In a 39 patient, single-centre pilot trial of phenobarbital as an adjunctive treatment for severe acute alcohol withdrawal syndrome (AAWS), the PHENOMANAL trial will demonstrate the ability to recruit, consent, and randomize participants and adhere to the assigned treatment protocol with minimal loss to follow up and infrequent adverse events. This pilot trial will provide data and proof of feasibility in preparation to conduct a multicentre randomized controlled trial of phenobarbital as an adjunctive treatment for severe AAWS. Patients will be recruited from the Emergency Department, Intensive Care Unit, Medical Step-Up unit, and wards of Unity Health Toronto - St. Michael's Hospital, a quaternary care centre located in downtown Toronto, Canada.
Recruitment Patients will be identified by research personnel during daytime hours from Monday to Friday using a multi-modal approach. With the assistance of the hospital informatics department, personnel will collect a daily census of patients who have been started on symptom-triggered benzodiazepine therapy using the Clinical Institute Withdrawal Assessment (revised) (CIWA-Ar) protocol.
Consent The investigators will use a hybrid consent model. Whenever possible, consent will be obtained from the patient or his/her substitute-decision-maker (SDM) at the time of recruitment. In this instance, Research Coordinators will ask one of the members within the 'circle of care' (e.g., physician, nurse, respiratory technologist, social worker, etc) to ask for permission from patient or their SDM for a research coordinator to contact him/her about participation in research.
Randomization To conceal allocation, central randomization will be conducted by the Pharmacy Department of St. Michael's Hospital using a list of random treatment assignments created on www.randomize.net.
Intervention As soon as possible after randomization, patients will receive either a single intravenous (IV) dose of phenobarbital (7.5 mg per kg of ideal body weight, prepared in 250 ml D5W; n=26) or placebo (250 ml D5W; n=13).
Safety During the infusion patients will be monitored with continuous oximetry, non-invasive blood pressure measurement q 30 minutes (+/-, and continuous cardiac monitoring in a high-acuity unit (e.g Intensive Care, Cardiac Intensive Care, Emergency, Step-Up). Phenobarbital achieves peak serum concentration within one hour of administration; however, study participants will be monitored with continuous pulse oximetry and q 30 minute non-invasive blood pressure measurement for a minimum of 3 hours post administration. Each patient will thus receive a total of four hours of intensive monitoring (i.e. one hour of monitoring during the study drug administration, and three hours of monitoring subsequently to ensure clinical stability). Eligible female trial participants \< 55 years of age will have a urine or blood ßhCG checked prior to treatment administration.
Data Collection Research personnel will collect data on (i) Patient Demographics including age; sex; housing; relevant comorbidities; time since last alcohol consumption; baseline laboratory investigations; and baseline CIWA-Ar scores. (ii) Treatment including maximal CIWA-Ar score; number, dose and type of benzodiazepine or alternative treatments administrated (e.g., benzodiazepines, propofol, clonidine, dexmedetomidine, haloperidol); vital signs; need for admission to a monitored setting (i.e. vitals recorded more often than every 4 hours, or need for continuous oximetry); use of non-invasive or invasive ventilation; and the presence of a bedside sitter. These data will be recorded for a span of 7 days or until hospital discharge (whichever comes first). (iii) Clinical Outcomes including adverse events (aspiration, intubation, seizures); monitored care and hospital length of stay; time to Addictions Medicine assessment; and vital status at hospital discharge.
Statistical Power Fifty patients will be recruited in the PHENOMANAL pilot trial. This sample size will ensure that clinicians gain experience with the treatment protocol, enable assessment of recruitment and consent practices, and provide preliminary estimates of treatment effect (safety and efficacy) to inform the design of a larger trial.
Statistical Analysis In the primary analysis, recruitment will be considered feasible if the investigators can recruit 2 or more patients per month, on average, over the 24-month study period. In secondary analyses, compliance rates greater than 80% will be considered acceptable in both study arms. Similarly, a cross-over rate of \< 10% from the control arm to the phenobarbital arm will be taken to be acceptable. Preliminary estimates of treatment effect will be compared between the alternative treatment strategies using the Chi-Squared test (alternatively, Fisher's Exact test) and Student's T-test (Alternatively, Wilcoxon Rank-Sum test) for binary and continuous data, respectively.
Discussion This pilot randomized controlled trial will assess feasibility of a multicenter trial to investigate intravenous phenobarbital as an adjunctive treatment for severe AAWS. Randomization of 39 patients will allow us to assess feasibility metrics, including clinicians' ability to comply with the protocol and cross-over rates, and generate preliminary estimates of safety and treatment effect.
This trial protocol has been designed in light of the unique challenges of studying patients with severe AAWS. It is important to clearly define the population of interest. Patients with mild alcohol withdrawal may not benefit from IV adjunctive treatment with phenobarbital, and may lead to over-sedation, complications (e.g., intubation, aspiration), and/or prolonged hospitalization. Oral phenobarbital administration may warrant further investigation in this patient population. In contrast, highly agitated patients suffering from AAWS may not show a clinically important response to the dose of IV phenobarbital that the investigators are evaluating. The investigators decision to limit enrollment to patients with a severely elevated CIWA score (\>16) after receiving at least 60 mg of diazepam or equivalent represents an attempt identify the patient population that is most likely to benefit from treatment with adjunctive IV phenobarbital. Base on pharmacokinetics, phenobarbital is most likely to be useful if administered early in the course of treatment and thus enrollment will be accordingly limited to the first 16 hours after patients are identified.
