When More Is Still Not Enough: A Case of Ceftazidime-Avibactam Resistance in a Burn Patient.

Burn patients have numerous risk factors for multidrug-resistant organisms (MDROs) and altered pharmacokinetics, which both independently increase the risk of treatment failure. Data on appropriate antimicrobial dosing are limited in this population and therapeutic drug monitoring (TDM) for beta-lactams is impractical at most facilities. Technology is available that can detect genetic markers of resistance, but they are not all encompassing, and often require specialized facilities that can detect less common genetic markers. Newer antimicrobials can help combat MDROs, but additional resistance patterns may evolve during treatment. Considering drug shortages and antimicrobial formularies, clinicians must remain vigilant when treating infections. This case report describes the development of resistance to ceftazidime-avibactam in a burn patient. The patient was a 54-year-old burn victim with a 58% total body surface area (TBSA) thermal burn who underwent multiple courses of antibiotics for various Pseudomonal infections. The initial Pseudomonal wound infection was sensitive to cefepime, aminoglycosides, and meropenem. A subsequent resistant pseudomonal pneumonia was treated with ceftazidime-avibactam 2.5 g every 6 hours due to the elevated MIC to cefepime (16 mcg/mL) and meropenem (>8 mcg/mL). Although the patient improved over 7 days, the patient again spiked fevers and had increased white blood counts (WBC). Repeat blood cultures demonstrated a multidrug-resistant (MDR) Pseudomonas with a minimum inhibitory concentration (MIC) to ceftazidime-avibactam of 16 mcg/mL, which is above the Clinical and Laboratory Standards Institute (CLSI) breakpoint of 8 mcg/mL. At first, resistance was thought to have occurred due to inadequate dosing, but genetic work demonstrated multiple genes encoding beta-lactamases.

A Cautionary Tale of Etanercept Use in Patients With Toxic Epidermal Necrolysis.

Toxic epidermal necrolysis (TEN) is a severe cutaneous reaction that can be life-threatening. In the United States, there are no established guidelines for the treatment of TEN. Supportive care including fluids and supportive therapies are the current recommendations. Research surrounding TEN involves mostly case studies or small, uncontrolled studies. Recent literature describes the use of tumor necrosis factor blockers in the treatment of TEN with positive results. These case reports describe decreased time to reepithelization, hospital length of stay, and minimal side effects. Conversely, we present three fatalities after the administration of etanercept.

Impact of Genitourinary Injuries on Patients Requiring an Emergency Laparotomy for Trauma.

Introduction In patients having emergency abdominal surgery for trauma, the presence of urologic injury tends to increase mortality and morbidity. Methods This retrospective study evaluated patients requiring emergency surgery for abdominal trauma at a Level 1 Trauma Center over 30 years (1980-2010). Special attention was given to patients with concomitant genitourinary (GU) injuries. Results Of 1105 patients requiring an emergency laparotomy for trauma, 242 (22%) had urologic injuries including kidney 178 (16%), ureter 47 (4%), and bladder 46 (4%). Of the 242 patients, 50 (20%) died early (<48 hours) and 13 (5%) died later, primarily due to infection. A concept of "seven deadly signs" of hypoperfusion was developed. In patients with GU injuries, the presence of any deadly sign of hypoperfusion increased the mortality rate from 4% (6/152) to 63% (56/90), p<0.001. Of the 53 patients having a nephrectomy, 36 (68%) had one or more deadly signs and 27 (75%) died. Of 17 without deadly signs, only 2 (12%) died (p=0.001). Of 167 GU patients receiving blood, 59 (35%) developed infection vs 3/75(4%) in those receiving no blood (p<0.001). Conclusions The presence of deadly signs of severe injury and hypoperfusion on admission was the major factor determining mortality. With a severely injured kidney plus any deadly signs of hypoperfusion, special efforts should be made to avoid a nephrectomy.

Intravenous pantoprazole utilization in a level 1 trauma center.

BACKGROUND: In recent years there has been a rapid increase in the use of proton pump inhibitors. Our institution has recently had several shortages of IV pantoprazole, each lasting 7-10 days. The purpose of our study was to evaluate in-patient usage of IV pantoprazole. We hypothesized that hospitalized patients with upper gastrointestinal bleeding (GIB) or risk for stress ulcers inappropriately received IV pantoprazole based on current literature. METHODS: This was a retrospective study of 165 consecutive in-patients identified as receiving pantoprazole from December 2004 to March 2005. Only patients receiving IV pantoprazole were included (n = 78). Data collected included demographics, indication and dosing of pantoprazole, admitting team (surgery vs. medicine), and risk factors for stress ulcers. RESULTS: Our study population had a mean age of 54 +/- 17 years and 62% were male. Overall, 45% (35/78) of patients receiving IV pantoprazole had an appropriate indication, and 19% (15/78) received the correct dose. Of the 78 patients, 43 (55%) were treated with pantoprazole for stress ulcer prophylaxis (SUP), and 35 (45%) patients were treated for GIB. We found that none of the 43 patients treated for SUP had an appropriate indication for pantoprazole, but all of the patients with GIB (35) had an appropriate indication. Of the 35 patients treated for GIB with pantoprazole, only 40% (14/35) received the correct dose. In all cases of incorrect dosing, the patients were underdosed. CONCLUSIONS: Pantoprazole is not being prescribed appropriately for stress ulcer prophylaxis in our patient population. Even in patients appropriately receiving pantoprazole the majority were prescribed an incorrect dose. Appropriate indications and dosing of pantoprazole could eliminate the shortages seen at our institution.

Gastric injury increases infections in trauma patients.

Some physicians feel gastric injury is not a significant contributing factor to the adverse outcome of trauma patients, but rather a marker of epigastric injury. We hypothesized the addition of a gastric injury to multiple injured trauma patients would increase infection rate. We conducted a retrospective study comparing 450 consecutive patients with full-thickness gastric injury with 983 patients without gastric injury during the same time period. Infection rate in patients with gastric injury was 44 per cent (200 of 455) and significantly higher than 36 per cent (357 of 983) seen without gastric injury (P = 0.006). Logistic regression revealed gastric injury was an independent risk factor for infection controlling for age, Injury Severity Scale, gender, mechanism of injury, shock, and associated injuries (P = 0.047). Requiring a transfusion, Injury Severity Scale, colon injury, age, pancreas injury, and emergency department shock were also independent risk factors for developing an infection. The addition of a gastric injury to a trauma patient appears to increase the risk for infection.

Post-traumatic hypotension: should systolic blood pressure of 90-109 mmHg be included?

It is generally accepted that patients with a systolic blood pressure (SBP)<90 mmHg are in "shock" and have a worse prognosis than patients with a higher SBP. Our objective was to determine if patients with a SBP of 90-109 mmHg have a worse outcome than patients with a higher SBP following trauma. Patients with gastric, small bowel, and/or diaphragm injuries were identified retrospectively through the trauma database from 1980-2003. All 2071 patients underwent emergent laparotomy at an urban, level one trauma center. The mortality rate of patients with a SBP of 90-109 mmHg in the ED or OR was 5% (17/354) and significantly higher than the 1% (12/1020) mortality seen in patients with a SBP of 110 mmHg or greater (P<0.001). The average length of stay of patients with a SBP of 90-109 mmHg was 15+/-14 days and was significantly longer than the 11+/-11 days seen in patients with a higher SBP. If the SBP was 90-109 mmHg, the infection rate was 39% (131/340), and this was significantly higher than the 22% (219/1016) infection rate seen in patients with higher SBP (P<0.001). Trauma patients with a systolic blood pressure of 109 mmHg or below are at increased risk for morbidity and mortality following trauma. Patients with a systolic blood pressure of 90-109 mmHg following trauma should be considered as a special group requiring aggressive resuscitation and surgery. Early operative control of hemorrhage in these patients can reduce mortality and infection.

Factors affecting prognosis in patients with gastric trauma.

Morbidity and mortality after gastric injury is usually the result of associated injuries. The authors conducted a retrospective study of 544 consecutive patients with gastric trauma requiring emergency surgery. Blunt injuries had the highest mortality and length of stay. The mortality of a proximal stomach injury was 43 per cent (9 of 21) and was significantly higher than the 19 per cent mortality seen in patients with more distal injuries (P < 0.01). The majority of gastric injuries were closed primarily (492 of 544 or 90%). The patients requiring more than a primary repair had a higher mortality (22 of 52 or 42% vs. 87 of 492 or 18%; P < 0.001), required more blood (16+/-16 U vs. 6+/-11 U; P < 0.001), had an increased rate of surgical site infections (17 of 52 or 33% vs. 75 of 492 or 15%; P = 0.001), and had an increased length of stay (20+/-30 days vs. 13+/-18 days; P = 0.024). There were 22 patients with an isolated gastric injury, and all of these patients survived. Patients with an associated arterial injury had the highest mortality (49%) and highest incidence of shock (64%). Patients with colon and gastric injuries had the highest (48 of 176 or 52%) surgical site infection rate. Isolated gastric injury is rare, but is associated with low morbidity and mortality. The mechanism of injury, location of injury, and type of repair used all affect patient outcomes with gastric injury.

Vascular endothelial growth factor modulates contractile response in microvascular lung pericytes.

BACKGROUND: Pericytes are capillary support cells that may play a role in regulating permeability by their contractile responses. Vascular endothelial growth factor (VEGF) may play a role in the increased permeability found in sepsis and other inflammatory conditions. The purpose of this study was to evaluate the role of VEGF in regulating pericyte contraction. METHODS: Rat microvascular lung pericytes were isolated according to previously described methods and cultured on collagen gel matrices. Cells were exposed to VEGF (10, 100, and 1000 pg/mL) for varying time periods (0, 10, 30, 60, and 120 minutes). The gels were released and their contractile responses digitally quantified. RESULTS: At all doses, VEGF induced initial pericyte relaxation (contraction 85% to 90% of controls; P < .001). This was followed-up by increased and sustained contraction (107% to 120% of controls; P < .01). CONCLUSIONS: VEGF modifies the contractile response of microvascular lung pericytes. This mechanism may play a role in the increased permeability demonstrated in inflammatory states.

Hydromorphone vs fentanyl for epidural analgesia and anesthesia.

BACKGROUND: Epidural analgesia/anesthesia is used during surgery because it dramatically relieves pain and attenuates the stress response. Because limited data exist regarding the relative merits of hydromorphone (HM) and fentanyl (FENT), the objective was to determine which was more safe and effective. METHODS: Prospective case-matched, observational study evaluated elective surgery patients: 30 HM and 60 FENT. Variables were measured perioperatively. RESULTS: Of the 90 patients, mean age was 52 years; simplified acute physiology score was 26 ± 10; and American Society of Anesthesiologists score was 2.4 HM vs 2.7 FENT, P = .03. HM patients were more apt to be excessively sedated (16% HM vs 1% FENT, P = .007) and have poor mental unresponsiveness (6% HM vs 0% FENT, P = .04). The incidence of hypotension was not different, 76% HM vs 80% FENT, not significant. CONCLUSIONS: In a closely case-matched population, FENT caused less excessive sedation and unresponsiveness. FENT patients had better intraoperative urine output and tended to have less repeated episodes of hypotension.

Impact of inappropriate initial antibiotics in critically ill surgical patients with bacteremia.

BACKGROUND: Bloodstream infections in critically ill patients are associated with mortality as high as 60% and a prolonged hospital stay. We evaluated the impact of inappropriate antibiotic therapy (IAAT) in a critically ill surgical cohort with bacteremia. METHODS: This retrospective study evaluated adults with intensive care unit admission greater than 72 hours and bacteremia. Two groups were evaluated: appropriate antibiotic therapy (AAT) vs IAAT. RESULTS: In 72 episodes of bacteremia, 57 (79%) AAT and 15 (21%) IAAT, mean age was 54 ± 17 years and APACHE II of 17 ± 8. Time to appropriate antibiotics was longer for IAAT (3 ± 5 IAAT vs 1 ± 1 AAT days, P = .003). IAAT was seen primarily with Acinetobacter spp (33% IAAT vs 9% AAT, P = .01) and Enterococcus faecium (26% IAAT vs 7% AAT, P = .03). If 2 or more bacteremic episodes occurred, Acinetobacter spp. was more likely, 32% vs 2%, P = .001. CONCLUSIONS: AAT selection is imperative in critically patients with bacteremia to reduce the significant impact of inappropriate selection. Repeated episodes of bacteremia should receive special attention.

Impact of minimizing diagnostic blood loss in the critically ill.

BACKGROUND: The use of a small-volume phlebotomy tube (SVPT) versus conventional-volume phlebotomy tube (CVPT) has led to a decrease in daily blood loss. Blood loss due to phlebotomy can lead ultimately to decreased rates of anemia and blood transfusions, which can be important in the critically ill patient. METHODS: We compared SVPT vs CVPT retrospectively in critically ill adult patients age ≥18 years admitted to a surgical intensive care unit for ≥48 hours. CVPT were evaluated from January 2011 to May 2011 and SVPT from June 2012 to October 2012. RESULTS: Amount of blood drawn for laboratory tests and transfusions were evaluated in 248 patients (116 SVPT vs 132 CVPT). When compared with CVPT, total blood volume removed (mean ± SD) with SVPT was less overall, 174 ± 182 mL vs 299 ± 355 mL, P = .001. Daily blood draws also were less, 22.5 ± 17.3 mL vs 31.7 ± 15.5 mL, P < .001. The units of packed red blood cells given were not significant, 4.4 ± 3.6 units vs 6.0 ± 8.2 units, P = .16. CONCLUSION: The use of SVPT blood sampling led to a decreased amount of blood drawn. Strategies that use SVPT in a larger cohort also may decrease the number of transfusions in selected patients. Every effort should be made to use SVPT.

Endothelial cells protect against lipopolysaccharide-induced caspase-3-mediated pericyte apoptosis in a coculture system.

BACKGROUND: Cells that comprise the pulmonary capillary walls, the pericytes and endothelial cells, may undergo apoptosis in inflammatory states. This study examined the effects of lipopolysaccharide (LPS) on apoptosis in pericytes and endothelial cells, both individually and grown together in a coculture system. METHODS: Pericytes and endothelial cells were isolated and cultured separately and in coculture as previously described. The cells were subsequently exposed to LPS for 12, 24, 48, and 72 hours. The cellular contents were then examined by Western blot analysis for products of apoptosis. TUNEL staining was also performed to analyze for apoptosis. RESULTS: Pericytes alone exposed to LPS showed increased levels of p11 and p17, which are activated fragments of capase-3, a cysteine effector protease involved in cleaving cytoskeletal and nuclear proteins to induce apoptosis. When grown in coculture with endothelial cells and exposed to LPS in coculture but harvested independently, pericytes showed decreased levels of p11 and p17 and increased levels of Bcl-xL, an antiapoptotic protein that protects the integrity of mitochondria, and prevents cytochrome c release and subsequent caspase-9 activation. CONCLUSIONS: In response to LPS, pericytes undergo apoptosis involving the caspase-3 pathway. Endothelial cells may decrease this effect through the expression of a soluble mediator.

Prophylactic and therapeutic inferior vena cava filters to prevent pulmonary emboli in trauma patients.

HYPOTHESIS: Insertion of inferior vena cava filters (IVCFs) can prophylactically reduce pulmonary embolism (PE) in trauma patients. DESIGN: Retrospective review. SETTING: Urban, level I trauma center. PATIENTS: Two hundred blunt trauma patients undergoing IVCF placement. INTERVENTIONS: In 122 patients who had already been diagnosed as having deep vein thrombosis (DVT) (112 patients) and/or PE (22 patients), the insertion of the IVCF was considered "therapeutic." In 78 patients who had no evidence of DVT or PE but who were considered to be at high risk for a PE, the IVCF was considered "prophylactic." MAIN OUTCOME MEASURES: Incidence of PE and related mortality and morbidity in therapeutic vs prophylactic IVCFs. RESULTS: The number of prophylactic IVCFs inserted increased significantly from only 4% (3/68 cases) from 1991 through 1996, up to 57% (75/132 cases) from 1997 to June 2001. Although the mean +/- SD age (51 +/- 20 years vs 41 +/- 15 years; P<.001) was higher in the therapeutic group, there was no difference in the mean +/- SD Injury Severity Scores (20 +/- 12 vs 21 +/- 11). Therapeutic filters were placed much later after injury (mean +/- SD time, 11 +/- 7 vs 3 +/- 2 days; P<.001). The mortality rate was 11% (13/122 patients) in patients having a therapeutic IVCF, as compared with only 3% (2/78 patients) in those placed prophylactically (P =.07). None of the patients who had placement of a prophylactic IVCF developed subsequent PE. The incidence of PE decreased in all blunt trauma patients from 0.29% before 1997 to 0.15% after January 1, 1997, when 57% of the IVCF inserted were prophylactic (P =.06). CONCLUSIONS: Prophylactic IVCFs should be inserted within 48 hours of injury in specific trauma patients at high risk for PE and with contraindications to anticoagulation.

Factors affecting the outcome of patients with splenic trauma.

This is a report of 546 consecutive patients with penetrating and blunt splenic trauma seen over a 17 1/2-year period (1980-1997). The etiology of the splenic injuries and the associated mortality rates were: blunt injuries 45 of 298 (15%), gunshot wounds 48 of 199 (24%), and stab wounds four of 49 (8%). The overall mortality rate was 97 of 546 (18%). The most significant risk factors for death were all associated with major blood loss: transfusion requirements > or = 6 units of blood, low initial operating room blood pressure, associated abdominal vascular injuries, and performance of a thoracotomy. The two most important organs injured in conjunction with the spleen that were significant predictors of postoperative infectious complications were colon and pancreas. The need for splenectomy was most significantly correlated with higher grades of splenic injury especially grades IV and V. The evolution in management of blunt splenic trauma has led to a significant improvement in splenic preservation and avoidance of laparotomy for many patients. Operative splenic salvage is reduced in patients subjected to laparotomy who are candidates for nonoperative treatment. Improved results with splenic injury should be obtained by rapid control of bleeding. This may require more liberal criterial in selecting patients with splenic trauma for early operative treatment.

Inhibition of heme oxygenase-1 in microvascular lung pericytes diminishes at high concentrations of an inflammatory mediator.

Post-traumatic inflammation and sepsis induce changes in the lung microvasculature causing increased permeability. Pericytes, contractile cells positioned abluminally to endothelial cells, play a role in regulating this response. An in vitro model of microvascular lung pericytes (MLP) was used to investigate the effect of inhibiting heme oxygenase-1 (HO-1), a stress-induced enzyme, in the presence of varying levels of lipopolysaccharide (LPS), a mediator in the initiation of inflammation, on pericyte contractility. Rat MLP were cultured on collagen gel matrices. Cells were exposed to three concentrations of LPS in the presence of zinc protoporphyrin IX (ZnPP-9), a known inhibitor of HO-1. After 24 hours, the surface area of the collagen disks was quantified, thereby measuring pericyte contraction. ZnPP-9 caused a significant attenuation of the LPS-induced relaxation of the pericytes (P < or = 0.003). The effects of ZnPP-9, however, depended on the concentration of LPS to which the pericytes were exposed. Greater concentrations of LPS decrease the attenuating power of ZnPP-9. The inhibition of HO-1 diminished MLP relaxation triggered by LPS. The effect of ZnPP-9, however, is dependent on the concentration of LPS to which the MLP are exposed, indicating its saturation. ZnPP-9 may antagonize the microvascular response to trauma.

Correlation of revised trauma score and injury severity score (TRISS) predicted probability of survival with peer-reviewed determination of trauma deaths.

Trauma deaths at our institution are evaluated by a multidisciplinary trauma committee. The purpose of this study was to evaluate preventable trauma deaths (PRDs) as determined by our review committee and correlate them with the Revised Trauma Score and Injury Severity Score (TRISS) probability of survival (PS). A total of 10,002 patients were identified. The PS was calculated using the TRISS method. The Z scores were calculated and the predicted number of deaths was established. The actual number of deaths was compared with the predicted number of deaths. PRDs were compared with the actual and predicted deaths. The Z score was 0.79, which meant we observed more deaths than predicted by TRISS. We had 281 deaths compared with 271 deaths predicted by TRISS. Peer review characterized 45 deaths as preventable. Although we performed well when our outcomes were compared with TRISS predicted outcomes our PRD rate was higher. The higher the PS the more likely the death was found preventable by peer review. We conclude that for our patient population the peer review process is very sensitive and may be more discerning in identifying PRD than TRISS.

The extent of cryosurgery increases the complication rate after hepatic cryoablation.

Although there have been many reports on the use of cryosurgery to ablate hepatic malignancies none have specifically examined the relationship of complication rates to the extent of cryoablation. A retrospective review from January 1997 to May 2002 identified 98 patients treated with hepatic cryotherapy. The extent of the cryosurgery was determined by the total number of lesions (TNL) and total estimated area (TEA) of the lesions from preoperative evaluation by CT scan and intraoperative evaluation by ultrasound. The major complication rate was 11 per cent. The 30-day mortality was 0 per cent, but the late procedure-related mortality was 2 per cent. Increasing the extent of cryotherapy measured by intraoperative ultrasound demonstrated significant increases in the complication rate and length of stay (LOS). With cryoablation of TEA > or = 30 cm2 there was a significant increase in the overall complication rate (56% vs 23%; P = 0.003) and LOS (8.8 +/- 6.9 vs 6.1 +/- 4.2; P = 0.022) compared with TEA < 30 cm2. Performance of concurrent procedures also led to a significant increase in complications (69% vs 29%; P = 0.010) and LOS (8.6 +/- 6.8 vs 6.0 +/- 4.0; P = 0.019). Multivariate analysis, however, showed intraoperative TEA > or = 30 cm2 to be the most significant independent predictor of increased complications and prolonged LOS.

Complications are increased with the need for an abdominal-assisted Kraske procedure.

The Kraske procedure offers a sphincter-saving alternative for surgical correction of rectal disease. This study was performed to investigate the complication rate with the traditional (transsacral) Kraske procedure versus an abdominal-assisted Kraske approach (laparoscopic or open). We conducted a retrospective review of all patients undergoing the Kraske procedure at Harper University Hospital over a 10-year period. A total of 54 patients were identified. Indications for surgery included rectal carcinoma (43), large villous adenomas (6), and other (5). Average post-operative follow-up was 40 +/- 25 months (mean +/- SD). Complications included rectocutaneous fistulae (9), perineal infections (13), and incontinence (8). In patients requiring an abdominal-assisted approach for colorectal mobilization, the fistula rate was significantly higher (33% vs 3%; P = 0.007), as were the rates of perineal infections (33% vs 17%) and of initial incontinence (25% vs 7%). The laparoscopic-assisted approach significantly reduced the operating time (272 +/- 72 minutes) compared to the open-assisted approach (498 +/- 138 minutes) (P < 0.001). The traditional Kraske procedure can be utilized in a safe, effective manner for treatment of rectal disease. Knowledge of the increased rate of complications with the abdominal-assisted Kraske approach can guide the patient and physician considering sphincter salvage.

Prognostic determinants in duodenal injuries.

A retrospective review of 222 consecutive patients with duodenal injuries admitted to an urban Level 1 Trauma Center who subsequently underwent laparotomy during the period July 1980 to April 2002 was performed in an effort to elucidate factors associated with mortality, infectious morbidity, and length of stay in these patients. Predictably, the patients were predominantly male (92.7%) and young (mean age, 31.6 years). The overall mortality rate was 22.5 per cent, with a mortality rate of 18 per cent seen in the first 48 hours. Penetrating trauma was suffered by 88.3 per cent of the patients. Multivariate analysis revealed the performance of a thoracotomy, initial emergency department (ED) systolic blood pressure (SBP) <90 mm Hg, final operating room (OR) core body temperature less than 35 degrees C, and presence of a splenic injury to be the most important predictors of mortality (all P < 0.05). Mortality in the patients undergoing a resuscitative thoracotomy was 88.9 per cent versus 13.3 per cent in those patients not requiring thoracotomy. An initial SBP in the ED <90 was associated with a 46 per cent mortality rate, as compared with 4 per cent in those patients not in shock. A final OR core body temperature of less than 35 degrees C led to a 60 per cent mortality rate versus 8.3 per cent for warmer patients. Patients with a concomitant splenic injury were noted to have a 62.5 per cent mortality rate; those without had a 19.4 per cent mortality rate. The mean length of stay among survivors greater than 48 hours was 16.0 +/- 24.7 days. Univariate analyses revealed lowest OR core body temperature below 35 degrees C, initial OR SBP <90, presence of infection, >5 transfusions, initial ED SBP <90, final OR core temperature <35 degrees C, colon injury, spleen injury, and an injury severity score (ISS) >25 all to be significantly associated with increased length of stay. Multivariate analysis revealed an initial operating room blood pressure of less than 90 mm Hg systolic, the presence of an infection, and greater than 5 blood transfusions to be the factors most significantly correlated with increased length of stay (all P < 0.02). Of 182 patients surviving 48 hours, 98 (54%) developed an infection. Fifty-seven (31%) patients were noted to have wound-related infections, 92 (51%) patients had nosocomial infections, and 50 (27%) patients had both types. The presence of an abdominal arterial injury, an ISS >25, pancreatic injury, and lowest OR core body temperature <35 degrees C were the factors identified on multivariate analysis most significantly correlated with infectious morbidity (all P < 0.05). This data suggests that early efforts to prevent shock and rapidly control bleeding are the most likely efforts to reduce mortality rates in these patients. Those patients with duodenal injury presenting in shock or requiring a thoracotomy for resuscitation did poorly. Splenic injury was the associated injury found on multivariate analysis to be most closely associated with increased mortality. Early control of bleeding and the prevention of infection provide the most significant opportunity for decreasing length of stay. Infections are common with duodenal injuries, and aggressive surveillance should especially be performed in those patients with an abdominal arterial injury, an ISS >25, pancreatic injury, or lowest OR core body temperature <35 degrees C.

Predictors of mortality in patients with traumatic diaphragmatic rupture and associated thoracic and/or abdominal injuries.

This is a retrospective review of 731 patients sustaining diaphragmatic trauma over a 22 year period (1980-2002) at an urban level I trauma center. Patients had an average injury severity score (ISS) of 22 +/- 12. The mortality rate (MR) was 23 per cent (168/731). There were a total of 460 left-sided diaphragmatic injuries (L-TDR), 263 right-sided diaphragmatic injuries (R-TDR), and 8 bilateral diaphragmatic injuries (B-TDR). There were no significant differences in mortality with L-TDR versus R-TDR. Shotgun wounds had the highest MR (42%) (P = 0.0028). Emergency thoracotomies were performed in 31 per cent (225) with a 62 per cent (140) MR. Bilateral thoracotomies had a significantly higher MR of 85 per cent (33/39) compared to the 58 per cent (107/186) for unilateral thoracotomies (P = 0.0028). Multivariate analysis revealed the most significant independent predictors of mortality to be the revised trauma score, transfusion of pRBCs > 10 units, and need for thoracotomy (P < 0.0001). The infection rate was 41 per cent. Multivariate analysis revealed blunt trauma, blood transfusions, ISS, and pancreatic injury as the most significant independent predictors of infection (P < 0.001). The initial physiologic presentation of the patient and the severity of hemorrhagic shock are the primary determinants for survival. Prompt identification of associated injuries with rapid control of bleeding is paramount to survival.