Delayed hemothorax after blunt thoracic trauma: an uncommon entity with significant morbidity.

OBJECTIVE: To describe the nature of delayed hemothorax occurring after blunt thoracic trauma and to identify the population at risk for this complication. METHODS: A retrospective review was conducted of 36 consecutive patients with hemothorax consequent to blunt trauma. Criteria for the definition of delayed hemothorax were established involving normal interval chest radiographs or computed tomographic scans during hospitalization. RESULTS: Twelve cases of delayed development of hemothorax were identified. Ninety-two percent of cases occurred in patients with multiple or displaced rib fractures. Presentation occurred from 18 hours to 6 days after injury. Eleven of the 12 cases were heralded by a prodrome of new pleuritic chest pain and dyspnea that occurred from 4 to 19 hours before treatment. CONCLUSION: Delayed hemothorax after blunt trauma is a unique entity occurring in patients with multiple or displaced rib fractures. Vigilance for the recognizable prodrome in the high-risk population should allow early remediation of this complication.

Long-term functional status and mortality of elderly patients with severe closed head injuries.

OBJECTIVE: To evaluate long-term clinical outcome of elderly patients with severe closed head injuries. DESIGN: Retrospective study. PATIENTS AND METHODS: All patients older than 65 years of age admitted to a regional trauma center with a diagnosis of closed head injury and an admission Glasgow Coma Scale (GCS) score of 8 or less. Using chi 2 analysis, Student's t test, and multiple logistic regression, we correlated age, sex, mechanism of injury, pupillary reactivity, alcohol and drug use, admission GCS score, Injury Severity Score, Revised Trauma Score, heart rate, and blood pressure to the main outcome measures, i.e. long-term functional outcome and mortality. RESULTS: Among 40 elderly patients who met the criteria, 27% were still alive at the end of 38 +/- 3 month follow-up. Eighty-five percent of patients who were discharged from the hospital were still alive long-term, but did not show significant neurologic improvement. In a univariate analysis, GCS and pupillary reactivity were predictive for long-term functional outcome and mortality. In a multivariate analysis, GCS and heart rate were predictive. All patients with an admission GCS score of 3 died in-hospital. All patients with an admission GCS score of 3 to 7 were either deceased or lived in persistent vegetative or dependent functional states. CONCLUSIONS: Elderly patients with severe closed head injuries have high in-hospital mortality. Those who survived the hospital stay had high long-term survival, but did not show significant functional improvement. Prediction of long-term functional status is vital to the trauma care of elderly patients with severe closed head injuries.

Gastric perforation in a case of gastric volvulus.

We describe a case of organoaxial gastric volvulus in which the axis was through the body of the stomach. The intraabdominal part of the volvulus was complicated by bleeding from two necrotic ulcers, one of which was perforated by the tip of a nasogastric tube.

Vehicular trauma triage by mechanism: avoidance of the unproductive evaluation.

An instrument was developed using routinely available field data to identify the sizable subgroup of stable vehicular trauma victims initially triaged to the trauma center by mechanism indicators alone who are in reality at minimal risk for serious injury. The six most common vehicular mechanism indicators seen at a level I trauma center were evaluated: rollover, head-on greater than 30 mph, intrusion, prolonged extrication, other death in same vehicle, and ejection. Review of 1235 consecutive trauma team activations yielded 349 victims with a qualifying vehicular mechanism. Outcome indicators were used to classify patients into two groups: Minor Injury (MI) and Severe Injury (SI). Nineteen common field data elements routinely reported on arrival by the regional Emergency Medical Service (EMS) personnel were then reviewed. Data patterns associated only with the MI group were sought. A checklist was developed for Mechanism vehicular trauma utilizing physiologic, anatomic, and neurologic elements. A single positive element would define trauma team activations. Retrospectively, use of this instrument would have excluded 56% of the MI group from unproductive trauma team referral, but nearly none of the SI group. We conclude that an identifiable subset of trauma patients referred by vehicular mechanism criteria alone could be safely evaluated on arrival in the emergency department as a form of secondary triage rather than by referral to the trauma team. The use of an appropriate exclusionary instrument can still preserve the sensitivity of trauma team activation for severely injured victims.

Preceptored introduction of laparoscopic techniques for cholecystectomy into a large university-affiliated medical center.

Faced with the task of introducing laparoscopic techniques for cholecystectomy into the practice of a large department composed of individuals with varied backgrounds and experience, our surgical staff decided to grant provisional provileges to five surgeons, two from the full-time faculty and three from the community, who had completed a formal course in laparoscopic cholecystectomy. These five surgeons agreed to assist one another through 10 cases a piece before performing any procedures on their own or serving as preceptors for additional surgeons. Other surgeons could obtain credentials for this procedure by satisfying the same course criteria and receiving assistance from one of the five original surgeons during their first 10 cases. In the 14 months after September 1990, 250 laparoscopic cholecystectomies were performed by 19 different attending surgeons at our hospital. One death from hemorrhage (0.4%) and two bile duct injuries (0.8%) occurred in these patients. One of the bile duct injuries occurred after conversion to open cholecystectomy, as did the hemorrhage, which was from a vessel within the parenchyma of the gallbladder bed which rebled even after temporary control through open ligature technique. The second bile duct injury, the result of injudicious application of hemoclips for hemostasis, was minor in degree and the only injury to occur in a procedure conducted exclusively through the laparoscope. This experience demonstrates that laparoscopic techniques can be safely introduced into an environment involving multiple surgeons by adherence to a careful protocol of preceptored assistance.

Systemic and regional hemodynamic effects of cyclo-oxygenase and thromboxane synthetase inhibition in normal and hyperdynamic endotoxemic rabbits.

We tested the hypothesis that prostaglandins (PGs) and thromboxane (Tx) A2 are important mediators of the hemodynamic derangements occurring in a rabbit model of hyperdynamic endotoxicosis. Rabbits were injected with either normal saline (NS) or Escherichia coli lipopolysaccharide (LPS; 1-3 micrograms/kg) and studied 6 hr later. Cardiac index (CI) and regional blood flow were determined using thermodilution and radioactive microspheres, respectively. Systemic and regional hemodynamics were determined before and 40 min after administering indomethacin (cyclo-oxygenase inhibitor; 5 mg/kg), UK38485 (Tx synthetase inhibitor; 10 mg/kg), or NS. LPS increased CI (P = .0024) and decreased mean arterial pressure (P = .0031) and systemic vascular resistance index (P = .0001). LPS increased flow to the heart and small intestine and decreased flow to the hepatic artery and pancreas. The systemic and regional hemodynamic effects of indomethacin were similar in NS- and LPS-treated rabbits. UK38485 decreased perfusion of skeletal muscle and diaphragm in both endotoxemic and control animals. This agent increased splenic perfusion only in NS-treated rabbits. Plasma levels of 6-keto PGF1 alpha (PGI2 metabolite) were typically undetectable in both NS- and LPS-treated rabbits. These data do not support the hypothesis that PG's or TxA2 are major determinants of the hemodynamic perturbations that occur in this endotoxicosis model.

Systemic and regional hemodynamic changes after intraperitoneal endotoxin in rabbits: development of a new model of the clinical syndrome of hyperdynamic sepsis.

Rabbits were injected intraperitoneally with sterile saline or Escherichia coli endotoxin (LPS; 50 or 500 micrograms/kg). Sixteen to 18 h later, the animals were anesthetized with ketamine and instrumented to permit measurement of mean arterial pressure, cardiac output (by thermodilution), and regional blood flow (using radioactive microspheres). The animals were allowed to waken fully in a plastic restraining cage prior to measuring systemic and regional hemodynamics. LPS had similar effects regardless of dose, and results from the two dosage groups have been combined. Compared to controls, administrations of LPS resulted in a 29.3% increase in cardiac output (P = .011) and a 22.8% decrease in systemic vascular resistance (P = .0009). Injection of LPS caused significant changes in blood flow to the heart (55.3% increase), small intestine (128.3% increase), portal vein (54.7% increase), and hepatic artery (65.0% decrease). The percentage of cardiac output perfusing the kidneys and hepatic artery was significantly decreased in the endotoxemic group (P = .037 and P = .002, respectively). Injecting LPS resulted in increased relative flow to the heart (P = .024), small intestine (P = .049), and portal vein (P = .041). We conclude that this model reproduces several of the systemic hemodynamic features of the sepsis syndrome in humans. In this model, the hyperdynamic state is associated with vasodilatation in mesenteric (small intestine and colon) and coronary beds and vasoconstriction in the hepatic artery.