Secondary Logo

Journal Logo

Critical Care, Trauma, and Resuscitation: Research Report

Risk Factors for Dexmedetomidine-Associated Hemodynamic Instability in Noncardiac Intensive Care Unit Patients

Ice, Calvin J. PharmD*; Personett, Heather A. PharmD; Frazee, Erin N. PharmD; Dierkhising, Ross A. MS; Kashyap, Rahul MBBS§; Oeckler, Richard A. MD, PhD§‖

Author Information
doi: 10.1213/ANE.0000000000001125
  • Free


In the intensive care unit (ICU), approximately two thirds of adult critically ill patients require sedation during mechanical ventilation.1,2 Traditionally, sedative choice has been limited to benzodiazepines and propofol, which have been associated with oversedation, respiratory depression, and delirium among other adverse effects.3–6 Recent guidelines suggest dexmedetomidine, a selective α-2 adrenergic agonist, as one of the preferred sedative options for mechanically ventilated adult ICU patients. Dexmedetomidine offers the benefit of producing light sedation with minimal respiratory depression when used in the absence of other sedative or analgesic agents.3

Hypotension and bradycardia have been associated commonly with dexmedetomidine therapy, occurring in 13% to 68% and 1% to 42% of patients, respectively.4–11 The variability in reported incidence may be partially attributed to inconsistent definitions and study populations. The significance of this hemodynamic instability is not only highlighted by its high incidence but also the need for corrective interventions. In one study, hemodynamic instability requiring clinical intervention occurred in nearly one third of ICU patients receiving dexmedetomidine.4 In comparison, similar rates of hypotension and bradycardia have been reported with propofol (13% and 10%, respectively) and benzodiazepines (12%–56% and 2%–19%, respectively) among critically ill patients.4,6,9

Although hemodynamic instability may negatively impact outcomes in the ICU, specific risk factors for the development of clinically significant hemodynamic instability in patients receiving dexmedetomidine are poorly characterized in the current literature.7,8,12,13 Although previous studies in focused populations have implicated dexmedetomidine dosing strategies, including initial loading infusions and titration frequency, the degree to which alternative patient-specific factors or concurrent interventions impact the risk of hemodynamic instability remains incompletely understood.7,8,12,13 The current study’s primary aims were to further characterize the incidence and clinical significance of hemodynamic instability during dexmedetomidine sedation in a broad, mixed medical and surgical ICU population and to identify patient- and treatment-specific risk factors predictive of dexmedetomidine-associated hemodynamic instability.



This study was approved by the Mayo Clinic IRB. All patients included in this study provided written informed consent to use their medical records for research purposes.

Study Setting and Patient Population

This retrospective cohort study included adult ICU patients at a tertiary-care academic medical center who were sedated with dexmedetomidine between November 1, 2012, and October 31, 2013. Patients were excluded if they were admitted to a cardiac surgical ICU or coronary care unit, received dexmedetomidine for procedural sedation only, had a cardiac pacemaker or automatic implantable cardioverter defibrillator, were admitted with a primary diagnosis of substance withdrawal, were pregnant, or were incarcerated.

At our institution, the dexmedetomidine order set specifies a standard initial rate of 0.2 μg/kg/h and a recommended maximum rate of 0.7 μg/kg/h. Providers may use their clinical judgment to exceed this threshold. The order set does not specify an absolute maximum infusion rate. Dose titration targets a Richmond Agitation Sedation Scale (RASS) levels of −2 to zero and is performed in 0.1 μg/kg/h increments to achieve target RASS levels or allow for institutional daily awakening protocols. Targeting RASS levels below −2 is uncommon because of the relatively light sedation produced by dexmedetomidine. The dexmedetomidine order set allows for initial loading infusions of 0.5 to 1.0 μg/kg, but these were used infrequently during the study time period. Invasive mechanical ventilation is not required for our patients to receive dexmedetomidine, but its use is restricted to ICUs or procedural areas. No changes were made to dexmedetomidine or other sedation practices during the study timeframe. Arterial catheters are placed routinely in our critically ill adults receiving sedation and typically remain in place throughout sedation titration. Automated monitoring systems recorded arterial blood pressure and heart rate at 15-minute intervals, and this information was preferentially collected from arterial catheter values unless unavailable.

Data Collection and Study End Points

Abstracted data from the electronic medical record included baseline demographics, presence and type of underlying cardiac disease, Acute Physiology and Chronic Health Evaluation (APACHE) III score at 24 hours from ICU admission, day 1 Sequential Organ Failure Assessment (SOFA) score, and hemodynamics before dexmedetomidine initiation.14,15 Additional data gathered included dexmedetomidine dose and the concomitant use of continuous infusion sedatives, antihypertensives, antiarrhythmics, and vasoactive support. Patient follow-up was limited to the initial course of dexmedetomidine sedation during the specified hospitalization.

We hypothesized that hemodynamic instability develops in patients who receive dexmedetomidine in part because of patient- or treatment-specific risk factors. The primary end point was the occurrence of at least 1 episode of clinically significant hemodynamic instability during dexmedetomidine therapy, defined as systolic blood pressure (SBP) <80 mm Hg, diastolic blood pressure (DBP) <50 mm Hg, or heart rate <50 beats per minute (bpm).4,7–9 To qualify as an event, the hemodynamic variable had to remain below the specified threshold for at least 2 consecutive readings (≥30 minutes of recorded hemodynamic instability). The identified events were used to evaluate the relationship between hemodynamic instability and collected predictors.

Secondary end points included the total number of hemodynamic instability events and the number of clinical interventions required for those events. To support the findings of hemodynamic instability, each event was evaluated to determine the likelihood of its association with dexmedetomidine by previously validated methods.16


Underlying cardiac disease was defined as structural heart disease, coronary artery disease, previous myocardial infarction, or chronic heart failure. Evaluated concomitant antihypertensive or antiarrhythmic medications were those most frequently used in our ICUs and included β-blockers, calcium channel blockers, amiodarone, and digoxin administered 24 hours before, or during, dexmedetomidine therapy. Low baseline arterial blood pressure was defined as SBP <100 mm Hg or mean arterial blood pressure (MAP) <70 mm Hg in the 60 minutes preceding dexmedetomidine initiation, and slow baseline heart rate was <70 bpm in the 60 minutes preceding dexmedetomidine initiation. Continuous infusion sedatives included fentanyl, hydromorphone, propofol, midazolam, lorazepam, or ketamine concomitantly administered with dexmedetomidine.

Interventions for hemodynamic instability were defined as the discontinuation of dexmedetomidine within 60 minutes or 1 or more of the following within 30 minutes of the event: dexmedetomidine dose decrease, vasopressor initiation or dose increase, a fluid bolus of at least 250 mL crystalloid or colloid equivalent, atropine administration, or transcutaneous pacing.17 Two investigators independently evaluated each hemodynamic instability event to determine its likelihood of association with dexmedetomidine by using an adverse drug reaction probability scale validated by Naranjo et al.16 A score of 0 to 10 was assigned to each event, with subsequent transformation of numerical risk rating to a categorical classification of definite, probable, possible, or doubtful. The results were then analyzed for interobserver variance. If the categorized association differed between the 2 investigators, a third adjudicator, blinded to the previous assessments, reviewed the case. In patients with multiple hemodynamic instability events, the total number of events was recorded, but assessment for interventions and adverse drug reaction evaluation were completed for only the first 6 events.

Data Analysis

Summary statistics were computed for the entire sample, as appropriate for the variable type. Cumulative incidence of hemodynamic instability was estimated using Kaplan-Meier methodology. For the adverse drug reaction agreement assessment, a weighted kappa statistic with 95% confidence interval (CI) was computed. To assess the associations of variables with risk of hemodynamic instability, univariate Cox proportional hazards models were fit for the outcome of time to first hemodynamic instability, and unadjusted hazard ratios (HRs), 95% CI, and P values were computed for each variable. Time 0 was defined as the initiation of dexmedetomidine, and last follow-up time was defined as 30 minutes after dexmedetomidine cessation. Variables measured after dexmedetomidine initiation were treated as time-dependent covariates. Functional form of variables and the proportional hazards assumptions were assessed using Martingale residuals and found to be satisfied for almost all variables.18 The only exception was low baseline arterial blood pressure, which exhibited a diminished effect on hemodynamic instability with increased follow-up time. To account for this, an interaction term of low baseline arterial blood pressure and time was included in the Cox models.

A multivariable Cox model was fit that included all potential predictors collected. Adjusted HR, 95% CI, and P values were computed for each variable. Data obtained for the first hemodynamic instability event only were included in the Cox model. For an approximately balanced binary risk factor, 66 patients with a hemodynamic instability event were needed to provide 80% power to detect a HR of 2. Given the range in dexmedetomidine-associated hemodynamic instability reported in previous literature, we assumed a cumulative incidence of 22%, indicating that a sample size of 300 patients should yield at least 66 patients with hemodynamic instability. Initially, 7 variables were specified a priori to assess as risk factors of hemodynamic instability; however, given the observed high incidence of hemodynamic instability, 17 variables were included in the multivariable Cox model. To adjust for propofol use specifically, a second Cox model was estimated that included concomitant propofol use as a predictor instead of the number of concomitant sedatives. This study was not designed to investigate any interactions between predictors for their effect on hemodynamic instability. Because it was not properly powered for this purpose, interactions among predictors were not considered in the Cox models. All analyses were performed with SAS version 9.2 statistical software (SAS Institute Inc., Cary, NC).


Five hundred thirty-six potentially eligible noncardiac critically ill adults received dexmedetomidine for ICU sedation during the selected study timeframe. Patients were randomly screened until the desired sample size of 300 patients was reached. Three hundred eighty-six patients were examined, and 86 were excluded (Fig. 1). The median age of the 300 study patients was 63 years (interquartile range [IQR], 52–75], and patients were primarily men (59%). The median APACHE III and SOFA scores were 74 (IQR, 54–89) and 7 (IQR, 5–10), respectively. Initial dexmedetomidine loading infusions were used in 62 patients (21%); however, the majority of these (56%) were partial loading infusions (<0.75 μg/kg). The median maximum dexmedetomidine infusion rate was 0.7 μg/kg/h, with only 27% of patients receiving infusion rates >0.7 μg/kg/h during therapy. The median MAP change was a decrease of 2.5 mm Hg (IQR decrease of 11 an increase of 4) from the baseline to within 60 minutes after dexmedetomidine initiation. The cardiac SOFA score, a secondary indicator of hemodynamic instability, worsened in 69 patients (23%) after the dexmedetomidine infusion began. Additional baseline demographics and dexmedetomidine therapy characteristics are summarized in Table 1.

Table 1:
Patient and Dexmedetomidine Therapy Characteristics
Figure 1:
Flow diagram of cohort patients. Other exclusion were incarceration (N = 1), initiated dexmedetomidine infusion at an outside facility (N = 1), and initiated dexmedetomidine outside of study timeframe (N = 1).

The primary end point of cumulative incidence of hemodynamic instability within 24 hours of dexmedetomidine initiation was estimated to be 71% (95% CI, 64%–77%) and was associated with a median time to event of 4.25 hours (Fig. 2). At least 1 episode of hemodynamic instability occurred in 197 patients, and the first hemodynamic instability events were further stratified by type to include hypotension in 182 patients, bradycardia in 12 patients, and both simultaneously in 3 patients. Patients experiencing bradycardia had a median reduction in heart rate of 20 bpm. Those with hypotension experienced a median reduction in SBP and MAP of 13 and 19 mm Hg, respectively (Table 2). The hypotension results were largely driven by low DBP, with 83% (154 of 185) of hypotensive patients meeting hemodynamic instability criteria solely because of DBP <50 mm Hg at the time of first event. The median MAP was 64 mm Hg (IQR, 58–68 mm Hg) at the time of first event, and 84% (156 of 185) of hypotensive patients also had a MAP <70 mm Hg at event onset. Patients experiencing hemodynamic instability received a median cumulative dexmedetomidine dose of 44 μg (IQR, 12–138 μ) at the time of the first event. In addition, they experienced a median of 1 dexmedetomidine dose titration (IQR, 0–1) and received a median dexmedetomidine infusion rate of 0.5 μg/kg/h (IQR, 0.3–0.7 μg/kg/h) within 60 minutes preceding the event.

Table 2:
Hemodynamic Instability Event Type and Intervention Requirement
Figure 2:
Kaplan-Meier curve for estimate of dexmedetomidine-associated hemodynamic instability cumulative incidence per hour of dexmedetomidine therapy. Hemodynamic instability was the composite of hypotension (systolic blood pressure <80 mm Hg or diastolic blood pressure <50 mm Hg) and/or bradycardia (heart rate <50 beats per minute). The solid line represents cumulative incidence, and the dotted lines represent the 95% confidence interval. Dexmedetomidine was associated with a median time-to-hemodynamic instability of 4.25 hours.

Patient- and treatment-specific characteristics were analyzed as predictors of hemodynamic instability. Univariate analysis revealed only age, vasopressor use at dexmedetomidine initiation, low baseline arterial blood pressure, and use of concomitant sedatives in addition to dexmedetomidine as significant predictors of hemodynamic instability (Table 3). In the multivariable analysis, both age (HR, 1.23 per 10 years; 95% CI, 1.10–1.38; P < 0.001) and low baseline arterial blood pressure (HR, 2.42 at dexmedetomidine initiation; 95% CI, 1.68–3.49; P < 0.0001) were predictors of dexmedetomidine-associated hemodynamic instability (Table 4). A significant interaction was noted between baseline arterial blood pressure and time (P = 0.04). The effect of low baseline arterial blood pressure on hemodynamic instability decreased over time, with its predictive ability dissipating by the ninth hour of dexmedetomidine therapy (HR, 1.41 at 9 hours; 95% CI, 0.92–2.16; P = 0.11). The majority of patients with low baseline arterial blood pressure were classified as hemodynamically stable at baseline per study definitions; however, a minority (17%) met criteria for hemodynamic instability in the 60 minutes preceding dexmedetomidine initiation and had subsequent worsening of hemodynamic variables. An additional analysis excluding patients with preexisting hemodynamic instability at dexmedetomidine initiation resulted in a median time to event of 5.75 hours and cumulative hemodynamic instability incidence of 69% at 24 hours. A multivariable sensitivity analysis excluding these patients was consistent with the primary study findings (data not shown).

Table 3:
Univariate Analysis of Patient and Treatment Characteristics
Table 4:
Multivariable Analysis of Patient and Treatment Characteristics

Our sample did not demonstrate an association between the risk for hemodynamic instability and the presence of cardiac disease, concomitant cardiac medications, concomitant sedative therapies, dexmedetomidine doses in excess of 0.7 μg/kg/h, or other evaluated characteristics (Table 4). To evaluate the specific impact of concomitant propofol itself, a separate multivariable Cox model including propofol use instead of the number of concomitant therapies showed that propofol was not significantly associated with hemodynamic instability (HR, 1.32; 95% CI, 0.87–2.00; P = 0.19).

Table 5:
Assessment of Dexmedetomidine Association with Hemodynamic Instability Using an Adverse Drug Reaction Probability Scale16

Secondary end point analysis revealed 485 hemodynamic instability events in the 300-patient sample during the course of dexmedetomidine therapy. Of the total cohort, patients experienced a median of 1 event (IQR, 0–2). All events up to the first 6 for each patient were completely evaluated. Thirteen patients experienced >6 hemodynamic instability events during the initial dexmedetomidine course. In these cases, intervention and adverse drug reaction probability scoring were recorded on the first 6 events only. The need for clinical intervention was highest for the primary event (64%) and also frequent in the total of 435 evaluated events (48%). Of these, dexmedetomidine dose reduction (36% of index events and 28% of total events) was the most common intervention (Table 2). Notably, only 9% of patients with hemodynamic instability had dexmedetomidine discontinued as a result of the first event. A validated adverse drug reaction probability scale revealed that the association of dexmedetomidine with the development of hemodynamic instability was high for the majority of events, with classification of definite or probable in 4% and 76% of events, respectively (Table 5). Agreement analysis indicated a high interrater reliability before adjudication with a weighted kappa statistic of 0.89 (95% CI, 0.84–0.95) for the primary events.


We found that the cumulative incidence of hemodynamic instability within 24 hours of initiating sedation with dexmedetomidine was 71% in this mixed medical and surgical ICU cohort. The majority of patients experienced at least 1 additional episode of hemodynamic instability after the initial event. The clinical significance of these events is evidenced by the need for intervention in 64% and 48% of initial and total events, respectively. In addition, advanced age and the presence of low baseline arterial blood pressure were identified as predictors of hemodynamic instability. The predictive value of low baseline arterial blood pressure decreased as a function of time from dexmedetomidine initiation and dissipated by the ninth hour of therapy. Other variables including concurrently administered cardiac medications or sedative therapies and dexmedetomidine doses >0.7 μg/kg/h did not predict the development of hemodynamic instability in our model.

The reported incidences of hypotension and bradycardia in dexmedetomidine-treated patients have varied significantly with ranges of 13% to 68% and 1% to 42%, respectively.4,6–11 We observed a profound cumulative incidence of hemodynamic instability, marginally higher than the majority of previous studies.4,6–9,11 This may be explained by inconsistencies in follow-up time and absent analysis for the effect of time on incidence in previous studies.4,6,9 Alternatively, because our institution lacks a fixed titration protocol, differences in dexmedetomidine dosing between institutions may have contributed to the observed high incidence. However, the use of full initial loading infusions and maximum infusion rates >0.7 μg/kg/h were infrequent in our patients, which reduces the probability that high dexmedetomidine doses were a contributing factor.

There is no widely accepted definition of dexmedetomidine-associated hemodynamic instability in the currently available literature. Although bradycardia criteria are well established and consistent with our study definition, dexmedetomidine-associated hypotension has been defined previously using a variety of SBP, DBP, and MAP targets.4,6–9,11,12 The Safety and Efficacy of Dexmedetomidine Compared with Midazolam trial conducted by Riker et al.,4 which compared dexmedetomidine with midazolam in a medical/surgical ICU population, used a hypotension definition most similar to this study. That trial demonstrated high rates of both hypotension (56%) and bradycardia (42%) and reported that more than one third of these events required clinical intervention. Their results are comparable with the hemodynamic instability incidence and clinical intervention requirement we report here. In our investigation, both SBP and DBP were used as indicators of hypotension, which led to a surprisingly high rate of diastolic-driven hypotensive events (85% of first events). Previous dexmedetomidine studies that use both SBP and DBP in their hypotension definition have not reported differences in systolic- versus diastolic-driven hypotensive events.4,7,9,12 Although it is unknown whether dexmedetomidine preferentially predisposes patients to systolic or diastolic hypotension, current evidence suggests an association between low DBP and poor clinical prognosis.19–22 A recent echocardiographic assessment of the cardiac function of patients receiving dexmedetomidine infusion did not demonstrate an impairment in systolic or diastolic function but did find a reduction in cardiac output because of a negative chronotropic effect of dexmedetomidine.23 Of note, these data were obtained in younger, otherwise-healthy patients undergoing procedural anesthesia for orthopedic surgery and, therefore, may not be translatable to our critically ill population. However, further investigation into the mechanism of dexmedetomidine-associated hemodynamic instability is warranted.

Evidence previously has established several dexmedetomidine dosing strategies as predictive of hemodynamic instability, including the use of initial loading infusions and frequent dose titrations.7,12 However, the literature remains inconclusive about the association between dexmedetomidine doses in excess of 0.7 μg/kg/h and incidence of hemodynamic instability.7,8,12 A retrospective review conducted in trauma patients compared standard-dose (≤0.7 μg/kg/h) and high-dose (>0.7 μg/kg/h) dexmedetomidine with propofol. The authors reported a significantly greater rate of hypotension when comparing high-dose dexmedetomidine with propofol, an observation absent in the standard-dose comparison. Notably, nearly half of these events were associated with dexmedetomidine loading infusions and may not be truly reflective of the maximum infusion rate.12 In contrast, a separate retrospective study demonstrated no difference in rates of hypotension and bradycardia in a direct comparison of standard-dose to high-dose dexmedetomidine.8 Although our findings support the results of the latter study, we note that dexmedetomidine infusions rates exceeded 0.7 μg/kg/h in only 26% of our cohort.

Although dexmedetomidine dosing characteristics have demonstrated an impact on the risk of hemodynamic instability, few data have evaluated patient characteristics and concurrent treatment considerations that impact the risk of hemodynamic instability with dexmedetomidine.7,8,12 One recent retrospective cohort study, published in an abstract form, identified APACHE II score and baseline MAP <70 mm Hg as predictors of dexmedetomidine-associated hypotension.13 Although our findings endorse low baseline arterial blood pressure as a risk factor of hemodynamic instability, our evaluation of APACHE III scores did not reveal increased hemodynamic instability risk. In addition, increasing age was associated with hemodynamic instability in our sample but was not a predictor in the previous study.13 The present investigation expanded risk factor assessment and found body mass index, slow baseline heart rate, cardiac disease, concomitant administration of cardiac medications, and/or sedative therapies do not alter the risk of hemodynamic instability.

The retrospective nature of this study limits its ability to implicate dexmedetomidine as a potential cause of hemodynamic instability. However, previous literature has firmly established the association of both hypotension and bradycardia with dexmedetomidine use. We sought to overcome this limitation through the inclusion of other potentially confounding causes of hemodynamic instability in a multivariable analysis, analyzing the time course of event development, and determining the likelihood of event association with a validated adverse drug reaction probability scale. Within the limits of retrospective research, these measures help to strengthen the association of dexmedetomidine with the observed events. In addition, we used the most common definition of dexmedetomidine-associated hemodynamic instability but recognize that there is no consensus definition yet. To clarify the clinical significance of our defined events, we evaluated additional pertinent hemodynamic considerations such as concurrent changes in the cardiac component of the SOFA score and intervention requirements. Although the approach to intervention varied by individual provider and context of the clinical situation, the frequent need to intervene further reinforced the impact of the hemodynamic instability we observed. In addition, this study is limited by its primarily male Caucasian population at a single center; however, the tertiary-care nature of the center and mixed medical/surgical population increase the generalizability of the results. Finally, our institution did not use a standard dexmedetomidine titration protocol, which may have contributed to a high incidence of hypotension. To address this issue, initial loading infusions and high-dose dexmedetomidine (>0.7 μg/kg/h) were included as potential factors in univariate and multivariable analyses and found not to be associated with hemodynamic instability. Furthermore, the lack of dosing restrictions increases the generalizability of the results.

In conclusion, hemodynamic instability is common in critically ill adults receiving dexmedetomidine for sedation. In our study, more than two thirds of patients developed hypotension and/or bradycardia within 24 hours of dexmedetomidine therapy, and nearly half of all instability events required a clinical intervention, most commonly a dexmedetomidine dose reduction. Increasing age and low baseline arterial blood pressure were associated with the development of hemodynamic instability after the initiation of dexmedetomidine. These findings suggest that clinicians should be aware of the risk of hemodynamic instability when using dexmedetomidine in patients of advancing age or with a baseline SBP <100 mm Hg or MAP <70 mm Hg.


Name: Calvin J. Ice, PharmD.

Contribution: This author helped design the study, conduct the study, collect the data, conduct data analysis, and prepare the manuscript.

Attestation: Calvin J. Ice has approved the final manuscript, attests to the integrity of the original data and the analysis reported in this manuscript, and is the archival author.

Name: Heather A. Personett, PharmD.

Contribution: This author helped design the study, conduct the study, collect the data, conduct data analysis, and prepare the manuscript.

Attestation: Heather A. Personett has approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Erin N. Frazee, PharmD.

Contribution: This author helped design the study, conduct the study, collect the data, conduct data analysis, and prepare the manuscript.

Attestation: Erin N. Frazee has approved the final manuscript and attests to the integrity of the original data and the analysis reported in this manuscript.

Name: Ross A. Dierkhising, MS.

Contribution: This author helped design the study, conduct data analysis, and prepare the manuscript.

Attestation: Ross A. Dierkhising has approved the final manuscript.

Name: Rahul Kashyap, MBBS.

Contribution: This author helped collect the data, conduct data analysis, and prepare the manuscript.

Attestation: Rahul Kashyap has approved the final manuscript.

Name: Richard A. Oeckler, MD, PhD.

Contribution: This author helped design the study, conduct the study, and prepare the manuscript.

Attestation: Richard A. Oeckler has approved the final manuscript.

This manuscript was handled by: Avery Tung, MD.


The authors thank the support of the Mayo Clinic Multidisciplinary Epidemiology and Translational Research in Intensive Care (METRIC) team in obtaining data from the electronic medical record. Study data were collected and managed using REDCap electronic data capture tools hosted at Mayo Clinic.24


1. Salgado DR, Favory R, Goulart M, Brimioulle S, Vincent JL. Toward less sedation in the intensive care unit: a prospective observational study. J Crit Care. 2011;26:113–21
2. Hughes CG, McGrane S, Pandharipande PP. Sedation in the intensive care setting. Clin Pharmacol. 2012;4:53–63
3. Barr J, Fraser GL, Puntillo K, Ely EW, Gélinas C, Dasta JF, Davidson JE, Devlin JW, Kress JP, Joffe AM, Coursin DB, Herr DL, Tung A, Robinson BR, Fontaine DK, Ramsay MA, Riker RR, Sessler CN, Pun B, Skrobik Y, Jaeschke RAmerican College of Critical Care Medicine. . Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med. 2013;41:263–306
4. Riker RR, Shehabi Y, Bokesch PM, Ceraso D, Wisemandle W, Koura F, Whitten P, Margolis BD, Byrne DW, Ely EW, Rocha MGSEDCOM (Safety and Efficacy of Dexmedetomidine Compared with Midazolam) Study Group. . Dexmedetomidine vs midazolam for sedation of critically ill patients: a randomized trial. JAMA. 2009;301:489–99
5. Ruokonen E, Parviainen I, Jakob SM, Nunes S, Kaukonen M, Shepherd ST, Sarapohja T, Bratty JR, Takala J“Dexmedetomidine for Continuous Sedation” Investigators. . Dexmedetomidine versus propofol/midazolam for long-term sedation during mechanical ventilation. Intensive Care Med. 2009;35:282–90
6. Jakob SM, Ruokonen E, Grounds RM, Sarapohja T, Garratt C, Pocock SJ, Bratty JR, Takala JDexmedetomidine for Long-Term Sedation Investigators. . Dexmedetomidine vs midazolam or propofol for sedation during prolonged mechanical ventilation: two randomized controlled trials. JAMA. 2012;307:1151–60
7. Gerlach AT, Dasta JF, Steinberg S, Martin LC, Cook CH. A new dosing protocol reduces dexmedetomidine-associated hypotension in critically ill surgical patients. J Crit Care. 2009;24:568–74
8. Jones GM, Murphy CV, Gerlach AT, Goodman EM, Pell LJ. High-dose dexmedetomidine for sedation in the intensive care unit: an evaluation of clinical efficacy and safety. Ann Pharmacother. 2011;45:740–7
9. Pandharipande PP, Pun BT, Herr DL, Maze M, Girard TD, Miller RR, Shintani AK, Thompson JL, Jackson JC, Deppen SA, Stiles RA, Dittus RS, Bernard GR, Ely EW. Effect of sedation with dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients: the MENDS randomized controlled trial. JAMA. 2007;298:2644–53
10. Bharati S, Pal A, Biswas C, Biswas R. Incidence of cardiac arrest increases with the indiscriminate use of dexmedetomidine: a case series and review of published case reports. Acta Anaesthesiol Taiwan. 2011;49:165–7
11. Erdman MJ, Doepker BA, Gerlach AT, Phillips GS, Elijovich L, Jones GM. A comparison of severe hemodynamic disturbances between dexmedetomidine and propofol for sedation in neurocritical care patients. Crit Care Med. 2014;42:1696–702
12. Devabhakthuni S, Pajoumand M, Williams C, Kufera JA, Watson K, Stein DM. Evaluation of dexmedetomidine: safety and clinical outcomes in critically ill trauma patients. J Trauma. 2011;71:1164–71
13. Gerlach AT, Blais D, Jones G, Burcham P, Stawicki S, Cook C, Murphy C. Abstracts of the 42nd Critical Care Congress. January 19–23, 2013. San Juan, Puerto Rico. Crit Care Med. 2012;40(12 suppl 1):1–328
14. Knaus WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG, Sirio CA, Murphy DJ, Lotring T, Damiano A. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;100:1619–36
15. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, Reinhart CK, Suter PM, Thijs LG. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22:707–10
16. Naranjo CA, Busto U, Sellers EM, Sandor P, Ruiz I, Roberts EA, Janecek E, Domecq C, Greenblatt DJ. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther. 1981;30:239–45
17. Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton RSAFE Study Investigators. . A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350:2247–56
18. Lin DY, Wei LJ, Ying Z. Checking the cox model with cumulative sums of martingale-based residuals. Biometrika. 1993;80:557–72
19. Benchekroune S, Karpati PC, Berton C, Nathan C, Mateo J, Chaara M, Riché F, Laisné MJ, Payen D, Mebazaa A. Diastolic arterial blood pressure: a reliable early predictor of survival in human septic shock. J Trauma. 2008;64:1188–95
20. Rigamonti F, Graf G, Merlani P, Bendjelid K. The short-term prognosis of cardiogenic shock can be determined using hemodynamic variables: a retrospective cohort study*. Crit Care Med. 2013;41:2484–91
21. Guichard JL, Desai RV, Ahmed MI, Mujib M, Fonarow GC, Feller MA, Ekundayo OJ, Bittner V, Aban IB, White M, Aronow WS, Love TE, Bakris GL, Zieman SJ, Ahmed A. Isolated diastolic hypotension and incident heart failure in older adults. Hypertension. 2011;58:895–901
22. Tringali S, Oberer CW, Huang J. Low diastolic blood pressure as a risk for all-cause mortality in VA patients. Int J Hypertens. 2013;2013:178780
23. Lee SH, Choi YS, Hong GR, Oh YJ. Echocardiographic evaluation of the effects of dexmedetomidine on cardiac function during total intravenous anaesthesia. Anaesthesia. 2015;70:1052–9
24. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42:377–81
© 2016 International Anesthesia Research Society