Scott Mueller

A randomized, double-blind, placebo-controlled dose range study of dexmedetomidine as adjunctive therapy for alcohol withdrawal.

Objectives:To evaluate dexmedetomidine as adjunctive therapy to lorazepam for severe alcohol withdrawal. Design:Prospective, randomized, double-blind, placebo-controlled trial. Setting:Single center; medical ICU. Patients:Twenty-four adult patients with a Clinical Institute Withdrawal Assessment score greater than or equal to 15 despite greater than or equal to 16 mg of lorazepam over a 4-hour period. Interventions:Patients received a symptom-triggered Clinical Institute Withdrawal Assessment protocol with lorazepam and were randomized to dexmedetomidine 1.2 &mgr;g/kg/hr (high dose), 0.4 &mgr;g/kg/hr (low dose), or placebo as adjunctive therapy for up to 5 days or resolution of withdrawal symptoms. Measurement and Main Results:High-dose and low-dose groups were combined as a single dexmedetomidine group for primary analysis with secondary analysis exploring a dose-response relationship. The difference in 24-hour lorazepam requirements after versus before study drug was greater in the dexmedetomidine group compared with the placebo group (–56 mg vs –8 mg, p = 0.037). Median differences were similar for high dose and low dose. The 7-day cumulative lorazepam requirements were not statistically different between dexmedetomidine and placebo (159 mg vs 181 mg). Clinical Institute Withdrawal Assessment or Riker sedation-agitation scale scores representing severe agitation (13% vs 25%) or moderate agitation (27% vs 22%) within 24 hours of initiating study drug were similar for dexmedetomidine and placebo groups, respectively. Bradycardia occurred more frequently in the dexmedetomidine group versus placebo group (25% vs 0%, p = not significant), with the majority of bradycardia occurring in the high-dose group (37.5%). Study drug rate adjustments occurred more often in the dexmedetomidine group compared with the placebo group (50% vs 0%, p = 0.02). Neither endotracheal intubation nor seizure occurred in any group while on study drug. Conclusions:Adjunctive dexmedetomidine for severe alcohol withdrawal maintains symptom control and reduces lorazepam exposure in the short term, but not long term, when using a symptom-triggered protocol. Monitoring for bradycardia is needed with dexmedetomidine but the occurrence may be lessened with low dose. Further study is needed to evaluate the clinical impact of dexmedetomidine.

Vasopressors and Inotropes

Medication errors and adverse drug events occur more frequently in the intensive care unit compared to general care units.1 Adverse drug events become more likely as patients receive more medications. Sentinel events and medication errors are more common as the number of failing organs increases.2,3 Vasopressors are frequently associated with adverse drug events and they are considered high-alert drugs by the Institute of Safe Medication Practices due to their increased potential to cause harm.4-6 Vasopressors and inotropes are used in patients with the highest acuity and under stressful situations which adds to the potential for errors. In addition, dosing guidelines, ranges and units are not always standardized across agents at specific institutions or for a certain agent across institutions. The literature is disparate with respect to dosing recommendations. With the exception of vasopressin, we report the dosing of these agents in a weight-based manner. This should enhance dosing consistency to help minimize errors. We encourage institutions to adopt this dosing scheme to reduce discrepancies associated with their administration. Moreover, using a weight based dosing strategy in an era of increasing obesity raises the question of whether actual, adjusted or ideal body weight should be used when administering vasopressors. No data are available to select an appropriate weight and trials evaluating the use of vasopressors rarely report the body weight used in cases of obesity. With the exception of milrinone, we encourage the use of ideal body weight for all weight-based dosing strategies because these agents possess short half-lives, rapid onsets, and low volumes of distribution but may be associated with severe adverse events when higher weights are used resulting in higher doses in heavier patients. Moreover, these agents are rapidly titrated to clinical response, so starting at lower doses based on ideal body weight is prudent.

Neuromuscular Blockade Resistance During Therapeutic Hypothermia

Objective: To report a case of neuromuscular blockade resistance to multiple agents during therapeutic hypothermia and discuss possible mechanisms of this resistance. Case Summary: A 64-year-old man with stage IV non–small-cell lung cancer and respiratory distress developed cardiac arrest in the emergency department. The man was quickly resuscitated and treated with therapeutic hypothermia. A chest tube was Inserted for pleural drainage of a large right-sided effusion that collapsed the right lung; this was unsuccessful in reinflating the lung. A bronchopleural fistula developed and independent lung ventilation was initiated due to persistent hypoxemia. Neuromuscular blockade was initiated after sedation and analgesia did not control shivering and was continued due to patient-ventilator dyssynchrony and persistent hypoxemia. Despite large doses of 3 different neuromuscular blocking agents and negligible response to train-of-four tests, clinical neuromuscular blockade, represented by ventilator synchrony, was not achieved until the patient was warmed. Discussion: Resistance to neuromuscular blocking agents has been reported in critically ill patients. Our case of neuromuscular blockade resistance occurred in a patient treated with therapeutic hypothermia, which generally requires a dose reduction of neuromuscular blocking agents. Resistance to neuromuscular blockade was quickly reversed upon warming of the patient as patient-ventilator synchrony was achieved at lower neuromuscular blocking agent doses. Conclusions: Clinicians should be aware of a potential blunted response to neuromuscular blocking agents during therapeutic hypothermia and difficulty with paralysis monitoring since train-of-four response may correlate poorly with clinical neuromuscular blockade during hypothermia. Further research is needed to elucidate the mechanism of this interaction, identify patients at risk, and evaluate alternative strategies to neuromuscular blockade for controlling shivering in patients undergoing therapeutic hypothermia.

Prefilter Bivalirudin for Preventing Hemofilter Occlusion in Continuous Renal Replacement Therapy

Objective: To describe a case of successful bivalirudin use as a prefilter anticoagulant in continuous venovenous hemofiltration (CVVH). Case Summary: A 30-year-old male was brought to the hospital by ambulance with an anterior communicating artery subarachnoid hemorrhage, signs of intraparenchymal hemorrhage, and hydrocephalus. During the patients complicated hospital course, he developed acute renal failure requiring CVVH, as well as hepatic insufficiency (Child-Pugh class B). Unfractionated heparin was used as a prefilter anticoagulant. After he had a positive heparin-induced thrombocytopenia (HIT) antibody test, prefilter heparin was discontinued in favor of bivalirudin. Filter survival and systemic activated partial thromboplastin time (aPTT) values were compared between prefilter heparin (n = 5) and bivalirudin (n = 4). Filter survival was similar (median 26 h with heparin vs 37 h with bivalirudin; p = 0.52). Prefilter bivalirudin 1–2.5 mg/hour (0.009–0.023 mg/kg/h) was effective in maintaining systemic aPTTs that were 1–1.4 times the reference range. Serotonin release assay and subsequent HIT antibodies were negative. The patients renal function improved and CVVH was discontinued. Discussion: Critically ill patients requiring CVVH often need regional or systemic anticoagulation to prevent filter occlusion. In some patient populations, such as those with HIT or liver failure, prefilter heparin and regional citrate, respectively, may not be options. Alternative anticoagulants may be needed to avoid complications of frequent filter occlusions. The direct thrombin inhibitors (DTIs) lepirudin and argatroban have been used to maintain hemofilter patency, in small studies. Bivalirudin may have pharmacokinetic advantages over other DTIs when used in patients with hepatic and renal impairment. In our patient, bivalirudin provided a safe alternative to heparin therapy and was effective in maintaining hemofilter patency during CVVH. Conclusions: Prefilter bivalirudin may be an option to prevent filter occlusion in patients requiring continuous renal replacement therapy. Future studies are needed to validate the safety and efficacy of bivalirudin as a prefilter anticoagulant.

Stability of dexmedetomidine in polyvinyl chloride bags containing 0.9% sodium chloride injection

PURPOSE The stability of dexmedetomidine in polyvinyl chloride (PVC) bags containing 0.9% sodium chloride injection was studied. METHODS Dexmedetomidine solutions (4, 8, 12, and 20 μg/mL; n = 6 for each) were prepared by removing 2, 4, 6, and 10 mL of 0.9% sodium chloride injection, respectively, from 50-mL PVC bags and injecting 2, 4, 6, and 10 mL of dexmedetomidine 100 μg/mL, respectively. To ensure a homogeneous mixture, the contents of each bag was manually mixed initially and before each sample was removed. All compounding was conducted by a single pharmacist using aseptic technique in a horizontal-laminar-airflow hood at 25 °C. Forced-degradation studies were conducted at 70 ± 1 °C. Stability samples were analyzed using high-performance liquid chromatography electrospray ionization-tandem mass spectrometry (LC/MS-MS) and high-performance liquid chromatography-ultraviolet-light (HPLC/UV) absorbance. Forced-degradation samples were monitored using LC/MS-MS, HPLC/UV, and gas chromatography-MS. RESULTS Dexmedetomidine solutions were very stable at 23 ± 2 °C at all four concentrations over the 48-hour testing period. As determined via LC/MS-MS and HPLC/UV methods, over 97% of the initial concentration of dexmedetomidine remained after 48 hours. Extensive HPLC/UV active degradation products could be observed in basic conditions; only minor UV active degradation products were observed in acidic, oxidative, and photochemical conditions. CONCLUSION Dexmedetomidine hydrochloride 4, 8, 12, and 20 μg/mL stored in PVC bags at 23 ± 2 °C was stable for 48 hours, despite a slight decrease in solution pH seen with increasing dexmedetomidine concentrations.

Fospropofol Disodium for Procedural Sedation: Emerging Evidence of its Value?

Background: Fospropofol is a phosphate-ester water soluble prodrug of propofol approved in the United States for the short-term sedation of adults undergoing diagnostic or therapeutic procedures with monitored anesthesia care. Fospropofol may be advantageous for procedural sedation because it is not a lipid formulation, it lacks pain on injection, it possesses a different pharmacokinetic profile due to in vivo conversion of fospropofol to propofol, and it offers rapid recovery from sedation. Objective: To review published pharmacokinetic data, clinical trials, and the safety profile of fospropofol and provide comment on its use for sedation. Clinical trials: Fospropofol has been studied for sedation in colonoscopy and bronchoscopy with fentanyl for analgesia. Results indicate that a 6.5 mg/kg bolus dose with repeated doses of 25% of the original bolus dose up to every 4 minutes produces an effective and safe level of sedation. Fospropofol comparisons to midazolam were not assessed. Patient satisfaction is high for those treated with the recommended dosage regimen. Most commonly reported adverse effects included pruritus and paresthesia. Sedation related side effects such as hypoxia and apnea were rarely reported and generally mild in scope. Conclusion: Fospropofol represents an alternative sedative agent to midazolam and propofol for short-term sedation. However, the efficacy and safety of fospropofol compared to midazolam or propofol has not been assessed, therefore use will be based on practitioner preference. Restrictive package labelling requiring MAC poses medico-legal and cost-benefit issues thereby not conferring a benefit over propofol which caries similar labelling. Additional studies designed to compare fospropofol to propofol and/or midazolam for procedural sedation are needed.

564: CHARACTERIZING THE USE OF 4-FACTOR PROTHROMBIN COMPLEX CONCENTRATE

www.ccmjournal.org Critical Care Medicine • Volume 46 • Number 1 (Supplement) Learning Objectives: Off-label medication use is common in critically ill patients in the United States. Kcentra® is a 4-factor prothrombin complex concentrate (4PCC) approved for warfarin reversal in patients with acute major bleeding or need for urgent surgery. As with other procoagulants, non-approved use occurs. However, there is a paucity of data describing the real-life utilization of 4PCC. The aim of this snapshot analysis is to characterize the FDA approved and non-approved uses of 4PCC in 5 medical centers over a specified period of time. Methods: This multi-center retrospective analysis of 4PCC use in critically ill patients occurred from January 1, 2016 through December 31, 2016. All adult ICU patients who received 4PCC at a participating center were eligible for inclusion. Data included: demographics, 4PCC dose, coagulation parameters before and after 4PCC administration, blood product and hemostatic agent administration as well as incidence of thromboembolic events (TEEs). Results: A total of 104 patients from 5 institutions were included. Anticoagulant use prior to 4PCC was present in 80 patients (76.9%), and 63 (78.8%) were warfarin. The type of bleeding includes: intracranial (n = 33), trauma (n = 23), cardiac surgery (n = 12), GI (n = 12), musculoskeletal (n = 2), other (n = 13). The mean 4PCC dose was 30 IU/kg. The median INR prior to and 12–24 hours following 4PCC was 2.2 and 1.5, respectively. Fortyeight patients (46.2%) received off-label 4PCC which included: massive hemorrhage on novel oral anticoagulants (29.2%), cardiac surgery (25%), acute coagulopathy (20.8%), non-cardiac surgery (10.4%), or other (14.6%). Of 92 patients, excellent and good hemostatic efficacy was achieved in 21 and 49 patients, respectively. Of the patients who achieved excellent hemostatic efficacy, 15 received 4PCC on-label and 6 received 4PCC off-label (p < 0.05). There were 11 TEEs. Conclusions: Although the predominant use of 4PCC was on label, many patients received offlabel 4PCC during the index study period. Further study is needed to identify the relative clinical impact, risk and cost associated with real-life utilization of 4PCC.

Patient nonadherence to filling discharge medication prescriptions from the emergency department: Barriers and clinical implications

PURPOSE Barriers to and clinical implications of patient nonadherence to filling discharge medication prescriptions from the emergency department (ED) were evaluated. METHODS This was a retrospective, observational analysis of patients discharged from the ED from April 2013 through May 2015 with medication prescriptions. Patients age 18-89 years who were seen in the ED and did not retrieve discharge medication prescriptions from the onsite, 24-hour ED discharge pharmacy were included in this study. Patients who did not pick up prescriptions were called and asked about barriers to prescription filling. These charts were then retrospectively reviewed and categorized. The primary study outcome was the frequency of nonadherence to filling discharge medications prescribed during the ED visit at the ED outpatient pharmacy. Secondary outcomes included identifying barriers to medication adherence, the rate of return ED visits within 30 days of ED discharge, and the rate of 30-day hospital admissions. Associations between patient and medication variables and the rates of return ED visits within 30 days of discharge and 30-day hospital admissions were analyzed. RESULTS Of the 4,444 patients discharged from the ED with a prescription to be filled at the satellite pharmacy, 510 were nonadherent. Of these patients, 505 had complete chart information available for evaluation. A large proportion of nonadherent patients revisited the ED within 30 days of ED discharge. Multivariate logistic regression found payer class, ethnicity, and sex were independently associated with return ED visits. CONCLUSION The majority of patients who received a prescription during an ED visit filled their discharge medications. Sex, ethnicity, and payer class were independently associated with nonadherence.