Step-by-Step Methods:
Population: The study will be conducted at University Hospital, the 716-bed county hospital for Bexar County and serves a varied population, including patients from medically underserved areas. The patient population will come from the medical intensive care unit and the cardiac intensive care unit, a 20 and 26-bed unit, respectively, at University Hospital that is the clinical practice and teaching site for one study investigator.
Screening: Consecutive patients admitted to the medical or cardiac intensive care unit at University Hospital with new onset atrial fibrillation in the setting of severe sepsis or septic shock will be screened for study inclusion. Participants will be 18 years or older. Patients who develop new-onset atrial fibrillation in the setting of severe sepsis or septic shock will be considered for study inclusion. Exclusion criteria apply to those younger than 18 years of age, those with a history of atrial flutter, baseline QTc (a measurement made on a electrocardiogram which measures the start of the Q wave to the end of the T wave) \> 500 msec, atrial fibrillation during the index admission or in their past medical history, second or third degree atrioventricular (AV) block, those recovering from cardiac surgery done during the same hospital admission, those who have received amiodarone or other Vaughan Williams class I or III antiarrhythmic drugs in the last 6 months, untreated/poorly controlled hypothyroidism, hyperthyroidism, acute or chronic hepatic failure, other requirements of antiarrhythmic drug therapy, recent cardiac surgery in the preceding 30 days, and those who are pregnant.
Intervention: Patients aged 18 years or older who develop new-onset atrial fibrillation in the setting of severe sepsis or septic shock and in whom the medical team deems appropriate to initiate amiodarone therapy in will be considered for study inclusion. All patients will receive a 150 mg IV bolus dose of amiodarone, followed immediately by a continuous infusion of 1 mg/min for the first six hours, with a recommended reduction to 0.5 mg/min subsequently. Conversion from IV to PO amiodarone will occur based on patient hemodynamic stability and physician/pharmacist discretion. Patients will be randomized to receive amiodarone at a full loading dose (≥ 5g IV or ≥10g PO +/- 20%) or a partial loading dose (\<4g IV or \< 8g PO). The assigned total loading dose will include all IV and PO amiodarone administered within 7 days from initiation of amiodarone. All doses will be compared in total PO amount (accounting for 50% bioavailability of PO versus IV amiodarone). The primary attending or fellow physician may over-ride randomization if the patient's amiodarone dose if the patient's severity of illness warrants a duration that differs from that assigned during randomization. In the full loading dose group, discontinuation of amiodarone will be at the discretion of the physician/pharmacists, after the pre-determined randomization loading dose has been provided.
Management Components: Facilities to be used throughout the project include the University Hospital and The University of Texas Health Science Center at San Antonio. Equipment use for direct patient care will be in keeping with standards-of-care, without any additional equipment being required for the clinical research study.
Data Collection: Data collection will include age, race, gender, body mass index (BMI), data for calculating an Acute Physiology and Chronic Health Evaluation (APACHE II) and Sequential Organ Failure Assessment (SOFA) scores, suspected source of infection, identified pathogens, pre-existing conditions that may predispose the patient to AF and the severity/classification of such illness (coronary artery disease, heart failure, chronic kidney disease, chronic obstructive pulmonary disease, diabetes mellitus); vital signs at the onset of AF (mean arterial pressure, systolic blood pressure, heart rate), lactate, pH (scale of acidity), and serum bicarbonate at the time of AF onset; fluid resuscitation volume for the 24 hours prior to AF, vasopressor use (including drug, dose, duration) in the preceding 24 hours; other medications used to stabilize hemodynamic status in AF (e.g. β-blockers, calcium channel blockers, digoxin), other attempted methods of cardioversion attempted prior to amiodarone use (e.g. direct current cardioversion, other antiarrhythmic drug therapy); amiodarone loading dose, cumulative study amiodarone dose; the duration of time spent in AF before conversion to normal sinus rhythm (NSR), the incidence of AF recurrence; and hemodynamic variables to be assessed at various time points after the initiation of amiodarone (-6 hours, at onset of AF, 2 hours, 6 hours, 12 hours, and every 6 hours thereafter until day 7 or death) including mean arterial pressure, central venous pressure, systolic blood pressure, heart rate, pH, serum bicarbonate, and serum lactate.
Monitoring for Efficacy and Safety:
Efficacy: Patients will be stratified based on response to therapy into one of three groups: conversion to NSR, non-converting AF that is hemodynamically stable or non-converting AF that is hemodynamically unstable. The primary efficacy outcome is the mean heart rate every 6 hours after initiation of the amiodarone infusion to day 7 or death. Secondary efficacy endpoints will include the percentage of time spent hemodynamically unstable in the first 7 days following amiodarone initiation, ICU LOS, hospital LOS and 28-day mortality along with comparisons of the individual portions of the hemodynamic endpoints \[mean arterial pressure (MAP), heart rate (HR), systolic blood pressure(SBP)\], cumulative vasopressor doses of norepinephrine and vasopressin, central venous pressure (CPR), conversion to NSR, maintenance of NSR, proportion of time spent in NSR after the start of the infusion to day 7 or death, pH, standard bicarbonate, serum lactate, and central venous oxygen saturation.
Safety: Safety outcomes include worsening of hypotension (largely related to the rate of infusion and the polysorbate 80 diluent), bradycardia, acute elevations of liver function tests ≥ 3 times the upper limit of normal, phlebitis, QTc prolongation, skin reactions, neuropathy, and pulmonary toxicity. We will also collect premature study discontinuation due to any adverse event. Because the experimental group of this study is a partial load of amiodarone, rather than a full load, it is likely that the experimental group will experience fewer amiodarone related adverse effects than those receiving the current practice. In the setting of worsening hemodynamic stability during amiodarone infusion (i.e. SBP \<90 mmHg OR MAP \< 70 mmHg, despite HR control of \< 120 bpm; need for fluid boluses ± vasopressors or dobutamine), which the medical team believes to be caused by IV amiodarone therapy:
1. the amiodarone infusion rate should be decreased by 50%, with efforts to increase back to the standard infusion rate after the patient is hemodynamically stable, or
2. switch from IV amiodarone to PO amiodarone therapy at a dose of 400 mg PO three times daily (TID), or
3. emergent electro-cardioversion at the discretion of the medical team (as is the standard of care)
Data Analysis Plan: Data will be collected and stored using Microsoft Excel or Access. After data organization and cleaning, data will be imported into JMP 12.0 or SPSS. Data will be analyzed using both descriptive and comparative statistics. The primary outcome will be analyzed using repeated measures ANOVA. Nominal and ordinal variables will be compared using Chi-square test or Fisher's Exact test, as appropriate. Continuous variables will be tested for normality using the Shapiro-Wilk W goodness of fit test. Continuous variables with normal distributions will be compared using the student's t-test, while non-parametric variables will be compared using the Wilcoxon rank sum test. Statistical significance will be defined as an alpha level less than 0.05.
All baseline demographics, including patient comorbidities, amount of fluid resuscitation, previous vasoactive therapies, and use of other interventions such as direct current cardioversion, will be entered into a multivariate regression model if the p value is less than 0.20 on univariate comparison between treatment groups. The multivariate regression model will be used to assess if baseline factors or previous treatment other than the intervention affected the study outcomes.
Based on extrapolation from available literature assessing HR control at 24 hours, we estimate that the full loading dose group will have a mean HR of 110 bpm over the course of 7 days, while the partial loading dose group will have a mean HR of 130 bpm over the course of 7 days. 7 day mean HR that is 20 bpm lower than the partial loading dose group. Therefore, we estimate that 250 patients will be necessary in each group to have an 85% power to detect a difference, assuming a 15% drop-out rate.
