BJD

E P I DE M I O L O G Y British Journal of Dermatology

Epidemiology of staphylococcal scalded skin syndrome in

U.S. children
A. Staiman,1 D.Y. Hsu1 and J.I. Silverberg iD 1,2,3
Departments of 1Dermatology, 2Preventive Medicine and 3Medical Social Sciences, Feinberg School of Medicine at Northwestern University, Chicago, IL 60611,
U.S.A.

Summary

Correspondence Background Staphylococcal scalded skin syndrome (SSSS) is a blistering dermatosis

Jonathan I. Silverberg. caused by exfoliative toxins released from Staphylococcus aureus.
E-mail: JonathanISilverberg@Gmail.com Objectives To describe the incidence, costs, length of stay (LOS), comorbidities
and mortality of SSSS in U.S. children.
Accepted for publication
23 October 2017
Methods The Nationwide Inpatient Sample 2008–2012 was analysed, including a
20% sample of U.S. hospitalizations and 589 cases of SSSS.
Funding sources Results The mean annual incidence of SSSS was 7
67 (range 1
83–11
88) per mil-
This publication was made possible with support lion U.S. children, with 45
1 cases per million U.S. infants age < 2 years. In
from the Agency for Healthcare Research and multivariable logistic regression models, SSSS was significantly associated with
Quality (AHRQ), grant number K12 HS023011,
the following (shown as adjusted odds ratio and 95% confidence interval):
and the Dermatology Foundation.
female sex (1
12, 1
00–1
25), age (2–5 years: 13
31, 11
82–14
99; 6–10 years:
Conflicts of interest 2
93, 2
35–3
66; 11–17 years: 0
44, 0
31–0
63); race/ethnicity (black: 0
69,
None declared. 0
58–0
84) and season (winter: 2
04, 1
66–2
50; summer: 3
47, 2
86–4
22;
autumn: 3
04, 2
49–3
70), with increasing odds over time (2010–2011: 2
28,
DOI 10.1111/bjd.16097 2
07–2
51; 2012: 2
98, 2
69–3
30). The geometric mean (95% confidence inter-
val) LOS and cost of hospitalization for patients with vs. without SSSS were 3
2
(3
0–3
4) vs. 2
4 (2
4–2
5) days and $4624
0 ($4250–$5030) vs. $1872
($1782
7–$1965). Crude inpatient mortality rates (with 95% confidence inter-
vals) were similar for children with vs. without SSSS (0
33%, 0
00–0
79% vs.
0
36%, 0
34–0
39%). SSSS was associated with other infections, including in the
upper respiratory tract and skin.
Conclusions The prevalence of SSSS appears to be increasing over time, and was
associated with a number of sociodemographic factors and other infections. Fur-
ther studies are needed to confirm these findings and reduce rising rates of SSSS.

What’s already known about this topic?

• Staphylococcal scalded skin syndrome is a rare but potentially life-threatening
infection in childhood.

What does this study add?

• Paediatric staphylococcal scalded skin syndrome is associated with prolonged hospi-
talization, increased costs of care, multiple infectious comorbidities, rising inci-
dence and significant racial/ethnic disparities in the U.S.A.

Staphylococcal scalded skin syndrome (SSSS) is a potentially released by Staphylococcus aureus.3,4 Little is known about the epi-
life-threatening disorder characterized by erythema and super- demiology of SSSS. Epidemiological studies of SSSS in France
ficial blistering of the skin.1,2 SSSS is a toxin-mediated syn- and Germany found the incidence to be between 0
09 and
drome that occurs secondary to exfoliative toxins A and B 0
56 per million persons.5,6 However, the incidence of SSSS in

704 British Journal of Dermatology (2018) 178, pp704–708 © 2017 British Association of Dermatologists

the U.S.A. remains unknown. Older studies found high mor-
tality rates for SSSS in children (3
6–11%).7,8 Since then, there
have been major advances in the antibiotic management of
staphylococcal infections, yet few large-scale studies have
recently examined mortality rates from SSSS in children.
Certain risk factors for SSSS are well established, such as
insufficient infection control practices in nursing homes and
hospitals, impaired renal clearance,9 and poor immunity to
staphylococcal toxins.10 While most SSSS is attributed to
methicillin-sensitive S. aureus, an increase in methicillin-resis-
tant bacteria has also been observed.11,12 One study found
higher rates of SSSS in rural vs. urban areas, suggesting there
may be sociodemographic and/or environmental risk factors
for SSSS.13 Furthermore, previous studies found conflicting or
limited evidence of differences of SSSS prevalence by sex and/
or race/ethnicity.14–16 However, few studies have examined
whether there are sociodemographic factors associated with
SSSS. In the present study, we sought to describe the inci-
dence, associations, comorbidities, mortality and financial
burden of SSSS in U.S. children.
Patients and methods
The Agency for Healthcare Research and Quality (AHRQ) devel-
oped the Nationwide Inpatient Sample (NIS) as part of the
Healthcare Cost and Utilization Project (HCUP) to inform deci-
sion making at the state, national and community levels. Each
year, the NIS contains cross-sectional data on roughly 7 million
hospital stays, with sample weights created by NIS that factor in
the sampling design of hospitals. All discharges of sampled hos-
pitals are included in the database and all patients are deidenti-
fied. The 2008–2012 NIS was analysed in this study, which
includes a 20% stratified sample of all hospitalizations in the
U.S.A., excluding rehabilitation and long-term acute-care hospi-
tals. This study was approved by the institutional review board
at Northwestern University Feinberg School of Medicine.
Staphylococcal scalded skin syndrome and comorbidity
identification
Hospital discharges with either a principal or secondary diagno-
sis of SSSS were identified using the International Classification
of Disease, Ninth Revision, Clinical Modification (ICD-9-CM)
code 695
81. The principal diagnosis in NIS was defined as the
primary reason for hospital admission. The structure of NIS
allows for a single principal diagnosis, up to 24 secondary diag-
noses, and no more than 15 procedures associated with each
hospitalization to be stored. The control group consisted of all
hospitalizations without a diagnosis of SSSS, yielding a cohort
representative of hospitalizations in the U.S.A. that excluded
newborn deliveries.
Comorbidities and procedures were classified by ICD-9-CM
codes and were determined by the principal and secondary
diagnoses and procedure codes. The diagnoses analysed were
chosen based on previously described associations with SSSS
or based on hypothesized contribution to hospitalization.
© 2017 British Association of Dermatologists
Staphylococcal scalded skin syndrome, A. Staiman et al. 705
Mortality and loss of function
Crude adjusted inpatient mortality, based on the composition
of the population of the U.S.A. from 2008 to 2012, was cal-
culated. Patients who were transferred to another acute-care
hospital upon discharge were excluded from mortality rate
determination, as such transfers may represent escalation of
care. Loss of function and mortality risk were determined by
the All Patient Refined Diagnoses Related Group classification,
which is based on algorithms performed by software devel-
oped by 3M (St Paul, MN, U.S.A.).
Statistical methods
All data processes and statistical analyses were performed using
the SURVEY procedures of SAS version 9.4 (SAS Institute Inc.,
Cary, NC, U.S.A.), allowing for adjustment due to the complex
weighting, sample clusters and strata of NIS. The unit of analysis
was an individual hospitalization. The weighted incidence of
hospitalizations with a diagnosis of SSSS was determined,
excluding both adults and patients who were transferred to
another hospital. The mean, 95% confidence interval (CI) for
mean, SD and sum regarding length of stay (LOS) and costs of
care were calculated. Costs of care were adjusted for inflation to
the year 2014 according to the consumer price index and were
based on the total charge of the hospitalization in combination
with the cost-to-charge ratio provided by HCUP.
Several different associations of hospitalization, length of
stay and cost of care due to SSSS were examined, including
age, sex, year, hospital region, season, race, zip code, income
quartile, urban location, insurance status, number of chronic
diseases, hospital bed size and hospital type. Survey logistic
regression models were constructed with SSSS as the depen-
dent variable and each of the aforementioned covariates as the
independent variables. To determine predictors of LOS and
cost of care, survey linear regression models were created with
each sociodemographic characteristic as the independent vari-
ables and log-transformed LOS or cost of care as the depen-
dent variables. Finally, survey linear regression models were
performed to determine the prevalence of SSSS in each of the
46 available states, where SSSS was the dependent variable and
hospital state was the independent variable.
Associations between each comorbidity and SSSS hospitaliza-
tion were examined. Row percentage prevalences with 95%
CIs were graphed on forest plots through Open Meta Analyst,
an open-source application supported by Brown University
and the AHRQ. Finally, the top 20 most frequent principal
diagnoses in patients with a secondary discharge diagnosis of
SSSS were examined.
Results
Incidence of staphylococcal scalded skin syndrome
Overall, 6 149 864 paediatric admissions were captured in the
NIS in 2008–2012, representing 29 533 677 weighted
British Journal of Dermatology (2018) 178, pp704–708
706 Staphylococcal scalded skin syndrome, A. Staiman et al.
(a)
Fig 1. (a) State-wide incidence of staphylococcal scalded skin syndrome (SSSS). Data are presented as the incidence per 100 000 children. SSSS
incidence was divided into tertiles and colour coded: tertile 1 in blue, tertile 2 in green, tertile 3 in red. (b) Incidence of SSSS stratified by region
and season.
hospitalizations. There were 589 hospitalizations with a diag-
nosis of SSSS (weighted frequency of 2833). The annual inci-
dence of SSSS was 7
67 (range 1
83–11
88) per million U.S.
children. The annual estimated incidences were 45
1 cases per
million U.S. infants age < 2 years and 20
9 cases per million
U.S. infants who were 1 year old.
There was considerable state-wide variation in the incidence
of SSSS (Fig. 1a), with the highest rates occurring in the Mid-
west and southern U.S.A. Moreover, the incidence of SSSS var-
ied by season, with the highest rates occurring in the summer
and autumn months (Fig. 1b).
Associations of staphylococcal scalded skin syndrome
In multivariable survey-weighted logistic regression models,
SSSS was significantly associated with multiple sociodemo-
graphic factors, including female sex, age, hospital location,
season, hospital size and teaching status, but inversely associ-
ated with black race, 2nd-quartile household income, public
insurance and more chronic conditions.
Moreover, in multivariable models, the odds of SSSS signifi-
cantly increased in the years 2010–2011 and 2012 compared
with 2008–2009 (Table S1; see Supporting Information).
Length of stay and cost of care
The mean (95% CI) LOS for patients with vs. without SSSS
was 3
2 (3
0–3
4) vs. 2
4 (2
4–2
5) days. The geometric
mean (95% CI) cost of hospital care for patients with SSSS
was dramatically higher than for those without SSSS [$4624
0
($4250
8–$5030
1) vs. $1871
7 ($1782
7–$1965
1)], with a
marginally significant increase between 2008 and 2012
[2008–2009: $4282
6 ($3691
3–$4968
6); 2010–2011:
$4450
3 ($3919
4–$5053
2); 2012: $5185
6 ($4450
6–
$6041
9); P = 0
07]. The mean annual total LOS and cost of
care in patients hospitalized with SSSS in 2008–2012 were
3174
7 days and $5 537 452
3, respectively (Table S2; see
Supporting Information).
British Journal of Dermatology (2018) 178, pp704–708
(b)
In multivariable survey-weighted linear regression models,
increased LOS was associated with ages 2–5, 6–10 and 11–17
vs. < 2 years, black and multiracial/other race/ethnicity, and
multiple chronic conditions (Table S3; see Supporting Infor-
mation). Cost of care was associated with age 11–17 years,
multiracial/other race/ethnicity, and multiple chronic
conditions.
Hospital course and disposition
Only a small subset of children with SSSS underwent intuba-
tion and ventilation (2
7%), physical therapy (0
5%), dialysis
(0
2%), skin debridement (1
7%) or graft (0
2%) (Table S4;
see Supporting Information). Most patients had only minor
(54
0%, 95% CI 49
5–58
4%) or moderate (36
3%, 32
0–
40
6%) loss of function and only minor mortality risk
(90
4%, 88
1–92
7%).
The crude inpatient mortality rate for patients with SSSS
(0
33%, 95% CI 0
00–0
79%) was similar to that in those
without SSSS (0
36%, 0
34–0
39%). Children with SSSS were
most commonly routinely discharged to home or self-care
(94
5%, 92
5–96
4%) and less commonly transferred to a
short-term hospital (3
8%, 2
2–5
4%) (Table S4).
Comorbidities of staphylococcal scalded skin syndrome
The most common comorbidity found in the cohort of
patients with SSSS was any skin infection [15 490, 95% CI
12 060–18 920 per hundred thousand children (PHTC)],
with other common comorbidities being upper respiratory
tract infection (6410, 4320–8500 PHTC), cellulitis (2290,
1020–3550 PHTC), sepsis (3650, 2190–5110 PHTC), pharyn-
gitis (2770, 1360–4190 PHTC) and any other upper respira-
tory infection (2050, 930–3180 PHTC). Patients with SSSS
were also commonly found to have fungal (3780, 2300–5250
PHTC) or viral infections (3228, 1780–4660 PHTC).
SSSS was significantly associated with any skin infection,
including cellulitis, as well as viral infection; fungal infection;
© 2017 British Association of Dermatologists

any upper respiratory tract infection, including nasopharyngi-
tis; pharyngitis; epiglottitis; other influenzas; other upper res-
piratory tract infection; empyema and sepsis; but inversely
associated with bronchitis, pneumonia and appendicitis
(Table S5; see Supporting Information). In multivariable sur-
vey logistic regression models of sufficiently powered comor-
bidities, adjusted for age and sex, SSSS remained significantly
associated with all of these covariates.
Children with SSSS were found to have these associated
comorbidities most commonly during the autumn and sum-
mer and less commonly in the spring. In particular, the high-
est rates occurred during autumn and winter in the southern
U.S.A., as well as in the Midwest and southern U.S.A. during
the summer. Less commonly, these infections occurred in
children with SSSS in the west, Midwest and northeast during
spring, as well as in the west and northeast during winter
(Fig. S1; see Supporting Information).
Discussion
The present study identified 589 cases of SSSS over the years
2008–2012 with a mean estimated annual incidence of 7
67
cases per million children. Previous studies found incidences
of 0
09–0
13 and 0
56 cases per million persons annually in
Germany and France, respectively.5,6 However, those studies
did not stratify incidence rates by paediatric vs. adult cases per
se. In particular, the incidence of SSSS in our study was 45
1
cases per million infants age < 2 years, which differed consid-
erably from previous estimates of 251
1 and 696
4 cases per
million children younger than 1 year in the Czech Republic
and China, respectively. Rates of SSSS significantly increased in
the years 2010–2011 and 2012 compared with 2008–2009.
This upward trend of SSSS incidence is consistent with previ-
ous reports.16–18 A study of 39 cases of SSSS in Chinese infants
found that only eight cases (21%) occurred between 2004
and 2008 and 31 (79%) occurred between 2008 and 2012.16
This upward trend is concerning because the overall incidence
of SSSS is thought to have dramatically declined in Europe and
other regions over the past couple of centuries with the advent
of better hygiene standards.5
Significant sociodemographic factors that were associated
with SSSS in this hospitalized cohort include female sex, age
group 2–5 years, and to a lesser extent age 6–10 years, but
there was an inverse association with age 11–17 years. The
significant difference in age groups compared with patients
who were < 2 years old is consistent with a study whose
mean age of patient with SSSS was 17
4 days.16 However, our
results indicate that SSSS is more commonly associated with
female than male sex, while other studies have reported that
male patients were more commonly affected.15,16 Interest-
ingly, black race, 2nd quartile household income, public
insurance and more chronic conditions were all inversely asso-
ciated with SSSS. A previous study also found that white chil-
dren may be more prone to SSSS than black children.14 The
lower rates and prolonged LOS for SSSS in black children may
© 2017 British Association of Dermatologists
Staphylococcal scalded skin syndrome, A. Staiman et al. 707
be due to underdiagnosis and/or delayed diagnosis in darker-
skinned infants due to masking of erythema by the pigment.
SSSS occurred at the highest rates in the summer and
autumn, consistent with previous studies.16,19 Some authors
have suggested that viral upper respiratory tract infections
occurring in autumn may predispose to SSSS in individuals
colonized with S. aureus enterotoxin-producing strains.6 Indeed,
we found that SSSS was associated with a multitude of infec-
tions of the upper respiratory tract, skin and other organ sys-
tems. Previous reports showed infections of the upper
respiratory tract, inner ear, conjunctiva and/or umbilical
cord,20 as well as pneumonia,16 pyomyositis21 and maternal
breast abscesses22 to precede SSSS.
The adjusted in-hospital mortality of SSSS was low (0
33%)
in the present study, which is consistent with results found in
two other studies.15,16 However, some studies discovered a
mortality rate of 3
6–11% in children with SSSS.5,7,8,23 The
main causes of death from SSSS were infectious complications
such as pneumonia, electrolyte imbalance and sepsis.24 In
addition, cases of SSSS were associated with more than double
the costs of a hospitalization without a diagnosis of SSSS (ex-
cess geometric mean LOS of 0
8 days and cost of $2752
3 per
hospitalization).
This study has several strengths, including the use of a
large, nationally representative sample of U.S. children and
their socioeconomic statuses from all states. Data on nearly 6
million hospitalizations were analysed, allowing for identifica-
tion of over 500 cases of SSSS in the years 2008–2012. SSSS is
a severe illness, for which affected individuals are almost
invariably hospitalized for management and captured in the
NIS. Our study also has some limitations. SSSS and comorbidi-
ties were identified by ICD-9-CM codes, and were not verified
by chart review. It is possible that milder cases of SSSS were
undiagnosed. In addition, we were unable to determine the
effectiveness of any treatments as the NIS does not contain
data on treatments or medications used during hospitalization.
Finally, this was a cross-sectional study, thereby precluding
determination of temporality of comorbidities, and direction
of association. It is possible that some of the examined comor-
bidities may have preceded the diagnosis of SSSS while others
were a result of underlying SSSS.
In conclusion, SSSS poses a significant healthcare burden,
with increased LOS and costs of care per hospitalization, and
increasing prevalence over the 2008–2012 study period.
Female sex; ages 2–5 and to a lesser extent 6–10 years; and
autumn, winter and summer seasons were associated with
increased odds of SSSS. Black race, 2nd quartile household
income, public insurance and more frequent chronic condi-
tions were all inversely associated with SSSS. In contrast, black
and multiracial/other race/ethnicity and multiple chronic con-
ditions were associated with increased LOS and costs of inpa-
tient care due to SSSS. Finally, the adjusted in-hospital
mortality of SSSS was low, and largely attributable to infec-
tious complications, such as pneumonia and sepsis. Future
studies are needed to confirm these associations and determine
British Journal of Dermatology (2018) 178, pp704–708
708 Staphylococcal scalded skin syndrome, A. Staiman et al.

optimal strategies aimed at reducing the prevalence of and
mitigating risk factors for SSSS.
References
1 Cribier B, Piemont Y, Grosshans E. Staphylococcal scalded skin
syndrome in adults. A clinical review illustrated with a new case. J
Am Acad Dermatol 1994; 30:319–24.
2 Mishra AK, Yadav P, Mishra A. A systemic review on staphylococ-
cal scalded skin syndrome (SSSS): a rare and critical disease of
neonates. Open Microbiol J 2016; 10:150–9.
3 Resnick SD. Staphylococcal toxin-mediated syndromes in child-
hood. Semin Dermatol 1992; 11:11–18.
4 Ladhani S. Understanding the mechanism of action of the exfolia-
tive toxins of Staphylococcus aureus. FEMS Immunol Med Microbiol 2003;
39:181–9.
5 Mockenhaupt M, Idzko M, Grosber M et al. Epidemiology of
staphylococcal scalded skin syndrome in Germany. J Invest Dermatol
2005; 124:700–3.
6 Lamand V, Dauwalder O, Tristan A et al. Epidemiological data of
staphylococcal scalded skin syndrome in France from 1997 to
2007 and microbiological characteristics of Staphylococcus aureus asso-
ciated strains. Clin Microbiol Infect 2012; 18:E514–21.
7 Melish ME, Glasgow LA. The staphylococcal scalded-skin syn-
drome. N Engl J Med 1970; 282:1114–19.
8 Lowney ED, Baublis JV, Kreye GM et al. The scalded skin syndrome
in small children. Arch Dermatol 1967; 95:359–69.
9 Schwartz RA, McDonough PH, Lee BW. Toxic epidermal necroly-
sis: part II. Prognosis, sequelae, diagnosis, differential diagnosis,
prevention, and treatment. J Am Acad Dermatol 2013; 69:187.
10 Kadam S, Tagare A, Deodhar J et al. Staphylococcal scalded skin
syndrome in a neonate. Indian J Pediatr 2009; 76:1074.
11 Liassine N, Auckenthaler R, Descombes MC et al. Community-
acquired methicillin-resistant Staphylococcus aureus isolated in
Switzerland contains the Panton-Valentine leukocidin or exfoliative
toxin genes. J Clin Microbiol 2004; 42:825–8.
12 Noguchi N, Nakaminami H, Nishijima S et al. Antimicrobial agent
of susceptibilities and antiseptic resistance gene distribution among
methicillin-resistant Staphylococcus aureus isolates from patients with
impetigo and staphylococcal scalded skin syndrome. J Clin Microbiol
2006; 44:2119–25.
13 Su J, Li J, Luo H et al. Demographic and clinical characteristics and
risk factors for staphylococcal scalded skin syndrome in Hunan.
Zhong Nan Da Xue Xue Bao Yi Xue Ban 2016; 41:417–21.
14 Rasmussen JE. Toxic epidermal necrolysis. A review of 75 cases in
children. Arch Dermatol 1975; 111:1135–9.
15 Lipovy B, Brychta P, Chaloupkova Z et al. Staphylococcal scalded
skin syndrome in the Czech Republic: an epidemiological study.
Burns 2012; 38:296–300.
16 Li MY, Hua Y, Wei GH et al. Staphylococcal scalded skin syndrome
in neonates: an 8-year retrospective study in a single institution.
Pediatr Dermatol 2014; 31:43–7.
17 Faden H. Neonatal staphylococcal skin infections. Pediatr Infect Dis J
2003; 22:389.
18 El Helali N, Carbonne A, Naas T et al. Nosocomial outbreak of
staphylococcal scalded skin syndrome in neonates: epidemiological
investigation and control. J Hosp Infect 2005; 61:130–8.
19 Handler MZ, Schwartz RA. Staphylococcal scalded skin syndrome:
diagnosis and management in children and adults. J Eur Acad Derma-
tol Venereol 2014; 28:1418–23.
20 Bailey CJ, Lockhart BP, Redpath MB et al. The epidermolytic (exfo-
liative) toxins of Staphylococcus aureus. Med Microbiol Immunol 1995;
184:53–61.
21 Wong GW, Oppenheimer SJ, Evans RM et al. Pyomyositis and
staphylococcal scalded skin syndrome. Acta Paediatr 1993; 82:113–
15.
22 Raymond J, Bingen E, Brahimi N et al. Staphylococcal scalded skin
syndrome in a neonate. Eur J Clin Microbiol Infect Dis 1997; 16:453–
4.
23 Patel GK, Finlay AY. Staphylococcal scalded skin syndrome: diag-
nosis and management. Am J Clin Dermatol 2003; 4:165–75.
24 Patel GK. Treatment of staphylococcal scalded skin syndrome. Expert
Rev Anti Infect Ther 2004; 2:575–87.
Supporting Information
Additional Supporting Information may be found in the online
version of this article at the publisher’s website:
Fig S1. Comorbidities of paediatric staphylococcal scalded
skin syndrome stratified by season.
Table S1 Associations with staphylococcal scalded skin syn-
drome in the Nationwide Inpatient Sample 2008–2012.
Table S2 Length of stay and cost of care in patient subsets
with staphylococcal scalded skin syndrome in the Nationwide
Inpatient Sample S 2008–2012.
Table S3 Associations of length of stay of cost of care in
patients with staphylococcal scalded skin syndrome in the
Nationwide Inpatient Sample 2008–2012.
Table S4 Procedures, loss of function, mortality risk, and
discharge status of patients with staphylococcal scalded skin
syndrome.
Table S5 Infectious comorbidities of staphylococcal scalded
skin syndrome.

British Journal of Dermatology (2018) 178, pp704–708 © 2017 British Association of Dermatologists