Introduction

Alcohol is the second most commonly abused drug after nicotine.1 Approximately 20% of patients who access medical care have an alcohol use disorder.2 Alcohol withdrawal syndrome (AWS) occurs when, after abrupt alcohol discontinuation, the patient becomes symptomatic. Minor AWS symptoms include insomnia, anxiety, nausea, headache and palpitations. If AWS is effectively prevented, these symptoms will resolve within one to two days after the alcohol discontinuation. Major AWS symptoms include seizures and delirium tremens (DTs). The seizures are usually singular and resolve spontaneously. Delirium tremens is the most dangerous complication of AWS, and it can result from under-treatment or lack of treatment of the preceding AWS symptoms. The overall mortality of DTs is reported to be 8%.3 The incidence of DTs in patients in the intensive care unit (ICU) is approximately 10%.4 Delirium tremens is an independent predictor of length of ICU and hospital stay with an increased duration of invasive ventilation. The related ICU costs are mainly due to room occupancy and mechanical ventilation, yet recently another driver of ICU cost was identified to be mortality before ICU discharge.5

The AWS symptoms occur when blood alcohol concentrations decline rapidly, within the first four to 12 hours after alcohol consumption has been stopped or reduced. The symptoms reach peak intensity on the second day and improve by the fourth or fifth day of alcohol suspension.6 The onset of DTs is typically three to five days after the last alcohol consumption.7 Symptoms include severe agitation, tremor, disorientation, hallucinations and exacerbation of autonomic symptoms. The symptoms may persist up to one week, and DTs duration is associated with an increased risk of mortality and cognitive dysfunction. According to evidence, early diagnosis and effective management of DTs results in reduced mortality.2

The goal of treatment for an agitated patient with AWS is sedation. Benzodiazepines (BZDs) have been the cornerstone of therapy for prevention and management of AWS and its complications. Unfortunately, some patients exhibit refractory AWS when given high doses of BZDs. Moreover, the use of BZDs is burdened by excessive sedation, ataxia, confusion, memory impairment and respiratory depression.8 Options for management of refractory AWS include the addition to BZD therapy of phenobarbital, propofol or, more recently, dexmedetomidine. Dexmedetomidine is an alpha-2 agonist that decreases the release of norepinephrine. It is eight times more potent that clonidine with a rapid onset of action (15 minutes) and short half-life (two hours). The agonist action on alpha-2 receptors is responsible for anxiolytic and sedative effects, decreasing the sympathetic tone.9 The possible advantage of dexmedetomidine compared to BZDs is that it does not cause respiratory depression, so the risk of intubation and hospitalization in the ICU should be reduced. The most common adverse drug events (ADEs) of dexmedetomidine are hypotension and bradycardia.10

This review will analyze the effectiveness of dexmedetomidine in comparison to BZDs, or BZDs adjunctive therapy (benzodiazepines plus propofol), as propofol may be useful in patients who do not clinically respond to first-line therapy with BZDs.11

Our systematic review differs from relatively recent ones. Firstly, the systematic review conducted by Woods et al.12 analyzed the effectiveness of dexmedetomidine only in comparison to benzodiazepines in ICU patients. In this proposed systematic review, the population will not be restricted to ICU patients. Furthermore, the outcome was measured by Woods et al.12 using the Clinical Institute Withdrawal Assessment Score - Revised (CIWA), the Ramsey scale, the Richmond Agitation-Sedation Scale (RASS) and the Confusion Assessment Method for the ICU (CAM-ICU). Instead, this review has pragmatic outcomes (respiratory depression requiring endotracheal intubation and ICU admission) rather than a score (CIWA, RASS, CAM-ICU), which does not provide concrete data on the effectiveness of the therapy related to augmented costs for ICU length of stay. Additionally, Woods et al.12 did not analyze the treatment safety of the most common dexmedetomidine-related ADEs. These ADEs are harmful and unintended consequences of medications; therefore, the reporting of ADEs is essential for monitoring and providing research on drug safety, although it has not been internationally standardized.13

In this systematic review, we will analyze dexmedetomidine safety and ADEs as secondary outcomes reporting the onset of the most common dexmedetomidine ADEs (bradycardia and hypotension).10 Other systematic reviews that assessed the effectiveness of dexmedetomidine in the treatment of AWS were conducted by consulting a single database (MEDLINE).10,11,14 Furthermore, some systematic reviews evaluating the role of dexmedetomidine in AWS management concluded that further studies are needed to develop evidence-based recommendations.15,16 Therefore, a systematic review, including several databases that also identify the most recent studies, is necessary to establish the role of dexmedetomidine in the treatment of AWS. The Cochrane Database of Systematic Reviews, the JBI Database of Systematic Reviews and Implementation Reports and PubMed were searched to identify previously published reviews on a similar topic.

Review objective

The purpose of this review is to evaluate the effectiveness and safety of dexmedetomidine as adjunctive treatment to the standard of care (BZD) compared to either the standard of care or other adjunctive treatment approaches (BZD plus propofol) used in the treatment of AWS.

Inclusion criteria

Participants

This review will consider studies including patients who meet the following criteria: 18 years or older with AWS diagnosed using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) or Diagnostic and Statistical Manual of Mental Disorders (DSM-4) criteria.17,18

The exclusion criteria will include a history of use of other psychoactive substances or withdrawal states and/or severe neurologic disorder (e.g. traumatic brain injury, acute stroke, severe dementia, seizure disorder).

Intervention

This review will evaluate the effectiveness of dexmedetomidine in the treatment of AWS. The range of dexmedetomidine dose will be 0.2-1.4 mcg/kg/h as registered in the medication package insert.

Comparator

The comparator will be management of AWS with BZDs alone, or association between BZDs and propofol.

Outcomes

The primary outcome will be the effectiveness of dexmedetomidine on the occurrence of respiratory depression (requiring endotracheal intubation) and/or ICU admission, compared to the other treatments.

The secondary outcome will be the occurrence of ADEs: hypotension (systolic blood pressure < 90 mmHg) and/or bradycardia (heart rate < 50 bpm), compared to the other treatments.

Types of studies

This review will include both randomized controlled trials and observational studies (including cohort and case-control studies) to assess the drug effectiveness and safety.

Methods

The proposed systematic review will be conducted in accordance with JBI methodology for systematic reviews of effectiveness evidence.19 The review protocol was registered on PROSPERO (International Prospective Register of Systematic Reviews, Registration Number CRD42018084370).20

Search strategy

The search strategy will consist of three steps, with the aim to find both published and unpublished studies.

Firstly, we will perform a search on MEDLINE and CINAHL, examining the text words contained in the title and abstract, and the index terms used to describe the articles. A search strategy, which will be adapted for each information source, will be developed. A proposed search strategy for PubMed is detailed in Appendix I. A second search will be conducted using all recognized keywords and index terms through all included databases. Searching for additional studies from gray literature from government departments, international agencies and academics institution repositories or websites will be conducted using similar keywords from the search strings. Thirdly, the reference list of all identified reports and articles will be searched for additional eligible studies. Studies published in English will be considered for inclusion in this review. We will not set a year limit to avoid the exclusion of pre-registration studies. If pre-registration studies are found, a sub-analysis will be conducted between pre-registration versus post-registration studies.

The searches will be completed a second time before the final analyses and additional studies retrieved for inclusion.

Information sources

The databases utilized will include PubMed, EMBASE and Cochrane Central Register of Controlled Trials (CENTRAL). Also, the trial registers to be searched will include: World Health Organization International Clinical Trials Registry Platform (ICTRP), U.S. National Library of Medicine and ClinicalTrials.gov.

The search for unpublished studies will include Conference Proceedings Citation Index via Web of Science, ProQuest Dissertations and Theses, Google Scholar, Networked Digital Library of Theses and Dissertations, OpenDOAR and OpenGrey. The bibliography of all identified reports and articles will be searched for additional studies; this will include the use of backward and forward citation tracking.

Study selection

Following the search, all identified citations will be loaded into EndNote VX8.0 (Clarivate Analytics, PA, USA) and duplicates removed. Two authors (GT and AA) will independently evaluate the titles and abstracts of potentially eligible studies. Data will be extracted independently according to the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) guideline, and the results will be cross-checked.21 Any disagreements on study eligibility or data extraction will be resolved with a third reviewer (MF).

Potentially relevant studies will be retrieved in full and their citation details will be imported into the JBI System for the Unified Management, Assessment and Review of Information (JBI SUMARI; Joanna Briggs Institute, Adelaide, Australia).19

The full text of selected citations will be assessed in detail against the inclusion criteria by two independent reviewers (GT and AA). Reasons for exclusion of full-text studies that do not meet the inclusion criteria will be recorded and reported in the systematic review. Any disagreements that arise between the reviewers at each stage of the study selection process will be solved through discussion or with a third reviewer (MF). The results of the search will be reported in full in the final report and the data will be presented in a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.22

Assessment of methodological quality

Before inclusion in the review, the quantitative papers will be assessed by two independent reviewers (GT and AA) for methodological validity using the standardized critical appraisal instruments from JBI.19 The authors of papers will be contacted to request missing or additional data, where clarification is required. Disagreements between the two reviewers will be resolved through discussion or with a third reviewer (MF). The results collected by the critical appraisal will be reported in narrative form and a table. Following critical analysis, studies that do not meet a certain quality threshold will be excluded. The decision to exclude will be based on cut-off scores of less than 70% of the items assessed for all JBI critical appraisal tools. This represents the following checklist amount of "yes" answers: fewer than six out of nine for quasi-experimental studies; fewer than nine out of 13 for randomized controlled trials; fewer than eight out of 11 for cohort studies; fewer than seven out of 10 for case-control studies and fewer than five out of eight for analytical cross-sectional studies.19

Data extraction

The collected data will be extracted independently by two reviewers (GTand AA), and the results will be cross-checked.

Quantitative data will be extracted from manuscripts included in the review using the standardized data extraction tool from JBI. The data extracted will include the authors, year of study enrollment, country of origin, outcome analyzed in the individual studies, details about the populations, methods and interventions.

Data synthesis

Data extracted from the included studies will, where possible (e.g. studies using uniform case definitions, the same measures of outcome, context and approaches), be pooled in a statistical meta-analysis using STATA v14 (StataCorp LLC, Texas, USA) or any other relevant software.

Risk differences will be used as the meta-analytic measure of association between the addition of dexmedetomidine and the requiring of tracheal intubation. For each study, the proportion of patients requiring tracheal intubation will be used to calculate risk difference and corresponding 95% confidence interval using a 2x2 table. The proportion of patients requiring intubation also will be calculated for each study. Heterogeneity between studies will be assessed by using Q statistic and I², which is the proportion of total variance observed between the studies attributed to the differences between studies rather than to sampling error. I² values of 25%, 50%, and 75% correspond to cut-off points for low, moderate, and high degrees of heterogeneity. A P-value of Q statistic less than 0.10 will be considered significant. If overall heterogeneity is significant, a random-effect model will be used; otherwise, a fixed-effect model will be used.

Subgroup analyses will be conducted, if appropriate, based on these items: i) pre-registration studies versus post-registration; ii) standard of care therapy (BZD) versus adjunctive standard of care therapy (BZD plus other sedative drugs). Furthermore, sensitivity analysis will be performed, in case of strong heterogeneity, excluding single studies from the final evaluation. A funnel plot will be generated to evaluate publication bias; if 10 or more studies are found, a meta-analysis will be conducted. Statistical tests for funnel plot asymmetry (Egger test, Begg test, Harbord test) will be performed, where appropriate. If statistical pooling is not possible, the results will be presented in a narrative form including tables and figures to aid in data presentation. The tables will report the authors, the year of study enrollment, the country of origin and the outcome analyzed in the individual studies.

Assessing certainty in the findings

The Grading of Recommendations, Assessment, Development and Evaluation (GRADE) approach will be used for grading the certainty of evidence, and a Summary of Findings (SoF) will be created using GRADEPro software (McMaster University, ON, Canada).23 The SoF will present the following information where appropriate: the absolute risks for the treatment and control, the estimates of relative risk, a ranking of the quality of the evidence based on the risk of bias, the directness, the heterogeneity, the precision and the risk of publication bias of the review results. All primary outcomes will be included in the SoF.

Acknowledgments

The authors would like to thank Dr. Anna Battimelli and Dr. Alessia Gatani from the University of Campania Luigi Vanvitelli for providing excellent bibliographic service and assistance.

Appendix I: Search strategy for PubMed

Search conducted on 11 November 2018

References