Management of infections in the trauma patient.

Infections play a leading role in the morbidity and mortality of injured patients. This article discusses risk factors that can increase the chances of a nosocomial infection. It also discusses common types of infection, causative organisms, and the approach to the febrile trauma patient.

Seroprevalence of human immunodeficiency virus, hepatitis B virus, hepatitis C virus, and rapid plasma reagin in a trauma population.

We evaluated the presence of human immunodeficiency virus (HIV), hepatitis B virus (HBV), hepatitis C virus (HCV), and rapid plasma reagin (RPR) among patients admitted to our trauma unit from April 15 to June 30, 1993. Of 984 patients tested, we found 255 (26%) had evidence of exposure to one or more of these agents: HIV, 4%; HBV, 20%; HCV, 14%; and RPR, 1%. Thirty-eight percent of patients had more than one positive serology, 75% of the HIV patients, 49% of the HBV patients, and 66% of the HCV patients. There was no difference between penetrating and nonpenetrating trauma with respect to any of the viruses. The risk factors for HIV-positive patients were non-White race, positive drug screen, positive alcohol screen, and city resident. Risk factors for HBV patients were non-White race, positive drug screen, and city resident. Risk factors for HBC patients were male sex, non-White race, positive alcohol screen, positive drug screen, and city resident. The risk of blood-borne infections in this group of patients is substantial.

Role of immunomodulator therapy in sepsis.

Endotoxin is released from the cell walls of gram-negative bacteria and causes severe systemic effects due to the release of cytokines. Monoclonal antibodies directed at endotoxin may be promising adjuncts to the standard therapeutic interventions of antibiotics and supportive measures used to treat patients with gram-negative sepsis. Monoclonal antibodies interfere with the bacteria's ability to trigger an unfavorable response. In recent clinical trials, two immunoglobulin M monoclonal antibodies have improved survival in certain small patient subgroups, although neither drug improved overall mortality in all septic patients treated. E5 murine monoclonal antibody reduced mortality in patients with gram-negative sepsis who were not in refractory shock. HA-1A human monoclonal antibody reduced mortality in patients with gram-negative infections who were bacteremic or in shock. The statistical significance and clinical importance of these benefits is not yet known. Results of these clinical trials are reviewed.

Ciprofloxacin pharmacokinetics in critically ill trauma patients.

The steady-state pharmacokinetics of ciprofloxacin 200 mg intravenously every 12 hours was examined in 10 critically ill trauma patients. The mean parameter estimates for total clearance, renal clearance, non-renal clearance, and volume of distribution were 30.08 liters/hour/1.73 m2, 16.62 liters/hour/1.73 m2, 13.46 liters/hour/1.73 m2, and 2.10 liters/kg. Although the mean values were similar to those previously reported, significant individual differences were observed, with the coefficient of variation ranging from 41 to 61 percent. Non-renal clearance appeared to have a bimodal distribution. The dosage studied appeared to provide adequate serum concentration profiles to treat most pathogens found in infected trauma patients. However, the use of higher doses and more frequent dosing may be required to treat patients with Staphylococcus aureus and Pseudomonas aeruginosa infections.

Prospective use of optimal sampling theory: steady-state ciprofloxacin pharmacokinetics in critically ill trauma patients.

We examined the use of optimal sampling theory to determine a sparse sampling design to estimate pharmacokinetic parameters of ciprofloxacin in patients who had sustained trauma. Two serum sampling strategies, consisting of six sampling times each, were derived on the basis of the patient's renal function (patients with creatinine clearance greater than or equal to 6 L/hr/1.73 m2 and patients with creatinine clearances less than 6 L/hr/1.73 m2). Two additional serum samples were obtained for other aspects to the study. A timed urine collection was also obtained. Pharmacokinetic parameter estimates were determined by comodeling the serum and urine data with a three-compartment open model (parameterized as microconstants) with a bayesian algorithm and by noncompartmental analysis. Bayesian-derived parameter estimates were total body clearance of drug from plasma, 29.8 L/hr/1.73 m2; renal clearance, 17.0 L/hr/1.73 m2; and nonrenal clearance, 12.7 L/hr/1.73 m2 and were not significantly different from noncompartmentally derived parameters (p = 0.80, p = 0.65 and p = 0.333, respectively). The study demonstrates the use of optimal sampling theory to determine an informative yet relatively sparse sampling strategy for a drug with a complex pharmacokinetic model.

The septic multiple-trauma patient.

Sepsis in the multiple trauma patient is being seen with increased frequency now that more of these patients are surviving the initial period. Traumatic destruction of tissue barriers, the placement of various tubes and drains, and surgical repair with debridement all provide conduits for colonization and infection with pathogens. Many components of the host immune system also become altered after trauma and surgery, predisposing this population to infectious complications. The site of infection can be cryptic in the moribund trauma patient; locating it may require many special diagnostic procedures. Continuing close surveillance is important to prevent or to identify infections at the earliest possible time. The liberal use of antibiotics should be discouraged so that development of resistant organisms and superinfection is kept to a minimum. Handwashing between patient contacts may be the most important prophylaxis against the spread of pathogens within a trauma unit.

The septic multiple-trauma patient.

Sepsis in the multiple trauma patient is being seen with increased frequency now that more of these patients are surviving the initial period. Traumatic destruction of tissue barriers, the placement of various tubes and drains, and surgical repair with debridement all provide conduits for colonization and infection with pathogens. Many components of the host immune system also become altered after trauma and surgery, predisposing this population to infectious complications. The site of infection can be cryptic in the moribund trauma patient; locating it may require many special diagnostic procedures. Continuing close surveillance is important to prevent or to identify infections at the earliest possible time. The liberal use of antibiotics should be discouraged so that development of resistant organisms and superinfection is kept to a minimum. Handwashing between patient contacts may be the most important prophylaxis against the spread of pathogens within a trauma unit.

Duration of preventive antibiotic administration for open extremity fractures.

The necessary duration of antibiotic administration after open fracture has not been established. In a double-blind prospective trial we randomized 248 patients with open fractures to receive one or five days of cefonicid sodium therapy or five days of cefamandole nafate therapy as part of the initial treatment. Rates of fracture-associated infections in the three groups were ten (13%) of 79, ten (12%) of 85, and 11 (13%) of 84, respectively. The 95% confidence limit for the difference in infection rates between the one-day group and the combined five-day groups was 0% to 8.3%. The actual difference was 0.2%. A brief course of antibiotic administration is not inferior to a prolonged course of antibiotics for prevention of postoperative fracture-site infections.

A randomized controlled trial of moxalactam versus clindamycin/tobramycin in the treatment of mixed anaerobic/aerobic infections.

Many infections are due to mixtures of facultative gram-negative bacilli and anaerobic bacteria. Moxalactam, a semisynthetic beta lactam antibiotic, is active against a wide range of anaerobic organisms, including most strains of Bacteroides fragilis, as well as many aerobic gram-negative bacilli. We performed a prospective, randomized controlled trial comparing moxalactam alone with the regimen of clindamycin and tobramycin for treatment of mixed aerobic/anaerobic infections. One hundred and six patients with presumed mixed infections were randomized to the study groups. The resultant groups were clinically and microbiologically comparable. The effectiveness of treatment was similar with both antibiotic regimens. Five of 25 patients tested in the moxalactam group had a prolongation of their prothrombin time and one of them developed clinically important bleeding. Two of the 23 patients tested in the clindamycin/tobramycin group had a prolonged prothrombin time with no bleeding. Decreases in hematocrit which could be "probably" or "possibly" related to antimicrobial use were seen in 6 of 48 moxalactam patients and none of 50 clindamycin/tobramycin patients (P = .03). Moxalactam, a potent antimicrobial for both anaerobic and aerobic organisms, demonstrated effectiveness in treating mixed anaerobic/aerobic infections similar to clindamycin/tobramycin but was associated with clinically important decreases in hematocrit.

Double-blind, prospective, multicenter trial comparing ceftazidime with moxalactam in the treatment of serious gram-negative infections.

Ceftazidime is a new antimicrobial agent possessing excellent in vitro activity against most members of the family Enterobacteriaceae and against Pseudomonas aeruginosa. We conducted a double-blind, prospective, multicenter trial to compare ceftazidime with moxalactam in the treatment of serious gram-negative infections. The overall favorable response rates for the two regimens were similar (93 of 106 [88%] and 84 of 97 [86%], respectively). Among these, the response rates of the 56 gram-negative bacteremias and the 23 P. aeruginosa infections were comparable. Both groups had similar incidences of subsequent infections with P. aeruginosa, enterococci, and yeasts. A total of 13% of the patients in the moxalactam group developed a prolonged prothrombin time (P less than 0.01), and three patients demonstrated clinical bleeding. These results suggest that although the overall efficacy of both regimens was similar, treatment with moxalactam resulted in a higher incidence of prolongation of prothrombin time with an attendant risk of bleeding. In nonneutropenic patients, ceftazidime as a single agent is safe and effective in gram-negative bacillary infections.

Identification and treatment of infections in multiply traumatized patients.

Trauma is the leading cause of death among young adults, and infection is a leading complication in multiply traumatized patients. All antibiotic use and all infections among 1,009 patients admitted to the Maryland Institute for Emergency Medical Services Systems over a six-month period were reviewed. The vast majority of patients had sustained high-speed automobile trauma and had blunt injuries. All antibiotics were given by the infectious diseases consultants under predetermined protocols. During this time period, 175 infections and 76 bacteremias were identified. Thirty-three percent of the antibiotic use was for prophylaxis. Prophylactic antibiotics were used for open fractures, in which a cephalosporin was used; for abdominal trauma, in which an aminoglycoside and clindamycin or cefoxitin alone was used; and for penetrating open fractures of the oral cavity, in which penicillin was used. As therapy, the aminoglycosides were used in 25 percent, the cephalosporins in 21 percent, the penicillins in 39 percent, and other antibiotics in 15 percent of the cases. The organisms identified as causing infection were Staphylococcus aureus (25 percent), Escherichia coli (18 percent), Enterobacter species (17 percent), Pseudomonas species (12 percent), and Klebsiella species (11 percent). The sites of infections were primary bacteremia (11 percent), vascular lines (21 percent), the central nervous system (3 percent), the lower respiratory tract (13 percent), the paranasal sinuses (6 percent), the urinary tract (19 percent), surgical wounds (11 percent), the abdomen (7 percent), and other sites (9 percent). More than 82 percent of the infections that occurred were nosocomial in origin and were related to the various procedures used for monitoring and therapy in these critically ill patients. Infections of the abdominal cavity and the lower respiratory tract accounted for eight of the 10 infection-related deaths in these patients.

Effect of saturable clearance during high-dose mezlocillin therapy.

As mezlocillin has been shown to display nonlinear pharmacokinetics in single-dose evaluations, we evaluated a crossover trial in patients with renal dysfunction the impact on serum clearance of fixed-dose versus fixed-interval administration of identical daily doses of the drug. In four patients with creatinine clearances of 0.00 to 1.78 liters/h per 1.73 m2, equal serum clearances were observed when the calculated daily total dose of mezlocillin was given either as a fixed dose of 5,000 mg at various intervals or every 4 h at various doses. We found that repetitive large daily doses that are equivalent to 30 g/day in patients with normal renal function can be administered to patients with impaired renal function as a reduced dose every 4 h instead of prolonging the dosing interval, as suggested by Mangione et al. (Antimicrob. Agents Chemother. 21:428-435, 1982). The observed serum clearances were equal for the two schedules, probably owing to the degree of continuing saturation of the nonlinear clearance mechanisms of mezlocillin.

Empyema occurring in the multiply traumatized patient.

During an 11-month period 31 cases of nosocomial empyema were identified in 29 of 741 multiply traumatized patients who remained in our unit for more than 3 days. Nosocomial empyema was defined as purulent culture-positive material drained from the pleural space after five days' hospitalization. All patients had fever and leukocytosis. Possible risk factors included previous aspiration in five patients but none developing pneumonia, prior respiratory tract infection in nine but none with the same pathogen as their empyema, prior antibiotic use in over 50% of the patients, and severe head or chest injury in two thirds of the patients. Thirty-eight pathogens were recovered: S. aureus, 14; beta-streptococci, three; Pseudomonas, six; Klebsiella, two; Enterobacter, two; E. coli, two; other Gram-negative bacilli, six; and anaerobes, three. Fourteen infections were polymicrobic and bacteremia occurred in 42% of the patients. Of these 29 patients, 27 had chest tubes inserted for fluid in the pleural cavity before development of empyema; nine for hemo- or pneumothorax secondary to chest trauma, 11 for pneumothorax while on ventilators, and seven for unexplained sterile pleural effusion. If empyema complicated a prior hemothorax it was usually caused by Staphylococcus aureus and occurred about 10 days after draining blood from the pleural cavity. If empyema was a complication of pneumothorax or serothorax it was usually due to Gram-negative organisms colonizing the upper respiratory tract and occurred within 4 days of draining the fluid. Sixteen per cent of all patients who had chest tubes placed for fluid in their pleural cavity subsequently developed empyema.(ABSTRACT TRUNCATED AT 250 WORDS)

Nosocomial Hemophilus pneumonia in patients with severe trauma.

In recent years, there have been increasing reports of Hemophilus pneumonia in adults. With few exceptions, these reports described Hemophilus as a cause of community acquired pneumonia. In the past three years, however, we have seen 15 instances of nosocomial Hemophilus pneumonia in patients with trauma who are intubated. This represented 11 per cent of the incidences of pneumonia that we diagnosed in this time period. The average age of the patient was 36 years and most patients had been in motor vehicle accidents. Six had received antibiotics and nine, steroids prior to their infection. Of note is that all incidences of Hemophilus pneumonia occurred within 11 days of hospitalization. This suggested to us that although the infections were clearly nosocomial, the organisms were probably not hospital acquired. Nevertheless, Hemophilus should be considered a potential cause of early pneumonia in the patients with trauma who are intubated.

Circulating thyroid hormone changes in acute trauma: prognostic implications for clinical outcome.

Alterations in circulating thyroid hormone concentrations occur in a variety of nonthyroidal disease states. In the present study, thyroid hormone levels were measured every 8 to 12 hours in 19 otherwise healthy individuals suffering acute severe trauma necessitating admission to the Maryland Institute for Emergency Medical Services Systems. Four fatalities occurred within 48 hours of admission. The mean total T3 level fell rapidly after the onset of trauma and remained low throughout the observation period. Reverse T3 rose concurrent with the fall in T3 but gradually returned to normal in the survivors. Total and free T4 levels remained normal in the survivors but fell below normal in the fatalities on the samples obtained preceding death. Changes in free T4 were consistent in three separate radioimmunoassay systems. Pharmacologic doses of glucocorticoids administered to seven of the 15 survivors and to the four fatalities did not result in an acute depression in total and free T4 levels in the survivors. Post-mortem examination of three fatalities did not reveal evidence of significant thyroid or pituitary disease. These results suggest that in acutely traumatized patients: 1) T3 declines rapidly and remains depressed throughout the illness; 2) continued fall of T4 to subnormal levels is associated with a poor prognosis; and 3) steroid therapy alone cannot explain the acute changes observed in hormone levels.

Response of traumatized splenectomized patients to immediate vaccination with polyvalent pneumococcal vaccine.

In recent years the syndrome of overwhelming post-splenectomy sepsis has been increasingly reported in adults. Since more than 50% of these infections are caused by pneumococcus these post-splenectomy patients are considered a suitable group to receive the pneumococcal vaccine. Previous studies of the response obtained in post-splenectomy patients have been conflicting and we found no study that looked at the response to immediate vaccination in this group of patients. Sixteen consecutive multitraumatized patients received polyvalent pneumococcal vaccine 0.5 ml IM within 72 hours of splenectomy and 10 normal controls were given 0.5 cc polyvalent pneumococcal vaccine. Patients received an average of 19.2 units of blood and blood products; seven were on steroids for concomitant head injury. Antibody was measured by the radioimmune assay. Most of the subjects of both groups responded to at least seven of the 12 measured antigens and no patient in the control group and only one in the splenectomized group responded to all 12 antigens. When rate of response to individual serotypes was compared no difference was found between the two groups. Comparison of geometric mean fold rise and fold rise between the two groups for each of the 12 serotypes revealed essentially no difference. We conclude the response to polyvalent pneumococcal vaccine among polytrauma splenectomized patients is similar to that of normal controls, and that the vaccine can be administered immediately post-splenectomy.

A randomized, controlled trial of cefoperazone vs. cefamandole-tobramycin in the treatment of putative, severe infections with gram-negative bacilli.

Cefoperazone was compared with the combination of cefamandole and tobramycin in a prospective, randomized study of putative, severe, gram-negative bacillary infections. We attempted to exclude patients with granulocytopenia or infections due to Pseudomonas species. A total of 118 isolates (94 gram-negative bacilli and 24 gram-positive cocci) caused infection in 99 of the 120 patients studied. Cefoperazone (16 micrograms/ml) was active against 93% of the organisms tested; cefamandole (16 micrograms/ml) and/or tobramycin (4 micrograms/ml) was active against 95%. Infection was cured or improved in 77% of cefoperazone-treated patients and 81% of cefamandole-tobramycin-treated patients. Bacteremia was cured or improved in 61% of cefoperazone-treated patients and in 63% of cefamandole-tobramycin-treated patients. Adverse reactions included five cases of probable antibiotic-associated nephrotoxicity in the cefamandole-tobramycin group; there were no such cases in the cefoperazone group. One patient given cefoperazone plus eight other drugs became granulocytopenic, but the condition resolved when all medications were stopped. This analysis suggests that cefoperazone alone may be as effective as cefamandole plus tobramycin in the treatment of severe infections with gram-negative bacilli and is less nephrotoxic. The role of cefoperazone in patients with granulocytopenia or infections due to Pseudomonas aeruginosa was not evaluated.

Infected false femoral artery aneurysms secondary to monitoring catheters.

Six patients who developed infected false femoral artery aneurysms secondary to monitoring catheters are reported. All aneurysms were infected and resulted in systemic sepsis. Initially the origin of the sepsis was not obvious. Findings which suggest this lesion include staphylococci. Appearance of distal petechial hemorrhages should lend a strong suspicion to the possibility of the lesion. Appearance of the pulsatile groin mass completes the diagnosis. Only aneurysmal resection combined with appropriate antimicrobial therapy is curative. For necessary revascularization procedures, autogenous vein grafts should be used when available.