International Journal of Infectious Diseases (2009) 13, 255—260

http://intl.elsevierhealth.com/journals/ijid

Necrotizing fasciitis in southeast Taiwan:

clinical features, microbiology, and prognosis
Ming-Jong Bair a, Hsin Chi a,b,c,*, Wei-Sheng Wang d, Ya-Chun Hsiao a,
Ruey- An Chiang e, Kuan-Yu Chang e

a
Department of Internal Medicine, Mackay Memorial Hospital, Taitung Branch, Taipei, Taiwan
b
Department of Pediatrics, Mackay Memorial Hospital, 92, Section 2, Chung Shan North Road, Taipei 10449, Taiwan
c
Mackay Medicine, Nursing and Management College, Taipei, Taiwan
d
Department of Infectious Disease, Mackay Memorial Hospital, Taipei, Taiwan
e
Department of Surgery, Mackay Memorial Hospital, Taitung Branch, Taipei, Taiwan

Received 30 January 2008; accepted 8 April 2008

Corresponding Editor: Craig Lee, Ottawa, Canada

KEYWORDS Summary
Liver cirrhosis; Objectives: To determine the spectrum of microorganisms, clinical features, and risk factors of
Monobacterial; necrotizing fasciitis in southeast Taiwan.
Risk factor; Methods: We retrospectively studied patients diagnosed with necrotizing fasciitis and fully
Streptococcus pyogenes treated in our hospital for the period January 1995 to December 2006.
Results: The mean age of the patients was 58.2  14.2 years. The affected anatomical sites were
primarily peripheral (91 patients, 85.8%). Sixty patients (56.6%) had a type 1 infection, 17
patients (16.0%) had type 2, and eight patients (7.5%) had type 3. Diabetes mellitus was the most
common comorbidity. A single pathogen was identified as the infectious agent in 64 patients
(60.4%), multiple pathogens were identified in 21 patients (19.8%), and no organism was
identified in 21 patients (19.8%). Streptococcus pyogenes was the most common pathogen.
The average hospital stay was 28.0  23.1 days. Patients received a mean of 2.3  1.2 debride-
ments, and five patients (4.7%) eventually underwent an amputation. The overall mortality was
17.0%. Predictors of mortality included advanced age, class C liver cirrhosis, ascites, higher serum
creatinine, and lower hemoglobin and platelet levels.
Conclusions: Monobacterial infections are more common in our patients. Accurate early diag-
nosis and extensive surgical debridement are essential for a favorable outcome.
# 2008 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.

Introduction

* Corresponding author. Tel.: +886 2 25433535; Necrotizing fasciitis is an uncommon life-threatening soft
fax: +886 2 25433642. tissue infection, which spreads rapidly and progressively
E-mail address: chi.4531@ms1.mmh.org.tw (H. Chi). along the deep fascia.1 It was named ‘necrotizing fasciitis’

1201-9712/$36.00 # 2008 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijid.2008.04.015
256 M.-J. Bair et al.

in 1952, from necrosis, which means death of a portion of Statistical Package for the Social Sciences (SPSS), version
tissue, and fascia, which refers to the fibrous tissues that 12.0.
enclose and connect the muscles.2 The predisposing condi-
tions are mainly diabetes mellitus, alcoholism, end-stage Results
renal disease, malignancy, liver cirrhosis, malnutrition, gout,
corticosteroid use, and trauma.3—6 Diagnosis may be delayed
A total of 112 patients diagnosed with necrotizing fasciitis
because of the initial minimal skin cellulitis, even as the
were reviewed; six patients were excluded due to their
subcutaneous tissue suffers extensive necrosis and pro-
transfer to another hospital. The remaining 106 patients
nounced systemic toxicity develops.7 The progression of
(77 men, mean age of 58.2  14.2 years) were included in
the disease is fulminant and the mortality is high, despite
this study (Table 1). The mean interval between diagnosis and
effective antibiotic treatment.8,9 The prognosis depends on
first operation was 60.4  55.6 hours. Most patients (84.9%)
accurate diagnosis and immediate treatment. In addition to
had documented comorbid conditions or risk factors; the
delay in diagnosis and treatment, several risk factors, such as
most common was diabetes mellitus in 37 patients (34.9%).
more than one underlying condition, thrombocytopenia,
Other risk factors included alcoholism (29 patients (27.4%))
anemia, and older age, have been shown to influence mor-
and smoking (28 patients (26.4%)). The affected location was
tality.10 Higher mortality in patients with streptococcal toxic
peripheral in 91 patients (a lower limb in 71 and an upper limb
shock syndrome and in disease associated with certain bac-
in 20) and central in 15 patients (trunk in seven, head and
teria has been reported.11—13 The purpose of our study was to
analyze the risk factors and microbiological characteristics of
Table 1 Demographic data of patients with necrotizing
this infection, and to evaluate the determinants of mortality
fasciitis (N = 106)
associated with necrotizing fasciitis in southeast Taiwan.
n %
Patients and methods Age (years), mean  SD 58.2  14.2
Gender (male) 77 72.6
All patients who were admitted to Taitung Mackay Memorial
Race (aboriginal) 44 41.5
Hospital with a diagnosis of necrotizing fasciitis during the
period of January 1995 to December 2006 were studied. The
diagnoses were made via operation and based on the pre- Comorbidities and risk factors
sence of subcutaneous necrotic tissue extending through the No comorbidity 24 22.6
fascial planes. Histopathological examinations of surgical Diabetes mellitus 37 34.9
specimens were used to confirm the diagnoses when avail- Alcoholism 29 27.4
able. Blood and pus cultures of the tissue samples that had Smoking 28 26.4
been obtained at the time of first operative debridement Liver cirrhosis 29 27.4
were analyzed. Child—Pugh class A 8 27.6
The disease was classified into three types according to Child—Pugh class B 12 41.4
the culture results.1,7,14 Type 2 necrotizing fasciitis is that Child—Pugh class C 9 31.0
caused by Streptococcus pyogenes alone or in combination Cardiovascular disease 22 20.8
other bacteria, mostly with staphylococci. Type 3 is that Gout 14 13.2
caused by Vibrio and Aeromonas species; all others are Renal insufficiency 13 12.3
categorized as type 1. The affected anatomical sites were Malignancy 3 2.8
defined as either central (trunk and head and neck) or
peripheral (upper and lower limbs). A total of six kinds of
Affected location
comorbidity were recorded, including diabetes mellitus,
Peripheral 91 85.8
liver cirrhosis, cardiovascular disease, gout, renal insuffi-
Lower limbs 71 67.0
ciency, and malignancy. The severity of liver cirrhosis was
Upper limbs 20 18.9
classified according to the Child—Pugh score. Personal his-
Central 15 14.2
tory, including information on smoking, alcoholism, and
trauma, were also analyzed. The time lag before the first
debridement, the number of operations undergone, the need Type of infection
for amputation, the duration of hospitalization, and the Type 1 60 56.6
mortality rate were also studied. Type 2 17 16.0
Variables that were examined in the present study Type 3 8 7.5
included age, gender, location and type of infection, type
of comorbidity, and laboratory findings including causative Microorganisms
organisms. If the patients were transferred to another hos- Single pathogen 64 60.4
pital, they were excluded from the study. Multiple pathogens 21 19.8
Continuous values are expressed as mean  standard Negative culture 21 19.8
deviation (SD) and were compared using the Student’s t-test.
Categorical values are expressed as absolute and relative Amputation 5 4.7
frequencies and were compared using the Chi-square test or Duration of hospitalization 28.0  23.1
Fisher’s exact test. Tests were two-tailed with a significance Mortality 18 17.0
level of 0.05. Statistical analyses were performed using the
Necrotizing fasciitis in southeast Taiwan 257

Table 2 Comparisons between types of necrotizing fasciitis

Type 1 (N = 60) Type 2 (N = 17) Type 3 (N = 8)

n % n % n %
Age (years), mean  SD 59.9  14.0 a 52.6  15.6 52.3  11.3
>70 14 23.3 3 17.6 1 12.5

Gender
Male 41 68.3 15 88.2 8 100
Female 19 31.7 2 11.8 0 0

Race
Non-aboriginal 35 58.3 6 35.3 6 75
Aboriginal 25 41.7 11 64.7 2 25

Risk factor
Alcoholism 13 21.7 4 23.5 7a 87.5
Smoking 16 26.7 5 29.4 3 37.5
Ascites 9 15 3 17.6 2 25

Comorbidity
Diabetes mellitus 26 a 43.3
Liver cirrhosis 13 21.7
Child—Pugh class A 3 5
Child—Pugh class B 5 8.3
Child—Pugh class C 5 8.3
Renal insufficiency 8 13.3
Malignancy 2 3.3
2 11.8
6 35.3
1 5.9
2 11.8
3 17.6
3 17.6
0 0
1 12.5
5a 62.5
2 25
3 37.5
0 0
0 0
1 12.5

Affected location
Upper limb 6 10 5 29.4 3 37.5
Lower limb 42 70 12 70.6 5 62.5
Central 12 a 20 0 0 0 0
Mortality rate 8 13.3 5 29.4 0 0
Recurrence 7 11.7 0 0 1 12.5
Amputation 4 6.7 1 5.9 1 12.5
a
p < 0.05.

neck in six, and Fournier-like gangrene in two). Sixty patients
were categorized as having type 1 necrotizing fasciitis, 17
patients as having type 2, and eight patients as having type 3.
Comparisons between the types of necrotizing fasciitis are
summarized in Table 2. Patients with type 1 infection were
older than those with type 2 and type 3 ( p = 0.027). All type 2
and type 3 necrotizing fasciitis occurred in peripheral sites,
while central site involvement occurred most in type 1
infections ( p = 0.018). Diabetes mellitus occurred more in
type 1 patients ( p = 0.003), and neither liver cirrhosis nor
alcoholism occurred in type 3 patients.
A single organism was identified in 64 patients (60.4%).
Among these, Gram-positive bacteria were isolated in 39
patients and Gram-negative bacteria in 25 patients. Multiple
pathogens were identified in 21 patients (19.8%) and no
organism was isolated in 21 patients (19.8%). Microorganisms
are summarized in Table 3. S. pyogenes was the most common
single isolate, followed by methicillin-sensitive Staphylococ-
cus aureus (MSSA) in 10 patients and Escherichia coli in seven.
S. aureus and E. coli were isolated mostly in patients with
type 1 infections. Aeromonas spp was isolated in six patients
and Vibrio spp was isolated in two; all of them survived.
Members of the Enterobacteriaceae such as Morganella mor-
ganii and Serratia marcescens were isolated in the polymi-
crobial pathogens. Blood culture was positive in 25 patients
(23.6%); among these S. pyogenes was found in five patients
and Aeromonas hydrophila in four.
The average hospital stay was 28.0  23.1 days, ranging
from 1 to 139 days. In order to control the infective process,
patients received a mean of 2.3  1.2 debridements, ranging
from one to eight. Thirty-five patients received more than
three debridements. Amputation was performed to control
the infection in five patients (4.7%). Eighteen patients died
and the mortality rate was 17.0%. Four patients died within
24 hours. Advanced age (>70 years), presenting with Child—
Pugh class C liver cirrhosis, and ascites were statistically
significant factors that led to a poor prognosis (Table 4).
Lower platelet count, anemia, and elevated creatinine were
258 M.-J. Bair et al.

Table 3 Microorganisms isolated in patients with necrotizing fasciitis

Microorganisms Monomicrobial Polymicrobial Total

Gram-positive 39 14 53
Streptococcus pyogenes 14 3 17
MSSA 10 1 11
MRSA 2 0 2
Enterococcus 3 3 6
CoNS 3 3 6
a-Hemolytic streptococcus 3 2 5
Streptococcus group non-ABD 2 1 3
Group B Streptococcus 1 0 1
Group D Streptococcus 1 0 1
Streptococcus pneumoniae 0 1 1
Gram-negative 25 28 53
Escherichia coli 7 1 8
Klebsiella pneumoniae 4 3 7
Morganella morganii 0 6 6
Serratia marcescens 1 5 6
Pseudomonas aeruginosa 3 3 6
Aeromonas hydrophila 5 1 6
Acinetobacter spp 0 3 3
Proteus mirabilis 1 2 3
Enterobacter cloacae 1 2 3
Proteus vulgaris 0 1 1
Vibrio vulnificus 1 0 1
Vibrio alginolyticus 1 0 1
Salmonella group B 1 0 1
Haemophilus influenzae 0 1 1
Anaerobic bacteria 0 1 1
Clostridium spp 0 1 1
CoNS, coagulase-negative Staphylococcus; MSSA, methicillin-sensitive Staphylococcus aureus; MRSA, methicillin-resistant Staphylococcus
aureus.

significantly associated with mortality. There were no sig-
nificant differences in affected sites, types of infection, or
sex between patients who died of necrotizing fasciitis and
those who survived. The mortality rate was lower in patients
with multiple pathogens isolated than those with a single
pathogen isolated (4.8% vs. 18.8%); however, this was not
statistically significant ( p = 0.093). The etiology of necrotiz-
ing fasciitis had no statistically significant impact on survival.
Discussion
Necrotizing fasciitis is a term that was first used in the 1950 s
to describe rapidly spreading gangrene of the skin and sub-
cutaneous tissues above the fascial layer.2 It is a rapidly
progressing, life-threatening infection located in the deep
fascia, and is categorized into three distinct groups according
to bacteriologic cultures.
Type 1 necrotizing fasciitis occurs in immunocompromised
hosts, and the infecting organisms are often introduced at
sites of surgery or trauma.15 In our patient series, type 1 was
the most common type of infection, and most of the affected
patients had underlying diseases, but surgery or trauma
history was not well defined. Unlike in the published data,16
two thirds of these were monomicrobial infections, and
among them S. aureus was the most common isolate. This
might be due to the lower isolation rate of anaerobic bacteria
or longer preceding use of antibiotics. Compared with pre-
vious reports,3,16 in our series we noted some organisms such
as M. morganii and S. marcescens that have rarely been
reported to be the cause of type 1 infection. This condition
might be due to the colonization by the above bacteria of the
gastrointestinal tract, resulting in an opportunistic infection.
We found that type 2 infections were mostly caused by S.
pyogenes alone and that all affected the limbs; this finding is
similar to those of previous reports.15,17 The association of
type 3 necrotizing fasciitis with chronic liver disease and
alcoholism in our study is similar to the results of previously
reported articles.18,19
Diabetes mellitus was the leading predisposing factor in
our patients, especially in type 1 infections, as found in
previous studies.3,20 The mechanisms are not only attributed
to the increased susceptibility to minor trauma resulting from
peripheral sensory polyneuropathy, but also to tissue hypoxia
caused by diabetic vascular disease and the underlying immu-
nodeficiency.21
The susceptibility of patients with liver cirrhosis to infec-
tion has been emphasized.5,22 Cirrhotic patients usually have
chronic edema of the lower limbs, which may predispose
them to minor trauma, resulting in an entry port for bacteria.
On the other hand, bacteremia may first occur via the
intestinal—portal route, because liver cirrhosis weakens
the barrier to the passage of bacteria from the intestine
Necrotizing fasciitis in southeast Taiwan 259

Table 4 Comparison between patients who survived and those who died

Patients who Patients who OR (95 % CI) p

survived (N = 88) died (N = 18)
Age (years) 57.3  14.1 64.4  13.9 0.06
>70 14 7 3.36 (1.11—10.17) 0.026
Gender (male) 65 11 0.56 (0.19—1.61) 0.27

Risk factor
Alcoholism 24 5 1.03 (0.33—3.18) 0.23
Smoking 23 5 1.09 (0.35—3.38) 0.22
Ascites 9 7 5.59 (1.88—16.60) 0.002

Comorbidity
Diabetes mellitus 30 7 1.23 (0.43—3.50) 0.70
Liver cirrhosis 22 6 1.50 (0.50—4.47) 0.46
Child—Pugh class C 4 5 22.5 (2.03—249.25) 0.006
Cardiovascular disease 19 3 0.73 (0.19—2.77) 0.27
Renal insufficiency 9 4 2.51 (0.68—9.27) 0.11
Gout 13 1 0.34 (0.04—2.77) 0.20
Malignancy 2 1 2.53 (0.22—29.49) 0.44

Type
Type 1 52 8 0.55 (0.20—1.54) 0.25
Type 2 12 5 2.44 (0.74—8.07) 0.13
Type 3 8 0 0.26 (0.02—2.14) 0.18
Time lag (h), mean  SD 64.2  58.2 41.6  36.6 0.039
Operation 24 h 30 8 1.55 (0.55—4.33) 0.40

Laboratory data
Hemoglobin 11 g/dl 44 14 3.50 (1.12—10.92) 0.03
White blood cell count 15  109/l 35 9 1.51 (0.55—4.19) 0.42
Platelet count 100  109/l 16 9 4.50 (1.54—13.13) 0.004
Albumin 2.4 mg/dl 45 12 1.91(0.66—5.55) 0.23
Creatinine 2.0 mg/dl 23 11 4.44 (1.62—12.18) 0.004
Aspartate aminotransferase 70 IU/l 18 6 1.94 (0.64—5.89) 0.23
Glucose 180 mg/dl 23 5 1.09 (0.35—3.38) 0.22
OR, odds ratio; CI, confidence interval.

to the systemic circulation. Bacteria in the bloodstream may
subsequently seed in the edematous soft tissue of the lower
limbs and then cause infection. The association of mortality
and liver cirrhosis and edema in our study is consistent with
previous studies.23,24
Necrotizing fasciitis is essentially polymicrobial, deriving
significant contributions from both Gram-negative and Gram-
positive bacteria.25 In our study, S. pyogenes was the most
common single pathogen isolated and was associated with a
higher mortality; this is similar to the results of a previous
study.26 Streptococcal toxic shock syndrome may be the most
severe risk factor for mortality,26 and the use of antibiotics to
reduce the production of toxin has been shown to improve
survival. Polymicrobial infections occurred in less than one
fourth of our cases. This is in contrast to recent studies in
which microbiological isolates have been found to be poly-
microbial in the majority of patients.4,8 This might be due to
the use of relatively unsophisticated techniques for the
collection, transfer, or culture of anaerobic specimens. S.
aureus was the second leading pathogen isolated in our
series, and like S. pyogenes can cause necrotizing fasciitis
alone, especially in high-risk patients. In contrast to another
report,27 most of our patients with disease caused by S.
aureus had MSSA. Highly virulent bacteria such as Aeromonas
species and Vibrio species have been described as the etio-
logic cause associated with high mortality.13,28 No patients
with these species died in our case series, and we think this
might be due to the high suspicion of such pathogens in
patients with liver cirrhosis and necrotizing fasciitis in our
hospital.
The mainstay of treatment is early and complete surgical
debridement, combined with antimicrobial therapy, close
monitoring, and physiologic support.1,29 Unlike in the pre-
vious study,29 in our study emergent surgery had no signifi-
cant impact on survival, despite repeated debridement being
performed to control the infection. This might be due to the
aggressive broad-spectrum antibiotic coverage combined
with adequate organ support used in our patients during
260
the acute phase of the disease. As in the study of Miller
et al.,27 S. aureus emerged as a monomicrobial cause in our
series. From our study, we think that appropriate antibiotic
therapy played an important role, and special consideration
should be made for S. pyogenes and S. aureus, as well as
Aeromonas species.
Advanced age is implicated in the increased risk of mor-
tality in patients.3,8 Underlying diseases reflect the immune
status of the patient. Renal insufficiency and ascites may
contribute to the invisible invasion of bacteria and poor
nutritional status. Regarding the site of infection, it is
assumed that infections of the trunk or head and neck region
are associated with higher mortality. In our study all patients
with infections of the trunk and head and neck region sur-
vived. The association of thrombocytopenia and elevated
creatinine with higher mortality found in our study is similar
that found in the study of Hung et al.,20 and this may be
correlated to clinical conditions such as disseminated intra-
vascular coagulopathy and toxic shock syndrome.
Clinical suspicion must prompt immediate surgical revi-
sion with aggressive debridement and appropriate antibiotic
therapy. In our patients, advanced age, class C liver cirrhosis,
ascites, higher serum creatinine, and lower hemoglobin and
platelet levels are negative prognostic factors for survival in
necrotizing fasciitis. S. pyogenes was the most common
pathogen. With a better understanding of the risk factors
and the microbiological data, the logical approach to diag-
nosis and treatment with early and complete debridement,
broad-spectrum antibiotic therapy, and adequate organ sup-
port may result in improved outcomes.
Conflict of interest: We all work at the Mackay Memorial
Hospital. There are no potential conflicts of interest among
us. The patients were diagnosed and treated in our hospital.
This task did not receive any grants.
References
1. Wong CH, Chang HC, Pasupathy S, Khin LW, Tan JL, Low CO.
Necrotizing fasciitis: clinical presentation, microbiology, and
determinants of mortality. J Bone Joint Surg Am 2003;85:
1454—60.
2. Wilson B. Necrotizing fasciitis. Am Surg 1952;18:416—31.
3. McHenry CR, Piotrowski JJ, Petrinic D, Malangoni MA. Determi-
nants of mortality for necrotizing soft-tissue infections. Ann Surg
1995;221:558—63.
4. Elliott DC, Kufera JA, Myers RA. Necrotizing soft tissue infec-
tions. Risk factors for mortality and strategies for management.
Ann Surg 1996;224:672—83.
5. Cheng NC, Tai HC, Tang YB, Chang SC, Wang JT. Necrotising
fasciitis: clinical features in patients with liver cirrhosis. Br J
Plast Surg 2005;58:702—7.
6. Yu KH, Ho HH, Chen JY, Luo SF. Gout complicated with necrotiz-
ing fasciitis–—report of 15 cases. Rheumatology (Oxford)
2004;43:518—21.
7. Swartz MN. Cellulitis and subcutaneous tissue infections. In:
Mandell GL, Bennett JE, Dolin R, editors. Principles and practice
of infectious diseases. 6th ed. Philadelphia: Elsevier Churchill
Livingstone; 2005. p. 1172—94.
M.-J. Bair et al.
8. Childers BJ, Potyondy LD, Nachreiner R, Rogers FR, Childers ER,
Oberg KC, et al. Necrotizing fasciitis: a fourteen-year retro-
spective study of 163 consecutive patients. Am Surg 2002;68:
109—16.
9. Taviloglu K, Cabioglu N, Cagatay A, Yanar H, Ertekin C, Baspinar I,
et al. Idiopathic necrotizing fasciitis: risk factors and strategies
for management. Am Surg 2005;71:315—20.
10. Liu YM, Chi CY, Ho MW, Chen CM, Liao WC, Ho CM, et al.
Microbiology and factors affecting mortality in necrotizing fas-
ciitis. J Microbiol Immunol Infect 2005;38:430—5.
11. Golger A, Ching S, Goldsmith CH, Pennie RA, Bain JR. Mortality in
patients with necrotizing fasciitis. Plast Reconstr Surg
2007;119:1803—7.
12. Meisel M, Schultz-Coulon HJ. Life-threatening necrotizing fas-
ciitis colli caused by Serratia marcescens. HNO 2007. [Epub
ahead of print] German. PMID: 17676289.
13. Tsai YH, Hsu RW, Huang TJ, Hsu WH, Huang KC, Li YY, et al.
Necrotizing soft-tissue infections and sepsis caused by Vibrio
vulnificus compared with those caused by Aeromonas species. J
Bone Joint Surg Am 2007;89:631—6.
14. Oliver JD. Wound infections caused by Vibrio vulnificus and other
marine bacteria. Epidemiol Infect 2005;133:383—91.
15. Green RJ, Dafoe DC, Raffin TA. Necrotizing fasciitis. Chest
1996;110:219—29.
16. Elliott D, Kufera JA, Myers RA. The microbiology of necrotizing
soft tissue infections. Am J Surg 2000;179:361—6.
17. Kotrappa KS, Bansal RS, Amin NM. Necrotizing fasciitis. Am Fam
Physician 1996;53:1691—7.
18. Howard RJ, Pessa ME, Brennaman BH, Ramphal R. Necrotizing
soft-tissue infections caused by marine vibrios. Surgery 1985;98:
126—30.
19. Uchiyama D, Fujimoto K, Hirohashi N, Miyazaki T. Vibrio vulni-
ficus-necrotizing fasciitis in patient with cirrhosis. Intern Med
2007;46:143.
20. Hung CC, Chang SC, Lin SF, Fang CT, Chen YC, Hsieh WC. Clinical
manifestations, microbiology and prognosis of 42 patients with
necrotizing fasciitis. J Formos Med Assoc 1996;95:917—22.
21. McArdle P, Gallen I. Necrotising fasciitis in diabetics. Lancet
1996;348:552.
22. Vilstrup H. Cirrhosis and bacterial infections. Rom J Gastroen-
terol 2003;12:297—302.
23. Kuo CH, Changchien CS, Yang CY, Sheen IS, Liaw YF. Bacteremia in
patients with cirrhosis of the liver. Liver 1991;11:334—9.
24. Corredoira JM, Ariza J, Pallares R, Carratala J, Viladrich PF, Rufi
G, et al. Gram-negative bacillary cellulitis in patients with
hepatic cirrhosis. Eur J Clin Microbiol Infect Dis 1994;13:
19—24.
25. Legbo JN, Legbo JF. Bacterial isolates from necrotizing fasciitis:
a clinico-pathological perspective. Niger J Med 2007;16:143—7.
26. Zurawski CA, Bardsley M, Beall B, Elliott JA, Facklam R, Schwartz
B, et al. Invasive group A streptococcal disease in metropolitan
Atlanta: a population-based assessment. Clin Infect Dis 1998;27:
150—7.
27. Miller LG, Perdreau-Remington F, Rieg G, Mehdi S, Perlroth J,
Bayer AS, et al. Necrotizing fasciitis caused by community-
associated methicillin-resistant Staphylococcus aureus in Los
Angeles. N Engl J Med 2005;352:1445—53.
28. Ko WC, Lee HC, Chuang YC, Liu CC, Wu JJ. Clinical features and
therapeutic implications of 104 episodes of monomicrobial Aero-
monas bacteraemia. J Infect 2000;40:267—73.
29. Anaya DA, Dellinger EP. Necrotizing soft-tissue infection: diag-
nosis and management. Clin Infect Dis 2007;44:705—10.