International Journal of Antimicrobial Agents 64 (2024) 107263

Contents lists available at ScienceDirect

International Journal of Antimicrobial Agents

journal homepage: www.elsevier.com/locate/ijantimicag

Review

Safety and efficacy of outpatient parenteral antimicrobial therapy: A

systematic review and meta-analysis of randomized clinical trials
Solomon Ahmed Mohammed a,b, Jason A. Roberts a,c,d,e, Menino Osbert Cotta a,d,
Benjamin Rogers f, James Pollard g, Getnet Mengistu Assefa a,b, Daniel Erku h,
Fekade B. Sime a,∗
a
UQ Centre for Clinical Research, The University of Queensland, Queensland, Australia
b
Department of Pharmacy, Wollo University, Dessie, Ethiopia
c
Department of Pharmacy and Intensive Care Medicine, Royal Brisbane and Women’s Hospital, Queensland, Australia
d
Herston Infectious Disease Institute (HeIDI), Metro North Health, Queensland, Australia
e
Division of Anaesthesiology Critical Care Emerging and Pain Medicine, Nimes University Hospital, University of Montpellier, Nimes, France
f
Centre for Inflammatory Disease, Monash University, Melbourne, Australia
g
Cabrini @ Home, Cabrini Health, Melbourne, Australia
h
Centre for Applied Health Economics, Griffith University, Queensland, Australia

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

Article history: Background: Outpatient parenteral antimicrobial therapy (OPAT) offers an alternative to inpatient (hos-
Received 16 April 2024 pital bed-based) treatment of infections that require intravenous administration of antimicrobials. This
Accepted 25 June 2024
meta-analysis aimed to summarise the evidence available from randomised controlled trials (RCTs) re-
garding the efficacy and safety of OPAT compared to inpatient parenteral antimicrobial therapy.
Editor: Professor Evangelos Methods: We searched the Cochrane Library, MEDLINE, Embase, PubMed, and Web of Sciences databases
Giamarellos-Bourboulis for RCTs comparing outpatient versus inpatient parenteral antimicrobial therapy. We included studies
Keywords: without restrictions on language or publication year. Eligibility was reviewed independently by two as-
Outpatient sessors, and data extraction was cross validated. We evaluated bias risk via the Cochrane tool and deter-
Inpatient mined the evidence certainty using GRADE. Meta-analysis was conducted using a random effects model.
OPAT The protocol of this review was registered on PROSPERO (CRD42023460389).
Parenteral antimicrobial therapy Result: Thirteen RCTs, involving 1,310 participants were included. We found no difference in mortality
Safety (Risk Ratio [RR] 0.54, 95% Confidence Interval [CI] 0.23 to 1.26; P = 0.93), treatment failure (RR 1.0, CI
Efficacy
0.59 to 1.72; P = 0.99), adverse reaction related to antimicrobials (RR 0.89, CI 0.69 to 1.15; P = 0.38), and
administration device (RR 0.58, CI 0.17 to 1.98; P = 0.87) between outpatient and inpatient parenteral
antimicrobial therapy. The overall body of evidence had a low level of certainty.
Conclusion: Existing evidence suggests OPAT is a safe and effective alternative to inpatient treatment. Fur-
ther RCTs are warranted for a thorough comparison of inpatient and outpatient parenteral antimicrobial
therapy with a high level of certainty.
© 2024 The Author(s). Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

1. Introduction

Globally, around one-third of hospitalised patients receive par-

Abbreviations: CI, Confidence interval; C-OPAT, Outpatient clinic or infusion enteral antimicrobials to treat serious infections [1]. Although in-
centre administration; GRADE, Grading of Recommendations Assessment, Devel-
opment, and Evaluation; HITH, hospital-in-the-home; H-OPAT, physician or nurse-
patient (bed-based) treatment of severe infection may be neces-
administered OPAT at patient home; IM, Intramuscular; IV, Intravenous; MD, Mean sary in some patients, hospital admission for parenteral antimicro-
differences; OPAT, Outpatient Parenteral Antimicrobial Therapy; PRISMA, Preferred bial therapy contributes to bed space competition and increased
Reporting Items for Systematic Reviews and Meta-Analyses; RCTs, Randomized risk of acquiring hospital-related infections [2]. While the option
controlled trials; RR, Risk ratio; SD, Standard deviation; S-OPAT, self, or carer-
of discharging patients with oral antimicrobials exists, its clinical
administered OPAT at patients’ home.
∗
Correspondence: Fekade B Sime, UQ Centre for Clinical Research, The University
feasibility is contingent upon several factors, including, among oth-
of Queensland, Building 71/918 RBWH Herston, Brisbane City, QLD 4006, ers, antimicrobial susceptibility, infection severity, infection type,
E-mail address: f.sime@uq.edu.au (F.B. Sime). gastrointestinal issues, and patient adherence [3]. Previously, intra-

https://doi.org/10.1016/j.ijantimicag.2024.107263
0924-8579/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

venous antimicrobial therapy was considered a barrier to hospi- any restrictions on language, publication year or publication for-
tal discharge. However, the development of outpatient parenteral mat. The specific search strategy for each electronic database is
antimicrobial therapy (OPAT) services, commonly administered in presented in the supplementary file. Additionally, we examined the
hospital-in-the-home (HITH) programs, allows patients to complete reference lists of the included studies and review articles to iden-
their treatment outside the hospital setting [4]. tify relevant studies.
The transition from exclusive inpatient treatment to outpatient
alternative for intravenous antimicrobial therapy began almost 50 2.3. Data collection and analysis
years ago in the United States [5]. Outpatient treatment offers
convenience, enabling patients to maintain their regular routines The eligibility screening of studies was conducted by two inde-
without requiring an overnight hospital stay. Recuperating at home pendent reviewers, SAM and GMA. Both reviewers examined the
with emotional support from loved ones contribute to a more pos- abstracts and full-text articles using predefined inclusion criteria.
itive healing experience, enhancing patient satisfaction and poten- In cases where there was disagreement, a discussion took place be-
tially speeding up recovery [6]. Furthermore, OPAT reduces the risk tween the two reviewers to reach a consensus. If necessary, a third
of hospital-acquired infections, which is of particular importance reviewer, FS, was involved in the discussion to assist in resolving
considering the global problem of infections by multidrug-resistant any disagreements. The level of agreement between the two re-
pathogens [7]. viewers was assessed and reported using the kappa value. To facil-
OPAT improves healthcare access through patient-centred intra- itate the screening process, we utilised Covidence (Veritas Health
venous treatment. Whether self-administered or supervised by a Innovation, Melbourne, Australia).
caregiver or visiting health professional, OPAT reduces the need for
2.4. Data extraction and management
frequent hospital visits, travel challenges, and waiting times. This
approach is cost-effective, alleviating the financial strain associated
We employed a standardised form for data extraction. One re-
with prolonged hospital stays [8].
viewer, SAM, extracted data on the characteristics of trials, includ-
The safety and efficacy of OPAT in comparison to inpatient
ing the year of the trial, study setting, trial design and author(s).
treatment have been evaluated in a limited number of reviews.
Additionally, data on the characteristics of study participants, such
The reviews by Bryant et al. [9], Mitchell et al. [10] and Sriskan-
as the number of participants, their age and gender, were ex-
darajah et al. [11] were based on observational studies, and the
tracted. Data on various treatment outcomes, including mortality,
search was conducted over 7 years ago. Wen and colleagues con-
treatment failure, antimicrobial-related adverse reactions, compli-
ducted a meta-analysis on the safety and efficacy of OPAT in pa-
cations related to the use of medication administration device and
tients with infective endocarditis, relying on observational stud-
duration of antimicrobial therapy were also extracted. To ensure
ies [12]. Pacheco and colleagues conducted a meta-analysis of ran-
accuracy and reliability, the extracted data was then checked by
domised controlled trials (RCTs) on the use of intravenous antimi-
the senior reviewer, FS. Data extraction was performed using per-
crobial therapy in the adult population [13]. However, studies with
protocol analysis, whenever possible, aiming to evaluate the safety
a comprehensive analysis encompassing both children and adult
and efficacy of parenteral antimicrobial therapy among patients
populations across diverse disease conditions are lacking. There-
who strictly adhered to the treatment protocol either inpatient or
fore, the aim of the present systematic review and meta-analysis
outpatient settings.
was to conduct a comprehensive evaluation and synthesise existing
evidence regarding the relative efficacy and safety of OPAT versus 2.5. Assessment of risk of bias
inpatient parenteral antimicrobial therapy.
One of the reviewers (SAM) evaluated the risk of bias using the
2. Methods Cochrane risk of bias assessment tool for randomised studies [15].

We conducted a review following the Preferred Reporting Items 2.6. Data analysis
for Systematic Reviews and Meta-Analyses (PRISMA) guideline [14].
The protocol was registered on PROSPERO (CRD42023460389). We calculated risk ratios for mortality, treatment failure, ad-
verse reactions related to antimicrobials and complications related
2.1. Study selection to the use of medication administration devices such as intra-
venous lines and infusion devices. The included studies defined
RCTs comparing conventional inpatient parenteral antimicrobial treatment failure differently; therefore, it was considered as re-
therapy with OPAT were included. The study population consisted ported by the studies. An adverse drug reaction is any reactions
of both adult and paediatric patients receiving parenteral antimi- related to the use of an antimicrobial regimen. Complications re-
crobial treatment in either outpatient or inpatient setting. Outpa- lated to medication administration devices encompasses any ad-
tient setting in this context refers to the administration of antimi- verse events resulting from the use of devices employed to admin-
crobials in any ambulatory setting, including fixed clinics, emer- ister the antimicrobials. Mean difference (MD) was calculated for
gency departments, and/or patient’s place of residence. The inter- the duration of antimicrobial treatment (days after the start of an-
vention was parenteral antimicrobial therapy in an outpatient set- timicrobial therapy). In cases where studies reported medians with
ting, while the comparator was parenteral antimicrobial therapy in a 95% confidence interval (CI), we calculated the SD accordingly.
an inpatient setting. The primary measure of outcome was mortal- However, for studies that only reported medians and ranges, data
ity. The secondary measures of outcome included treatment failure, was extracted as presented and subsequently incorporated into the
adverse drug reactions related to antimicrobials, complications re- narrative. All estimates and their 95% CI were generated using the
lated to the use of medication administration devices, and duration R software (version 2023.03.1 + 446) (Posit Software, Vienna, Aus-
of antimicrobial treatment. tria).
We employed a random effects model (Mantel–Haenszel
2.2. Search strategy method) for analyses [16], considering the heterogeneity within
the included studies. Heterogeneity was assessed using the I-
We conducted searches in the Cochrane Library, MEDLINE, Em- squared statistic, where a value exceeding 50% indicates substan-
base, PubMed and Web of Sciences databases without imposing tial heterogeneity [17]. Subgroup analysis was conducted to assess

2
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

Figure 1. Study eligibility screening.

variations across populations (children and adult) and models of process and the reason for excluding 5 RCTs [32–36], is provided
OPAT care (S-OPAT, self or carer-administered OPAT at patients’ in Figure 1.
home; H-OPAT, physician or nurse-administered OPAT at patient
home; C-OPAT, outpatient clinic or infusion centre administration). 3.2. Characteristics of included studies
We prioritised the subgroup value when the overall heterogene-
ity test had a P-value of less than 0.05. Furthermore, we utilised a Characteristics of the included studies [19–31] are provided in
funnel plot to examine publication bias within the included stud- Table 1. The studies were published between 1997 and 2019. In
ies. A sensitivity analysis was conducted to assess the robustness total, 1310 participants (1363 treatment episodes) were involved,
of the primary analysis. This analysis included examining the influ- with the number of participants per trial ranging from 20 to 200.
ence of disease conditions and antimicrobials, where we restricted Of the total participants, 697 (53.2%) were male. The participants’
the analysis to the same disease condition and antimicrobials used. ages ranged from 6 months [23] to 94 years [21]. Specifically, two
The Grading of Recommendations Assessment, Development, and RCTs included children (under 18 years) [19,23], while six RCTs in-
Evaluation (GRADE) approach [18] was utilised to summarise the cluded adults (over 18 years) [20,22,26–29]. Additionally, one RCT
certainty of evidence obtained within this review. considered participants under 21 years of age [24] as children for
analysis. In the remaining RCTs, participants aged over 14 years
[31] and 16 years [21,25,30] were categorised as part of the adult
3. Results population for analysis.
The included RCTs involved the treatment of patients with
3.1. Search results febrile neutropenia [19,24,26,28,29], cellulitis [21,23], severe in-
jecting drug use-related infections [22], peritonsillar abscess [25],
Out of 6,619 studies identified, 18 studies were deemed eligi- Staphylococcus aureus bacteremia and infective endocarditis [27],
ble for a full-text review. Finally, 13 RCTs [19–31] were included in cystic fibrosis [30] and other infectious diseases [20,31]. Two RCTs
the final analysis (Fig. 1). The concordance rate between reviewers, [22,27] included patients diagnosed with methicillin-resistant S.
both during title and abstract screening as well as full-text review, aureus infection. In one of these trials, patients were treated
was 98.49% and 100%, respectively. The flow of the study selection with either vancomycin or daptomycin [27]. One RCT reported

3
 S.A. Mohammed, J.A. Roberts, M.O. Cotta et al.
Table 1
Characteristics of included studies.

Author and Country Design Type of disease OPAT Discharge Population % Male Intervention Adverse Cost of Readmission rate
year condition model time event rate treatment per
patient (mean)

Inpatient Outpatient

Ahmed et al., Egypt Single-centre, Febrile C-OPAT After 72 Children 59.7 Imipenem 80–100 Ceftriaxone Outpatient Outpatient Not indicated
2007 [19] open-label neutropenia hours mg/kg/day every 6 100 mg/kg/day 13.11%, $1,208,
hours, IV plus Amikacin Inpatient Inpatient
15 mg/kg/day, 6.9% $1,783
IV
Caplan et al., Australia Single centre Acute illnessa H-OPAT After 24 Adult 45.0 Not indicated Ceftriaxone, Outpatient Not indicated Outpatient 5.9%,
1999 [20] hours Gentamicin, 15.7%, Inpatient 10.2%
and Inpatient readmission within
Vancomycin, IV 22.45% 28 days
Corwin et al., New Single centre Cellulitis H-OPAT After 48 Adult 67.5 Cephazolin, Cephazolin 2 g, Not indicated Not indicated Outpatient 11.22%,
2004 [21] Zealand hours Flucloxacillin, and IV Inpatient 3.13%
other (not indicated), readmission within
IV one month
4

Fanucchi United Single-centre, Severe C-OPAT After 48 Adult 30.0 Not indicated Not indicated Outpatient Not indicated Outpatient 30%,
et al., 2019 States of superiority injecting drug hours 0%, Inpatient Inpatient 20%
[22] America trial use related 10% readmission within
infections 12 weeks
Ibrahim Australia Single-centre, Cellulitis H-OPAT, After the Children 54.3 Flucloxacillin 50 mg/kg Ceftriaxone 50 Not indicated Outpatient Outpatient 2.25%
et al., 2019 open-label, telemedicine first dose every 6 hours, IV mg/kg/day, IV AUD 530,
[23] non-inferiority Inpatient AUD
trial 1,297
Orme et al., Australia Single centre, Febrile H-OPAT After 48 Children 63.0 Cefepime 50 mg/kg Outpatient 0%, Not indicated Outpatient

International Journal of Antimicrobial Agents 64 (2024) 107263

2014 [24] open-label neutropenia hours every 12 hours, IV Inpatient 0% 31.58%
Ozbek et al., Turkey Single centre Peritonsillar Not not Adult 56.9 Ampicillin/Sulbactam 1 Clindamycin Not indicated Not indicated Not indicated
2005 [25] abscess indicated indicated g every 6 hours, IV 600 mg every
12 hours, IM
Rapport Six coun- Open-label, Febrile S-OPAT After Adult 41.7 Ceftriaxone 2 g/day Not indicated Not indicated Not indicated
et al., 1999 triesb multicentre, neutropenia 48–72 plus (Gentamicin,
[26] non-inferiority hours Netilmicin 4.5–6.5
trial mg/kg, or Amikacin 20
mg/kg), IV
(continued on next page)
 S.A. Mohammed, J.A. Roberts, M.O. Cotta et al.
Table 1 (continued)

Author and Country Design Type of disease OPAT Discharge Population % Male Intervention Adverse Cost of Readmission rate
year condition model time event rate treatment per
patient (mean)

Inpatient Outpatient

Rehm et al., USA Single-centre, Staphylococcus S-OPAT, After 9 Adult 59.5 Nafcillin, Oxacillin, or Daptomycin 6 Outpatient Not indicated Outpatient 17.48%
2009 [27] open-label, aureus H-OPAT, days flucloxacillin 2 g every mg/kg/day, IV 46.6%,
non-inferiority bacteraemia C-OPAT 4 hours or Vancomycin Inpatient
and infective 1 g every 24 hours, 53.61%
endocarditis and Gentamicin 1
mg/kg every 8 hours,
IV
Santolaya Chile Multicentre Febrile C-OPAT After 24 to Children 45.8 Ceftriaxone 100 Not indicated Outpatient Not indicated
et al., 2004 neutropenia 36 hours mg/kg/d every 24 £308,
[28] hours and Teicoplanin Inpatient
20 mg/kg/d every 12 £752
hours for the first day
then 10 mg/kg/d every
24 hours, IV
5

Talcott et al., United Multicentre Febrile H-OPAT Not Adult 46.0 Aminoglycoside and Not indicated Not indicated Outpatient
2011 [29] States of neutropenia indicated Vancomycin or 8.51%
America Ceftazidime alone; for
allergic patients’
Imipenem alone or an
Aztreonam-containing
regimen, IV
Wolter et al., Australia Single centre, Cystic fibrosis S-OPAT After 2–4 Adult 32.3 Ceftazidime and Outpatient 0%, Outpatient 0% short-term
1997 [30] two-factor days Tobramycin, Imipenem Inpatient 5.56% $190.46, readmission

International Journal of Antimicrobial Agents 64 (2024) 107263

mixed design used on allergy or Inpatient rate
failure, IV $440.3
Wolter et al., Australia Single centre Infectious Home After the Adult 40.2 Not indicated Not indicated Outpatient Outpatient Outpatient 15.9%,
2004 [31] diseasesc infusion first dose 11.36%, £3686, Inpatient 10.52%
Inpatient Inpatient readmission within
13.16% £6936d 30 days
a
Pneumonia, urinary tract infections, cellulitis, endocarditis, osteomyelitis, deep vein thrombosis, congestive cardiac failure, acute back pain, faecal impaction, acute myocardial infarct, shingles, anaemia, cerebrovascular accident.
b
South Africa, Colombia, Israel, Perú, Argentina, Spain.
c
Urinary tract infection, cellulitis, pneumonia, osteomyelitis, mycobacterial infection, cystic fibrosis, bronchiectasis.
d
Total cost of treatment.Abbreviations: AUD, Australia Dollar; S-OPAT, self or carer-administered OPAT at patients’ home; H-OPAT, physician or nurse-administered OPAT at patient home; C-OPAT, outpatient clinic or infusion
centre administration; IV, intravenous; IM, intramuscular.
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

Table 2
Risk of bias of the included studies.

+, low risk of bias; -, high risk of bias; ?, unclear risk of bias.

the incidence of healthcare-associated infections among hospi- RCTs in adults [20,21,25–27,29,31]. It is worth noting that a higher
talised patients. In this study, four patients were colonized with number of patients in the inpatient groups experienced treatment
vancomycin-resistant enterococci and one with methicillin-resistant failure compared to outpatients. However, no statistically signifi-
S. aureus [31]. Nine RCTs reported readmission rates [20–24,27,29– cant difference in the risk of treatment failure was observed (RR
31], which ranged from 0% to 31.58% among OPAT patients, and 0% 1.0, 95% CI 0.59 to 1.72; P = 0.99; I2 38%, 1,312 participants; low-
to 20% for inpatient patients. The cost of parenteral antimicrobial certainty evidence) (Fig. 3).
treatment has been documented in five RCTs [19,23,28,30,31]. All
of these trials have reported the effectiveness of the OPAT program 3.6. Adverse drug reactions related to antimicrobials
compared to inpatient treatment.
Adverse drug reaction related to parenteral antimicrobials was
3.3. Risk of bias reported by 6 RCTs; 2 in children [19,24] and 4 in the adult pop-
ulation [20,27,30,31]. One RCT in each of the children [24] and
The risk assessment for the included studies is provided in adult population [30] reported absence of adverse reactions in ei-
Table 2. Some of the RCTs lacked adequate information on se- ther treatment groups. Although the risk of an adverse drug re-
quence generation, allocation concealment, and participant blind- action related to parenteral antimicrobials was higher in the in-
ing. A summary of the risk assessment and the certainty ratings patient group, no significant evidence of a difference between the
of the evidence are presented in Supplementary Tables S1 and S2, two groups was found (RR 0.89, 95% CI 0.69 to 1.15; P = 0.38; I2
respectively. 0%, 569 participants; low-certainty evidence) (Fig. 4).

3.4. Mortality 3.7. Complications related to the medication administration devices

Two RCTs in children [19,28] and six RCTs in adult All of the RCTs that assessed complications related to the use
[20,22,26,27,29,30] assessed the outcome of mortality in both of administration devices were conducted in the adult population
OPAT and inpatient treatment across children and adult popula- [20,22,30,31]. Despite a higher number of complications among pa-
tions. Two RCTs [29,30] in the adult population reported no death tients receiving parenteral antimicrobial therapy in an inpatient
in either treatment settings. Although the number of deaths in setting, no difference was observed when compared to the out-
both the children and adult population was higher among hospi- patient group (RR 0.58, 95% CI 0.17 to 1.98; P = 0.39; I2 0%, 233
talised patients, there was no significant difference in mortality participants; low-certainty evidence) (Supplementary Fig. S1).
between outpatient and inpatient parenteral antimicrobial therapy
(RR 0.54, 95% CI 0.23 to 1.26; P = 0.16; I2 18%, 812 participants; 3.8. Duration of antimicrobial treatment
low-certainty evidence) (Fig. 2).
Three RCTs in children [19,24,28] reported duration of antimi-
3.5. Treatment failure crobial treatment. The duration did not significantly differ between
the outpatient and inpatient groups (MD -0.16, 95% CI -1.35 to 1.02;
The efficacy of parenteral antimicrobial treatment was assessed P = 0.79; I2 59%, 305 participants). In the adult population, five
across 11 RCTs, with 4 focusing on children [19,23,24,28] and 7 RCTs compared the duration of antimicrobial treatment, but only

6
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

Figure 2. A forest plot comparison of mortality between outpatient and inpatient antimicrobial therapy.

Figure 3. A forest plot comparison of treatment failure between outpatient and inpatient antimicrobial therapy.

7
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

Figure 4. A forest plot comparison of adverse drug reactions related to antimicrobials during outpatient versus inpatient antimicrobial therapy.

two were included in the meta-analysis. Notably, these two RCTs This meta-analysis provides evidence that OPAT is non-inferior
revealed a significant difference, with a shorter duration of antimi- to the traditional hospital-bed based intravenous antimicrobial
crobial treatment obtained in the inpatient group compared to the therapy, given no significant differences were observed either in
outpatient group (MD 14.03, 95% CI 8.12 to 19.94; P < 0.01; I2 mortality, treatment failure, or rate of adverse events. This find-
60%, 220 participants) (Supplementary Fig. S2). The three excluded ing is generally consistent with a prior review by Pacheco and col-
studies reported the range and median duration of antimicrobial leagues’ [13] albeit that methodological differences limit compari-
treatment, therefore pooling of data was not possible [26,30,31]. son across all outcome measures investigated. Their study included
The outcomes of different OPAT models of care (S-OPAT, C-OPAT, only six RCTs in adult patients; however, the current review in-
H-OPAT) employed in the RCTs were analysed to see if any model cluded an additional three RCTs in adults and four RCTs in chil-
is superior to the others. There was no significant difference in dren. Our review also included the assessment of additional out-
terms of mortality, treatment failure, adverse reactions associated come measures of adverse reactions related to drug administration
with the antimicrobial agent, or complications related to medica- devices and the duration of antimicrobial treatment.
tion administration devices (Supplementary Figs. S3–S7). Multiple factors contribute to the efficiency of OPAT program
To confirm the consistency of the primary analysis results, and subsequently to the observed no difference between OPAT and
a sensitivity analysis was conducted, restricting the analysis to inpatient treatment in terms of adverse reactions associated with
RCTs that specifically evaluated the same type of treatment condi- either antimicrobial or administration device use. In particular, the
tions and interventions. The results for all outcomes remained un- fact that OPAT is indicated for carefully assessed and selected pa-
changed, as detailed in the Supplementary Table S3. Additionally, tients minimises the risk of adverse events that might surface in
the funnel plot for the outcome of mortality (Supplementary Fig. the abscess of close monitoring afforded in hospital bed-based
S8) indicated no evidence of publication bias. This was further sup- therapy [37]. The administration of antimicrobials to these pa-
ported by Egger’s regression test estimate, which yielded a P-value tients occurs through intravenous infusion or bolus injection, with
of 0.066, signifying no significant evidence of publication bias. the choice of delivery devices considering critical factors including
compatibility with the antimicrobials and treatment duration. Ad-
ditionally, the choice of antimicrobials is guided by the health ser-
4. Discussion vice’s antimicrobial stewardship agenda. Furthermore, patient ed-
ucation and proper use of the administration device enhances the
This review evaluated the relative safety and efficacy of par- safety of OPAT [4,38].
enteral antimicrobial treatment in inpatient versus outpatient set- The comparative safety and efficacy of OPAT, relative to inpa-
tings. A thorough quantitative analysis was conducted to assess tient parenteral antimicrobial therapy, suggest that clinically sta-
various critical outcomes, including mortality, treatment failure, ble patients, who may be hospitalised for parenteral antimicro-
adverse drug reactions related to antimicrobials, complications bial therapy, can be discharged from hospital and transition to
arising from the use of drug administration devices, and dura- OPAT [4]. This transition offers enhanced patient comfort and an
tion of antimicrobial therapy. Major findings of the study include improved overall quality of life, fostering a more patient-centred
firstly, there was no significant difference in mortality benefit be- approach. Importantly, the program alleviates the strain on inpa-
tween OPAT and inpatient intravenous antimicrobial therapy. Sec- tient facilities by freeing up hospital beds and allowing for more
ondly, the rate of treatment failure was comparable between OPAT efficient resource allocation [39]. Moreover, OPAT plays a crucial
and inpatients therapy suggesting that the setting of parenteral an- role in antimicrobial stewardship efforts by promoting appropri-
timicrobial administration by itself does not significantly affect the ate use of antibiotics. Its role in reducing the risk of hospital-
effectiveness of the antimicrobial agent. Thirdly, there was no sig- acquired infections is significant, especially in the wake of the cur-
nificant difference in the rates of adverse reactions associated with rent global challenge posed by multidrug-resistant nosocomial in-
either the antimicrobial agent or the administration devices used. fections [7]. OPAT can either avoid hospitalisation or enable earlier

8
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

discharge, thereby reducing the risk of hospital acquired infection ferent agents in inpatient and outpatient settings. Moreover, cer-
and the overall healthcare cost. Furthermore, the program’s ability tain regimens used in these studies may no longer reflect current
to expand treatment access contributes significantly to equitable clinical practice. Sixth, the included studies exhibited heterogene-
healthcare delivery, ensuring more patients have access to neces- ity in the types of disease conditions evaluated. Many of the in-
sary treatments, particularly for those requiring prolonged antimi- cluded studies investigated a less frequently treated disease condi-
crobial therapy [8]. tion in OPAT, specifically Febrile Neutropenia. Consequently, trans-
The spectrum of clinical conditions effectively managed in out- lating these findings to more traditional OPAT practice may present
patient settings under the OPAT program has continued to evolve. a challenge. Nonetheless, sensitivity analyses that were restricted
A diverse range of infections, including skin and soft tissue in- to the same disease conditions and antimicrobial treatments con-
fections, bone and joint infections, bacteremia, wound infections, firmed the primary analysis. The overall statistical heterogeneity
pneumonia, urinary tract infections, intra-abdominal infections, en- also remained low.
docarditis and central nervous system infections, have been treated
through the OPAT program [40]. Furthermore, OPAT plays a pivotal
5. Conclusion
role in managing complex cases that pose a challenge to traditional
treatment modalities, particularly in scenarios involving drug users
Existing evidence suggests a comparative safety and efficacy of
[41], patients experiencing failure with oral therapies or individuals
OPAT relative to conventional inpatient (bed-based) parenteral an-
deemed unsuitable for oral regimens due to suspected noncompli-
timicrobial treatment. OPAT serves as a safe and effective alter-
ance or poor reliability [3].
native to conventional inpatient care for patients in need of par-
Advancements in infusion device technology have enabled the
enteral antimicrobial therapy. Further studies with robust method-
outpatient administration of antimicrobials previously considered
ologies are crucial to enable a more comprehensive comparison
impractical for such settings. Given adequate resources, most an-
between parenteral antimicrobial treatment administered in inpa-
timicrobials can now be part of OPAT for treating various gram-
tient (bed-based) and outpatient settings (HITH/OPAT programs).
positive or gram-negative infections [42]. Antimicrobials for OPAT
To this effect, we suggest establishing national or international
are selected based on various factors, including the OPAT model
OPAT database(s) that enable ongoing monitoring of the safety and
of care, the likely infecting organism, the pharmacodynamic and
efficacy of the program.
pharmacokinetic properties of potential drugs, and drug stability
[38]. The mode of administration typically involves parenteral de-
livery. The frequency of dosing schemes varies depending on fac- CRediT authorship contribution statement
tors such as the antibiotic’s pharmacokinetics, the severity of the
infection, and the patient’s response to treatment. Stability consid- Solomon Ahmed Mohammed: Conceptualization, Methodol-
erations are crucial to ensure the effectiveness and safety of the ogy, Investigation, Formal analysis, Data curation, Writing – orig-
antimicrobial during storage and administration [38,42]. inal draft. Jason A. Roberts: Supervision, Writing – review & edit-
In this meta-analysis, inpatient treatment was observed to have ing. Menino Osbert Cotta: Supervision, Writing – review & edit-
a shorter duration of antimicrobial therapy compared to OPAT. ing. Benjamin Rogers: Writing – review & editing. James Pollard:
However, the evidence available for this outcome remains of very Writing – review & editing. Getnet Mengistu Assefa: Investigation.
low certainty. Therefore, the difference between the duration of Daniel Erku: Supervision, Writing – review & editing. Fekade B.
outpatient and inpatient antimicrobial treatment may deviate from Sime: Conceptualization, Methodology, Supervision, Writing – re-
the estimated effect, highlighting the need for further research and view & editing.
more robust data to draw definitive conclusions.
Declarations
Strength and limitations
Funding: This research did not receive any specific grant from
Our review followed the PRISMA guidelines, ensuring a robust funding agencies in the public, commercial or not-for-profit sec-
and transparent methodology. A comprehensive search was con- tors.
ducted across multiple databases, employing a broad search strat-
egy that minimised the risk of inadvertently excluding potentially Competing interests: The authors declare no potential conflict of
relevant studies. Another strength lies in the inclusive nature of interest.
the review, which encompassed both children and adult popula-
tions, followed by a sub-group analysis. Furthermore, the quality Ethical approval: Not applicable.
of evidence was assessed using the GRADE approach, ensuring a
rigorous and systematic evaluation of the available data.
Acknowledgements
However, the study is not without limitations. First, some of
the included RCTs inadequately described or executed processes
J.A. Roberts would like to acknowledge funding from the Aus-
such as randomization, blinding, and allocation, thereby potentially
tralian National Health and Medical Research Council for a Centre
compromising the reliability of the findings. Second, majority of
of Research Excellence (APP20 070 07) and an Investigator Grant
the studies included smaller sample sizes, leading to wider con-
(APP2009736) as well as an Advancing Queensland Clinical Fellow-
fidence intervals and reduced precision in the estimates. Third,
ship. Fekade Sime acknowledges support from Australian National
some eligible studies reported limited data; for example, the out-
Health and Medical Research Council (NHMRC) Investigator Grant
come of duration of antimicrobial treatment was reported as me-
(APP1197866).
dian and restricted the possibility of conducting a more compre-
hensive meta-analysis. Fourth, the included RCTs lack consistency
in reporting readmission rates, which hinders a clear understand- Supplementary materials
ing of rehospitalization patterns. Fifth, the diverse geographic lo-
cations of the included RCTs may limit the generalisability of find- Supplementary material associated with this article can be
ings, reflecting variations in social, economic, cultural, and health- found, in the online version, at doi:10.1016/j.ijantimicag.2024.
care systems. Fifth, some of the included studies employed dif- 107263.

9
S.A. Mohammed, J.A. Roberts, M.O. Cotta et al. International Journal of Antimicrobial Agents 64 (2024) 107263

References [23] Ibrahim LF, Hopper SM, Orsini F, Daley AJ, Babl FE, Bryant PA. Efficacy and
safety of intravenous ceftriaxone at home versus intravenous flucloxacillin in
[1] Versporten A, Zarb P, Caniaux I, Gros M-F, Drapier N, Miller M, et al. Antimi- hospital for children with cellulitis (CHOICE): a single-centre, open-label, ran-
crobial consumption and resistance in adult hospital inpatients in 53 coun- domised, controlled, non-inferiority trial. Lancet Infect Dis 2019;19:477–86.
tries: results of an internet-based global point prevalence survey. Lancet Glob [24] Orme LM, Babl FE, Barnes C, Barnett P, Donath S, Ashley DM. Outpatient
Health 2018;6:e619–ee29. versus inpatient IV antibiotic management for pediatric oncology patients
[2] Monegro AF, Muppidi V, Regunath H. Hospital-Acquired Infections. [Updated with low risk febrile neutropenia: a randomised trial. Pediatr Blood Cancer
2023 Feb 12]. In: StatPearls [Internet]. Treasure Island, FL: StatPearls Publish- 2014;61:1427–33.
ing; 2024. Available from: https://www.ncbi.nlm.nih.gov/books/NBK441857/. [25] Ozbek C, Aygenc E, Unsal E, Ozdem C. Peritonsillar abscess: a comparison of
[3] Kim J, De Jesus O. Medication routes of administration. StatPearls. Treasure Is- outpatient i.m. clindamycin and inpatient i.v. ampicillin/sulbactam following
land, FL: StatPearls Publishing; 2021. needle aspiration. Ear Nose Throat J 2005;84:366–8.
[4] Chapman AL, Patel S, Horner C, Green H, Guleri A, Hedderwick S, et al. Up- [26] Rapoport BL, Sussmann O, Herrera MV, Schlaeffer F, Otero JC, Pavlovsky S,
dated good practice recommendations for outpatient parenteral antimicro- et al. Ceftriaxone plus once daily aminoglycoside with filgrastim for treatment
bial therapy (OPAT) in adults and children in the UK. JAC Antimicrob Resist of febrile neutropenia: early hospital discharge vs. standard in-patient care.
2019;1:dlz026. Chemotherapy 1999;45:466–76.
[5] Rucker RW, Harrison GM. Outpatient intravenous medications in the manage- [27] Rehm S, Campion M, Katz DE, Russo R, Boucher HW. Community-based out-
ment of cystic fibrosis. Pediatrics 1974;54:358–60. patient parenteral antimicrobial therapy (CoPAT) for Staphylococcus aureus bac-
[6] Durojaiye OC, Bell H, Andrews D, Ntziora F, Cartwright K. Clinical efficacy, teraemia with or without infective endocarditis: analysis of the randomized
cost analysis and patient acceptability of outpatient parenteral antibiotic ther- trial comparing daptomycin with standard therapy. J Antimicrob Chemother
apy (OPAT): a decade of Sheffield (UK) OPAT service. Int J Antimicrob Agents 2009;63:1034–42.
2018;51:26–32. [28] Santolaya ME, Alvarez AM, Avilés CL, Becker A, Cofré J, Cumsille MA, et al.
[7] Barr DA, Semple L, Seaton RA. Outpatient parenteral antimicrobial therapy Early hospital discharge followed by outpatient management versus continued
(OPAT) in a teaching hospital-based practice: a retrospective cohort study hospitalization of children with cancer, fever, and neutropenia at low risk for
describing experience and evolution over 10 years. Int J Antimicrob Agents invasive bacterial infection. J Clin Oncol 2004;22:3784–9.
2012;39:407–13. [29] Talcott JA, Yeap BY, Clark JA, Siegel RD, Loggers ET, Lu C, et al. Safety of early
[8] Tan SJ, Ingram PR, Rothnie AJ, Whitmore TJ, Robinson JO, Hatch JB, et al. Suc- discharge for low-risk patients with febrile neutropenia: a multicenter ran-
cessful outpatient parenteral antibiotic therapy delivery via telemedicine. J An- domized controlled trial. J Clin Oncol 2011;29:3977–83.
timicrob Chemother 2017;72:2898–901. [30] Wolter JM, Bowler SD, Nolan PJ, McCormack JG. Home intravenous therapy in
[9] Bryant PA, Katz NT. Inpatient versus outpatient parenteral antibiotic therapy cystic fibrosis: a prospective randomized trial examining clinical, quality of life
at home for acute infections in children: a systematic review. Lancet Infect Dis and cost aspects. Eur Respir J 1997;10:896–900.
2018;18:e45–54. [31] Wolter JM, Cagney RA, McCormack JG. A randomized trial of home vs hospi-
[10] Mitchell E, Murray CC, Meads D, Minton J, Wright J, Twiddy M. Clinical and tal intravenous antibiotic therapy in adults with infectious diseases. J Infect
cost-effectiveness, safety and acceptability of community intravenous antibi- 2004;48:263–8.
otic service models: CIVAS systematic review. BMJ Open 2017;7:e013560. [32] Ibrahim LF, Huang L, Hopper SM, Dalziel K, Babl FE, Bryant PA. Intravenous
[11] Sriskandarajah S, Hobbs J, Roughead E, Ryan M, Reynolds K. Safety and ef- ceftriaxone at home versus intravenous flucloxacillin in hospital for children
fectiveness of ‘hospital in the home’ and ‘outpatient parenteral antimicrobial with cellulitis: a cost-effectiveness analysis. Lancet Infect Dis 2019;19:1101–8.
therapy’ in different age groups: a systematic review of observational studies. [33] Peña DA, Viviani ST, Le Corre PN, Morales MV, Montecinos BC, Gajardo SC.
Int J Clin Pract 2018;72:e13216. [Treatment of urinary tract infections in febrile infants: experience of outpa-
[12] Wen W, Li H, Wang C, Chen C, Tang J, Zhou M, et al. Efficacy and safety of tient intravenous antibiotic treatment]. Rev Chilena Infectol 2009;26:350–4.
outpatient parenteral antibiotic therapy in patients with infective endocarditis: [34] Rappo U, Gonzalez PL, Puttagunta S, Akinapelli K, Keyloun K, Gillard P, et al.
a meta-analysis. Rev Esp Quim 2022;35:370. Single-dose dalbavancin and patient satisfaction in an outpatient setting in the
[13] Pacheco RL, de Oliveira Cruz Latorraca C, Dos Santos APP, Martimbianco ALC, treatment of acute bacterial skin and skin structure infections. J Glob Antimi-
de Fátima Carreira Moreira R, Logullo P, et al. Efficacy and safety of home- crob Resist 2019;17:60–5.
-based intravenous antibiotic therapy among adults: a systematic review. Int J [35] Richards DA, Toop LJ, Epton MJ, McGeoch GRB, Town GI, Wynn-Thomas SMH,
Antimicrob Agents 2022;59:106555. et al. Home management of mild to moderately severe community-acquired
[14] Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. pneumonia: a randomised controlled trial. Med J Aust 2005;183:235–8.
The PRISMA 2020 statement: an updated guideline for reporting systematic [36] Wilimas JA, Flynn PM, Harris S, Day SW, Smith R, Chesney PJ, et al. A ran-
reviews. Int J Sur 2021;88:105906. domized study of outpatient treatment with ceftriaxone for selected febrile
[15] Higgins JP, Savović J, Page MJ, Elbers RG, Sterne JA, et al. Assessing risk of bias children with sickle cell disease. N Engl J Med 1993;329:472–6.
in a randomized trial. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, [37] Chapman AL, Seaton RA, Cooper MA, Hedderwick S, Goodall V, Reed C,
Page MJ, et al., editors. Assessing risk of bias in a randomized trial Cochrane et al. Good practice recommendations for outpatient parenteral antimicro-
handbook for systematic reviews of interventions. London, United Kingdom: bial therapy (OPAT) in adults in the UK: a consensus statement. J Antimicrob
Cochrane Collaboration; 2019. p. 205–28. Chemother 2012;67:1053–62.
[16] Kuritz SJ, Landis JR, Koch GG. A general overview of Mantel-Haenszel [38] Tice AD, Rehm SJ, Dalovisio JR, Bradley JS, Martinelli LP, Graham DR, et al.
methods: applications and recent developments. Annu Rev Public Health Practice guidelines for outpatient parenteral antimicrobial therapy. Clin Infect
1988;9:123–60. Dis 2004;38:1651–71.
[17] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in [39] Gaynor M, Anderson GF. Uncertain demand, the structure of hospital costs, and
meta-analyses. BMJ 2003;327:557–60. the cost of empty hospital beds. J Health Econ 1995;14:291–317.
[18] Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, et al. GRADE guide- [40] Williams DN, Baker CA, Kind AC, Sannes MR. The history and evolution
lines: 1. Introduction—GRADE evidence profiles and summary of findings ta- of outpatient parenteral antibiotic therapy (OPAT). Int J Antimicrob Agents
bles. J Clin Epidemiol 2011;64:383–94. 2015;46:307–12.
[19] Ahmed N, El-Mahallawy HA, Ahmed IA, Nassif S, El-Beshlawy A, El-Haddad A. [41] Suzuki J, Johnson J, Montgomery M, Hayden M, Price C. Outpatient parenteral
Early hospital discharge versus continued hospitalization in febrile pediatric antimicrobial therapy among people who inject drugs: a review of the litera-
cancer patients with prolonged neutropenia: a randomized, prospective study. ture. Open Forum Infect Dis 2018;5:ofy194.
Pediatr Blood Cancer 2007;49:786–92. [42] Norris AH, Shrestha NK, Allison GM, Keller SC, Bhavan KP, Zurlo JJ, et al.
[20] Caplan GA, Ward JA, Brennan NJ, Coconis J, Board N, Brown A. Hospital in the 2018 Infectious Diseases Society of America clinical practice guideline for the
home: a randomised controlled trial. Med J Aust 1999;170:156–60. management of outpatient parenteral antimicrobial therapy. Clin Infect Dis
[21] Corwin P, Toop L, McGeoch G, Than M, Wynn-Thomas S, Wells JE, et al. Ran- 2019;68:e1–e35.
domised controlled trial of intravenous antibiotic treatment for cellulitis at
home compared with hospital. BMJ 2005;330:129.
[22] Fanucchi LC, Walsh SL, Thornton AC, Nuzzo PA, Lofwall MR. Outpatient par-
enteral antimicrobial therapy plus buprenorphine for opioid use disorder and
severe injection-related infections. Clin Infect Dis 2020;70:1226–9.

10