Full Text
Full Text
Full Text
Author(s)
Citation
Issued Date
URL
Rights
2012
http://hdl.handle.net/10722/173742
By
Declaration
I, Sinn Ting Ting, Maria , declare that this dissertation represents my own work and
that it has not been submitted to this or other institution in application for a degree,
diploma or any other qualifications.
I, Sinn Ting Ting , Maria also declare that I have read and understand the guideline
on What is plagiarism? published by The University of Hong Kong (available at
http://www.hku.hk/plagiarism/) and that all parts of this work complies with the
guideline.
Table of Contents
Page
1. Abstract
2. Introduction
4. Results
10
5. Discussion
21
6. Conclusion
31
7. Appendix
32
8. References
33
Abstract
Background
Necrotizing fasciitis(NF) is a severe form of soft tissue infection that primarily
involves the superficial fascia. The purpose of this study is, by reviewing all the NF
cases in our hospital, to see if any areas that can be done to optimize the outcome of
this group of patients.
Objective
To study the clinical features, risk factors, spectrum of organisms, treatment and
outcomes of necrotizing fasciitis cases which were admitted to Tseung Kwan O
Hospital.
Methodology
A retrospective study including 28 patients who were discharged with the diagnosis of
Necrotizing Fasciitis (NF) or Fourniers Gangrene (FG) in Tseung Kwan O
Hospital from June 2010 to May 2012 were recruited. Data regarding co-morbidities,
laboratory variables, micro-organisms and surgical treatment were collected. Fishers
exact test and Mann-Witney U test were utilized for comparing variables.
Results
A total of 27 patients but 28 episodes were included in this study. The mean age of
hospital survivors was 56.6 years(17.9 years), while for the non-survivors was 60.6
years(20.6 years). Diabetes mellitus (35.7%) and hypertension (46.4%) were the
most frequent co-mobidities. Most (16, 57.1%) of them were in stage 1 NF, five
(17.9%) in stage 2 and seven (25%) in stage 3. Less than half (39.3%) were febrile on
admission. Fifty percent had one lower limb involvement on presentation, 32.1% had
upper limb involvement and the rest (17.9%) were central NF, ie Fourniers gangrene
(FG), primarily involve the perineum on admission. Among the 23 NF cases, 39.1 %
grew vibrio vulnificus, 21.7% grew streptococcus pyogenes ( in which one of them
grew mixed streptococcus pyogenes and E.coli), 17.4% grew mixed flora. Among the
5 FG cases, all but one grew mixed flora. Of the 22 extremities NF cases who
underwent operation, 10(45.5%) had amputation done. Mean number of operation was
4.1( range 1-16). The mean length of hospital stay for survivors were significantly
longer than that of non-survivors (54.15 days [SD 49.66] vs 4.11 days[SD 4.11] ,
p <0.001). The overall hospital mortality was 32.1%. The presence of septic shock,
disseminated intravascular coagulation and acute kidney injury were significantly
associated with mortality.
Conclusion
Necrotizing fasciitis runs a rapid deteriorating course and is associated with high
morbidity and mortality. The success in management requires prompt diagnosis, early
and aggressive surgical debridement, as well as appropriate antibiotics.
Background
Objective:
To study the clinical features, risk factors, spectrum of organisms, treatment and
outcomes of necrotizing fasciitis cases which were admitted to Tseung Kwan O
Hospital.
Design:
This is a retrospective study.
Methods:
A computer-based search in the Computer Management System was utilized to
identify patients who were discharged with the diagnosis of Necrotizing Fasciitis
(NF) or Fourniers Gangrene (FG) in Tseung Kwan O Hospital from June 2010 to
May 2012.
Diagnosis was defined by either specific physical signs( for example, rapidly
progressive soft tissue inflammation with necrosis, bullae or gas formation in the deep
tissue) , or clinical suspicion( such as rapidly progressive tissue inflammation but
without necrosis, bullae or gas) further confirmed during operation or by biopsy
findings. Characteristic CT findings were accepted if operation was not done.
Demographic data collected were their sex, age, social habits (smoking ,drinking, and
intravenous drug abuse ) . Comorbidities in particular history of diabetes mellitus
hypertension and other cardiovascular diseases, liver and renal diseases, malignancies,
gout, as well as the use of steroids or immunosuppressants were recorded. Etiology,
site of infection on admission and the subsequent area involved were recorded. The
presentation on admission, including temperature (fever is defined by temperature
38.3 OC, hypothermia < 36 OC , normothermia is the temperature in between) ,
presence or absence of shock (systolic blood pressure < 90 mm Hg) , the presence of
pain, the condition of the lesions( presence of erythema, swelling, tenderness, warmth,
crepitus, blister/ bullae, haemorrhagic bullae, skin anesthesia, gangrenous changes,
ulcer or purulent/serous discharge ) as well as the stage of disease were recorded. The
definitions of hypotension, fever and hypothermia were classified according to the
2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference.
(3)Whereas, the stage of disease is defined according to the scheme adopted by Wong
and Wang (4): table 1
Stage 2 (Intermediate)
Blister or bullae (serous
fluid)
Skin fluctuance
Skin induration
Stage 3 (Late)
Hemorrhagic bullae
Skin anesthesia
Tissue crepitus
Skin necrosis with dusky
discoloration progressing
to frank gangrene
The time from onset of symptoms to admission was recorded. The time of operation
was defined by the duration between admission and the time of operation. The time of
diagnosis was defined as the duration between admission and the time when diagnosis
of NF was highly suspected, for example when the surgeon was consulted for
suspected NF, or the operation was arranged for NF. The initial admission diagnosis
was also recorded.
The month of admission and the initial specialty on admission were recorded.
Concerning the investigations, laboratory data like complete blood picture, renal and
liver function, clotting, C-reactive protein, glucose level were collected. Any presence
of gas in XR was recorded. The need of CT scanning to make the diagnosis or
delineate the extent of involvement was also recorded.
Concerning the treatment of NF, the type of empirical antibiotics used prior to the
diagnosis NF and the definitive antibiotics for NF were recorded. The number of
operations and any amputation done were recorded.
Microbiological cultures of tissue were obtained at the time of operation. The
microbiological and blood culture results were recorded.
Data of critical care support included time of ICU admission, length of ICU stay, the
need of inotropes, mechanical ventilation and dialysis were recorded.
Statistical Analyses
Statistical analyses were performed using Statistical Package for the Social Sciences
software(SPSS), version 16.0. Continuous variables were expressed as mean with
standard deviation ( SD), and they were compared using the Mann-Witney U test.
Fishers exact test was utilized for comparing categorical variables. Tests were twotailed with a significance level of 0.05.
Results:
Table 1: Ages
All
Patients
Parameter
AGE
28
Nonsurvivors
Mean(SD)
57.85(18.49)
Survivors
Mean(SD)
Mean(SD)
p Value
60.55(20.6)
19
56.57(17.86)
0.69
Table 2
No. of patients
Nonsurvival
N (%)
60.55(20.6)
0.69
Male
19
6(31.5)
>0.09
Female
3(33.3)
Yes
2(40)
No
13
1(7.6)
Yes
2(50)
No
15
3(20)
Yes
10
4(40)
No
18
5(27.7)
Yes
3(42.8)
No
20
6(30)
Yes
0(0)
No
26
9(34.6)
Yes
13
5(38.4)
No
15
4(26.6)
Yes
1(33.3)
No
25
8(32)
Yes
0(0)
No
27
9(33.3)
Yes
0(0)
No
27
9(33.3)
Yes
0(0)
No
26
9(34.6)
Yes
14
3(21.4)
No
14
6(42.8)
Age
Sex
Smoker
Drinker
DM
Liver diseases
Renal diseases
HT
Malignancies
Gout
Use of steroid
Immunosuppressants
Predisposing
wound/injury
p Value
0.172
0.272
0.677
0.653
>0.09
0.689
>0.09
>0.09
>0.09
>0.09
0.42
Most (16, 57.1%) of them were in stage 1 NF, five (17.9%) in stage 2 and seven(25%)
in stage 3. The most consistent physical finding was swelling (25, 92.6%, n=27).
Erythema was present in 20 patients( 87.0%, n= 23). Blisters/ bullae were uncommon
(6 patients, 21.4%), so as gangrenous changes (6 patients, 21.4%), tissue crepitus(1
patient, 3.6%) and haemorrhagic bullae formation(3 patients, 10.7%).
Hypotension was present in 7 (25%) patients upon admission, but 9 more patients
(total 16, 57.1%) developed shock during the subsequent hospital course. Table 3
Percentage %
Erythema (n=23)
20
87
Swelling (n=27)
26
96.2
Ulcer (n=28)
21.4
10
35.7
Tenderness (n=21)
21
100
Warmth (n=12)
58.3
Crepitus (n=28)
14.2
blisters/bullae (n=28)
21.4
14.2
Anesthesia (n=28)
3.6
17.8
stage of disease: 1
16
57.1
17.9
25
Temp : Hypothermia
17.9
Normothermia
12
42.9
Hyperthermia
Hypotension on admission
(n=28)
11
39.2
25.0
p/s discharge : purulent /serous discharge; n might not equal to 28 because the
signs/symptoms were not documented in case notes.
The initial diagnoses were mostly cellulitis(11 patients, 39.3% , n= 28).Table 4 The
diagnosis of NF was already suspected on admission in 4 patients (17.4%, n= 23),
whereas that of fourniers gangrene in 2 (40%, n=5). The other common diagnosis
was abscess (4 patients, 14.3%). One patient had recurrent carcinoma(CA) of rectum
and presented with buttock pain. The admission diagnosis was bone pain secondary to
the carcinoma. It was only when the patient developed fever and shock, the lower
limb already progressed to stage 3 NF with haemorrhagic bullae, crepitus, gangrenous
change on the 4th day of admission, a CT was done and the diagnosis of NF was
confirmed.
Out of the 23 NF cases, 14 (60.9%) were admitted to orthopaedics, 7 (30.4%)to
medical, 1 (4.3%)to surgical and 1(4.3%) to paediatrics. All of the 5 FG cases were
admitted to surgical ward. Or put it in another way, out of the 7 stage 3 NF, 6 were
admitted to the orthopaedic ward, only 1 was admitted to the medical ward. Of the 4
stage 2 NF cases, all were admitted to the orthopaedic ward.
n
2
7
2
1
2
2
1
2
2
1
1
3
2
White blood cell count was normal in 8 (28.6%) cases and low in 1 (3.6%) case, while
the rest showed raised count. Table 5
Non-Survival
Survival
Mean(SD)
Mean(SD)
Mean(SD)
p-value
WBC (cells/mm )
28
15.18(7.96)
11.45(6.48)
19
16.94(8.13)
0.09
Hb (dL)
28
12.13(2.97)
12.62(2.26)
19
11.91(3.29)
0.68
28 203.17(111.98)
178.55(140.98)
19
214.84(97.61)
0.24
CRP (mg/L)
106.51(89.99)
101.93(97.69)
0.62
Cr (umol/L)
28 189.25(223.44)
177.88(88.76)
19
194.63(267.01)
0.23
Na (mmol/L)
28
134.21(4.96)
133(3.9)
19
134.78(5.4)
0.39
Albumin (g/dL)
27
33.85(9.07)
30.62(7.32)
19
35.21(9.56)
0.28
Non-Survival
Survival
Mean(SD)
Mean(SD)
Mean(SD)
p-value
Onset of Symptoms
to admission (day)
28
3.53(5.03)
3.25(4.43)
19
3.66(5.4)
0.74
Hrs to OT from
admission(hr)
27
27.11(46.55)
16.93(10.12)
19
31.39(55)
0.30
15
53.03(158.3)
16.56(10.58)
94.71(233.54)
0.16
Hrs to Diagnosis
of NF (hr)
27
67.44(135.57)
19.06(23.03)
19
87.81(157.68)
0.54
28
5.43(10.8)
3.02(4.72)
19
6.57(12.68)
0.22
Length of hospital
stay(day)
28
38.07(47.09)
4.11(4.7)
19
54.15(49.66)
<0.001
X-ray findings were helpful only in one case (3.6%) by showing subcutaneous gas
formation. In fact, tissue crepitation can be elicited clinically in that case. CT scan
imaging was done in 8 patients (28.6%) to make the diagnosis or to delineate the
extent of involvement.
Microbiological findings
Blood culture was done in 24 cases and of which, only six (25%) had positive growth.
In contrast, tissue cultures had a much higher yield. All 26 cases sent had positive
growth. Among the 23 NF cases, 9 grew vibrio vulnificus(39.1%), 5(21.7%) grew
streptococcus pyogenes ( in which one of them grew mixed streptococcus pyogenes
and E.coli), 4(17.4%) grew mixed flora, and one (4.3%) grew photobacterium
damsela. Among the 5 FG cases, all but one grew mixed flora, the remaining one
grew E.coli. Table 7
Among the 9 vibrio cases, one occurred in July and August respectively, three in
September, two in October and November respectively. Five (55.6%) of them
involved the upper limb and the remaining the lower limb. Only five (55.6%) of them
recalled history of injury: 3 were injured by fish, 1 by crab, one had lower limb injury
after a fall from stairs.
Comorbidities
Initial
site
Injury/
Extent
involve
wound
d
L arm,
chest ,
L hand
lower
/
jaw,RU
L
Organisms
photobacterium
damsela
Initial antibiotics
flagyl, ampicillin ,
cloxacillin
Definitive
antibiotics
Death
clindamycin,g
entamicin,cirp
ofloxacin,
vancomycin
Yes
E.coli, Klebsiella,
enterococcus,
bacteriodes fragilus,
corynebacterium
L hand
forearm crab
vibrio vulnificus
L ankle
L leg
vibrio vulnificus
L calf
vibrio vulnificus
L thigh
and
/
groin,
abdomen
GAS
zinacef, flagyl
pen G and
clindamycin
Yes
augmentin,
flagyl
Yes
rocephin,
ciprofloxacin,
minocycline
augmentin
ciprofloxacin,
cloxacillin
DM, anaemia of
L foot
chronic disease
M/47
F/88 DM
M/73
flagyl,
rocephin, ampicillin , timentin,
cloxacillin, flagyl
ciprofloxacin,
clindamycin
ampicillin, cloxacillin,
flagyl, clindamycin, rocephin,
rocephin,
ciprofloxacin
ciprofloxacin,
and doxycline
doxycycline
rocephin,
ampicllin,
ciprofloxacin
cloxacillin,flagyl,
and
levofloxacin
minoxycline
rocephin,
ampicllin, cloxacillin,
levofloxacin
pen G
and doxycline
F/29 /
M/39
recurrent CA
rectum for
conservative
management
R calf
R buttock and
thigh
F/42
hepatitis B
carrier
L hand
fish fin
vibrio vulnificus
M/84 HT
L leg
L foot
contusion
10
M/58 HT
L foot
and leg
pseudomonas
aeruginosa and
serratia species
11
F/66 anaemia
L hand
L
fish
forearm
vibrio vulnificus
L
fish bone
forearm
vibrio vulnificus
GAS
ampicillin, cloxacillin
12
13
CHF,AF,post
F/84 RAI
L hand
hypothyroidism
ESRF with renal
M/44 transplant done, R leg
HT, gout
rocephin,
ciprofloxacin,
minocycline, flagyl
ampicillin, cloxacillin,
flagyl
zinacef,cloxacillin
rocephin,
ampicillin, cloxacillin,
ciprofloxacin,
flagyl
minocycline
rocephin,
ampicillin, cloxacillin ciprofloxacin,
minocycline
14
M/59
HT, hepatitis B
L thigh
carrier
L
/
buttock
streptococcus
pneumoniae, ESBL
E.coli
15
M/57
HT, alcoholic
liver disease
L leg
contusion
vibrio vulnificus
16
M/36 /
R hand
abrasion
erysipelothrix
rhusiopathiae
17
F/52 HT
bilateral
feet
scald
GAS
18
M/49
L calf
L thigh /
vibrio vulnificus
augmentin,
klebsiella pneumoniae clindamycin,
ciprofloxacin
alcoholic liver
disease
19
M/75 DM
L thigh
L
buttock,
/
liver
abscess
20
HT,
F/68 hyperlipidaemia, L hand
obesiy
L
/
forearm
vibrio vulnificus
20
F/8
GAS
R thigh
insect bite
pen G and
clindamycin
claforan,
ampicillin, cloxacillin,
ciprofloxacin,
gentamicin, flagyl
minocycline
augmentin ,
zinacef, flagyl,
flagyl,
gentamicin
gentamicin
pen G,
augmentin,
pipercillin,
clindamycin,
amikan,
cloxacillin
vancomycin
cloxacillin
Yes
Yes
Yes
Yes
cloxacillin
Yes
rocephin
R hand
HT, hepatitis B
and
carrier
forearm
20
M/59
20
HT,
F/73 hyperlipidaemia, R foot
pseudogout
20
20
20
20
20
M/47 DM
R
perianal
DM,HT, IHD,
ESRF on HD,
R
M/56
CA thyroid with scrotum
OT done, gout
R groin,
M/56 DM
L chest
wall
perianal
CA rectum, post
M/62
and
chemotherapy
scrotum
DM,
perianal
HT,hepatitis B
M/87
and
liver cirrhosis,
scrotum
beta-thal trait
R leg
abrasion
bilateral
posterior
/
thigh,
back
perineu
/
m
R thigh /
upper
thigh
cloxacillin,
clindamycin
rocephin,
flagyl ,
levofloxacin
tienam
strep.anginosus/milleri
, bacteroides fragilis, meropenem
E.coli
meropenem
strep.anginosus/milleri
, bacteroides fragilis, augmentin
klebsiella
strep.anginosus/milleri
, bacteroides fragilis, augmentin
E.coli
augmentin,
clindamycin,
amikan
augmentin,
cloxacillin ,
flagyl
E.coli
augmentin,
flagyl,
levofloxacin
stap. Aureus
ampicillin, cloxacillin
proteus,
strep.anginosus/milleri rocephin, clinamycin,
, bacteroides fragilis, levofloxacin
enterococcus
streptococcus
anginosus, ESBL
zinacef, flagyl
E.coli,
peptostreptococcus
augmentin, flagyl
Surgical treatment
Apart from the afore-mentioned case of recurrent CA rectum, all patients (96.4%)
underwent operation. None of the FG or central NF cases received amputation. Of the
22 NF cases who underwent operation, 10(45.5%)had amputation done. Mean number
of operation was 4.1( range 1-16).
Critical care support and complications
ICU admission was necessary in 15 (53.6%) patients. Inotropes were required in 18
(64.3%) patients. Fifteen(53.6%) patients required mechanical ventilation. Thirteen
(46.4%) of them had DIC and 14 (50%) had acute kidney injury. Eight (28.6%) of
them undergone dialysis. The presence of septic shock, DIC and AKI were
significantly associated with mortality. Table 8
Table 8: Complications
Septic shock
DIC
AKI
Compartment syndrome
No. of patients
n
16
13
14
1
Nonsurvival
n (%)
9(56.2)
9(69.2)
9(64.2)
1(100)
p Value
0.002
<0.01
0.001
0.321
Yes
GIB
pneumonia
ARDS
4
2
2
1(25)
0(0)
2(100)
>0.09
>0.09
0.087
Outcome
The overall hospital mortality was 32.1%( n=9). Among the 23 NF cases, hospital
mortality was 34.8% (n=8) . Among the remaining 5 FG cases, hospital mortality was
20%(n= 1). Among the 19 survivors, 4 were transferred to other hospitals ( 3 to
rehabilitation hospitals, 1 to renal unit of the hospital patient used to attend), the
remaining were discharged home. Six of them were discharged with limb loss.
Discussion
Necrotizing fasciitis is a surgical emergency. Even for virulent organism as vibrio
vulvificus, mortality rate decreases from 23.0 % to 4.9% if surgical debridement and
fasciotomy are done within 24 hours of onset of symptoms (6). Therefore, early
diagnosis, early and aggressive debridement or amputation, coupled with appropriate
antibiotics is essential for survival. However, the diagnosis of NF is notoriously
difficult to make. The main differential is cellulitis of which the infection begins at the
junction between the dermis and superficial fascia. Whereas the primary pathology of
NF is at the superficial fascia. Proliferation of bacteria results in angio-thrombotic
microbial invasion and liquefactive necrosis of the superficial fascia. Later, occlusion
of the perforating vessels to skin caused skin ischemia and subsequent necrosis of
subcutaneous fat, dermis and epidermis, resulting in bullae formation, ulceration and
gangrenous changes. Clinically the skin changes can be categorized into stages as
proposed by Wong and Wang(see Table 1 )(4). It is hard to differentiate stage 1 NF
with cellulitis when there is only erythema, swelling and tenderness. However, in NF,
the margins of tissue involvement are often poorly defined and tenderness often
extend beyond the apparent area of involvement. In other words, pain is often out of
proportion to the clinical appearance. Also, lymphangitis is rarely seen in NF. In stage
2, blister or bulla formation already signifies the onset of critical skin ischemia. It is
rarely seen in cellulitis and its presence should ring the bell for the diagnosis of NF. In
stage 3, the presence of tissue necrosis is manifested by the appearance of
hemorrhagic bullae, skin anesthesia and gangrene. In our cohort, 3 and 2 of the
patients with stage 3 and stage 2 diseases respectively were diagnosed to have
cellulitis on admission. The use of this clinical staging can heighten our alertness
towards the diagnoses of more serious soft tissue infection rather than just cellulitis,
and thereby arrange surgical debridement earlier. For those with stage 1 disease, if
serial clinical monitoring reveals progression towards higher stage despite the use of
antibiotics, early surgical exploration is warranted. Luckily in our cohort , all but one
of the stage 2 and 3 NF cases were triage under the specialty of orthopaedic, thereby
preventing further delay in surgical treatment.
Besides clinical appearance, a bedside procedure finger test can be performed that
aids diagnosis. A 2-cm incision down to the deep fascia is made under local
anaesthesia. Then the level of superficial fascia is probed with a gloved finger. Lack
of bleeding, foul-smelling dishwater pus and minimal tissue resistance to finger
dissection represent a positive finger test and are diagnostic of NF (7). However, this
procedure is seldom performed in our hospital.
Most of the patients with NF had some chronic illnesses such as diabetes mellitus,
alcohol abuse, or renal impairment. Among these, the presence of diabetes and
immunocompromised state were associated with mortality (8,9). Diabetes was present
in 35.7 % in our cohort and only 10 % enjoyed good past health. However, our study
was underpowered to detect any association between co-morbidities and death.
Fever was present in only 39.3 % of this cohort, therefore it per sc is not a sensitive
marker for NF. Hypotension on admission was present in 25 % of them. The result
was consistent with the review by Wong et al of 89 consecutive patients, with fever
only present in 53 % and hypotension in 18 % at presentation, especially in
immunocompromised patients such as those diabetics (10). Therefore these patients
might appear systemically quite well at least on presentation.
Concerning the laboratory data, raised white cell count (raised in 67.9% of patients)
was not a sensitive marker for NF.
Though a low admission serum sodium level( < 135 mEq/L) was not associated with
hospital mortality in this study, it was not so in studies like that of Arezou et al (11).
There are some possible mechanisms. Firstly, sepsis leads to increase muscle glucose
uptake, increase in ratio of muscle membrane permeabilities to Na+ and K+ and
increase intracellular Na+ concentration, mediated by complement activation.
Secondly, sepsis is associated with an increase in antidiuretic hormone and with
adrenal insufficiency. Thirdly, severe sepsis can induce marked third spacing of fluids,
which may be replaced by free water. All these can give rise to hyponatremia. In fact,
hyponatremia was one of variables in the LRINEC( Laboratory Risk Indicator for
Necrotizing fasciitis) score developed by Wong et al to discriminate between
necrotizing soft tissue infection from its non-necrotizing counterpart (12) [Appendix
1] . A score of 6 has a positive predictive value of 92.0% and negative predictive
value of 96.0% for the diagnosis of NF. However, the turnaround time of C-reactive
protein (CRP) was too long in our hospital to render it useful in making the diagnosis.
In fact, only 7 of our patients had their CRP checked on admission. The LRINEC
score can only be calculated in 5 of them because of missing data( no admission
glucose) and the scores are 0,0, 3,6 and 8. Therefore, the sensitivity for the diagnosis
of NF was only 33.3%.
In this cohort, Vibrio vulnificus was the most common culprit (32.1%). Vibrio species
are natural inhabitants of brackish water and salt water worldwhile. For vibrio
vulnificus, a water salinity of 0.7% to 1.6% and warm temperatures(>20 OC) facilitate
its growth (13). As these organisms are ubiquitous in the coastal waters at the
temperate zones, they tend to cause illness in the warmer months of the year when
their concentrations become high enough. Since the mean sea surface temperature is
26 OC during the summer months in Hong Kong, the disease should be more prevalent
in summer (14). The bacterium is frequently found in oysters, crustaceans and
shellfish. Up to 50% of oysters and 11 % of crabs were cultured positive for vibrio
vulnificus during summer months (15) . Human infection can be acquired through
contamination of wound by these marine organisms or through consumption of
contaminated raw or undercooked seafood. In our cohort, patients were infected
between July to November, in fact most of them occurred between September to
November which was not quite summer already. This might reflect global
rewarming resulting in lengthening of the summer months. Four out of 5 of the upper
extremities NF cases had contact history of crab/ fish, whereas one of the 4 lower
extremities NF cases had history of seafood contact.
Most patients with vibrio vulnificus infections are immunocompromised and they
mostly suffer from chronic liver disease, diabetes, adrenal insufficiency, malignancies,
haemochromatosis, thalassaemia ,chronic renal insufficiency or alcoholism (14). In
our cohort, 3 out of the 9 patients got diabetes, one got alcoholic liver disease and one
was a hepatitis B carrier. As v. vulnificus is endemic in this locality, wearing globes is
In this cohort, the second most prevalent organism was streptococcus pyogenes. Four
patients were lower limb NF and the remaining one involved the hand. Three( 60%)
had history of injury and the mortality was 40%. In the report by Brown et al, almost
40 % of those with invasive group A streptococcal (GAS)infection had no
predisposing illnesses or risk factors and in up to 45 % of cases there is no direct
injury (20). The most prevalent types of GAS in Hong Kong were M1(15%) and
M12(21%) (21). They were usually found in skin and throat isolates and commonly
associated with invasive disease (64 % of cases). The virulence of M1 type may be
related to its ability to enhance adhesion of streptococci to tissues and thus preventing
phagocytosis by neutrophils. Moreover, patients with invasive GAS disease had lower
level of antibodies against M-protein and superantigens than those with noninvasive
disease, implying poor humoral immunity against GAS virulence factors (22). The
mortality of GAS induced NF was around 20% (23).
All but one of the patients underwent operation and 19 (70.4%) were done within 24
hours after admission. In the study by Wong et al, a delay in surgery of greater than
24 hours was associated with an increase risk of death (relative risk = 9.4, p<0.05)
(10). In fact, multiple studies have shown the single most common factor associated
with increased mortality was delay to operative debridement (24,25,26,27). Other
parameters that have been associated with increase mortality include age > 50 years,
extent of infection, lactic acidosis, degree of organ dysfunction at admission,
hypotension, immune compromise, and a white cell count > 30,000/mm3
(11,24,25,26,27,28). In this study, the presence of septic shock, AKI and DIC were
significantly associated with mortality. However, the mean time to operation was
paradoxically longer in those survivors (31.39 hr vs 16.93 hrs in non-survivors),
though the difference was not statistically significant. The reason was that some
patients with less virulent organisms had a much longer time to operation but better
survival.
Concerning the antibiotics therapy, it is best to start with a combination of broadspectrum antibiotics in those with necrotizing soft-tissue infections. It is refined later
based on culture results and clinical response. According to the IMPACT guideline
The usefulness of hyperbaric oxygen (HBO) therapy has been conflicting (32). None
of our patients has been sent to the Recompression Treatment Centre on Stonecutters
Island, where the location rendered it too risky to send our most haemodynamically ill
patients for this therapy.
In our cohort, 53.6% of the patients required ICU care. The rate was much lower than
that from another local study which was 93% (7). However, the amputation rate was
similar, ours was 45.5% and theirs was 46%.
It has been found that an average of 3 debridements, spaced 12 to 36 hours apart, is
needed to control gross infection (10,25,33). In this study, the mean number of
operation was 4.1 ( range 1-16).
Conclusion
Necrotizing fasciitis runs a rapid deteriorating course and is associated with high
morbidity and mortality. The success in management requires prompt diagnosis, early
and aggressive surgical debridement, as well as appropriate antibiotics. Preventive
measures include use of gloves in handling seafood especially in
immunocompromised patients. The public awareness of this condition should be
heightened and seeking medical advice early can certainly improve the outcome.
Score
0
4
0
1
2
0
1
2
0
2
0
2
0
1
References
10. Wong CH, Chang HC, Pasupathy S, et al. Necrotizing fasciitis: clinical presentation,
microbiology, and determinants of mortality. J Bone Joint Surg Am. 2003; 85(8):
1454- 1460.
11. Arezou Yaghoubian, Christian de Virgilio, Christine Dauphine, et al. Use of admission
serum lactate and sodium levels to predict mortality in necrotizing soft-tissue
infections. Arch Surg 2007; 142(9): 840-846.
12. Wong CH, Khin LW, Heng KS, et al. The LRINEC (Laboratory Risk Indicator for
Necrotizing Fasciitis) score: a tool for distinguishing necrotizing fasciitis from other
soft tissue infections. Crit Care Med 2004; 32(7)1535-1541.
13. Renato Finkelstein, Ilana Oren. Soft tissue infections caused by marine bacterial
pathogens: epidemiology, diagnosis, and management. Curr Infect Dis Rep 2011;
13:470-477.
14. Vincent Hau, CO Ho. Necrotizing fasciitis caused by Vibrio vulnificus in the lower
limb following exposure to seafood on the hand. Hong Kong Med J 2011; 17:335-7.
15. Chiang SR, Chuang YC. Vibrio vulnificus infection: clinical manifestations,
pathogenesis, and antimicrobial therapy. J Microbiol Immunol Infect 2003; 36:81-8.
16. Dechet AM, Yu PA, Koram N, et al. Nonfoodborne Vibrio infections: an important
cause of morbidity and mortality in the United States, 1997-2006. Clin Infect Dis
2008;46:970-6.
17. Bisharat N, Agmon V, Finkelstein R, et al. Clinical, epidemiological, and
microbiological features of Vibrio vulnificus biogroup 3 causing outbreaks of wound
infection and bacteraemia in Israel. Israel Vibrio Study Group. Lancet 1999; 354:1421-
4.
18. Howard RJ, Bennett NT. Infections caused by halophilic marine Vibrio bacteria. Ann
Surg 1993;217:525-31.
19. Goodell KH, Jordan MR, Graham R, et al. Rapidly advancing necrotizing fasciitis
caused by Phytobacterium( vibrio) damsela: A hyperaggressive variant. Crit Care Med
2004;32:278-281.
20. Brown CN, Pollard TCB, Iyer S, et al. Invasive group A streptococcal infection. An
update on the epidemiology and orthopaedic management. J Bone Joint Surg 2010;92B:763-9.
21. Ho PL, Johnson DR, Yue AW, et al. Epidemiologic analysis of invasive and
noninvasive group A streptococcal isolates in Hong Kong. J Clin Microbiol
2003;41:937-42.
22. Sangeeta Mehta, Allison McGeer, Donald EL, et al. Morbidity and mortality of
patients with invasive group A streptococcal infections admitted to the ICU. Chest
2006;130: 1679-1686.
23. Mulla ZD. Treatment options in the management of necrotizing fasciitis caused by
Group A streptococcus. Expert Opin Pharmacother 2004;5:1695-700.
24. McHenry CR, Piotrowski JJ, Petrinic D, et al. Determinants of mortality for
necrotizing soft-tissue infections. Ann Surg 1995; 221:558-565.
25. Elliot DC, Kufera JA, Myers RA. Necrotizing soft tissue infections. Risk factors for
mortality and strategies for management. Ann Surg 1996; 224:672-683.
26. Anaya DA, McMahon K, Nathens AB, et al. Predictors of mortality and limb loss in