Dexmedetomidine During Noninvasive Ventilation
October 1, 2014
ABSTRACT & COMMENTARY
Dexmedetomidine During Noninvasive Ventilation
By Eric C. Walter, MD, MSc
Pulmonary and Critical Care Medicine, Northwest Permanente and Kaiser Sunnyside Medical Center, Portland
Dr. Walter reports no financial relationships relevant to this field of study.
SYNOPSIS: In this randomized, controlled trial, continuous infusion dexmedetomidine started early during noninvasive ventilation did not improve tolerance or success of noninvasive ventilation and may have caused harm.
SOURCE: Devlin JW, et al. Efficacy and safety of early dexmedetomidine during noninvasive ventilation for patients with acute respiratory failure. Chest 2014;145:1204-1212.
Noninvasive ventilation (NIV) is an increasingly viable treatment option for patients with acute respiratory failure. However, intolerance of NIV due to mask discomfort, anxiety, or claustrophobia is common. Sedatives can be used to improve tolerance but run the risk of decreasing respiratory drive and causing delirium. Dexmedetomidine is an alternative sedative that is theoretically advantageous as it produces sedation without respiratory depression. Devlin and colleagues tested this theory in this blinded, randomized, controlled trial comparing early implementation of dexmedetomidine vs placebo to improve NIV tolerance.
Adult patients requiring NIV for acute respiratory failure were eligible. Patients had to have been using NIV for ≤ 8 hours prior to randomization. Notable exclusion criteria were age ≥ 85 years, systolic blood pressure ≤ 90 mmHg, heart rate ≤ 50 beats/min, and acute decompensated heart failure with an ejection fraction ≤ 25%. The primary study outcome was NIV intolerance as measured using a 4-point scale: 1) tolerating NIV (patient comfortable), 2) mild intolerance (occasionally grabbing at mask, 3) moderate intolerance (discomfort most of the time and frequently grabbing at mask), and 4) severe intolerance (inability to leave mask in place). NIV failure was defined as need for intubation or death on NIV. Also monitored were the patient’s degree of sedation, pain, and delirium.
Despite a study period from 2008-2012, only 61 patients were evaluated for participation. After exclusions, 36 patients were randomized to either a continuous infusion of dexmedetomidine or placebo. The infusions were started early, approximately 4 hours after time from NIV initiation, with no significant difference between groups. Infusions were titrated up every 30 minutes until the desired level of sedation was obtained. Every 3 hours a single dose of IV midazolam (0.5-1.0 mg) and a single dose of IV fentanyl (25-50 µg) could be given as needed for agitation or pain. Haloperidol (0.5-1.0 mg IV) could be used every 6 hours as needed for delirium. Study infusions were stopped if the patient remained stable off NIV for ≥ 2 hours, after 72 hours of continuous administration, or at time of intubation. During data analysis, it was noted that among all patients at all time points, < 5% of intolerance scores were > 2. Therefore, the analysis was dichotomized as either NIV tolerant (intolerance score = 1) or NIV intolerant (intolerance score = 2, 3, or 4).
Dexmedetomidine was not associated with increased tolerance of NIV (odds ratio 1.44; 95% confidence interval, 0.44-4.7; P = 0.54) nor a greater percentage of time tolerating NIV (P = 0.56). The intubation rate (31% vs 29%) and duration of mechanical ventilation did not differ between the two groups. No patients died while on NIV. Dexmedetomidine use was not associated with better pain control or sedation level and the use of midazolam, fentanyl, and haloperidol did not differ between the groups. Despite this, more patients on dexmedetomidine had episodes of deep sedation (25% vs 0%; P = 0.04) and spent more time on NIV (median 37 hours vs 12 hours; P = 0.03) than patients on placebo.
COMMENTARY
There are many reasons why patients may not tolerate NIV, including mask fit, a sense of claustrophobia, and anxiety. Cautiously used sedatives may be able to overcome these intolerances. Worsening respiratory distress is another reason for NIV intolerance, a situation where sedative use is potentially unsafe. Therefore, the authors should be commended for embarking on this randomized, controlled trial to help answer these questions. That it took 4 years to recruit 36 patients suggests there were many challenges, but using observational methods to answer these questions would be subject to multiple biases. It is unclear if the slow recruitment affected the results but it does make one wonder about patient selection.
The early initiation of a continuous infusion of dexmedetomidine did not improve tolerance or success of NIV. The small sample size may have limited the ability to detect a significant difference, but the bulk of the data suggest no benefit. The small sample size also makes it challenging to make conclusions about secondary outcomes. However, there is a suggestion that dexmedetomidine use may have caused harm. Patients on dexmedetomidine infusion were more deeply sedated (despite no difference in the need for anxiolytics or narcotics for "breakthrough" anxiety or pain) and required NIV three times as long as those on placebo. Most patients were tolerating NIV at study entry (62% on dexmedetomidine and 71% on placebo). Among patients not tolerating NIV at baseline, dexmedetomidine increased the time spent tolerant of NIV but the difference was not significant (P = 0.46) and extremely few patients were in each group (n = 6 on dexmedetomidine and n = 5 on placebo). The authors’ conclusion that dexmedetomidine may be beneficial in these patients is not supported by data from this study.
These data do not support the routine use of dexmedetomidine in patients on NIV. Despite the negative result, this study provides useful information to help guide critical care practitioners’ decisions about the use of sedation and NIV. This highlights the importance of publishing negative trials.