100-602-1-pb

7/30/2019 100-602-1-PB

1/8

17

Treatment Modalities: Stevens-Johnson syndrome and toxic epidermal necrolysis

2011 Volume 4 No 1Wound Healing Southern Arica

Stevens-Johnson syndrome and toxic epidermal necrolysis:

topical treatment infuencing systemic response

Widgerow AD, MBBCh, FCS(SA) (Plast), MMed(Wits), FACS

Honorary Proessor, Plastic Surgery, University o the Witwatersrand and Irvine, Caliornia

Correspondence to: Alan Widgerow, e-mail: [email protected]

Keywords: Stevens-Johnson syndrome, toxic epidermal necrolysis

Introduction

Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis

(TEN, Lyells syndrome) can be considered to be among the most

devastating and lie-threatening o the adverse skin reactions.

The two conditions are defned according to severity, based on the

extent o skin involvement. SJS is considered to be a minor orm

o TEN, characterised by less than 10% total body-surace area

(TBSA) o skin detachment, and an average reported mortality o

1-5%, whereas TEN is defned as more than 30% skin detachment,

and an average reported mortality o 25-35%.1 Both conditions

are characterised by keratinocyte apoptosis, and cleavage o the

epidermis rom the dermis, resulting in the exposure o large areas

o dermis akin to superfcial and partial thickness burn injuries. In

this paper, TEN will be used to denote both conditions.

Drugs most commonly associated with TEN are antibiotics such

as sulphonamides, tetracyclines and quinolones; anticonvulsants

such as phenytoin, phenobarbital and carbamazapine; antiretroviral

drugs; nonsteroidal anti-inammatory drugs; and allopurinol.1

TEN is a serious systemic disorder with the potential or severe

morbidity, and even death. Missed diagnosis is common. Although

sepsis is currently accepted as the main cause o mortality, much

o the morbidity and subsequent threat to lie is orchestrated by

an exaggerated inammatory response, with major outpouring

o cytokines and destructive matrix metalloproteinases (MMPs).

These mediators and proteinases cause intense destruction o the

extracellular matrix, major uid shits and a systemic inammatory

response (SIRS) that has lie-threatening consequences.2 Thus

modern therapeutic interventions, aside rom resuscitative eorts,

have sought to control not only the potential sepsis, but also the

marked inammatory response in relation to this pathology.

Nanocrystalline silver (NCS) is one such agent. It has exceptional

antimicrobial efcacy, but is also eective in lowering MMP levels, a

critical part o the disease process.3-5

In the South Arican environment, a number o doctors are using NCS

silver dressings to treat patients with TEN. However, no signifcant

data relating to clinical outcomes and resource utilisation or this

treatment modality are available. Twelve cases o confrmed TEN,

treated with NCS, were analysed with a view to assessing efcacy

and setting logical guidelines or this devastating condition.

Method

Cases were included where details were available and consented to

by patients, sta and amilies. Details were obtained rom medical

records, rom doctors treating the patients, and rom patients and

amilies o patients. Inormation was collated and analysis o this

inormation is presented.

Abstract

Toxic epidermal necrolysis (TEN) and Steven-Johnson syndrome (SJS) are rare, but potentially lie-threatening diseases, which relate to a

variety o medications, and which are characterised by widespread epidermal necrosis. Although sepsis is currently accepted as the main

cause o mortality, much o the morbidity and subsequent threat to lie is orchestrated by an exaggerated inammatory response with major

outpouring o cytokines and destructive matrix metalloproteinases (MMPs). These mediators and proteinases cause intense destruction o

the extracellular matrix, major uid shits, and a systemic inammatory response (SIRS) that has lie-threatening consequences. Modern

therapeutic interventions, aside rom resuscitative eorts, have sought to control not only the potential sepsis, but also the marked inammatory

response in relation to this pathology. Nanocrystalline silver (NCS) is one such agent. It has exceptional antimicrobial efcacy, but is also

eective in lowering MMP levels, a cr itical part o the disease process. Twelve cases o confrmed TEN, treated with NCS, were analysed with

a view to assessing efcacy and setting logical guidelines or this devastating condition. Eleven out o the 12 cases were categorised in the

TEN diagnostic group, with total body-surace area involvement ranging rom 22-100% (mean 64%). Two patients in the series died (16.6%),

but both had a severe co-morbid disease background. Topical application o NCS (Acticoat, Smith and Nephew Hull, UK) dressings, as part

o a comprehensive dressing protocol, proved successul in the management o this devastating disease. Guidelines or the management o

TEN are suggested.

Medpharm Wound Healing Southern Africa 2011;4(1):17-24

7/30/2019 100-602-1-PB

2/8

18



Twelve cases were included in the study. Eleven out o 12 cases

were categorised in the TEN diagnostic group. One patient (22%

TBSA) was classifed in the intermediate area between SJS and TEN.

No reactions were o the milder SJS orm and TBSA involvement

ranged rom 22-100% (mean 64%). Two patients in the series died

(16.6%). All patients had a clinically confrmed diagnosis o TEN.

The dressing routine was as ollows: wounds were cleansed with

water, saline, or diluted chrlorhexidine and dry skin ragments were

removed. Operating room debridement was undertaken where

deemed necessary by the physician. Intrasitegel (Smith and Nephew

Hull, UK) was applied to wounds and Acticoat (Smith and Nephew

Hull, UK) was applied over the gel. In the majority o cases, classic

Acticoat (a three-layered dressing) was used. The gel created a

moist environment in which to activate the Acticoat dressing, and

to encourage moist wound healing and epithelialisation. A secondary

absorbent dressing, usually oam such as Allevyn oam or Exudry

(Smith and Nephew Hull, UK, or equivalent), was applied i excess

moisture or exudate was anticipated. Most dressings were changed

every three days, unless excess exudate dictated daily dressings

or the frst ew days (three out o 12, 25% o cases). Acticoat

dressings were stopped when epithelialisation was judged to be 90%

complete. I the wound was considered to be deep partial thickness

or deeper, or where scarring or contracture was considered to be

a real possibility, a biological skin substitute was introduced to the

dressing regimen, such as Biobrane (Smith and Nephew Hull, UK).

Operative debridement was carried out using Versajet apparatus

(Smith and Nephew, Hull UK) where deemed necessary.

ResultsThe group comprised seven emale patients and fve male patients.

Age varied rom eight to 52 years (mean 31 years). Background

diseases included diabetes, tuberculosis, immunocompromised state,

epilepsy, varied respiratory problems, head injuries, depression, and

addiction rehabilitation (see Table I). The average number o days

rom the onset o SJS/TENS to diagnosis was 4.6 (+3.8), median 3.0,

range 1-10. In the majority o cases sulphonamides (50%, six out

o 12) or carbamezapine (33.3%, our out o 12) were administered

as causative agents. Epilepsy prophylaxis occurred in 42% o cases

(fve o 12), with epilepsy being diagnosed in only one o these cases.

The mean number o days o exposure to the causative agent was

5.8 days (range one to 24). All cases presented with superfcial-to-

partial thickness (one case, 6% deep partial thickness within 22%

TBSA) wounds. No ull thickness wounds resulted, and no skin

grating was necessary. Two deaths occurred in the series, related to

respiratory ailure (16.6%).

The treatment regimen ollowed routine lines o resuscitation. Nine

out o 12 cases were treated in burn wound acilities, with the

majority o cases (10 out o 12, 86%) having passed through the

intensive care unit (ICU) or high-care acilities. Initial resuscitation

(correct uid loss assessed, as or burn injuries), immediate

withdrawal o the oending agent on recognition and varied pain

control (analgesics to narcotics) was unremarkable. Intravenous

steroids were used in one case, and intravenous immunoglobulins

in two o the 12 cases (16%).

The TBSA o skin disruption varied rom 22-100% (mean 59%). The

most common isolated organisms were Staphylococcus aureus,

Pseudomonas aeruginosa, and Enterobacteriaceae species, with

blood-borne septicaemia disease confrmed in only one case.

Mouthwashes were prescribed where oral mucosal involvement was

evident. A dermatologist or plastic surgeon conducted the diagnosis

and management in seven cases (58%), and an ophthalmologist was

consulted in our cases (33%). Operating room debridement was

undertaken in 33% o cases. Most dressings were changed every

three days, except where copious exudate dictated daily dressings

or the frst ew days in three o the 12 cases (25%).

Only six out o 12 cases were treated in the operating room; two

o these or a tracheotomy procedure, rather than or debridement,

and dressing change (one patient had tracheotomy and dressings

perormed in the operating room). Dressings were carried out in

ICU, high care, or in the ward in eight out o 12 cases. Operative

procedures involved debridement and dressings in our cases, and

three tracheotomy procedures. No dressing application problems

were reported. Only one case reported residual scarring and

pigmentation related to the TEN, but these resolved without sequelae.

No unctional problems were reported in this patient. This was the

only contracture that was reported that related to a tracheotomy

procedure. Following discharge, it resolved over a ew months.

Commonly reported TEN complications were noted in this series (see

Table II), with the exception o sepsis and Septicaemia, that were

not commonly observed. Only two cases (both 100% exposure)

experienced signifcant sepsis problems. Two cases o septicaemia

were seen, one o which resolved completely over a ew days.

Long-term ever, or other symptoms and signs o inection, were not

apparent ollowing the application o the dressing regimen, other

than in the one recorded case.

Acticoat dressings were stopped when epithelialisation was judged

to be 90% complete. Time to healing ranged rom 12 days to 65 days

(mean 23 days). A biological skin substitute (Biobrane) was added

to the dressing regime in three o the 12 cases (25%) and Versajet

debridement was carried out in three cases (25%).

Table I: Primary diagnosis, causative actors and co-morbidities (n = 12)

n %

Primary diagnosis/causative factor

Toxic epidermal necrolysis: sulphonamide allergy 6 50

Toxic epidermal necrolysis: carbamazepine allergy/epilepsy 4 33.3

Toxic epidermal necrolysis: lamictin allergy/epilepsy 1 8.3

Toxic epidermal necrolysis: chickenpox/varicella/acute tonsillitis 1 8.3

Co-morbidities

None 2 16.7

Turberculosis and type 2 diabetes 1 8.3

aDiabetes and immunocompromised 1 8.3

aTurberculosis and immunocompromised 1 8.3

Addiction, dependency, depression 1 8.3

a = deaths

7/30/2019 100-602-1-PB

3/8

19



Discussion

First described in 1922, SJS is an immune-complex-mediated

hypersensitivity disease. TEN and SJS are rare, but potentially lie-

threatening diseases that relate to a variety o medications, and are

characterised by widespread epidermal necrosis.1,2,6 The majority

o cases appear to be related to idiosyncratic drug reactions,

and the drugs most commonly cited include antibiotics such as

sulphonamides, tetracyclines and quinolones; anticonvulsants such

as phenytoin, phenobarbital and carbamazapine; antiretroviraldrugs; nonsteroidal anti-inammatory drugs; and allopurinol.1,2 In

this series, sulphonamide antibiotics (50%) and carbamezapine

drugs (33%) were responsible or the vast majority o cases. Skin

biopsy is the defnitive diagnostic investigation, demonstrating

subepidermal bullae, epidermal cell necrosis, and perivascular areas

infltrated with lymphocytes.1 However, in this series, as in others,

the clinical diagnosis was made relatively easily without impacting

on the prognosis, and without skin biopsy.

Aside rom the devastating consequences o the skin separation,

TEN can involve inammation and subsequent necrosis o mucous

membranes, including oral, nasal, ophthalmic, vaginal, urethral,

gastrointestinal and lower respiratory tract tissues, with severe

internal maniestations o this disease. Three cases in this series

eatured oral mucous membranes, which contributed considerably

to the morbidity seen in these cases (see Figure 1). Enteral eeds

(three patients) were ound to be better tolerated and more eective

than parenteral eeds (one patient, longest recovery).

It is likely that lower respiratory tract involvement was a major

actor in the two deaths in this series. Both these cases died o

respiratory distress, and both were immunocompromised [human

immunodefciency virus (HIV)], one case being compounded by

pulmonary tuberculosis. The TBSA in these cases was 35% and

60%, and the wounds did not appear to present a major problem in

management o these cases.

Presentation

Patients reported a non-specifc upper respiratory tract inection in

most cases. The prodrome o up to 14 days was characterised by

ever, sore throat, headache and malaise. Mucocutaneous lesions

and blistering then began in specifc regions (see Figure 2). The initial

presentation was very similar to an acute burn wound, dependent on

the TBSA exposure: hypotension, ever, tachycardia and dehydration.

In three cases, oral lesions prevented patients rom eating or

drinking. In three cases, ocular involvement was documented.

The SCORTEN scale is a severity-o-illness scale with which

the severity o certain bullous conditions can be systematically

determined (see Table III). It was originally developed or TEN.7 In the

SCORTEN scale, seven independent risk actors or high mortality are

systematically scored, so as to determine the mortality rate or that

particular patient.

Table II: Complications relating to toxic epidermal necrolysis (n = 12)

Anaemia (as defned by the World Health Organization) requiring packed

cells or blood transusion5

Mucous membrane involvement, e.g. oral, nasal and gastrointestinal

involvement, oesophageal strictures4

Ocular involvement, e.g. symblepharon, keratoconjunctivitis,pseudomembranous conjunctivitis

4

Pneumonia or lower respiratory tract involvement (inormation rom

treating physician)4

Renal ailure (inormation rom treating physician) 3

Septicaemia (positive blood cultures) 2

Genital involvement, e.g. penile scarring, vaginal stenosis (inormation

rom treating physician)2

Ater resolution: scarring, cosmetic deormity, contractures, unctional

problems2

Postinammatory hyperpigmentation 1

Pulmonary embolus 1

Recurrence o inection through slow-healing wounds -

Figure 1: Toxic epidermal necrolysis cases with oral mucous membrane involvement Figure 2: Appearance o mucocutaneous lesions and blistering

Table III: The SCORTEN scale

Risk factor 0 1

Age < 40 years > 40 years

Associated malignancy No Yes

Heart rate (beats/minute) < 120 > 120

Serum blood, urea, nitrogen (BUN) (mg/dl) < 27 > 27

Detached or compromised body surace < 10% > 10%

Serum bicarbonate (mEq/l) > 20 < 20

Serum glucose (mg/dl) < 250 > 250

7/30/2019 100-602-1-PB

4/8

20



The more risk actors that are present, the higher the SCORTEN

score, and the higher the mortality rate, as shown in Table IV.7

In this series, the SCORTEN system proved to be only partially useul.

Prognostic actors were signifcantly increased in the two cases that

died (scored 3 and 4 respectively), but the chart did not accurately

reect the signifcance o the background diseases that appeared to

play a dominant role in the death o these patients. Although the HIV

status does not necessarily coner added risk (this will be discussed

later), it appears that HIV, combined with another background

illness, severely compromises the prognosis. In this series, the

two patients who died were both HIV positive, one compounded by

severe pulmonary tuberculosis, the other with advanced diabetes.

Had these actors been taken into account by changing the risk

parameter to associated malignancy, or immunocompromised or

co-existent severe chronic disease, both these cases would have

been classifed as 5 or more in the prognostic scale, with a > 90%

mortality. This is even more apparent when one considers that the

patient with severe HIV and pulmonary tuberculosis was a young

adult, had 35% TBSA wounds, but was overwhelmed by respiratory

system ailure.

Intravenous treatment and debridement

Although TEN is a serious systemic disorder with the potential or

severe morbidity and even death, no efcacious pharmaceutical

treatments have been established as standard therapy.1.2 When

easible, admission to a burns or intensive care unit is considered

crucial.1,2,8 Immediate withdrawal o the suspected drug, uid

and electrolyte replacement and topical wound care orms the

cornerstone o frst-line therapy.1,8 Interestingly, corticosteroid usage

has been abandoned (used in one case in this series), and the role

o immunosuppressants in this context is not well understood, and is

not considered standard.1 Having noted this, the role o intravenous

immunoglobulins (IVIG) is worthy o urther discussion.

The largest o the studies involving IVIG was a multicentre study o

48 patients with TEN. Success was concluded based upon an 88%

survival rate, but also the rapid (mean o 2.3 days) cessation o skin

and mucosal detachment in 89.6% o the patients. Mortality was

associated with a lower dose o IVIG, longer time o onset to IVIG use,

co-existing underlying chronic conditions, older age, and greater

body surace area involved.9-13 Brown et al12 reported on 45 patients

with a TBSA epidermal detachment o 45%. Twenty-one were treated

with standard therapy, and 24 were treated with IVIG. Mortality was

42% in the IVIG group, vs 29% in the standard-care group. Mortality

in the IVIG-treated group was also higher than that predicted by

SCORTEN (38%). These surprising results were discussed by French

et al.14 Several eatures o this study were noted as unusual. Wound

debridement, a practice that is no longer routinely recommended,

was perormed on all patients at admission.1 The authors suggest

that the association o IVIG treatment and debridement may have

impacted on the poor outcome in the study. Additionally, the lowdose and late administration o IVIG may have contributed to the

surprising results.1

These are interesting observations. In this series, our cases

underwent debridement in the operating room (18, 30, 30 and 65

days healing time). Two cases had IVIG therapy (18 and 24 days

healing time), one o which underwent operating room debridement.

Interestingly this case involved a TBSA o only 22%, the smallest

surace area in this series, but the deepest recorded wounds (18

days healing time). The size o this series precludes us rom making

any defnite decisions in this regard. However, some observations

are appropriate. IVIG did not seem to have an impact on prognosisin these cases. Debridement did not seem to aect healing time or

prognosis, but numbers are small or defnitive statements. One may

argue that cases reerred to plastic surgeons may have been more

severe, and thus may have warranted debridement in the operating

room. However, the other parameters o these cases (TBSA, systemic

symptoms and signs) do not confrm this supposition. Thereore,

debridement and IVIG therapy do not appear to be constants in a

TEN therapeutic regimen. This is especially complicated in HIV-

positive patients. These patients are more prone to TEN when their

CD4 counts are high (able to mount a response). Administration o

IVIG may help to douse the immune response, and theoretically haltthe process o skin separation. However, it is unknown whether

debridement and exposure o large areas o wounds in patients who

have decreased CD4 counts now, makes them more susceptible to

inective complications. An alternative therapeutic agent that halts

the skin separation process, without systemic immune modulation,

would be preerable. This may be possible with the right choice o

dressing, an important ocus o this paper.

Additionally, a total o six blood transusions were administered.

Three o our patients were debrided, and three o eight were

not debrided. The trend appears to demonstrate a likelihood o

debridement being ollowed by blood replacement, although othervariables may be possible. Thus, aggressive debridement should be

restricted unless absolutely necessary. In most cases, a controlled

gentle washing or debridement should sufce.

The use o systemic steroids remains controversial.1 Treatment with

systemic steroids has been associated with an increased prevalence

o complications. The ophthalmology literature contains several

papers that advocate systemic and topical steroids to minimise

ocular morbidity.15-19 Treatment with steroids appears to be limited

to that directed against ocular complications.

Causative drugs and human immunodeciency virusUse o therapeutic drugs is reported in over 95% o patients with

TEN. A strong association between drug ingestion and development

Table IV: Risk actors in relation to mortality ra te

Risk factors Mortality rate

0-1 3.2%

2 12.1%

3 35.3%

4 58.3%

5 or more > 90%

7/30/2019 100-602-1-PB

5/8

21



o the cutaneous eruption is observed in 80% o cases.1-3 Many

drugs have been identifed as causative agents, and the drugs in

this series appear to be typical o those reported previously. 1,8,20.21

However, with the advent o greater numbers o HIV-positive cases,

antiretrovirals have come into ocus. More particularly, nevirapine

has been reported as a cause o TEN in some reports.22,23 Metry etal reported Stevens-Johnson syndrome in two HIV patients treated

with nevirapine, and mentioned one other in the literature.24Although

sulphonamide drugs have been the most common cause o adverse

skin reactions in HIV-inected patients or many years, nevirapine

has become the leading cause o severe skin reactions in these

patients in the past decade.23 Although the incidence o nevirapine-

related rash is reported to be as high as 36.0%, the requency o

TEN described in HIV patients on ART is relative low, between 0.3-

1.5%.22-26 Risk actors or nevirapine-related rash include the emale

sex, a high baseline CD4 cell count, a history o drug allergy, lower

body weight, a high nevirapine plasma level, and probably certain

human leukocyte antigen types.22-26

Allied to this, many patients are treated or complications o HIV

inection, particularly respiratory complications and pneumonia

[(Pneumocystis carinii pneumonia (PCP)] with sulphonamide

antibiotics. This increases the risk o TEN considerably, particularly

in cases with high CD4 counts. The overlap o sulphonamide and

nevirapine therapy in these patients may make identifcation o

the causative agent more difcult. It is important that physicians

are aware o the potential problems associated with these drugs

in this group o patients. Sulphonamide antibiotics appear to be

particularly troublesome in this regard, and it may be time to seek

alternatives, or to strictly control the time o exposure to these

drugs when dealing with respiratory complications o HIV disease.

Physicians should consider the risk o the lie-threatening cutaneous

reactions when prescribing a change to a nevirapine-based regimen

in patients with a high CD4 count. In this series, it is important to

note that the HIV status o many o the patients on sulphonamide

antibiotics was not known, or disclosed. In keeping with South

Arican laws o confdentiality, this inormation was not orthcoming.

Thus, it is possible that more o these cases were HIV positive. It is

noteworthy that o the two cases who died, both were HIV positive,

one with advanced HIV disease and pulmonary tuberculosis, and the

other, with advanced diabetic disease. The signifcance o severe co-

morbidities is stressed.

In dealing with HIV patients in many situations, the current regimen

advocates using sulphonamide antibiotics as prophylaxis against

PCP when CD4 counts are low. Once antiretrovirals are started, and

the CD4 counts rise to an acceptable level, the sulphonamides are

stopped. Although difcult to ascertain rom this study, it is likely that

as CD4 counts increase, patients are probably going to be susceptible

to TEN i they are on sulphonamide antibiotics. It would be worthwhile

considering withdrawing the sulphonamides as soon as the CD4

count starts responding and rising, rather than waiting or it to reach

high levels, especially i they are being used prophylactically. This

is worthy o urther discussion with physicians. O course, this will

help i the TEN is a result o sulphonamide exposure. I nevirapine

is the cause, the problem will persist. It would appear that in cases

o immunosupression, patients appear to be more susceptible to

TEN outbreaks as their immune status improves. It is during this

transition o immunity that careul drug choices and usage, or

limitations, should be borne in mind by managing physicians.

An additional observation that warrants discussion concerns the use

o antiepileptic prophylaxis, or the use o these drugs in patients

without documented epilepsy. In this series, only one patient was

reported to have diagnosed epilepsy. The other our cases were

prescribed antiepileptic drugs or head injury, meningitis and

depression, as part o a rehabilitative programme. Circumspection

is needed when choosing drugs, especially when they are selected

prophylactically.

Infection

With cleavage o the epidermal rom the dermal plane, impaired

barrier unction o the skin results, with inevitable bacterial

colonisation. Inection is generally considered to be the major

actor associated with morbidity and mortality in TEN.1,7-10 However,

although ever was present in many cases periodically, sepsis and

Septicaemia were not the major components o morbidity in 10 o

12 cases in this series. It would appear that the efcacy o NCS

against bacterial burden and gross sepsis had the desired eect o

controlling any overwhelming septic incidents. The one exception to

this was the case involving 100% o the TBSA. This was a severely

ill patient, with a delayed transer to the surgeon, and where

healing took 65 days, the longest in the series. In this case, multiple

organisms were isolated, including Candida albicans. Interestingly,this was the frst case, dating back eight years, in which NCS was

used. Excluding this case, mean healing time in this series would

have been 15 days.

Additionally, when sta were questioned on the tipping point to

wound healing (when skin separation appeared to stop, and the toxic

symptoms and signs began to subside), 11 o the 12 related this

to the start o the NCS dressing regimen. The single case where

extended healing and septic episodes occurred is described above.

The two cases that died were not reported to have signifcant wound

healing problems. Sepsis did not appear to play a role in their

deaths, which were due to respiratory ailure, against a background

o severe chronic disease and compromised immunity. Overall

isolated organisms included Staphylococcus aureus, Pseudomonas

aeruginosa, Enterobacteriaceaeand in one case, Candida albicans

organisms. It is also important to note that prophylactic antibiotics

were prescribed in 90% o cases.

Nanocrystalline silver

Pathologically, cellular cleavage o the epidermis rom the dermis

occurs. The death receptor, Fas, and its ligand, FasL, have been

linked to the process, as has tumour necrosis actor-alpha (TNF-

alpha). Researchers have ound increased soluble FasL levels in the

sera o patients with SJS and TEN beore skin detachment, or the

inset o mucosal lesions.1,27 Others have also linked inammatory

7/30/2019 100-602-1-PB

6/8

22



cytokines to the pathogenesis.28 The process o epidermal basal cell

separation, with detachment o this layer, is very suggestive o a

proteolytic process thought to involve MMPs.28 Keratinocytes have

also been shown to express metalloproteinases such as gelatinase

A (MMP-2) and gelatinase B (MMP-9), which are able to degrade

components o the basement membrane.28 Increased gelatinaseactivity in theculture medium o skin rom TEN and SJS patients,

maintained in organ culture and in blister uid, may be responsible

or the detachment o the epidermis in this drug-induced necrolysis.28

As with many other inammatory conditions, the control o MMPs

and their destructive eects, appears to be critical. Additionally, the

two processes described above are likely to be interrelated. Recent

papers have suggested that the apoptosis o cells (endothelial and

neural) in varying disease processes may be as a direct result

o MMP-9 release, that, in turn, causes a release o extracellular

soluble FasL.43,44 I this is the case, the control o MMP-9 should

aect the entire cleavage and apoptotic process that is seen in TEN.

This is where NCS has become a very exciting therapeutic modality

or control o these conditions. Acticoat NSC dressings exert rapid

(within 30 minutes) and potent (releasing active silver at a level

o 70-80 g/ml) antibacterial action that is sustained or a period

o three days.31 Not only does NCS have a broad, eective, well-

accepted potency against bacterial and ungal organisms, but the

Ag0 ion appears to oer signifcant anti-inammatory eects.3-5,32-34

This is particularly evident against MMP-9, the destructive gelatinase

responsible or much o the destruction in the wound milieu o

chronic non-healing wounds, chronic wound uid corrosive eects,

and in some acute (dermatitic) conditions.35-37

MMP-9 coordinates and eects multiple events that are involved

in the process o epithelial regeneration.35,38 In an acute surgical

wound, MMP-9 is transiently expressed.38 In contrast, MMP-9 is

persistently elevated in chronic wounds. Increased MMP-9 is seen

in decubitus ulcers, venous stasis ulcers and non-healed burn

wounds. As these wounds heal, MMP-9 disappears.35-37 MMP-9

is not expressed in normal, uninjured skin.38 Reis et al speculate

that the excess MMP-9 ound in the centre o burn wounds, or in

the chronic wound milieu, prevents reconstitution o the normal

dermal and epidermal structures.38 According to these authors, the

complicated interplay between the matrix attachment molecules,cytokines, inhibitors and activators in the wound environment

becomes disjointed and unbalanced in the wound that ails to heal.38

The TEN epidermal cleavage eect appears to ft well into this model

o MMP-9 excess.3-5,35

Most o the research that unearthed the anti-MMP-9 eect o NCS

started with studies that were carried out in a contact dermatitis

porcine model. Contact dermatitis is associated with potent

inammation and raised MMP-9 levels. NCS was shown to be

more eective than traditional anti-inammatory agents, such

as silver nitrate, in lowering inammatory mediators. 3-5,35 The

decreased inammation in the NCS-treated group was associated

with increased inammatory cell apoptosis, a decreased expression

o proinammatory cytokines, and decreased gelatinase (MMP-9)

activity. These data oer support that a species o silver (Ag 0) that

is uniquely associated with NCS may be responsible or the anti-

protease activity and improvement in healing.4 Results suggest

that at higher pHs, more o the total silver in solution is actively

anti-inammatory, likely to be related to Ag0 clusters.4,35 The Ag+

released into the solution reacts with hydroxyl ions, orming calciumhydroxide in the solution, and causing precipitation, resulting in

a greater amount o Ag0 species being available to aect its anti-

inammatory properties.39

Considering these acts related to the disease entity and the

therapeutic eects o NCS, it would appear that this is an ideal agent

or topical application to acute TEN wounds. The anti-inammatory

eects and efcacious bioburden-diminishing abilities appear to halt

the destructive process and decrease complications. Although this is

a small series, the use o NCS appears to have been very eective

in achieving these goals. This report echoes similar fndings in other

reported series or case reports.8,40

Recommended treatment

Although this series involved Smith and Nephew dressings, the

real ocus is on the use o NCS in a unique spectrum o pathology.

From the authors perspective, it is important that this study is not

misconstrued as one that recommends the products mentioned here

only. Many orms o oams, exudate absorbent dressings and other

hydrogels may well be acceptable to use in these cases. In act, it

is believed that the use o Versajet is not indicated in most TEN

cases. However, NCS (and to an extent, Biobrane) appear to be real

advances in the management o TENS.

Based on the fndings o 12 cases in this series, a provisional

guideline can be suggested or the management o TEN. This is

aimed at limiting disease progression and preventing inection.

An acute awareness o the condition is important, particularly in the

light o increased disease prevalence o specifc diseases, where

the disease process and the therapy put the patient at increased

risk o TEN occurrence. This will prevent unnecessary delays in

diagnosis. Consideration should be given to stopping prophylactic

sulphonamide antibiotics in HIV patients on antiretrovirals as soon as

CD4 counts are seen to be rising.

Once the diagnosis is made, this syndrome should be treated as

an acute burn, with admission to a burn wound centre i possible.

Full burn resuscitative measures are instituted. Anticoagulation with

low-dose heparin should be considered in all cases, as one would do

when treating comparable burn injuries.

Initial and continued debridement is controversial, and i considered

necessary by the surgeon, debridement with Versajet should be

gentle, at a very low setting, removing blisters and involved epidermis

without damaging the underlying dermis. This is an aspiration,

vacuum mode, rather than an excisional mode.41 (see Figure 3). This

is extremely important: i over-aggressive debridement is carried out,

the patient has a good chance o experiencing the added morbidities

o bleeding, sepsis, delayed healing and scarring. In most cases,

7/30/2019 100-602-1-PB

7/8

23



the nature o the injury appears to suit a mechanical gentle wash,

rather than an aggressive debridement. Thus Versajet is probably

not indicated in most cases.

Many cases can be dressed directly without debridement, ollowing

a gentle cleansing in the ICU/high-care environment. The NCS

dressing should applied as soon as possible. Initial application o

an hydrogel may acilitate the application o the Acticoat (NCS)

dressing. Choices are available, o Acticoat or Acticoat Flex,

making it easier to choose an appropriate dressing suited to the

wound circumstance. The amount o exudate, anatomical areas

involved, and nature o the wound, will determine which Acticoat

dressing would be used. Ideally, the dressing should be changed

every third day, or longer, i possible. This has the distinct advantage

o encouraging healing with minimum exposure o the wounds. It is

also advantageous in terms o pain control, an exceedingly important

component o the healing process. I deemed necessary, a secondary

absorptive dressing can be applied.

Should the wound be assessed to be deeper (deep partial thickness),

or in contracture-prone anatomical areas, a biological skin

substitute is an excellent addition to the process. This aids in dermal

regeneration and can prevent potential scarring and contractures in

anatomically prone areas.42 Additionally, a biological skin substitute

also reduces exudate loss and reduces pain, and decreases the

number o dressings, so it may be chosen as a primary dressing in

combination with Acticoat in many cases, even where depth is not

in question, but where anatomical areas are prone to contractures

(the neck, hands, eet and elbows). (See Figure 4.) In this series,

Biobrane was successully used in appropriate cases. The skin

substitute is placed on the wound frst, ollowed by the NCS dressing.

The combination dressings are changed when they appear saturated

with wound exudate, or i the wound appears to have dried. The

regimen is continued until 90% healing is achieved. Most o these

areas may then be let exposed with moisturising agents that are

used to prevent desiccation.

Systemic therapeutic interventions are applied on an individual

basis, according to circumstances. This includes intravenous

antibiotics, blood replacement and systemic respiratory or cardiac

support agents (inotropes, volume expanders and diuretics).

Enteral eeds are preerred to parenteral eeds.

Conclusion

TEN is a devastating disease, with signifcant mortality i not

diagnosed and managed early and aggressively. It is likely that therewill be increasing numbers o TEN due to more immunocompromised

patients, as a result o disease or treatment modalities.

In keeping with newer strategies to inuence systemic outcome

by targeting the local wound interace, newer dressings are being

used to aid healing and to control sepsis, and now to decrease the

destructive inammatory component o the disease. NCS is one such

agent that has an impressive antibacterial spectrum, but also has

the potential o dousing the excessive inammatory component o

the disease. As in previous reports, the dressing appears to have

been eective in this series o patients in terms o sepsis control and

Figure 3: Gentle debridement with Versajet. This is only when deemed

necessary. A gentle wash is preerable.

Figure 4: Application o a biological skin substitute (Biobrane) prior to the

Acticoat application

Figure 5: Surviving cases in the series underwent eective non-scarring

epithelialisation

7/30/2019 100-602-1-PB

8/8

24



initiating the tipping point o a halt to epidermal cleavage and the

start o re-epithelialisation (see Figure 5).8,40 Together with biological

skin substitutes, NCS can serve as an eective means to promote

healing, control pain and prevent contractures in a potentially

devastating disease process. I this can be achieved by local dressing

application, rather than systemic immune-modulating agents whichhave inherent problems or host and disease, the choice would be a

very logical one.

Acknowledgements

Acknowledgement is given to the ollowing clinicians or help with

clinical details relating to the cases discussed above: Drs Struwig,

Steyn, Schroder, Altena, Mokgosi, Moledi, Hartslie, French, Naude

and Sacoor.

References

1. French LE. Toxic epidermal necrolysis and Stevens-Johnson syndrome: our currentunderstanding. Allergol Int. 2006;55(1):9-16.

2. Murata J, Abe R, Shimizu H. Increased soluble Fas ligand levels in patients with Stevens-

Johnson syndrome and toxic epidermal necrolysis preceding skin detachment. J Allergy

Clin Immunol. 2008;122(5):992-1000.

3. Bhol KC, Alroy J, Schechter PJ. Anti-inammatory eect o topical nanocrystalline silver

cream on allergic contact dermatitis in a guinea pig model. In: International investigative

dermatology meeting. Miami Beach; 2003.

4. Nadworny PL, Wang JF, Tredget EE, Burrell RE. Anti-inammatory activity o nanocrystalline

silver in a porcine contact dermatitis model. Nanomedicine NB. 2008;4:241-251.

5. Wright JB, Lam K, Buret A, et al. Early healing events in a porcine model o contaminated

wounds: eects o nanocrystalline silver on matrix metalloproteinases, cell apoptosis, and

healing. Wound Repair Regen. 2002;10:141-145.

6. Stevens AM, Johnson FC. A new eruptive ever associated with stomatitis and ophthalmitis.

Report o two cases in children. Am J Dis Child. 1922;526-533.

7. Bastuji-Garin S, Fouchard N, Bertocchi M, et al. SCORTEN: a severity-o-illness score or

toxic epidermal necrolysis. J Invest Dermatol. 2000;115(2):149-153.

8. Dalli RL, Kumar R, Kennedy P, et al. Toxic epidermal necrolysis/Stevens-Johnson syndrome:

current trends in management. ANZ J Surg. 2007;77:671-676.

9. Schneck J, Fagot JP, Sekula P et al. Eects o treatments on the mortality o Stevens-

Johnson syndrome and toxic epidemal necrolysis: a retrospective study on patients

included in the prospective EuroSCAR Study. J Am Acad Dermatol. 2008;58(1):33-40.

10. Pehr K. The EuroSCAR study : cannot agree with the conclusions. J Am Acad

Dermatol. 2008;59(5):898-899; author reply 899-900.

11. Hebert AA, Bogle MA. Intravenous immunoglobulin prophylaxis or recurrent Stevens-

Johnson syndrome. J Am Acad Dermatol. 2004;50(2):286-288.

12. Brown KM, Silver GM, Halerz M, et al. Toxic epidermal necrolysis: does immunoglobulin

make a dierence? J Burn Care Rehabil. 2004;25:81-88.

13. Brett AS, Philips D, Lynn AW. Intravenous immunoglobulin therapy or Stevens-Johnson

syndrome. South Med J. 2001;94(3):342-343.

14. French LE, Trent JT, Kerdel FA. Use o intravenous immunoglobulin in toxic epidermal

necrolysis and Stevens-Johnson syndrome: our current understanding. Int

Immunopharmacol. 2006;6(4):543-549.

15. Sotozono C, Ueta M, Kinoshita S. Systemic and local management at the onset o Stevens-

Johnson syndrome and toxic epidermal necrolysis with ocular complications. Am J

Ophthalmol. 2010;149(2):354; author reply 355.

16. Araki Y. Successul treatment o Stevens-Johnson syndrome with steroid pulse therapy at

disease onset. Am J Ophthalmol. 2009;147(6):1004-1011.

17. Koh MJ, Tay YK. Stevens-Johnson syndrome and toxic epidermal necrolysis in Asian

children. J Am Acad Dermatol. 2010;62(1):54-60.

18. Shammas MC, Lai EC, Sarkar J S, et al. Management o acute Stevens-Johnson syndrome

and toxic epidermal necrolysis utilising amniotic membrane and topical corticosteroids. Am

J Ophthalmol. 2010;149(2):203-213.

19. Tseng SC. Acute management o Stevens-Johnson syndrome and toxic epidermal

necrolysis to minimise ocular sequelae. Am J Ophthalmol. 2009;147(6):949-951.

20. Mockenhaupt M, Messenheimer J, Tennis P, et al. Risk o Stevens-Johnson syndrome and

toxic epidermal necrolysis in new users o antiepileptics. Neurology. 2005;64(7):1134-1138.

21. Cunha BA. Antibiotic side-eects. Med Clin North Am. 2001;85(1):149-185.

22. Namayanja GK, Nankya JM, Byamugisha JK, et al. Stevens -Johnson syndrome due to

nevirapine. Ar Health Sci. 2005;5(4): 338-340.

23. Logrena SM, Stevens WS, Bartlett JA, Crumpa JA. Epidemic Stevens-Johnson syndrome

in HIV patients in Guinea-Bissau: a side-eect o the drug-supply policy? Correspondence

AIDS. 2010,24:783-787.

24. Metry DW, Lahart CJ, Farmer KL, Herbert AA. Stevens-Johnson syndrome caused by the

antiretroviral drug nevirapine. J Am Acad Dermatol. 2001;44(2 Suppl):354-357.

25. Fagot JP, Mockenhaupt M, Bouwes-Bavinck JN, et al. Nevirapine and the risk o Stevens-

Johnson syndrome or toxic epidermal necrolysis. AIDS. 2001;15:1843-1848.

26. De Maat MM, Ter Heine R, Mulder JW, et al. Incidence and risk actors or nevirapine-

associated rash. Eur J Clin Pharmacol. 2003;59(5-6): 457-462.

27. Morel E, Escamochero S, Cabanas R, et al. CD94/NKG2C is a killer eector molecule inpatients with Stevens-Johnson syndrome and toxic epidermal necrolysis. J Allergy Clin

Immunol. 2010;125(3):703-710.

28. Nassi A, Moslehi H, Le Gouvello S, et al. Evaluation o the potential role o cytokines in toxic

epidermal necrolysis. J Invest Dermatol. 2004;123:850-855.

29. Xu J, Liu H, Wu Y, et al. Proapoptotic eect o metalloproteinase 9 secreted by trophoblasts

on endothelial cells. J Obstet Gynaecol Res. 2011;37(3):187-194. Epub 2010.

30. Fuyuan Qiao Kiaei M, Kipiani K, Calingasan NY, et al. Matrix metalloproteinase-9 regulates

TNF-alpha and FasL expression in neuronal, glial cells and its absence extends lie in a

transgenic mouse model o amyotrophic laterial sclerosis. Exp Neurol. 2007;205(1):74-81.

31. Dunn K, Edwards-Jones V, editors. The role o Acticoat with nanocrystalline silver in the

management o burns. Burns. 2004;30(Suppl 1).

32. Wright JB, Lam K, Hansen D, Burrell RE. Efcacy o topical silver against ungal burn wound

pathogens. Am J Inect Control. 1999;27:344-350.

33. Yin HQ, Langord R, Burrell RE. Comparative evaluation o the antimicrobial activity o

ActicoatTM antimicrobial barrier dressing. J Burn Care Rehabil. 1999;20(3):195-200.

34. Wright JB, Lam K, Burrell RE. Wound management in an era o increasing bacterial antibiotic

resistance: a role or topical silver treatment. Am J Inect Control. 1998;26(6): 572-577.

35. Widgerow AD. Nanocrystalline silver, gelatinases and the clinical implications. Burns.

2010;36(7):965-974.

36. Ladwig GP, Robson MC, Liu R, et al. Ratios o activated matrix metalloproteinase-9 to tissue

inhibitor o matrix metalloproteinase-1 in wound uids are inversely correlated with healing

o pressure ulcers. Wound Repair Reg. 2002;10 26-37.

37. Rayment EA, Upton Z, Shooter GK. Increased matrix metalloproteinase-9 (MMP-activity

observed in chronic wound uid is related to the clinical severity o the ulcer. Br J Dermatol.

2008;158:951-961.

38. Reiss MJ, Han Y-P, Garner WL. a1-Antichymotrypsin activity correlates with and may

modulate matrix metalloproteinase-9 in human acute wounds. Wound Repair Regen.2009;17(3):418-26.

39. Johnston HL, Cuta F, Garrett AB. The solubility o silver oxide in water, in alkali and in alkaline

salt solutions. The amphoteric character o silver hydroxide. JACS. 1933;55:2311-2325.

40. Asz J, Asz D, Moushey R, et al. Treatment o toxic epidermal necrolysis in a pedia tric patient

with a nanocrystalline silver dressing. J Pediatr Surg. 2006:41,E9-E12.

41. Dillon CK, Lloyd MS, Dzeiwulski P. Accurate debridement o toxic epidermal necrolysis using

Versajet PBurns. 2010;36(4):581-584.

42. Whitaker LS, Worthington S, Jivan S, Phipps A. The use o Biobrane by burn units in the

United Kingdom: a national study. Burns. 2007;33(8):1015-1020.

100-602-1-pb

Documents