Clinical Microbiology and Infection 23 (2017) 674.e1e674.

e5

Contents lists available at ScienceDirect

Clinical Microbiology and Infection

journal homepage: www.clinicalmicrobiologyandinfection.com

Original article

Factors associated with time to clinical stability in complicated skin

and skin structure infections
€a
I.H. Ja €inen 1, *, L. Hagberg 2, E. Forsblom 1, A. Ja
€skela €rvinen 1
1)
Department of Infectious Diseases, Inflammation Center, Helsinki University Central Hospital Helsinki University and University of Helsinki, Helsinki,
Finland
2)
Institute of Biomedicine, Department of Infectious Diseases, Sahlgrenska Academy, University of Gothenburg, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: Factors associated with the time to clinical stability in patients with complicated skin and
Received 5 October 2016 skin structure infection (cSSSI) were analysed in a retrospective population-based study.
Received in revised form Methods: All hospitalized patients (n¼402) with cSSSI in two Nordic cities during a 4-year period were
27 February 2017
included. Patient, disease, and treatment related factors were analysed in relation to early (0e3 days) or
Accepted 27 February 2017
Available online 6 March 2017
late (4 days) clinical stability. Clinical stability was assessed as improvement of infection related local
and systemic signs. Furthermore, the effect of antimicrobial and other treatment on achievement of
Editor: M. Paul clinical stability was studied.
Results: Clinical stability was reached within 0e3 days by 59% (239/402) of patients. In multivariable
Keywords: analysis later clinical stability was associated with admission to ICU (OR 10.1, 95% CI 4.01e25.3), post-
Complicated skin and skin structure traumatic wound infection (OR 3.17, 95% CI 1.31e7.69), bacteraemia (OR 3.09, 95% CI 1.36e7.02), surgical
infections intervention after diagnosis (OR 2.64, 95% CI 1.36e5.11), diabetes (OR 2.33, 95% CI 1.28e4.25), and initial
Evaluation of treatment response broad-spectrum antibiotic therapy (OR 3.03, 95% CI 1.43e6.40). Early stabilization within 3 days was
Retrospective
associated with previous hospitalization (OR 0.47, 95% CI 0.22e0.99) and empirical antimicrobial therapy
Use of resources
covering the initial pathogens (OR 0.38, 95% CI 0.18e0.80). Patients with clinical stability within 3 days
Population-based
were less likely to have treatment modifications and antimicrobial changes and had shorter hospital stay
and antimicrobial treatment than those who stabilized later.
Conclusions: This study suggests that late treatment response depends on several baseline characteristics
of patients and disease related factors other than treatment related factors. I.H. Ja €a
€skela€inen, Clin
Microbiol Infect 2017;23:674.e1e674.e5
© 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All
rights reserved.

Introduction
Skin and skin structure infections (SSSI) are among the most
frequent human bacterial infections and an increasing indication
for antimicrobial treatment both in the hospital and in the outpa-
tient setting [1e3]. SSSI is generally regarded as complicated (cSSSI)
if it involves deep subcutaneous tissues, needs surgery in addition
to antimicrobial therapy or affects a patient with severe comor-
bidities [4]. Studies in hospitalized patients with cSSSI have shown
high rates of initial treatment failure leading to increased length of
* Corresponding author. I. Ja €a
€skela
€inen, Helsinki University Central Hospital,
Department of Infectious Diseases, PL 348, 00029 HUS, Helsinki, Finland.
€€
E-mail address: iiro.jaaskelainen@hus.fi (I.H. Ja €inen).
askela
hospital stay (LOS), higher total inpatient costs, and higher mor-
tality [5e7].
The majority of SSSI are caused by Gram-positive cocci [8e10]
and practically all new antimicrobial agents against Gram-
positive bacteria are studied in patients with SSSI before
licensing. Historical evidence from placebo-controlled trials sug-
gests that the therapeutic effect of antibiotic treatment is most
prominent during the first days of therapy [11]. Therefore, the US
Food and Drug Administration (FDA) recommended in their recent
guidance for the development of antimicrobials used in acute
bacterial skin and skin structure infections (ABSSSI) the treatment
response to be evaluated at 48 to 72 hours after initiation of the
therapeutic agent instead of traditional end-of-treatment evalua-
tion [12]. Applicability of this early clinical response by day 3 has
been tested in a clinical trial [13] and also in studies with

http://dx.doi.org/10.1016/j.cmi.2017.02.033
1198-743X/© 2017 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
 €a
I.H. Ja €skela
€inen et al. / Clinical Microbiology and Infection 23 (2017) 674.e1e674.e5
retrospective observational design [14,15]; and several factors other
than antimicrobial treatment were found to have an association
with time to treatment response. As opposed to early responders,
late responders were more likely to have diabetes, wound in-
fections, more severe infections, Gram-negative or anaerobic bac-
teria detected, or recurrent infections [14,15]. Later treatment
response was also associated with longer antibiotic treatment
duration and LOS [14,15]. The time point for first assessment of
clinical stability might not only affect the clinical studies made in
cSSSI but also real-life clinical practice; timely evaluated treatment
response may, for example, prevent unnecessary changes in anti-
microbial treatment.
The abovementioned studies were made either in a subset of
patients recruited into a clinical study or in selected hospitals and
might thus not reflect the entire disease population. We have
previously conducted a population-based real-life study in areas
with low incidence of antibiotic resistance and observed that 41% of
the patients reached clinical stability after 3 days [16]. In this paper,
we wanted to study the feasibility of early treatment response
criteria in a population-based setting and to evaluate factors
associated with the time to clinical stability and the association of
early response with clinical outcomes.
Materials and methods
The study design was an observational retrospective cohort
study and the study population consisted of all hospitalized adult
residents of two cities, Helsinki, Finland and Gothenburg, Sweden,
who were treated because of cSSSI during 2008e2011. Detailed
study protocol is presented in the primary publication of this study
[16]. In short, to be included in the final analysis population pa-
tients were required to have an infection affecting deeper soft tis-
sue, infection that required significant surgical intervention,
developed on a lower extremity in a subject with diabetes mellitus
or peripheral vascular disease, and to have systemic sign(s) of
infection.
The main outcome measure was clinical stability which was
defined as signs of improvement of infection and assessed from
patient records by two infectious disease physicians. In particular,
improvements in systemic symptoms such as fever and vital signs
(pulse rate, blood pressure) were used along with local signs of
infection. In line with the FDA recommendation [12] patients were
divided into two groups, according to time from diagnosis of cSSSI
to clinical stability, within 0e3 days or over 3 days. In addition, a
subgroup of 4e5 days to clinical stability was defined to test
whether this subgroup of patients shared similar characteristics
and outcomes with the 0e3 days or 6 days groups. Follow-up data
were collected until 1 year post cSSSI diagnosis.
Statistical analysis
Data are presented as absolute values and percentages, as mean
(and standard deviation) or as median and interquartile ranges
(IQR, 25th and 75th percentiles). On univariable analysis categorical
variables were compared with the Likelihood ratio- or the Pear-
son's c2 test, whereas continuous variables were analysed with the
Mann-Whitney U test. OR were calculated with 95% CI. Multivari-
able logistic regression analysis with backward selection was per-
formed including 1) all clinically relevant variables and 2) those
having univariable p-values less than 0.15, and 3) were not multi-
collinear [17]. The model with lowest Akaike information criteria
(AIC) [18] was the final multivariable model. All tests were two-
tailed and p <0.05 was considered to be significant. SPSS version
21.0 (SPSS Inc., Chicago, IL, USA) was used for all analyses.
674.e2
Results
Patients were recruited from two cities with a similar number of
inhabitants (approximately 600 000): in total, 219 patients were
from Helsinki, Finland and 241 patients from Gothenburg, Sweden.
After exclusion of patients who died before they reached clinical
stability (n¼10) or those with day of clinical stability unknown
(n¼49), the analysis of time to clinical stability included 402 pa-
tients in total, of whom 239 (59%) reached stability in 0e3 days and
163 (41%) after 3 or more days (Table 1). Time to clinical stability
varied between 0 and 50 days. Of the 402 patients, five patients
died within the first 30 days after the diagnosis of cSSSI and they
were included in the analysis, as they had reached clinical stability
before death.
Clinical stability on 0e3 vs. 4 days
Patient's age was not associated with time to clinical stability,
whereas male patients were found to stabilize more often on 4
days than females on univariable analysis (Table 1), although no
statistically significant difference was detected on multivariable
analysis. On multivariable analysis, posttraumatic wound infection,
bacteraemia, diabetes, and a short (<2 days) time from onset of
symptoms to diagnosis of cSSSI were associated with clinical sta-
bility on 4 days (Table 1). Similarly, admission to the intensive
care unit (ICU), a surgical intervention after diagnosis of cSSSI, and
initial treatment with a broad-spectrum antimicrobial agent (car-
bapenem or piperacillin-tazobactam) were found to be associated
with clinical stability on 4 days on multivariable analysis
(Table 1).
Previous hospitalization or invasive surgery within the 3
months before infection and infections with abscess formation,
initially caused by staphylococci or initially treated by penicillin
with staphylococcal effect or other (usually clindamycin) antimi-
crobial were statistically significantly associated with stabilization
on 0e3 days on univariable analysis (Table 1). However, only pre-
vious hospitalization and initial antimicrobial treatment covering
initial pathogens were the factors that remained statistically
significantly associated with stabilization on 0e3 days on multi-
variable analysis (Table 1).
Clinical stability on 0e3 vs. 4e5 days
A small group of patients (52 patients, 13%) stabilized first after 5
days and a high proportion (48%) of them had been treated in ICU,
had polymicrobial aetiology (37%), and they were treated on
average with 5.5 different antimicrobials (Table 1). These patients
might represent a subpopulation with a more complicated course.
Therefore, we analysed separately patients who reached clinical
stability on days 4e5 (n¼111, 28%) and compared them with those
with clinical stability within 3 days (n¼239, 59%). On univariable
analyses, clinical stability on days 4e5 was associated with ICU
treatment (OR 7.05 (3.15e15.77), p <0.001), bacteraemia (OR 3.90
(1.95e7.83), p <0.001), diabetic foot infection (OR 2.47 (1.33e4.59),
p 0.003), diabetes (OR 2.14 (1.35e3.38), p 0.001), male gender (OR
2.00 (1.24e3.22), p 0.004), and cellulitis/fasciitis (OR 1.69
(1.07e2.66), p 0.024). Abscess (OR 0.45 (0.28e0.72), p 0.001) or
surgery within the prior 3 months (OR 0.50 (0.25e0.98), p 0.039)
were the factors that were associated with clinical stability within 3
days on univariable analysis.
Treatment and the use of resources
Treatment modifications and the use of resources were associ-
ated with the time point of clinical stability (Table 2). Patients with
674.e3 €a
I.H. Ja €skela
€inen et al. / Clinical Microbiology and Infection 23 (2017) 674.e1e674.e5

Table 1
Baseline and disease characteristics, microbiological diagnosis, and antimicrobial treatment of 402 patients with complicated skin and skin structure infection (cSSSI) cate-
gorized according to time to clinical stability

Variable Clinical stability Clinical Clinical Clinical stability Clinical stability 4 vs. 0e3 days
0e3 days stability stability 4 vs. 0e3 days Multivariable analysisa
4e5 days 6 days

Full analysis population (N¼402) 239 111 52 OR (95% CI) pb OR (95% CI) P

Baseline characteristics, n (%)

Male gender (n¼240) 127 (53) 77 (69) 36 (69) 1.99 (1.31e3.03) 0.001 e
Age >60 years (n¼253) 146 (61) 68 (61) 39 (75) 1.22 (0.80e1.84) 0.35 e
Injection drug abuse (n¼31) 22 (9) 9 (8) 0 (0) 0.58 (0.26e1.29) 0.17
Alcohol abuse (n¼33) 13 (5) 12 (11) 8 (15) 2.43 (1.17e5.04) 0.015 e
Congestive heart disease (n¼29) 16 (7) 7 (6) 6 (12) 1.21 (0.57e2.58) 0.63
Respiratory disease (n¼31) 15 (6) 9 (8) 7 (13) 1.63 (0.78e3.39) 0.2
Chronic renal failure (n¼25) 13 (5) 10 (4) 3 (6) 1.51 (0.68e3.34) 0.31
Liver disease (n¼21) 10 (4) 9 (8) 2 (4) 1.66 (0.69e4.00) 0.26
Cancer/malignancy (n¼31) 22 (9) 7 (6) 2 (4) 0.58 (0.26e1.29) 0.17
HIV infection (n¼7) 4 (2) 3 (3) 0 (0) 1.10 (0.24e4.50) 0.9
Other immunosuppressive disease (n¼14) 10 (4) 2 (2) 2 (4) 0.58 (0.18e1.87) 0.34
Diabetes (n¼158) 81 (34) 58 (52) 19 (37) 1.75 (1.16e2.63) 0.007 2.33 (1.28e4.25) 0.006
Peripheral vascular disease (n¼97) 49 (21) 31 (28) 17 (33) 1.62 (1.02e2.57) 0.041 1.82 (0.93e3.58) 0.08
Hospitalization within 3 months (n¼76) 53/234 (23) 16/110 (15) 7/51 (14) 0.58 (0.34e0.99) 0.04 0.47 (0.22e0.99) 0.047
(data available for 395 patients)
Invasive surgery within 3 months (n¼67) 47/236 (20) 12/109 (11) 8/51 (16) 0.57 (0.32e1.01) 0.047 e
(data available for 396 patients)
Disease characteristics, n (%)
Abscess (n¼172) 121 (51) 35 (32) 16 (31) 0.44 (0.29e0.67) <0.001 0.55 (0.29e1.04) 0.07
Cellulitis/fasciitis (n¼171) 86 (36) 54 (49) 31 (60) 1.94 (1.29e2.91) 0.001 e
Posttraumatic wound (n¼42) 19 (8) 12 (11) 11 (21) 1.90 (1.00e3.62) 0.05 3.17 (1.31e7.69) 0.011
Postsurgical wound (n¼76) 46 (19) 20 (18) 10 (19) 0.95 (0.57e1.58) 0.83
Decubitus/pressure ulcer (n¼10) 8 (3) 2 (2) 0 (0) 0.36 (0.08e1.71) 0.16
Diabetic foot/leg ulcer (n¼53) 24 (10) 24 (22) 5 (10) 1.94 (1.08e3.47) 0.026 e
Peripheral vascular disease ulcer (n¼35) 22 (9) 11 (10) 2 (4) 0.86 (0.42e1.75) 0.67
Bite (n¼2) 2 (1) 0 (0) 0 (0) 0.99 (0.98e1.00) 0.15
Antibiotic treatment before diagnosis (n¼108) 68 (28) 27 (24) 13 (25) 0.82 (0.52e1.29) 0.38
Symptoms <2 days before diagnosis (n¼117) 56/234 (24) 40/110 (36) 21/52 (40) 1.95 (1.26e3.02) 0.003 2.01 (1.02e3.94) 0.043
(data available for 396 patients)
Bacteraemia (n¼50) 15 (6) 23 (21) 12 (23) 4.08 (2.15e7.77) <0.001 3.09 (1.36e7.02) 0.007
Admission to intensive care unit (n¼58) 9 (4) 24 (22) 25 (48) 11.0 (5.21e23.2) <0.001 10.1 (4.01e25.3) <0.001
Surgical intervention after diagnosis (n¼211) 114 (48) 61 (55) 36 (69) 1.61 (1.08e2.41) 0.02 2.64 (1.36e5.11) 0.004
Initial microbiological diagnosis, n (%)
Staphylococci (n¼99) 70 (29) 18 (16) 11 (21) 0.52 (0.32e0.85) 0.008 e
Streptococci (n¼90) 48 (20) 26 (23) 16 (31) 1.38 (0.86e2.22) 0.18
Gram-negative bacteria (n¼24) 15 (6) 7 (6) 2 (4) 0.87 (0.37e2.05) 0.75
Polymicrobial infection (n¼79) 40 (17) 25 (23) 14 (27) 1.57 (0.95e2.57) 0.08 e
Other microbe (n¼12) 5 (2) 4 (4) 3 (6) 2.10 (0.66e6.74) 0.21
Negative/unknown (n¼98) 61 (26) 31 (28) 6 (12) 0.86 (0.54e1.37) 0.52
Initial antimicrobial treatment (n¼400), n (%) (n¼237) (n¼111) (n¼52)
Broad-spectrumc (n¼79) 34 (14) 31 (28) 14 (27) 2.28 (1.38e3.75) 0.001 3.03 (1.43e6.40) 0.004
Cephalosporinsd (n¼202) 119 (50) 57 (51) 26 (50) 1.03 (0.69e1.53) 0.89
Penicillin with staphylococcal effecte (n¼46) 36 (15) 8 (7) 2 (4) 0.37 (0.18e0.76) 0.004 e
Penicillin without staphylococcal effectf (n¼38) 21 (9) 10 (9) 7 (13) 1.20 (0.61e2.35) 0.6
Other antimicrobialsg (n¼35) 27 (11) 5 (5) 3 (6) 0.40 (0.18e0.91) 0.02 e
Initial pathogen(s) covered by initial 155/180 (86) 65/84 (77) 36/46 (78) 0.56 (0.31e1.01) 0.05 0.38 (0.18e0.80) 0.011
antimicrobial treatment (n¼256)
(data available for 310 patients)

Data are no. (column-%) of patients unless otherwise specified. OR, odds ratio; HIV, human immunodeficiency virus.
a
Logistic regression analysis (method: backward/likelihood ratio/Akaike information criteria, n¼308).
b
Likelihood ratio test.
c
Carbapenem and piperacillin-tazobactam.
d
Cefadroxil, cefotaxim, ceftriaxone, cefuroxime, and cephalexin.
e
Cloxacillin, flucloxacillin, and other b-lactamase-stable penicillin.
f
Amoxicillin, benzylpenicillin, and phenoxymethylpenicillin.
g
Clindamycin, doxycyclin, fluoroquinolone, fusidic acid, linezolid, metronidazole, cotrimoxazole, tobramycin, and vancomycin.

clinical stability within 3 days were less likely to have their initial
treatment modified and had shorter length of hospital stay and
antimicrobial treatment duration and fewer different antibiotic
courses and clinics visited compared with patients with later
clinical stability (Table 2).
Because patients were recruited from two countries, we
repeated analyses with country as one explanatory factor on
multivariable analyses and the results were practically identical
(data not shown).
Discussion
Seventy-two hours has been suggested as a breakpoint to assess
the treatment response in SSSI and a new primary endpoint in
clinical studies [11]. This time point seemed to divide patients fairly
even to early and late responders in our study as 59% of patients
had their condition stabilised before the fourth day. However, later
clinical stability was associated not only with treatment related
factors but also with patient's baseline conditions and with disease
 €a
I.H. Ja €skela
€inen et al. / Clinical Microbiology and Infection 23 (2017) 674.e1e674.e5 674.e4

Table 2
Clinical outcomes and resource use of 402 patients with complicated skin and skin structure infections (cSSSI) categorized by time to clinical stability

Variable Clinical Clinical Clinical Clinical stability Clinical stability Clinical stability
stability stability stability  4 vs. 0e3 days 4e5 vs. 0e3 days  6 vs. 4e5 days
0e3 days 4e5 days  6 days

Full analysis population (N¼402) 239 111 52 OR (95% CI) p OR (95% CI) p OR (95% CI) p

Hospitalized again because of cSSSI 39/233 (17%) 19/109 (17%) 8/49 (16%) 1.03 (0.60e1.76) 0.93a 1.05 (0.58e1.92) 0.87a 0.92 (0.37e2.28) 0.87a
(n¼66, data available for 391 patient)
Initial treatment modificationc 37/236 (16%) 54/110 (49%) 25/51 (49%) 5.18 (3.25e8.27) <0.001a 5.19 (3.11e8.66) <0.001a 1.00 (0.51e1.94) 0.99a
(n¼116, data available for 397 patient)
Length of hospital stay, days 7 (4,13) 16 (10, 31) 33 (20, 59) e <0.001b e <0.001b e <0.001b
(n¼378, median (IQR25, 75)) (n¼232) (n¼103) (n¼43)
Duration of antimicrobials, days 12 (7, 23) 27 (16, 43) 29 (16, 48) e <0.001b e <0.001b e 0.61b
(n¼395, median (IQR25, 75)) (n¼234) (n¼110) (n¼51)
No. of antibiotic courses 2.8 (1.4) 4.1 (2.1) 5.5 (2.7) e <0.001b e <0.001b e 0.001b
(n¼401, mean (SD)) (n¼239) (n¼110) (n¼52)
No. of clinics during hospital stay 1.3 (0.7) 2.0 (1.4) 2.4 (1.4) e <0.001b e <0.001b e 0.048b
(n¼402, mean (SD)) (n¼239) (n¼111) (n¼52)

Data are number of patients unless otherwise specified. OR, odds ratio; IQR, interquartile range.
a
Pearson's X2 test.
b
Mann-Whitney U test.
c
Only iv to iv modifications.

characteristics. Later treatment response was associated with
longer and more often modified antibiotic treatment, longer hos-
pital stay, and more clinics taking part in the treatment.
The major strength of this study was its population-based
approach, enabled by the high affinity to public health care in
Nordic countries, that is if patients had the correct diagnostic
ICD10-codes we were able to catch all cSSSI patients from hospital
databases. Naturally, we may have missed some patients because of
coding errors. The population-based approach may give compre-
hensive real-life data in contrast to clinical trials in which patients
with the most severe illness commonly have been excluded. The
data were collected in two countries which make them more
generalizable. This study was also conducted in an area of low
prevalence of MRSA (2011: Finland 2.8%, Sweden 0.8%) [19,20],
ruling out one possible confounding factor. However, the retro-
spective design relying on medical records not initially designed for
research purposes is the major limitation of this study. Insufficient
recording of local signs of infection made it impossible to fully
evaluate the patients' treatment response in the manner recom-
mended by the FDA [12]. The retrospective nature led also to
missing data in some parameters but the main reason for missing
data was the common problem in all SSSI studies: microbiological
documentation is not possible in all patients, but we managed to
retrieve it from 77% of patients. To control for the missing data, we
repeated multivariable analysis without covariate “Initial patho-
gen(s) covered by initial antimicrobial treatment” and found
practically no changes in the results.
Diabetes was more common among late responders in our study
as well as in another European study by Garau et al. [14]. Patients
with wound infection were found to have more initial treatment
failures in two earlier studies, which is in line with our findings
[6,15]. We repeated our analyses by comparing patients with early
clinical stability to those who stabilized on days 4e5. The same risk
factors for late response (Table 1) were all significant also in this
analysis, except peripheral vascular disease, posttraumatic wound
and surgical interventions which did not reach statistical signifi-
cance, but a tendency towards later response was detected. Our and
previous findings suggest that these background and infection
related factors are universal and ought to be controlled when time
to clinical stability is used for comparison of various treatment
options. These findings might also be helpful in clinical practice by
reducing unnecessary early antimicrobial treatment modifications
in patients with these risk factors for later clinical stability.
Early treatment response had significant association with the
outcome. Early responders had shorter median lengths of hospital
stay (7 vs. 21 days) and antimicrobial treatment (12 vs. 28 days)
compared with late responders (after 3 days). The association of
early treatment response to LOS and duration of antimicrobial
treatment now demonstrated in this population-based real-life
study are consistent with the findings of a previous European study
[14] and studies from the USA [5e7,15].
The use of time to clinical stability as the primary endpoint for
clinical success was proposed by the FDA to be able to show
differences between the new antimicrobial agents and the
reference treatments [12]. Indeed, if the empiric antimicrobial
treatment covered the initial causative bacterium, clinical sta-
bility was reached earlier also in our study (OR 0.38 (0.18e0.80)
on multivariable analysis). However, initial pathogens were
covered by empirical treatment in the vast majority of our pa-
tients (83% of patients with microbiological diagnosis) and the
actual difference in those who stabilized early (86%) and late
(78%) does not seem to be clinically significant. The high coverage
of empiric treatment resulted from the dominance of Gram-
positive bacteria (62% of patients with microbiological diag-
nosis), low proportion of methicillin-resistant Staphylococcus
aureus (MRSA, 1%), and high use of antimicrobials comprising also
Gram-negative coverage.
In contrast, initial treatment with a broad-spectrum antibiotic
(20%) was associated with a late treatment response. Similar trends
were detected by Garau et al., who observed that patients treated
with piperacillin-tazobactam or ampicillin-sulbactam were less
likely to have an early response [14]. Together these data might
suggest that broad-spectrum empiric coverage would not guar-
antee early response but might even postpone the treatment
response. Furthermore, a recent study from the USA found signif-
icant increases in the total use of broad-spectrum antibiotics,
including b-lactam/b-lactamase-inhibitor combinations and car-
bapenems [21]. Our population based data suggest that broad-
spectrum coverage is seldom needed in cSSSI and these infections
might be one possible option in antimicrobial stewardship.
In conclusion, this retrospective real-life population-based
study suggests that time to treatment response depends on several
baseline characteristics of patients and disease related factors other
than treatment related factors. Interestingly, initial broad-spectrum
antimicrobial treatment was associated with late clinical response.
Patients with no treatment response within 72 hours after
674.e5 €a
I.H. Ja €skela
€inen et al. / Clinical Microbiology and Infection 23 (2017) 674.e1e674.e5
diagnosis eventually had more antimicrobial treatment changes
and longer antimicrobial and in-hospital treatment duration.
Transparency declaration
This study was sponsored by AstraZeneca Nordic-Baltic. IJ has
received financial support from Gilead for attending symposia. LH
has received speaker's honoraria from Glaxo and Pfizer. EF has no
conflict of interest. AJ has received speaker's honorarium from
Astellas, LeoPharma, MSD, OrionPharma and Ratiopharm and
conference invitation from OctaPharma and has recent consul-
tancies with MSD and Ratiopharm.
Acknowledgements
We greatly thank PhD Jukka Ollgren for his valuable statistical
advice.
References
[1] Edelsberg J, Taneja C, Zervos M, Haque N, Moore C, Reyes K, et al. Trends in US
hospital admissions for skin and soft tissue infections. Emerg Infect Dis
2009;15(9):1516e8.
[2] Hayward A, Knott F, Petersen I, Livermore DM, Duckworth G, Islam A, et al.
Increasing hospitalizations and general practice prescriptions for community-
onset staphylococcal disease, England. Emerg Infect Dis 2008;14(5):720e6.
[3] Currie CJ, Berni E, Jenkins-Jones S, Poole CD, Ouwens M, Driessen S, et al.
Antibiotic treatment failure in four common infections in UK primary care
1991e2012: longitudinal analysis. BMJ 2014;349:g5493.
[4] Center for Drug Evaluation and Research, Food and Drug Administration.
Guidance for industry: uncomplicated and complicated skin and skin struc-
ture infectionsddeveloping antimicrobial drugs for treatment. 1998. http://
www.fda.gov/ohrms/dockets/98fr/2566dft.pdf.
[5] Edelsberg J, Berger A, Weber DJ, Mallick R, Kuznik A, Oster G. Clinical and
economic consequences of failure of initial antibiotic therapy for hospitalized
patients with complicated skin and skin-structure infections. Infect Control
Hosp Epidemiol 2008;29(2):160e9.
[6] Berger A, Oster G, Edelsberg J, Huang X, Weber DJ. Initial treatment failure in
patients with complicated skin and skin structure infections. Surg Infect
(Larchmt) 2013;14(3):304e12.
[7] Zilberberg MD, Shorr AF, Micek ST, Chen J, Ramsey AM, Hoban AP, et al.
Hospitalizations with healthcare-associated complicated skin and skin struc-
ture infections: impact of inappropriate empiric therapy on outcomes. J Hosp
Med 2010;5(9):535e40.
[8] Garau J, Ostermann H, Medina J, Avila M, McBride K, Blasi F, et al. Current
management of patients hospitalized with complicated skin and soft tissue
infections across Europe (2010e2011): assessment of clinical practice pat-
terns and real-life effectiveness of antibiotics from the REACH study. Clin
Microbiol Infect 2013;19(9):E377e85.
[9] Jeng A, Beheshti M, Li J, Nathan R. The role of beta-hemolytic streptococci in
causing diffuse, nonculturable cellulitis: a prospective investigation. Medicine
(Baltimore) 2010;89(4):217e26.
[10] Jenkins TC, Sabel AL, Sarcone EE, Price CS, Mehler PS, Burman WJ. Skin and
soft-tissue infections requiring hospitalization at an academic medical center:
opportunities for antimicrobial stewardship. Clin Infect Dis 2010;51(8):
895e903.
[11] Talbot GH, Powers JH, Fleming TR, Siuciak JA, Bradley J, Boucher H, et al.
Progress on developing endpoints for registrational clinical trials of
community-acquired bacterial pneumonia and acute bacterial skin and skin
structure infections: update from the Biomarkers Consortium of the Foun-
dation for the National Institutes of Health. Clin Infect Dis 2012;55(8):
1114e21.
[12] U.S. Food and Drug Administration, Center for Drug Evaluation and Research
(CDER). Guidance for Industry: Acute Bacterial Skin and Skin Structure In-
fections: Developing Drugs for Treatment. 2013. http://www.fda.gov/
downloads/Drugs/.../Guidances/ucm071185.pdf.
[13] Friedland HD, O'Neal T, Biek D, Eckburg PB, Rank DR, Llorens L, et al. CANVAS 1
and 2: analysis of clinical response at day 3 in two phase 3 trials of ceftaroline
fosamil versus vancomycin plus aztreonam in treatment of acute bacterial
skin and skin structure infections. Antimicrob Agents Chemother 2012;56(5):
2231e6.
[14] Garau J, Blasi F, Medina J, McBride K, Ostermann H. Early response to anti-
biotic treatment in European patients hospitalized with complicated skin and
soft tissue infections: analysis of the REACH study. BMC Infect Dis 2015;15:78.
[15] Amara S, Adamson RT, Lew I, Huang X. Clinical response at Day 3 of therapy
and economic outcomes in hospitalized patients with acute bacterial skin and
skin structure infection (ABSSSI). Curr Med Res Opin 2013;29(7):869e77.
[16] Jaaskelainen IH, Hagberg L, From J, Schyman T, Lehtola L, Jarvinen A. Treat-
ment of complicated skin and skin structure infections in areas with low
incidence of antibiotic resistance-a retrospective population based study from
Finland and Sweden. Clin Microbiol Infect 2016;22(4):383.e1e383.e10.
[17] Hosmer DW, Lemeshow S, Sturdivant RX. Applied Logistic Regression. 3rd ed.
Hoboken, New Jersey: John Wiley & Sons, Inc.; 2013.
[18] Akaike H. A new look at the statistical model identification. IEEE Trans
Automat Contr 1974;19(6):716e23.
[19] Jaakola S, Lyytika €inen O, Rimhanen-Finne R, Salmenlinna S, Vuopio J,
Roivainen M, et al. Tartuntataudit Suomessa 2011. 2012. ISBN:
978e952e245e658e8, http://urn.fi/URN.
[20] Johnson AP. Methicillin-resistant Staphylococcus aureus: the European land-
scape. J Antimicrob Chemother 2011;66(Suppl 4):iv43e8.
[21] Baggs J, Fridkin SK, Pollack LA, Srinivasan A, Jernigan JA. Estimating National
Trends in Inpatient Antibiotic Use Among US Hospitals From 2006 to 2012.
JAMA Intern Med 2016;176(11):1639e48.