Michael J. Hershman

HLA-DR antigen expression on peripheral blood monocytes correlates with surgical infection

Monocyte human leukocyte antigen-DR (HLA-DR) expression has correlated closely with clinical outcome in severely injured patients at high risk for infection. Monocytes from 77 asymptomatic volunteers expressed HLA-DR antigen with minimal variability in respect to age, gender, race, time of day or year, or serum alcohol level. Patients who developed infection after elective laparotomy had a significantly lower mean percentage of monocytes expressing HLA-DR antigen and a lower mean fluorescent intensity than uninfected patients (p less than 0.05). Severely infected nonsurgical patients had significantly lower values than normal volunteers (p less than 0.01), and the mean fluorescent intensity of those who died from infection was significantly lower than that of those who survived (p less than 0.05). Patients on immunosuppressive regimens after renal transplantation had levels of HLA-DR expression similar to those of the volunteers. Monocyte HLA-DR expression was found to be a reliable marker of clinical infection and showed remarkable reproducibility within the normal uninfected study population.

An outcome predictive score for sepsis and death following injury

Injury is an important cause of both morbidity and mortality, particularly in the young. Scoring systems have been developed to establish guidelines of transfer and compare patient outcome, but no scoring system as yet has been constructed that focuses upon immune capability of these patients. We report an outcome predictive score (OPS) which appears to distinguish good outcome from sepsis in patients who survive, and between patients with sepsis who survive from patients with sepsis who die. The score is based on (1) Injury severity score (ISS) expressed as percentage of the 50 per cent lethal dose of injury for age (%LD50), (2) Degree of bacterial contamination at initial injury, (3) The patients monocyte DR antigen expression. Fifty-one severely injured patients were divided into three groups: (1) A group without infection (N = 15), (2) A major sepsis group which survived (N = 24), (3) A group who died (N = 12). There was no difference between the ISS of these groups. The mean OPS of the good outcome group was significantly less than the mean OPS of both the septic (P less than 0.0002) and dead (P less than 0.00001) groups. The mean OPS of the septic group was also significantly less than the mean of the group that died (P less than 0.002). Identification of high risk patients may be important to determine priority of patient care and to institute additional therapeutic measures.

The capacity of serum to support neutrophil phagocytosis is a vital host defense mechanism in severely injured patients.

The opsonic capacity of patient serum was studied in 43 trauma patients of whom 13 recovered uneventfully, 21 developed major infection, and nine died, mostly of infection. Blood samples were taken within 24 hours of injury. Fifteen patients were studied serially of whom 14 developed severe infection and/or died. Opsonic capacity was determined by flow cytometry and measured as the ability of normal neutrophils to phagocytose killed bacteria previously incubated with patient serum. The most dilute sera reflected changes for better and worse most clearly. On initial assessment, those who died of sepsis showed a 61% mean fluorescent intensity (MFI), which was significantly lower than the 99% MFI for those who survived infection (p < 0.01) and the 78% MFI of those who developed no infection (p < 0.05). Serial samples demonstrated a super serum response in four of seven patients surviving major sepsis but in none of the seven who died of infection.

Association of Cytomegalovirus Infection with Increased Morbidity Is Independent of Transfusion

Forty-three consecutive trauma patients with an injury severity score greater than 20 were studied prospectively for evidence of cytomegalovirus (CMV) infection. Twenty-one patients had serologic conversion: 3 with primary CMV infections, 18 with reactivation of CMV infection (CMV group). Twenty-two patients had no serologic conversion (no CMV group). To differentiate the effects of CMV and transfusion, the CMV group and the no CMV group were each divided into high (more than 10 units) and low (less than 10 units) transfusion subgroups. Similar fever peaks, leukocyte counts, lymphocyte counts, and incidence of major bacterial sepsis were recorded for the four subgroups. Several factors were significantly associated with CMV infection independent of transfusion, including increased duration of major bacterial sepsis and number of septic episodes per patient; prolonged duration of anergy; increased duration of intensive care unit and hospital stay; increased duration of ventilatory assistance and rate of tracheostomy; and increased suppressor cells, decreased helper: suppressor ratios, increased functional suppressor cells, and increased natural killer cells. Although mortality was not increased with CMV infection, our data suggest that such infection after trauma may delay recovery from major bacterial infection, often resulting in a major increase in morbidity.

Effects of interferon-γ treatment on surgically simulated wound infection in mice

Interferon-γ (IFN-γ) has been shown to have immunoregulatory properties and is able to modulate resistance to several microbial infections. This study was designed to determine the efficacy of IFN-γ treatment in a mouse model of infection that simulates many clinical conditions associated with surgical wound infection: viz, a bacteria laden suture and tissue injury. The test bacteria were Klebsiella pneumoniae. Groups of CBA/J mice received either IFN-γ or RPMI1640 medium (controls) subcutaneously. IFN-γ was administered daily at a dose of 7500 units for 5 days prior to bacterial challenge. Mice treated with IFN-γ survived significantly longer than controls. Systemic bacterial recovery was significantly reduced in the IFN-γ treated group but local bacterial recovery was unaffected.

Enhancement of survival from murine polymicrobial peritonitis with increased abdominal abscess formation

Muramyl dipeptide (MDP), a purified synthetic immune adjuvant, has been shown to increase murine intraabdominal abscess formation in a monomicrobial model using Bacteroides fragilis. This effect required live bacteria and was abolished by appropriate antibiotics. A polymicrobial model of peritonitis and abdominal abscess formation using Streptococcus fecalis, Escherichia coli, and B. fragilis was initially used to determine mortality rates at various concentrations and obtain an appropriate LD50. Animals were then pretreated with MDP or its inert buffer and underwent intraperitoneal injection of the appropriate bacterial suspension. Mortality and abdominal abscess formation were then assessed at 2 weeks after injection. There was a significant reduction in mortality (P less than 0.03) in mice treated with MDP compared to the controls. In surviving animals, there was also a significant increase in the number of animals forming abscesses (P less than 0.05) following treatment with MDP. This study has shown that nonspecific immune stimulation by MDP provided enhanced protection against a polymicrobial intraperitoneal challenge and paradoxically increased the formation of abdominal abscesses at the same time. This may be regarded as enhancement of the natural history of survival from peritonitis via bacterial containment through intraabdominal abscess formation, a manifestation of beneficial outcome in experimental peritonitis.

Monocyte HLA-DR Antigen Expression: Its Reproducibility in Asymptomatic Volunteers and Correlation with Clinical Infection

Trauma victims are an immunodepressed group of patients at high risk for infection because many facets of the immune response are depressed following severe injury [1–3]. These include delayed hypersensitivity [4], immunoglobulin production [5], and serum opsonic capacity [6], in addition to macrophage, lymphocyte, and neutrophil function [7, 8].

Relevance of Protein Binding to Cephalosporin Antimicrobial Activity in vivo

Protein binding, serum kinetics and minimum inhibitory concentrations (MICs) for Staphylococcus aureus were determined for cefoxitin, cefazolin, ceftazidime and ceftriaxone in the rabbit. MICs of cefazolin and cefoxitin were also measured for Escherichia coli. Varying concentrations of the bacteria were administered intradermally to create areas of cellulitis, which were quantified as mean erythematous areas (EAs). Despite large differences in protein binding of the antibiotics (range 12-88%) and antibiotic dosing to allow serum concentrations to drop below the respective MICs, there was no statistical difference in the mean EAs of the animals after bacterial challenge. Antibiotic protein binding did not alter the course of cellulitis nor correlate with bacterial MIC in this model.

The use of a new assay for detecting antibody to Staphylococcus aureus in severely injured patients.

Investigation of the antibody response to Staphylococcus aureus infection has been hindered by lack of a simple and specific assay. We report an enzyme-linked immunosorbent assay (ELISA) using a strain of S. aureus devoid of Protein A, a frequent cause of false positive results in ELISAs, and have used this assay to study antibody responses of 23 severely injured patients. This IgM ELISA had a diagnostic sensitivity for major Staphylococcal infection of 70% (seven of ten patients with major infection) and a specificity of 92% (12 of 13 patients without major infection). Of three patients with major Staphylococcal infections who mounted no IgM response, two died and the other developed severe chronic Staphylococcal infection, and hence prompt initiation of appropriate therapy was necessary. Furthermore, the ability to mount IgM response to Staphylococcal infections appears to contribute to an orchestrated host defense response against this organism.

Evaluation of Copovithane as a Nonspecific Immunomodulator in Surgically Simulated Sepsis

Copovithane (CPV), a synthetic polymer, has been shown to have antitumor activity and also to reduce mortality in experimental murine peritonitis. The purpose of this study was to compare CPV with muramyl dipeptide (MDP), an immunomodulator of proven efficacy in simulated surgical infection. Six groups of CBA/J mice were compared; they received intramuscular injections of normal saline (controls), MDP (100 micrograms), or CPV (100, 200, and 400 mg/kg) 24 h prior to bacterial challenge. The challenge consisted of a Klebsiella-impregnated thigh suture. The first experiment assessed survival after bacterial challenge. The MDP and the CPV groups both had median survival times of 3 days, significantly longer than the control group (1 day, p less than .05). In the second experiment, animals were sacrificed at 6, 24, and 48 h following bacterial challenge, and blood and infected muscle were taken for quantitative bacteriology. At 6 h, there was no difference between groups. Both the MDP and CPV groups had significantly (p less than .05) lower blood bacterial counts than the control group at 24 and 48 h. Both the MDP and CVP groups had significantly lower local bacterial recovery than controls at 48 h (p less than .05), and local bacterial recovery of the MDP group was significantly lower than the control group at 24 h (p less than .05). CPV improved survival and reduced local and systemic bacterial recovery compared with controls. Although the effect of CPV was similar to MDP in this model, it consistently was of lower magnitude and had a narrow dose range.