The Journal of Infectious Diseases

SUPPLEMENT ARTICLE

Antimicrobial Treatment Duration in Sepsis and Serious

Infections
Lindsay M. Busch and Sameer S. Kadri
Critical Care Medicine Department, National Institutes of Health Clinical Center, Bethesda, Maryland, USA

Sepsis mortality has improved following advancements in early recognition and standardized management, including emphasis on

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

early administration of appropriate antimicrobials. However, guidance regarding antimicrobial duration in sepsis is surprisingly
limited. Decreased antibiotic exposure is associated with lower rates of de novo resistance development, Clostridioides difficile-
associated disease, antibiotic-related toxicities, and health care costs. Consequently, data weighing safety versus adequacy of shorter
treatment durations in sepsis would be beneficial. We provide a narrative review of evidence to guide antibiotic duration in sepsis.
Evidence is significantly limited by noninferiority trial designs and exclusion of critically ill patients in many trials. Potential chal-
lenges to shorter antimicrobial duration in sepsis include inadequate source control, treatment of multidrug-resistant organisms,
and pharmacokinetic alterations that predispose to inadequate antimicrobial levels. Additional studies specifically targeting patients
with clinical indicators of sepsis are needed to guide measures to safely reduce antimicrobial exposure in this high-risk population
while preserving clinical effectiveness.
Keywords.  sepsis; infection; antibiotic; duration; length.

Sepsis mortality has declined significantly over the past 30 treatment trials (predominantly in pneumonia [19–21], and
years, driven largely by improvements in early recognition and intraabdominal [22] and urinary tract infections [23]) with
standardized management approaches [1, 2]. While the nuances limited representation of patients with sepsis and septic shock.
of some management strategies in sepsis such as fluid resuscita- In principle, the optimal duration of antibiotic therapy in
tion [3], serial laboratory monitoring [4–6], and corticosteroids sepsis would be one that maximizes clinical effectiveness while
[7, 8] are still being debated, the timely initiation of appropriate minimizing the antibiotic-associated risks such as toxicities,
antibiotic therapy remains an uncontested hallmark of suc- Clostridioides difficile-associated disease, and emergence of re-
cessful sepsis treatment. Myriad studies have highlighted the sistance, as well as health care costs. There are many host- and
value of appropriate (in vitro-active) empiric antibiotic choices pathogen-specific determinants impacting the required du-
in sepsis [9–11] and their early initiation, especially in septic ration of antibiotic therapy in sepsis, and extrapolation from
shock [12–15], and have even led to inclusion of early antibiotic healthier populations may be overly simplistic. Conspicuously
administration in national quality metrics to compare hospital few studies have investigated the optimal duration of antibiotic
performance [16, 17]. However, guidance is surprisingly lim- therapy in critically ill populations. Indeed, even the landmark
ited regarding the optimal duration of therapy for patients with sepsis trials which have shaped sepsis management over the
sepsis. The current Surviving Sepsis Campaign (SSC) guideline last 2 decades [4–6, 24–26] did not report any specific antibi-
makes a general recommendation that 7 to 10 days of antibi- otic regimens, durations, or evidence of microbiologic cure in
otic coverage is likely sufficient for most serious infections as- populations with culture-positive sepsis. As such, it is not sur-
sociated with sepsis and septic shock, although this course may prising that usual care durations of antibiotic therapy for sepsis
be lengthened in some scenarios (eg, undrained foci of infec- and serious infections remain highly variable [27]. A survey of
tion, Staphylococcus aureus bacteremia, and neutropenia) or health care professional users of a sepsis crowdsourcing applica-
shortened in others (eg, pyelonephritis and spontaneous bac- tion recently revealed an average reported duration of intrave-
terial peritonitis) [18]. The recommendation is graded as weak, nous antibiotic therapy for sepsis of more than 10 days for 17%,
with low-quality evidence, supported specifically by data from 7–10 days for 40%, 5–7 days for 27%, and 3–5 days for 13% of
respondents [28].
 
The mortality risk in sepsis is substantial and the margin
Correspondence: Sameer S. Kadri, MD, MS, FIDSA, Critical Care Medicine Department, for error small. Bedside providers have until recently been rel-
Clinical Center, National Institutes of Health, 10 Center Drive, Bethesda, MD 20892 (Sameer.
kadri@nih.gov). atively complacent with longer courses of therapy, potentially
The Journal of Infectious Diseases®  2020;222(S2):S142–55 due to the false sense of security it may offer for sicker patients.
Published by Oxford University Press for the Infectious Diseases Society of America 2020. However, a paradigm change has occurred in recent years [29]
This work is written by (a) US Government employee(s) and is in the public domain in the US.
DOI: 10.1093/infdis/jiaa247 and the importance and need for antibiotic stewardship is well

S142 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

recognized across the spectrum of providers and disciplines. In reduced treatment durations for each of these conditions, which
a retrospective cohort study of 7118 patients with severe sepsis have significantly changed practice in the last decade. These
or septic shock, Teshome et al [30] reported a 4% increased data present consistent themes of preserved treatment efficacy
risk of de novo antibiotic resistance for each additional day of with fewer antibiotic days and reduced adverse events. Below
antipseudomonal β-lactam exposure, highlighting the impor- are a few studies that have significantly contributed to the para-
tance of striving to determine and implement the minimum digm change toward shorter durations of antibiotic therapy. We
necessary duration of therapy, even in sepsis. Furthermore, have limited the focus of this review predominantly to clinical
recent data from Rhee et al demonstrated that among patients trials rather than observational studies given that observational
with culture-proven sepsis treated with adequate empiric anti- studies are biased towards better outcome in those with early
biotics, treatment with overly broad-spectrum antibiotics was discontinuation (see Table 1 for additional details on relevant
associated with a 20% increase in the odds of death, with a me- studies).

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

dian (interquartile range [IQR]) duration of treatment of 4 (2–6)
days for both antipseudomonal β-lactams and carbapenems • Ventilator-associated pneumonia (VAP): In 2003, Chastre et al
[31]. Given that the evidence base is currently insufficient to [19] published a landmark trial demonstrating that 8 days of
perform a systematic review in search of the optimal duration antibiotic treatment for VAP was noninferior to 15 days for
of antibiotic treatment in sepsis, we instead provide a narra- 28-day mortality and infection recurrence for all organisms
tive review of the existing literature, which can be leveraged to except for nonfermenting gram-negative bacilli including
inform the current practice of antibiotic therapy in sepsis, fo- Pseudomonas aeruginosa.
cusing specifically on the optimal duration rather than choice • Community-acquired pneumonia (CAP): Dunbar et al
of antimicrobial therapy. [37] demonstrated noninferior clinical response of short-
course, higher-dose levofloxacin (750 mg for 5 days) com-
IS SHORTER ALWAYS BETTER? pared with a longer course at a lower dose (500 mg for 10
The first step to determining the optimal duration of therapy days) as well as improved symptoms and defervescence
in sepsis and serious infections is to understand the origins of in the intervention arm. This study was followed by sev-
our usual care standard. Much of our modern antibiotic pre- eral others investigating other regimens. In a 2016 study
scribing practice has been based largely on expert opinion and of adults hospitalized for CAP, Uranga et al [38] found
influenced to an extent by historical lessons learned from the that a short course of 5 days of physician-determined anti-
treatment of tuberculosis. For the latter, success was directly biotics was noninferior to a physician-determined longer
linked to duration and resistance could occur in setting of inap- course with regard to clinical cure at 10 and 30 days, CAP
propriate dosing or monotherapy [32]. Early studies in patients symptoms, and 30-day mortality, and also found that the
with cystitis noted that single-dose therapy was suboptimal shorter-course patients had fewer readmissions within 30
compared to multiday therapy [33], establishing that most se- days. The current guidelines on diagnosis and treatment
rious infection would presumably at least require multiple days of CAP have incorporated these data into their recom-
of antibiotics. Regimens for acute bacterial infections evolved mendation of 5 days of antibiotic duration for all patients
to prolonged courses with the rationale of reducing relapses provided that they have demonstrated clinical improve-
and emergence of resistance from undertreated infections [34]. ment and were not diagnosed with either P. aeruginosa or
However, this evolution was based on a weak evidence base methicillin-resistant Staphylococcus aureus, for which they
(small studies, heterogenous populations, and subjective met- recommend 7 days [48].
rics for clinical response) and was often arbitrary with a pe- • Complicated urinary tract infection (cUTI): In a 2008 study
culiar penchant for 7-day increments [35]. This led to many of patients with acute pyelonephritis or cUTI, 5 days of high-
previous iterations of practice guidelines recommending itera- dose levofloxacin was noninferior to 10 days of ciprofloxacin
tive courses such as 1–2 weeks for community-acquired pneu- for microbiologic eradication and clinical success [40].
monia [36], 2 weeks for pyelonephritis [33], and 3–4 weeks for Subsequently, in a study of women with community-acquired
bacteremia [34]. pyelonephritis, 7 days of ciprofloxacin was noninferior to 14
Recent years have seen a consistent trend toward shorter days in clinical and microbiologic efficacy measures and
antibiotic treatment durations for many infectious syndromes longer-course therapy was associated with more oral candi-
including pneumonia (community-acquired and nosocomial), diasis [41].
cystitis, complicated urinary tract infections, intraabdominal • Neutropenic fever: In a 2017 study of high-risk neutropenic
infection, acute bacterial sinusitis, cellulitis and soft tissue infec- fever patients without a microbiologically diagnosed infec-
tion, septic arthritis, and chronic osteomyelitis [29, 35]. Some tion, empiric antimicrobial therapy was safely discontinued
examples of these studies are listed in Table 1. In fact, there after 72 hours of apyrexia in the intervention group com-
are now several examples in the “shorter is better” literature of pared to the control group in which empiric antimicrobial

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S143

 Table 1.  Trials of Reduced Antimicrobial Durations in Multiple Infectious Syndromes

Short Course Long Course

Infectious Primary No. of Patient Antibiotic: Antibiotic:
Syndrome Author Study Design Outcome Patients Inclusion Criteria Exclusion Criteria Location Severity of Illness Duration, d Duration, d Outcomes Comments

VAP Chastre Multicenter, 28-d mortality; 197 MV > 48 h; clinical SAPSII > 65; immu- ICU Short course: Adequate Adequate Primary ARRs: Noninferiority met
et al noninferiority RCT microbi- suspicion of VAP; nosuppression or SAPSII 45 (SD, abx per abx per all-cause
2003 ologically positive distal long-term corticosteroid 15), SOFA 7.3 (4); physician physician mortality 1.6
[19] documented airway culture; therapy; concomitant vasporessors 33% discretion: discretion: (90%
PNA recur- appropriate abx extrapulmonary infection long course: SAPSII 8 15 CI, −3.7
rence; abx- within 24 h of requiring >8 d abx 45 (15), SOFA 7.4 to 6.9);
free days culture (4); vasopressors pulmonary
35%; mechanical infection
ventilation 100% recurrence
2.9 (− 3.2
to 9.1); abx-
free days
4.4 (3.1–5.6)

S144 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

CAP Dunbar Multicenter, Clinical response 530 Mild-to-severe CAP Levofloxacin-resistant Inpatient PSI class I/II 58%; Levofloxacin Levofloxacin Clinical Noninferiority met;
et al double-blind, at follow-up; organism; previous or class III/IV 42% 750 mg: 5 500 mg: success rate short course
2003 noninferiority RCT 7–14 d post quinolone treatment outpa- 10 92.4% vs (higher dose) group
[37] medication failure; life expectancy < tient 91.1% defervescence
completion 72 h; neutropenia or HIV; earlier than that
empyema or effusion longer course
requiring chest tube
CAP Uranga Multicenter, Clinical cure at 312 Hospitalization for ICU admission before ran- Ward PSI short 81.8 (SD, Adequate Adequate abx Clinical cure No difference in any
et al noninferiority RCT 10 d; clinical CAP domization; immunosup- 33.8); PSI long 83.7 abx per and dura- 10 d: 53.6% primary outcomes;
2016 cure at 30 d; pression; HCAP; specific (33.7); vasopressors physician tion per vs 48.6%; significant reduction
[38] CAP symp- indication for longer dura- 1.6%; mechanical discre- physician clinical in duration of anti-
toms at 5 d tion; required chest tube ventilation 1% tion: 5 discretion cure 30 d: biotics and hospital
and 10 d 91.9% vs readmissions by
88.6%; CAP 30 d
symptoms
5 d: 27.2 vs
24.7; CAP
symptoms
10 d: 17.9 vs
18.6
CAP; Vaughn Multicenter, retro- Rate of excess 6481 Adult medical ICU admission; MV; severe Ward qSOFA short >2 9.8% Adequate Adequate Median excess Excess duration was
HCAP et al spective cohort antibiotic patients with immunocompromise; qSOFA long >2 abx per abx per duration: only associated with
2019 treatment community-onset Legionella or fungal 8.8% physician physician CAP 2 d patient-reported
[39] pneumonia (CAP pathogen; bacteremia or discretion: discretion: (IQR, 0–4), events (diarrhea,
or HCAP) empyema 5–7 > 5–7 HCAP 1 d GI distress, thrush
(0–3) most common)
cUTI Peterson Multicenter, Microbiologic 1093 Acute pyelonephritis Complete obstruction; Inpatient NR Levofloxacin Ciprofloxacin Microbiologic Noninferiority met;
et al double-blind, eradication or cUTI surgery or lithotripsy or 750 mg: 5 400/500 eradication: clinical success
2008 noninferiority RCT post therapy within 7 d; abx therapy outpa- mg: 10 79.8% vs comparable
[40] for concurrent infection; tient 77.5% between groups
quinolone-resistant path-
ogen; abscess, prosta-
titis, epidymitis

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

 Table 1.  Continued

Short Course Long Course

Infectious Primary No. of Patient Antibiotic: Antibiotic:
Syndrome Author Study Design Outcome Patients Inclusion Criteria Exclusion Criteria Location Severity of Illness Duration, d Duration, d Outcomes Comments
cUTI Sandberg Multicenter, Clinical and 156 Women with Systemic abx within 72 h; Inpatient NR Ciprofloxacin Ciprofloxacin Clinical cure: Noninferiority met;
et al double-blind, bacteriologic diagnosis of indwelling or intermittent or 500 mg: 7 500 mg: 97% vs long course signifi-
2012 noninferiority RCT efficacy 10–14 community- bladder catheterization; outpa- 14 96% cantly higher rate of
[41] d after treat- acquired CrCl < 0.5 mL/s tient oral candidiasis
ment pyelonephritis
NF Aguilar- Multicenter, open, Number of EAT- 157 Hematologic malig- Microbiologic diagnosis of Ward NR 72 h Apyrexia, Mean EAT-free Mean fever days and
Guisado superiority RCT free days nancies or HSCT infection or noninfectious apyrexia, symptom days: 16.1 all-cause mortality
et al with febrile neu- etiology for fever; CrCl symptom resolution, (SD, 6.3) vs was not different;
2017 tropenia without < 30 mL/min; receiving resolu- normal 13.6 (7.2) control group had
[42] microbiologic antibiotics for any reason tion and vital signs more grade 3–4
diagnosis prior to NF onset normal AND neu- adverse events than
vital signs tropenia the short course
resolved
BSI Daneman Multicenter, open Feasibility (re- 115 Positive blood Immunocompromise; ICU APACHEII 22 (IQR, Adequate Adequate Median recruit- 90-d mortality 15%,
et al pilot RCT cruitment, culture result with prosthetic heart valve 18–26); vasopres- abx per abx per ment rate ICU mortality 7%,
2018 adherence) pathogenic bac- or endovascular grafts; sors 52% physician physician 0.7 patients/ hospital mortality
[43] teria while in ICU established requirement discre- discretion: mo (IQR, 13%; duration MV
for extended treatment; tion: 7 14 0.3– 1.5); 8 d (3–21); relapse
Staphylococcus aureus or median BSI 4%; CDI 4%;
fungal BSI adherence secondary AMR
71% (50%– infection 9%
85%)
BSI Yahav Mulicenter, open, Composite: 90-d 604 Hospitalized Uncontrolled source; Ward; Presentation SOFA: Adequate Adequate Primary com- Noninferiority met;
et al noninferiority RCT mortality, adults with polymicrobial infection; ICU short course 2 (IQR, abx per abx per posite: risk secondary endpoints
2019 clinical failure, gram-negative immunosuppression 1–3), long 2 (1–3) physician physician difference not different except
[44] readmission, bacteremia Randomization SOFA: discre- discretion: −2.6 (CI, − time to return to
or LOS >14 d surviving to day 7 short 1 (0–2), long tion: 7 14 10.5 to 5.3) baseline activity,
of treatment and 2 (0–2) duration of antibiotic
clinically stable therapy, and total
antibiotic days (P <
.001)
IAI Sawyer Multicenter, open, Composite: 518 Complicated IAI Inadequate source control; NR APACHE II: 10.1 ± 0.3 Adequate Adequate Primary Secondary: no
et al superiority RCT surgical site having undergone high likelihood of death (range 0–29) abx per abx per composite: difference except for
2015 infection, an intervention for within 72 h; SBP physician physician ARR −0.5% duration of therapy
[22] recurrent IAI, source control discretion: discretion: (CI, −7.0% and abx-free days
30-d mortality 4 after 2 after to 8.0%;
source resolution P = .92)
control of SIRS
ABSSTI Prokocimer Multinational, Early clinical 667 Skin or soft Uncomplicated ABSSTI or NR NR Tedizolid Linezolid: 10 Early clinical Noninferiority was
et al double-blind, response tissue infection association with prosthetic PO: 6 response: met for primary and
2013 noninferiority RCT at 48–72 h accompanied device or vascular catheter 79.5% vs secondary endpoints
[45] assessment by regional or site, gram-negative 79.4%
systemic signs pathogen suspected
of infection; or documented (unless
gram-posi- wound infection); any
tive organism necrotizing process; septic
suspected/ shock or severe sepsis;
documented immunosuppression

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S145

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020
 therapy was continued until 72 hours of apyrexia and res-

Abbreviations: ABSSTI, acute bacterial skin and soft tissue infection; abx, antibiotics; AMA, against medical advice; AMR, antimicrobial resistance; APACHE II, acute physiology + age points + chronic health points score; ARR, absolute risk reduction; BSI, bloodstream
secondary endpoints

infection; CAP, community-acquired pneumonia; CI, confidence interval; CrCl, creatinine clearance; cUTI, complicated urinary tract infection; CVC, central venous catheter; DFI, diabetic foot infection; EAT, empiric antibiotic therapy (consisted of antipseudomonal β-lactam
monotherapy or in combination with other agents per institutional protocol); EOT, end of therapy; HCAP, health care-associated pneumonia; HIV, human immunodeficiency virus; HSCT, hematopoetic stem cell transplantation; IAI, intraabdominal infection; ICU, intensive
care unit; IQR, interquartile range; ITT, intention to treat; LOS, length of stay; MV, mechanical ventilation; NF, neutropenic fever; NR, not reported; PD, peritoneal dialysis; PNA, pneumonia; PSI, pneumonia severity index; qSOFA, quick sequential organ failure assessment
met for primary and
olution of neutropenia. The intervention group had sig-

Noninferiority was

Noninferiority met
Comments
nificantly greater antibiotic-free days while mean fever
days and all-cause mortality was not different between the
groups [42].
• Bloodstream infection (BSI): In a recent study of hospital-
ized patients with gram-negative bacteremia surviving and
Outcomes

Clinical cure:
response:
Tedizolid IV Linezolid: 10 Early clinical

90.9% vs
clinically stable at day 7, 7 days of antibiotic therapy was
85% vs

90.9%
83%

noninferior to 14 days for the composite endpoint of 90-day

mortality, clinical failure, or hospital length of stay [44].
Short Course Long Course

discretion: discretion:
Duration, d Duration, d
Antibiotic:

Even in such highly morbid infections as S. aureus bacte-

physician

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

abx per
Adequate
remia there has been a sequential reduction in the number

84
of weeks recommended for therapy. S. aureus bacteremia
physician was historically treated for a standard 4–6 weeks of intra-
Antibiotic:

to PO: 6

abx per
Adequate

venous therapy [49], until a subgroup of “uncomplicated”

42
S. aureus bacteremia was identified for whom 2–4 weeks
became accepted [50]. Now, a trial is underway to evaluate
Severity of Illness

just 7 days of therapy in uncomplicated S. aureus bacte-

remia [51]. In another approach, Holland et al recently pub-
NR

NR

score; RCT, randomized controlled trial; SAPS II, simplified acute physiology score; SBP, spontaneous bacterial peritonitis; SOFA, sequential organ failure assessment score.

lished a successful approach to protocolizing the treatment

durations of multiple clinically diverse staphylococci BSIs.
This approach resulted in a noninferior rate of clinical suc-
Location
Patient

cess paired with 29% reduction in median antibiotic dura-

NR

NR

tion without any increase in infection-related adverse events

device or vascular catheter

necrotizing process; septic

associated with prosthetic

within 1 wk of treatment

[52]. Although it should be noted that this trial enrolled

bacterial infection; death
fungal, brucellar, myco-

Number of patients diagnosed with sepsis not reported but number of patients for whom sepsis was the reason for MV was reported.
shock or severe sepsis;
Uncomplicated ABSSTI or

or documented (unless
Exclusion Criteria

both S. aureus and coagulase-negative staphylococci infec-

wound infection); any

immunosuppression
Life expectancy < 1 y;
pathogen suspected
site; gram-negative

tions, which have very different clinical outcomes, and the

study was not powered to adequately study individual sub-
groups. Additionally, the data on which the protocol meth-
odology were based were often low-quality evidence due to
the limited availability of randomized trials testing antibiotic
treatment durations in BSI [53]. Despite these limitations,
Inclusion Criteria

tissue infection

confirmed pyo-
systemic signs

genic vertebral
359 Microbiologically
by regional or

tive organism

osteomyelitis
accompanied

documented

this approach provides another potential tool in the stew-

of infection;

suspected/
gram-posi-
666 Skin or soft

ardship toolkit.
• Intraabdominal infections (IAIs): The 2015 STOP-IT trial
[22] significantly impacted the practice for managing IAIs.
Patients
No. of

Patients with IAI undergoing source control intervention

were randomized to receive antibiotics for either 4 days after
noninferiority RCT posttreatment
Multicenter, open, Clinical cure 1 y

source control (intervention arm) or 2 days after resolution

response at
noninferiority RCT 48–72 h as-
Outcome

sessment
Primary

Early clinical

of systemic inflammatory response symptoms, which ended

up being a median of 8 (IQR, 5–10) days. There was no differ-
ence in the primary composite endpoint of mortality, surgical
site infection, or recurrent IAI, but the intervention arm re-
Study Design

double-blind,

ceived significantly shorter duration of therapy with greater

Multinational,

antibiotic-free days. Of note, the mean age in this population

was 52 years, and the mean APACHE II score (acute physi-
ology + age points + chronic health points) was relatively low
at 10.1 (predicted mortality approximately 10%) compared
Table 1.  Continued

Author

Acute py- Bernard

2014

2015
et al

et al
Moran

[46]

myelitis [47]

to the average hospitalized patients with abdominal sepsis.

• Acute bacterial skin and soft tissue infections (ABSSTIs):
Syndrome
Infectious

ogenic
osteo-

The ESTABLISH studies investigated the use of tedizolid for

ABSSTI

ABSSTIs. In both studies, 6 days of tedizolid (either oral or

S146 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

 intravenous to oral) was noninferior to 10 days of linezolid antibiotic administration,” and ARISE [6] reported a median
for clinical response [45, 46]. time to antibiotic administration of 70 and 67 minutes in their
• Acute bacterial osteomyelitis: Bernard et al [47] demonstrated experimental and control groups, respectively. Furthermore,
noninferiority of 6 vs 12 weeks of antibiotics in the primary critically ill patients are underrepresented in trials evaluating
analysis of 1-year clinical cure. However, the noninferiority the optimal duration of antibiotic treatment in organ-specific
margin was not met for the subgroup analyses of age over 75, infections such as pneumonia and urinary tract infections. In
non-S. aureus infection, immunosuppression, diabetes, and the absence of truly representative data, we must ask ourselves
presence of neurologic signs, abscess, or endocarditis. 2 questions: Do sicker patients in fact warrant longer courses
of antibiotics? And is it reasonable to extend the findings of
Despite these exciting results, it must be noted that many of these studies in patients with infection to those with sepsis? Infection
data are derived from noninferiority studies and with broad is necessary but not sufficient for the definition of sepsis. Due

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

exclusion criteria that tended to restrict the final study popu- to the complexity of organ dysfunction in sepsis, observed mor-
lation to those with milder acute illness and fewer underlying bidity does not bear a linear relationship with microbial burden,
high-risk illnesses (Table 1). While a noninferiority trial design and the risk of mortality is not entirely mitigated by optimal
may be a reasonable means for investigating new antibiotic dur- antimicrobial management. Unfortunately, relatively few trials
ations in select circumstances, it has important limitations [54, have been conducted specifically in the critically ill or in se-
55]. Historical trial data used to establish the magnitude of the rious infections with a high likelihood of systemic manifest-
effect of standard therapy or active control (vs placebo) relies ations. One example is the aforementioned Chastre et al study
on the “constancy” assumption. However, historical data may of antibiotic duration in VAP [19]. Inclusion criteria required
not reflect the current landscape of patient complexity and pa- ICU admission and mechanical ventilation for at least 48 hours,
tient care practices [56]. Furthermore, noninferiority of a new and approximately one-third of the patients received vaso-
therapy to an active control does not necessarily confirm supe- pressor support. The authors found 8 days of antibiotics to be
riority of the new therapy over placebo, and the sample size for noninferior to 15 days with regard to all-cause mortality and
noninferiority trials is unfortunately often influenced by arbi- infection recurrence, which greatly changed treatment guide-
trary thresholds of clinical importance and trial sponsor budget. lines [21, 57, 58]. Although it should be noted that here, too,
Importantly, the majority of the antibiotic treatment dura- exclusion criteria included a simplified acute physiology score
tion studies either specifically excluded patients with sepsis or (SAPS II) greater than 65 (which correlates to approximately
intensive care unit (ICU) admission or did not provide dem- 75% mortality), and immunocompromising conditions such as
ographic information such as the frequency of sepsis diag- neutropenia, AIDS, and immunosuppressant or long-term cor-
nosis, vasopressor or ventilatory support, or severity of illness ticosteroid therapy. Yahav et al [44] conducted an open-label
scores. Therefore, by limiting the inclusion criteria to a pa- noninferiority study of hospitalized patients with uncompli-
tient population with a lower severity of illness, the event rates cated gram-negative bacteremia receiving 7 vs 14 days of an-
for mortality or serious complications are decreased, and the tibiotic therapy. The noninferiority margin was met; however,
trial may be biased toward noninferiority, particularly if the the mean baseline sequential organ failure assessment (SOFA)
prespecified margin is large. Additional common exclusion cri- score was lower than would be expected in gram-negative BSI
teria limiting applicability of these data include renal dysfunc- in both groups. Additionally, in order to be randomized at day
tion, immunocompromising conditions, and recent antibiotic 7, the patients in this study had to be clinically stable, thus there
use, which are all relatively common in real-world critically ill were no patients in shock or mechanical ventilation at that time,
populations. and these frequencies were not reported at presentation. More
specific to the critically ill population, Daneman et al [43] have
CAN ORGAN-SPECIFIC INFECTION TREATMENT published a pilot study of bacteremic ICU patients with high
DURATIONS BE EXTRAPOLATED TO SEPSIS?
median APACHE II scores (22; IQR, 18–26) and vasopressor
There is a notable lack of trials on the duration of antibiotic support (52%), in which they demonstrated feasibility and
therapy in sepsis. As previously mentioned, none of the land- good adherence to the study protocol. We anxiously await the
mark sepsis trials which have shaped current sepsis manage- results of their complete randomized controlled trial appropri-
ment [4–6, 24–26] reported any specific antibiotic regimens, ately powered to examine the 7 vs 14 day treatment duration in
durations, or microbiologic data. The PROWESS [24] and bacteremic shock for noninferiority in the primary outcome of
PROWESS-SHOCK [25] protocol did not call for any stand- 90-day mortality and several relevant secondary outcomes.
ardized approach to critical care management, including anti- Logically, it seems safer to discontinue antibiotics earlier in
biotics, and no data were provided on the frequency, classes, septic patients who demonstrate clinical stability by the time
or duration of prescribed antibiotics. Later, the ACCESS [26] culture results are available compared to those who remain un-
and ProCESS [4] trials only reported high rates of “appropriate stable. However, given the limited data in critically ill patients

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S147

on this topic to date, there are several important factors to these procedures reduce microbial burden and facilitate antibi-
be considered before routinely accepting shorter antibiotic otic penetration into sequestered sites, which could otherwise
courses, even for clinically stable septic patients. Some of these serve as reservoirs of persistent infection and acquisition of drug
will be examined below. resistance. The importance of source control is weighted in the
SSC guidelines as a best practice statement, with emphasis on
Severity of Illness early implementation as soon as medically and logistically fea-
In multiple treatment guidelines, severity of illness is used as a sible [18]. Inability to achieve control of a known source is an
tool to guide the choice and timing of the initial empiric anti- accepted indication for extending duration of therapy, and in-
biotic regimen whereas recommendations on ultimate duration deed nearly all trials of shortened treatment durations have ex-
are based on the organism cultured and the primary organ- plicitly excluded patients with an uncontrolled source or those
system involved [21, 48, 59]. However, in clinical practice, al- requiring active drainage.

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

lowance is often given for the patient to demonstrate signs of Expanding on the traditional concept of source control, there
clinical improvement before an ultimate duration is chosen, is now evidence that the immunologic milieu of sepsis results
which itself predisposes to longer treatment durations associ- in an immune dysregulated state characterized by an inability
ated with greater severity of illness [60]. The 2019 CAP guide- to clear septic foci, as well as widespread lymphocyte apoptosis,
lines’ recommendation on duration states that most patients reduced inflammatory cytokine production, and increased sus-
should be treated for a minimum of 5 days, with discontinuation ceptibility to secondary infections [66, 67]. An autopsy study of
at that point considered only if the patient has been achieved 235 ICU patients with sepsis or septic shock demonstrated an
clinical stability [48]. However, the authors endorse longer unresolved septic focus in nearly 80% of subjects [68]. This sug-
courses for pneumonia complicated by deep-seated infections gests that even in patients lacking an overt uncontrolled source
as well as less common organisms not covered by the guideline of infection, there may yet be occult foci. The precise clinical
(eg, Burkholderia pseudomallei, Mycobacterium tuberculosis, impact of this finding is not known, but may be most significant
and endemic fungi). Interestingly, Aliberti et al [61] evaluated in patients with prolonged critical illness, increased age, and
whether the 2005 recommendations (similar to those above) comorbidities associated with increased infection risk such as
were utilized by treating physicians to tailor duration of therapy diabetes mellitus, and further question our ability to extrapolate
based upon disease severity or clinical response. The mean ± what is optimal antibiotic duration from studies on healthier
standard deviation treatment duration was 11 ± 4.7 days, with patients [69, 70]. Additional evidence of this sepsis-induced im-
42% of patients receiving a course of 10–14 days. Significantly, munosuppression includes the high rate of reactivation of cyto-
time to clinical stability was not associated with total length of megalovirus in otherwise immune competent patients [71] as
therapy, but it was associated with the duration of intravenous well as documented secondary infections with relatively lower-
therapy. This is likely related to sicker patients spending longer virulence organisms [72, 73]. Numerous observational studies
in the hospital, during which time the default route of adminis- of detailed immunophenotyping in septic patients have been
tration is generally intravenous. Interestingly, while severity of published in the last decade [74–77], but these have not yet
illness scores were not associated with length of therapy, sur- been correlated to treatment outcomes. Due to the paucity of
rogate markers including admission to the ICU, hypotension, clinical data in this arena, it is not clear whether the relative im-
and acidemia were associated with significantly longer dur- munosuppression of sepsis could limit the efficacy of shortened
ations. Earlier transition to oral step-down therapy was also antibiotic treatment durations, but it is a host factor worth con-
found to be safe in a recent observational study of patients with sidering while deciding when to discontinue antibiotics until
Enterobacteriaceae bacteremia who attained clinical stability by additional evidence is available.
day 5 [62]. As such, there is not a clear-cut association between
severity of illness and required length of therapy, but providers Microbial Characteristics
appear to have more confidence in transitioning to oral therapy Pseudomonas spp., notably P. aeruginosa, have been long rec-
earlier in less severely ill patients. ognized as a difficult-to-treat pathogen. This is largely due to
many intrinsic and acquired resistance mechanisms as well as
Source Control—Overt and Occult a predilection for high-risk hosts, which can make eradica-
Source control of septic foci has long been recognized as a key tion very difficult. Indeed, P. aeruginosa infections are associ-
intervention in the nonantimicrobial management of sepsis ated with substantial mortality risk [78], and clinical decision
[63–65], and typically refers to procedures such as draining in- making often changes when faced with these infections com-
fected fluid collections, debriding infected tissues, removing in- pared with other organisms. For example, following the land-
fected devices or foreign materials, and correction of anatomic mark trial by Chastre et al, which has been previously discussed,
abnormalities which either predispose to microbial contamina- standard treatment duration for VAP was reduced from 2 weeks
tion or reduce antimicrobial exposure. On a macroscopic level, to just 8 days for most patients [19]. However, due to a high rate

S148 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

of relapse from nonfermenting gram-negative bacilli (predomi- as the ability to provide an appropriate and reliable dose of the
nantly P. aeruginosa), some clinicians did not reduce treatment antibiotic that yields therapeutic drug levels in the blood and
duration for P. aeruginosa infections for many years. It should other affected infection sites. Unfortunately, in the critically ill
be noted, though, that mortality was not different between population, there are numerous competing factors that may im-
the groups and several subsequent studies did not reproduce pact effective dosing, including increased or decreased renal
this finding, leading to the 2016 Infectious Disease Society of blood flow, organ dysfunction (particularly renal and hepatic),
America and American Thoracic Society guidelines on man- changing volume of distribution, and initiation of mechanical
agement of hospital-acquired and VAP to recommend a 7-day support devices such as continuous renal replacement therapy
treatment course for all patients, regardless of organism [21]. or extracorporeal membrane oxygenation. Although the bac-
Indeed the potential recurrence of infection must be weighed tericidal property of the antibiotic has been traditionally con-
against the probable development of resistance with additional sidered an important factor in treatment success against serious

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

antibiotic exposure, leading some providers to adhere to the infections, a recent meta-analysis of 56 trials suggest there may
short course recommendation for sensitive organisms and lean be no intrinsic advantage of bactericidal over bacteriostatic
toward longer courses when multidrug resistance is present agents and that drug dosing and other pharmacokinetic/phar-
[79]. macodynamic properties may be more important drivers of
S. aureus infections are complex owing to both potential drug efficacy [83]. For these reasons and many others, critical care
resistance and the invasive nature of the bacterium, with signifi- pharmacists are a crucial resource when designing an effective
cant rates of endovascular and distant site complications such as antimicrobial regimen for septic patients [84, 85].
endocarditis, abscess, and vertebral osteomyelitis [80]. While a Please refer to the article by Tam et al in this Supplement for
subset of “uncomplicated” S. aureus bacteremia patients (no ev- additional information on the topic [86].
idence of endocarditis or metastases, no prostheses, rapid clear-
ance of cultures, and defervescence) has now been identified NOT ALL SEPSIS IS CREATED EQUAL—SPECIAL
CONSIDERATIONS FOR SPECIFIC POPULATIONS
that can be treated with shorter courses of therapy [50], com-
plication rates remain high for this infection and diligence is Neutropenic Sepsis
needed to prevent undertreatment, relapse, and morbidity [81]. The prevalence and phenotypes of immunocompromising
Like P. aeruginosa and S. aureus, many pathogens such as conditions have increased over the last several decades and
Acinetobacter spp., Stenotrophomonas spp., Enterobacteriaceae, may increase susceptibility to sepsis from a variety of typical
and even the yeast Candida present a clinical challenge due or opportunistic infections, which may warrant specific man-
to their propensity to form biofilm and seeding of secondary agement strategies. However, neutropenia particularly in-
infection sites, which can induce antimicrobial tolerance and creases vulnerability to serious acute infections and sepsis, and
impair eradication. Treatment of sepsis due to gram-negative notably increases morbidity and mortality risk. Roughly half
pathogens harboring difficult-to-treat resistance (ie, resistance of neutropenic fever episodes may be complicated by sepsis or
to all first-line high-efficacy, low-toxicity antibiotics, namely septic shock, with an attendant mortality of 35% to 50% [87].
β-lactams [including carbapenems] and fluoroquinolones) [82], According to the Infectious Diseases Society of America neutro-
necessitates use of second- and third-line agents such as poly- penic fever guidelines, for patients with a clinically or microbi-
myxins, aminoglycosides, and tigecycline or newer agents yet to ologically documented infection, appropriate antibiotic therapy
be studied specifically in sepsis such as ceftazidime/avibactam. should be given at least until resolution of neutropenia (abso-
Guidance is limited for optimal duration of therapy for such lute neutrophil count > 500 cells/mm3) or longer if clinically
infections and difficult-to-treat resistance is a poor prognostic necessary. In some instances, if an appropriate treatment course
marker. Consequently, most providers currently err on the side has been completed prior to resolution of neutropenia, patients
of longer courses for these infections. Furthermore, given the may resume oral prophylaxis until marrow recovery [88]. They
complexity of the patients who contract highly resistant patho- make no specific recommendations for duration of antimicro-
gens, a detailed consideration of all host, pathogen, source, bial regimens based upon disease severity. Interestingly, the
clinical response trajectories, and treatment-related factors 2017 study by Aguilar-Guisado et al [42] challenged the dogma
are needed to define an adequate course for these complicated of continuing antimicrobials in neutropenic fever until neutro-
infections. phil recovery. In patients with neutropenic fever without a mi-
crobiologic diagnosis of infection, discontinuation of empiric
Pharmacokinetic/Pharmacodynamic Issues antibiotics after 72 hours without fever resulted in no difference
The success of a defined antimicrobial course in sepsis is con- in mean fever days or all-cause mortality. However, it should
tingent not only on the in vitro activity of the designated agent be noted that the included population was hospital ward pa-
against the pathogen and the adequacy of source control, but tients; they did not include patients with septic shock and did
also on pharmacokinetic/pharmacodynamic properties such not report how many patients met criteria for sepsis. Extended

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S149

duration of therapy is most likely to be recommended in the set- NARROWER VERSUS SHORTER—LESSONS FROM
ting of neutropenic sepsis due to highly resistant gram-negative DE-ESCALATION TRIALS
organisms, mold infections, or endovascular seeding. If the data regarding shortened duration of therapy inadequately
address critically ill populations, antimicrobial de-escalation may
Culture-Negative Sepsis be another tactic to reduce the adverse effects associated with
Culture-negative sepsis poses a number of its own unique is- prolonged broad-spectrum antibiotic use. The SSC guidelines
sues with regard to antimicrobial management. First, we em- endorse de-escalation for patients initially prescribed multiple
phasize that a large proportion of patients (17% in 1 study) agents (ie, combination therapy) once the patient’s condition has
admitted with an initial clinical diagnosis of sepsis in whom a improved and/or cultures become available, and recommend for
pathogen is not ultimately identified are subsequently found all septic patients that potential for de-escalation be assessed daily
to have a sepsis “mimic” (another noninfectious etiology

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

[18]. In order to evaluate the evidence behind this practice, Tabah
for their illness) and do not require antibiotic therapy [89]. et al published a systematic review and meta-analysis of anti-
Restricting our discussion to those patients with true culture- microbial de-escalation specifically in septic ICU patients [92].
negative sepsis (due to antecedent antibiotics, low culture While the definition of de-escalation varied over the 14 included
sensitivity, fastidious organisms, lack of molecular diagnostic studies, all studies described a narrowing of the spectrum of cov-
testing available, etc.), multiple studies have documented ap- erage. Thirteen studies de-escalated by decreasing the number of
proximately a third of sepsis cases as culture negative [24, prescribed antimicrobials and 4 included a shortening of the du-
90, 91]. Based on recent estimates of national sepsis inci- ration of therapy. Documentation of culture data, a lower baseline
dence [2], this could account for over 500 000 cases annually, severity of illness, and clinical improvement increased the rate of
meaning that earlier discontinuation of antibiotics in culture- de-escalation. Pertinently, infection with a multidrug-resistant
negative sepsis is likely to have a tremendous reduction in organism significantly reduced the likelihood of de-escalation in
patient and population-level antibiotic pressure. However, several studies, as did polymicrobial infection and infections with
determining appropriate antimicrobial management in these a risk of undiagnosed pathogens (eg, IAI). Similarly, a prospective
patients is a challenge. Without an organism against which cohort study by Salahuddin and colleagues [93] found that failure
to direct therapy, treatment courses tend to remain broad to de-escalate was predicted by SAPS II score, hematologic ma-
and there is no clear guidance for discontinuation. A large lignancy, and isolation of multidrug-resistant organisms. None
multicenter retrospective cohort study by Kethireddy et al of the 14 studies reported worsened survival with de-escalation,
[91] recently reported that culture-positive vs negative sepsis and in the pooled mortality analysis provided there was a pro-
have similar survival, which is contingent on timely admin- tective effect of de-escalation (relative risk, 0.68; 95% confidence
istration of appropriate antibiotics. However, the authors interval, .52–.88), with moderate heterogeneity (I2 = 44%) [92].
did not report on mean duration of antibiotic therapy pre- Interestingly, de-escalation was not associated with decreased
scribed or frequency of de-escalation from empiric regimen. duration of therapy, although 1 study did report fewer days of
In a separate single-center retrospective study, Lockhart et al antipseudomonal β-lactam and broad-spectrum gram-positive
investigated the duration of antibiotic treatment received by antibiotics associated with de-escalation [94]. Indeed, a recently
culture-negative sepsis survivors [60]. Groups were stratified published European position statement discussed the conflicting
into less than or equal to 3 days, 4 to 7 days, and greater than data regarding de-escalation and duration of therapy [95]. Much
7 days. Greater severity of illness (as measured by APACHE of the difficulty in interpretation comes from the preponderance
II scores, Charlson comorbidity index, and mechanical ven- of observational study designs (that bias de-escalation towards
tilatory support) was associated with increasing duration of better outcome as it tends to occur in patients who are already
treatment. Specific sites of infection (pneumonia, urinary doing better) and the variable definitions of de-escalation used
tract, joint space, and central nervous system) were associ- by investigators. Based upon the current data it seems that de-es-
ated with longer duration, while unknown or undocumented calation and duration should be assessed separately, as they may
sites of infection correlated with shorter duration. These data have overlapping but unique roles in antimicrobial stewardship
support an organ systems-based approach to antibiotic du- efforts.
ration in culture-negative sepsis. The SSC guidelines do not
provide a specific recommendation of a defined duration for BIOMARKER-BASED GUIDANCE FOR
ANTIMICROBIAL TREATMENT
treatment of culture-negative sepsis. However, close scrutiny
of host and disease characteristics and trajectories of fever, Procalcitonin has been the most extensively studied biomarker
vasopressor dependence, biomarkers, etc. may guide whether for use in the diagnosis of bacterial infections and guidance of
a patient may be a reasonable candidate for a duration of antibiotic therapy. Procalcitonin is a short-lived hormone (pre-
therapy shorter than the standard, albeit not evidence-based, cursor to calcitonin) that is rapidly induced by the inflamma-
7- to 10-day recommendation for all patients with sepsis [18]. tory cytokines associated with bacterial infection. The short

S150 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

half-life of procalcitonin and the correlation of its kinetics with duration of antibiotic therapy in sepsis, we hope that this narra-
the intensity of stimulus are desirable properties for any can- tive review will provide a call to action for conducting random-
didate biomarker to guide both the initiation and duration of ized control trials to specifically address the question of how
antibiotic therapy in septic patients [96]. A comprehensive dis- long to treat in culture-positive and culture-negative sepsis.
cussion on this topic can be found in the article by Gilbert et
al in this Supplement [97], but we will highlight a few salient Notes
points with regard to antimicrobial duration. Financial support. This work was supported by the National
Several large, multicenter trials have now been conducted Institutes of Health Clinical Center Intramural Research
in ICU populations: Procalcitonin to Reduce Antibiotic Program.
Treatments in Acutely Ill Patients (PRORATA) [98], Supplement sponsorship.  This supplement is sponsored
Procalcitonin Guided Antibiotic Rational Decision Making by bioMérieux, the Gordon and Betty Moore Foundation and

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

in ICU Patients (ProGUARD) [99], and Stop Antibiotics on Beckman Coulter.
Procalcitonin Guidance Study (SAPS) [100]. In a meta-analysis Potential conflicts of interest. Both authors: No reported
of these 3 trials and 7 others, procalcitonin-guided patients conflicts of interest. Both authors have submitted the ICMJE
had shorter antibiotic courses compared to controls, with Form for Disclosure of Potential Conflicts of Interest. Conflicts
no adverse impact on mortality or ICU length of stay [101]. that the editors consider relevant to the content of the manu-
However, a recent meta-analysis of 16 randomized controlled script have been disclosed.
trials reported that decreased antibiotic utilization associated
with procalcitonin-guided antibiotic discontinuation in criti- References
cally ill patients represented low-certainty evidence with a high 1. Gaieski DF, Edwards JM, Kallan MJ, Carr BG. Benchmarking
risk of bias and was primarily observed in studies without high the incidence and mortality of severe sepsis in the United
protocol adherence and in those with algorithms combining States. Crit Care Med 2013; 41:1167–74.
procalcitonin and C-reactive protein [102]. The SSC guidelines 2. Rhee C, Dantes R, Epstein L, et al. Incidence and trends of
[18] rank procalcitonin use to decrease antibiotic duration in sepsis in US hospitals using clinical vs claims data, 2009–
sepsis as a weak recommendation with uncertain risk-benefit 2014. JAMA 2017; 318:1241–9.
profile limiting its universal adoption for antibiotic discontinu- 3. Self WH, Semler MW, Bellomo R, et al; CLOVERS Protocol
ation in sepsis across US hospitals and providers today. Committee and NHLBI Prevention and Early Treatment of
Acute Lung Injury (PETAL) Network Investigators. Liberal
CONCLUSIONS
versus restrictive intravenous fluid therapy for early septic
The evidence regarding optimal antimicrobial treatment strat- shock: rationale for a randomized trial. Ann Emerg Med
egies for sepsis is conspicuously lacking and thus guideline 2018; 72:457–66.
recommendations remain necessarily vague despite universal 4. ProCESS Investigators, Yealy DM, Kellum JA, et al. A ran-
acceptance of the importance of antibiotics. Unfortunately, ret- domized trial of protocol-based care for early septic shock.
rospective and observational studies are significantly limited in N Engl J Med 2014; 370:1683–93.
their capacity to accurately capture the full treatment course of 5. Mouncey PR, Osborn TM, Power GS, et al; ProMISe Trial
antimicrobials prescribed for sepsis (eg, outpatient administra- Investigators. Trial of early, goal-directed resuscitation for
tion of intravenous antibiotics, transfer to subacute rehabilita- septic shock. N Engl J Med 2015; 372:1301–11.
tion centers to complete courses of therapy, and de-escalation 6. ARISE Investigators; ANZICS Clinical Trials Group; Peake
to oral regimens), contributing to the incomplete data on this SL, et al. Goal-directed resuscitation for patients with early
topic. There are a handful of on-going trials investigating treat- septic shock. N Engl J Med 2014; 371:1496–506.
ment duration for some of the clinical syndromes associated 7. Venkatesh B, Finfer S, Cohen J, et al; ADRENAL Trial
with sepsis that are addressed here, but high-quality studies in Investigators and the Australian-New Zealand Intensive
this clinically heterogenous syndrome will remain difficult to Care Society Clinical Trials Group. Adjunctive glucocorti-
conduct. Optimal antimicrobial duration in sepsis is likely to coid therapy in patients with septic shock. N Engl J Med
remain best determined through close collaboration between 2018; 378:797–808.
intensivists, infectious disease specialists, and other multidis- 8. Annane D, Renault A, Brun-Buisson C, et al; CRICS-
ciplinary providers in order to weigh the relative contributions TRIGGERSEP Network. Hydrocortisone plus
of the many factors addressed in this review. Infectious disease fludrocortisone for adults with septic shock. N Engl J Med
consultation has been shown to improve patient outcomes in 2018; 378:809–18.
many studies of serious infections [103–106] and sepsis [107, 9. Kumar A, Ellis P, Arabi Y, et al; Cooperative Antimicrobial
108], and this practice should be encouraged. Although we are Therapy of Septic Shock Database Research Group.
unable to provide specific data-driven recommendations for Initiation of inappropriate antimicrobial therapy results in a

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S151

 fivefold reduction of survival in human septic shock. Chest Infectious Diseases Society of America and the American
2009; 136:1237–48. Thoracic Society. Clin Infect Dis 2016; 63:e61–e111.
10. Ibrahim EH, Sherman G, Ward S, Fraser VJ, Kollef MH. The 22. Sawyer RG, Claridge JA, Nathens AB, et al; STOP-IT
influence of inadequate antimicrobial treatment of blood- Trial Investigators. Trial of short-course antimicrobial
stream infections on patient outcomes in the ICU setting. therapy for intraabdominal infection. N Engl J Med 2015;
Chest 2000; 118:146–55. 372:1996–2005.
11. Paul M, Shani V, Muchtar E, Kariv G, Robenshtok E, 23. Eliakim-Raz N, Yahav D, Paul M, Leibovici L. Duration of
Leibovici L. Systematic review and meta-analysis of the ef- antibiotic treatment for acute pyelonephritis and septic uri-
ficacy of appropriate empiric antibiotic therapy for sepsis. nary tract infection—7 days or less versus longer treatment:
Antimicrob Agents Chemother 2010; 54:4851–63. systematic review and meta-analysis of randomized con-
12. Kumar A, Roberts D, Wood KE, et al. Duration of hypoten- trolled trials. J Antimicrob Chemother 2013; 68:2183–91.

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

sion before initiation of effective antimicrobial therapy is 24. Bernard GR, Vincent JL, Laterre PF, et al; Recombinant
the critical determinant of survival in human septic shock. Human Protein C Worldwide Evaluation in Severe Sepsis
Crit Care Med 2006; 34:1589–96. (PROWESS) Study Group. Efficacy and safety of recombi-
13. Kumar A, Haery C, Paladugu B, et al. The duration of hypo- nant human activated protein C for severe sepsis. N Engl J
tension before the initiation of antibiotic treatment is a crit- Med 2001; 344:699–709.
ical determinant of survival in a murine model of Escherichia 25. Ranieri VM, Thompson BT, Barie PS, et al; PROWESS-
coli septic shock: association with serum lactate and inflam- SHOCK Study Group. Drotrecogin alfa (activated) in adults
matory cytokine levels. J Infect Dis 2006; 193:251–8. with septic shock. N Engl J Med 2012; 366:2055–64.
14. Liu VX, Fielding-Singh V, Greene JD, et al. The timing 26. Opal SM, Laterre PF, Francois B, et al; ACCESS Study
of early antibiotics and hospital mortality in sepsis. Am J Group. Effect of eritoran, an antagonist of MD2-TLR4, on
Respir Crit Care Med 2017; 196:856–63. mortality in patients with severe sepsis: the ACCESS ran-
15. Seymour CW, Gesten F, Prescott HC, et al. Time to treat- domized trial. JAMA 2013; 309:1154–62.
ment and mortality during mandated emergency care for 27. Corona A, Bertolini G, Ricotta AM, Wilson A, Singer M,
sepsis. N Engl J Med 2017; 376:2235–44. Wilson A. Variability of treatment duration for bacteraemia
16. Septimus EJ, Coopersmith CM, Whittle J, Hale CP, Fishman in the critically ill: a multinational survey. J Antimicrob
NO, Kim TJ. Sepsis national hospital inpatient quality Chemother 2003; 52:849–52.
measure (SEP-1): Multistakeholder Work Group recom- 28. ESCAVO. Results of ESCAVO’s survey on antibiotic use in
mendations for appropriate antibiotics for the treatment of sepsis. https://www.escavo.com/2019/07/18/antibiotic-use-
sepsis. Clin Infect Dis 2017; 65:1565–9. in-sepsis-survey/. Accessed 16 January 2020.
17. The Joint Commission. National Hospital Inpatient 29. Spellberg B, Rice LB. Duration of antibiotic therapy: shorter
Quality Reporting Measures Specificaiton Manual. Chart is better. Ann Intern Med 2019; 171:210–1.
Abstracted Measures. https://www.jointcommission.org/ 30. Teshome BF, Vouri SM, Hampton N, Kollef MH, Micek ST.
en/measurement/specificationmanuals/chart-abstracted- Duration of exposure to antipseudomonal β-lactam anti-
measures/. Accessed 12 January 2020. biotics in the critically ill and development of new resist-
18. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis ance. Pharmacotherapy 2019; 39:261–70.
campaign: international guidelines for management of 31. Rhee C, Kadri SS, Dekker JP, et al; CDC Prevention
sepsis and septic shock: 2016. Intensive Care Med 2017; Epicenters Program. Prevalence of antibiotic-resistant
43:304–77. pathogens in culture-proven sepsis and outcomes associ-
19. Chastre J, Wolff M, Fagon JY, et al; PneumA Trial Group. ated with inadequate and broad-spectrum empiric antibi-
Comparison of 8 vs 15 days of antibiotic therapy for otic use. JAMA Netw Open 2020; 3:e202899.
ventilator-associated pneumonia in adults: a randomized 32. Various combinations of isoniazid with streptomycin or
trial. JAMA 2003; 290:2588–98. with P.A.S. in the treatment of pulmonary tuberculosis;
20. Choudhury G, Mandal P, Singanayagam A, Akram AR, seventh report to the Medical Research Council by their
Chalmers JD, Hill AT. Seven-day antibiotic courses have Tuberculosis Chemotherapy Trials Committee. Br Med J
similar efficacy to prolonged courses in severe community- 1955; 1:435–45.
acquired pneumonia—a propensity-adjusted analysis. Clin 33. Warren JW, Abrutyn E, Hebel JR, Johnson JR, Schaeffer
Microbiol Infect 2011; 17:1852–8. AJ, Stamm WE. Guidelines for antimicrobial treatment of
21. Kalil AC, Metersky ML, Klompas M, et al. Management uncomplicated acute bacterial cystitis and acute pyelone-
of adults with hospital-acquired and ventilator-associated phritis in women. Infectious Diseases Society of America
pneumonia: 2016 clinical practice guidelines by the (IDSA). Clin Infect Dis 1999; 29:745–58.

S152 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

34. Pezzani MD, Be G, Cattaneo P, et al. Evidence based re- 46. Moran GJ, Fang E, Corey GR, Das AF, De Anda C,
view on optimal duration of antibiotic therapy for bacterial Prokocimer P. Tedizolid for 6 days versus linezolid for
infections to support antimicrobial stewardship recom- 10 days for acute bacterial skin and skin-structure in-
mendations. https://www.who.int/selection_medicines/ fections (ESTABLISH-2): a randomised, double-blind,
committees/expert/22/applications/ABWG_optimal_dura- phase 3, non-inferiority trial. Lancet Infect Dis 2014;
tion_AB.pdf. Accessed 16 January 2020. 14:696–705.
35. Spellberg B. The new antibiotic mantra—“shorter is better”. 47. Bernard L, Dinh A, Ghout I, et al; Duration of Treatment for
JAMA Intern Med 2016; 176:1254–5. Spondylodiscitis (DTS) Study Group. Antibiotic treatment
36. Bartlett JG, Breiman RF, Mandell LA, File TM Jr. for 6 weeks versus 12 weeks in patients with pyogenic verte-
Community-acquired pneumonia in adults: guidelines for bral osteomyelitis: an open-label, non-inferiority, random-
management. The Infectious Diseases Society of America. ised, controlled trial. Lancet 2015; 385:875–82.

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

Clin Infect Dis 1998; 26:811–38. 48. Metlay JP, Waterer GW, Long AC, et al. Diagnosis and treat-
37. Dunbar LM, Wunderink RG, Habib MP, et al. High-dose, ment of adults with community-acquired pneumonia. An
short-course levofloxacin for community-acquired pneu- official clinical practice guideline of the American Thoracic
monia: a new treatment paradigm. Clin Infect Dis 2003; Society and Infectious Diseases Society of America. Am J
37:752–60. Respir Crit Care Med 2019; 200:e45–67.
38. Uranga A, España PP, Bilbao A, et al. Duration of antibi- 49. Nolan CM, Beaty HN. Staphylococcus aureus bacteremia.
otic treatment in community-acquired pneumonia: a Current clinical patterns. Am J Med 1976; 60:495–500.
multicenter randomized clinical trial. JAMA Intern Med 50. Liu C, Bayer A, Cosgrove SE, et al; Infectious Diseases
2016; 176:1257–65. Society of America. Clinical practice guidelines by the
39. Vaughn VM, Flanders SA, Snyder A, et al. Excess antibiotic Infectious Diseases Society of America for the treatment
treatment duration and adverse events in patients hospital- of methicillin-resistant Staphylococcus aureus infections in
ized with pneumonia: a multihospital cohort study. Ann adults and children. Clin Infect Dis 2011; 52:e18–55.
Intern Med 2019; 171:153–63. 51. Thorlacius-Ussing L, Andersen CØ, Frimodt-Møller N,
40. Peterson J, Kaul S, Khashab M, Fisher AC, Kahn JB. A dou- Knudsen IJD, Lundgren J, Benfield TL. Efficacy of seven
ble-blind, randomized comparison of levofloxacin 750 mg and fourteen days of antibiotic treatment in uncomplicated
once-daily for five days with ciprofloxacin 400/500 mg twice- Staphylococcus aureus bacteremia (SAB7): study protocol
daily for 10 days for the treatment of complicated urinary tract for a randomized controlled trial. Trials 2019; 20:250.
infections and acute pyelonephritis. Urology 2008; 71:17–22. 52. Holland TL, Raad I, Boucher HW, et al; Staphylococcal
41. Sandberg T, Skoog G, Hermansson AB, et al. Ciprofloxacin Bacteremia Investigators. Effect of algorithm-based therapy
for 7 days versus 14 days in women with acute pyelo- vs usual care on clinical success and serious adverse events
nephritis: a randomised, open-label and double-blind, in patients with staphylococcal bacteremia: a randomized
placebo-controlled, non-inferiority trial. Lancet 2012; clinical trial. JAMA 2018; 320:1249–58.
380:484–90. 53. Perencevich EN, Malani PN. Treatment algorithms for
42. Aguilar-Guisado M, Espigado I, Martín-Peña A, et al. staphylococcal bacteremia: improving clinical care and
Optimisation of empirical antimicrobial therapy in patients enhancing antimicrobial stewardship. JAMA 2018;
with haematological malignancies and febrile neutropenia 320:1243–4.
(How Long study): an open-label, randomised, controlled 54. Powers JH, Fleming TR. Noninferiority trials: clinical
phase 4 trial. Lancet Haematol 2017; 4:e573–83. understandings and misunderstandings. Clin Investig
43. Daneman N, Rishu AH, Pinto R, et al; Canadian Critical (Lond) 2013; 3:215–8.
Care Trials Group. 7 versus 14 days of antibiotic treatment 55. Kaul S, Diamond GA. Good enough: a primer on the anal-
for critically ill patients with bloodstream infection: a pilot ysis and interpretation of noninferiority trials. Ann Intern
randomized clinical trial. Trials 2018; 19:111. Med 2006; 145:62–9.
44. Yahav D, Franceschini E, Koppel F, et al; Bacteremia 56. Head SJ, Kaul S, Bogers AJ, Kappetein AP. Non-inferiority
Duration Study Group. Seven versus 14 days of antibiotic study design: lessons to be learned from cardiovascular
therapy for uncomplicated gram-negative bacteremia: a trials. Eur Heart J 2012; 33:1318–24.
noninferiority randomized controlled trial. Clin Infect Dis 57. Torres A, Niederman MS, Chastre J, et al. International ERS/
2019; 69:1091–8. ESICM/ESCMID/ALAT guidelines for the management of
45. Prokocimer P, De Anda C, Fang E, Mehra P, Das A. Tedizolid hospital-acquired pneumonia and ventilator-associated
phosphate vs linezolid for treatment of acute bacterial skin pneumonia: guidelines for the management of hospital-
and skin structure infections: the ESTABLISH-1 random- acquired pneumonia (HAP)/ventilator-associated pneu-
ized trial. JAMA 2013; 309:559–69. monia (VAP) of the European Respiratory Society (ERS),

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S153

 European Society of Intensive Care Medicine (ESICM), 72. van Vught LA, Klein Klouwenberg PM, Spitoni C, et al;
European Society of Clinical Microbiology and Infectious MARS Consortium. Incidence, risk factors, and attributable
Diseases (ESCMID) and Asociacion Latinoamericana del mortality of secondary infections in the intensive care unit
Torax (ALAT). Eur Respir J 2017; 50:1700582. after admission for sepsis. JAMA 2016; 315:1469–79.
58. American Thoracic Society; Infectious Diseases Society of 73. Otto GP, Sossdorf M, Claus RA, et al. The late phase of sepsis
America. Guidelines for the management of adults with is characterized by an increased microbiological burden
hospital-acquired, ventilator-associated, and healthcare- and death rate. Crit Care 2011; 15:R183.
associated pneumonia. Am J Respir Crit Care Med 2005; 74. Xia Q, Wei L, Zhang Y, Sheng J, Wu W, Zhang Y. Immune
171:388–416. checkpoint receptors Tim-3 and PD-1 regulate monocyte
59. Mazuski JE, Tessier JM, May AK, et al. The Surgical Infection and T lymphocyte function in septic patients. Mediators
Society revised guidelines on the management of intra- Inflamm 2018; 2018:1632902.

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

abdominal infection. Surg Infect (Larchmt) 2017; 18:1–76. 75. Liu Q, An L, Qi Z, Zhao Y, Li C. Increased expression of
60. Lockhart GC, Hanin J, Micek ST, Kollef MH. Pathogen- programmed cell death-1 in regulatory T cells of patients
negative sepsis-an opportunity for antimicrobial steward- with severe sepsis and septic shock: an observational clin-
ship. Open Forum Infect Dis 2019; 6:ofz397. ical study. Scand J Immunol 2017; 86:408–17.
61. Aliberti S, Blasi F, Zanaboni AM, et al. Duration of antibiotic 76. Boomer JS, To K, Chang KC, et al. Immunosuppression in
therapy in hospitalised patients with community-acquired patients who die of sepsis and multiple organ failure. JAMA
pneumonia. Eur Respir J 2010; 36:128–34. 2011; 306:2594–605.
62. Tamma PD, Conley AT, Cosgrove SE, et al; Antibacterial 77. Huang X, Venet F, Wang YL, et al. PD-1 expression by
Resistance Leadership Group. Association of 30-day mor- macrophages plays a pathologic role in altering microbial
tality with oral step-down vs continued intravenous therapy clearance and the innate inflammatory response to sepsis.
in patients hospitalized with Enterobacteriaceae bacte- Proc Natl Acad Sci U S A 2009; 106:6303–8.
remia. JAMA Intern Med 2019; 179:316–23. 78. Curran CS, Bolig T, Torabi-Parizi P. Mechanisms and tar-
63. Marshall JC, al Naqbi A. Principles of source control in the geted therapies for Pseudomonas aeruginosa lung infection.
management of sepsis. Crit Care Clin 2009; 25:753–68, Am J Respir Crit Care Med 2018; 197:708–27.
viii–ix. 79. Sun HY, Fujitani S, Quintiliani R, Yu VL. Pneumonia due to
64. Lagunes L, Encina B, Ramirez-Estrada S. Current under- Pseudomonas aeruginosa: part II: antimicrobial resistance,
standing in source control management in septic shock pa- pharmacodynamic concepts, and antibiotic therapy. Chest
tients: a review. Ann Transl Med 2016; 4:330. 2011; 139:1172–85.
65. Martínez ML, Ferrer R, Torrents E, et al; Edusepsis Study 80. Tong SY, Davis JS, Eichenberger E, Holland TL, Fowler VG
Group. Impact of source control in patients with severe Jr. Staphylococcus aureus infections: epidemiology, patho-
sepsis and septic shock. Crit Care Med 2017; 45:11–9. physiology, clinical manifestations, and management. Clin
66. Hotchkiss RS, Monneret G, Payen D. Sepsis-induced im- Microbiol Rev 2015; 28:603–61.
munosuppression: from cellular dysfunctions to immuno- 81. Fowler VG Jr, Olsen MK, Corey GR, et al. Clinical identi-
therapy. Nat Rev Immunol 2013; 13:862–74. fiers of complicated Staphylococcus aureus bacteremia. Arch
67. Hotchkiss RS, Moldawer LL, Opal SM, Reinhart K, Turnbull Intern Med 2003; 163:2066–72.
IR, Vincent JL. Sepsis and septic shock. Nat Rev Dis Primers 82. Strich JR, Kadri SS. Difficult-to-treat antibiotic-resistant
2016; 2:16045. gram-negative pathogens in the intensive care unit: epide-
68. Torgersen C, Moser P, Luckner G, et al. Macroscopic post- miology, outcomes, and treatment. Semin Respir Crit Care
mortem findings in 235 surgical intensive care patients with Med 2019; 40:419–34.
sepsis. Anesth Analg 2009; 108:1841–7. 83. Wald-Dickler N, Holtom P, Spellberg B. Busting the myth of
69. Hotchkiss RS, Monneret G, Payen D. Immunosuppression “static vs cidal”: a systemic literature review. Clin Infect Dis
in sepsis: a novel understanding of the disorder and 2018; 66:1470–4.
a new therapeutic approach. Lancet Infect Dis 2013; 84. Leape LL, Cullen DJ, Clapp MD, et al. Pharmacist partici-
13:260–8. pation on physician rounds and adverse drug events in the
70. Hawkins RB, Raymond SL, Stortz JA, et al. Chronic crit- intensive care unit. JAMA 1999; 282:267–70.
ical illness and the persistent inflammation, immunosup- 85. Wang T, Benedict N, Olsen KM, et al. Effect of critical care
pression, and catabolism syndrome. Front Immunol 2018; pharmacist’s intervention on medication errors: a system-
9:1511. atic review and meta-analysis of observational studies. J
71. Limaye AP, Kirby KA, Rubenfeld GD, et al. Cytomegalovirus Crit Care 2015; 30:1101–6.
reactivation in critically ill immunocompetent patients. 86. Tam et al. Optimizing PK/PD management in patients with
JAMA 2008; 300:413–22. sepsis. J Infect Dis.

S154 • jid 2020:222 (Suppl 2)  •  Busch and Kadri

87. Patel A, Gruber P. Severe infections in neutropenic patients. 99. Shehabi Y, Sterba M, Garrett PM, et al; ProGUARD Study
Curr Opin Crit Care 2015; 21:586–92. Investigators; ANZICS Clinical Trials Group. Procalcitonin
88. Freifeld AG, Bow EJ, Sepkowitz KA, et al. Clinical practice algorithm in critically ill adults with undifferentiated in-
guideline for the use of antimicrobial agents in neutropenic fection or suspected sepsis. A randomized controlled trial.
patients with cancer: 2010 update by the Infectious Diseases Am J Respir Crit Care Med 2014; 190:1102–10.
Society of America. Clin Infect Dis 2011; 52:e56–93. 100. de Jong E, van Oers JA, Beishuizen A, et al. Efficacy and
89. Heffner AC, Horton JM, Marchick MR, Jones AE. Etiology safety of procalcitonin guidance in reducing the duration
of illness in patients with severe sepsis admitted to the of antibiotic treatment in critically ill patients: a random-
hospital from the emergency department. Clin Infect Dis ised, controlled, open-label trial. Lancet Infect Dis 2016;
2010; 50:814–20. 16:819–27.
90. Rangel-Frausto MS, Pittet D, Costigan M, Hwang T, Davis 101. Iankova I, Thompson-Leduc P, Kirson NY, et al. Efficacy

Downloaded from https://academic.oup.com/jid/article-abstract/222/Supplement_2/S142/5874160 by guest on 05 August 2020

CS, Wenzel RP. The natural history of the systemic inflam- and safety of procalcitonin guidance in patients with sus-
matory response syndrome (SIRS). A prospective study. pected or confirmed sepsis: a systematic review and meta-
JAMA 1995; 273:117–23. analysis. Crit Care Med 2018; 46:691–8.
91. Kethireddy S, Bilgili B, Sees A, et al; Cooperative 102. Pepper DJ, Sun J, Rhee C, et al. Procalcitonin-guided an-
Antimicrobial Therapy of Septic Shock (CATSS) Database tibiotic discontinuation and mortality in critically ill
Research Group. Culture-negative septic shock compared adults: a systematic review and meta-analysis. Chest 2019;
with culture-positive septic shock: a retrospective cohort 155:1109–18.
study. Crit Care Med 2018; 46:506–12. 103. Burnham JP, Olsen MA, Stwalley D, Kwon JH, Babcock
92. Tabah A, Cotta MO, Garnacho-Montero J, et al. A system- HM, Kollef MH. Infectious diseases consultation reduces
atic review of the definitions, determinants, and clinical 30-day and 1-year all-cause mortality for multidrug-
outcomes of antimicrobial de-escalation in the intensive resistant organism infections. Open Forum Infect Dis
care unit. Clin Infect Dis 2016; 62:1009–17. 2018; 5:ofy026.
93. Salahuddin N, Amer L, Joseph M, El Hazmi A, Hawa H, 104. Goto M, Schweizer ML, Vaughan-Sarrazin MS, et al.
Maghrabi K. Determinants of deescalation failure in crit- Association of evidence-based care processes with mor-
ically ill patients with sepsis: a prospective cohort study. tality in Staphylococcus aureus bacteremia at veterans
Crit Care Res Pract 2016; 2016:6794861. health administration hospitals, 2003–2014. JAMA Intern
94. Paskovaty A, Pastores SM, Gedrimaite Z, Kostelecky N, Med 2017; 177:1489–97.
Riedel ER, Seo SK. Antimicrobial de-escalation in septic 105. Turner RB, Valcarlos E, Won R, Chang E, Schwartz J.
cancer patients: is it safe to back down? Intensive Care Impact of infectious diseases consultation on clinical
Med 2015; 41:2022–3. outcomes of patients with Staphylococcus aureus bacte-
95. Tabah A, Bassetti M, Kollef MH, et al. Antimicrobial de-escala- remia in a community health system. Antimicrob Agents
tion in critically ill patients: a position statement from a task Chemother 2016; 60:5682–7.
force of the European Society of Intensive Care Medicine 106. Schmitt S, McQuillen DP, Nahass R, et al. Infectious dis-
(ESICM) and European Society of Clinical Microbiology and eases specialty intervention is associated with decreased
Infectious Diseases (ESCMID) Critically Ill Patients Study mortality and lower healthcare costs. Clin Infect Dis 2014;
Group (ESGCIP). Intensive Care Med 2020; 46:245–65. 58:22–8.
96. Rhee C. Using procalcitonin to guide antibiotic therapy. 107. Madaline T, Wadskier Montagne F, Eisenberg R, et al. Early
Open Forum Infect Dis 2017; 4:ofw249. infectious disease consultation is associated with lower
97. Gilbert D. The role of procalcitonin in sepsis management mortality in patients with severe sepsis or septic shock
and stewardship. J Infect Dis. who complete the 3-hour sepsis treatment bundle. Open
98. Bouadma L, Luyt CE, Tubach F, et al; PRORATA trial Forum Infect Dis 2019; 6:ofz408.
group. Use of procalcitonin to reduce patients’ exposure 108. Viale P, Tedeschi S, Scudeller L, et al. Infectious diseases
to antibiotics in intensive care units (PRORATA trial): a team for the early management of severe sepsis and septic
multicentre randomised controlled trial. Lancet 2010; shock in the emergency department. Clin Infect Dis 2017;
375:463–74. 65:1253–9.

Sepsis Antimicrobial Treatment Duration   •  jid 2020:222 (Suppl 2) • S155