A Randomized, Double-Blind Study of Placebo vs. Ketamine For Use During Dressing Changes in Critically Ill Burn Patients

Research Methods

1. Outcome Measure(s):

This is a randomized, double-blind study designed to evaluate the effectiveness of ketamine versus placebo in addition to usual care used during dressing changes in 30 thermally injured subjects. The primary outcomes are the opioid and benzodiazepine requirements surrounding dressing changes, pain profiles, hemodynamic changes, and pain-related anxiety scores related to ketamine use versus placebo. Secondary outcomes are exploratory and will evaluate the inflammatory markers, IL-1β and IL-6, and their presence in regards to ketamine versus placebo as well as their associations with ASD, depression, and PTSD.

1. Description of Population to be Enrolled:

The study methodology will follow the CONSORT guidelines. A total of 30 critically ill burn patients with a total body surface area (TBSA) ≥ 5% requiring dressing changes that will be performed in the hydro procedure room will be enrolled over 18-24 months from the 9-bed burn-trauma ICU (B-TICU) at the UCH. Patients will only be enrolled if they can receive dressing changes in the burn treatment room because these patients are alert and oriented, not sedated, and are able to communicate with health care providers and respond to questions regarding their perceptions of pain and anxiety. Subjects will be randomized in a double-blind manner to receive placebo (n=15) or ketamine (n=15) in addition to usual care used before, during, and after dressing changes.

1. Assessment of Outcomes

Primary Hypothesis 1: Adjunctive ketamine safely reduces short-term pain and anxiety surrounding dressing change events in conscious thermally injured patients compared to usual care.

Aim 1a: Comparatively evaluate the opioid and benzodiazepine requirements before, during, and after dressing changes between placebo and ketamine.

When a subject is scheduled for a dressing change in the burn treatment room, the subject's active medication list will be evaluated for analgesic, anxiolytic, and neuropathic medications. Every hour, starting 4 hours prior to the beginning of the dressing change procedure, each drug, dose, and route will be recorded for administered medications. Opioid medications will be converted to fentanyl equivalents and benzodiazepine medications will be converted to midazolam equivalents as fentanyl and midazolam represent the most commonly administered agents. Documentation will continue every hour up to 4 hours after the end of the procedure. This will be repeated for every dressing change.

Aim 1b: Comparatively evaluate patients' perceived pain levels before, during, and after dressing changes between placebo and ketamine.

When a subject is scheduled for a dressing change in the burn treatment room, the subject will be asked to rate their pain score using the Visual Analogue Scale (VAS) [anchored by 0 representing no pain and 10 being the worst pain the patient has ever experienced]. The VAS instrument will be administered by the investigator or designee within 4-12 hours prior to the dressing change procedure. Subjects will then be asked to rate their pain using the VAS within 4-12 hours after the end of the procedure. This will be repeated for every dressing change.

Aim 1c: Comparatively evaluate patients' pain-related anxiety surrounding dressing changes between placebo and ketamine.

When a subject is scheduled for a dressing change in the burn treatment room, the subject will be asked to rate their pain-related anxiety score using the Burn Specific Pain Anxiety Scale (BSPAS). The BSPAS instrument will be administered by the investigator or designee within 4-12 hours prior to the dressing change procedure. Subjects will then be asked to rate their pain-related anxiety score using BSPAS again within 4-12 hours after the end of the procedure. The BSPAS is a 5-item assessment with questions relating to: feeling of worry, emotions surrounding dressing changes, anticipation or anxiety related to pain surrounding procedures, and feelings of being "on edge" due to experiencing pain. Each question is scored on a visual analog line starting at 0 and ends at 100, two anchors of reference for the subject. This tool was designed to detect anticipatory anxiety in burn patients and has been used in studies with the ability to distinguish acute anxiety in burn patients related to dressing changes. This will be repeated for every dressing change.

Aim 1d: Comparatively evaluate the occurrence of hypo- or hypertension, tachycardia, neurologic agitation / disassociation, over-sedation, and the need for open-label ketamine between placebo and ketamine.

Blood pressure and heart rate will be monitored hourly for four hours before, during, and after the end of the procedure. Hypotension will be defined as a systolic blood pressure ≤90 mmHg or a decrease of systolic blood pressure of ≥40 mmHg, hypertension will be defined as a systolic blood pressure ≥180 mmHg or an increase of systolic blood pressure of ≥40 mmHg, and tachycardia is defined as a heart rate ≥120 beats/minute or an increase of ≥20 beats/minutes. The administration of fluid boluses or vasopressor therapy will also be recorded. It is anticipated that these therapies will not be required. The Richmond Agitation Sedation Scale (RASS) score will be used to determine neurologic disassociation (RASS score of +2 to +4 indicating agitation or combative) and over-sedation (RASS score of -3 to -5 indicating moderate to deep sedation). The RASS score is performed as routine care every hour and it will be recorded for four hours before and after the end of the procedure. Open label ketamine may be administered by the burn ICU physician if patients have received both doses of study drug and subjects report pain scores of ≥7 (out of 10) despite the administration of ≥400 mcg fentanyl equivalents (this represents the 75th percentile of fentanyl required in the medication usage evaluations conducted at UCH) or a RASS score of -4 to -5, indicating levels of sedation deep enough to warrant caution against administering additional doses of opioids or anxiolytics. The use of open label ketamine will be recorded and comparatively assessed.

Secondary Hypothesis 2: Adjunctive ketamine administration alters cytokine levels, specifically interleukins IL-1β and IL-6, during dressing changes compared to usual care.

Aim 2a: Comparatively evaluate the magnitude of change in serum concentrations of IL-1β and IL-6 levels during dressing changes between placebo and ketamine.

When a subject is scheduled for a dressing change in the burn treatment room, venous blood samples of 10 mL will be obtained from an indwelling catheter 4-12 hours before and after the procedure to evaluate the level of IL-1β and IL-6 in the serum. Blood samples will be centrifuged for 15 minutes at 3000rpm with plasma immediately separated and stored at -80°F. Enzyme-linked immunosorbent assays (ELISA) will be used to determine IL-1β and IL-6 concentrations after all subjects have completed the study.

Secondary Hypothesis 3: Ketamine when used as an adjunct during burn dressing changes reduces the incidence of long-term psychometric outcomes when compared to usual care.

Aim 3a: Comparatively evaluate the incidence of ASD between placebo and ketamine after dressing changes.

Within seven days of the final dressing change procedure, an investigator will facilitate the completion of the Stanford Acute Stress Reaction Scale (SASRS). The SASRS instrument will be administered by the investigator or designee. The SASRS consists of 30 items. Ten questions are related to dissociation, six to re-experiencing of trauma, six to anxiety and increased arousal, six to avoidance of the trauma, and two to impairment in functioning. The questions can be modified to pertain to a specific event (i.e. burn dressing changes). Each question is scored from 0 to 5. Also, the instrument contains an additional three questions specifically associated to ASD. They involve an event description, how traumatic the event was, and for how long the subject had his or her most severe symptoms in relation to the event. In order to have a positive score for ASD, the subject must score a 3 or higher for at least three of the dissociative symptoms and at least one symptom of each of the following: re-experiencing, increased anxiety and arousal, avoidance, and impairment of functioning. This scale has demonstrated reliability and validity and has been used in burn-related studies to assess for ASD with positive results.

Aim 3b: Comparatively evaluate the incidence of anxiety and depression between placebo and ketamine prior to hospital discharge.

Within 3-7 days prior to the projected date of discharge, the investigator or designee will administer the Hospital Anxiety Depression Scale (HADS). The HADS instrument will be administered by the investigator or designee only. The HADS consists of 14 questions, seven for anxiety and seven for depression. Each item is scored from 0 to 3, with a cut-off cumulative score of 11 for both subscales indicative of anxiety or depression. This scoring tool has been used for 30 years, possesses excellent reliability and validity, and avoids reliance conditions that are also common somatic symptoms of illness such fatigue, insomnia, and hypersomnia.

Aim 3c: Comparatively evaluate the incidence of PTSD between placebo and ketamine prior to hospital discharge.

Within 3-7 days prior to the projected date of discharge, the investigator or designee will administer the Impact of Events Scale-Revised (IES-R). The IES-R instrument will be administered by the investigator or designee only. The IES-R evaluates subjective distress caused by traumatic events and will be used to assess manifestations of PTSD. It is not diagnostic but possesses excellent reliability and validity for manifestations of PTSD. The IES-R has three subscales (eight items on intrusion, eight items on avoidance, and six items on hyperarousal). Each item is scored on a four point scale: 0 = "not at all," 1 = "a little bit," 2 = "moderately often," 3 = "quite a bit," and 4 = "extremely often." The total score of each subscale may be averaged and a cumulative score of 30 is indicative of the presence of PTSD.

Aim 3d: Assess possible associations between the cytokine levels and ASD, anxiety, depression, and PTSD.

Statistical associations between IL-1β and IL-6 values with SASRS, HADS, and IES-R will be investigated.

1. Data Analysis Plan:

A total of 30 subjects will be evaluated (15 in each study group) based on a power of 0.8 and a significance level of 0.05. Data obtained from medication usage evaluations at the UCH were used to estimate the mean and standard deviations for the change in cumulative fentanyl equivalent doses for the four hour period before vs. after dressing changes. Using standard deviations of 130mcg, 12 subjects are required in each study group to show a desired difference of 150mcg between groups when comparing the magnitude of change in the fentanyl dose before vs. after the procedure. While all dressing changes will be evaluated, it was assumed for this analysis that only one dressing change per subject will occur. An additional three subjects per study group will be enrolled to ensure a total of 24 subjects assuming an upper limit of 20% for a possible drop-out rate. Thirty subjects provides power of 0.8 at a significance level of 0.05 to detect a 50% decline in patients reporting moderate pain (by VAS or BSPAS) or anxiety (by SASRS) assuming the baseline rates exceed 80%.

Mean, median, standard deviation, and interquartile ranges will be determined for all continuous data. The primary analyses will include all randomized subjects based on intent to treat regardless of whether they completed the study. Additional analyses will be conducted based on subjects completing the protocol (i.e. those withdrawn from the study prior to completing the protocol or receiving open-label ketamine will be excluded) and those completing treatment. Statistical analysis of continuous data between groups will use the student T test or the Mann-Whitney U for parametric and nonparametric data, respectively. Statistical analyses of continuous data within a study group will use the paired student T test or the Wilcoxan Rank test for parametric and nonparametric data, respectively. Statistical analysis of proportions will use the Chi-square test or Fisher's Exact test for parametric and nonparametric data, respectively, with appropriate corrections for small sample size. Pearson product or Spearman Rank Order will be used for correlation analyses of IL-1β or IL-6 with the results of the psychometric scoring tools. The study is designed to investigate the primary hypothesis and the related specific aims. All secondary hypotheses are exploratory to investigate consistency of data and analyses will be interpreted cautiously.