T h e NE W E NGL A ND JOU R NA L o f M E DICINE

ORIGINAL ARTICLE
ki, R. Askari, M.A.
Wilson, L.M.
Tria Napolitano, N.
Namias, P.R. Miller,
l of E.P. Dellinger, C.M.
Sho Watson, R. Coimbra,
D.L. Dent, S.F.
rt- Lowry,* C.S.
Cocanour, M.A. West,
Cou K.L. Banton, W.G.
Cheadle,
rse P.
Anti A.
Li
mic ps
ett

The authors full

robi ,
C.
names, academic
de-grees, and
al A.
G
affiliations
listed in
are
the The ui
Appendix. dr
Address
requests to Dr.
reprint rapy y,
an
Sawyer at the
Department of
for d
K.
Surgery,
University of Intr Po
aab
Virginia Health
po
System, Box
800709, vs
Charlottesville, VA
22908-0709, or at
do ky
rws2k@virginia.e
du. min ABS TR
*Deceased. al AC T
N Engl J
Med Inf
BACKGROUND
2015;372:
1996-
2005.
ecti
The
treatment
successful
of
DOI:
10.1056/N
on
intraabdominal
EJMoa141
infection requires a
1162 R.G. combination of ana-
Copyright 2015 Sawye tomical source control
Massachusetts
Medical Society. r, J.A. and antibiotics. The
Claridg appropriate duration of
e, A.B. antimicrobial therapy
Nathen remains unclear.
s, O.D.
Rotstei METHODS
n, T.M. We randomly assigned
Duane, 518 patients with
H.L. complicated
Evans, intraabdominal
C.H. infection and adequate
Cook, source control to
P.J. receive antibiotics until
ONeill, 2 days after the
J.E. resolution of fever,
Mazus leukocytosis, and ileus,
 with arecurren duration of antibiotic
maximum oft therapy was 4.0 days
10 days ofintraabd (interquartile range, 4.0
therapy ominal to 5.0) in the
(control infectio experimental group, as
group), or ton, or compared with 8.0
receive adeath days (interquartile
fixed courseoccurre range, 5.0 to 10.0) in
of antibioticsd in 56 the control group
(experimental of 257 (absolute difference,
group) forpatients 4.0 days; 95% CI,
41 calendarin the 4.7 to 3.3; P<0.001).
days. Theexperim No signifi-cant
primary ental between-group
outcome wasgroup differences were found
a composite(21.8%) in the individual rates
of surgical-, as of the compo-nents of
site infection,compare the primary outcome or
recurrent d with in other secondary
intraab- 58 of outcomes.
dominal 260
infection, orpatients CONCLUSIONS
death withinin the In patients with
30 days aftercontrol intraabdominal
the indexgroup infections who had
source- (22.3%) undergone an adequate
control proce-(absolut source-control
dure, e procedure, the outcomes
according todifferen after fixed-duration
ce, 0.5 antibiotic therapy
treatment
percenta (approxi-mately 4 days)
group.
were similar to those
Secondary ge
point; after a longer course of
outcomes
antibiotics (ap-
included the95%
proximately 8 days) that
duration ofconfide
extended until after the
therapy andnce
resolution of
rates ofinterval
physiological
subsequent [CI],
7.0 to abnormalities. (Funded
infections.
8.0; P= by the National
0.92). Institutes of Health;
RESULTS
STOP-IT ClinicalTrials
Surgical-site The
infection, median .gov number,
NCT00657566.)
.No ri
other g
uses ht
without s
permiss re
1996 ion. s
N ENGL J
MED Copyri er
372;21 ght v
NEJM.ORG 2015 e
MAY 21,
2015
Massac d.
husetts
Medica
TheNewEnglandJournalofMedicine
Downloadedfromnejm.orgon l
January30, Society
2016.For .All
ANTIMICROBIAL THERAPY FOR INTRAABDOMINAL INFECTION
adequate source We
control, a shorter tio
C
conducted the
OMPLICATED
course of 3 to 5 randomized nal
INTRAABDOMINALdays should Study to Opti- str
INFEC-tion continues tosuffice for cure9 mize Peritoneal ate
be a common problemand could de- Infection
worldwide. gy
Approximately 300,000crease the risk of Therapy (STOP- of
cases of appendicitisantimicrobial IT) trial to ad
occur each year in theresistance. Cur-
United States,1 compare two mi
and at least twice thatrently used strategies nis
many cases of non- guidelines, guiding the tra
appen-diceal infectionincluding those duration of tio
published jointly antimicrobial
require management.2 n
by the Surgical therapy for the
Morbidity ranges from of
5% among patientsInfection Society management of ant
evaluated in broad(SIS) and the complicated ibi
Infectious
observational studies2-4 intraabdominal oti
Diseases Society
to close to 50% in infection. We cs
of America
some cohorts, such as hypoth-esized un
(IDSA),
the elderly or critically that the til
recommend a
ill.5,6 De-spite the administration 2
treatment course
diversity of specific of fixed- da
of 4 to 7 days,
processes in these duration ys
de-pending on
infections, the basic antibiotic aft
the clinical
tenets of management therapy (4 days)
7,8 er
are similar: resuscitateresponse. after source the
patients who have the Despite these control would
systemic inflammatoryrecommendatio res
lead to ol
response syndromens, equivalent
(SIRS), control theobservational uti
outcomes and a on
source ofstudies show
shorter duration
contamination, removethat therapy is of
most of the infected ortypically of therapy as the
necrotic material, andadministered for compared with ph
administer anti-10 to 14 the tradi- ysi
microbial agents todays.4,10,11 One ol
eradicate residualreason that og
7,8
pathogens. shortening -
Antimicrobial therapy has ica
therapy for thebeen difficult is l
management ofthe 20% rate of ab
intraabdominal clini-cally no
infections continues tosignificant rm
evolve. Publishedinfectious
ali
guidelines includecomplications
tie
recommendations forafter treatment.3
s
appropriate These
antimicrobial agentssubsequent
rel
on the basis of high-complications, ate
d
quality evidence.7,8how-ever, are
to
The appropriate dura-often due to
tion of therapy,progression of SI
however, remainsthe original RS
unclear. Tradi-tionally,disease or .
practitioners haveinadequate
treated patients untiloriginal source METHODS
all evidence of SIRScontrol and may
has resolved, typicallynot be ST
U
for 7 to 14 days. Morepreventable with
DY
recently, it has beenantimicrobial P
suggested that withtherapy alone. O
PULATION procedure ganization paid
Patients were eligible(experimental by the sponsor.
for enrollment in thegroup) or to Each study site
study if they were 16receive
years of age or older;antimicrobial
if they pre-sentedtherapy until 2
with a complicateddays after the
intraabdominal infec-resolution of the
tion with either feverphysiological
(temperature abnormalities
38.0C), leu-related to SIRS
kocytosis (11,000(con-trol group).
peripheral white cellsThis resolution
per cubic millimeter),was defined as a
or gastrointestinalbody
dysfunction due totemperature of
peritonitis precludingless than 38.0C
intake of more thanfor 1 entire
half their normal diet;calendar day,
and if they hadnormalization of
undergone anthe peripheral
intervention towhite-cell count
achieve sourceto less than
control. 11,000 per cubic
Source control,millimeter, and
defined as proceduresthe patients
that eliminateability to
infectious foci,consume more
control factors thatthan half of his
pro-mote ongoingor her regular
infection, and correctdiet without
or control anatomicaladverse effects.
derangements to All the
restore normal phys-patients or their
iological function, islegal surrogates
critical to theprovided written
management of anyinformed
infection.12 Theconsent. The
adequacy of sourcespecific choice
control wasof antimicrobial
confirmed by theagents was not
local investigator anddictated by the
the principalprotocol but
investigator of therather was
overall study (see theconsidered to be
full protocol,acceptable if it
available with the fullwas consistent
text of this article atwith published
NEJM.org). SISIDSA A
guidelines.7,8 Qui
STUDY DESIGN Randomizati ck
AND OVERSIGHT on was Tak

In this investigator-performed with e

su
initiated, open-label,the use of a mm
multi-center trial, weWeb-based data ary
randomly assignedsystem managed is
participants in a 1:1by Merge ava
ratio to receive 4 fullHealthcare, an ilab
le
days of antimicrobialindependent
at
therapy after theircontract NE
index source-controlresearch or- JM.
org
MAY 21, 17
2015
N ENGL J MED 9
372;21
NEJM.ORG 9
ew Medic
o ht2015
T Engla ine n Massachusetts
h nd
Downloadedfrom J Medical
e Journ
nejm.org a Society.All
N alof n rightsreserved.
 T h e NE W E NGL A ND JOU R NA L o f M E DICINE
used a unique randomization sequence. Only 20
patients with appendiceal disease were permitted
in each sequential block of 200 randomly assigned
patients. Patients were enrolled at 23 sites through-
out the United States and Canada, and the study
was coordinated through the SIS and managed
by the University of Virginia team. The protocol
was approved by the institutional review board at
each site. All the authors vouch for the accuracy
and completeness of the data, the analyses re-
ported here, and the fidelity of the study to the
protocol.
DURATION OF THERAPY AND ADHERENCE
TO THE PROTOCOL
After randomization, patients were followed to
assess their clinical course and adherence to the
protocol. In the control group, adherence to the
protocol was defined as the receipt of appropri-
ate antimicrobial agents until 21 calendar days
after the first day that the patient had a maxi-
mum temperature of less than 38.0C for 1 whole
calendar day, less than 11,000 peripheral white
cells per cubic millimeter, and the ability to
meet more than half their nutritional needs en-
terally. A maximum of 10 days of therapy was
allowed for treatment of the initial intraabdomi-
nal infection.
In the experimental group, adherence to the
protocol was defined as the receipt of effective
antimicrobial agents for 41 calendar days after
the index source-control procedure. Patients were
followed for 30 days after the initial source-con-
trol procedure and assessed for infectious com-
plications, use of antimicrobial therapy, and death
from any cause. Patients hospitalized for more
than 30 days were followed to determine hospi-
tal length of stay. A resistant pathogen was de-
fined as methicillin-resistant Staphylococcus aureus,
any vancomycin-resistant enterococcus species,
or a gram-negative organism that was resistant
to all members of a major class of antimicrobial
agents.
STATISTICAL ANALYSIS
The primary analysis compared the proportion of
patients in each group in whom a surgical-site in-
fection or recurrent intraabdominal infection de-
veloped or in whom death occurred within 30 days
after the index source-control procedure. Analy-
ses were performed with the use of the chi-square
test with Yates correction, Students t-test, or Wil-
1998 N ENGL J MED 372;21 NEJM.ORG MAY 21, 2015
TheNewEngland
Journalof
Medicine
Downloadedfromnejm.orgonJanuary30,2016.
coxon signed-rank test, where applicab
intention-to-treat population, which co
all patients who provided consent and
randomization. KaplanMeier curves f
an event were constructed and compar
the use of the log-rank test.
Secondary analyses assessed the du
antimicrobial therapy for the index inf
overall exposure to antimicrobial agen
subsequent extraabdominal infection,
herence to the protocol. Prespecified s
for analysis included patients with an Ac
ology and Chronic Health Evaluation (A
score of 10 or higher (on a scale of 0 t
higher scores indicating an increased r
death), patients with a health careasso
infection, patients with an appendiceal
index infection and those with a non-a
ceal source of index infection, and pat
an index infection treated by surgical d
and those whose index infection was t
percutaneous drainage. Multivariate lo
gression was performed to determine a
tions with the composite outcome, inc
demographic variables and treatment-g
signments.
According to the original sample-si
tion to define equivalence between the t
we calculated that a sample of 505 pat
group would be required to give the st
power to detect a 10% difference in co
rates, assuming a 30% complication ra
controls and assuming a dropout rate o
an alpha level of 0.05. After the first in
analysis showed nearly identical outco
two cohorts, a competitive renewal req
continued support to reach the targeted
levels was not funded owing largely to
for futility.
R ESULTS
PATIENT CHARACTERISTICS AND INDEX IN
From August 2008 through August 201
of 518 patients underwent randomizati
patient withdrew consent after random
Thirty-day follow-up was completed i
the patients (Fig. 1).
The mean (SE) age of the pati
52.21.0 years (range, 16 to 88); most
tients were male, and the composition
and ethnic groups was similar in the tw
Forpersonaluseonly.Nootheruseswithout
permission.
Copyright2015Massachusetts
MedicalSociety.Allrights
reserved.
 ANTIMICROBIAL THERAPY FOR INTRAABDOMINAL INFECTION

518 Patients were enrolled and underwent randomization

260 Were assigned to control group
189 Received assigned intervention
71 Did not receive assigned intervention
26 Received more treatment than required
on basis of physiological findings, but
for <10 days total
35 Received treatment for >10 days
6 Had new surgical-site infection
8 Had recurrent intraabdominal infection
1 Had infectious disease owing to extra-
abdominal infection
20 Had no identifiable reason to receive
>10 days of treatment
10 Received therapy for too few days for
no identifiable reason
258 Were assigned to experimental group
211 Received assigned intervention
47 Did not receive assigned intervention
owing to >5 days of treatment
16 Had ongoing elevated white-cell count
12 Had gastrointestinal dysfunction
6 Had no identifiable reason to receive
>5 days of treatment
2 Had persistent fever
2 Had new surgical-site infection
7 Had recurrent intraabdominal infection
1 Had extraabdominal infection
1 Withdrew consent after receiving antibiotic
therapy

260 Were included in 30-day follow-up 257 Were included in 30-day follow-up

260 Were included in primary intention-to-treat 257 Were included in primary intention-to-treat
analysis analysis

189 Were included in evaluation of patients who 211 Were included in evaluation of patients who
adhered to study protocol adhered to study protocol

Figure 1. Enrollment, Randomization, and Treatment.

All enrolled patients had complicated intraabdominal infection and adequate source control. The control group con-
sisted of patients who received antibiotics until 2 days after resolution of fever, leukocytosis, and ileus, with a maxi-
mum of 10 days of therapy, and the experimental group received a fixed course of antibiotics for 41 calendar days.
372;21

days,
The mean APACHE II score for the index including 7
infec-tion was 10.10.3 (range, 0 to 29), the patients
most com-mon origin of the infection was the who
colon or rec-tum, and one third of the continued
infections were treated with a percutaneous to have an
procedure (Table 1). Com-plete demographic elevated
data are provided in Table S1 in the white-cell
Supplementary Appendix, available at NEJM count and 4
.org. There were no differences between the patients
two study groups, at a significance level of who had
less than 0.05, with respect to any ongoing
demographic variable. gastrointest
inal
ADHERENCE TO THE PROTOCOL dysfunctio
A total of 211 of 258 patients in the n when
experimental group (81.8%), as compared therapy
with 189 of 260 pa-tients in the control group was
(72.7%), ultimately re-ceived antimicrobial discontinue
therapy for the protocol-specified duration (P d. Among
=0.02) (Fig. 1). In the control group, 10 the other
patients received therapy for too few pa-tients
who did
not receive
N ENGL J MED
the assigned therapy, 26
received therapy for less
than 10 days (but lon-
ger than required
according to their
physiological findings)
and 35 received therapy
for more than 10 days.
In some patients in the
control group, the initial
course of antimicrobial
therapy was extended
owing to a second
infection, and they were
considered to be in
violation of the proto-
col; 6 of these patients
had a surgical-site infec-
tion, 8 had a recurrent
intraabdominal
infection, and 1 had an
infectious disease owing
to extra abdominal
infection.
In the experimental
group, all 47 patients
who did not adhere to
the protocol received a
treat-
NEJM.ORG MAY 21, 2015 1999
Downloadedfromnejm.orgonJanuary Nootheruseswithout Society.Allrightsreserved.
TheNewEngland 30,2016.For permission.Copyright
JournalofMedicine
personaluseonly. 2015MassachusettsMedical
 T h e NE W E NGL A ND JOU R NA L o f M E DICINE

PRIMARY AND MAJOR SECONDARY OUTCOMES

Table 1. Baseline Demographic and Clinical Characteristics, According to
Study Group.* The composite primary end point of surgical-site
infection, recurrent intraabdominal infection, or
Control Experimental death occurred in 56 of 257 patients in the experi-
Group Group
Variable (N=260) (N=258) mental group (21.8%), as compared with 58 of 260
patients in the control group (22.3%) (absolute
Age yr 52.21.0 52.21.0
difference, 0.5 percentage point, 95% confidence
Male sex no. (%) 145 (55.8) 144 (55.8) interval [CI], 7.0 to 8.0; P=0.92) (Table 2). Ka-
Race or ethnic group no. (%) planMeier analysis showed no significant be-
White 208 (80.0) 196 (76.0) tween-group difference in the time to the com-
Black 43 (16.5) 51 (19.8) posite primary outcome (Fig. 2).
Asian 5 (1.9) 6 (2.3) No significant between-group differences were
seen in the rates of the individual primary-end-
American Indian or Alaskan Native 2 (0.8) 1 (0.4)
point components of surgical-site infection (abso-
Hispanic no. (%) 20 (7.7) 15 (5.8)
lute difference, 2.2 percentage points; 95% CI,
Other 2 (0.8) 4 (1.6) 2.4 to 7.0; P=0.43), recurrent intraabdominal in-
Characteristics of index infection fection (absolute difference, 1.8 percentage points;
APACHE II score 9.90.4 10.30.4 95% CI, 4.5 to 7.8; P=0.67), and death (absolute
3
Maximum white-cell count per mm 15,6000.4 17,1000.7 difference, 0.4 percentage point; 95% CI, 1.7 to
Maximum body temperature C 37.80.1 37.70.1
2.7; P=0.99). Diagnosis of surgical-site infection
and recurrent intraabdominal infection, but not
Organ of origin no. (%)
death, occurred significantly later in the control
Colon or rectum 80 (30.8) 97 (37.6)
group than in the experimental group (Table 2).
Appendix 34 (13.1) 39 (15.1) Death occurred in 5 of the 518 patients in the
Small bowel 31 (11.9) 42 (16.3) two groups combined (1.0%, 2 patients in the
Source-control procedure no. (%) control group and 3 patients in the experimental
Percutaneous drainage 86 (33.1) 86 (33.3) group) at a mean of 18.70.4 days after the index
source-control procedure. All the deaths were
Resection and anastomosis or closure 69 (26.5) 64 (24.8)
judged by the site principal investigator and the
Surgical drainage only 55 (21.2) 54 (20.9)
study principal investigator to be related to under-
Resection and proximal diversion 27 (10.4) 37 (14.3) lying coexisting diseases (principally cancer and
Simple closure 20 (7.7) 12 (4.7) cardiovascular disease) and not to the initial in-
Surgical drainage and diversion 3 (1.2) 4 (1.6) traabdominal infection.
The median duration of antimicrobial treat-
* Plusminus values are means SE. There were no significant differences be- ment for the index intraabdominal infection was
tween the groups (P<0.05).
Race and ethnic groups were reported by the patient or surrogate. 4.0 days (interquartile range, 4.0 to 5.0) in the
Acute Physiology and Chronic Health Evaluation (APACHE) II scores range experimental group, as compared with 8.0 days
from 0 to 71, with higher scores indicating an increased risk of death. (interquartile range, 5.0 to 10.0) in the control
group (absolute difference, 4.0 days; 95% CI,
4.7 to 3.3; P<0.001). There were significantly
ment course that was longer than the duration fewer median antimicrobial-free days at 30 days
specified in the protocol. Of these patients, 16 had (including all antimicrobial therapy) in the con-
an ongoing elevated white-cell count, 2 had per- trol group than in the experimental group. No
sistent fever, and 12 continued to have gastroin- significant between-group differences were found
testinal dysfunction that prevented enteral intake. with respect to the rates of extraabdominal infec-
A new infection leading to extension of antimi- tion, Clostridium difficile infection, or secondary in-
crobial therapy for more than 4 days occurred in fections with resistant pathogens (Table 2).
10 patients, including 2 with a surgical-site infec-
tion, 7 with a recurrent intraabdominal infection, OUTCOMES IN PRESPECIFIED SUBGROUPS
and 1 with an extraabdominal infection. Two pa- The occurrence of the primary composite outcome
tients with recurrent intraabdominal infection also was similar in the two study groups in all pre-
had a persistently elevated white-cell count. specified subgroups, including patients who were

2000 N ENGL J MED 372;21 NEJM.ORG MAY 21, 2015

TheNewEnglandJournalofMedicine
Downloadedfromnejm.orgonJanuary30,2016.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2015MassachusettsMedicalSociety.Allrightsreserved.
 ANTIMICROBIAL THERAPY FOR INTRAABDOMINAL INFECTION

Table 2. Primary and Major Secondary Outcomes.*

Control Experimental
Group Group
Variable (N=260) (N=257) P Value
Primary outcome: surgical-site infection, recurrent intraabdominal 58 (22.3) 56 (21.8) 0.92
infection, or death no. (%)
Surgical-site infection 23 (8.8) 17 (6.6) 0.43
Recurrent intraabdominal infection 36 (13.8) 40 (15.6) 0.67
Death 2 (0.8) 3 (1.2) 0.99
Time to event no. of days after index source-control procedure
Diagnosis of surgical-site infection 15.10.6 8.80.4 <0.001
Diagnosis of recurrent intraabdominal infection 15.10.5 10.80.4 <0.001
Death 19.01.0 18.50.5 0.66
Secondary outcome
Surgical-site infection or recurrent intraabdominal infection with 9 (3.5) 6 (2.3) 0.62
resistant pathogen no. (%)
Site of extraabdominal infection no. (%)
Any site 13 (5.0) 23 (8.9) 0.11
Urine 10 (3.8) 13 (5.1) 0.65
Blood 3 (1.2) 5 (1.9) 0.71
Lung 3 (1.2) 3 (1.2) 0.99
Area of skin other than surgical site 1 (0.4) 4 (1.6) 0.36
Vascular catheter 0 (0) 2 (0.8) 0.47
Clostridium difficile infection no. (%) 3 (1.2) 5 (1.9) 0.71
Extraabdominal infection with resistant pathogen no. (%) 6 (2.3) 2 (0.8) 0.29
Duration of outcome days
Antimicrobial therapy for index infection <0.001
Median 8 4
Interquartile range 510 45
Antimicrobial-free days at 30 days <0.001
Median 21 25
Interquartile range 1825 2126
Hospitalization after index procedure 0.48
Median 7 7
Interquartile range 411 411
Hospital-free days at 30 days 0.22
Median 23 22
Interquartile range 1826 1626

* Plusminus values are means SE.

Some patients had extraabdominal infections at more than one site.
intention-to-treat
tary Appendix). analysis, with the
treated per protocol (33 of 189 patients in the control Differences in the exception of the
group [17.5%] and 37 of 211 patients in theduration of therapy subgroup of patients
experimental group [17.5%]; absolute differ-ence,among patients in all who were not treated
0.1 percentage point; 95% CI, 7.4 to 7.6; P = 0.72) subgroups were similar according to the
(Fig. 3, and Table S2 in the Supplemen- to those seen in the
N ENGL J MED 372;21 NEJM.ORG MAY 21, 2015 2001
Downloadedfromnejm.orgon personaluse permission.Copyright
TheNewEngland only.Noother 2015MassachusettsMedical
January30,
Journalof
2016.For useswithout Society.Allrightsreserved.
Medicine
 T h e NE W E NGL A ND JOU R NA L o f M E DICINE

Traditionally, physicians have administered

1.0 antimicrobial therapy in patients who have in-
0.9
Control group traabdominal infections until clinical and labo-

0.8 ratory evidence suggests that the infection has

Experimental group resolved. They reasoned that ongoing sepsis was
EventsNoofProbability

0.7
0.6 indicative of ongoing replication of pathogens.
0.5
More recent experimental data, however, suggest
that a prolonged SIRS may be more a reflection
0.4
of host immune activity than an indication of
0.3

0.2 such, efforts have

the presence begun
of viable to shorten the duration
microorganisms. 13-15
As
P=0.96 by log-rank test
0.1 of antimicrobial therapy in the presence of tra-
0.0 ditional markers of sepsis. These efforts have
0 5 10 15 20 25 30 already been successful in other severe infec-
Days to Composite Outcome tions such as ventilator-associated pneumonia.16
No. at Risk Currently, the average duration of antibiotic
Control 260 255 243 228 219 210 205 therapy for intraabdominal infection is 10 to 14
group
Experimental 258 253 227 214 208 203 202 days.4,10,11 The results of smaller studies of the
group
effect of an abbreviated course of antimicrobial
therapy have been published. Schein et al. reported
Figure 2. KaplanMeier Time-to-Event Curves for the Composite Primary
on an uncontrolled study in which 23 consecutive
Outcome, According to Treatment Group.
patients with diffuse peritonitis were assigned
The composite primary outcome was surgical-site infection, recurrent in- to receive 3 to 5 days of antibiotics.9 Infections
traabdominal infection, or death.
developed in 22% of these patients; these rates
were similar to those seen in a historical cohort.
protocol, for whom the median duration was 11 Basoli et al., who randomly assigned 90 patients
days in both groups. Results of the logistic- with mild-to-moderate intraabdominal infection
regression analysis are provided in Table S3 in to either 3 days or 5 or more days of ertapenem
the Supplementary Appendix; there was no sig- therapy, found no between-group difference in
nificant interaction between treatment groups infectious outcomes.17 Those findings, however,
and the primary composite outcome. are not generalizable to the majority of patients
with intraabdominal infection, since half the pa-
tients had appendiceal disease and the overall rate
DISCUSSION of infectious complications was less than 10%. As

In this randomized study involving patients with
complicated intraabdominal infections, a fixed
duration of 4 days of antibiotic treatment re-
sulted in outcomes that were similar to those of
a traditional, longer course that was based on
resolution of physiological abnormalities, and
the shorter course was associated with signifi-
cantly fewer days of antibiotic exposure. These
data provide support for the concept that after
an adequate source-control procedure, the ben-
eficial effects of systemic antimicrobial therapy
are limited to the first few days after interven- to affect infectious outcomes. Truly clinically sig-
tion. In addition, it might be argued that the
delay in manifestation of infectious complica-
tions in the control group was itself an adverse
outcome, since the time to recognition of these
events and, therefore, the overall time to resolu-
tion of all infections was prolonged.
compared with these studies, the STOP-IT trial
had several advantages, including a larger sample
size, randomized design, and enrollment of pa-
tients with a broader range of severity of illness.
The rate of infectious complications was more
than 20% in both groups of our study, and most
of these complications were recurrent intraab-
dominal infections. Given the large difference in
the number of days of treatment in the two study
groups, neither shortening nor lengthening the
duration of antimicrobial therapy appears likely
nificant improvement in the management of this
disease, therefore, probably awaits more effective
technical or immune responsemodifying inter-
ventions. However, the observed mortality in our
study was only 1%; this is better than anticipat-
ed among patients with a mean APACHE II score

2002 N ENGL J MED 372;21 NEJM.ORG MAY 21, 2015

 TheNewEnglandJournalofMedicine
Downloadedfromnejm.orgonJanuary30,2016.Forpersonaluseonly.Nootheruseswithoutpermission.
Copyright2015MassachusettsMedicalSociety.Allrightsreserved.
 ANTIMICROBIAL THERAPY FOR INTRAABDOMINAL INFECTION

No. of Days of Antibiotic

Subgroup Patients Therapy Proportion with Composite Outcome
median
(interquartile range)
Adhered to protocol
Control 189 7 (510)
Experimental 211 4 (45)
Did not adhere to protocol
Control 71 11 (717)
Experimental 47 11 (819)
APACHE II score 10
Control 120 8 (510)
Experimental 122 4 (45)
Health careassociated infection
Control 94 8 (510)
Experimental 102 4 (45)
Percutaneous drainage
Control 86 8 (510)
Experimental 86 4 (45)
Surgical drainage
Control 174 8 (510)
Experimental 171 4 (45)
Appendiceal source
Control 34 8 (510)
Experimental 39 5 (46)
Non-appendiceal source
Control 226 8 (510)
Experimental 218 4 (45)
0.0 0.1 0.2 0.3 0.4 0.5

Figure 3. Primary Composite Outcome in Key Subgroups.

The median proportions of patients with the composite outcome are shown. I bars indicate the interquartile range.
equivalence cannot be
these populations claimed, although the
of 10 and is perhaps related to the reduction in would benefit from a 95% confidence
mortality from many serious infections notedlonger duration of interval for the
after the initiation of the Surviving Sepsistherapy. Second, the difference in event
Cam-paign.18 rate of nonad-herence rates was 7.0 to 8.0
Strengths of the trial include the inclusionto the protocol was and provides
of 23 centers and the similarity in outcomes moderately high, encouraging evidence
between the control group and theincluding 18% of that the true difference
experimental group in a wide variety of patients in the is less than 10%. In
important, prespecified subgroups. Forexperimental group; other words, the null
example, although rates of complicationsthis created bias toward hypothesis of equal
differed according to the severity of illness and the null hypoth-esis of efficacy can-not be
the method of source control (percutaneous vs. no difference in rejected.
surgical drainage), the effect size was similartherapy, though the fact In conclusion,
in the two study groups. These findings suggest that the per-protocol outcomes in patients
that the results may well be generalizable population had similar with in-traabdominal
across a wide array of patient populations, care com-plication rates infections who have
settings, and interventions, as long as adequate between the undergone a successful
source control is first achieved. experimental and source-control
Several limitations to the interpretation of control groups is procedure and receive a
these data are worth noting. First, patientsreassuring. Finally, the fixed, 4-day course of
without adequate source control were excluded, original calculated antimicrobial therapy
and only a small number of patients with sample size to assert ap-pear to be generally
immunosuppres-sion were included; it remainsequivalence be-tween similar to outcomes in
unclear whether groups was not pa-
achieved and proof of
N ENGL J MED 372;21 NEJM.ORG MAY 21, 2015 2003
Downloadedfromnejm.orgon personaluse permission.Copyright
TheNew only.Noother 2015MassachusettsMedical
January30,
EnglandJournal
2016.For useswithout Society.Allrightsreserved.
ofMedicine
T h e NE W E NGL A ND JOU R NA L o f M E DICINE
full text of this
for serving on article at NEJM.org.
tients in whomadvisory boards
from AstraZeneca,
systemic Cubist, Merck, and
antimicrobial Pfizer, consulting
agents arefees from Bayer,
lecture fees from
administered Merck, and grant
until after thesupport from
AstraZeneca, Bayer,
resolution of
Merck, and Pfizer;
signs andand Dr. Dellinger,
symptoms offees for serving on
advisory boards
sepsis. from Merck,
Baxter, Ortho-
Presented in partMcNeil, Targanta
at the 34th AnnualTherapeutics,
Meeting of theSchering-Plough,
Surgical InfectionAstellas,
Society, Baltimore,CareFusion, Durata
May 13, 2014. Therapeutics,
Supported byPfizer, Rib-X
grants Pharmaceuticals,
(R01GM081510 andAffinium
T32 AI078875) fromPharmaceuticals,
the NationalTetraphase Phar-
Institutes of Health. maceuticals, and R-
Dr. SawyerPharm, and lecture
reports receivingfees from Applied
consulting fees fromMedi-cal. No other
3M and Pfizer; Dr.potential conflict of
Duane, lecture feesinterest relevant to
from Pfizer; Dr.this article was
ONeill, fees forreported.
serving on a surgical Disclosure forms
review panel forprovided by the
Cubist; Dr. Mazuski,authors are
fees available with the

Napolitano M.D., (J.A.C.); the Department of

, M.D., Christophe Surgery, University of
Nicholas
APPENDIX r A.Toronto, Toronto (A.B.N.,
The Namias, Guidry, O.D.R.); the Depart-ment
authors fullM.D., M.D., andof Surgery, University of
names andPreston R. Kimberley Washington, Seattle
academic Miller, Popovsky, (H.L.E., E.P.D.); the
degrees areM.D., E. B.S.N. Department of Surgery,
as follows:Patchen The Beth Israel Deaconess
Robert G.Dellinger, authors Medical Center (C.H.C.),
Sawyer, M.D., affiliations and the Department of
M.D., Christophe are asSurgery, Brigham and
Jeffrey A.r M. follows: theWomens Hospital (R.A.)
Claridge, Watson, Department both in Boston; the
M.D., AveryM.D., Raul of Surgery,Department of Surgery,
B. Nathens,Coimbra, University Mari-copa Integrated
M.D., OriM.D., of VirginiaHealth System, Phoenix,
D. Rotstein,Daniel L. Health AZ (P.J.O.); the
M.D., Dent, System, Department of Surgery,
Therese M.M.D., Charlottesvi Washington University, St.
Duane, Stephen F. lle (R.G.S.,Louis (J.E.M.); the Depart-
M.D., Lowry, C.A.G., ment of Surgery, VA
Heather L.M.D., K.P.); thePittsburgh Healthcare
Evans, Christine Department System, Pittsburgh (M.A.
M.D., S. of Surgery,Wilson); the Department of
Charles H.Cocanour, Virginia Surgery, University of
Cook, M.D.,M.D., Commonwe Michigan, Ann Arbor
Patrick J.Michaela alth (L.M.N.); the Department
ONeill, A. West, University, of Surgery, University of
M.D., M.D., Richmond Miami Miller School of
Ph.D., JohnPh.D., (T.M.D.); Medicine, Miami (N.N.);
E. Mazuski,Kaysie L. the the Department of Surgery,
M.D., Banton, Department Wake Forest School of
Ph.D., RezaM.D., of Surgery,Medicine, Winston-Salem,
Askari, William G. Case NC (P.R.M.); the
M.D., MarkCheadle, Western Department of Surgery,
A. Wilson,M.D., Reserve University of South
M.D., LenaPamela A. University, Carolina, Columbia
M. Lipsett, Cleveland (C.M.W.); University of
California, California, Science Newark (S.F.L.); the
San Diego,San Center atDepartment of Surgery,
San DiegoFrancisco, San University of Minnesota
(R.C.), theSan Antonio, Medical School,
Department Francisco San Minneapolis (K.L.B.); the
of Surgery,(M.A. Antonio Department of Surgery,
UC DavisWest) all (D.L.D.); University of Louisville
Medical in the School of Medicine,
Center, California; Department Louisville, KY (W.G.C.);
Sacramento the of Surgery,and the Department of
(C.S.C.), andDepartment University Surgery, Johns Hopkins
the of Surgery, of MedicineUniversity School of
Department University and Medicine, Baltimore
of Surgery,of Texas Dentistry of(P.A.L.).
University ofHealth New Jersey,
intra- 5:417- o n B,
abdom- 24. X, Cha
REFER inal 5. Chr al. Enterococci Sn inf bok
ENCES infectio increase theche ect A,
n is istou et
morbidity andz io
1. DeF associa NV,
mortality Gar ns: al.
rances ted Barie Cha
associated with ca an
CJ, with PS, nge
Delling severe in-tra-M, an
Cullen intrinsi Bas al s in
KA, c risk er EP,abdominal the
infections in sett ysi
Kozak factors Wayma aero
elderly pa-tients i s
LJ. rather ck JP, bic
M, ba
National than Stone hospitalized in the faec
HH. intensive careet se
Hospital the al. d al
Dischar source Surgica unit. J Antimicrob flor
Chemother 2011; Saf on
ge of in- l Infec- a of
66: 2379-85. ety fiv
Survey: fection. tion pati
and e
2005 Surger Society ents
tole Eu
annual y intra- 7. Solomkin JS,
rabi ro- trea
summar 2009;1 abdomi Mazuski JE, lity ted
pe
y with46:654- nal Bradley JS, et al. of wit
an
detailed 61. infectio Diagnosis and tige h
ob
diagnosi n management of cy- anti
s and 4. Ric study: com-plicated
ser
biot
va
procedurcio prospecintra-abdominal clin ics
e tio
e data.LM, tive infection in na for
Popovs evaluat adults for
Vital and l acut
the
Health ky KA, ion ofchildren:
stu e
Stat 13Hranje manageguidelines by the trea intr
tme di
2007;16 c T, et -ment Surgical In- es. a-
5:1-209. al. techniq fection Society nt abd
J
Associ ues andand the Infectious of
2. Sarte ation of outcom Diseas-es Society co
A omi
nti nal
lli M,excessi e. Archof America. Surg mpl infe
mi
Catena F,ve Surg Infect (Larchmt) icat ctio
cr
Ansaloni duratio 1993;1 2010;11:79-109. ed n.
L, et al.n ski ob
of 28:193-
Complic antibiot 8. 8. Solomkin JS,n Ch Sca
nd J
ated Mazuski JE,and e
ic
intra- therapy
6. Du Bradley JS, et al.soft m Infe
ponDiagnosis ot ct
abdomin for and-
t H,management he Dis
al intra- oftiss
Fri com-plicated r 201
infection abdomi ue
gge intra-abdominal 20 2;4
s nal in- and
ri 13 4:8
worldwidfection infection in adultsintr
A, ;6 20-
e: thewith and children:a-
To 8: 7.
definitivesubseq guidelines by theabd
uze
data ofuent
au
Surgical In-omi Su 12
pp
the extra- J, fection Societynal
l . M
CIAOW abdomi et and the Infectious 2:i arsh
Study. nal Diseases Society
World Jinfectio of America. Clin i3 all
Emerg n and Infect Dis 2010; 7- JC,
Surg 50:133-64. ii4 Mai
death:
2014;9: a study 4. er
9. Schein M,
37. of
Assalia A, 1 RV,
Jim
3. Inui 2,552
con-
Bachus H. Mini- 1. ene
T, mal antibiotic Sa z
secutiv
Haridas therapy after m M,
e
M, emergency ue Dell
infectio
Claridge abdominal lss inge
ns.
JA, surgery: a on r
Surg
Malango prospective A, EP.
Infect
ni MA. study. Br J Surg Isa Sou
(Larch
Mortalit 1994;81:989-91. ks rce
mt)
y for
2014;1 10. Guira so cont
rol in :Suppl mito- me and quantify do lev M
the :S513- chondr inflam-matory not els. a
man- S526. ial tissue injury in cha Sh k
agemen DNA nonhuman pri- nge ock
t of
13. disting mates. Shock mor 20 S
severe Sursal uish 2013;39:55-62. talit 04; K
sepsis T, bacteri y in 21: ,
and Stearn al sep-
14. Turnb
mic 12 e
septic s- sis ull IR, Javadi P,e 1- t
shock: Kuros from Buchman TG,with 5. a
an awa sterile Hotchkiss RS,mar l
evidenc DJ, system Karl
15.
IE,kedl .
e-based Itaga- ic Coopersmith CM.y Wong
YY
review. ki K, inflam Antibiotics elev
et al. improve survival ,
Crit matory ated
Plasm in sepsis in- Wo
Care respon inter
Med a se dependent ofleuk ng
PN
2004;32 bacteri syndro injury severity butin 6 ,
al and
J 2015 m.orgon dical
MED
372; January30, Society.
21 TheNewEnglandJournalofMedicine
2016.For Allrights
NEJM Downloadedfrom
personal reserved.
2004 .ORG
N MAY useonly.
ENGL 21, Noother
ANTIMICROBIAL THERAPY FOR INTRAABDOMINAL INFECTION
intraabdomina
l infection. J
nia in adults: a 18. Dellinger
Gastrointest
Persistent sterile peritonealrandomized RP, Levy MM,
Surg
inflammation after cathetertrial. JAMA Rhodes A, et
2008;12:592-
removal for refractory bac-2003;290:2588 al. Surviving
600.
terial peritonitis predicts full--98. sepsis
blown en-capsulating17. Basoli A, campaign:
peritoneal sclerosis. Perit DialChirletti P, interna-tional
Int 2013;33:507-14. Cirino E, et al. guidelines for
16. Chastre J, Wolff M, FagonA prospective, management of
JY, et al. Comparison of 8 vs 15double-blind, se-vere sepsis
days of antibiotic therapy for multicenter, and septic
ventilator-associated pneumo- randomized shock: 2012.
trial comparing Crit Care Med
ertapenem 3 vs 2013;41:580-
>or=5 days in 637.
community- Copyright 2015
acquired Massachusetts
Medical Society.
N ENGL J MED 372;21 NEJM.ORG MAY 21, 2015 2005
New Medicine
on Forpersonaluse Massachusetts
T England Januar only.Nootheruses MedicalSociety.
Downloadedfrom
h Journal
nejm y30, withoutpermission. Allrights
e of .org 2016. Copyright2015 reserved.