7-[L- 2'0

JOURNAL OF SURGICAL RESEARCH 61,17-22 (1996)ARTICLE NO. 0074

Neutrophil Activation and Tissue Neutrophil Sequestration in a Rat,1""" Model of Thermal Injury

JOHN F. HANSBROUGH, M.D., T. WIKSTRbM, M.D.,* MAGNUS BRAIDE, M.D., PH.D.,* MAYER TENENHAUS, M.D.,OLIVER H. RENNEKAMPFF, M.D.,l VERENA KIESSIG, B.S., AND LARs M. BJURSTEN, M.D., PH.D. t

Department of Surgery, University of California, San Diego Medical Center, San Diego, California 92103; *Department of Anatomy andCell Biology, University of Goteborg, Goteborg, Sweden; and tThe Department of Experimental Research,

Lund University, Malmo, Sweden

Submitted for publication February 2, 1995

clinical outcomes by limiting remote tissue injury.@ 1996 Academic Press, loc.Neutrophil (PMN) deposition in tissues (leuko-

sequestration) alter shock may produce local tissueinjury from proteases and oxygen intermediarieswhich are released from sequestered PMNs. We quan-tified leukosequestration in tissues in burned rats us-ing two methods of analysis: 1), measurement of lungmyeloperoxidase (MPO)j 2), measurement of radiola-beled PMNs and erythrocytes deposited in multiple tis-sues. Mter tracheostomy and venous cannulation, ratsreceived 17% TBSA full-thickness contact burns andwere resuscitated with 20 cc intraperitoneal saline.Lung PMNs were estimated by measuring MPO in lungtissue. PMN influx into lung, liver, spleen, gut, skin,muscle, kidney, and brain was determined by remov-ing (preburn) and differentially radiolabeling PMNsr11ln) and erythrocytes (51Cr), reinfusing cells 4.5 hrpostburn, and measuring tissue radioactivity 5 hrpostburn. Tissue edema was measured by determiningextravasation of 1251-labeled albumin in tissues. Pe-ripheral blood PMNs were analyzed for intracellularH2O2 content utilizing a fluorescent dye which reactswith H2O2 coupled with analysis of cell fluorescenceby flow cytometry. MPO was elevated in lungs 8 hrpostburn (P < 0.05). PMN influx into lung tissues wasconfirmed by histologic examination. Radioisotopestudies demonstrated significant (P < 0.05) leuko-sequestration into lung, gut, kidney, skin, and braintissues at 5 hr postburn. Respiratory burst activity ofperipheral blood PMNs was increased 5 hr postburn(P < 0.05). Flow cytometric analysis indicated that pe-ripheral blood PMNs were capable ofproducing mark-edIy increased H2O2 levels 5 hr postburn. Tissueedema, manifested by radiolabeled albumin influx,was not seen in any tissues. Since others have shownthat sequestration of metabolically active PMNs mayinduce remate tissue injury, therapies which blockpostburn leukosequestration may be able to improve

INTRODUCTION

In the United States, the gradual and increasinglywidespread adoption of burn wound excision and ag-gressive wound coverage in most burn centers [1-3]has dramatically decreased the incidence of burnwound sepsis which previously was the major causeof mortality in burn patients [4]. The major source ofmortality following burn injury has now become pulmo-nary failure and pneumonia [5]. Pulmonary complica-tions are usually manifested as acute respiratory dis-tress syndrome (ARDS) and pneumonia.

There is evidence that the development of ARDS fol-lowing injury, sepsis, and shock is closely related toaccumulation of neutrophils (PMNs) in the lungs, fol-lowed by local release of toxic mediators including pro-teases and high energy oxygen metabolites [6-11].Priming, adherence, and activation of PMNs may bemediated by endotoxin [10, 12, 13] in association withrelease of multiple cytokines, which may elicit bothsystemic and local effects.

Our previous experiments verified that 32% TBSAburn injury in mice resulted in transient leukoseques-tration in the lungs [14, 15]. We have proceeded tostudy PMN activation and leukosequestration in multi-pIe tissues in a rat burn injury model, since the use ofthe larger animal facilitates more detailed studies oflocal and systemic effects of injury and shock.

MATERIALS AND METHODS

Animals and burn injury. Rats were maintained in accordancewith the guidelines ofthe U.C.S.D. Animal Research Committee onthe Care and Use of Laboratory Animals of the Institute of Labora-tory Animal Resources, National Research Council. The radiolabel-ing studies, performed in Sweden, were approved by the GoteborgEthical Review Committee on Animal Experiments. Male Wistarrats, weighing 200-300 g, were anesthetized by intraperitoneal (i.p.)injection with Nembutal (Abbott Labs, N. Chicago, IL), 2 mg/lOO gbody weight (b.w.), and Diazepam (Elkins and Sinn, Cherry Hill,

1 Supported by a grant from "Deutsche Forschungsgemeinschaft."

Funding is alBo acknowledged from the Swedish Medical ResearchCouncil (00663), the Goteborg Medical Society, and the Wilhelm andMartina Lundgrens' Research Foundation.

~ --0- 11 P :A/ ..4- IJ 17 /l ~ 0022-4804/96 $18.00~-, ~ 11: ~~~ «c t'- 9 ~ ~ Copyright C> 1996 by Academic Press, Inc.

AlI rights of reproduction in any forro reserved.-" r--- -~ ~~"'X & r..-c.c~J;: ~ /~..,.4~

a.< Jo o-1l- ~ tf , o... .t" ~ ---!.- ~ --.,. ~ -~ -oS&- '---7'-'c. ~ c..- /i.4"'~~~ , ~ t::. ~ ,

~

~

>- ~~,

FEBRUARY 15, 1996JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO.

18

the blood in the pulmonary circulation is immobilized [20]. The radio-activities (disintegrations per minute (dpm» of the blood samples,the excised lungs, kidneys, spleen, and sections of liver, intestine,skeletal muscle, nonburned skin, and brain were measured. Tissue-specific leukocyte sequestration was calculated using the leukocytetransit time factor as a measure. The transit factors were calculatedaccording to the relation: Transit factor = (Regional lllIn dpm X51Cr dpm/ml venous blood)/Regional 51Cr activity X ll1In activity/rnlvenous blood) [20]. There were five animals in each of the control

and burn groups.Measurement of tissue edema. Tissue edema was measured by

injection of 1-10 Jiol of 125I-Iabeled human albumin (37 MBq/rnl, Am-ersham) together with the radiolabeled cells. The ratio of albuminto erythrocytes was calculated in the same manner as for the leuko-cytes, and used as a measure of tissue edema in the various tissues

and organs.Measurement of mixed uenous blood gases. Animals were anes-

thetized and a catheter placed by cutdown into either the femoralor jugular vein. The animals were then burned and permitted toawaken. At appropriate time points, venous blood samples were with-drawn into a heparinized tube and blood gas analysis was performedimmediately utilizing a Corning 168 PhlBlood Gas Analyzer (CorningGlassworks, Medfield, MA) to determine the base deficit. Five toeight animals were utilized for each time point.

Histologic preparation of lung tissue. Representative samples oflungs of control and burned animals were taken. Tissues were fixedin formalin, embedded in paraffin, and then stained with hematoxy-

lin and eosin.Measurement of PMN respiratory burst. We used a modification

of the technique developed to quantify intracellular content of H2O2[21] with certain modifications to utilize flow cytometry [22-24].The dye which was utilized, 2,7 -dichlorodihydro-fiuorescein diacetate(DCFH-DA), crosses the cell membrane and changes its fiuorescenceemission spectrum after reacting with intracellular H2O2.

Blood was drawn from the vena cava of anesthetized rats andplaced in a heparinized tube. A 100-Jiol aliquot of blood was dilutedwith 850 Jiol of PBS and 8 Jiol of 5 mM DCFH-DA, (Molecular Probes,Eugene, OR) was added. Following 15 min of incubation, 30 Jiol ofLDS-751 (1.67 mg/ml) (Exciton, Dayton, OH) was added just priorto fiow cytometric analysis of cell fluorescence. For F-met-Ieu-phe(FMLP)-induced reactive oxygen species measurement, DCFH-DAlabeled samples were incubated with 25 Jiol of 20 mM FMLP (Sigma)at 37°C for 30 mino Samples were analyzed at O min and 30 min ona Becton-Dickinson FACStar dual-channel flow cytometer (BectonDickinson Immunocytometry Systems, San Jose, CA). The instru-ment was set up to measure linear forward scatter, which is a mea-sure of particle size, linear 900 light scatter, which is a measure ofcell granularity, and green fiuorescence at 535 nm (corresponding toDCFH) and red fluorescence (LDS-751) at 620 nro. LDS-751 wasused to label the nuclei of living cells and permit their identificationin whole blood without interference from erythrocytes. Nucleated(red fiuorescent) cells were analyzed for PMN mean channel fiuores-cence at 535 nro. There were five animals in each group.

Statistical analysis of data. MPO data are expressed as units per100 mg of lung tissue. For the MPO, base-deficit and respiratoryburst experiments, statistical comparisons between experimentalgroups were performed by multiple analysis ofvariance (when therewere more than 2 groups) and the e-test (Instat software, San Diego,CA), with significance assumed at P < 0.05. For the radioisotopestudies, the results obtained for the various organs in the controland burned groups of animals were compared using the Mann- Whit-ney U-test for non-parametric data, with the chosen level of signifi-cance at P < 0.05, two tailed distribution. AlI data are shown withthe mean and standard error of the mean indicated.

NJ), 0.4 mg/100 g b.w. Animals were shaved and depilated usingNair(Carter-Wallace Inc., New York, NY). The animals then underwenttracheostomy and were allowed to breathe spontaneously throughthe tracheostomy tube (3/4-in. polyethylene tube, 2 mm i.d.). Thefemoral or internal jugular vein was then cannulated by cutdownwith polyethylene tubing (PE 50). Line patency was maintained with.2 cc of a 10 U/cc heparin in .9% NaCI solution. Animals were main-tained normothermic by being placed on a warm (37°C) heating pad.

A convex, body-conforming stainless steel template (approximately8 cm x 5.8 cm x .4 cm) was heated to 250°C using a thermistor forcalibration, and the template was applied to the depilated dorsumof the anesthetized animal for 7 seco The animal was resuscitatedwith 20 cc of .9% NaCI i.p. immediately postburn. The templateproduced a burn covering approximately 17% of the body surfacearea. Slight changes in template size were utilized for various b. W.sto produce burn sizes ofuniform percentage total body surface areas;surface area was calculated from a published formula for small ani-

mals [16].Duration of the experiments was up to 24 hr postburn. Animals

were euthanized and viscera were immediately procured and assayedat appropriate time points. For the radioisotope studies, rats werekept anesthetized for 5 hr so that the animals could not removeintravenous catheters. For the analysis of lung myeloperoxidase(MPO), animals were permitted to awaken postburn and they were

then killed at indicated time points.Measurement oftissue MPO contento At various time points aQ:.er

burn injury, animals were killed by cervical dislocation. The chestcavity was opened and the lungs were flushed with normal salinevia cardiac puncture ofthe still-beating heart. Lungs were harvested,weighed, and snap frozen at -70°C for subsequent batch processing.MPO was assayed by measuring the H2O2-dependent oxidation of 0-dianisidine. Lung tissue samples were homogenized and resus-pended in .45% saline. After centrifugation the supernatant was dis-carded and the pellet resuspended in 1 ml/50 mg tissue of .05 Mpotassium phosphate buffer, pH 6.0, containing .55 mixed trimethy-lammonium bromide (Sigma, Sto Louis, MO). This was followed bya three time freeze-thaw cycle and additional sonication. After cen-trifugation at 4°C, .033 mI of the supernatant was mixed with .967mI of .05 M potassium phosphate buffer, pH 6.0, containing .167 mg/mI of o-dianisidine hydrochloride (Sigma, Sto Louis, MO) and .4 JLl/mI of 3% H2O2, and absorbance was immediately measured at 460nm (DU-7 Spectrophotometer, Beckmann Instruments, Fullerton,CA). A standard curve was created using human MPO (Sigma, StoLouis, MO). MPO is a heat-stable enzyme, and by heating the tissuecertain factors which can interfere with the assay are eliminated[17]. There were four to six animals in each control and experimental

group.Measurement ofradiolabeled PMN and erythrocyte contenta in tis-

sues. Leukosequestration in various tissues during burn shock wasstudied using radiolabeled autologous PMNs and erythrocyteS. Im-mediately before the burn injury was created, l°ml of venous bloodwas drawn from the indwelling catheter into a heparinized tube forpreparation ofthe cells. The PMN fractiqn was obtained by centrifug-ing diluted whole blood on a two-step.Percoll (Pharmacia- Biotech,Sollentuna, Sweden) density gradient [18]. The PMN fraction wasretrieved and labeled with lllIn as described [19]: 1 rol of the cellsuspension (.10-3 x 106 cells) was mixed with 1-5 JLI ofll1ln (indiumoxide) solution (37 MBq/ml, Amersham International plc, Amer-sham, UK) and incubated for 30 min at ro'om temperature. Cellswere washed twice and resuspended in 1 mI of phosphate-bufferedsaline (PBS). Radioactivity was measured in a gamma counter before

reinjection via the venous catheter.Erythrocyte labeling was performed on the leukocyte-free fraction

obtained at the separation procedure described above. Cells weresuspended in 1 mI of Hanks' Balanced Salt Solution and incubatedwith 20 JLI of51Cr sodium chromate solution (37 MBq/ml, Amersham)for 30 min at room temperature. Cells were washed twice, pooledwith the PMNs, and injected 4.5 hr aft.er the burnwas performed.

Al-rol blood sample was drawn from the venous catheter 5 hraft.er the burn and immediately before the rats were sacrificed byrapid installation of 5 mI glutaraldehyde (2.5% in .05 M sodiumcacodylate) through the tracheostomy. The procedure insures that

RESULTS

HANSBROUGH ET AL.: LEUKOSEQUESTRATION AFTER BURN INJURY 19

~~u.tw~¡:1-iñz

~1-W

~~:;)W-J

o 6 12 18 24

HOURS POSTBURN

FIG. 1. Base-deficit profile in rats foIlowing 17% fuIl-thicknessbody surface area burn injury. There were 5 to 8 animals in eachgroup.

Lung Liver Spleen Gut Kidney Muscie Skin Brain

FIG. 3. Tissue neutrophil flux was measured by determinationofradiolabeled neutrophils and erythrocytes in tissues, 5 hr following17% TBSA burn injury. There were 5 animals in each experimentaland control group.

Evaluation of radiolabeled albumin extravasation intissues showed no detectable edema in any of the tis-sues or organs, and no differences in edema betweenthe burned and nonburned animals.

DISCUSSION

Polymorphonuclear leukocytes (PMNs) play an im-portant role in host defenses by releasing oxygen radi-cals and lysosomal enzymes to kill engulfed microor-ganisms. Although PMNs are important for killing mi-croorganisms, the accumulation of PMNs at sites oftissue injury coupled with excessive release of prote-ases and oxygen metabolites may produce tissue injurywhich can lead to organ dysfunction and organ failure[6-13, 25]. The hydroxyl radical has algO been shownto actívate complement, which leads to further undesir-able inflammatory sequelae following shock [26]. Theseevents may be related to the actions of multiple in-flammatory mediators following thermal injury, whichmay have effects at both local and distant sites [25,27-29].

It has become clear that events associated with dis-tant tissue ischemia and reperfusion can induce localtissue injury. These effects can be mediated by the in-duced expression of multiple cell adhesion moleculeson the surfaces of PMNs and endothelial cells [7, 13].

burn (Fig. 1). At 18 and 24 hr postburn, the base deficitvalues were unchanged from those of control animals.

Mortality after burn injury ranged from 20% to 50%.No mortalities occurred in the unburned groups. Themortalities were high and indicated that the burn in-jury was extremely deep. Previous studies in our labo-ratory showed that increasing the amount ofresuscita-tion fluid did not improve animal survival. Tissues fromanimals which expired spontaneously were not utilizedfor obtaining any of the data in these studies.

Figure 2 shows the results of MPO determinationsin lung tissue at various time periods after bum injury.MPO values were increased from control animallevelsat the 8- through 12-hr time points postburn.

PMN sequestration as determined by the leukocytetransit factor showed significantly higher values inburned animals for the lungs, kidneys, liver, and non-burned skin than for the control group (Fig. 3). Nostatistical difference was observed for skeletal muscleor spleen, although numerically the values were mark-edly increased in the burned animals in these two tis-sues.

Histologic slides of the lung tissues were evaluatedand showed marked accumulation of leukocytes at 5hr postburn.

Analysis of the PMN respiratory burst (Fig. 4)showed that intracellular H2O2 was significantly in-creased in rats 5 hr postburn, compared to control rats.

NO BURN BURN

FIG.4. Intracellular levels ofH2O2 were quantified in rat periph-eral blood neutrophils utilizing flow cytometry and a dye which altersits fluorescente after reacting with intracellular H2O2. There were5 animals in each group. Numbers and dashed lines indicate P valuescomparing various experimental groups.

CONTROL 4 8 12 24

I-HOURS POSTBURN-I

FIG. 2. Myeloperoxidase (MPO) was measured in lung tissue inrats following 17% body surface area bum. Lungs were flushed insitu by injecting saline in the beating heart, to eliminate circulatingPMNs from lungs before harvesting lung tissue. There were 4 to 6animals in each experimental group.

20 JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996

These molecules include integrins, selections, and in-tracellular adhesion molecules [30, 31]. It is now be-lieved that the ischemic/reperfused gastrointestinaltract may serve as a priming bed for circulating PMN s,where upregulation oí surface receptors via the releaseof various inflammatory mediators appears to occur[32-34]. Most PMN emigration occurs in specializedregions of the vascular tree, i.e. in postcapillary ve-nules, where leukocytes are normally only loosely teth-ered to the vessel wall as they roll along the surfaceon the endothelium. PMN recruitment to sites of in-flammation occurs in several discrete steps involvingrolling of cells on activated endothelium, activation ofPMNs, adhesion ofPMNs to endothelium, and migra-tion of PMNs into surrounding tissues [35]. AlthoughPMN adherence appears to be influenced by the CDI8/CDll integrin, as well as endothelial selectins [13, 36],PMN extravasation and superoxide release appearmore specifically governed by the CDllb (Mac-l, Mol,CRa) subunit. Thus the primed state of the PMN maybe a decisive factor in the production of tissue injury[37]. Evidence suggests that trauma including burnsmay upregulate the baseline and stimulus.;mediat~dproduction oftoxic oxygen intermediates by PMNs [11,31, 38]. Thus, recruitment of PMNs into tissue sitesfollowed by release of toxic mediators may induce tis-sue injury.

MPO is a constituent enzyme of azurophilic neutro-philic granules, constituting 5% of PMN dry weight[39,40]. Assessment oftissue MPO activity provides anindirect assessment ofPMN numbers, which correlateswell with morphometric evaluations [41-43] and withradioisotope evaluation of sequest~ring cells [44].Nonetheless, there are theoretical concerns with utiliz-ing MPO values to estimate tissue PMN content, pri-marily related to the possible release of MPO by acti-vated PMNs into the plasma, prior to or during theiradherence to endothelium. Another concern relates tothe relative insensitivity of the MPO assay. Back-ground levels of tissue MPO are at the barely detect-able range of the assay, and therefore the actual in-crease in tissue MPO content may be far greater thanreflected in the experiments. .

While removal of leukocytes from the animal fol-lowed by their radiolabeling in vitro may permit moresensitive measurement ofleukocyte adherence tó endo-thelium and subsequent extravasation into tissues, theremoval, manipulation, and reintroduction of leuko-cytes could theoretically result in their activation andsubsequent exaggerated responses in tissues. Thus theMPO results serve as a backup method of analysis ofpulmonary leukosequestration in thes.e studies. Whilethe spectrophotometric assay which was utilized herealBo measures the activity of erythrocyte catalases andperoxidases and ofMPO in PMNs in residual blood inthe lung tissue, blood peroxidase contributes only about7% oftotal measured lung peroxidase activity in blood-perfused lung [41]. By using the improved assay forMPO, [17] erythrocyte enzymes and other interferingproteins were eliminated by heat inactivation since

MPO is highly stable to exposure to higher tempera-tures.

While PMN adherence to endothelium is an im-portant component of normal host defenses [35, 45],reports have demonstrated that PMN adherence to en-dothelium may be dissociated from PMN-induced tis-sue injury [33, 46] which occurs primarily via releaseof toxic oxygen intermediates and proteases into thelocal tissue environment [7-10, 32,47]. This eventmayrequire activation ofPMNs by a second stimulus [48].Although PMN adherence appears to facilitate endo-thelial injury by creating a local protective environ-ment for the cytotoxic agents which are released bystimulated PMN s, PMN adherence may algo occur andbe unrelated to release of injurious mediators.

Our studies demonstrate that leukosequestrationinto multiple tissues occurs during acute burn shockin this rat burn modelo Absence ofleukocyte sequestra-tion in the gut has been previously reported followingburn injury by others in the rat [49], and our own labo-ratory found no gut leukosequestration in burned mice[15]. Leukosequestration may be a result of upregula-tion of adhesion receptors by capillary endothelium.The differential relationship between integrins andleukocyte adherence to lung and skin endothelium hasbeen reported following burn injury [31], but we foundsimilar levels of leukosequestration in those two tis-sues. This could suggest that multiple mechanismscould contribute to the events. Leukosequestration inBorne tissue beds could be secondary to severe vasocon-striction during acute burn shock, with microvascularstasis and resultant opportunistic leukocyte adherenceto endothelium. This process could be accentuated bythe known process of PMN aggregation, a processwhich has been shown to be primarily a CD11b-depen-dent event [50], and aggregation between PMNs andplatelets, an event which appears primarily dependenton P-selectin [51]. In addition, results from filter stud-ies performed by Borne ofthe investigators contributingto this paper indicate that non-viscous properties ofleukocytes (elasticity and adhesiveness) may becomeimportant determinators for the passage of leukocytesthrough narrow fiow channels at low driving pressures[52]. The importance ofrheological properties ofleuko-cytes for their pulmonary margination at low fiow wasshown in previous experiments on isolated, artificiallyperfused rat lungs [18]. Flow-dependent capillary re-tardation ofleukocytes has alBo been shown in isolated,perfused skeletal muscle [53].

Further investigations into the mechanisms relatedto leukosequestration may yield therapies which willprotect the tissues from injury, particularly the lungs,during acute burn shock. Protection may alBo be desir-able during extensive surgical procedures followingburn injury, since our studies in mice suggest that earlyburn wound excision may exacerbate pulmonary leuko-sequestration {15]. Such therapies could include theuse of blocking agents for various intracellular adhe-sion molecules or selectins [31, 54].

The analysis ofPMN respiratory burst activity dem-

HANSBROUGH ET AL.: LEUKOSEQUESTRATION AFTER BURN INJURY 21

onstrated that burn injury significantly upregulatedbaseline and FMLP-stimulated intraceIlular H2O2 con-tent. Our studies with radiolabeled albumin demon-strated that at this"..early time point (5 hr) after burninjury, tissue edema could not be demonstrated. Al-though Mulligan et al. [31] demonstrated edema in tis-sues foIlowing burn injury in rats, a much larger burninjury was utilized in their report (25 to 30% total bodysurface area) than in our studies. In addition, our fail-ure to determine edema formation may relate to theexperimental model which we utilized; the radiolabeledalbumin was injected only 30 min prior to sacrifice ofthe animals. Future experiments wiIl evaluate edemaformation over a longer time line.

Although PMNs are key participants in the develop-ment of tissue injury in a variety of acute inflammatorydiseases as well as foIlowing trauma, the mere presenceoftissue PMNs does not prove tissue injury. However,the increased respiratory burst activity of peripheralblood PMN s in burned rats, as demonstrated in ourstudies, suggests that leukosequestration of metaboli-caIly active PMN s is occurring in the burned animals.Further studies are underway to study the relativeroles of tissue leukosequestration and PM;N activationleading to tissue damage foIlowing thermal injury.

REFEREN CES

1. Burke, J. F., Bondoc, C. C., and Quinby, W. C. Primary burnexcision and immediate graf'ting: A method shortening illness.J. Trauma 14: 389, 1974.

2. Sheridan, R. L., Tompkins, R. G., and Burke, J. F. Managementoy the burn wound by prompt excision and immediate closure.J. Intensive Care Med. 9: 6, 1994.

3. Tompkins, R. G., Burke, J. F., Schoenfeld, D. A., Bondoc, C. C.,Quinby, W. C., Behringer, G. C., and Ackroyd, F. W. Prompteschar excision: A treatment system contributing to reducedburn mortality: A statistical evaluation ofburn care at the Mas-sachusetts General Hospital (1974-1984). Ann. Burgo 204: 272,1986.

4. Hansbrough, J. F. Burn wound sepsis. J. Intensive Care Med.2: 313,1987.

5. Peck, M. D., and Heimbach, D. M. Does early excision of burnwounds change the pattern of mortality? J. Burn Care Rehab.10: 7, 1989.

6. Schmeling, D. J., Caty, M. G., Oldham, K. T., Guice, K. S.,and Hinshaw, D. B. Evidence for neutrophil-related acute lunginjury af'ter intestinal ischemia-reperfusion. Surgery 106: 195,1989.

7. Albelda, S. M., Smith, C. W., and Ward, P. A. Adhesion mole-cules and inflammatory injury. FASEB J. 8: 504, 1994.

8. Till, G. O., Johnson, K J., Kunkel, R. G., and Ward, P. A.Intravascular activation of complement and acute lung injury:dependency on neutrophils and toXic oxygen metabolites. J.G/in. Invest. 69: 1126, 1982.

9. Tate, R. M., and Repine, J. E. Neutrophils and the adult respira-tory distress syndrome. Am.Rev. Respir. Dis. 128: 552, 1983.

10. Smedley, L. A., Tonnesen, M. G., Sandhaus, R. A., Haslett, C.,Guthrie, L. A., Johnston, R. B., Jr., Henson, P. M., and Worthen,G. S. Neutrophil-mediated injury to endothelial cells: enhance-ment by endotoXin and essential role of neutrophil elastase. J.G/in. Invest. 77: 1233, 1986.

11. Maderazo, E. G., Woronick, C. L., Albano, S. D., Breaux,S. P., and Pock, R. M. Inappropriate activation, deactivation,

and probable auto-oxidative damage as a mechanism of neutro-phillocomotory defect in trauma. J.lnfect. Dis. 154: 471,1986.

12. Worthen, G. S., Haslett, C., Rees, A. J., Gumbay, R. J., Henson,J. E., and Henson, P. M. Neutrophil-mediated pulmonaryvascu-lar injury: synergistic effect oftrace amounts oflipopolysaccha-ride and neutrophil stimuli on vascular permeability and neu-trophil sequestration in the lungs. Annu. Rev. Resp. Dis. 136:19, 1987.

13. McMillen, M. A., Huribal, M., and Sumpio, B. ComInon pathwayof endothelial-leukocyte interaction in shock, ischemia, and re-perfusion. Am. J. Burgo 166: 557, 1993.

14. Tenenhaus, M., Hansbrough, J. F., Zapata-Sirvent, R., andNeumann, T. Treatment ofburned mice with intermittent hy-perbaric oxygen reduces mesenteric bacteria colonies but doesnot decrease néutrophil deposition in lungs. Arch. Burgo 129:1338, 1994.

15. Rennekampff, H. O., Tenenhaus, M., Hansbrough, J. F., andKiessig, V. Leukosequestration following burn injury in micemeasured by an improved myeloperoxidase assay: A time coursestudy. Am. Burn Assoc. 27th Annu. Mtg., Albuquerque, April1995.

16. Walker, H. L., and Mason, A. D. Jr. A standard animal bum.J. Trauma 8: 1049, 1965.

17. Schierwagen, C., Bylund-Fellenius, A.-C., and Lundberg, C. Im-proved method for quantification of tissue PMN accumulationmeasured by myeloperoxidase activity. J. Pharmacol. Methods23: 179, 1990.

18. Wikstrom, T., Braide, M., Bagge, U., and Risberg, B. Erythro-cyte-Ieukocyte interactions in the vascular bed of isolated per-fused rat lungs. Int. J. Microcirc. Clin. Exp. 12: 17, 1992.

19. Gunter, K. P., Lukens,J. N., Clanton,J.A., Morris,P. J., Janco,R. L., and English, D. Neutrophillabeling with Indium-111:Tropolone ve. oxine. Radiology 149: 563, 1983.

20. Wikstrom, T., Braide, M., Bagge, U., and Risberg, B. Leukocytemargination during hemorrhagic shock correlates to preshockmargination and is reduced by fucoidin. Shock 3: 40, 1995.

21. Keston, A. S., and Brandt, R. The fluorometric analysis ofultra-micra quantities of hydrogen peroxide. Anal. Biochem. 11: 1,1965.

22. Hansbrough, J. F., Zapata-Sirvent, R. L., and Cooper, M. L.Effects oftopical antimicrobial agents on the human neutrophilrespiratory burato Arch. Burgo 126: 603, 1991.

23. Zapata-Sirvent, R. L., Hansbrough, J. F., Cox, M. C., and Car-ter, W. H. Immunologic alterations in a murine model ofhemor-rhagic shock. Crit. CaTe Med. 20: 508, 1992.

24. Himmelfarb, J., Hakin, R. M., Holbrook, D. G., Leeber, D. A.,and Ault, K. A. Detection of granulocyte reactive oxygen speciesformation in whole blood using flow cytometry. Cytometry 13:83, 1992.

25. Baue, A. T. Multiple organ failure, multiple organ dysfunctionsyndrome, and the systemic inflammatory response syndrome-where do we stand? Shock 2: 385, 1994.

26. Turnage, R. H., Magee, J. C., Guice, K. S., Myers, S. l., andOldham, K T. Complement activation by the hydroxyl radicalduring intestinal reperfusion. Shock 2: 445, 1994.

27. Harms, B. A., Wenger, H., and Demling, R. H. Prostaglandinrelease and altered microvascular integrity after burn injury.J. Burgo Res. 31: 274, 1981.

28. Demling, R. H., Wenger, H., LaLonde, C., Hechtman, H., Wong,C., and West, K. Endotoxin-induced prostanoid production inthe burn wound can cause distant lung dysfunction. Surgery99: 421, 1986.

29. Rodriguez, J. L., Miller, C. G., Gamer, W. L., Till, G. O., Guer-rero, P., Moore, N. P., Corridore, M., Normolle, D. P., Smith,D. J., and Remick, D. G. Correlation ofthe local and systemiccytokine response with clinical outcome following thermal in-jury. J. Trauma 34: 684, 1993.

30. Mulligan, M. S., Polley, M. J., Bayer, R. J., Nunn, M. F., Paul-

22 JOURNAL OF SURGICAL RESEARCH: VOL. 61, NO. 1, FEBRUARY 15, 1996

43. Schierwagen, C. A., Bylund-Fellenius, A. B., and Lundberg, C.Improved method for quantification of tissue PMN accumula-tion measured by myeloperoxidase activity. J. Pharmacol.Methods 23: 179, 1990.

44. Lundberg, C., and Arfors, K.-E. Polymorphonuclear leukocyteaccumulation in infiammatory dermal sites as measured by51Cr-Iabeled cells and myeloperoxidase. lnflammation 7: 247,1983.

45. Harlan,J.M., Vedder,N.B., Winn, R. K,andRice, C. C.Mecha-nisms and consequences of leukocyte-endothelial interactions.West. J. Med. 155: 365, 1991.

46. Hill, J., Lindsay, T., Rusche, J., Valeri, C. R., Shepro, D., andHechtman, H. B. A MAC-l antibody reduces liver and lunginjury but not neutrophil sequestration alter intestinal isch-emia/reperfusion. Surgery 112: 166, 1992.

47. Granger, D. N. Role of xanthine oxidase and granulocytes inischemia-reperfusion injury. Am. J. Physiol. 255: H1269, 1988.

48. Anderson, B. O., Bensard, D. D., Brown, J. M., Repine, J. E.,Shanley, P. F., Leff, J. A., Terada, L. S., BaneIjee, A., andHarken, A. H. FNLP injures endotoxin-primed rat lungs byneutrophil dependent and independent mechanisms. Am. J.Physiol. 260: R413, 1991.

49. Trop, M., Schriffrin, E. J., and Carter, E. A. Role ofneutrophilsin the intestinal alterations associated with thermal injury.Burns 16: 343, 1990.

50. Anderson, D. C., Miller, L. J., Schmalstieg, F. C., Rothlein, R.,and Springer, T. A. Contributions of the Mac-l glycoproteinfamily to adherence-dependent granulocyte functions: struc-ture-function assessments employing subunit-specific mono-clonal antibodies. J. lmmunol. 137: 15, 1986.

51. Buttrum, S. M., Hatton, R., and Nash, G. B. Selectin-mediatedrolling of neutrophils on immobilized platelets. Blood 82: 1165,1993.

52. Braide, M., Nannmark, U., and Bagge, U. Bolus filtration ofleukocytes: A method to study the pressure dependence of leu-kocyte filter pore passage time distributions. Clin. Hemorrheol.11: 369, 1991.

53. Braide, M., Nannmark, U., and Bagge, U. Quantitative studieson the influence of leukocytes on the vascular resistance in askéletal muscle preparation. Microvasc. Res. 27: 331, 1984.

54. Morisaki, T., Goya, T., Toh, H., Nishihara, K., and Torisu, M.The anti Mac-l monoclonal antibody inhibits neutrophil seques-tration in lung and liver in a septic murine modelo Clin. lmmu-noto lmmunopathol. 61: 365, 1991.~

son, J. C., and Ward, P. A. Neutrophil-dependent acute lunginjury. Requirement for P-selectin (GMP-140). J. Clin. Invest.90: 1600, 1992.

31. Mulligan, M. S., Till, G. O., Smith, C. W., Anderson, D. C.,Miyasaka, M., Ta;nafani, T., Todd, R. F., III, Issekutz, T. B.,and Ward, P. A. Role of leukocyte adhesion molecules in lungand dermal vascular injury after thermal trauma of skin. Am.J. Pathol. 144: 1008, 1994.

32. Cioffi, W. G., Burleson, D. G., and Pruitt, B. A. Jr. Leukocyteresponses to injury. Arch. Surg. 128: 1260, 1993.

33. Anderson, B. O., Moore, E. E., Moore, F. A., Leff, J. A., Terada,L. S., Harken, A. H., and Repine, J. E. Hypovolemic shock pro-motes neutrophil sequestration in lungs by a xanthine oxidase-related mechanism. J. Appl. Physiol. 71: 1862, 1991.

34. Koike, K., Moore, E. E., Moore, F. A., Carl, V. S., Pitman,J. M., and Banerjee, A. Phospholipase A2 inhibition decoupleslung injury from gut ischemia-reperfusion. Surgery 112: 173,1992.

35. Butcher, D. C. Leukocyte-endothelial cell recognition: three (ormore) steps to specificity and diversity. Cell 67: 1033, 1991.

36. Aiteri, D. C., and Edgington, T. S. A monoclonal antibody re-acting with distinct adhesion molecules defines a transition inthe function state of the receptor CD11b/CD18 (Mac-1). J. Im-munol. 141: 2656, 1988.

37. Vercelotti, G. M., Yin, H. Q., Gustafson, K. S., Nelson, R. D.,and Jacob, H. S. Platelet activating factor primes neutrophilresponse to agonists: role in promoting neutrophil mediatedendothelial damage. Blood 71: 1100, 1988.

38. Cioffi, W. G., Jr., Burleson, D. G., Jordan, B. S., Mason, A. D.,Jr., and Pruitt, B. A., Jr. Granulocyte oxidative activity follow-ing thermal injury. Surgery 112: 860, 1992.

39. Kinkade, J. M., Pember, S. O., Barnes, K. C., Shapira, R., Spitz-nagel, J. K., and Martin, L. E. Differential distribution of dis-tinct forros of myeloperoxidase in different azurophilic granulesubpopulations from human neutrophils. Biochem. Biophys.Res. Commun. 114: 296, 1983.

40. Schultz, J., and Kaminker, K. Myeloperoxidase ofthe leukocyteof normal human blood.. l. Content and localization. Arch. Bio-chem. 96: 465, 1962.

41. Anderson, B. O., Brown, J. M., Shanley, P. F., Bensard, D. D.,and Harken, A. H. Marginating neutrophils are reversibly ad-herent to normallung endothelium. Surgery 109: 51, 1991.

42. Goldblum, S. E., Wu, K., andJay, M. Lung rnyeloperoxidase asa measure ofpulmonary leukostasis in rabbits. J. Appl. Physiol.59: 1978, 1985.