DOI: https://doi.org/10.21203/rs.3.rs-2027467/v1
Background: There is a debt whether prophylactic antibiotic use may further hamper prognosis of patients with OHCA. In this study, we have performed a meta-analysis to assessed the effect of prophylactic antibiotic administration.
Methods: Electronic databases were searched for trials in which prophylactic antibiotic had been administered to adults with OHCA. The predefined primary outcome was the incidence of pneumonia.
Results: The included 6 trials enrolled a total of 5061 patients. Prophylactic antibiotic administration was not associated with decreased incidence of pneumonia and early-onset pneumonia (OR 0.44; 95%CI (0.19, 1.02); p= 0.056; I2=95.9% vs. OR 0.54; 95%CI (0.22, 1.32); p= 0.175; I2=46.8%). No adverse effect on mortality was found among trials (OR 1.17; 95%CI (0.46, 2.97); p= 0.748; I2=87.8%).
Conclusion: Given this conflicting collection of limited quality, no difference in the incidence of pneumonia (including early-onset pneumonia) and mortality was found when receiving prophylactic antibiotic administration in patients suffering out-of-hospital cardiac arrest. Randomized trials are warranted to define the best prophylactic antibiotic protocol.
PROSPERO registration number: CRD42022341601. 07 March, 2022 retrospectively registered.
Out-of-hospital cardiac arrest (OHCA) is one of the major public health challenges, with an average incidence of 55 adults of presumed cardiac cause per 100,000 person-years [1]. Infections were common in OHCA survivors [2], and the respiratory tract is the most common source [3]. Previous studies reported an increased length of sedation, time to extubation and intensive care unit (ICU) stay in patients with infectious complication [4,5].
The Brain Trauma Foundation recommended peri-intubation prophylactic antibiotic administration in patients with severe traumatic brain injury (Level II recommendation) [6]. However, there is a debt whether prophylactic antibiotic therapy can improve the chain of survival in the post resuscitation period among OHCA. A multicenter, double-blind, randomized, placebo-controlled trial [7] involved adult patients after OHCA presented benefits of antibiotic prophylaxis with a lower incidence of early ventilation-associated pneumonia. Paradoxically, the result was not replicated in the trial of Kim and co-workers [8].
Due to a paucity of consolidated data regarding the benefits of antibiotic prophylaxis among OHCA, we presented this meta-analysis intended to exam the associations between antibiotic prophylaxis and OHCA, with a primary outcome of the incidence of pneumonia.
This study was performed and prepared according to the guidelines proposed by Cochrane Collaboration in the Cochrane Handbook for Systematic Reviews of Interventions (http://www.cochrane handbook.org) and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [9,10] (e-Figure 1).
Search strategy
We searched articles of all languages published from inception to March 27, 2022 in PubMed, Embase, Ovid, Medline and Cochrane Central Register of Controlled Trials by a combination of MESH terms and keywords (e-Appendix 1). Additional potentially relevant references were identified through manually forward and backward citation tracking of included papers.
Study selection criteria
Inclusion and exclusion criteria
Observational and interventional trials included a control group were eligible for review inclusion, if they compared the effect of prophylactic antibiotics administration with delayed/ clinically-driven/ even no administration of antibiotics in adult patients following OHCA. Trials reported only as abstracts were excluded. Case reports and case series were not eligible for inclusion.
Data extraction
Data were independently extracted by the first and the second authors. Extracted data consisted of the name of first author, year of publication, study design, participants, interventions, clinical parameters and adverse events. We resolved disagreements through discussions until a consensus was reached. We did not contact study authors for further data.
Outcome measurements and definitions
The primary outcome was the incidence of pneumonia (including early-onset pneumonia). Secondary outcomes included survival and survival with good neurological outcome, length of ICU and hospital stay, incidence of sepsis and bacteremia. Early-onset pneumonia was considered pneumonia occurs within the first 5 days when recruited.
Assessment of risk of bias
We used the Cochrane Collaboration tool to assess the risk of bias in randomized controlled trials (RCTs) [10, 11]. Domains containing random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of the outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other bias were assessed. The remaining observational trials were assessed using the ROBINS-I tool [12]. Domains include bias due to confounding, bias in selection of participants into the study, bias in classification of interventions, bias due to deviation from intend intervention, bias due to missing data, bias in measurement of outcomes, and bias in selection of the reported results. We rated each domain of the trials as low risk, unclear, or high risk. Trials were considered low risk when each independent domain was rated as low risk. Any domain rated as unclear or high risk increased the overall risk score.
Statistical analysis
Data were analyzed using Statistics/Data Analysis 15.1. The results of dichotomous data were presented as forest plots through the odds ratios (ORs) with 95% confidence intervals (CIs). Forest plots using Weighted Mean Difference (WMD) with 95% CI were performed for the assessment of continuous data. We used random (M-H heterogeneity) model to assess the effects since clinical heterogeneity including study setting, study design, intervention was high. A Eegger test was conducted to assess the publication bias. A p value ≤ 0.05 was considered statistically significant, except when otherwise specified.
Assessment of the certainty of the evidence
Grading of Recommendations Assessment, Development and Evaluate system (GRADE) was used to evaluating the quality of evidence, based on the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness and publication bias) [13]. The first and the second authors independently rated the quality of evidence for the outcomes. The level of evidence was classified into four categories as high, moderate, low or very low. We resolved disagreements through discussions until a consensus was reached.
Literature search
Through database and manual search (Fig. 1), 21882 records were retrieved. We screened both titles and abstracts according to the inclusion and exclusion criteria, leaving 28 records that were deemed suitable for inclusion. After reviewing the full texts, 6 records [7,8,15-18] met the inclusion criteria and were finally included (Table 1). Studies were excluded for the following reasons: lack of control groups compared the effect of prophylactic antibiotics administration with delayed/ clinically-driven/ even no administration of antibiotics (n= 14); reported only as abstracts (n= 6); patients following in-hospital cardiac arrest (n= 1); protocol (n= 1) (e-Appendix 2).
Study characteristics
The six eligible studies comprised two RCTs and four observational trials, recruited a total of 5061 patients. Studies were published from 2012 to 2020. The characteristics of each trial are shown in Table 1. The sample sizes ranged from 60 to 2803, among which all received targeted temperature management (TTM) therapy.
Risk of bias and quality of evidence
The risk of bias was summarized in Fig. 2. Five trials were considered low risk. The trial performed by Ribaric and colleagues [16] despite being classed as a randomized study was deemed a high risk of bias as investigators were not blinded to the patient assessment.
Meta-analysis results
Five trials [7,8,14,15,17] compared the incidence of pneumonia between Prophylactic antibiotics administration (PRO)-group and control group, among them two trials concerning about early-onset pneumonia [7,8]. Results disclosed that PRO was not associated with decreased incidence of pneumonia and early-onset pneumonia (OR 0.44; 95%CI (0.19, 1.02); p= 0.056; I2=95.9%, Fig. 3; vs. OR 0.54; 95%CI (0.22, 1.32); p= 0.175; I2=46.8%, Fig. 4).
No adverse effect on mortality was found among trials [7,8,16,17] (OR 1.17; 95%CI (0.46, 2.97); p= 0.748; I2=87.8%, Fig. 4). No differences in other outcome parameters could be demonstrated, including CPC1-2 (OR= 1.18; 95%CI (0.91, 1.54); p= 0.212; I2=0.0%, Fig. 4) and ICU LOS (SMD= 0.05; 95%CI ((-0.20, 0.29); p= 0.706; I2=0.0%, Fig. 5).
A Eegger test was employed to assess the publication bias of the included trials, and the result suggested a minimal publication bias (p = 0.762).
Sensitivity analysis was performed through the removal of each individual trial and reanalysis the remaining trials. When excluding the trial performed by Kim [8], the impact of PRO on the incidence of pneumonia altered.
Overall certainty of evidence
Evidence quality was assessed as low, due to study limitations (including high risk of bias over studies, clinical heterogeneity) (Table 2). Future evidence is likely to change the estimated effect.
This meta-analysis and systematic review disclosed prophylactic antibiotic administration had no effect on clinically-important outcomes with a low evidence quality, such as the incidence of pneumonia (including early-onset pneumonia), mortality, CPC1-2 and ICU LOS.
The main outcome of this trial was the incidence of pneumonia, and no differences could be demonstrated when applied prophylactic antibiotic. Several trials have revealed that infectious complications were common in survivors of cardiac arrest, as expected, pneumonia was the most common type of infection [19-20]. The risk of pneumonia after OHCA is increased by aspiration of gastric contents, post-resuscitation immune paralysis, high blood endotoxin concentration, intestinal mucosal disruption due to ischemic-reperfusion injury, mechanical ventilation, pulmonary contusions and chest wall dysfunction from rib and sternal fractures [21,22,23]. Emesis was documented in 32% of patients suffering non-traumatic OHCA, implying a high risk of aspiration [24]. In a RCT investigated benefits of prophylactic antibiotics in comatose survivors of OHCA, tracheobronchial aspiration was reported in 23 patients (28%) [16]. Immediate bronchoscopic airway cleaning together with prophylactic antibiotic administration might be efficient if aspiration is detected. Regrettably, there were only 2 trials excluding patients with aspiration [7,16] in this meta-analysis, leading a bias to the incidence of pneumonia.
This ambiguity can be further complicated by TTM, as it was deemed to be an increased risk of early-onset pneumonia [25]. Paradoxically, results from a meta-analysis drawn a conclusion that no significant difference was found in pneumonia events [26]. The discrepancy could be explained when concerning the level and the period of TTM. Studies reported higher risk of pneumonia when TTM was used over a longer period of time (≥48-72 hours) and at a temperature of 33 ℃ when compared with 36 ℃ [27,28]. All the patients included in this meta-analysis were treated with TTM, however, none of the trial detailed the level and the period.
Because of confounding factors, an accurate early diagnosis of pneumonia after OHCA remains a challenge. First of all, the post-cardiac arrest state which is characterized by high levels of circulating cytokines and adhesion molecules, the presence of plasma endotoxin and dysregulated leukocyte production of cytokines, recall the immunological profile found in patients with sepsis [29,30]. In the post-hoc analysis of TTM-trial [31], Josef and co-workers recorded prospective data on infectious complications and considered the diagnostic value of both C-reactive protein and procalcitonin as low during the first three days after cardiac arrest as they were shown to be non-discriminatory for patients with infections and OHCA. The introduction of TTM complicates the diagnosis of infection. During the temperature-controlled period, body temperature changes and the presence of fever cannot be easily employed. All that which has been discussed above confuses the diagnosis of pneumonia and ulteriorly contributes the bias of trials recruited in the meta-analysis.
In post-hoc analysis of the TTM cohort, Harmon and colleagues found that pathogens isolated after OHCA comprised of both gram-negative (61.1%) and gram-positive (38.2%) micro-organisms, with most of the pathogens isolated in patients with pneumonia was Staphylococcus (23%) [14]. Gajic [32] had similar results with their most reported pathogens being Staphylococcus aureus (31%). Taken together, Harmon recommended a cephalosporin of 2nd or 3rd generation as a reasonable empiric approach [14].
Mortensen [33] reported that 95% of in-hospital cardiac arrest (IHCA) and 82% of OHCA patients received antimicrobial treatment in the post-cardiac arrest period. The risk of resistance occurring as a negative consequence of antibiotic prophylaxis use need to be weighed against the benefits. Hoth and colleagues [34] retrospectively reviewed patients contracted nosocomial pneumonia and revealed that when receiving prolonged (>48h) prophylactic antibiotics, the causative organisms were more likely to be resistant or Gram-negative bacteria, and the incidence of antibiotic complications were two times greater than for patients who did not receive extended antibiotic prophylaxis. In line with this, Kroupa [35] disclosed that initial antibiotic therapy was changed in 52.8% of OHCA patients, alarming an increasing frequency of multi-resistant strains of pathogens. Contrarily, a recent RCT [7] comparing intravenous amoxicillin–clavulanate with placebo in patients suffering OHCA identified no increase in resistant bacteria. In trial of Kim [8], Methicillin-resistant Staphylococcus aureus was found in 2 patients in the prophylactic antibiotic group, and Klebsiella pneumoniae was found in 1 patient in the control group.
Couper and co-workers presented a meta-analysis concerning prophylactic antibiotic use in patients suffering cardiac arrest (including in-hospital and out-of-hospital cardiac arrest). Inpatients are likely to have pre-existing infections and/ or already be on antibiotics at the time of their cardiac arrest. The response time from trained personnel would be significantly shorter, as would the time to intubation, thereby potentially reducing the chance of aspiration [18]. Taken together, we only exam the association between prophylactic antibiotic administration and OHCA.
Limitations
This meta-analysis had several weaknesses that should be noted. Firstly, only 2 RCTs were recruited in this meta-analysis, of which one is under high risk. Secondly, the initiation and the treatment course of intervention group varied from trials. The missing data handling remained major limitations for this paper. We look forward to the publication of the full results of Ceftriaxone to PRevent pneumOnia and inflammaTion aftEr Cardiac arresT (PROTECT) trial [36] to clarify the effect furtherly.
Given this conflicting collection of limited quality, no difference in the incidence of pneumonia (including early-onset pneumonia), mortality, CPC1-2 and ICU LOS was found when receiving prophylactic antibiotic administration in patients suffering OHCA.
OHCA: Out-of-hospital cardiac arrest
ICU: Intensive care unit
RCT: Randomized controlled trials
CPC1-2: Cerebral performance category with 1-2
OR: odds ratios
CI: confidence interval
WMD: Weighted Mean Difference
GRADE: Grading of Recommendations Assessment, Development and Evaluate system
TTM: Targeted temperature management
Pro-group: Prophylactic antibiotics administration group
IHCA: In-hospital cardiac arrest
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information files.
Competing interests
The authors declare that they have no competing interests.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Author contributions
WY/ LJD developed the search strategy and performed the literature search. WY/LJD did the study selection and data extraction for the systematic review. WY wrote the first draft of the manuscript. All authors contributed to the interpretation of data and critical revision of the manuscript, and approved the final manuscript. All authors confirm to the accuracy or integrity of the work.
Table1: Characteristic of studies included
author |
year |
design |
participants |
intervention |
clinical parameters |
Harmon |
2020 |
Multicenter observational trial |
PRO-group n = 345; No PRO-group n = 351 |
|
Incidence of pneumonia, bacteremia in PRO-group â |
François |
2019 |
Multicenter RCT |
PRO-group n = 99; No PRO-group n = 95 |
PRO-group: amoxicillin–clavulanate starting less than 6 hours after OHAC at a dose of 1 g and 200 mg tid for 2 days |
Incidence of early-VAP â No differences in the incidence of late-VAP and 28-day mortality, ICU LOS, days with antibiotic use and ventilator |
Gagnon |
2015 |
Multicenter observational trial |
PRO-group n = 416; No PRO-group n=1428 (including antibiotic for treatment n=604) |
|
Incidence of serious infectious, pneumonia and sepsis in PRO-groupâ No differences in CPC1-2 and ICU LOS |
Ribaric |
2016 |
single-center RCT |
PRO-group n = 30; No PRO-group n = 30 |
PRO-group: Amoxicillin–Clavulanic acid 1.2 g q8h |
Incidence of positive mini-BAL on day 3 in PRO-groupâ No differences in days of MV and tracheal intubation, ICU LOS, ICU survival and CPC1-2 |
Tagami |
2016 |
Multicenter observational trial |
PRO-group n = 1272; No PRO-group n = 1531 |
|
Incidence of pneumonia in PRO-groupâ
|
Kim |
2016 |
single-center observational trial |
PRO-group n = 48; No PRO-group n = 20 |
PRO-group: Prophylactic antibiotic therapy within the first 24 hours |
No differences in the incidence of early-onset pneumonia, duration of MV, ICU LOS, in-hospital mortality, CPC3-5 |
Abbreviations: PRO group: Prophylactic antibiotics group; No PRO-group: Delayed/ clinically-driven/ even no administration of antibiotics; RCT: Randomized controlled trial; OHCA: Out-of-hospital cardiac arrest; VAP: Ventilator-associated pneumonia; LOS: Length of stay; CPC1-2: Cerebral performance category with 1-2 indicating good neurological recovery; Mini BAL: Mini bronchoalveolar lavage; MV: Mechanical ventilation
Table 2: Quality of evidence using GRADE approach.
Outcomes |
Studies (participants) |
Quality assessment |
Summary of findings |
|||||||
Risk of bias |
Inconsistency |
Indirectness |
Impression |
Publication bias |
Overall quality of evidence |
WMD or OR (95%CI) |
Heterogeneity |
|||
I2(%) |
P value |
|||||||||
Pneumonia |
5 (5001) |
Serious |
Very serious |
Not serious |
Not serious |
Not serious |
ÅOOO Very low |
0.44 (0.19,1.02) |
95.9 |
0.056 |
Mortality |
4 (3125) |
Very serious |
Very serious |
Not serious |
Not serious |
Not serious |
OOOO Very low |
1.17 (0.46,2.97) |
87.8 |
0.748 |
CPC1-2 |
3 (1083) |
Very serious |
Not serious |
Not serious |
Not serious |
Not serious |
ÅÅOO Low |
1.18 (0.91,1.54) |
0 |
0.212 |
ICU LOS |
2 (254) |
Serious |
Not serious |
Not serious |
Not serious |
Not serious |
ÅÅÅO Moderate |
0.05 (-0.20, 0.29) |
0 |
0.706 |