laparotomy, and following the progress of injury resolution. Overall, 34% of blunt liver injuries were managed nonoperatively. As of 1993, however, available data confirms that 51% of adult reported blunt hepatic injuries have been treated nonoperatively. Rigid adherence to the described guidelines may allow the majority of blunt hepatic injuries to be treated nonoperatively. It should be stressed, however, that this method of patient management should only be undertaken at institutions where the appropriate resources necessary to deal with this patient population are readily available

Two unusual cases of liquid caustic ingestion that resulted in gangrene of the duodenum and adjacent colon, and burns of the esophagus, stomach, and pancreas are presented. The routine evaluation of the oropharynx, esophagus, and stomach after liquid caustic ingestion can seriously underestimate the extent of injury to distal portions of the gastrointestinal (GI) tract, such as the colon and pancreas, that are not usually included in the initial evaluation of ingestion injuries. In stable patients managed nonoperatively, the entire upper GI tract, including the duodenum, must be visualized either by endoscopy or, less preferably, by barium series. Double-contrast computed tomography should be performed when significant duodenal injuries are present in order to inspect the colon, pancreas, and small bowel. With this approach, life-threatening, multi-organ, subdiaphragmatic ingestion injuries can be identified and treated early

Complex hepatic surgery often requires occlusion of the portal triad in order to decrease parenchymal bleeding. This study was undertaken to evaluate the effects of topical hypothermia and intravenous steroids on liver ischemia by measuring adenosine triphosphate (ATP) levels within the hepatic parenchyma. Forty New Zealand white rabbits were divided into four experimental and four control groups. All experimental animals underwent laparotomy and ligation of the porta hepatis. Serial liver biopsy specimens were obtained at predetermined time intervals. Group I received no further intervention. Group II were topically cooled until intrahepatic temperature reached 30 degrees C. Group III received preligation intravenous methylprednisolone (30 mg/kg). Group IV received both steroids and topical hypothermia. The corresponding control groups underwent laparotomy and isolation of the porta without ligation. Adenosine triphosphate was extracted from the liver parenchyma and quantified by high-performance liquid chromatography (HPLC). The data were analyzed using a three-factor mixed analysis of variance (ANOVA). There was a statistically significant protective effect on ATP levels provided by topical hypothermia at 15 and 30 minutes of ischemia (p < 0.01), but not at 60 minutes (p > 0.05). Steroids were not found to have any protective effect on ATP levels at any time point. The combination of steroids and topical hypothermia provided significant preservation of hepatic parenchymal ATP levels, although less than that of hypothermia alone, at 15 and 30 minutes of ischemia (p < 0.01)

OBJECTIVE: The authors reviewed the experimental evidence, surgical technique, complications, and results of clinical trials evaluating the role of autologous splenic transplantation for splenic trauma. SUMMARY BACKGROUND DATA: Splenorrhaphy and nonoperative management of splenic injuries have now become routine aspects in the management of splenic trauma. Unfortunately, not all splenic injuries are readily amenable to conventional spleen-conserving approaches. Heterotopic splenic autotransplantation has been advocated for patients with severe grade IV and V injuries that would otherwise mandate splenectomy. For this subset of patients, splenic salvage by autotransplantation would theoretically preserve the critical role the spleen plays in the host's defense against infection. METHODS: The relevant literature relating to experimental or clinical aspects of splenic autotransplantation was identified and reviewed. Data are presented on the experimental evaluation of autogenous splenic transplantation, methods and complications of autotransplantation, choice of anatomic site and autograft size, and results of clinical trials in humans. RESULTS: The most commonly used technique of autotransplantation in humans involves implanting tissue homogenates or sections of splenic parenchyma into pouches created in the gastrocolic omentum. Most authors have observed evidence of splenic function with normalization of postsplenectomy thrombocytosis, immunoglobulin M levels, and peripheral blood smears. Some degree of immune function of transplanted grafts has been demonstrated with in vivo assays, but the full extent of immunoprotection provided by human splenic autotransplants is currently unknown. CONCLUSIONS: Multiple human and animal studies have established that splenic autotransplantation is a relatively safe and easily performed procedure that results in the return of some hematologic and immunologic parameters to baseline levels. Some aspects of reticuloendothelial function are also preserved. Whether this translates into a real reduction in the morbidity or mortality rates from overwhelming bacterial infection is unknown and requires further investigation