Animal study
In total, 60 male Wistar rats were studied. Their mean ± SD weight was 253.4 ± 22.4 g. The study received permission from the Veterinary Directorate of the Perfecture of Athens, according to the Greek legislation, in conformance to the Council Directive of the European Community. Rats were housed in metal cages and had access to tap water and standard balanced chow ad libitum. Temperature ranged between 18°C and 22°C; relative humidity, between 55% and 65%; and the light/dark cycle was 6 AM to 6 PM. One multidrug-resistant (MDR) isolate of P. aeruginosa from the blood of a patient with severe sepsis was applied. Minimal inhibitory concentrations (MICs) of piperacillin/tazobactam, ceftazidime, imipenem, meropenem, ciprofloxacin, and amikacin, determined with a microdilution technique, were 256/4, 256, 128, 64, 512, and 512 μg/ml, respectively. The isolate was stored as multiple aliquots in skim milk (Oxoid Ltd, London, UK) under -70°C. One aliquot was removed from the refrigerator before each experiment. Single colonies were incubated at 37°C in 10 ml of Mueller-Hinton broth (Oxoid Ltd) for 8 hours to yield a log-phase inoculum that was applied for bacterial challenge.
Animals were divided into two groups for treatment, as follows:
Group A (n = 15), sham-operated; injected intraperitoneally with 1 ml of Mueller-Hinton; and
Group B (n = 45), animals injected intraperitoneally with 1 ml of the prepared inoculum of the test isolate yielding a challenge of 5 × 106 cfu/rat.
Survival was recorded at 12-hour intervals for four animals of group A and for 10 animals of group B. To identify the peak time point of lipid peroxidation, animals of group B were killed at serial time intervals with six rats per interval. At every time of sacrifice, a midline abdominal incision was performed. Intestines were displaced to the left, and the inferior vena cava was recognized and punctured with a 19-gauge needle. Five milliliters of blood was collected into pyrogen-free syringes and applied for the measurement of tumor necrosis factor-alpha (TNF-α) and of malondialdehyde (MDA). Animals were then killed with an intramuscular injection of pentothal.
These experiments were repeated in six animals per group at the defined peak time point of serum MDA. Under sterile conditions, segments from the liver, spleen, lower lobe of the right lung, the heart, the right kidney and of the abdominal aorta were taken and placed into separate sterile plastic containers. In these experiments, 5 ml of blood was sampled after venipuncture of the lower vena cava under aseptic conditions and collected into EDTA-containing tubes (Becton Dickinson, Cockeysville, MD, USA) for the measurement with flow cytometry of an oxidative burst on neutrophils. Tissue segments were weighed and homogenized by using 1 ml of sterile phosphate-buffered saline (PBS; pH 7.2). For the measurement of tissue bacterial outgrowth, aliquots of 0.1 ml of the homogenates were diluted 1:10 into sterile 0.9% NaCl for four consecutive times. Aliquots of 0.1 ml of each dilution were plated onto McConkey agar and incubated at 35°C for 3 days. The number of viable colonies was counted by multiplying with the appropriate dilution factor. The lower limit of detection was 10 cfu/g. The number of viable cells in tissues was expressed by log10 values.
Lipid peroxidation was estimated in serum and in tissue homogenates by the concentration of MDA, as already described [
12]. In brief, a 0.1-ml aliquot of each sample was mixed with 0.9 ml of trichloroacetic acid 20% (Merck, Darmstadt, Germany) and centrifuged at 12,000
g and 4°C for 10 minutes. The supernatant was removed and incubated with 2 ml of thiobarbituric acid, 0.2% (Merck) for 60 minutes at 90°C. After centrifugation, a volume of 10 μl of the supernatant was injected into a high-performance liquid chromatography system (HPLC; Agilent 1100 Series, Waldbronn, Germany) with the following characteristics of elution: Zorbax Eclipse XDB-C18 (4.6 × 150 mm, 5 μm) column under 37°C; mobile phase consisting of 50 m
M K
3PO
4 (pH 6.8) buffer, and methanol 99% at a 60:40 ratio with a flow rate of 1 ml/min, fluorometric detection with signals of excitation at 515 nm and emission at 535 nm. The retention time of MDA was 3.5 minutes, and it was estimated in millimoles per milliliter with a standard curve created with 1,1,3,3-tetramethoxy-propane (Merck). All determinations were performed in duplicate. The lowest limit of detection was 0.1 mmol/ml, and the interday variation of the assay was 1.01%. Tissue concentrations of MDA were expressed as millimoles per gram.
Concentrations of serum TNF-α were measured in duplicate with an enzyme immunoassay (Diaclone, Marseille, France). The lower limit of detection was 12 pg/ml.
For the measurement of the oxidative burst on neutrophils, 0.1 ml of EDTA-blood was aliquoted into separate sterile tubes. A volume of 20 μl of phorbol-myristate (PMA, Sigma, St. Louis, MO, USA; final concentration, 10 μΜ) was added in the second tube, and the first tube was left untreated. Both tubes were incubated for 10 minutes at 37°C in a water bath. Then 10 μl of DHR (dihydrorhodamine; Sigma, final concentration 100 μΜ) was added. After 10 minutes of incubation at 37°C in a water bath, 1 ml of VersaLyse was added in both tubes for the lysis of red blood cells (Immunotech, Marseille, France). After 15 minutes of incubation in the dark, tubes were analyzed with the FC500 counter (Beckman Coulter, Miami, FL, USA). Oxidative burst was expressed as the mean fluorescence intensity (MFI) of DHR on granulocytes after gating by the characteristic FS/SS scattering by using PMA-untreated cells as negative controls.
Statistical analysis
For animal experiments, survival was estimated with Kaplan-Meier analysis. Concentrations of MDA and TNF-α were expressed by their means ± SEM. Comparisons of concentrations of MDA and of TNF-α in serum between serial times of sampling were done with one-way analysis of variance (ANOVA) with post hoc Bonferroni corrections. Comparisons of tissue MDA between groups A and B were performed with the Mann-Whitney U test. Correlations between tissue MDA and bacterial growth were according to Spearman's rank of order.
For the clinical study, the MDA of the first day in serum was expressed as median and 95% confidence intervals. Comparisons between patients with organ failures and those without any organ failures were done with the Mann-Whitney U test, analyzing also for outliers. To demonstrate the impact of lipid peroxidation in the pathogenesis of liver dysfunction and of renal dysfunction, all patients with acute hepatic dysfunction and all patients with acute renal dysfunction were grouped together as having hepatic dysfunction and renal dysfunction, respectively, whether or not they had other organ failures. MDAs of consecutive days were expressed as mean ± SEM. Comparisons between survivors and nonsurvivors were done with the Mann-Whitney U test. For all comparisons, any value of P < 0.05 after adjustment for multiple comparisons with Bonferroni was considered significant.