Laser Doppler imaging in a paediatric burns population

Laser Doppler imaging in a paediatric burns population

Julie Mill a,*, Leila Cuttle a, Damien G. Harkin b, Olena Kravchuk c, Roy M. Kimble a

aRoyal Children’s Hospital Burns Research Group, University of Queensland, Department of Paediatrics and Child Health,

Royal Children’s Hospital, Herston Rd, Herston, Queensland 4029, Australiab School of Life Sciences and Institute of Health and Biomedical Innovation, Queensland University of Technology,

Kelvin Grove, Queensland 4059, AustraliacSchool of Land, Crop and Food Science, University of Queensland, St Lucia, Queensland 4072, Australia

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

a r t i c l e i n f o

Article history:

Accepted 26 November 2008

Keywords:

Laser Doppler

Paediatric

24 h

Fast scan

Grafting

Scar management

Re-epithelialisation

a b s t r a c t

Objective: Laser Doppler imaging (LDI) was compared to wound outcomes in children’s

burns, to determine if the technology could be used to predict these outcomes.

Methods: Forty-eight patients with a total of 85 burns were included in the study. Patient

median age was 4 years 10 months and scans were taken 0–186 h post-burn using the fast,

low-resolution setting on the Moor LDI2 laser Doppler imager. Wounds were managed by

standard practice, without taking into account the scan results. Time until complete re-

epithelialisation and whether or not grafting and scar management were required were

recorded for each wound. If wounds were treated with SilvazineTM or ActicoatTM prior to the

scan, this was also recorded.

Results: The predominant colour of the scan was found to be significantly related to the re-

epithelialisation, grafting and scar management outcomes and could be used to predict

those outcomes. The prior use of ActicoatTM did not affect the scan relationship to out-

comes, however, the use of SilvazineTM did complicate the relationship for light blue and

green scanned partial thickness wounds. Scans taken within the 24-h window after-burn

also appeared to be accurate predictors of wound outcome.

Conclusion: Laser Doppler imaging is accurate and effective in a paediatric population with a

low-resolution fast-scan.

# 2008 Elsevier Ltd and ISBI. All rights reserved.

avai lable at www.sc iencedi rec t .com

journal homepage: www.elsevier.com/locate/burns

1. Introduction

Traditional ways for assessing acute burns and subsequent

scarring rely largely on qualitative methods such as total burn

surface area calculation via the Lund and Browder chart

estimation [1] and the Vancouver General Hospital Scar Scale

[2]. While such techniques have been useful, they are unable to

guide us on the clinical status of a burn when decisions about

most appropriate treatments are being made.

Laser Doppler technology originally existed as flowmetry.

Reports at that time found that laser Doppler flowmetry could

* Corresponding author. Tel.: +61 7 3636 9067; fax: +61 7 3365 5455.E-mail address: [email protected] (J. Mill).

0305-4179/$36.00 # 2008 Elsevier Ltd and ISBI. All rights reserved.doi:10.1016/j.burns.2008.11.016

be used as a clinical predictor tool with burns displaying the

lowest perfusion values on days 0–3 requiring skin grafting or

taking longer than 21 days to re-epithelialise [3,4]. However, as

a small probe was required to be in contact with the burn in

order to take the measurements, this technique was con-

sidered painful and too localised and now laser Doppler

imaging (LDI) which can scan the whole burn area in a non-

contact manner is more popular.

LDI produces a colour-coded image of skin blood perfusion.

A low-intensity red laser light beam penetrates the full dermis

and is reflected by both moving red blood cells and the static

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1 825

tissue. The instrument then assigns perfusion units (PU) to

quantify the movement of blood cells flowing through the

vessels of the skin and the perfusion units are translated into

colours which are transposed on a greyscale digital image of

the scanned area. This provides a valuable indicator of burn

depth and thus severity.

Histological biopsy assessment has long been considered

to be the gold standard of burn depth assessment. LDI has

been shown to correlate well with histopathological markers

of wound depth [5–8], however, this has not been validated in

a paediatric population. Also, biopsies may be inaccurate in

the first few days as the burn wound progresses and more

tissue dies, or recovers. The optimum time that laser Doppler

scanning should be performed post-burn is still a contro-

versial topic. Laser Doppler is widely reported to give

accurate results when performed between 48 and 72 h

[7,9]. Holland et al. 2002 reported that scans at 24 h may

be inaccurate as burned areas which initially appear deep

can later improve. This was based on scans shown in Niazi’s

landmark paper in 1993 [5]. However, another study showed

that laser Doppler flowmetry performed within 24 h post-

burn gave similar flux values to scans taken three days after-

burn [10], indicating that perhaps scans within the 24-h

window may still give accurate readings. This is a theory

supported by others [11] who have also done serial

measurements using LDI.

For this study, we were interested to see if LDI might

provide a reliable indicator (via colour of scan) of burn depth

and therefore outcome when used as early as 0–24 h and

continuing up to 120 h after-burn. We also sought to

determine if a standard resolution (fast) laser Doppler scan

for the paediatric population was accurate. The effect of the

prior application of nanocrystalline silver dressing ActicoatTM

(Smith and Nephew, Hull, UK) and SilvazineTM cream (1%

silver sulphadiazine with 0.2% chlorhexidine digluconate,

Smith and Nephew, Clayton, Australia) to the scan was also

examined.

2. Materials and methods

2.1. Patients

The human research ethics committee of the Royal Children’s

Hospital (RCH) approved the study and all patients had

consent obtained prior any procedures. Subjects were children

who presented to the Stuart Pegg Paediatric Burns Centre,

whose parents consented or the patient assented to the study.

Patients were recruited at any time from 0 to 190 h after burn

injury. Both outpatients and inpatients were enrolled in the

study.

2.2. Laser Doppler scanning

The scanning was done using the MoorLDI-2 laser Doppler

imager (Moor Instruments Limited, Devon, UK), with a laser

wavelength of 633 nm (visible red), scan range (distance from

the wounds) of 35–80 cm, fast scan resolution of 256 � 64 pix-

pixels, and a scan duration of 80 s. The scanner head was

aimed slightly off perpendicular to the skin, to prevent

reflection and non-valid areas, as recommended by the

product manual (p54-55), rather than directly perpendicular

to the skin as suggested by others [12]. Software used was v1.5.

Using the standard colour palette setting XE, the amount of

blood flow (expressed as perfusion units in the range 0–1000)

was translated into specific colours: dark blue (<125 PU), light

blue (125–250 PU), green (250–440 PU), yellow (440–628 PU),

pink (628–812.5 PU), and red (812.5–1000 PU).

LDI was performed with a green drape placed under the

patient, using the standard (fast) scan. The laser Doppler

scan was performed either in the designated laser Doppler

room, burns clinic bathroom or operating theatre. All rooms

are maintained at �25–318 C. Patients wore the standard

goggles supplied by Moor during the scan (if age-appro-

priate), and the door was closed during scanning. For

patients who were non-compliant with the goggles (toddlers

and babies), a green drape was clipped to the scanner head

and the scanner head was lowered into position so that

child was unable to see the scanner red point light, thus

shielding their eyes completely during the scan. Each body

part that was injured was scanned separately and in some

cases lateral and medial surfaces were scanned for

circumferential burns, as suggested by others [9]. Patients

were scanned each time they returned to the outpatients

department, if this fell within the first 200 h.

Use of digital photography to document the wounds was

also utilised as an adjunct to the laser scan, as the colour

photograph representation on a laser Doppler image is

traditionally poor quality compared to a digital photograph

[9]. A digital photograph was taken at 3.0 megapixels with a

Fujifilm E510 Finepix 5.2 Megapixel camera with the flash on,

using the normal automatic setting.

2.3. Patient management

Patients were given standard pain relief used in the burns

outpatient department, most commonly oxycodone and

paracetomol (acetaminophen) orally and for inpatients most

commonly intravenous morphine. Patients who showed signs

of uncontrolled pain during dressing down procedures and

who would benefit from immediate redressing were not

scanned and a digital image only was taken. Age-appropriate

toys were used for distraction to enable to child to relax and

keep still for the scan.

If patients presented out of hours, they were clinically

assessed by the burns registrar without the use of laser

Doppler, a digital photograph was taken and dressings were

applied after discussion with the burns consultant. Routine

dressing care for the RCH Stuart Pegg Paediatric Burns Centre

is to gently cleanse the wound area with 0.1% chlorhexidine

digluconate mixed in warm water, removing any creams, or

dead skin that has lifted.

SilvazineTM cream was the mainstay of treatment for the

RCH Burns Centre until November 2001. Daily cleaning was

required due to the necessity to reapply the cream. The advent

of nanocrystalline silver dressings such as ActicoatTM has

changed the face of burns treatment in our centre. Dressings

are applied and left intact for up to 4 days for ActicoatTM and 7

days for Acticoat7TM [13,14]. The centre also pricks blisters and

expresses the fluid, leaving the blister skin intact until it

Table 1 – The total number of burns included in thestudy, some were on the same patient.

Burn sites Number of patients

1 27

2 10

3 4

4 3

Table 2 – The time post-burn of the laser Doppler scans.

Scan time post-injury (h) Number of wounds

0–24 13

25–48 22

49–72 15

73–120 19

>120 2

Unable to scan 11

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1826

further denudes at later dressing changes [15]. ActicoatTM is

applied using a simple water irrigation system [13,14]. Patients

were then instructed, if managed as outpatients, to return to

the burns outpatient clinic for dressing every 3–7 days.

Patients were recruited immediately into the study if they

presented during business hours and less than 200 h post-

burn. If they presented out of hours, they were recruited if they

returned to the centre for further treatment during this time

frame. A note of dressings such as SilvazineTM and ActicoatTM

used prior to scanning was taken.

The RCH Burns Centre includes five paediatric burn

surgeons. The general consensus amongst our surgeons is

that if a wound takes longer than 2–3 weeks to heal, it will

require a skin graft. It is well recognised that prediction of

burn wound outcome remains especially difficult in chil-

dren, due in part to the prevalence of mixed depth scald

burns, children’s thin skin and their unpredictable response

to injury [12]. If a burn wound appeared to be deep and could

be predicted by the surgeons to take more than 2–3 weeks to

heal, it was grafted as soon as possible. Indeterminate depth

burns were monitored over approximately a 2-week period

to determine if they needed grafting. The patients were

reviewed by the burns consultant and the laser Doppler

image was shown to the medical staff for their interest only,

however, the usual protocol for grafting remained regard-

less of the laser Doppler image. All patients who take longer

than 2–3 weeks to heal and all those who undergo skin

grafting are treated with active scar management, which

includes custom measured pressure garments and silicone

products.

2.4. Database

Patient and injury details were recorded prospectively. The

data included: date and time of burn, mechanism of injury,

provision and nature of dressing treatments, description of

treatments prior to scanning, and total burn surface area

(TBSA). The laser Doppler scan was taken and reviewed by the

first author. The laser Doppler scan was categorised by the

worst scan colour. The outcome of the burn wound was

recorded in days taken to completely re-epithelialise or time

until skin graft and whether or not the wound required

grafting, and/or scar management.

2.5. Statistical analysis

All the healing outcomes and diagnostics data were

analysed using the Minitab Software Package (Minitab,

Release 15, Minitab Inc., Chicago 2006). The associations

between the laser Doppler diagnostics and the wound

healing outcomes were modelled with logistic multiple

regressions (binary or nominal regressions with the logit

link function), in which various patient characteristics such

as age, mechanism of burns, the time to scan and others

were included. In cases where some categories were over- or

under-presented so that the execution of logistic regres-

sions was not possible, the chi-square and exact Fisher’s

test were conducted instead. The significance level was set

at 5% for individual effects; the p-values of the tests are

additionally reported in the text.

3. Results

3.1. Patient and subjects scanning data

There were 48 patients enrolled in the study, with a total of 82

separate burns. Many patients had multiple burn sites as

described in Table 1. The initial laser Doppler scan occurred at

various time points, from 0 to 186 h after-burn. Table 2 details

the number of burns scanned during each timeframe. A large

percentage of burns (42%) were only able to be scanned outside

the recommended 24–72 h timeframe after-burn, and some

burns (11 of 82) were unable to be scanned (13%).

3.2. Patient burn data

The patients ranged in age from 9 months to 14.6 years, with

the median age of 3 years 6 months and the mean age 5 years 5

months. The spread of mechanisms of burn was typical to that

seen in our total burn population, with 41.5% scalds, 22.0%

contact burns, 28.1% flame burns, 6.1% hot oil burns and 2.4%

friction burns. As expected, the colour of the laser scan was

found not to be associated with the patient age, TBSA, burn

mechanism or the time of the scan after-burn. There were a

total of 11 burns (from 9 patients) which were unable to be

scanned. These burns were not significantly different from the

scanned burns in terms of the healing time, grafting/scar

management outcomes, TBSA or age of the patients.

3.3. Grafted burns

There was a significant relationship between the colour of the

scan and whether the wounds were grafted (p < 0.001),

indicating that the colour could be used as a good predictor

for grafting. Fig. 1 shows the predominant scanned colour for

the grafted and non-grafted wounds. All five dark blue

scanned wounds were grafted, none of the 20 green scanned

wounds were grafted and only one of the yellow/pink/red

scanned wounds were grafted. This yellow/pink/red wound

was judged by the consultant to be deep and grafting was

performed as a precautionary measure as the patient lived

Fig. 1 – The predominant colour scanned and perfusion

units (PU) for the wounds and whether or not they were

grafted. All dark blue scanned wounds required grafting,

while no green scanned wounds were grafted. Light blue

scanned wounds were predominantly not grafted and one

yellow/pink/red wound was grafted erroneously.

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1 827

remotely. However, a biopsy of the tissue removed during the

grafting procedure found that the tissue was viable and the

surgeon’s prediction of deep burn depth was wrong. For the

light blue scanned wounds, the decision not to graft was

slightly more prevalent (58%). The light blue wounds that were

not grafted were predominantly contact and flame burns. The

light blue grafted burns were predominantly scald burns, with

only one out of 8 contact burns and 3 out of 10 flame burns

grafted. This mechanism effect was significant (p = 0.015),

demonstrating that prior knowledge of burn mechanism does

affect clinical judgement of the requirement for grafting,

especially for indeterminate depth burns.

3.4. Scar management

The need for active scar management was also found to be

significantly associated with the colour of the laser scan

(p = 0.003), indicating that the laser scan colour can be used for

predicting the need for scar management. The scan colour of

the wounds treated with scar management is shown in Fig. 2.

All dark blue wounds received scar management, whereas

only 50% of green wounds also received this treatment. Patient

Fig. 2 – The predominant colour scanned for the wounds

and whether or not they received active scar management.

All dark blue scanned wounds required scar management,

while 87% of light blue scanned wounds, 50% of green

wounds and 20% of yellow/pink/red wounds received scar

management.

age and TBSA were found to be significant factors influencing

scar management in the green group (p = 0.004), with younger

patients with greater burn TBSA more likely in need of scar

management. For the light blue wounds, the mechanism

significantly effected the requirement for scar management

(p = 0.038), in particular the majority of flame burns required

scar management (although they did not require grafting).

3.5. Effect of dressings on the scan

Prior to scanning, most wounds (72 of 82) received treatment

with SilvazineTM or ActicoatTM at either the primary referral

centre or in the RCH. Of these, 45.1% received SilvazineTM,

57.3% received ActicoatTM and 14.6% received both.

Statistical analysis showed that although the prior use of

ActicoatTM did not affect the relationship between scan colour

and expected outcome, the prior use of SilvazineTM may have

affected this relationship. If wounds that scanned green or

light blue were treated with SilvazineTM prior to the scan, they

tended to not need grafting (for light blue) or scar management

(for green), whereas wounds pre-treated with ActicoatTM had

significant trends for requiring grafting and scar management.

This may indicate that pre-treatment with SilvazineTM makes

the wound scan at a deeper depth than it really is, and it may

heal slightly better than expected from the scan colour. Or it

may indicate that prior use of SilvazineTM alters the clinical

appearance of the wound and clinicians tend to underestimate

the need for grafting or scar management, although the LDI

may indicate they are required.

SilvazineTM is not used for burn treatment in this hospital,

however, almost all burns are treated with ActicoatTM. Here,

ActicoatTM gave no observed detrimental effect on scanning

quality, despite the nanocrystalline silver deposition from this

dressing. An example of how ActicoatTM treatment did not

interfere with laser scanning can be seen in Fig. 3. This patient

was scanned at 0 h and then treated with ActicoatTM. At 0 h,

some dark blue areas can be seen. He was scanned again at

200 h post-burn, after two ActicoatTM dressing changes and

the dark blue area was still present and more demarcated. At

200 h, there is no evidence of silver deposition into the wound

and no apparent decrease in efficacy of the scan. At 22 days

this dark blue area was grafted, with good results.

3.6. Re-epithelialisation

For re-epithelialisation results, the grafted wounds were

excluded as the true time for complete re-epithelisation could

not be determined for these wounds. Importantly, there was a

significant relationship between time taken for re-epithelia-

lisation and scar management, with wounds that healed

slowly requiring scar management more often (p < 0.001).

There was also found to be a significant relationship between

the scan colour and the time taken for re-epithelialisation

(p < 0.003) in wounds both pre-treated with, and without

Silvazine. Silvazine treatment appeared to make the light blue

wound outcome less conclusive with three out of 15 light blue

wounds in that group, however, this effect was not found to be

significant (p > 0.10). Similarly, although light blue scanned

wounds that were pre-treated with Silvazine had a mean re-

epithelialisation of 19.3 days, compared to wounds that did not

Fig. 3 – Laser Doppler scanning of a hot tea scald treated with ActicoatTM. (A) The scan and picture of the burn taken at 0 h

post-burn. The patient had no SilvazineTM prior to this scan. He was dressed with ActicoatTM from this time. (B) The scan

and picture of the burn taken at 200 h, after 2 subsequent changes of ActicoatTM dressing. Note that there is no evidence of

silver deposition onto the wound and no apparent decrease in the efficacy of the scan after ActicoatTM has been used. (C)

The outcome of the burn at 50 days. The patient was grafted at 22 days, in the areas only where the laser Doppler scan

showed dark blue perfusion units. The wound outcome could be accurately predicted at the time of the 0 h scan; however,

the usual course of 2–3 weeks of dressing treatment was followed before grafting.

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1828

have Silvazine applied prior to scanning, with an average of

22.7 days, that difference was not significant (p > 0.10).

The mean times for re-epithelialisation (of non-grafted

wounds) for each colour are shown in Table 3. The re-

epithelialisation times reported here are similar to those

reported in the laser Doppler product manual (v2.0).

Fig. 4 shows the time to re-epithelialisation when wounds

are sorted by scan colour. The majority of light blue wounds

(72.2%) took >14 days to re-epithelialise or required grafting.

The light blue wounds that healed in <15 days were mostly

flash flame burns where the initial swelling may have

impeded the scan accuracy. The green wounds healed

around 14 days (mean was 13.8 days). The yellow/pink/red

wounds all healed within 15 days, except for one wound

which was erroneously grafted. The burns which were

unable to be scanned were not related to the time taken for

complete re-epithelialisation

3.7. Prediction of wound outcome

The data from this study comply well with the outcomes

suggested in the laser Doppler product manual (v2). We

Fig. 4 – The times taken for complete re-epithelialisation of

the wounds when they are sorted based on their

predominant scan colour. The majority of the light blue

wounds took >14 days to re-epithelialise or required

grafting. Green wounds did not require grafting but many

still required up to 21 days to re-epithelialise. All except

one of the yellow/pink/red wounds healed within 14 days,

this wound was grafted erroneously.

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1 829

support that predictions of the wound outcomes can be made,

based on the scan colour of the wound as shown in Table 4.

3.8. Time to scan

In this study, scans were taken at 0–186 h after the burn. The

predictability of wound outcome based on scan colour was not

significantly affected by the time of the scan. For scans taken

less than 24 h post-burn, the association between scar

management, time to heal and the colour of the scan did

not change significantly compared to other time points.

However, a larger sample size is needed to confirm that scans

are completely reliable during the first 24 h.

4. Discussion

This paediatric study found significant relationships between

LDI scan colour, time for complete re-epithelialisation and

requirement for grafting and scar management. Based on the

data from this study, laser Doppler scanning was used as an

accurate prediction tool for determining wound depth and

outcome and assisting with clinical management. In this

observational study it is important to note that the surgeons

were not blinded to the laser Doppler scan, and this may have

introduced some bias to the resulting clinical judgment and

course of treatment. However, we believe, as do others [7,12]

that the use of the laser Doppler prevents unnecessary surgery

or expedites treatment such as skin grafting and scar

management, leading to decreased hospital stays and there-

fore costs to the patient and hospital. Here we found that the

consultants would often wait, erring on the side of caution and

wounds which scanned as light or dark blue were commonly

left to heal as long as possible, when we could now predict that

they would not heal without grafting and/or scar manage-

ment. This is similar to work by others, who have also found

laser Doppler to be very accurate for assessing wound depth

compared to clinical assessment alone [7,9,12,16]. As clin-

icians become more comfortable with the use of this device

Table 4 – Predictions of wound outcome based on scan colourwith the prediction ability—for the light blue wounds, althoughre-epithelialisation time was <14 days for 3 out of 12 Silvazin

Colour Skin grafting Scar manag

Dark blue <125 PU Yes Yes

Light blue 125–250 PU Yes Yes

Green 250–440 PU No Yes/no (shou

Yellow/pink/red 440–628–1000 PU No No/yes (may

Table 3 – The mean times for re-epithelialisation (of non-graftethe laser Doppler product manual.

Colour of scan Time to re-epithelialisein this study

Light blue 20.5 � 8.3 days

Green 13.8 � 4.9 days

Yellow 10.0 � 2.6 days

Pink 10.0 � 2.6 days

Red 10.0 � 2.6 days

and its accuracy, this may also make it easier to explain to

patients and carers that grafting is warranted for the best long-

term outcome of their burn.

In our experience using the standard colour palette in v1.5 of

the Moor software, we found that if the wound scanned yellow

(440–628 PU),pink (628–812.5 PU) orred (812.5–1000 PU), it would

heal within 14 days and not require grafting or scar manage-

ment. If the wound was green (250–440 PU) it would heal in

approximately 14 days without requiring skin grafting, but

may require scar management. If the wound was light blue

(125–250 PU) it would take approximately 19 days to re-

epithelialise, may require skin grafting and would definitely

requirescar management. If thewound was dark blue (<125 PU)

of the wound. The prior application of Silvazine interferedthe prediction for scar management was 100% correct, the

e treated wounds.

ement Re-epithelialisation Success % of predictions

NA 100

>14 days or grafted 83

ld need) <21 days 100

need) <14 days 100

d wounds) for each colour as reported in this study and in

Time to re-epithelialise as reportedin Moor Manual v2.0

>21 days

�21 days

14–21 days

�14 days

<14 days

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1830

it would require skin grafting and scar management. Thus, the

predominant scan colour can be used to accurately predict

wound outcome and optimal clinical management. Interest-

ingly, for the light blue wounds, there is no way to accurately

predict a need for grafting. Certainly, prior experience with burn

mechanisms by the clinician can assist with clinical judgment

of the requirement for grafting. However, flash/flame burns can

be difficult to clinically judge as initially they may appeardeeper

in depth and here we saw that although the majority of light

blue flame wounds required scar management, they were not

grafted. The light blue wounds took approximately 3 weeks to

heal, and almost all required scar management, suggesting that

perhaps they could have all been grafted. However, there is no

evidence to date that wounds which take more than 3 weeks to

heal would have had a better cosmetic outcome if they had been

grafted. We were also surprised to see that three yellow/pink/

red wounds required scar management even though they all

healed in�14 days. Burns treatment is not an exact science and

rules as to which children require skin grafting and scar

management should not be rigorously applied. Surgeons and

therapists will always use their judgement based on experience

and some children will receive scar management even though

their burns heal within 2 weeks. This is because we know that

the occasional wound will ‘‘break the rules’’ and scar with

resulting contractures without appropriate management.

There were some situations for which LDI did not seem to

be entirely accurate. We found, similar to others [17] that

blisters are unable to be scanned and wet wounds may also

result in reflection artefacts. Interestingly, we found that pre-

treatment with SilvazineTM may make clinical interpretation

more difficult. Possibly this is due to the residual pseudo-

eschar layer that SilvazineTM can leave behind on a wound

even after it is removed. In a previous study comparing

SilvazineTM and ActicoatTM [14], we found that significantly

many more children who received SilvazineTM as their

treatment went on to be skin grafted (25.6% vs 15.4%,

p = 0.001). These patients were treated in our unit before

acquisition of the LDI device; however, perhaps treatment of

their wounds would have been different with the LDI tool as an

adjunct to management.

Unlike others [11], we did not have a problem with scans

after the use of ActicoatTM dressing. Although in that

publication they report visible silver deposition on the wounds

causing interaction with the laser light, we encountered no

such problems, possibly due to our unique irrigation techni-

que, which keeps the dressing moist [13,14]. All scans after the

use of ActicoatTM were accurate for predicting wound depth

and therefore outcome. Although it has been shown that laser

Doppler scanning is not possible through the actual ActicoatTM

dressing [18], this does not interfere with our current

treatment protocol as we are able to scan patients every 3–4

days at the dressing change. It is good to know, however, that

scanning is still accurate through plastic cling wrap [18], as

many patients who suffer from pain during dressing changes

are often alleviated by the simple use of plastic wrap placed

onto the wound.

At what point LDI imaging is deemed not only accurate but

also predictive of the clinical outcome varies within the

literature. In general, 24–72 h is considered the optimal time

frame after-burn. However, in this study approximately half of

the patients were initially only able to be scanned outside of

the 24–72-h window after-burn. The issue, then of when laser

Doppler scanning is still accurate is important. In this study,

we have found that scans taken within 24 h after-burn were as

accurate in predicting clinical outcome as burns taken within

the 24–72 h timeframe. Scans of wounds taken after 72 h were

also accurate. Although burn wound dynamics may continue

to change over 72 h after-burn, making biopsies (which

measure tissue viability) inaccurate, this technology (which

measures tissue blood flow and vessel patency) still seems to

accurately predict wound outcome outside of this timeframe.

It may be that LDI is a more reliable indicator of depth of injury,

especially in the first few hours post-burn. However, as this

area still remains controversial, more work with a larger burn

population is warranted.

In this study there were some patients or burns on patients

which were unable to be scanned. These burns were not

different to the scanned burns in terms of TBSA, depth,

mechanism, anatomical position or patient age. Burns were

often unable to be scanned because the patients were in pain

and we wanted to expedite the dressing. In our centre, we use

age-appropriate toys for distraction, as well as the use of a

musical play specialist. We also have access to other novel

augmented reality distraction tools [19,20], which supple-

ment the use of analgesics. Unfortunately, in our unit we do

not have access to nitrous oxide, which others have found to

greatly benefit their scanning incidence rate [12]. However, in

this study the low-resolution setting was used for all scans

and this takes a maximum of 80 s to complete. We found that

these fast scans were accurate for predicting wound outcome,

and were therefore an excellent option for increasing the

chances of obtaining a scan even when the paediatric patient

was in pain. Other techniques used for younger children

included using a green drape around the scanner head so they

were unable to see the scan in progress, or using a drape to

block their view. These techniques all helped to ensure that

patient treatment proceeded as quickly and pain free as

possible.

LDI is proving to be a useful technology, and many

researchers and clinicians are still trying to work out

specifically how scanned perfusion units relate to wound

outcome. However, we have found that different systems from

different companies and even different versions of the same

system allocate different perfusion units to the same colour

palette. This can make it difficult to compare outcomes

between studies. Having different systems which seem to

arbitrarily assign colour palettes and researchers who do not

report perfusion units in their publications (and only colours

or ‘‘low’’ or ‘‘high’’ flux) make it difficult to clarify this

important research.

In conclusion, we found laser Doppler imaging to be

accurate for predicting wound outcome and we believe it to be

an essential tool for effective wound management. It can be

used accurately as early as within the first 24 h post-burn and

using the low-resolution/fast-scan setting. Laser Doppler

scans are also accurate despite the prior use of ActicoatTM

and SilvazineTM, meaning it can be well integrated into current

wound management regimes. This technology is simple to use

and consistently proving to be a valuable clinical management

tool for paediatric burns.

b u r n s 3 5 ( 2 0 0 9 ) 8 2 4 – 8 3 1 831

Conflict of interest statement

There are no conflicts of interest for any of the authors.

Acknowledgements

This project was supported by funding received from the CASS

Foundation Science and Medicine Grants scheme, Queensland

University of Technology, and Tissue Therapies Ltd.

r e f e r e n c e s

[1] Lund CC, Browder NC. The estimation of areas of burns.Surg Gynecol Obstet 1944;79:352–8.

[2] Sullivan T, Smith J, Kermode J, McIver E, Courtemanche DJ.Rating the burn scar. J Burn Care Rehabil 1990;11(May–June(3)):256–60.

[3] Atiles L, Mileski W, Purdue G, Hunt J, Baxter C. LaserDoppler flowmetry in burn wounds. J Burn Care Rehabil1995;16(July–August (4)):388–93.

[4] Atiles L, Mileski W, Spann K, Purdue G, Hunt J, Baxter C.Early assessment of pediatric burn wounds by laserDoppler flowmetry. J Burn Care Rehabil 1995;16(November–December (6)):596–601.

[5] Niazi ZB, Essex TJ, Papini R, Scott D, McLean NR, Black MJ.New laser Doppler scanner, a valuable adjunct inburn depth assessment. Burns 1993;19(December (6)):485–9.

[6] Watts AM, Tyler MP, Perry ME, Roberts AH, McGrouther DA.Burn depth and its histological measurement. Burns2001;27(March (2)):154–60.

[7] Pape SA, Skouras CA, Byrne PO. An audit of the use of laserDoppler imaging (LDI) in the assessment of burns ofintermediate depth. Burns 2001;27(May (3)):233–9.

[8] Riordan CL, McDonough M, Davidson JM, Corley R, Perlov C,Barton R, et al. Noncontact laser Doppler imaging in burndepth analysis of the extremities. J Burn Care Rehabil2003;24(July–August (4)):177–86.

[9] Holland AJ, Martin HC, Cass DT. Laser Doppler imagingprediction of burn wound outcome in children. Burns2002;28(February (1)):11–7.

[10] Mileski WJ, Atiles L, Purdue G, Kagan R, Saffle JR, HerndonDN, et al. Serial measurements increase the accuracy oflaser Doppler assessment of burn wounds. J Burn CareRehabil 2003;24(July–August (4)):187–91.

[11] Jeng JC, Bridgeman A, Shivnan L, Thornton PM, Alam H,Clarke TJ, et al. Laser Doppler imaging determines need forexcision and grafting in advance of clinical judgment: aprospective blinded trial. Burns 2003;29(November (7)):665–70.

[12] La Hei ER, Holland AJ, Martin HC. Laser Doppler imaging ofpaediatric burns: burn wound outcome can be predictedindependent of clinical examination. Burns 2006;32(August(5)):550–3.

[13] Cuttle L, Mill J, Kimble RM. Acticoattrade mark: a cost-effective and evidence-based dressing strategy. Burns2008;34(June (4)):578–9.

[14] Cuttle L, Naidu S, Mill J, Hoskins W, Das K, Kimble RM. Aretrospective cohort study of Acticoat versus Silvazine in apaediatric population. Burns 2007;33(September (6)):701–7.

[15] Rockwell WB, Ehrlich HP. Should burn blister fluid beevacuated? J Burn Care Rehabil 1990;11(January–February(1)):93–5.

[16] Devgan L, Bhat S, Aylward S, Spence RJ. Modalitiesfor the assessment of burn wound depth. Journal[serial on the Internet]. 2006 Feb 15; 5: available from:<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=16921415>.

[17] Droog EJ, Steenbergen W, Sjoberg F. Measurement of depthof burns by laser Doppler perfusion imaging. Burns2001;27(September (6)):561–8.

[18] Holland AJ, Ward D, Farrell B. The influence of burnwound dressings on laser Doppler imaging assessmentof a standardized cutaneous injury model. J Burn Care Res2007;28(November–December (6)):871–8.

[19] Mott J, Bucolo S, Cuttle L, Mill J, Hilder M, Miller K, et al. Theefficacy of an augmented virtual reality system to alleviatepain in children undergoing burns dressing changes: arandomised controlled trial. Burns 2008;34(March (6)):803–8.

[20] Miller K, Bucolo S, Patterson E, Kimble RM. The emergenceof multi-modal distraction as a paediatric painmanagement tool. Stud Health Technol Inform2008;132:287–92.