SKRIFTLIG RAPPORT
Läkarprogrammet, självständigt arbete (30 hp)
Reaction after the use of xenograft treatment after burn injury in children– infection or inflammation?
Author: Linnea Gerhardsson, medical student, Uppsala medical school, Uppsala University.
Burn Center, Department of Plastic and Maxillofacial Surgery, Uppsala University Hospital, Uppsala, Sweden.
Supervisor: Fredrik Huss, MD, Associate Professor. Department of Surgical Sciences, Plastic Surgery, Uppsala University and the Burn Center, Department of Plastic and Maxillofacial Surgery, Uppsala University Hospital, Uppsala, Sweden
2018-01-08
2 Index
Popular science summary in Swedish ... 3
Word list/Definitions ... 4
Abstract ... 5
Background ... 6
Treatment ... 6
Xenografts ... 7
Inflammation ... 8
Material and methods ... 10
Statistics ... 10
Data extracted ... 11
Antibiotics ... 11
Ethical approval ... 11
Results ... 12
Study population ... 12
Microbial cultures ... 13
Fever ... 14
Antibiotic treatment ... 16
Case examples ... 19
Discussion ... 20
Study population and data registrations ... 20
Readmission ... 21
Hours from injury ... 21
Time frames ... 21
Microbial cultures ... 22
Fever ... 22
Antibiotic treatment ... 23
Case examples ... 23
Conclusions ... 24
Future research ... 24
Acknowledgements ... 25
References ... 25
3
Popular science summary in Swedish
Populärvetenskaplig sammanfattning
I brännskadebehandling används ofta biologiska förband (djur/människa) för att tillfälligt täcka sår med något som liknar frisk hud. Grishud är det vanligaste alternativet, och ingår i standardbehandlingen vid Brännskadecentrum i Uppsala. Biologiska förband kan orsaka en inflammatorisk reaktion, vilket även brännskadan som sådan gör. Vi har sett att en del barn som behandlats med grishud efter ett par dagar får hög feber och allmänpåverkan. Detta kan bero både på en sårinfektion, vilket inte är ovanligt och uppträder inom samma tidsram, men också på den inflammatoriska reaktionen från skadan eller grishuden. Dessa barn läggs ofta in och får antibiotika i blodet (iv), vilket är korrekt om infektion är skälet till symtomen. Vi har dock också sett att symtomen ibland går över utan antibiotika. Vi ville därför studera hur många patienter som behandlats med grishud som får feber efter behandlingen och hur det korrelerar med bakterieodlingar och inflammatoriska biomarkörer (CRP), för att se om det är en inflammatorisk eller infektiös reaktion. Förhoppningen är att kunna skilja ut de som behöver antibiotika, för att minska onödig antibiotikaanvändning, och få en minskad selektion av antibiotikaresistenta bakterier. I denna studie analyserade vi data från 57 brännskadepatienter som behandlats med grishud på Brännskadecentrum, Akademiska sjukhuset, de senaste åren.
Resultaten visar att 73 % av patienterna får feber efter behandlingen. Av dessa behandlas 46 % med antibiotika. Vi påvisade skillnad i temperatur och CRP mellan de som fick iv antibiotika och de som inte fick det. Detta tyder på att antibiotikabehandlingen till patienter som antas ha en infektiös orsak till temperaturstegringen är korrekt. Andra delar av studien, som utveckling av kroppstemperatur över tid och bakterieodlingar, tyder dock på att det kan finnas en inflammatorisk komponent som orsak till symtomen. Vidare studier krävs för att kunna avgöra om detta är en infektiös eller inflammatorisk reaktion.
4
Word list/Definitions
(partly adopted from https://medical-dictionary.thefreedictionary.com/biologic+dressing, accessed 171226)
Allograft “A graft transplanted between genetically nonidentical individuals of the same species.
Autograft “Tissue or organ transferred into a new position in the body of the same person.”
Biological dressing
“A dressing used in treatment of a burn or other large denuded area of skin to eg. prevent infection and fluid loss”
CRP C-Reactive Protein
Fever Body temperature >38° C
Haxa Hours after xenograft application
iv. Intravenous
Lyophilization Freeze-drying
Xenograft “A graft transferred from an animal of one species to one of another species”
5
Abstract
Porcine xenograft is a part of routine treatment for burns. It is known that transplantation of non-species- specific tissue can cause an inflammatory reaction within days. We have noticed that many children treated with porcine xenograft present with fever 2-4 days post-treatment, and are often subsequently treated with iv antibiotics on suspicion of wound infection and/or septicemia. Infections in burns are common and symptomatic within the same time frame. This study analyzed how many patients treated with porcine xenografts that presented with fever post-treatment, and how this correlated with CRP-levels and microbial cultures, to determine whether this symptomology was due to an infectious or inflammatory reaction. The results show that 73% of the patients had fever after xenografting. Of the patients with fever, 46% was given antibiotic treatment. The time frame when patients present with fever, median 12-18 hours after xenografting, as well as the prolonged fever duration, suggests the fever is of inflammatory origin. However, the difference in both temperature and CRP between the patients with iv.
antibiotics and those without, indicates that patients receiving iv antibiotics have an infectious origin of the fever. Further studies are required to distinguish between infection or inflammation, as cause to the fever.
6
Background
Burns
Skin burns are injuries arising due to the temperature in the skin exceeding 42-43° C. This occurs when the skin is exposed to heat from e.g. fire, hot liquids, or hot surfaces. Often also injuries caused by exposure to cold, friction, radiation, or chemicals are sorted under “burns”
even though the pathophysiology and treatment can differ1.
Burns are classified according to their depth into 1) superficial (1st degree, involving the epidermis only); 2) superficial dermal (2nd degree, involving the epidermis and superficial parts of the dermis); 3) deep dermal (2nd degree, involving the epidermis and deeper parts of the dermis); and 4) full thickness (3rd degree, where all skin layers are destroyed, and the injury may extend into the subcutaneous tissue involving bones, muscle, tendons, or neurovascular structures). Sometimes the superficial dermal and deep dermal burns are grouped under the term partial thickness burns. Scalds can be of all depths stated above1-5.
Burn injuries are one of the injuries causing the most suffering with more than 180 000 deaths yearly worldwide, most commonly in low and middle income countries6-9. Burns can also cause severe physical disability and pain, as well as psychological trauma. Burn injuries are especially prevalent in children, being the 11th leading cause of death for children between one and nine years of age9-11. This is partly due to young children’s urge to explore their surroundings though being less aware of the risks involved with fire and hot liquids/surfaces, and partly due to their slower withdrawal reflexes and thinner skin12.
Yearly, approximately 0.4% of Scandinavians are treated for burn wounds, most being superficial and deep dermal burns. Children, 0-4 years old, is the largest group admitted to hospital constituting 27% of all Swedish patients admitted to hospital for burn wounds. Males dominate in all age groups and constitute 69% of burns treated in hospital13,14. Scalds are the most prevalent burn in patients ≤ 5 years of age, accounting for over 65% of the cases in that age group15.
Treatment
Burn injuries are one of the most expensive injuries to treat, with a total health care cost ranging from approximately 704 to 717,306 USD, median 44,024 USD, per burn patient treated in hospital in high income countries16. Most scalds are treated conservatively, as most are of superficial dermal depth that heal on their own within two to three weeks with good wound care. However, the gold standard for treating deep dermal and full thickness burns is an early tangential excision of damaged tissue and autografting with split-thickness skin grafts5,17. The immediate treatment starts with cooling of the burnt area as soon as possible after injury.
The rational is to lower the temperature of the skin to <43° C to stop the pathophysiologic process. Cooling is also analgesic and reduces wound edema18. However, to avoid hypothermia and further tissue damage this treatment should not be too extensive and tempered water should be used to cool only the burn, not the patient4,19. If the superficial burn is adequately managed, and free from complications such as dehydration or infection, it will heal spontaneously within three weeks’ time4,20. Wound toilet and proper dressings with adequate dressing changes is most often all treatment that is required, sometimes with the addition of a topical antimicrobial treatment5,17,21.
7
The ideal dressings for burns should create an optimal micro-environment for wound healing allowing epithelial cell migration, be nonantigenic, antibacterial, and analgesic22. They should also permit wound respiration, maintain moisture level, and temperature. Their use should be without social stigma and have no possibility of microbial transmission. Preferably the dressings should also be readily available, have a long shelf life and be inexpensive19. Besides protecting the wound the dressing should also be durable and easily applicable19,22. The human allografts (ie skin from deceased donors; fresh, lyophilized, glycerolized, etc) holds several of the characteristics mentioned above19,23.
Xenografts
Due to the high cost and limited supply of human allografts, alternative dressings have been developed24. Xenografts have been used since the 17th century, and grafts from an array of different animals have been tried25,26. The porcine xenografts, introduced in the 1960s, are now the most commonly used aside from allografts, due to its multitude of benefits27. Porcine xenografts are thus a part of the routine treatment of partial thickness burns and are also used as protective dressings after excision of full thickness burns before applying the autograft28. We could find no studies supporting the theory of graft rejection when using porcine xenografts in burn treatment, but as the dermis re-epithelializes underneath the graft gradually sloughs off29. Porcine xenografts fulfill most criteria of an optimal burn dressing. Besides having the same positive characteristics as human allografts, they are also less expensive, more readily available, and more easily applied. They adhere well to the surface of the wound and both give an antimicrobial protection as well as a barrier function. The wound of the patient keeps its humidity and temperature as well as protein and electrolyte losses are minimized22,30. There is no direct manipulation of the wound surface proper when changing the protective dressings over the xenograft, making the dressing-change procedure less painful. Applying xenograft is also known to be analgesic as it covers exposed nerve endings, making it suitable in the care of burnt children as it considerably reduces the amount of sedatives and analgesics needed for wound care22,31. It has also been observed that porcine xenografts reduce the local inflammatory response in wounds as well as systemically, normalizing temperature and other vital signs compared with other dressings, though it should be stated that xenografts have no antibiotic properties29,32. Other studies have shown porcine xenografts to stimulate growth factors necessary for collagen synthesis as well as stem cell proliferation and differentiation33.
Porcine xenografts have a long shelf life due to the multitude of preservation methods possible.
The most common preservation (by glycerol) makes the graft nonviable by chemically dehydrating it, and also reduces the grafts antigenicity and microbial contamination (glycerol is antibacterial and antiviral) as well as allowing refrigerated storage. If refrigerated, for shorter time periods, the graft can also be used fresh and potentially viable30. Other methods of preservation are treatment with gamma radiation, ethylene oxide, hydrogen peroxide plasma, aldehyde cross-linking, and lyophilization. Preserving the graft in these ways may though reduce its adherence to the wound bed and thereby its efficacy but several studies have shown this does not affect the outcome34-37.
There are several different porcine grafts available on the market, and EZ Derm® (Molnlycke Health Care, US, LLC, Norcross, GA) is the one used in our clinic. It has a reduced antigenicity due to its preservation process with collagen being cross linked with an aldehyde group30,34.
8
It was evaluated in 2013 by examining the frequency of different complications. In this study it was concluded that EZ Derm® is a cost-effective xenograft with a minimal complication frequency, including incomplete epithelialization, hypertrophic scarring, and infections32. The common practice, in hospital, of treating superficial dermal burns conservatively with xenograft, is to sedate the patient to clean the burn wound thoroughly. Xenograft is then applied to the wound bed and may be fixed with sutures, staples, tape, or the like. A variety of dressings can then be applied over the xenograft. We prefer to use one layer of Vaseline gauze followed by dry compresses and an elastic wrap or pre-fabricated garments. A few hours postoperatively the patient is free to ambulate ad lib and in many cases, can be discharged. One to four days postoperatively the outer dressings are removed, and the adherence of the xenograft is inspected. If the xenograft is well attached to the wound the surface is cleaned with soap and water, and new Vaseline gauze and outer dressings are applied. Two to three times per week the outer dressings are changed until the xenograft spontaneously sloughs off (within approximately 2 weeks). If the xenograft has not attached properly by the first dressing change this indicates either that the graft was put on undamaged skin, deeper burn, or an infection has occurred. The xenograft is then removed wherever it has not attached. The wound is cleaned and alternative dressing is applied17.
Inflammation
It is known that transplantation of tissue from one species to another causes an inflammatory reaction and even though porcine xenografts are nontoxic the product specification on the EZ Derm® states that mainly children can present with a hyperthermic reaction after xenograft application38,39. Bacterial colonization and infection in burn wounds are also a common occurrence and often symptomatic two to four days after the initial burn40,41. Sepsis is the leading cause of death among burn patients and the most common bacteria causing septicemia is Staphylococcus aureus42,43.
Inflammatory processes are always induced at time of burn, and the levels of inflammatory markers have been shown to correlate with the severity of the burn and the subsequent clinical outcome for the patient. If the trauma is severe the levels of these markers initially decrease due to plasma leakage, starting to rise after 48 hours and sometimes staying elevated for months44-
47. Elevated levels of CRP are common in children with burn wounds 6 to 24 hours after the burn, as well as significantly elevated levels of cytokines correlated to elevated levels of CRP48. In our burn center we have clinically observed that children treated with porcine xenografts often present with a hyperthermic reaction two to four days after xenograft application, and on the suspicion of septicemia caused by a bacterial infection in the burn wound, they are subsequently treated with intravenous antibiotics49. Little research has been conducted on children who have received xenografts after burn injuries and we therefore wanted to investigate this hyperthermic reaction further as it is not clear whether this is an infectious or inflammatory reaction. If found that the hyperthermic reaction is due to an inflammatory reaction this discovery could lead to a more targeted aftercare, and thus more cost-effective, with less suffering for the patients. Furthermore, (iv) antibiotics would not be necessary to treat an inflammatory reaction, and thus antibiotic use could be reduced with a possible subsequent reduction in the selection of multi resistant bacteria in both patients and milieu.
9
Thus, we analyzed;
How many of the patients treated with xenografts after burn injury that present with a hyperthermic reaction post-treatment, and when it appears?
How many of these patients received antibiotic treatment, and what kind?
How does the hyperthermic reaction correlate with microbial cultures and inflammatory biomarkers such as CRP?
10
Material and methods
This is a retrospective clinical evaluation generating new hypothesis for further research. The aim was to include data from all patients (n=121) treated with the EZ Derm® xenograft during the years 2010-2017 at the Burn Center, Department of Plastic and Maxillofacial Surgery, Uppsala University Hospital. The years were divided among several persons to extract data from medical records according to a pre-made spreadsheet. Due to limited time available clinical data was extracted from 62 burn patients treated during the periods January 1-December 31 for the years 2017, 2014, 2012, 2011, July 1-Decemer 31 2016, and one patient treated in 2015. The number of patients per periods as above was 14, 29, 4, 2, 12, and 1 respectively.
There were initially no exclusion criteria, but as this study focuses on hyperthermic reactions after xenograft treatment, correlating to microbial cultures and CRP, five patients were excluded due to having none of these parameters registered, leaving in total 57 patients. In the subsequent data analyze patients lacking the current parameter were excluded from that specific analysis. Patients having a fever prior to the xenograft application were not excluded, as there was a possibility this could be caused by a bacterial infection in the burn wound before xenograft application. Detected bacteria were counted only once per patient regardless of times testing positive for that same bacterium, as microbial cultures are often taken repeatedly from the same sites.
Statistics
The data extracted was enteredinto Microsoft® Excel® (2016 MSO (16.0.8730.2046) 32-bit.
Redmond, USA) for further analysis. Statistical calculations were performed using Microsoft® Excel® and a p <0.05 was considered statistically significant. For normally distributed data one- tailed Student’s t-test was used for temperature differences at 0-24 haxa and 60-84 haxa. Two- tailed Student’s t-test was used to compare the groups ± (fever + iv antibiotics +CRP >100), as well as body temperature in the groups ± iv antibiotics.
The time frame for the CRP elevation was set to 96 hours as it takes up to 24 hours for the plasma levels to rise, hence the added 24 hours to the 48-72 hours spectrum (for infection to show). All patients with extensive burn injuries have an elevated CRP level, due to the inflammatory reaction caused by the injury. Therefore, the CRP level was set to 100 to indicate a bacterial infection. Furthermore, a CRP < 100 in a burn patient is seldom clinically relevant.
11
Time between xenografting and fever was calculated in intervals using the formula:
Mean x´= (∑fx)/y
Median = L + (n /2-c) /FxW
L = the number between the middle and next lower interval, n=total number of data points,
c=accumulated number of data points below the mid-interval, F=number of data points in mid-interval,
W=the width of the interval.
Data extracted
Gender
Age at time of burn Local hospital
Date and time of burn
Date and time of admission to the burn center Length of stay at the burn center
Location of burn Burn depth
Date and time of xenograft application
Microbial cultures; date, time, type, and location Growth in microbial culture samples
Temperature at times documented. Maximum once per hour Plasma- C Reactive Protein
Type of antibiotic treatment
Antibiotics
Route of administration of antibiotic treatment was not registered in the medical records.
Therefore, the division into iv and oral treatment was based on the antibiotic substance used.
Flukloxacillin is not possible to administer iv. Imipenem, Piperacillin/Tazobactam, and Cefotaxim, are administered iv-only. Klindamycin was the only substance possible to administer both orally and iv.
Ethical approval
N/A for this student’s report of clinical evaluation.
12
Results
Study population
The patients included in this study are described in Table 1. The majority (67%) of the patients were boys under the age of four (38/57 patients). Only one patient (2%) was over the age of four. Superficial dermal burns were the most common injuries (n=42, 74 %) among the 57 patients that had a documented burn depth.
Table 1. Distribution of age and burn depth among boys (n=39) and girls (n=18).
Age
Superficial dermal, Boys/Girls
Deep dermal, Boys/Girls
Superficial dermal + deep dermal, Boys/Girls
Unknown depth Boys/Girls 0-11
months 9 (82%) / 2 (18%) 1 (100%) / 0 1 (100%) / 0 1 (50%) / 1 (50%) 1-4
years 20 (67%) / 10 (33%) 1 (100%) / 0 3 (43%) / 4 (57%) 2 (67%) / 1 (33%) 5-12
years 1 (100%) / 0
Total 30 (71%) / 12 (29%) 2 (100%) / 0 4 (50%) / 4 (50%) 3 (60%) / 2 (40%)
Burns to the thorax and arm/hand were the most common injuries, representing 74% and 70%
respectively (Figure 1). Most patients had injuries at multiple locations making the total number of injury sites 143.
Figure 1. Distribution of burns among the 57 patients with 143 injury-sites.
13
Microbial cultures
As demonstrated in Figure 2, among 57 patients analyzed, Staphylococcus aureus was the most commonly detected bacteria representing 17/48 (35%) of positive cultures in this study population. The second most common finding was Enterobacter species, consisting of 5/48 (10%) positive cultures. Other common bacteria were Methicillin Resistant Staphylococcus Aureus (MRSA) and Bacillus species, constituting 9% of the positive cultures each. Coagulase Negative Staphylococcus (CNS), S. pneumoniae, P. Aeruginosa, Group C Streptococcus (GCS), Klebsiella species and Enterococcus species, were each found in 2/48 (4%) of the positive cultures. Extended Spectrum Beta-lactamase class A (ESBL class A), E. coli, Acinetobacter, E. faecalis and Group B Streptococcus (GBS) were each found in 1/48 (2%) of the positive cultures. Several patients had samples taken multiple times from the same site.
Figure 2. Distribution of positive cultures sampled after xenografting among the studied patients (n=48). Each bacterium is registered only once per patient regardless of number of positive cultures. S. aureus was the most commonly detected bacteria, Enterobacter species the second. Other findings were MRSA, Bacillus species, CNS, S. pneumoniae, P. aeruginosa, GCS, Klebsiella species, Enterococcus species, ESBL class A, E. coli, Acinetobacter, E. faecalis and GBS.
14
The majority, 50/104 (48%), of bacterial cultures after xenografting were sampled from wound exudate. Three blood cultures were found, only one being positive (ie < 1%). The second most common location was “unknown origin”, consisting of 16/104 (15%) of cultures. Other common locations were urine, throat, nasopharynx, and the perineum (Figure 3).
Figure 3. Distribution of locations of microbial cultures sampled after xenograft application (n=104).
Fever
At any point post-grafting, 41 out of 56 patients (73%) presented with fever (Figure 4).
Figure 4. Frequency of body temperature >38° C (fever) at any time point post-grafting (n=56).
15
There was a wide range in hours from xenograft application to first registered fever episode (1-151) hours. Both mean and median temperatures though were found within 72 haxa (Table 2).
Table 2. Hours from xenograft treatment to hyperthermic reaction. (n=40)
We could find no statistical significant difference in body temperature between the two time intervals (p = 0,6). The highest mean temperature < 24, and 60- 84 haxa was 38.5 °C and 38.7 °C, respectively. Data shown in Table 3.
Table 3. The highest temperatures registered <24 and 60- 84 haxa (n=32).
Highest temperature
registered 0-24 haxa 60-84 haxa
Mean 38.5 °C 38.7 °C
Median 38.2 °C 38.8 °C
Range 36.5-40.5 °C 36.5-40.5 °C
Data shows that 36/41 (88%) and 27/41 (66%) of the patients had a hyperthermic reaction
< 72 haxa and > 72 haxa respectively (Figure 5). Only 4/41 (10%) had the first temperature
>38° C registered >72 haxa.
Figure 5. Number and percent of patients having fever at <72 and, >72 haxa, and fever debute >72haxa (n=41).
Hours (h) to first registered temperature above 38°C (n=40)
<6 h ≥6<12h ≥12<18 h ≥18<24 h ≥24<30 h ≥30<36 h 36-42 h ≥42<48 h ≥48<54 h ≥54<78 h ≥78<84 h ≥84<148 h ≥148<154 h
Number of patients 8 5 10 5 2 2 2 2 2 0 1 0 1
% of patients 20% 12.5% 25% 12.5% 5% 5% 5% 5% 5% 0% 2.5% 0% 2.5%
Mean 24 h
Median 12-18 h
Range 0-151 h
16
Table 4. Differences in body temperature between the groups ± iv antibiotic treatment (n=53.)
Temperature
With iv. antibiotic treatment
Without iv. antibiotic treatment
Mean 40 °C 38.5 °C
Median 40.1 °C 38.3 °C
Range 38.1-41.0 °C 36.1-40.6 °C
The two tailed Student’s t -test showed a statistically significant difference in body temperature between the groups ± iv antibiotics (p <0.001).
Antibiotic treatment
After xenografting, 33/57 (58%) patients had a positive microbial culture registered at any time.
Out of those 33, 17 (52%) received antibiotic treatment (ie 30% of all patients). Among the patients without registered positive microbial culture post-xenografting 2/24 (8%) received antibiotic treatment (ie 4% of all patients). In total 19 out of 57 (33%) received antibiotic treatment (Figure 6).
Figure 6 Showing number (and %) of patients receiving antibiotic treatment (n=57) ± having positive microbial cultures (n=17).
17
The variety of antibiotics that the 19 patients received is shown in Figure 7. Since some patients had their treatment changed a total of 26 antibiotic treatments were administered. Cefotaxim was the most commonly used antibiotic treatment, constituting 70 % of the iv treatments, and 54% of the antibiotic treatments in total.
Figure 7 Distribution of antibiotics administered to the study population (number of antibiotic treatments in total n=26, n=19)
Forty six percent (19/41) of patients with fever received antibiotic treatment after xenograft application. No antibiotic treatment was given to 49% of these patients. In 5% of the patients with fever no data whether antibiotics was given or not could be found. The routes of administration were; 7% oral treatment only, 29% iv treatment only, and 10% iv treatment followed by oral treatment. In total, 16/19 (84%) of patients receiving antibiotic treatment, received iv treatment at some point (Figure 8).
Figure 8. Number (and %) of patients with a body temperature > 38° C receiving antibiotic treatment (n=41.)
18
Out of the 19 patients who at some point had fever and any antibiotic treatment, 11 (61%) also had a CRP ≥ 100. The two tailed Student’s t-test showed a statistically significant difference in CRP between the patients receiving and not receiving iv antibiotic treatment (p < 0.001) (Figure 9, Table 5).
Figure 9. Distribution of patients with fever receiving antibiotics having CRP < and ≥ 100 (n=19).
Table 5. CRP levels in patients with or without receiving iv antibiotic treatment (n=34).
CRP
With iv antibiotic treatment
Without iv antibiotic treatment
Mean 142 48
Median 129 34
Range 13-302 1-187
Out of the 15 patients without fever, 9 had no CRP levels registered. The remaining 6 patients had a mean CRP-level of 8.6 (median 3.7, range 0.36-35).
Table 6. CRP-levels in patients not having fever (n=6).
CRP in patients
without fever(n=6) CRP
Mean 8.6
Median 3.7
Range 0.36-35
19
Table 7. Signs and symptoms in 23 patients within and after 72 haxa..
Signs and symptoms
(n=23) Within 72 haxa After 72 haxa
Fever 19 (83%) 18 (78%)
Positive microbial
culture 12 (52%) 10 (43%)
iv. antibiotics 3 (13%) 10 (43%)
CRP≥100 9 (39%) 6 (26%)
This analysis includes the 23 patients who had body temperature, CRP-levels, and microbial cultures registered. All other patients were excluded. There were 17/19 (89%) patients having fever both before and after 72 haxa. One patient had the first registered fever episode > 72 haxa.
All patients receiving iv antibiotic treatment within 72 haxa still had fever > 72 haxa.
After 72 haxa, 4/23 (17%) patients showed a combination of fever, positive microbial culture, and CRP ≥ 100. In 75% of these, the combination presented > 72 haxa. In one patient (25%) the combination presented within 72 haxa. All these patients received iv antibiotic treatment.
Case examples
The only patient in this study having a positive microbial blood culture, had it registered within 72 haxa and was positive for S.aureus. This patient also had fever, and CRP ≥ 100 within 72 haxa. This patient received iv antibiotic treatment > 72 haxa but continued having fever, even though the CRP-level went < 100 > 72 haxa.
The patient with a negative blood culture had samples from wound exudate positive for S.
aureus, both before and after 72 haxa, CRP <100, but fever both before and after 72 haxa. This patient received iv antibiotic treatment > 72 haxa
20
Discussion
In this retrospective clinical evaluation of 57 children treated with the EZ Derm® xenograft after burn injury, 73% presented with fever at some point after xenografting, whereof 88% within the first 72 haxa. However, this study could not conclude whether this was due to an infectious or inflammatory reaction. In total 33% received antibiotic treatment. In patients with fever, 46%
received antibiotics. In the group of 23 patients who had body temperature, CRP-levels and microbial cultures registered, 58% of the patients had fever within 72 haxa and a positive microbial culture sampled within 72 haxa. The combination of having a fever, with a CRP ≥ 100 within 96 haxa was found in 47% of these patients.
Study population and data registrations
As expected, the majority, 67%, of the patients were boys under the age of four50. Only one patient, 2%, was older than four. Superficial dermal injuries were the most common in our study population. We found that injuries to the thorax and arm/hand are the most common injuries, 74% and 70% respectively. As these data are consistent with previous research we believe this is a representative study population despite the large exclusion15,51.
As we did not have any information of length of stay in hospital, readmission, or follow ups, five patients were excluded due to lacking registrations of body temperature, microbial cultures and CRP, as they could have been discharged and readmitted either to our department, or to their local hospital for further treatment. If we had included these five patients as patients having neither a fever, nor an elevated CRP, the results would have been quite different.
According to the standard protocol at the Burn center, parameters are supposed to be registered at least daily, as long as the patient is treated in hospital17. As this is a retrospective study on data not initially intended for research, two major issues presented themselves, the first issue being that the registration of the parameters is not systematic. If the temperature is not systematically registered on certain time intervals, “hours to fever” can be very inaccurate. The patient may have a hyperthermic reaction immediately after xenograft application, but as there may be several hours until first registration of temperature, or sometimes days, this reaction may go unnoticed until much later. Temperatures are also not likely to be accurate, as the patients are most likely given some sort of antipyretic treatment when having a fever.
The second major issue in this study was the homogenous study population. There is a tendency to examine patients (and record data) who are at a more critical clinical state more extensively than patients who are well. -For example, when analyzing how body temperature changes over time, this analysis only included patients having temperatures registered at these time intervals, being the patients still staying hospital at these hours. Patients not having any complications are most likely discharged and therefore not included. As we could only analyze the parameters being registered, patients who have very few parameters registered, and patients who are discharged, are not represented as well as the patients who are more severely ill. Most patients had parameters registered without any gaps in time though, indicating that they have not been
21
discharged. We do presume that patients who are discharged are doing well, but this could not be confirmed. There was a group of 23 patients who had body temperature, CRP-levels and microbial cultures registered, all parameters ideally wanted for this study. When comparing their results to the results in the larger study population, one must remember that these 23 patients most likely are the 23 patients who are the most unwell out of the 57 patients we have analyzed.
Readmission
We were interested in examining how many patients who were readmitted on the suspicion of septicemia after being discharged, compared with how many patients who had no hospital contact before the planned follow-up. We were also interested in if the patients transferred within the hospital, needing intensive care treatment at some point. Unfortunately, we had no access to information on readmissions, in-hospital transfers, or follow-ups. It is not likely that any of the patients included in this study had been admitted to another ward at Uppsala University Hospital for treatment of septicemia without us knowing. It however may be possible that they could have been admitted to their local hospital without us knowing.
Hours from injury
When attempting to analyze hours from injury to first fever episode, it was discovered that the registered time of injury could not be correct in a minimum of 12% of the patients, as 7/57 patients were registered as being admitted to the burn center before the registered time of injury.
Due to secrecy and the limited time available, it was not possible to revisit the medical records for control. Therefore, this perspective has not been analyzed further in this study. There has though been no reason to question the registered time of xenograft application, which was focus of this study.
Time frames
One-tailed Student’s t-test was performed comparing the mean highest temperature registered within 24 haxa with the mean highest temperature 60- 84 haxa. We wanted to compare the two time frames where the hyperthermic reaction most likely would be due either to infection or to inflammation. As there are almost no previous studies on children receiving the EZ Derm® xenograft after burn injury, it was difficult to choose a time frame for the possible inflammatory reaction. The only source, the EZ Derm® product information, does not give a more specific time frame than “post treatment”. The time interval 0-24 haxa was chosen as there has been previous research on elevated inflammatory biomarkers within the first 24 haxa. As we wanted to possibly detect an infection we chose the other time interval to be 60-84 haxa as infections usually are symptomatic within 48-96 hours after the initial burn. Though the data on time of burn was questionable, we concluded this would be a fairly adequate approximation.
22
Microbial cultures
When suspecting an infection originating from an open wound, collecting microbial cultures from wound exudate is a part of the standard protocol, but if suspecting septicemia microbial cultures should also be collected from blood before the administration of iv antibiotic treatment17,49. This aims at the possibility of narrowing the spectrum of the antibiotic treatment after determining the pathogen to optimize the clinical response as well as avoiding the selection of antibiotic resistant bacteria.
This study however shows, that though it is a part of the standard protocol, microbial cultures are not routinely collected from blood before treating with iv. antibiotics in this patient group.
In this study only one patient receiving antibiotic treatment had a microbial culture collected from blood to confirm the pathogen. Many positive microbial cultures were collected from patients not receiving antibiotic treatment as a part of the routine screening for MRSA and VRE (Vancomycin Resistant Enterococcus). Therefore, they are not likely to be collected on the suspicion of septicaemia, making the percentage of microbial cultures taken when suspecting septicaemia even less common.
As S. aureus is the most commonly detected bacteria (35%), it is not surprising that when detected in combination with fever there is a tendency to prescribe iv. antibiotics, as S. aureus is the most common pathogen in wound infections and can possibly cause fatal septicaemia42,49,52. It could be possible that if having a microbial culture positive for S. aureus previous to the patient presenting with fever, there is a tendency not to go through with further testing.
Cultures collected from wound exudate are problematic to interpret as all wounds are colonized within 72 hours after injury, often including pathogens potentially causing septicemia such as S. aureus. Therefore, it is impossible to conclude anything from these microbial cultures, if complimentary microbial cultures are not taken from other sites, preferably blood. No antibiotic prophylaxis is given at the time of burn trauma or xenograft application, making the microbial cultures diagnostically accurate. However, a positive microbial culture collected from wound exudate only confirms that the wound is colonised and a possible site of infection. It cannot conclude that the patient has septicaemia, as most patients are colonised without suffering from septicaemia.
Fever
As there are no previous studies on children receiving EZ Derm® xenograft after burn injury, which is the focus of this study, it is not possible to compare our results to any other study.
Many of the analyses on temperature however suggest that the hyperthermic reaction we saw in 73% of the patients is caused by an inflammatory reaction. One of these findings is the fact that 88% of the patients presenting with fever do so within 72 haxa, median 12-18 hours, well before a bacterial infection would be symptomatic. However, one can argue that the wound cleaning procedure, could elicit a microbial seeding from the wound to the blood stream. Fever presented within 72 haxa remains in 56% of the patients. This prolonged fever duration also suggests this is an inflammatory reaction, mean highest temperature within 24 haxa compared to 60-84 haxa being 38.5° C, and 38.7° C, respectively. In only 10% of the cases, fever was registered for the first time after 72 haxa. This is however impossible to draw any conclusions
23
from, since we have not been able to analyze time from injury to first registered fever episode, making the time frame for a possible wound infection prior to xenograft application very inexact. The range in time to first registered fever episode is also quite wide, 0-151 haxa, in many cases overlapping the time frame when an infectious origin of the hyperthermic reaction is more likely.
Antibiotic treatment
Route of administration of antibiotic treatment was not registered in the medical journals.
Therefore, the division into iv and oral treatment was based on the antibiotic substance used.
Flukloxacillin is not possible to administer iv. Imipenem, Piperacillin/Tazobactam and Cefotaxim, are administered iv only. Klindamycin was the only substance possible to be administered both orally and iv. As we aimed to study how many patients who received antibiotic treatment on the suspicion of septicemia we assumed that Klindamycin was administered iv, as the patients receiving this treatment were admitted to hospital, and Flukloxacillin is the first line oral treatment for wound infections in outpatients. There may have been some patients given Klindamycin due to allergy to some other antibiotic substance, but the number in this study was deemed minor. The oral treatment was noted as “not receiving intravenous antibiotics”, as oral treatment most likely was not prescribed on the suspicion of septicemia. The fact that the type of antibiotic treatment these patients received was not certain makes this analysis somewhat inaccurate. However, the antibacterial spectrum is the same regardless route of administration.
Since all pharmaceuticals, including antibiotics, can lead to adverse events/reactions, some even potentially lethal (eg toxic epidermal necrolysis), it is important to treat only treatable situations (infections) with antibiotics. Furthermore, unnecessary use of antibiotic drugs facilitates the selection of antibiotic resistant bacteria. Endangering not only the patient treated but also the whole milieu and thus all patients in need of antibiotic treatment53,54. When suspecting septicemia in pediatric burn patients, broad spectrum antibiotics such as Tienam, Piperacillin/Tazobactam, Cefotaxim and Klindamycin are often drugs of first choice55,56. To avoid selection of resistant bacteria as narrow spectrum antibiotics as possible should be used.
Thus, it is important to confirm that an infection is present and what microbe is most likely causing it by adequate microbial sampling eg from blood57,58.
Some findings indicate that the patients receiving iv antibiotic treatment have an infectious origin of the hyperthermic reaction. There was a statistically significant difference in both temperature as well as in CRP between the patients receiving and not receiving iv antibiotics.
Case examples
There were two patients of special interest within the group having all parameters registered,
“A” having a positive blood culture, and “B” having a negative blood culture. Each patient may represent a hyperthermic reaction due to infection, and inflammation respectively. Both patients had a hyperthermic reaction within the same time frame, 72 haxa, and fever remaining thereafter. Both patients also received iv antibiotics after 72 haxa. Interesting though, both patients were positive for S. aureus in their wound exudate. Patient “A”, S. aureus positive in
24
blood culture, had a CRP ≥ 100 and most likely a S. aureus infection originating from the burn wound. Patient “B” was S. aureus positive in wound exudate (but negative in blood culture), both before and after 72 haxa. This indicates that “B” may represent a patient having a hyperthermic reaction due to inflammation, as this patient had no elevated CRP and was only colonized with S. aureus.
Conclusions
We found that the time frame when patients present with fever, as well as the prolonged fever duration despite antibiotic treatment, suggests the hyperthermic reaction we see in our patients to be of inflammatory origin. However, the difference in both temperature as well as in CRP between the patients receiving, and not receiving, iv. antibiotics, indicate that the patients receiving iv antibiotics have an infectious origin of the hyperthermic reaction. It was shown that the standard protocol when suspecting septicemia is not followed regarding how to collect the microbial cultures. Therefore, it is not possible to accurately determine the pathogen responsible, evaluate the choice of antibiotic treatment, or to conclude if the patients suffered from septicemia or not. Further studies are needed to determine whether this hyperthermic reaction is an infectious or inflammatory reaction.
Future research
This retrospective clinical evaluation analyses many aspects of the outcome after xenograft treatment, leading to a broad spectrum of suggestions for further research. We would like to continue this research by doing a prospective randomized controlled study. Temperature and inflammatory biomarkers (CRP), as well as all microbial routine screenings should be systematically registered on all patients regardless of clinical status. Microbial cultures should be collected from multiple sites when registering a temperature over a set cutoff limit. One arm would randomize patients to “common practice”, treating patients as of today. In this arm each doctor would decide when antibiotic treatment should be administered. The other arm would have stricter criteria on the administration of antibiotic treatment, completely avoiding prescription If possible. Ideally, patients could continue to register the parameters studied at home after discharge, staying in the study. This study had no access to information on readmission, in hospital transfer or follow up. This could be also examined further in the future.
25
Acknowledgements
Special thanks to Karin Engblom Nurse assistant, Kristina Karlsson Nurse assistant, Marie Lindblad RN, ICU- and research nurse, for providing the data analyzed in this study.
References
1. Tintinalli, Judith E. Emergency Medicine: A comprehensive study guide. 7th edition. New York: McGraw-Hill Companies; 1374–1386 p. (Emergency Medicine; vol. 2010).
2. H.Cleland. Thermal burns: assessment and acute management in the general practice setting. Fam Physician. 2012(41):372–375.
3. Classification of Burns - Health Encyclopedia - University of Rochester Medical Center [Internet]. [cited October 25th 2017].
https://www.urmc.rochester.edu/encyclopedia/content.aspx?ContentTypeID=90&Content ID=P09575
4. Deitch EA, Wheelahan TM, Rose MP, Clothier J, Cotter J. Hypertrophic burn scars:
analysis of variables. J Trauma. 1983;23(10):895–8.
5. Papini R. Management of burn injuries of various depths. BMJ. 2004;329(7458):158–60.
6. Drago DA. Kitchen scalds and thermal burns in children five years and younger.
Pediatrics. 2005;115(1):10–6.
7. Alnababtah K, Khan S. Socio-demographic factors which significantly relate to the prediction of burns severity in children. Int J Burns Trauma. 2017;7(5):56–63.
8. Delgado J, Ramírez-Cardich ME, Gilman RH, Lavarello R, Dahodwala N, Bazán A, et.
al. Risk factors for burns in children: crowding, poverty, and poor maternal education. Inj Prev 2002;8(1):38–41.
9. Snapshot [Internet]. [cited December 08 2017].
http://www.who.int/mediacentre/factsheets/fs365/en/
10. 9789241563574_eng.pdf [Internet]. [cited October 13 2017].
http://apps.who.int/iris/bitstream/10665/43851/1/9789241563574_eng.pdf
11. Hutchings H, Barnes PM, Maddocks A, Lyons R, James-Ellison MY. Burns in young children: a retrospective matched cohort study of health and developmental outcomes.
Child Care Health Dev. 2010;36(6):787–94.
12. Argirova M, Hadzhiyski O. Treatment of palm burns in children. Ann Burns Fire Disasters. 2005;18(4):190–3.
13. Nieminen S, Laaksonen V, Viljanto J, Pakkanen A. Burn injuries in Finland. Scand J Plast Reconstr Surg. 1977;11(1):63–7.
26
14. Svee A, Jonsson A, Sjöberg F, Huss F. Burns in Sweden: temporal trends from 1987 to 2010. Ann Burns Fire Disasters. 2016;29(2):85–9.
15. Barrow RE, Spies M, Barrow LN, Herndon DN. Influence of demographics and inhalation injury on burn mortality in children. J Burns 2004;30(1):72–7.
16. Hop MJ, Polinder S, van der Vlies CH, Middelkoop E, van Baar ME. Costs of burn care:
a systematic review. Wound Repair Regen. 2014;22(4):436–50.
17. Brannskadekompendium.pdf [Internet]. [cited December 26 2017].
http://www.akademiska.se/Global/Neuro/Plastikkirurgi%20och%20k%C3%A4kkirurgi/D okument/Br%C3%A4nnskadecentrum/Brannskadekompendium.pdf
18. Wasiak J, Cleland H. Burns: dressings. BMJ Clin Evid [Internet]. [cited October 24 2017];2015. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4501337/
19. Sheridan RL, Tompkins RG. Skin substitutes in burns. J Burns 1999;25(2):97–103.
20. Smahel J. The problem of dehydration and healing of burn wounds. J Burns 1993;19(6):511–2.
21. Hayward PG MW. Current concepts in wound dressings. Aust Prescr. 1996(19):11–3.
22. Chiu T, Burd A. ”Xenograft” dressing in the treatment of burns. Clin Dermatol.
2005;23(4):419–23.
23. Austin RE, Merchant N, Shahrokhi SE, Jeschke MG. A comparison of BiobraneTM and cadaveric allograft for temporizing the acute burn wound: cost and procedural time. J Burns 2015;41(4):749–53.
24. Brusselaers N, Pirayesh A, Hoeksema H, Richters CD, Verbelen J, Beele H, et.al. Skin Replacement in Burn Wounds: J Trauma 2010;68(2):490–501.
25. Peters WJMD. Biological Dressings in Burns-A Review. Ann Plast Surg. 1980;4(2):133–
7.
26. Switzer WE, Moncrief JA, Mills W, Order SE, Lindberg RB. The use of canine heterografts in the therapy of thermal injury. J Trauma.1966;6(3):391–8.
27. P. Solovey, O. Kyryk, V. Barchuk. Xenografts—liophilized pig skin as a burn wound cover. Burn. 33, 2007:S85–6.
28. Still J, Donker K, Law E, Thiruvaiyaru D. A program to decrease hospital stay in acute burn patients. J Burns1997;30:23(6):498–500.
29. Ersek RA, Denton DR. Silver-impregnated porcine xenografts for treatment of meshed autografts. Ann Plast Surg. 1984;13(6):482–7.
30. Chiu T, Pang P, Ying SY, Burd A. Porcine skin: friend or foe? J Burns 2004;30(7):739–
41.
27
31. Burkey B, Davis W, Glat PM. Porcine xenograft treatment of superficial partial-thickness burns in paediatric patients. J Wound Care. 2016;25(2):S10-15.
32. Troy J, Karlnoski R, Downes K, Brown KS, Cruse CW, Smith DJ, et.al. The Use of EZ Derm® in Partial-Thickness Burns: An institutional review of 157 patients. Eplasty.
2013;13:e14.
33. Chen X, Shi Y, Shu B, Xie X, Yang R, Zhang L, et.al. The effect of porcine ADM to improve the burn wound healing. Int J Clin Exp Pathol. 2013;6(11):2280–91.
34. Pruitt BA. The evolutionary development of biologic dressings and skin substitutes. J Burn Care Rehabil. 1997;18(1 Pt 2):S2-5.
35. Rappaport I, Pepino AT, Dietrick W. Early use of xenografts as a biologic dressing in burn trauma. Am J Surg. 1970;120(2):144–8.
36. Elliott RA, Hoehn JG. Use of commercial porcine skin for wound dressings. Plast Reconstr Surg. 1973;52(4):401–5.
37. Tavis MJ, Thornton JW, Harney JH, Woodroof EA, Bartlett RH. Graft adherence to de- epithelialized surfaces: a comparative study. Ann Surg. 1976;184(5):594–600.
38. Yang Y-G, Sykes M. Xenotransplantation: current status and a perspective on the future.
Nat Rev Immunol. 2007;7(7):519–31.
39. Hajská M, Dragúňová J, Koller J. Cytotoxicity testing of burn wound dressings: first results. Cell Tissue Bank. 2017;18(2):143–51.
40. Mir MA, Khurram MF, Khan AH. What should be the antibiotic prescription protocol for burn patients admitted in the department of burns, plastic and reconstructive surgery. Int Wound J. 2017;14(1):194–7.
41. Sårinfektioner - Behandling och diagnos - Infektionsguiden [Internet]. [cited October 13 2017]. http://www.infektionsguiden.se/vardpersonal/infektioner/hud-
ochmjukdelsinfektioner/sarinfektioner/
42. Greenhalgh DG. Sepsis in the burn patient: a different problem than sepsis in the general population. Burns Trauma [Internet]. December 2017 [cited December 08 2017];5(1).
http://burnstrauma.biomedcentral.com/articles/10.1186/s41038-017-0089-5
43. de Oliveira THC, Amorin AT, Rezende IS, Barbosa MS, Martins HB, Brito AKP, et.al.
Sepsis Induced by Staphylococcus aureus: Participation of biomarkers in a murine model.
Med Sci Monit 2015;21:345–55.
44. Barrett M. The clinical value of acute phase reactant measurements in thermal injury.
Scand J Plast Reconstr Surg Hand Surg. 1987;21(3):293–5.
45. Oldham KT, Guice KS, Till GO, Ward PA. Activation of complement by hydroxyl radical in thermal injury. Surgery. 1988;104(2):272–9.
28
46. Faymonville ME, Micheels J, Bodson L, Jacquemin D, Lamy M, Adam J, et.al.
Biochemical investigations after burning injury: complement system, protease- antiprotease balance and acute-phase reactants. J Burns 1987;13(1):26–33.
47. Friedl HP, Till GO, Trentz O, Ward PA. Roles of histamine, complement and xanthine oxidase in thermal injury of skin. Am J Pathol. 1989;135(1):203–17.
48. Sikora JP, Kuzański W, Andrzejewska E. Soluble cytokine receptors sTNFR I and sTNFR II, receptor antagonist IL-1ra, and anti-inflammatory cytokines IL-10 and IL-13 in the pathogenesis of systemic inflammatory response syndrome in the course of burns in children. Med Sci Monit 2009;15(1):CR26-31.
49. Svår_sepsis_och_septisk_chock_dec_2013_new.pdf [Internet]. [cited October 12 2017].
http://www.infektion.net/sites/default/files/Sv%C3%A5r_sepsis_och_septisk_chock_dec_
2013_new.pdf
50. Akerlund E, Huss FRM, Sjöberg F. Burns in Sweden: an analysis of 24,538 cases during the period 1987-2004. J Burns 2007;33(1):31–6.
51. Biscegli TS, Benati LD, Faria RS, Boeira TR, Cid FB, Gonsaga RAT. Profile of children and adolescents admitted to a Burn Care Unit in the countryside of the state of São Paulo.
Rev Paul Pediatr 2014;32(3):177–82.
52. Infekterade sår 2008.pdf [Internet]. [cited December 31 2017].
https://www2.sahlgrenska.se/upload/SU/omrade_oss/hud/Infekterade%20s%C3%A5r%20 2008.pdf
53. Piperacillin/Tazobactam Sandoz - FASS Vårdpersonal [Internet]. [cited January 04 2018].
https://www.fass.se/LIF/product?userType=0&nplId=20060622000011 54. Läkemedelsreaktioner och hud [Internet]. [cited January 04 2018].
http://www.internetmedicin.se/page.aspx?id=988
55. Neveus T, Ågren J. Terapikompendium Pediatrik. 2014. Uppsala;
56. Neonatal sepsis och meningit [Internet]. [cited January 04 2018].
http://www.internetmedicin.se/page.aspx?id=1724
57. Resistenta bakterier - Infektionsguiden.se [Internet]. [cited Januari 02 2018].
http://www.infektionsguiden.se/om-antibiotika/resistenta-bakterier/
58. Antibiotika - 1177 Vårdguiden - sjukdomar, undersökningar, hitta vård, e-tjänster [Internet]. [cited January 02 2018]. https://www.1177.se/Fakta-och-rad/Rad-om- lakemedel/Antibiotika/
