b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1

Clostridial collagenase aggravates the systemic inflammatoryresponse in rats with partial-thickness burns

Zafer Dokumcu a, Orkan Ergun a, Handan Ak Celik b, Sohret Aydemir c,Murat Sezak d, Geylani Ozok a, Ahmet Celik a,*aDepartment of Paediatric Surgery, Ege University Faculty of Medicine, Izmir, TurkeybDepartment of Biochemistry and Clinical Biochemistry, Ege University Faculty of Medicine, Izmir, TurkeycDepartment of Microbiology and Clinical Microbiology, Ege University Faculty of Medicine, Izmir, TurkeydDepartment of Pathology, Ege University Faculty of Medicine, Izmir, Turkey

a r t i c l e i n f o

Article history:

Accepted 3 January 2008

Keywords:

Clostridial collagenase

Systemic inflammatory response

Cytokines

Second-degree burn

a b s t r a c t

Aim: Clostridial collagenase A (CCA) has been shown effective in degrading collagen in

eschar tissue and promoting healing in partial-thickness burns. As there are also reports of

fever, leukocytosis, increased C-reactive protein (CRP) levels and septic complications

during treatment with CCA, we aimed to determine in rats whether CCA aggravates the

systemic inflammatory response.

Methods: Rats with partial-thickness burns were randomly divided into groups with either

no dressing (ND), povidone-iodine dressing (PID) or CCA dressing (CCAD). Body weights and

temperatures, blood leukocyte counts, and serum levels of CRP, interleukin-1b (IL-1b),

interleukin-6 (IL-6) and tumour necrosis factor-a (TNF-a), were measured at 0, 3, and

24 h and days 3 and 7 from burn. Wounds were cultured on days 1, 3 and 7 and burn depth

was evaluated on day 1.

Results: Body weights for all groups were significantly lower after burn, with highest loss

(25.5%) in the CCAD group. At 3 h a significant drop in rectal temperature was noted in all

groups. The CCAD group had higher rectal temperature levels than the PID group on days 3

and 7 ( p < 0.05). Changes in serum levels of CRP, IL-1b, IL-6 and TNF-a were not significant in

the ND and PID groups; the CCAD group showed a significant rise in serum levels of CRP on

day 1, of IL-6 on day 3 and of TNF-a on day 7. Wound infection was more common in CCAD

group and increased on days 3 and 7, but this was insignificant.

Conclusion: CCA aggravated the systemic inflammatory response in rats with partial-thick-

ness burns, which is accompanied by a higher risk of infection.

# 2008 Elsevier Ltd and ISBI. All rights reserved.

avai lab le at www.sc iencedi rec t .com

journal homepage: www.e lsev ier .com/ locate /burns

1. Introduction

Severe burn has hazardous effects on skin cell structure and

function; these effects vary with the intensity and duration of

the trauma and the thickness of the effected area. The eschar,

i.e. the zone of coagulation necrosis formed in the area of

contact, is surrounded by an injured but not yet necrotic area,

* Corresponding author. Tel.: +90 232 390 2800.E-mail address: ahmet.celik@ege.edu.tr (A. Celik).

0305-4179/$34.00 # 2008 Elsevier Ltd and ISBI. All rights reserved.doi:10.1016/j.burns.2008.01.001

the zone of stasis, and a peripheral hyperaemic area. In the

presence of factors such as tissue hypoxia and the actions of

specialised substances, progressive necrosis may be inevitable

in the zone of stasis.

When the skin is traumatised in any way, the process of a

natural defence mechanism (so-called acute inflammation) is

triggered and numerous interactions occur between immune

b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1936

cells via macromolecules. It was discovered that these

mediators not only act locally but also systemically by

activation of inflammatory pathways and of the immune

system, involving and modulated by cytokines induced shortly

after injury, growth factors, vascular endothelial and other cells

[1]. In certain circumstances, like a ‘domino effect’, inflamma-

tory response becomes widespread, because overproduction of

cytokines and activation of leukocytes and endothelial cells

lead to overproduction of inflammatory substances. This may

end in SIRS (systemic inflammatory response syndrome), ARDS

(acute respiratory distress syndrome), MODS (multiple organ

dysfunction syndrome) or even death [2,3]. The roles of tumour

necrosis factor-a (TNF-a), interleukin-1b (IL-1b) and interleu-

kin-6 (IL-6) are also important in the inflammatory response.

These three cytokines originate principally from macrophage

activation and have been described as indicators of severity of

injury, as well as of infection [4–7].

The liver plays a major role in the body’s response to injury,

with critical changes in its biosynthesis profile. Synthesis of

acute phase proteins (APP) is regulated by the liver and

numerous physiological, biochemical and behavioural

changes may take place [8]. As one of the best-known APP’s,

C-reactive protein (CRP) is released from liver 6–8 h after

injury, reaching highest levels at 24–48 h. CRP acts as a

scavenger with an affinity for chromatin released during

cellular necrosis due to inflammation. IL-6 and TNF-a induce

CRP release from the liver and have been identified as

important indicators of a poor prognosis [9].

Eschar provides a medium for the growth of microorgan-

isms that may lead to serious infections, and should be

removed to start the wound healing process. Homogenate of

Clostridium hystoliticum is known to be one of the most

efficacious agents for disintegration of all types of collagen,

by hydrolisation of the 3-helical structure [10]. Since the 1990s,

clostridial collagenase A (CCA) has been used as an effective

debriding agent in numerous studies [11–14]. In 1995,

Hansbrough et al. demonstrated in a prospective, randomised,

controlled, multi-institutional study that superficial second-

degree burns treated with collagenase ointment achieved a

clean wound bed free of necrotic debris in a shorter period of

time, and healed faster, than burns treated with silver

sulfadiazine cream [12]. In 2002, the first study in the English

literature of successful enzymatic burn debridement with

collagenase in the treatment of children with burns was

documented by our institution. In this report CCA was shown

to be as effective as surgical excision, with advantages of

shorter hospital stay and less need for blood transfusion and

grafting. CCA has also been shown to reduce excision count

and area in cases where CCA treatment had to be converted to

surgical excision, because of wound infection or manifest

need for grafting [14]. However, CCA was suspected to cause

fever, tachycardia, leukocytosis or leukocytopenia and eleva-

tion in serum CRP levels even in the absence of infection. In

our preliminary study, we reviewed the hospital records of 32

children treated with CCA for second-degree burn; 46% of

these (15/32) developed these clinical signs predominantly

between days 3 and 5, without detectable wound infection

(unpublished data). The present study is designed to show the

relationship between use of CCA ointment and inflammatory

responses in rats with partial-thickness burns.

2. Methods

With the approval of the local animal research Ethics

Committee, the experimental research centre at Ege Uni-

versity Faculty of Medicine supplied 24 male Wistar Albino

rats weighing between 200 and 280 g. The rats were housed in

separate cages at 25 � 3 8C, with a 12-h light/12-h dark cycle,

for at least 5 days before the experiment, and were fed freely

with standard rat chow and water.

On the day of the experiment, the rats were anaesthetised

with intramuscular ketamine hydrochloride (75 mg/kg) and

xylazine hydrochloride (10 mg/kg). Hair on the dorsum, i.e. on

20% of the total body surface area (TBSA), was shaved and the

area was scrubbed with povidone-iodine and cleaned with

70% ethanol. Partial-thickness scald was created by submer-

ging the shaved dorsum of the animals in water at 70 8C for

30 s, as previously described [15]. All animals received lactated

Ringer’s solution (20 ml/kg) and dolantine (1 mg/kg/day)

intraperitoneally for fluid resuscitation and pain control.

Rats were randomly divided in to three groups according to

wound dressings. Mechanical debridement with saline was

not performed in any of the groups. Povidone-iodine (0.5 g/

cm2) and CCA (0.5 g/cm2) were applied in groups povidone-

iodine dressing (PID) and CCA dressing (CCAD), respectively,

no dressing (group ND) served as a control group with no

dressing. All dressings were changed every 24 h under sterile

conditions with ether anaesthesia.

Body weight and temperature of each rat were noted at

specified time intervals after burn (0, 3, and 24 h and days 3

and 7), when blood samples were obtained via intracardiac

puncture for leukocyte count and biochemical analysis. Test

tubes with ethylenediamine tetra-acetic acid were used to

prevent clotting. Leukocyte counts were performed on Thoma

lames after the samples had been diluted with Turk solution.

For measurement of serum levels of CRP, IL-1b, IL-6 and TNF-a,

serum specimens were collected after centrifugation and were

stored in ultracold until analysed.

In CRP assay, CRP in the serum sample reacted with anti-

CRP bodies adsorbed on polystyrene microtitre wells. After

removal of unbound serum proteins by washing, anti-CRP

antibodies conjugated with horseradish peroxidase were

added. These enzyme-labelled antibodies formed complexes

with the previously bound serum CRP. Following further

washing, the enzyme bound to the immunosorbant was

assayed by addition of a chromogenic substrate, 3,30,5,50-

tetramethylbenzidine. The quantity of bound enzyme varied

directly with the concentration of CRP in the sample tested;

thus, the adsorbance at 450 nm was a measure of the

concentration in the test sample. The quantity of CRP in

the test sample could be extrapolated from the standard

curve constructed from the standards and corrected for

serum dilution.

To estimate IL-1b, the BioSource Rat Interleukin-1b (rt IL-

1b) ELISA (Biosource, Camarillo, CA), was used. An antibody

specific for rt IL-1b was coated onto the wells of the microtitre

strips provided. Samples, including standards of known rt IL-

1b content, control specimens and unknowns, were pipetted

into these wells. During the first incubation, the rt IL-1b

antigen bound to the immobilised antibody on one side. After

washing, a biotinylated antibody specific for rt IL-1b was

Fig. 1 – Mean body weights of the groups. Error bars

represent S.E.M.; stars: p < 0.05 on comparison to initial

values for each group.

b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1 937

added. During the second incubation, this antibody bound to

the immobilised rt IL-1b captured during the first incubation.

After removal of excess secondary antibody, streptavidin-

peroxidase (enzyme) was added. This bound to the biotiny-

lated antibody to complete the four-member sandwich. After a

third incubation and washing to remove all unbound enzyme,

a substrate solution was added and was acted upon by the

bound enzyme to produce colour. The intensity of this

coloured product was directly proportional to the concentra-

tion of rt IL-1b present in the original specimen. The plates

were read spectrophotometrically in a microplate reader at

450 nm. The ELISA was performed using BioSource Rat IL-6 kit

(BioSource, Camarillo, CA). The method in the assay was the

same as outlined in IL-1b.

An ELISA was also performed using the BioSource Rat TNF-

a kit (BioSource, Camarillo, CA). The method was the same as

outlined for 1b, but incubation times and reagent concentra-

tion differed. For precision, final results of the ELISA assays

were the mean of three repeated calculations.

Tissue culture samples, collected on days 1, 3 and 7 after

burn, were placed in transport media with thiogluconate. After

processing with tissue crusher, samples were incubated and

identified using conventional and automatic systems.

To verify exact burn depth, tissue biopsies were taken on

day 1 after burn and laid in 10% formaldehyde; histological

evaluation was undertaken to examine epidermal detach-

ment, oedema, homogenisation in collagen and necrosis.

Data were collected and processed using SPSS 14.0 for

Windows. Numerical values such as body weights, rectal

temperatures and serum levels were analysed with the

Kruskal–Wallis test and the Mann–Whitney U-test. The

changes of these values with time intervals within each group

were analysed with Friedman and Wilcoxon tests. Tissue

cultures results between groups were compared with Fisher’s

exact test.

Fig. 2 – Mean rectal temperatures of the groups. Error bars

represent S.E.M.; stars: p < 0.05 on comparison of CCAD

and PID groups on 3rd and 7th days.

3. Results

Rats in the ND group had a crusty covering on the burned area

from day 2 or 3 after burn. No sign of epithelialisation or

infection was noted in any case after 7 days. In the PID group,

eschar was preserved in all animals and signs of infection

were noted in 50% of subjects, particularly on day 7 post burn.

In the CCAD group, peripheral lysis of the eschar was noted on

day 3 post burn and on day 7 nearly all the rats showed signs of

infection.

Throughout the experiment, there was no significant diffe-

rence between initial and subsequent values for body weights

between the groups. Initial mean body weight for the ND group

was 233.9� 32.6 g and 17.6% loss was noted during the

experiment (p = 0.001). The PID group (mean initial weight

225.5 � 29.7 g) and the CCAD group (mean initial weight

218.7 � 19.9 g) lost means of 20.7% (p < 0.05) and 25.5%

(p = 0.001) of their body weight, respectively (Fig. 1).

In all groups rectal temperature tended to decrease at 3 h

after burn. When compared with the PID group, the CCAD

group had significantly higher rectal temperatures on days 3

and 7 (p < 0.05). Rectal temperatures in the PID group showed

no statistically significant change throughout the study (Fig. 2).

Changes in blood leukocyte counts within the groups and

comparisons between the groups were statistically insignif-

icant at p > 0.05 (Fig. 3). Only the CCAD group showed a

statistically significant change in CRP serum levels, which rose

after 24 h and were significantly different on day 7 after burn

when compared with the PID group (p < 0.05) as shown in

Fig. 4. Changes in serum IL-1b levels were not statistically

significant in or between any of the groups (p < 0.05), depicted

in Fig. 5. Only the CCAD group showed a statistically

significant change in IL-6 levels, which rose significantly after

24 h and achieved significant difference at days 3 and 7 post

burn when compared with the PID group (p < 0.05) as

illustrated in Fig. 6. The CCAD group also showed significantly

higher serum levels of TNF-a on day 7 post burn when

compared with the other two groups (p < 0.05), displayed in

Fig. 7. On days 1 and 3, only 1/9 and 4/9 of CCAD tissue cultures

were positive. On day 7, tissue cultures were positive in 1/7, 4/6

and 6/8 in groups ND, PID and CCAD, respectively. The CCAD

group had a tendency to higher rates of positive cultures but

the difference remained insignificant (Fig. 8).

4. Discussion

CCA has been used in partial-thickness burns since 1998 at our

institution. According to our unpublished data, inflammatory

reactions such as fever, leukocytosis or leukocytopenia and

increased CRP levels in sera are often observed in a significant

proportion of burn victims treated with CCA ointment,

although synchronous wound swab cultures showed no signs

of infection. This study was designed to evaluate systemic

inflammatory response to CCA treatment among rats with

second-degree scalds.

Fig. 7 – Median of TNF-a levels of the groups. Error bars

represent S.E.M.; star: p < 0.05 on comparison of CCAD–PID

and CCAD–ND groups.

Fig. 3 – Mean blood leukocyte values of the groups. Error

bars represent S.E.M.; WBC: white blood cell.Fig. 6 – Median of IL-6 levels of the groups. Error bars

represent S.E.M.; stars: p < 0.05 on comparison of CCAD

and PID groups.

Fig. 4 – Median of CRP levels of the groups. Error bars

represent S.E.M.; stars: p < 0.05 on comparison to initial

values for CCAD group; #: p < 0.05 on comparison of CCAD

and PID groups.

b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1938

There are previous reports on the inflammatory effects of

collagenase in the literature. Collagen treated with collage-

nase from bacterial origin is degraded into small peptides, and

the chemo-attractive effect of these cleavage products has

been reported by several groups studying various cells [16].

Postlethwaite and Kang demonstrated this effect on macro-

phages and precursor monocytes, which are key promoters in

the repair process [17]. The same effect was shown on human

fibroblasts and human neutrophils [18–20]. Recent experi-

ments indicated that collagenase itself might attract fibro-

blasts and granulocytes as well as keratinocytes [21,22].

Experiments by Abatangelo and co-workers indicated that

fibroblast proliferation was stimulated in a dose-dependent

fashion when treated with peptides resulting from the

cleavage of human collagen [23,24]. Similar observations

followed when digestion products from human burn eschar

Fig. 5 – Median of IL-1b levels of the groups. Error bars

represent S.E.M.

were used [25]. This experimental setting was found to be

closely related to the clinical setting.

Radice and colleagues subcutaneously inoculated small

amino acidic fragments, obtained from digestion of human

burn eschar tissue with bacterial collagenase, into both

human dermal fibroblast cultures and polyvinyl alcohol

sponges in rats. Cell cultures treated with eschar fragments

duplicated significantly faster than controls, and biochemical

and histological data from sponge implants showed that the

inflammatory response was augmented by eschar-derived

fragments at postoperative day 2 [26]. All these data may

address systemic effects of collagenase from bacterial origin.

Research findings indicate that clostridial collagenase may

expedite debridement not only by attacking native collagen,

Fig. 8 – Distribution of positive tissue cultures. Stars:

p = 0.08 on comparison of the CCAD–ND and CCAD–PID

groups on day 3; #: p = 0.06 on comparison of the CCAD–

ND groups on day 7.

b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1 939

but also by enhancing macrophage chemotaxis and activation

within the wound itself, thus acting to some extent as an

immunomodulating agent and wound conditioner.

Time intervals for sampling were based on previous studies

on inflammatory cytokine and acute phase protein response

after thermal injury [1,3,7,27–31]. Common time points for peak

levels of cytokines were selected and repeated samplings on the

same subject were minimised, which may be considered a weak

point of the study. However, this report presents valuable

information on effects and outcome of burn and CCA treatment

by evaluating the data of the same subjects at different time

intervals up to 7 days.

In numerous studies, burn was shown to cause weight loss

due to increased inflammatory cytokines and production of

catabolic hormones, insulin resistance, protein disintegra-

tion and reduced fat production [32–34]. Bacterial colonisa-

tion was another factor for weight loss in the rats [35].

Although the relationship between collagenase treatment

and weight loss was not clear, weight loss of the subjects in

our experiment seemed to be a consequence of the injury

itself rather than the possible effect of CCA or wound

infection, since the difference between the groups with or

without dressing was statistically insignificant. However, the

highest rate of weight loss was observed in the CCAD group,

which had the highest rate of infection; this might be due to

increased inflammatory response and prolonged effect of the

high serum TNF-a level.

Body temperature in all groups dropped significantly early

after burn, which may be related to the early hypometabolic

phase mentioned in the study of Kataronovski et al. [27]

Increased body temperature had been previously reported in

large burns [1,30]. Caldwell et al. showed a positive correlation

between increased serum IL-6 levels and body temperature in

rats. In our experiment, the CCAD group had a tendency to

rising rectal temperatures after day 3 post burn, with

significantly higher values when compared with the PID group

on days 3 and 7, which may reflect the systemic inflammatory

effect of collagenase.

Blood leukocyte count and function are known to increase

because of the stimulation of colony-growing factors in bone

marrow under the control of IL-6 and TNF-a in the first 12 h

post burn [1,36]. Although mean blood leukocyte counts did

not reveal any significant difference between groups through-

out the study, the CCAD group had a tendency to leukocyto-

penia which might contribute to septic complications as well

as depressed immunity. This scene is supported by our

previously mentioned clinical experience in which two thirds

of the cases with progressing inflammatory reaction devel-

oped leukocytopenia rather than leukocytosis; blood leuko-

cyte counts of these children dramatically improved with

change of type of dressing. With serum IL-6 and TNF-a levels

also increased in the CCAD group, an inflammatory effect of

CCA may be indicated, needing elucidation by further studies.

CRP is known to be one of the acute phase proteins to

increase in first 24 h after burn. As inflammatory equilibrium

is achieved, serum levels of CRP begin to decrease in 2–3 days

after severe burn. Increased CRP levels were reported in cases

with large burns or infection, due to the stimulating effects of

TNF-a and IL-6 [1]. In our study, serum CRP levels started to

increase after day 1 post burn and reached significantly higher

values only in the CCAD group, which also displayed high

serum IL-6 and TNF-a levels.

There are reports indicating that IL-1b is the first cytokine

to increase in sera following injury [7,27,29]. Cannon et al.

stated that serum IL-1b levels were at upper limits a few days

post burn [37]. However, in another study group which was

similar in age and gender, no significant difference was noted

[38]. Two years later, these varying results were thought to

depend on a natural inhibitor named IL-1 receptor antagonist

(IL-1 RA) by the same group [39]. Drost et al. showed significant

increase in serum levels of IL-1b of rats which were thermally

injured compared with controls, but the difference between

groups that were both burned and infected and groups that

were solely burned was insignificant [40]. In our study, the

changes in serum levels of all groups were also not statistically

significant. According to these data, IL-1b seems to be

unrelated to infectious complications. The lack of any

detectable peak in serum IL-1b levels after burn may be due

to inhibition of IL-1b before the sampling time.

After burn, serum TNF-a levels rise and then decrease

rapidly [28,29]. Lipopolysaccharide fragments found in the

wall structure of Gram-negative bacteria are known to be

among the most stimulating factors of TNF-a. In our study,

serum levels of TNF-a for all groups peaked at 3 h after burn

and the difference between groups was insignificant. After

burn day 3, TNF-a levels in the CCAD group were significantly

higher than in the PID and ND groups.

IL-6 is also an important mediator in the acute phase

response. There are previous reports indicating a relationship

between fever and IL-6, as one of the mechanisms in body

temperature regulation [41]. Pro-inflammatory effects such as

leukocytosis and induction of APPs are also reported, and a

positive correlation was detected between IL-6 and CRP levels

during inflammation [27]. Like the CRP and TNF-a levels in our

study, serum IL-6 levels among the CCAD group kept rising

and, from day 3 post burn, were significantly higher than in the

other two groups where no significant change was noted.

The statistical differences between the CCAD and ND

groups (e.g. p = 0.08 on day 3, p = 0.06 on day 7) may be

considered to indicate a higher tendency to infection or a pro-

inflammatory effect of collagenase. When compared with the

ND group (low infection rate and low serum cytokine levels)

and the PID group (higher infection rate but low serum

cytokine levels), the CCA group did exhibit a higher infection

rate and higher serum cytokine levels. However, aminoacids

with other biomatrix fragments provide a good medium for

bacterial growth, and over-exudation of the wound and risk of

bacterial colonisation limits CCA use in large burns [42]. In our

study, the difference between infection rates of the groups

was statistically insignificant. This may indicate an inflam-

matory effect of collagenase itself rather than infection.

5. Conclusion

On evaluation of all data provided, in our standardised study

group (exactly similar age, gender, body weight and burn

depth among the rats) the CCAD group had the highest weight

loss, rectal temperatures, tendency for leukocytopenia, serum

levels of CRP, IL-6 and TNF-a and rates of wound infection.

b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1940

Although rates of positive tissue cultures among the PID

groups were similar to the CCAD group, significant differences

in rectal temperatures and serum CRP, IL-6 and TNF-a levels

were noted between these two groups. This suggests that

CCA at least aggravates the systemic inflammatory response

among rats with partial-thickness burns.

Acknowledgement

This study was supported by Ege University Branch Office of

Research Projects (Project Number:05TIP011).

Conflict of interest

The authors are responsible for the reported research. We

have participated in the concept and design; analysis and

interpretation of data; drafting or revising of the manuscript,

and we have approved the manuscript as submitted.

We are disclosing any affiliation, financial agreement, or

other involvement of any of us with any company whose

product figures prominently in the submitted manuscript so

that the editors can discuss with the affected authors whether

to print this information and in what manner.

r e f e r e n c e s

[1] Dehne MG, Sablotzki A, Hoffmann A, Muhling J, Dietrich FE,Hempelmann G. Alterations of acute phase reaction andcytokine production in patients following severe burninjury. Burns 2002;28:535–42.

[2] Sparkes BG. Immunological responses to thermal injury.Burns 1997;23:106–13.

[3] Arturson G. Pathophysiology of the burn wound andpharmacological treatment. The Rudi Hermans lecture,1995. Burns 1996;22:255–74.

[4] Yamada Y, Endo S, Inada K. Plasma cytokine levels inpatients with severe burn injury with reference to therelationship between infection and prognosis. Burns1996;22:587–93.

[5] Endo S, Inada K, Yamada Y, Kasai T, Takakuwa T, Nakae H,et al. Plasma tumour necrosis factor alpha (TNF-a) levels inpatients with burns. Burns 1993;19:124–7.

[6] Ueyama M, Maruyama I, Osame M, Sawada Y. Markedincrease in plasma interleukin-6 in burn patients. J Lab ClinMed 1992;120:693–8.

[7] Drost AC, Burleson DG, Cioffi Jr WG, Mason Jr AD, Pruitt JrBA. Plasma cytokines after thermal injury and theirrelationship to infection. Ann Surg 1993;218:74–8.

[8] Gabay C, Kushner I. Acute-phase proteins and othersystemic responses to inflammation. N Engl J Med1999;340:448–54.

[9] Schindler R, Mancillia J, Endres S, Ghorbani R, Clark SC,Dinarello CA. Correlations and interactions in theproduction of interleukin-6 (IL-6) IL-1 and tumournecrosis factor (TNF) in human blood mononuclearcells: IL-6 suppresses IL-1 and TNF. Blood 1990;75:40–7.

[10] Bond MD, Van Wart HE. Characterization of the individualcollagenases from Clostridium hystoliticum. Biochemistry1984;23:3085–91.

[11] Soroff HS, Sasvary DH. Collagenase ointment andpolymixin B/bacitracin spray versus silver sulfadiazine. JBurn Care Rehab 1994;15:13–7.

[12] Hansbrough JF, Achauer B, Daeson J, Dawson J, Himel H,Luterman A, Slater H, et al. Wound healing in partialthickness burn wounds treated with collagenase ointmentversus silver sulfadiazine cream. J Burn Care Rehab1995;16:241–7.

[13] Jung W, Winter H. Considerations for the use of clostridialcollagenase in clinical practice. Clin Drug Invest1998;15:245–52.

[14] Ozcan C, Ergun O, Celik A, Corduk N, Ozok G. Enzymaticdebridement of burn wound with collagenase in childrenwith partial-thickness burns. Burns 2002;28:791–4.

[15] Till GO, Beauchamp C, Menapace D, Tourtellotte Jr W,Kunkel R, Johnson KJ, et al. Oxygen radical dependent lungdamage following thermal injury of rat skin. J Trauma1983;23:269–77.

[16] Hatz RA. Mechanisms of action of collagenase in woundrepair.. In: Altmeyer P, Abatangelo G, editors. Woundhealing and skin physiology. Berlin: Springer; 1995. p. 227.

[17] Postlethwaite AE, Kang AH. Collagen- and collagen peptide-induced chemotaxis of human blood monocytes. J Exp Med1976;143:1299–307.

[18] Postlethwaite AE, Seyer JM, Kang AH. Chemotacticattraction of human fibroblasts to type I, II, and III collagensand collagen-derived peptides. Proc Natl Acad Sci USA1978;75:871.

[19] Albini A, Adelmann-Grill BC. Collagenolytic cleavageproducts of collagen type I as chemoattractants for humandermal fibroblasts. Eur J Cell Biol 1985;36:104.

[20] Walter RJ. Clostridial collagenase. A chemoattractant forhuman neutrophils. Inflammation 1986;10:347.

[21] Ranzati C, Zahn W, Thom H. Bacterial collagenase andcollagen breakdown products exert chemotactic effects invitro: implication for wound debridement and woundhealing [abstract]. In: Fourth annual meeting of theEuropean Tissue Repair Society; Aug 25–28, 1994.

[22] Herman IM. Stimulation of human keratinocyte migrationand proliferation in vitro: insights into the cellularresponses to injury and wound healing. Wounds 1996;8:33.

[23] Cortivo R, Radice M, Brun P, Abatangelo G. Biologicalactivity of human collagen breakdown products onfibroblasts. Wounds 1995;7:38A.

[24] Radice M, Cardarelli L, Cortivo R, et al. Chemotacticproperties of human collagen breakdown products inwound healing. In: Abatangelo G, Donati L, Vanscheidt W,editors. Proteolysis in wound repair. Berlin, Heidelberg,New York: Springer; 1996. pp. 51–61.

[25] Radice M, Cardarelli L, Cortivo R, Abatangelo G. Enzymaticescharolysis as possible source of local chemotactic factors[abstract]. In: Second joint meeting of the Wound HealingSociety and the European Tissue Repair Society; 1996.p. 15–9.

[26] Radice M, Brun P, Bernardi D, Fontana C, Cortivo R,Abatangelo G. Clostridial collagenase releases bioactivefragments from extracellular matrix molecules. J Burn CareRehab 1999;20:282–91.

[27] Kataronovski M, Magic Z, Pejnovic N. Early inflammatorycytokine and acute phase protein response under the stressof thermal injury in rats. Physiol Res 1999;48:473–82.

[28] Fang CW, Yao YM, Zhai HX, Yu Y, Wu Y, Lu LR, et al. Tissuelipopolysaccharide-binding protein expression in ratsafter thermal injury: potential role of TNF-a. Burns 2004;30:225–31.

[29] Schwacha MG. Macrophages and post-burn immunedysfunction. Burns 2003;29:1–14.

[30] Caldwell FT, Graves DB, Wallace BH. Pathogenesis of feverin a rat burn model: the role of cytokines andlipopolysaccharide. J Burn Care Rehab 1997;18:525–30.

b u r n s 3 4 ( 2 0 0 8 ) 9 3 5 – 9 4 1 941

[31] Deveci M, Eski M, Sengezer M, Kisa U. Effects of ceriumnitrate bathing and prompt burn wound excision on IL-6and TNF-a levels in burned rats. Burns 2000;26:41–5.

[32] Meyer NA, Barrow RE, Herndon DN. Combined insulin-likegrowth factor-1 and growth hormone improves weight lossand wound healing in burned rats. J Trauma 1996;41:1008–12.

[33] Wolfe RR. Relation of metabolic studies to clinicalnutrition—the example of burn injury. Am J Clin Nutr1996;64:800–8.

[34] Yasuhara S, Kaneki M, Sugita H, Sugita M, Asai A, Sahani N,et al. Adipocyte apoptosis after burn injury is associatedwith altered fat metabolism. J Burn Care Res 2006;27:367–76.

[35] Steinstraesser L, Burkhard O, Fan MH, Jacobsen F,Lehnhardt M, Su G, et al. Burn wounds infected withPseudomonas aerations trigger weight loss in rats. BMCSurg 2005;17:19.

[36] Peter FW, Schuschke DA, Barker JH, Fleishcher-Peter B,Pierangeli S, Vogt PM, et al. The effect of severe burn injuryon proinflammatory cytokines and leukocyte behaviour: itsmodulation with granulocyte colony-stimulating factor.Burns 1999;25:477–86.

[37] Cannon JG, Friedberg JS, Gelfand JA, Tompkins RG,Burke JF, Dinarello CA. Circulating interleukin-1 betaand tumor necrosis factor alpha concentrations afterburn injury in humans. Crit Care Med 1992;20:1414–9.

[38] Wogensen L, Jensen M, Svensson P. Pancreatic beta-cellfunction and interleukin-1 beta in plasma during the acutephase response in patients with major burn injuries. EurClin Invest 1993;23:311–9.

[39] Mandrup-Poulsen T, Wogensen L, Jensen M, Svensson P,Nilsson P, Emdal T, et al. Circulating interleukin-1receptor antagonist concentrations are increased inadult patients with thermal injury. Crit Care Med 1995;23:26–33.

[40] Drost AC, Larsen B, Aulick LH. The effects of thermal injuryon serum interleukin 1 activity in rats. LymphokineCytokine Res 1993;12:181–5.

[41] Caldwell FT, Graves DB, Wallace BH. The effect ofindomethacin on the cytokine cascade and bodytemperature following burn injury in rats. Burns1999;25:283–94.

[42] Klassen HJ. A review on the nonoperative removal ofnecrotic tissue from burn wounds. Burns 2000;26:207–22.