Organization for Advancing Critical Care Monitoring
Critical Care Pharmacotherapy Literature Update (last updated 10/23/2012)
Bleeding After Initiation of Multiple Antithrombotic Drugs, Including Triple Therapy, in Atrial Fibrillation Patients Following Myocardial Infarction and Coronary Intervention: A Nationwide Cohort Study
BENEFICIAL ASSOCIATION OF BETABLOCKER THERAPY ON RECOVERY FROM SEVERE ACUTE HEART FAILURE TREATMENT: DATA FROM THE SURVIVAL OF PATIENTS WITH ACUTE HEART FAILURE IN NEED OF INTRAVENOUS INOTROPIC SUPPORT TRIAL
Kwiatt ME, Patel MS, Ross SE, et al. J Trauma Acute Care Surg. 2012;73:625-28.
Study Question: Can the safety of low-molecular-weight heparin (LMWH) for venous thromboembolism (VTE) prophylaxisin blunt intracranial injury be demonstrated?
Study Description: The Western Trauma Association (WTA) Multicenter Trials Committee conducted a multi-center, retrospective, cohort study which included patients who presented with intracranial hemorrhage caused by blunt injury. Patients with Abbreviated Injury Scale (AIS) score of ≥3, intracranial hemorrhage, blunt mechanism of injury, age >18 years, head CT at admission, and at least one follow-up CT scan of the head were included. Patients with previous thromboembolic disease, on pre-injury anticoagulation with warfarin or therapeutic LMWH, on heparin for VTE prophylaxis, hospitalized <48 hours, or required emergent thoracic, abdominal, or vascular surgery at admission were excluded. Patients were divided into two groups: those who received LMWH during their hospitalization (LMWH group) and those who did not (control group). The primary outcome measured was progression of intracranial hemorrhage documented by repeated head CT scan. Multivariate logistic regression analysis was performed to identify potential independent risk factors for intracranial hemorrhage progression.
Results: A total of 1,215 patients were included in the study. Of these, 220 patients (18.1%) received LMWH for VTE prophylaxis. The remaining 995 did not receive anticoagulants (81.9%) and served as the control group. Patients who received LMWH had more severe injury and lower GCS score at admission with a longer ICU and hospital length of stay. At presentation, patients in the LMWH group frequently required operation for their intracranial hemorrhages.
Ninety-three (42%) patients in the LMWH group were found to have progressive bleed on follow-up head CT scans. In the control group, 239 (24%)patients were found to have progression on follow-up CT scans (p<0.001). There was no difference in the rate of hemorrhage progression after receiving LMWH regardless of timing of LMWH initiation. Factors found to be independent risk factors for hemorrhage progression include age >55 years, male sex, INR >1.3, intra-axial hemorrhage, GCS<9, and LMWH for deep vein thrombosis prophylaxis. LMWH was found to be the strongest risk factor for intracranial hemorrhage progression (OR 2.41; 95% CI 1.65-3.53). The LMWH group had 20 episodes of VTE while the control group had 31 episodes of VTE (9.1% vs. 3.1%, p<0.001). However, only 42% LMWH patients and 11% control patients had lower-extremity duplex ultrasounds.
Conclusion(s): Patients receiving LMWH were at higher risk for hemorrhage progression. LMWH was not demonstrated to be safe for VTE prophylaxis in patients with blunt traumatic brain injury. The risk of using LMWH may exceed its benefit.
Perspective: The use of LMWH for VTE prophylaxis in patients with blunt traumatic brain injury was not demonstrated to be safe according to the authors of this study. Nevertheless, the results of this study should be interpreted with caution. There was variability observed among the participating centers with respect to use of LMWH. Additionally, the baseline characteristics between the LMWH and control groups were significantly different in that the LMWH group was more injured and more likely to require neurosurgical intervention in the first 24 hours. The indications and timing for performing repeated head CT scans for evaluation of progression were variable among the participating centers. The retrospective design of this study precludes making any definitive conclusions regarding the safety of VTE prophylaxis in patients with blunt traumatic brain injury.
Bleeding After Initiation of Multiple Antithrombotic Drugs, Including Triple Therapy, in Atrial Fibrillation Patients Following Myocardial Infarction and Coronary Intervention: A Nationwide Cohort Study
Lamberts M, Olesen JB, Ruwald MH, et al.
Study Question: What is the risk and time frame of bleeding associated with triple therapy (vitamin K antagonist (VKA) + aspirin + clopidogrel), and is there any thromboembolic benefit to this regimen?
Study Description:This study used data from nationwide registries in Denmark to identify patients with atrial fibrillation who had been hospitalized for myocardial infarction (MI) or percutaneous coronary intervention (PCI). Patients were included 7 days after discharge if they met 3 criteria: ongoing antithrombotic treatment, age ≥ 30 years, and no record of bleeding, MI, or ischemic stroke during the 7-day waiting period. Based on prescription claims, patients were classified into 5 categories of antithrombotic regimens: triple therapy; VKA + single antiplatelet; dual antiplatelet; VKA monotherapy; or single antiplatelet. The primary outcome was fatal or nonfatal bleeding. For each antithrombotic regimen, bleeding risk was determined for early (0-89 days from inclusion) and delayed (90-360 days from inclusion) time periods. The secondary outcome was occurrence of thromboembolic events: cardiovascular death, death from ischemic stroke, nonfatal MI, or nonfatal ischemic stroke.
Results:A total of 11,480 patients were included in the study; 13% received triple therapy, 16% VKA + single antiplatelet, 27% dual antiplatelet, 7% VKA monotherapy, and 36% single antiplatelet therapy. Approximately 24% of study patients met inclusion criteria following PCI with no preceding MI. Of the 76% who had an MI, 17.3% of those patients had undergone PCI within one week. Nearly 10% of patients had a history of bleeding prior to study inclusion. The average HAS-BLED score was 2.1 and the average CHADS2 score was 1.5.
Within the first 30 days of antithrombotic therapy, there were 22.6 major bleeding events/100 person-years for triple therapy, 20.3 events/100 person-years for VKA + single antiplatelet, and 14.3 events/100 person-years for dual antiplatelet therapy. Nine of the 10 fatal bleeding events were intracranial or gastrointestinal. Bleeding rates for patients on triple therapy remained higher than rates for any other regimen throughout the 360-day study period. For all regimens except VKA monotherapy, bleeding risk was highest in the first 90 days of therapy.
The secondary endpoint of cardiovascular death, MI, and ischemic stroke was observed in 2,534 (22.1%) events during the study period. The rate of this endpoint was similar among patients on triple therapy and VKA + single antiplatelet therapy.
Conclusions:Triple antithrombotic therapy with a VKA, aspirin, and clopidogel creates a high risk of bleeding after MI or PCI in patients with atrial fibrillation.
Perspective:Triple antithrombotic therapy, limited to as short a time as possible, is currently recommended for patients with atrial fibrillation who present with an MI or undergo PCI. This study suggests that the bleeding risk with triple therapy is evident even within the first 30 days of treatment. Although the risk decreases over time, it is still greater than that seen with less intense antithrombotic regimens. In addition to the bleeding risk, this study found that a triple therapy regimen does not provide thromboembolic benefit over a VKA + single antiplatelet regimen. The results of this study suggest that triple antithrombotic therapy be prescribed only after careful evaluation of bleeding risk.
Pontes-Arruda A, Cesarino dos Santos MCF, Martins LF, et al. J Parenter Enteral Nutr. 2012;36:574-86.
Study Question: Do commercially available terminally sterile total nutrient admixtures (TNA) influence the risk of blood stream infections (BSI) in critically ill patients receiving parenteral nutrition (PN)?
Study Description: This was a multicenter, prospective, open-label, controlled trial that compared the rates of BSI among patients randomized to commercially available terminally sterile multichamber bags (MCB) compared to locally compounded PN (COM). COM were further randomized to either olive oil-based fat emulsion (COM1) compared to medium-chain triglyceride/ long-chain triglyceride-based fat emulsion (COM2). BSI was defined as any positive blood culture within 28 days of PN initiation. All patients had blood cultures sampled 72 hours after the start of PN and additional cultures were collected at the physician’s discretion. Patients received PN through either a subclavian or jugular single lumen catheter exclusive for PN (periperhally inserted central lines were not used).
Results: A total of 406 patients were included; 202 in MCB, 103 in COM1, 101 in COM2. Baseline characteristics, including severity of illness and rates of malnourishment, were comparable between treatment groups. Ileus was the most common reason to start PN.
Despite similar duration of catheter days, the incidence of BSI was lower in the MCB PN group compared to COM PN (16.8% vs. 22.5%, p=0.03). The observed raw BSI rate was 35.3% higher in patients who received COM PN compared to patients who received MCB PN. A. baumannii and S. aureus were more commonly isolated in the COM group. There were no significant differences between groups for 28 day all-cause mortality, development of severe sepsis or septic shock, or ICU/hospital length of stay. Although patients provided MCB PN received nutrition therapy faster (4 hrs vs. 12 hrs, p<0.001), patients who received COM PN reached goal energy requirement quicker (2 days vs. 3 days, p<0.001) and received more calories from protein (87 g vs. 80 g, p<0.001). The type of fat emulsion didn’t appear to influence rates of BSI.
Conclusion(s): COM was associated with a higher incidence of BSI suggesting that the use of MCB PN may play a role in reducing the incidence of BSI in patients who receive PN.
Perspective: PN-associated BSI are estimated to cost $22,383 per infection in the US. If MCB PN is associated with a 5.7% absolute reduction in rates of BSI compared to COM, it can be estimated that 12,500 cases of PN-associated BSI could be avoided with an overall savings of more than $279 million per year in the United States alone. Still, the difference in rates of BSI was not associated with worse outcomes and must be balanced with the ability of COM to provide individualized nutrition support therapy.
Drenger B, Fontes ML, Miao Y, et al. Circulation. 2012;126:261-69.
Study Question: Is there an association of ACE inhibitor (ACEI) administration and on-pump coronary artery bypass graft (CABG) surgery with clinical outcomes?
Study Description: This study was a prospective multinational observational study of 4,224 patients undergoing CABG surgery (nonvalvular) treated with any dose ACEI (ARBs not included). Patients were divided into the following groups: continuation (on an ACEI pre-op and post-op), withdrawal (on an ACEIpre-op but stopped post-op), addition (not on ACEI pre-op but started post-op), and no ACEI (ACEI never started). The primary outcome was a composite of cardiac, cerebral, and renal events and in-hospital mortality.
Results: Patients in the no ACEI group had fewer comorbidities, while patients treated with ACEI prior to surgery had longer on-pump times. The continuation group had a 31% lower odds of the composite outcome (p=0.009) versus no ACEI, however the continuation group required more non-routine inotropes (28.1% vs 13.8%, p<0.001) despite having similar cardiac indices. The continuation group had a 50% lower odds of the composite outcome (p<0.001) versus the withdrawal group as well as significantly lower use of transfusion and cardiac assist devices (p<0.001). The addition group had a 44% lower odds of the composite outcome (p=0.004) versus the no ACEI group.
Conclusion: Discontinuation of ACEIfollowingCABG is associated with nonfatal in-hospital ischemic events.
Perspective: The results of this study should serve as a launching point for a large randomized controlled trial to definitively dictate appropriate ACEI (or even RAAS blocking) therapy in these patients. We can see parallels to the benefits seen with beta-blockers in this population, yet must give pause with non-protocolized postoperative management. At minimum the results should require us to analyze the current practice of discontinuation of preoperative ACEI.
Janda S, Swiston J.CHEST.2012;142:401-11.
Study Question: What is the role of fibrinolytic therapy in the management of parapneumonic effusions and empyemas?
Study Description: This systematic review and meta-analysis utilized multiple databases to identify randomized controlled trials comparing fibrinolytic therapy to placebo in the treatment of parapneumonic effusions and empyemas. Only studies involving adults (> 19 years of age) were included. Data included: study publication year, country of origin, study design, number of patients in each arm, chest tube size, type, dose, and duration of fibrinolytic given, need for thoracic surgery, death, and treatment failure.
Results: A total of 682 citations were identified. Based on title and abstract alone, 669 were excluded and 13 were retrieved for full review. One study was then excluded based on lack of a placebo arm and five because the study population was made up of mainly tuberculous pleural effusions. Seven studies were included in the final analysis yielding a total of 801 patients (384 in fibrolytic group; 417 in placebo group). The overall quality of the studies was determined to be good using the Jadad Scale and Cochrane allocation approach. Doses of fibrinolytics utilized in the studies include: streptokinase 250,000 IU daily or BID x 3 or 7 days, urokinase 100,000 IU daily x 3 days, and alteplase 10 mg BID x 3 days +/- DNase 5mg BID x 3 days. Fibrinolytic therapy was found to be beneficial for the outcome of treatment failure (need for surgical intervention or death) (RR 0.50; 95% CI 0.28-0.87) as well as to decrease the need for surgical intervention alone (RR 0.61; 95% CI 0.45-0.82). There was no difference in mean duration of hospital stay with fibrinolytic therapy compared to placebo (standard mean difference0.69; 95% CI -1.54-0.16) or a reduction in death (RR 1.14; 95% CI 0.74-1.74). A subgroup analysis of loculated pleural effusions showed a statistically significant decrease in surgical interventions with fibrinolytics (RR 0.41; 95% CI 0.26-0.65) but did not decrease overall treatment failure (RR 0.55; 95% CI 0.30-1.01). A sensitivity analysis was performed, which did not change the statistical outcomes.
Conclusion(s):Fibrinolytic therapy may be beneficial in the management of parapneumonic effusions and empyemas in the adult population, but evidence is insufficient to support routine use. There may be a role for fibrinolytics in patients with loculated pleural effusions to decrease surgical interventions.
Perspective: Despite no definitive recommendations regarding treatment, fibrinolytics have been utilized in the setting of pleural infection since the 1950s. This meta-analysis attempts to identify their place in therapy, but ultimately bases its conclusions on only 7 trials. Of note, the two most recent trials (MIST1 and MIST2) showed no benefit in terms of surgical intervention, mortality, duration of hospitalization, or radiographic improvement with fibrinolytic therapy. However, the adverse effect profile of fibrinolytics in this setting has been shown to be favorable and the subgroup analysis demonstrating benefit in the setting of loculated pleural effusions suggests they may be an option in elderly patients or non-surgical candidates. Unfortunately, further studies are still required to definitively address the impact of fibrinolytic therapy on clinical outcomes as well as the role of DNase in this setting.
Pharmacologically Dosed Oral Glutamine Reduces Myocardial Injury in Patients Undergoing Cardiac Surgery: A Randomized Pilot Feasibility Trial
Sufit A, Weitzel LB, Hamiel C, et al. J Parenter Enteral Nutr 2012;36:556-61.
Study Question: Is the administration of pharmacologically dosed preoperative oral glutamine (GLN) safe and feasible as an option to attenuate myocardial injury in cardiac surgery patients?
Study Description: This study was designed as a randomized pilot feasibility study at U.S. hospitals to determine if the use of pharmacologically dosed GLN was both safe and feasible to administer. Study inclusion criteria included undergoing elective cardiac surgery (CABG, valvular surgery, or pulmonary vein repair) in which cardiopulmonary bypass (CPB) was utilized. Notable study exclusion criteria included patients with preexisting kidney or liver dysfunction, history of HIV, hepatitis B or C or ongoing signs of myocardial ischemia. Subjects were randomized to either GLN or maltodextrincontrol (CONT) taken at home starting 3 days prior to surgery. GLN was dosed at 25 grams orally twice daily, with a final dose 2 hours prior to induction of anesthesia. Patients in the CONT group received a corresponding dose of maltodextrin. Patient compliance was assessed using daily reminder calls from the study nurse and required empty package returns. Investigators and clinical caregivers were blinded to study assignment. Myocardial injury markers (Troponin I and CKMB), plasma glutamine and heat shock protein (HSP) levels were assessed at baseline (time of consent), 6, 24, 48 and 72-hours postoperatively. Blood was also collected for HSP and plasma glutamine levels prior to anesthesia.
Results: A total of 14 patients met study inclusion criteria and were enrolled. Four patients were not included in the analysis (3 withdrawn consent [GLN] and 1 intraoperative death [CONT]) leaving a total of ten patients included in the study analysis (GLN-4, CONT-6).
Plasma troponin I levels peaked in both groups 6-hours postoperatively. Patients in the GLN group had significantly decreased troponin I levels at 24, 48 and 72-hours postoperatively, compared to patients in the control group. The GLN group also had decreased CK-MB levels at 24 and 48-hours. There were no differences noted in HSP levels throughout the study period. Plasma glutamine levels were similar between the 2 groups prior to anesthesia. Additionally, both groups demonstrated a statistically significant decrease in plasma glutamine levels postoperatively compared to prior to anesthesia.
Conclusion(s): The use of GLN is both feasible and safe when given in pharmacologic doses to preoperative cardiac surgery patients. These data support the need for a larger, definitive, randomized controlled trial of GLN therapy to reduce myocardial injury and improve clinical outcome in cardiac surgery.
Perspective: The use of pharmacologically dosed GLN preoperatively in cardiac surgery appears to be safe and feasible, perhaps paving the way for larger, randomized studies. While not designed to deliver definitive answers on the topic, the study did offer insight about the potential benefits of a pharmacologically dosed regimen of GLN in cardiac surgery patients.
Gordon AC, Wang N, KR Walley, et al. CHEST 2012; 142:593–605.
Study Question: Is there a difference in cardiac output and other measures of hemodynamics between vasopressin- and norepinephrine-treated patients with septic shock?
Study Description: This study analyzed data from the VASST trial, a multi-center, double-blind, randomized controlled trial of vasopressin vs. norepinephrine in addition to standard vasopressors for the treatment of septic shock. Between July 2001 to April 2006, patients >16 years old with septic shock (≥2 SIRS criteria with proven or suspected infection, new dysfunction of at least one organ, and hypotension despite adequate fluid resuscitation requiring vasopressor support) were randomized to receive vasopressin (0.01-0.03 units/min) or norepinephrine (5-15 mg/min). Exclusion criteria included patients with unstable coronary syndromes, severe chronic heart disease (NYHA class III and IV), and vasospastic diathesis.
Results: 779 patients were randomized and infused with the blinded study drugs. Cardiac output monitoring was used 241 (31%) patients (vasopressin, 123 patients; norepinephrine, 118 patients; P=0 .97). Patients who had a PA catheter had several markers of more severe organ dysfunction than patients who did not have a PA catheter. Vasopressin use resulted in a significant reduction in norepinephrine requirements (p<0.001). There was a rapid and significant drop in heart rate after starting the vasopressin infusion (p<0.001), which was more pronounced in the less severe shock stratum. There was also a significantly greater use of inotropic drugs in the vasopressin group compared to norepinephrine. There was no difference in cardiac index, stroke volume index, markers of oxygen delivery or left ventricular stroke work index associated with vasopressin in the whole population or in either the more or the less severe shock strata.
Conclusion(s): Vasopressin treatment in septic shock is associated with a significant reduction in heart rate but no change in cardiac output or other measures of perfusion.
Perspective: This study used data from the VASST trial to study the hemodynamic effects of vasopressin and norepinephrine therapy, but PA catheters were only used in 31% of VASST patients. Vasopressin use was not associated with a significant decrease in cardiac output compared with norepinephrine. However, there was greater use of inotropic drugs that augment cardiac output in the vasopressin group than in the norepinephrine group.
Havey TC, Fowler R, Daneman N. Crit Care. 2011;15:R267.
Study Question: What is the optimal duration of antibiotic therapy for patients with non-Staphylococcus aureus bacteremia?
Study Description: A search through MEDLINE, EMBASE, and COCHRANE databases was conducted to identify trials which randomized patients to shorter (from 5-7 days) versus longer duration (from 7-21 days) of antibiotics for bacteremia or infections that commonly cause bacteremia, including catheter-related blood stream infections (CRBSIs); intra-abdominal infections; pneumonia; pyelonephritis; and skin and soft tissue infections (SSTIs). Eligible trials randomized patients to two durations of the same antibiotic regimen to evaluate clinical cure, microbiologic cure, or survival. Trials were excluded for duration of therapy based on physician discretion, clinical improvement, or biomarker measurements.
Results: Twenty-four trials were included in the meta-analysis yielding outcome data for a total of 155 patients with positive blood cultures. One trial focused on patients with bacteremia, and the other 23 evaluated other infections causing bacteremia (0 CRBSI, 3 intra-abdominal infections, 6 pyelonephritis, 1 SSTI, and 13 pneumonia). There were no significant differences overall for clinical cure (87% vs. 96%, risk ratio 0.88, 95% CI 0.77-1.01, p = 0.37), microbiologic cure (100% vs 94%, risk ratio 1.05, 95%, CI 0.91-1.21, p = 0.78), or survival (88% vs 89.6%, risk ratio 0.97, 95% CI 0.76-1.23, p = 0.36). No significant heterogeneity between studies was detected for clinical cure (I2=5%), microbiologic cure (I2=0%), or survival (I2=3%).
Conclusion(s): The authors cautiously concluded there are no significant differences in clinical cure, microbiologic cure, or survival for shorter versus longer antibiotic therapy in the treatment of bacteremia. This indicates shorter duration of therapy may be efficacious for treatment of bacteremia, though a larger randomized trial is needed to further support of these findings. The limited number of patients represented in the outcomes suggest future trials should be completed to consider a change in antibiotic duration in bacteremic patients.
Perspective: This article emphasizes the lack of evidence that exists for optimal duration of antibiotic treatment of various illnesses. The possibility of shorter antibiotic durations for bacteremia seems promising, but it is important to keep in mind that this meta-analysis only included one trial focused specifically on bacteremic patients.
Moreno R, Sprung CL, Annane D, et al. Int Care Med. 2011;37:1765-72.
Study Question: Does hydrocortisone (HC) therapy in patients with severe sepsis and septic shock lead to faster resolution of organ failure when compared to placebo?
Study Description: This is a pre-defined secondary analysis of the previously published CORTICUS trial, a prospective, multicenter, randomized, double-blind placebo-controlled trial comparing HC for 11 days to placebo. (The primary outcome of the CORTICUS study was 28-day mortality.) The Sequential Organ Failure Assessment (SOFA) score defined organ failure according to each of six organ systems. A score of 3 or 4 points in any one system was defined as failure and achievement of reversal occurred if the patient achieved a score of < 3 after treatment.
Results: The study included 499 participants who were enrolled within 72 hours of septic shock diagnosis. The groups did not differ in baseline total SOFA scores (HC 10.8 ± 3.2 vs placebo 10.8 ± 3.1; p = 0.55) nor in any of individual components of the score. Serial SOFA assessments from day 0 to day 7 demonstrated a faster decrease in the cardiovascular (p < 0.0001) and hepatic (p < 0.0001) organ failure scores favoring the hydrocortisone group. Significant deviation between treatment groups appeared to begin at approximately day 3. There were no other significant differences between groups for other organ systems in the first seven days. Over the course of 28 days, however, cardiovascular failure resolution (p = 0.043) and renal failure resolution (p = 0.039) both favored hydrocortisone, whereas liver failure resolution proved no different (p = 0.42).
Conclusion(s): Despite an inability to demonstrate improvements in 28-day mortality, the study suggests morbidity improvements in the hydrocortisone group, particularly with respect to cardiovascular failure.
Perspective: Selection of the most optimal endpoint for critical care studies is challenging, but these findings suggest some value in continuing the approach of conducting post hoc analyses. Unfortunately, the practical applications of the approach are less clear. To what extent can the findings be applied in care and to what extent should they be relegated to hypothesis generation? Critical care practitioners are unfortunately left to determine where priority should be placed.
Kutsogiannis J, Alberda C, Gramlich L, et al. Crit Care Med. 2011;39:2691-9.
Study Question: Does the early use of supplemental parenteral nutrition (PN) have any impact on clinical outcomes in critically ill patients?
Methods: Data from two international, prospective, observational studies conducted in 2007 and 2008 were combined to provide a robust sample. Over 225 intensive care units (ICUs) from 29 countries were included, and eligible patients were those who: were receiving mechanical ventilation (MV); remained in the ICU for greater than 72 hours; and received early enteral nutrition (EN), which was defined as being initiated within 48 hours of admission.
Results: Two thousand nine hundred twenty patients were included: 2,562 (87.7%) received early EN alone, 188 (6.4%) received early supplemental PN, and 170 (5.8%) received late supplemental PN. Patients receiving PN were more likely to be surgical patients, to have acute respiratory distress syndrome, and to have been in the hospital greater than 1 day prior to ICU admission when compared to patients receiving early EN alone. Among patients receiving PN, those receiving early therapy were significantly older and had lower BMIs, while those receiving late therapy were more likely to have experienced GI dysfunction for 2 or more days. Adequacy of calories and protein was highest in the early PN group (81.2% and 80.1%, respectively) and lowest in the early EN group (63.4% and 59.3%, p < 0.001). Sixty-day mortality was significantly lower in patients receiving early EN (27.8%) compared to those receiving early PN (34.6%) or late PN (35.3%, p = 0.02). Additionally, patients receiving early EN had significantly fewer days on MV and shorter ICU and hospital lengths of stay. The rate of discharge alive from the hospital was lower among patients receiving early PN and late PN compared to those receiving early EN. No change in outcomes was demonstrated by adjusting for confounders.
Conclusions: Supplemental PN was associated with improved protein and calorie provisions but did not demonstrate a clinical benefit.
Perspective: These findings are consistent with five randomized, controlled trials comparing EN alone to EN with supplemental PN, as well as additional observation and retrospective studies. Given this data and known complications of PN, the routine use of supplemental PN in critically ill patients cannot be recommended at this time.
Devabhakthuni S, Pajoumand M, Williams C, et al. J Trauma. 2011;71:1164-1171.
Study Question: What is the comparative safety and outcomes of prolonged infusion of standard dose dexmedetomidine (SDD, ≤ 0.7 mcg/kg/hr), high dose dexmedetomidine (HDD, > 0.7 mcg/kg/hr), and propofol in trauma patients receiving MV?
Study Description: This retrospective, single-center cohort study (n = 127) included adult trauma patients admitted to the ICU who received the study drug(s). Patients were excluded if they received the study drug(s) for less than 24 hours, received dexmedetomidine with propofol for greater than 6 hours, or were not receiving MV. The primary endpoints were the occurrence of hypotension, hypertension, bradycardia, and tachycardia using SEDCOM criteria. Secondary endpoints were overall mortality, ICU or hospital LOS, duration of MV, and concomitant administration of analgesics, sedatives, and antipsychotics.
Results: Baseline characteristics were matched overall with a few notable differences: more patients who received dexmedetomidine had thoracic injury, spinal cord injury and/or hepatic impairment. Duration of therapy was longer in the HDD arm (6 vs 2 vs 3 days, p < 0.001). Injury severity score was similar across groups. Patients who received HDD experienced more hypotension than those who received SDD or propofol (98% vs 86% vs 78%, respectively, p = 0.02). There was no difference in the incidences hypertension, bradycardia, or tachycardia. Both the HDD and SDD groups had longer ICU and hospital LOS, though no mortality difference was observed. Oxycodone, midazolam, and haloperidol administration was greatest in the HDD arm.
Conclusion(s): HDD may be associated with a higher incidence of hypotension, longer LOS, and concomitant sedative, analgesic and antipsychotic use in critically ill trauma patients receiving MV.
Perspective: Limitations include a heterogenous trauma population, lack of standardized sedation goals or a routine assessment of sedation, and the inability to control for provider preference with concomitant medication prescribing. Regardless, HDD likely increases risk of hypotension, particularly in the hypovolemic trauma patient.
Rice TW, Wheeler AP, Thompson BT, et al. JAMA. 2011;306:1574-81.
Study Question: Does supplementation of omega-3 fatty acid, γ-linolenic acid (GLA), and antioxidants increase ventilator-free days in patients with acute lung injury (ALI)?
Methods: This study was a randomized, double-blind, placebo-controlled trial was conducted at 44 hospitals of the ARDS Clinical Trials Network between January 2008 and February 2009. Eligible patients were those: receiving mechanical ventilation; who had bilateral pulmonary infiltrates; who had a PaO2-to-FiO2 ratio < 300; who had no clinical evidence of left atrial hypertension; and in whom the treating physician intended to initiate enteral nutrition. Patients were stratified by whether they had shock at baseline and randomized to twice-daily enteral supplementation of omega-3 fatty acids, GLA, and antioxidants or an isocaloric-isovolemic carbohydrate-rich control until the earliest of 21 days, 48 hours of unassisted breathing, or extubation. All patients were managed using lung-protective ventilation and fluid-conservative hemodynamic management protocols.
Results: Two-hundred seventy-two patients were enrolled, with roughly half randomized to the omega-3 supplement group and half to the control group. The study was stopped for futility after the first interim analysis when the observed ventilator-free day (VFD) difference was unfavorable for omega-3 supplementation by -3.2 days. With respect to the primary outcome, the omega-3 supplement group had significantly fewer VFDs (14 vs. 17.2 days; 95% CI -5.8 to -0.7, p = 0.02) compared to patients receiving placebo. In addition, patients receiving omega-3 supplementation had fewer ICU-free days (14 vs. 16.7 days, p = 0.04). The adjusted mortality prior to day 60 or hospital discharge did not differ significantly between the groups (25.1% omega-3 supplement vs. 17.6% control, p = 0.11). Patients receiving omega-3 supplementation experienced more diarrhea compared with controls (28.7% vs. 20.9%, p = 0.001), but other adverse effects and rates of development of new infections did not differ between groups.
Conclusions: Twice-daily enteral supplementation of omega-3 fatty acids, GLA, and antioxidants does not improve clinical outcomes and may actually be harmful in patients with ALI.
Perspective: The current study was powered to detect differences in clinical outcomes, unlike previous studies. It also addresses several limitations of prior studies, including the lack of intention-to-treat analysis, use of a non-standard “pulmonary” control formula, and lack of control for evidence-based therapeutic interventions known to provide benefit in ALI. Of note, the higher protein provision of the control could potentially account for some benefit compared to the omega-3 supplement.
Makris D, Manoulakas E, Komnos A, et al. Crit Care Med. 2011;39:2440-2446.
Study Question: Could the addition of statin therapy reduce the frequency of ventilator-associated pneumonia (VAP) in statin-naïve, critically ill patients?
Study Description: This study was an open-label, randomized, controlled trial that enrolled adult patients receiving mechanical ventilation > 48 hours. Exclusion criteria included: previous statin therapy or concomitant medications known to interact with pravastatin; elevated serum creatine kinase; malabsorption; active pneumonia; and pregnancy. Patients were randomized to receive either pravastatin 40 mg daily or placebo. The frequency of VAP was assessed during the 30-day treatment period.
Results: A total of 152 patients were randomized. There was no statistically significant difference in VAP frequency found between the pravastatin and placebo group (22.5% vs 34.5%, p = 0.11). No differences were detected in regards to VAP frequency at ICU discharge or 30-day mortality. However, the overall ICU mortality was reduced in the pravastatin arm (14.1% vs 29.1%, p = 0.03). A subgroup analysis of patients with APACHE scores ³ 15 had a lower incidence of VAP during their ICU stay (p = 0.04) and increased probability of survival during the 30-day treatment period (p = 0.04) in the pravastatin group.
Conclusion(s): Although statins were not beneficial in reducing VAP rates in the overall patient sample, they may have a beneficial effect on ICU mortality in patients who have more severe illness on admission.
Perspective: The pleiotropic effects of statins, including potential anti-inflammatory and immunomodulatory characteristics, make this drug class theoretically appealing in the prevention or treatment of infectious and inflammatory processes. Larger, blinded studies with a better-defined patient population are necessary to determine if adjunctive statin therapy can affect VAP rates in the critically ill.
Schneider EB, Efron DT, MacKenzie EJ, et al. J Trauma. 2011;71:815-9.
Study Question: Is pre-injury statin use associated with improved survival and functional outcome in adults > 65 years of age who experience moderate or severe head trauma?
Study Description: This multicenter, retrospective study of patients with traumatic brain injury (TBI) and Abbreviated Injury Score of > 3 and age > 65 years of age were included. Exclusion criteria included presentation of fixed dilated pupils and patients who died within 24 hours of admission.
Results: A total of 523 met inclusion criteria. Statin users (22%) were more likely to: have a cardiovascular comorbidity or risk equivalent such as hypertension, heart failure, or diabetes; taking more medications pre-injury; and receiving beta-blocker therapy. Statin users were had a 76% lower adjusted risk (RR 0.24; 95% CI 0.08-0.69) of hospital mortality. There was no difference in Extended Glasgow Outcome Scale scores at 3 months. However at 12 months, statin users had a 13% higher likelihood of good recovery (RR 1.13; 95% CI 1.01-.26). The presence of cardiovascular comorbidity obviated the protective effect.
Conclusion(s): The study’s authors conclude that premorbid statin use in older adults and TBI is associated with reduced risk of hospital death and improved functional recovery at 12 months. Statin users with cardiovascular comorbidity did not derive benefit.
Perspective: The benefits of premorbid statin use have been suggested in ischemic stroke, sepsis, and trauma. However, the benefit of stain use post-subarachnoid hemorrhage has not been conclusively demonstrated. Continuation of statin use seems reasonable post injury. It would be desirable to see a larger randomized study examining the benefit of continuing or initiating statin use post-TBI.
Devlin J, Skrobik Y, Riker R, et al. Crit Care. 2011. 15:R215.
Study Question: Does antipsychotic therapy with quetiapine affect the duration and time to first resolution of individual delirium symptoms?
Study Description: In this post-hoc analysis, delirium symptoms evaluated by the Intensive Care Delirium Screening Checklist (ICDSC) were compared between groups randomized to receive either 50 to 200 mg q12h quetiapine or placebo. Eighteen patients with delirium (ICDSC ≥ 4) were enrolled in each group and delirium assessments were conducted every eight to 12 hours. Both groups received as-needed intravenous haloperidol. Baseline symptoms, symptom resolution, and time to first symptom resolution were compared between groups. P-values of ≤ 0.010 was considered statistically significant.
Results: Data was available for 29 of 36 patients. There were no differences between individual symptoms at baseline between the quetiapine and placebo group. There were a higher proportion of patients with resolution of symptom fluctuation in the quetiapine group (p = 0.009) and these patients had shorter time to first resolution of symptom fluctuation (4 h vs. 14 h; p = 0.004) and disorientation (48 h vs. 204 h; p = 0.10). Patients in the quetiapine group had longer time to first resolution of agitation (84 h vs. 36 h; p = 0.07) and hyperactivity (120 h vs. 24 h; p = 0.04). Quetiapine-treated patients tended to spend a smaller percentage of time with inattention (47% vs. 78%; p = 0.03), hallucinations (0% vs. 28%; p = 0.10), and symptom fluctuation (47% vs. 89%; p = 0.04).
Conclusion(s): The authors conclude that quetiapine use may be associated with more rapid resolution of many common ICU delirium symptoms compared to placebo; however, agitation and hyperactivity took longer to resolve with quetiapine.
Perspective: This analysis is perhaps the first to measure the response of individual delirium symptoms to antipsychotic therapy in this setting. The authors did find that quetiapine is associated with a shorter time to resolution of symptom fluctuation and disorientation. Nonetheless, patients taking quetiapine had longer time to resolution of hyperactive delirium symptoms. In the trial’s initial publication, patients in the placebo group trended toward receiving more haloperidol than in the quetiapine group; the extent to which that trend impacts this post-hoc analysis’ findings is unclear.
Spencer DD, Jacobi J, Juenke JM, et al. Pharmacotherapy. 2011;31:934-41.
Study Question: What is the optimal dosing regimen, maintaining a serum concentration within 6 and 20 mcg/mL, of levetiracetam for seizure prophylaxis following subarachnoid hemorrhage, subdural hematoma, or traumatic brain injury?
Study Description: This was a single center, prospective, open-label, steady-state pharmacokinetic study. Patients admitted to adult neurocritical care unit received intravenous levetiracetam therapy for seizure prophylaxis following hemorrhagic stroke, subarachnoid hemorrhage, or traumatic brain injury. Doses were initiated at 500 mg every 12 hours and serum levels were measured at specified times after at least 4 doses had been administered. Analysis of samples and pharmacokinetics parameters were described in detail. Monte Carlo simulations were used for multiple dosing regimens including 500, 1000, 1500, and 2000 mg IV every 12 hours as well as 500, 1000, and 1500 mg IV every 8 hours.
Results: A total of 12 patients were included in the study. Indications for IV levetiracetam included subarachnoid hemorrhage (n=10), subdural hematoma (n=1), and traumatic brain injury (n=1). None of the patients experienced a seizure during the study. The probabilities of the dosing regimens achieving trough concentrations within the suggested range were as follows: the highest probability was 1000 mg IV every 8 hours (57.1%), followed by 2000 mg every 12 hours (51.2%), 1500 mg every 12 hours (51.1%), and 500 mg every 8 hours (48.1%).
Conclusion(s): The investigators concluded that the pharmacokinetics of levetiracetam are altered in neurocritically ill patients compared to healthy volunteers. These changes suggest that higher and/or more frequent dosing may be needed to achieve a goal serum concentration between 6 and 20 mcg/mL.
Perspective: The relationship of serum concentrations and clinical efficacy has not been established, although a goal concentration of 6 – 20 mcg/mL has been suggested. It is important to note that none of the patients in the study seized while receiving a dose of 500 mg IV q 12 hours.
Viasus, D, Pano-Pardo JR et al. Chest. 2011;140:1025-32.
Study Question: Does the timing of oseltamivir administration affect outcomes of patients hospitalized with confirmed influenza A (H1N1) viral infection?
Study Description: This was a multi-center, prospective observational trial of patients with confirmed H1N1 infection receiving oseltamivir at any time following hospital admission. Patients were divided into four groups based on timing of administration. Outcomes of interest included: duration of fever, hospital length of stay (LOS), need for mechanical ventilation, and hospital mortality. Neither ICU admission criteria nor treatment decisions were standardized.
Results: The study included 538 patients who were: mostly male; a median of 39 years old; and had no previous H1N1 vaccination or exposure to oseltamivir. Concomitant steroids were used in 28.3% of the population. The median duration of oseltamivir therapy was 5 days, and the study’s authors found that each one-day delay in initiating oseltamivir was associated with: increased duration of fever (OR 1.10; 95% CI, 1.02-1.19) and LOS (OR 1.07; 95% CI, 1.00-1.15) beyond the sample medians; a need for mechanical ventilation (statistical significance not maintained on multivariate analysis); and hospital mortality (OR, 1.2; 95% CI, 1.06-1.35). Similar results were found on the aforementioned subgroup analysis.
Conclusion(s): The study’s authors conclude that delays in oseltamivir administration following symptoms of influenza are associated with increased duration of fever, LOS, and hospital mortality.
Perspective: While the idea that timing of therapy is directly related to outcomes is not new in infectious diseases, this study provides further support for and underscores the necessity of timely administration of antiviral drugs in H1N1 influenza infection.
Oudemans-van Straaten HM, Endemen H, et al. Crit Care. 2011;15:R240.
Study Question: Is enterally administered tobramycin for selective decontamination of the gut significantly absorbed systemically and, if so, is there a clinically significant effect on the kidneys?
Study Description: This prospective, observational cohort study was conducted in a 20-bed general ICU in the Netherlands where, per hospital standard, all patients with an expected ICU length of stay > 2 days receive SDD with an enteral solution q6h and an oral paste, each containing polymixin E, tobramycin, and amphotericin B. Patients had blood and urine tobramycin measured during the first day of ICU admission only (day 1). The primary endpoint of the study was the proportion of patients with tobramycin leakage to systemic circulation (at least one serum tobramaycin concentration ≥ 0.050 mg/L) vs. no tobramycin leakage. Secondary endpoints included the concentration of tobramycin in serum and urine, and the relation between tobramycin leakage and markers of circulation, renal, and other organ function.
Results: A total of 105 patients were included, 83% had at least one detectable tobramycin concentration. The median highest serum concentration per patient was 0.120 (IQR 0.063-0.232) mg/L. Ninety-nine percent of patients had at least one urine sample with detectable tobramycin, and 49% had a urine concentration that was above the therapeutic trough level (> 1mg/L). The highest tobramycin concentration correlated positively with CRP, urea, and bilirubin on admission; and to highest dopamine dose, noradrenalin dose, urinary output on day 1, and RIFLE based on creatinine during the study period (p = 0.03). The highest tobramycin concentration was significantly higher in patients without AKI compared to those with any degree of AKI according to RIFLE and was not significantly related to total SOFA or APACHE scores. The differences in markers of organ failure were most pronounced on day of ICU admission.
Conclusion(s): The study authors concluded that the high percentage of patients with detectable serum tobramycin suggests gut barrier failure. At risk patients should have serum concentrations monitored during prolonged SDD use to prevent possible toxicity.
Perspective: While the direct clinical utility of this study is limited in geographical areas where SDD is not a widespread standard of care, it does provides further evidence that gut permeability is altered during the course of critical illness.
Skolnick BE, Shenouda M, Khutoryansky NM, et al. Anesth Analg. 2011;113:703-10.
Study Question: Can recombinant factor VIIa (rFVIIa) be used to mitigate spontaneous or traumatic bleeding in clopidogrel-treated patients?
Study Description: This study was a single-center, placebo-controlled, double-blind, dose escalating trial that randomized 40 healthy participants to receive IV placebo or one of five rFVIIa doses (5, 10, 20, 40, or 80 mcg/kg). Participants were given clopidogrel 300 mg PO x 1 then 75 mg PO Daily for 2 days prior to rVIIa administration. The study’s authors employed a method called the punch biopsy that is meant to measure bleeding duration (BD) and blood volume (BV) in healthy participants. Researchers performed the biopsy at 4 different time periods: prior to clopidogrel administration; four days after initiating clopidogrel; two and a quarter hours after study drug (rFVIIa or placebo) administration; and five hours after study drug administration. A platelet inhibition assay measured response to clopidogrel, and clot dynamics were assessed using thromboelastography (TEG).
Results: The study was halted when, midway through, the sponsor (NovoNordisk A/S) decided to allocate resources to studies evaluating rVIIa use only for FDA-approved indications. Data was collected for patients who received placebo or 5, 10, or 20 mcg/kg of rFVIIa. Treatment with rFVIIa had no effect on the primary endpoint of a reduction in BD. Compared to placebo, rVIIa 10 and 20 mcg/kg doses significantly reduced BV. TEG evaluation showed a significant reduction in time to clot onset and an increase in clot angle for all patients who received rFVIIa compared to placebo. No other TEG assessments were different between groups. There were no thromboembolic complications and all side effects were considered mild or moderate.
Conclusion(s): The study’s authors conclude that 10 and 20 mcg/kg rFVIIa may improve BD and BV in patients treated with clopidogrel.
Perspective: There important limitations – both shortfalls in the study’s internal validity and barriers to its generalizeability. First, the authors noted larger variation in BV and BD in patients who received punch biopsies by one of the two physicians, suggesting problems in methods standardization. They also did attempt to account for this difference statistically and any additional detail (e.g., values) beyond a single statement. The study excluded patients who were on additional antiplatelet therapy, limiting its widespread applicability; as the authors themselves note, it is not uncommon for patients to receive dual antiplatelet therapy with aspirin and clopidogrel. Overall, the utilization of rFVIIa for a reduction in BD was not very convincing at the doses studied. The results might have been significantly different if investigators had evaluated the effect of the 40 or 80 mcg/kg doses of rFVIIa, but the trial was halted before those doses were evaluated.
Chung SB, Lee SH, Kim ES, Eoh W. J Trauma. 2011;71:867-71.
Study Question: What is the incidence of deep vein thrombosis (DVT) and the effect of mechanical prophylaxis in patients with spinal cord injury (SCI)?
Study Description: This was a single-center, observational study of consecutive SCI patients admitted over a one-year period for either acute (within 1 week of injury) traumatic or non-traumatic SCI. Patients were included if they had no contraindications to mechanical prophylaxis and were excluded if they had: multi-organ injury; known abnormality of the coagulation system; or a history of anticoagulation management. In all patients, mechanical prophylaxis for thromboembolic events entailed a combination of: gradient elastic stockings; external, sequential pneumatic compression applied to lower limbs; and early mobilization within 30 days. Pharmacologic prophylaxis was not used. Patients were routinely checked for DVT with color Doppler ultrasonography.
Results: Sixteen (43%) of the total sample of 37 were found to have a DVT between two and 49 days’ (mean 17.2 days’) follow up: ten patients showed new thrombosis within 7 days; 3 patients after 2-3 weeks; 3 patients more than 1 month after injury. Two patients had symptomatic pulmonary embolism. A total of 10 of 17 (59%) patients with complete motor palsy (American Spinal Injury Association Impairment Classification [ASIA] A or B) and 6 of 20 (30%) with incomplete motor palsy (ASIA C or D) had DVT (p = 0.078). There were eight DVTs each in the traumatic and non-traumatic groups (p = 0.886).
Conclusion(s): These findings are different than some past studies showing lower rates of DVT in SCI patients in a similar geographic region of Southeast Asia. This study supports the western literature regarding the high rate of DVT in SCI patients. Mechanical prophylaxis for DVT without medical treatment may be inadequate for preventing DVT. The authors recommend using anticoagulant prophylaxis for DVT prophylaxis in SCI unless there is a contraindication.
Perspective: Although this is a small study, it does provide important data. Patients with SCI without anticoagulant prophylaxis had a very high rate of DVT, and most DVTs in this study were found early after injury. Patients with more severe ASIA grades trended toward a higher rate of DVT. The results from this study provide further support that mechanical prophylaxis alone, regardless of whether multiple types are used in combination, is inadequate in this patient population and that early pharmacologic prophylaxis is warranted.
Briguori C, Visconti G, Focaccio A, et al. Circulation. 2011;124:1260-9.
Study Question: Does a device that ensures a balance of both high urine output and venous fluid repletion – together with an IV antioxidant – lower the risk of contrast-induced acute kidney injury (CI-AKI)?
Study Description: This REMEDIAL-II study was a multicenter, randomized controlled trial that enrolled 294 non-dialysis patients scheduled to undergo angiography or angioplasty who had baseline chronic kidney disease (a glomerular filtration rate of ≤ 30 mL - min-1 - 1.73 m-2 as estimated by a Modification of Diet in Renal Disease formula) and a risk of CI-AKI in excess of 26% as determined by a confluence of weighted risk factors, including: hypotension; presence of an intra-aortic balloon pump; heart failure; age > 75 years; diabetes mellitus; anemia; and contrast media (CM) volume. Investigators assigned participants in a 1:1 ratio to one of two groups. The active treatment group patients received IV normal saline, N-acetylcysteine (NAC), and furosemide titrated to a urine output of ≥ 300 mL/h by a closed-loop, fluid-management system known as RenalGuard; patients in the control group received 154 mEq/L of sodium bicarbonate in dextrose and water and PO NAC, both in a manner consistent with standard practice, plus 1,200 mg NAC IV during the procedure.
Results: The incidence of the primary endpoint of CI-AKI (need for dialysis or a rise in serum creatinine ≥ 0.3 mg/dL 48 h after CM administration) was nearly halved in the active treatment group (11% vs. 20.5%; odds ratio 0.44; 95% confidence interval 0.24 to 0.92). One of the most important secondary outcomes was the rate of in-hospital dialysis, which trended lower in the active therapy group (0.7% vs. 4.1%; p = 0.056). Increases in levels of serum cystatin C, a kidney function biomarker, were lower in the active treatment group as well (p = 0.001), indicating more favorable renal findings.
Conclusion(s): The study’s authors conclude that their active treatment strategy of forced diuresis and repletion with saline and IV NAC is superior to conventional antioxidative CI-AKI prophylactic therapy in high-risk patients undergoing angiography or angioplasty with iodinated CM.
Perspective: The thought behind this strategy is that increasing renal fluid transit while maintaining relative systemic euvolemia ramps up elimination of CM, thereby reducing the net uptake by renal tubular cells. The strategy appears promising and is theoretically sound. As suggested in an accompanying editorial by McCullough et al., desirable future directions include: replication in a larger, independent sample and measurement or estimation of remaining CM in the kidneys and the amount excreted in the urine. In the mean time, it is worth bearing in mind that the decision to employ a strategy of forced diuresis necessitates balancing its accompanying risks of adverse outcomes, including pulmonary edema and electrolyte imbalances.
ACT Investigators; Circulation 2011, 124:1250-1259:
Study Question: Is acetylcysteine effective in preventing contrast-induced nephropathy (CIN)?
Study Description: This was a multicenter randomized, placebo controlled trial of patients receiving contrast for angiography with at least one risk factor for CIN. Patients were allocated to placebo (including similar smell to study drug) or acetylcysteine 1200mg orally every 12 hours, two doses prior to and two doses after contrast. The primary outcome was a 25% elevation of serum creatinine between 48 and 96 hours after angiography.
Results: A total of 2308 patients were randomized, with no difference found in the primary outcome (12.7% and 12.7%, p=0.97), and no differences detected in all secondary outcomes.
Conclusion(s): Acetylcysteine does not reduce the risk of CIN in at-risk patients undergoing angiography.
Perspective: Being the largest and most methodologically sound study on this subject to date, the controversy may be finally laid to rest. Important limitations include a lack of North American study sites (all Brazilian), use of serum creatinine as the primary outcome marker, and fluid therapy that was not controlled, albeit similar between groups.
Matthay MA, Brower RG, Carson S, et al. The National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. Am J Respir Crit Care Med 2011;184:561-8.
Study Question: Does β-agonist therapy accelerate the resolution of alveolar edema and improve clinical outcomes in patients with acute lung injury (ALI) and acute respiratory distress syndrome (ARDS)?
Methods: A prospective, randomized, placebo-controlled trial was conducted at 33 hospitals of the ARDS Clinical Trials Network between August 2007 and July 2008. Eligible patients were those receiving mechanical ventilation who had bilateral pulmonary infiltrates, a PaO2 to FiO2 ratio ≤ 300, and no clinical evidence of left atrial hypertension. Patients were randomized to 5mg aerosolized albuterol sulfate or saline placebo every 4 hours for 10 days or 24 hours after extubation, whichever was shorter. Dose reduction to 2.5mg or study drug discontinuation occurred in patients experiencing tachycardia beyond pre-specific age-specific limits.
Results: 282 patients were randomized: 152 to the albuterol group and 130 to the saline group. The study was stopped for futility after the second interim analysis when the observed ventilator-free day (VFD) difference was unfavorable for albuterol by -2.2 days. There was no difference between the albuterol and saline groups with respect to the primary outcome of VFDs to Day 28 (14.4 vs. 16.6 days, respectively; 95% CI -4.7-0.3, p=0.087). Patients receiving albuterol had more intensive care unit-free days compared to saline (13.5 vs. 16.2 days, p=0.023); however, there were no significant difference between the groups with regard to mortality before hospital discharge to Days 60 and 90 and the number of organ failure-free days. Heart rates were higher in patients receiving albuterol, but new onset atrial fibrillation and other dysrhythmias were not significantly different.
Conclusions: Aerosolized β2-agonist therapy does not improve clinical outcomes in mechanically ventilated patients with ALI or ARDS. Perspective: Due to the study being stopped early for futility and the fact that no significant differences were demonstrated between the groups with respect to ventilator-free days, organ failure-free days, or mortality, the routine use of aerosolized albuterol therapy for the treatment of ALI cannot be recommended.
Visschers RGJ, Damink SW, Gehlen JM, Winkens B, Soeters PB, vanGemert WG. JPEN J Parenter Enteral Nutr. 2011;35:610-5.
Study Question: Is there a beneficial effect of short term intravenous lipid (IL) withdrawal from the parenteral nutrition (PN) regimen in patients developing hypertriglyceridemia?
Study Description: The investigators retrospectively studied patients from 2002 to 2006 who required PN and had hypertriglyceridemia defined as a triglyceride (TG) level above 450 g/dL. The primary endpoint of the study was TG concentration following IL withdrawal.
Results: Of 961 patients receiving PN over the study period 73 (7.6%) met the study criteria, of which 16 were excluded for elevated TG levels prior to PN initiation. Following other exclusions only 40 patients remained for the analysis. Patients remained on PN for a median of 30 days and were fully dependant on PN for the duration of the study. Patients had a median (range) of 5 (1-23) days without IL. A statistically significant reduction in TG level (approximately 44%; p<0.001) was observed following IL withdrawal. Patients with a reintroduction of IL had repeat elevations in TG level requiring withdrawal of IL for the remainder of the PN period. Among secondary measures worth noting, following IL discontinuation there was a significant decrease in aspartate aminotransferase and leukocyte count, and an increase in serum albumin.
Conclusion(s): Acutely ill patients receiving PN who develop hypertriglyceridemia exhibit an improvement in laboratory parameters following IL withdrawal.
Perspective: This small retrospective study offers some evidence as to the beneficial effect of IL withdrawal following development of hypertriglyceridemia in patients on short term (ie, not life long) PN. While a reduction in TG level was not unexpected, the observed recurrence of TG rise in patients resuming IL may be of some concern. There was a slight positive effect on a few markers of inflammation with IL withdrawal. Unfortunately, this study did not evaluate other options for managing hypertriglyceridemia in this patient population.
Sandiumenge A, et al. CHEST 2011; 140(3):643-651.
Study Question: Does antimicrobial diversity affect resistance of Enterococcus faecium, Staphylococcus aureus, Klebsiella species, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter species (ESKAPE) pathogens in ventilator-associated pneumonia (VAP).
Study Description: Single center, prospective, interventional study of adult patients admitted to a medical-surgical ICU. Excluded patients were those admitted for less than 48 hours, immunocompromised, or less than 18 years of age. Only documented microbiological VAPs were included. Included patients then received three different antimicrobial strategies for empirical treatment of VAP; a 10 month patient-specific period, a 24 month scheduling period composed of prioritization (12 months) & restriction (12 months) periods, and a 10 month mixing period. Antibiotic prescription patterns were then determined to be homogeneous or diverse according to the antibiotic heterogeneity index (AHI) for each period.
Results: 127 documented microbiological VAPs occurred in 119 patients. Patient-specific (AHI, 0.88) and mixing (AHI, 0.87) periods were significantly more antimicrobial diverse than the scheduling (AHI, 0.65) period (p<0.01). No difference was observed in the incidence of ESKAPE organisms between the three periods but resistant ESKAPE strains significantly increased in the scheduling period compared to the patient-specific period (RR, 2.67; 95% CI, 1.01-7.08) and the mixing period (RR, 3.84; 95% CI, 1.55-12.9). Mortality of patients with resistant ESKAPE VAP was significantly higher compared to patients without ESKAPE VAP (RR, 2.25; 95% CI, 1.67-9.48).
Conclusion(s): Implementation of antimicrobial diversity for empiric VAP treatment may help prevent emergence of resistant ESKAPE organisms.
Perspective: This study suggests using antimicrobial strategies that are more homogeneous than heterogeneous as defined by AHI promote the rise of VAP caused by resistance ESKAPE. The rise in resistance may be due to the selective antimicrobial pressure caused by homogeneous strategies. Practitioners should remain cognizant of patient specific co-morbidities and previous antibiotic use when determining appropriate empirical therapy.
Lahtinen P, Pitkanen O, Polonen P, Turpeinen A, Kiviniemi V, Uusaro A. Crit Care Med. 2011;39:2263-2270.
Study Question: Does levosimendan decrease the incidence of heart failure after cardiopulmonary bypass (CPB) in postoperative cardiac surgery patients?
Methods: A prospective, single-center study was performed in adult patients undergoing elective heart valve or combined heart valve/coronary artery bypass graft (CABG) surgery requiring CPB. Patients were randomized to receive either levosimendan (bolus: 24 mcg/kg IV over 30 minutes, infusion: 0.2 mcg/kg/min x 24 hours) or placebo. The primary outcome was incidence of heart failure. Secondary outcomes included in-hospital and 6-month mortality in addition to major organ failure.
Results: The incidence of heart failure, defined as CI < 2 L/min/m2 or inotrope administration within 2 hours postoperatively after CPB, was statistically significant in favor of the levosimendan group (levosimendan: 15% vs placebo: 58%, p<0.001). There were no significant differences in secondary outcomes between cohorts. Levosimendan-induced hypotension occurred frequently in the study with 82% of patients requiring norepinephrine postoperatively compared to 51% in the placebo group (p<0.001).
Conclusions: Levosimendan may be a useful option for patients undergoing high-risk cardiac surgery if trouble weaning from CPB is anticipated.
Perspective: Levosimendan is a unique agent with calcium sensitizing properties and inotropic/vasodilatory effects. Although theoretical advantages may exist with levosimendan based on its mechanism of action, this study was not powered to show a difference in patient outcome data between groups. The use of an arbitrary definition of heart failure, a small sample size and single center design are limitations of the current study.
Jensen JU, Hein L, Lundgren B, et al. Crit Care. 2011;39:2048-58.
Study Question: Does daily monitoring of procalcitonin (PCT) improve management and outcomes of infections?
Study Description: This study (“PASS”) was a multi-center, open-label randomized controlled trial (RCT) conducted in medical/surgical ICUs in Danish academic, tertiary care hospitals. The study intervention consisted of using daily PCT levels to guide escalation of monitoring and antimicrobial coverage in an algorithm/protocol for management of infections in the ICU.
Results: There was no difference in the primary endpoints of 28-day mortality 31.5% (190 of 604) in the PCT arm compared to 32.0% (191 of 596) of patients in the standard-of-care-only arm. Inclusion of PCT levels (vs. standard algorithm) resulted in significantly greater length of MV (65.5% vs. 60.7% of days), median ICU LOS (6 vs. 5 days), median antimicrobial exposure (6 vs. 4 days), and total number of cultures performed (7,874 vs. 6,641), with no apparent advantage in time to appropriate antimicrobial therapy.
improve the management or outcomes for protocol-based therapy of infection in ICU patients.Perspective: This study represents a challenge to the promise of earlier studies suggesting that PCT is an effective biomarker identifying and separating infection and treatment failure from other ICU complications. The negative findings may stem from an emphasis, in study design, on using PCT to unilaterally escalate care, without the apparent balance of using levels to guide de-escalation or cessation of therapy. Finally, the relatively low sensitivity (59%) of PCT for infection was surprising and serves as reminder that the role of PCT in the ICU remains to be elucidated
Peacock WF, Varon J, Baumann BM, et al. Crit Care. 2011:15;R157.
Study Question: To determine the safety and efficacy of nicardipine infusion vs. IV bolus labetalol for the management of acute hypertension in the emergency department (ED).
antihypertensive agents was discouraged during the first 30 minutes following study drug administration. The primary endpoint was number of patients within target SBP range (goal SBP ± 20 mmHg; determined by the treating physician) at 30 minutes.
Results: Researchers enrolled 226 patients. Baseline SBP was 212 mmHg and 210 mmHg for the nicardipine and labetalol groups, respectively. There was no significant difference in median initial SBP target (nicardipine 169 mmHg; labetalol 165 mmHg). More patients in the nicardipine group reached their target range at 30 minutes (91.7% vs. 82.5%; p = 0.039). At study completion, median SBP was 163 mmHg for nicardipine and 168 mmHg in the labetalol group (95% CI of the difference = -13.3 to -2.0). The mean number of titrations was 2.2 for nicardipine and 1.3 for labetalol (p < 0.001). There was no difference in the proportion of patients with blood pressure lowered below the target range. Patients receiving labetalol had a greater reduction in heart rate.
Conclusion(s): In this clinical trial, patients receiving nicardipine infusion were more likely to reach target SBP range within 30 minutes than patients receiving IV labetalol bolus.
Perspective: The authors note that separation of effect curves was seen at 15 minutes, a time period which allows for re-dosing of labetalol; however, the mean number of labetalol doses was 1.3, indicating that the majority of patients did not receive a second bolus. The decision not to re-dose the patient was left to the discretion of the treating physician and could be a confounding factor. Also, use of other antihypertensive agents was discouraged, but not prohibited, during the first 30 minutes after study drug administration. The authors did not report the nature or extent of additional antihypertensive use.
Bourdeaux CP, Brown JM. Neurocrit Care. 2011;15:42-5.
Study Question: What are the effects of equi-osmolar (EO) doses of 8.4% sodium bicarbonate and 5% NaCl on elevated intracranial pressure (ICP) after a traumatic brain injury (TBI)?
Study Design: This study was a single-center RCT of adult patients requiring sedation, ventilation, and ICP monitoring following TBI. For each episode of unprovoked ICP > 20 mmHg for > 5 min, patients were randomized to receive EO doses of either hypertonic NaCl (100 ml, 5%) or sodium bicarbonate (85 ml, 8.4%) over 30 minutes.
Results: Twenty episodes of elevated ICP were identified in 11 patients. Both treatments reduced ICP effectively, causing a decrease from baseline at all time points (P < 0.001). ICP fell below 20 mmHg after 30 minutes in all treatment episodes. There were no significant differences in ICP over time between patients receiving NaCl vs. sodium bicarbonate.
Conclusion(s): Both 8.4% sodium bicarbonate and 5% NaCl were effective in decreasing ICP in patients with TBI, and there were no significant differences in measures between either of the agents. The study’s authors suggest that hypertonic sodium bicarbonate may be a desirable therapeutic option in patients with or at risk of developing hyperchloremic acidosis.
Perspective: While the agents appear equally as effective in decreasing ICP, it is worth bearing in mind that this study was not designed to detect differences in clinical outcomes. Nonetheless, both agents may be viable options in lowering ICP following TBI, with NaCl retaining its first-line status in patients who do not have or are not at risk of hyperchloremia.
Minshall CT, Eriksson EA, Leon SM, et al. J Trauma. 2011;71:396-400.
Study Question: What is the optimal timing and type of chemical prophylaxis, UFH vs. LMWH, in patients with TBI?
Study Description: This was a 42 month retrospective review of adult trauma patients admitted to the ICU with a Head Abbreviated Injury Severity score (HAIS) > 2 and ICU LOS > 48 hours. Patients were excluded if they had isolated neck injury. All included patients were treated with sequential compression devices initially and throughout hospital stay; chemical prophylaxis was initiated once a patient’s ICH was considered stable. The prescribing physician determined the type of prophylaxis: enoxaparin 30 mg SC BID or UFH 5,000 units SC TID. Data collected included worsening ICH, clinical finding of DVT confirmed with duplex ultrasound, and clinical finding of PE confirmed by 128-slice helical CT pulmonary angiogram.
Results: A total of 386 patients were included in the review. Most patients received chemical prophylaxis at some point during their stay: 40.9% received LMWH; 44.3% received UFH; and 14.8% received only SCDs. HAIS and ISS scores were significantly higher in patients who received UFH compared to LMWH, although patients who did not receive chemical prophylaxis had the highest HAIS. There were no significant differences between ICU and hospital LOS in the LMWH (eight and nineteen days) compared to UFH group (eleven and seventeen days) (patients in the SCD only group had much shorter LOSs of two and four days). The UFH group had a significantly higher rate of DVT (1%) and PE (3.7%) than those treated with LMWH (1% and 0; p < 0.05). There were no hemorrhagic complications requiring transfusion in either chemical prophylaxis group.
Conclusion(s): The authors conclude that early administration of chemical prophylaxis in patients with severe TBI had a low rate of hemorrhagic complications requiring surgical intervention and that LMWH is more effective than UFH in preventing thrombotic complications. Further prospective studies are needed for comparison and determination of optimal chemical prophylaxis in this high risk population.
Perspective: To date, this is the largest comparison of LMWH and UFH chemical prophylaxis in patients with severe blunt head trauma. Limitations of this study include those inherent to a retrospective review, including potential selection bias. Of note, the small number of patients included in this review who did not receive chemical prophylaxis included those who were discharged or died within 96 hours of admission. The results suggest that patients with less severe injuries were more likely to get LMWH, while those with more severe injuries were treated with UFH or no agent. The low rates of DVT are likely due to a lack of a routine screening protocol.
Phung OJ, Kahn SR, Cook DJ, et al. Chest. 2011;140:374-81.
Study Question: Is there a difference between BID and TID dosing of unfractionated heparin (UFH) in thrombosis and major bleeding rates for hospitalized medical patients?
Study Description: This was a pooled analysis of sixteen RCTs and 27,667 non-surgical patients. Eligible studies compared UFH BID, UFH TID, or LMWH to one another or an inactive group.
Results: There was a significant reduction in DVT rates in patients who received thromboprophylaxis as compared to the inactive therapy but no significant difference between active therapy groups. However, there were no statistically significant differences in rates of PE, death, or major bleeding between any of the thromboprophylaxis groups and the inactive groups.
Conclusion(s): The authors conclude that there is no difference in thrombosis or bleeding rates between BID and TID dosing of UFH for medical patients admitted to the hospital.
Ruth YY Wan, Moneesha Kasliwal, Catherine A McKenzie, et.al. Crit Care. 2011;15:R159.
Study Question: Is quetiapine safe and effective when administered to patients with hyperactive or mixed delirium refractory to other management strategies?
Study Description: This is a retrospective, single-center case series of quetiapine-naïve ICU patients. Refractory hyperactive or mixed delirium was defined as a RASS > 1 for longer than 48 hours, electronic medical record (EMR) documentation of an acute confusional state, and use of two or more agents prior to introduction of quetiapine.
Results: Seventeen patients were included in this series, the majority of which were male (88%) and on mechanical ventilation (MV) (94%). Both medical (59%) and surgical patients (41%) were included, and median (range) ICU length of stay (LOS) was protracted at 41 days (16-235). Quetiapine was initiate 15 (3-48) days into the course of delirium and a combination of several agents including haloperidol, lorazepam, propofol, opioids, and clonidine were used prior to and during quetiapine therapy. Patients experienced a resolution of delirium within 4 (2-29) days after starting quetiapine. Adverse effects included QTc prolongation (n = 1), excessive somnolence (n = 1), and transient hypotension (n = 2).
Conclusion: There is a temporal association between the commencement of quetiapine therapy and the resolution of refractory hyperactive or mixed delirium.
Giraldo E, Khalid A, Zand R. Neurocrit Care. 2011;15:76-9.
Study question: To determine the safety of IV thrombolysis within the first 4.5 hours of stroke symptom onset in patients later determined not to have had a cerebral infarction.
Study description: This article described a retrospective study of patients treated with IV recombinant tissue plasminogen activator (rt-PA) within 4.5 h of stroke symptom onset. Patients had pre-thrombolytic head CT as well as post-thrombolytic multi-modal brain MRI with diffusion-weighted imaging (DWI) and head and neck MRA. Patients who did not have a cerebral infarction were diagnosed retrospectively with a TIA or stroke mimic based on the clinical presentation, hospital course, and negative post-thrombolytic DWI for cerebral infarction.
Results: Eighty-nine patients were treated with rt-PA within 4.5 h of stroke symptom onset. One patient expired shortly after rt-TPA infusion due to ruptured aortic aneurysm and was excluded from the study. Twenty-six percent of patients had negative DWI for cerebral infarction within 24 h of admission. Of those, 15.7% had a TIA and 10.1% had a stroke mimic (complicated migraine, seizure with Todd’s paralysis, or somatoform disorder). The patients with negative DWI were significantly younger (52 vs. 62 years old; p < 0.01) and had: a lower rate of arterial hypertension (52% vs. 80%; p = NR); a higher rate of psychiatric disease (17% vs. 3%; p < 0.05); a lower median NIHSS score (6 vs. 11; p < 0.001); and a shorter hospital LOS (4 vs. 9 days; p < 0.01). Twenty-six percent of both the DWI-negative and -positive patients were treated within 3-4.5 h after stroke symptom onset. Among the patients who had a confirmed diagnosis of ischemic stroke (n = 66), 4 (6.1%) had symptomatic ICH (and one received rt-PA within the 3-4.5 h timeframe). None of the DWI-negative patients had symptomatic or asymptomatic ICH on follow-up imaging.
Conclusion: The study authors suggest that administration of rt-PA within 4.5 hours of stroke symptom onset is safe in patients with suspected stroke even if post-thrombosis stroke imaging does not demonstrate cerebral infarction.
Perspective: This study suggests that delaying rt-PA to rule out TIA or stroke mimic is not warranted in terms of safety. Although the application of this study is limited by its retrospective nature and small sample size, it adds to the literature supporting the current practice of allowing rt-PA administration up to 4.5 hours after stroke symptom onset.
Krinsley JS, Schultz MJ, Spronk PE, et al. Crit Care. 2011;15:R173.
Study Question: Is mild hypoglycemia, defined as blood glucose (BG) < 70 mg/dl, associated with an increased mortality risk?
Study Description: Data from two observational cohorts, one in Stamford, CT (ST; 1 hospital), and one in the Netherlands (NL; 3 hospitals), and from the prospective GLUCONTROL trial (GL; 19 European hospitals, 7 countries) were analyzed. Patients were included if they had at least 3 BG values during their ICU stay. A total of 3,263 patients were included in the ST cohort, with a glycemic target of 80-125 mg/dl. The NL cohort included 2,063 patients who were treated with either loose (BG < 150 mg/dl) or intensive insulin therapy (IIT; BG 80-110 mg/dl). The GL cohort included 914 patients treated with either IIT targets (BG 80-110 mg/dl) or loose targets (140-180 mg/dl).
Results: A total of 6,240 patients were included. Overall, the RR of death for patients with minimum BG < 40, 40-54, 55-69, and 70-79 mg/dl compared to those patients with a minimum BG 80-110 mg/dl was 3.55, 2.70, 2.18, and 1.43, respectively, with all differences being statistically significant. This association was independent of medical or surgical diagnosis, presence of diabetes, ICU LOS, and frequency of BG checks. Even patients with only a single incident of mild hypoglycemia (OR 1.31; p=0.01) had a higher risk of mortality.
Conclusion(s): A BG < 70 mg/dl was associated with an increased risk of mortality in ICU patients, even in patients with a single hypoglycemic episode, and this risk increased with worsening degree of hypoglycemia.
Perspective: Earlier observed associations have demonstrated an increased risk of mortality with severe hypoglycemia (BG < 40 mg/dl). This study confirms the association is also present with milder degrees of hypoglycemia. It is interesting that this association was present in a large, diverse patient population, exposed to different insulin protocols. Providers and institutions should strive to prevent hypoglycemia in ICU patients.
Defne Altintas N, Aydin K, Aybar Turkoglu M, et al. Nutr Clin Pract. 2011;26:322-9.
Study Question: Is there a difference between EN and PN in the likelihood of developing of VAP?
Study Description: This study was a single-centered randomized non-blinded trial comparing EN to PN in patients admitted to a medical ICU and requiring MV for at least 72 hours.
Results: Thirty patients received EN and 41 received PN with no statistically significant differences in baseline characteristics between the two groups. VAP was identified in 16.7% in the EN group and 26.8% in the PN group (p = 0.311). There were no statistically significant differences in mortality, catheter infections, sepsis, or LOS between groups. More patients receiving PN attained nutrition targets.
Conclusion(s): There were no differences in outcomes between EN and PN patients receiving MV in a medical ICU.
Perspective: This small trial is the first to randomize MV patients to EN vs PN in a medical ICU. Unfortunately, a power analysis was not reported, making it difficult to draw conclusions given the lack of difference between the two groups in the study’s major endpoints. The study’s authors admit that the study’s impact is limited to hypothesis generation.
Cahill NE, Murch L, Jejeebhoy K, et al. JPEN J Parenter Enteral Nutr. 2011;35:160-8.
Study Question: When early EN is not possible in critical illness, is early PN beneficial?
Study Description: Two identical multicenter, prospective, observational trials were conducted in medical ICU patients in 2007 and 2008. Daily nutrition data were collected for a period of 12 consecutive days from MV adult patients with an expected ICU stay of at least 72 hours. Defining early versus late nutrition initiation at the 48-hour mark, the study’s authors looked at three groups: Group 1) early PN + late EN; Group 2) late PN + late EN; and Group 3) no PN + late EN.
Results: A total of 703 medical patients from 193 ICUs were included. The majority of patients (n = 541; 77%) received late EN only (no PN). Of the remaining patients, 83 (11.8%) were given early PN and 79 (11.2%) received late PN. A significant difference in adequacy of calories and protein – defined as the number of times the prescribed nutritional therapy was received within the first 12 ICU days – was observed between Groups 1 and 3 (74% versus 42%, respectively; p < 0.001). This apparent benefit notwithstanding, the proportion of patients who died or remained in the hospital at 60 days was significantly higher in Group 1 compared to Group 3, although the difference disappeared in multivariate analysis. No differences in outcomes between groups were noted in the subgroup of patients with a low BMI, defined as less than 25 kg/m2.
Conclusion(s): In medical ICU patients who receive EN starting more than 48 hours after ICU admission, there is no overall benefit to supplementing with PN. Conversely, a trend toward worse outcomes is observed in patients treated with early PN.
Comment: In some nutrition circles, the practice of starting PN early in the ICU is common. While good data are lacking, an oft-cited meta-analysis (Simpson et al., Intensive Care Med 2005) and ESPEN guidelines for nutrition in the ICU (Clinical Nutrition 2009) support early initiation of PN. Although the present is an observational study, the lack of observed benefit and trend toward possible harm calls for a thoughtful reconsideration of starting early PN in ICU patients who do not receive early EN or those in whom titrating to goal EN is difficult. Further prospective, randomized trials might yield additional insight.
High-Dose Selenium Substitution in Sepsis: a Prospective Randomized Clinical Trial.
Valenta J, Brodska H, Drabek T, et al. Intensive Care Med. 2011;37:808-15.
Study Question: Does the administration of high dose IV selenium (Se) improve mortality, markers of inflammation, and serum Se concentration in critically ill adult patients?
Study Description: A prospective, randomized, open-label study comparing outcomes in 150 ICU patients admitted with SIRS, sepsis, or septic shock who received either high dose Se (1,000 mcg on day one followed by 500 mcg/day for 5-14 days or until discharge) or standard supplementation of < 75 mcg/day.
Results: There was a significant increase in plasma Se levels in patients who received high dose as opposed to standard supplementation. In the high dose supplementation group, there was a significant increase in the selenoenzyme glutathione peroxidase, which protects against reactive oxygen species. There was no difference in overall mortality between groups, and subgroup analysis dividing patients by SOFA or APACHE II scores also indicated no significant difference in mortality.
Conclusion(s): The study’s authors conclude that while high dose Se supplementation did improve plasma Se levels and increase select antioxidant enzymes, that improvement was not associated with any significant effect on mortality.
Comment: This study included patients with a wide range of sepsis severity, and further evaluation of more critically ill patients may demonstrate a more profound effect on mortality. However, at this time, there is limited evidence to support high dose Se supplementation in critically ill patients. The study’s authors also comment that the initial response to rapid Se administration may cause a pro-oxidant state. As a result, research centering on the optimal dose and time of delivery of Se may provide further insight into whether or not high dose selenium supplementation is beneficial.
Effect of Corticosteroids on the Clinical Course of Community-Acquired Pneumonia: a Randomized Controlled Trial.
Fernández-Serrano S, Dorca J, Garcia-Vidal C, et al. Critical Care. 2011;15:R96.
Study Question: Can the administration of steroids to patients admitted to the hospital with CAP-induced respiratory failure lessen the need for MV and improve clinical outcomes?
Study Description: This was a single-center, randomized, double blind study that enrolled patients aged 18 to 74 admitted for CAP with extensive radiographic findings and a pO2/FiO2 < 300. Patients were excluded if they had: a clear need for corticosteroid treatment; a recent hospital admission; the presence of shock; a report of aspiration pneumonia; or the need for MV prior to study inclusion. All patients received ceftriaxone with levofloxacin and either placebo or methylprednisolone (MPDN) 200 mg IV given 30 min prior to antibiotic administration and followed by 20 mg IV q6h with a 9-day taper.
Results: Fifty-six patients were included in the intent-to-treat analysis, and between the two groups, there were no statistically significant differences in the need for MV, duration of MV, duration of hospital or ICU LOS, or mortality. Time to resolution of morbidity, as determined by a semi-quantitative score incorporating clinical and radiological findings, was significantly improved in the MPDN group. There was an improvement in the inflammatory response, demonstrated by a significant reduction in serum IL-6, IL-8, and C-reactive protein in patients given MPDN. Also, radiographic resolution and pO2/FiO2 measurements were better in patients who received MPDN.
Conclusion(s): The authors conclude that administration of MPDN may reduce the inflammatory process in patients with CAP and lead to lower MV requirements.
Comment: This study is limited not only by its small sample size but also its strict exclusion criteria.
Early Corticosteroids in Severe Influenza A/H1N1 Pneumonia and Acute Respiratory Distress Syndrome .
Brun-Buisson C, Richard J-C, Mercat A, et al. Am J Respir Crit Care Med. 2011;183:1200-6.
Study Question: Does the addition of a steroid to standard therapy impact outcomes in cases of severe influenza A/H1N1 pneumonia?
Study Description: This study was a retrospective analysis of registry data from critically ill patients hospitalized for severe influenza A/H1N1 infection with a primary outcome of in-hospital mortality.
Results: Data from 208 patients with ARDS and confirmed influenza infection were analyzed. Eighty-three patients received steroids at a median dose of 270 mg (IQR 200-400) hydrocortisone equivalents and a median duration of 11 (IQR 6-20) days. Unadjusted results showed a higher incidence of mortality in the steroid (33.7%) comopared to the no-steroid (16.8%) group, with a HR 2.39 (95% CI 1.32-4.31; p = 0.004) that remained significant following adjustment for propensity score. The steroid treated group also had significantly higher rates of ICU acquired infections and pneumonia.
Conclusion(s): Routine early use of corticosteroids in the treatment of ARDS associated with viral pneumonia should be discouraged. There may be a benefit with delayed administration of steroids; however, this use should be limited to experimental settings.
Comment: Use of corticosteroids in the treatment of ARDS has been associated with detrimental outcomes, and this finding appears to hold for the subset of patients with confirmed influenza infection. The authors of the present study suggest that early administration of steroids may aid viral replication.
Clinical and Microbiologic Outcomes in Trauma Patients Treated for Stenotrophomonas maltophilia Ventilator-Associated Pneumonia. Czosnowski QA, Wood GC, Magnotti LJ, et al. Pharmacotherapy. 2011;31:388-45.
Study Question: What are clinical success rates in the antimicrobial treatment of trauma ICU patients with Stenotrophomonas maltophilia VAP?
Study Description: This was a retrospective, single center, medical record review of 101 patients with S. maltophilia VAP from 1997 through 2007. The study population was predominately male (76%) with a median age of 40 years old, and patients were included if they had documented VAP with S. maltophilia after at least 48 hours of MV. Diagnosis of VAP required bacterial growth of at least 105 CFU/mL from a BAL and new or changing infiltrate on chest radiograph plus at least two of the following: abnormal temperature (> 38°C or < 36°C); abnormal white blood cell count (> 10 x 103 cells/mm3 or < 4 x 103 cells/mm3, or > 10% immature neutrophils); and macroscopically purulent sputum. Once culture results returned, alteration of therapy in accordance with susceptibilities was considered the standard of care. As all included patients had S. maltophilia isolated, 86% of the population was transitioned to monotherapy with trimethoprim/sulfamethoxazole, with the most common dose being 12 mg/kg/day.
Results: The primary outcome measured was clinical success rate, defined as an improvement in signs and symptoms of VAP that allowed for the discontinuation of antimicrobial therapy without relapse. This outcome was met by 88/101 (87%) of patients. The mortality rate in the patient sample was 13%, and the average LOS was 62 days.
Conclusion(s): Patients with S. maltophilia VAP appeared to have similar rates of clinical success and mortality when compared with previous reports of patients with other Gram-negative bacilli from this study site.
Comment: These data suggest that patients who have S. maltophilia VAP do not experience lower clinical success rates despite a delay in initiating the treatment of choice, trimethoprim/sulfamethoxazole, while cultures and sensitivities are pending.
Linezolid vs Glycopeptide Antibiotics for the Treatment of Suspected Methicillin-Resistant Staphylococcus aureus Nosocomial Pneumonia. Walkey AJ, O’Donnell MR, Wiener RS. Chest. 2011;135:1148-55.
Study Question: Is linezolid as efficacious as glycopeptides antibiotics (vancomycin, teicoplanin) for the treatment of nosocomial pneumonia with suspected MRSA?
Study Description: Eight randomized, controlled trials of patients age > 12 with nosocomial pneumonia were included in this meta-analysis (n = 1,641). Studies were excluded if: they were non-English; did not provide study details; or clinical success was not an endpoint. The primary endpoint, clinical success, was defined as resolution of signs and symptoms of pneumonia compared to baseline at the test-of-cure (TOC) visit in the clinically evaluable (CE) population. Secondary endpoints were: clinical success at the TOC in the ITT population; clinical success at end-of-therapy in the CE population; microbiologic success; all-cause mortality; and adverse drug events. Two studies compared linezolid to teicoplanin, while the remainder compared it to vancomycin.
Results: The average Jadad score, which is taken from a six-point (0 - 5) scale that evaluates methodological quality, was 3 ± 0.93. No differences were noted between linezolid and glycopeptide antibiotics for the primary and any of the secondary endpoints. For all endpoints, the I2 was < 0.25, indicating that the included studies engendered a relatively low degree of heterogeneity. Thrombocytopenia occurred more frequently in the linezolid arm, though it was not significantly different from comparators. Clinical success in the subgroup of patients with confirmed MRSA pneumonia (RR 1.23, 95% CI 0.97-1.53; p = 0.09) was not different from those patients who did not have confirmed MRSA pneumonia (RR 0.95, 95% CI 0.83-1.09; p = 0.48).
Conclusion(s): Linezolid is not superior to glycopeptide antibiotics for the treatment of nosocomial pneumonia with or without confirmed MRSA.
Comment: The meta-analysis and the studies it included had a number of limitations. Vancomycin was dosed at 1 g IV q12h in five of the six vancomycin trials and no drug levels were reported, raising the question of whether vancomycin might have outperformed linezolid if dosed pharmacokinetically. For patients with confirmed MRSA pneumonia, vancomycin MICs were not reported in the meta-analysis. While awaiting detailed results from the ZEPHYR study in a peer-reviewed publication, local resistance patterns, availability, and cost should continue to drive empiric antibiotic selection.
Predicting High Vancomycin Minimum Inhibitory Concentration in Methicillin-Resistant Staphylococcus aureus Bloodstream Infections. Lubin AS, Snydman DR, Ruthazer R, et al. Clin Infect Dis. 2011;52:997-1002.
Study Question: Which factors are associated with elevated vancomycin MICs in patients with MRSA bacteremia?
Study Description: This is a prospective cohort study of adults with MRSA bacteremia at a single medical center. A high MIC for vancomycin was defined as ≥ 2 mcg/mL. Factors associated with high vancomycin MICs were compiled into a scoring system to predict which patients have a higher likelihood of infection with organisms with high vancomycin MICs.
Results: Out of 296 separate MRSA bacteremic events from 272 patients, there were 57 cases (19%) with high vancomycin MICs. Variables associated with high MICs in the final predictive model were: age > 50 years (3 points); vancomycin administered for > 48 hours in the previous week (2 points); chronic liver disease (2 points); history of MRSA bacteremia (2 points); and nontunneled central line (2 points). A score cutoff of ≥ 4 had a sensitivity of 75% and specificity of 59%.
Conclusion(s): A simple predictive tool can help identify which patients are likely to have high-vancomycin-MIC MRSA isolates.
Comment: This study suggests that using this tool will enhance the ability to predict high MIC isolates, despite its modest performance in positive prediction. The tool is limited by the relatively small sample size and the elimination of other significant variables from the tool such as: prior daptomycin use; presence of septic shock; and recent surgery or ICU exposure. These factors, among others, were associated with high MICs found both in this study and previous ones. This tool may be useful in concert with other variables found to be associated with high vancomycin MICs.
Comparison of Hypertonic Saline and Mannitol on Whole Blood Coagulation in vitro Assessed by Thromboelastometry.
Luostarinen T, Niiya T, Schramko A, et al. Neurocrit Care. 2011;14:238-43.
Study Question: Is blood coagulation affected by differing strengths of hypertonic saline (HS) versus mannitol and how do they compare?
Study Description: Hyperosmolar therapy with mannitol and, more recently, HS has been used for the treatment of elevated ICP in neurosurgical patients with traumatic brain injury as well as in those with brain tumors or subarachnoid hemorrhage. Previous studies suggest that both agents interfere with coagulation, so this study sought to compare the anticoagulation effects of the two therapies when equiosmolar and equivolemic solutions of each were used. Healthy volunteers (n = 10) donated venous blood that was diluted in citrated solution in 0, 10, and 20 volume % concentrations. The solutions studied included 0.9% NaCl, 2.5% NaCl, 3.5% NaCl, and 15% mannitol (which is equiosmolar with 2.5% NaCl). Thromboelastometry devices were used to make coagulation analyses, and parameters measured included clotting time (CT), clot formation time (CFT), maximum clot firmness (MCF), alpha angle, PLTs, Hb, and HCT.
Results: Notable results include: a prolonged CT and weaker MCF for mannitol when compared to control (p < 0.05); a delay in CFT with mannitol when compared to all strengths of NaCl (p < 0.05); and a decreased alpha angle, which measures the rate of formation of a solid clot, in the mannitol group (p < 0.05). No other parameter was significantly different between solutions.
Conclusions: Based on the results, authors conclude that 15% mannitol leads to a slower coagulation process when compared to various saline concentrations and that clot formation may be weaker with mannitol therapy. When comparing results between different NaCl solutions, higher concentrations were associated with weaker coagulation processes, although no significant difference was observed in such a small sample.
Comment: Although a handful of results were statistically significant, the clinical significance may be questionable (i.e., difference in clotting time of mannitol compared to control reported to be ~30 seconds). As unimpaired coagulation is often necessary in an interventional neurosurgical setting, this study may make HS look more favorable than mannitol to treat elevated ICPs. However, the decision to do so should continue to be driven, in part, by the risks posed by the serious side effects of each therapy (e.g., AKI, hypernatremia, metabolic acidosis, etc.) when considering using one form of therapy over another.
Intravenous Erythropoietin in Patients with ST-Segment Elevation Myocardial Infarction (REVEAL: A Randomized Controlled Trial).
Najjar SS, Rao SV, Melloni C, et al. JAMA. 2011;305:1863-72.
Study Question: Does erythropoietin have a cardioprotective effect in STEMI patients?
Study Description: Investigators conducted a randomized, double-blind, placebo-controlled trial involving an initial dose-escalation safety phase (15,000-, 30,000-, or 60,000-unit epoetin alfa [EPO] doses) followed by a single-dose efficacy phase at 28 American study sites. In the efficacy phase, patients with STEMI were randomized to receive either a single dose of IV EPO 60,000 units or placebo within four hours of successful PCI. To assess outcomes, patients underwent cardiac magnetic resonance (CMR) imaging between 2 and 6 days post-intervention and then again at 12 weeks. Safety endpoints included patients enrolled in both dose-escalation and efficacy phases.
Results: A total of 138 patients were evaluated for the efficacy phase. There was no difference between EPO and placebo groups in the primary end point of infarct size at the region of the culprit artery based on first CMR. In subgroup analysis, the mean infarct size was 41.2% larger in the EPO group in patients aged 70 years and older (n = 21). Safety outcomes were assessed using both cohorts (n = 223), and the EPO group had a higher risk of death, MI, stroke, or stent thrombosis compared to placebo (4% vs. 0%; p = 0.04).
Conclusion(s): EPO, administered in STEMI patients following successful PCI, does not reduce cardiac infarct size. In fact, in older patients, treatment with EPO is associated with increased infarct size. The incidence of severe adverse effects, primarily thromboembolic, was higher in EPO-treated patients.
Comment: Despite its theorized physiologic benefits and recent animal trials suggesting a cardioprotective role, EPO currently has no place in post-ischemic injury care given the lack of benefit and association with adverse outcomes demonstrated in this trial.
RANDOMIZED, PLACEBO CONTROLLED TRIAL OF CANDESARTAN IN THE TREATMENT
OF ACUTE STROKE: RESULTS FROM THE SCAST STUDY.
Sandset EC, Bath PMW, Boysen G, et al. Lancet. 2011;377:741-50.
Study Question: In patients with acute stroke and elevated BP, will antihypertensive treatment with the angiotensin receptor blocker candesartan have beneficial effects on vascular endpoints and functional status during the first six months following a stroke?
Study Description: This article described a multicenter, double-blind study in which adult patients were eligible for enrollment if they: had a clinical diagnosis of stroke; presented within 30 hours of symptom onset; and had a SBP of greater than 140 mm Hg. A total of 2,029 patients were randomized in a double‐blind fashion to receive either a fixed dose escalation of candesartan or a matched placebo for a total of seven days. All patients continued to receive standard treatment for stroke, including the possibility of additional antihypertensive agents when clinically warranted. The co‐primary endpoints defined were: 1) a composite of vascular death, non‐fatal MI, and nonfatal stroke, and 2) functional status; both endpoints were measured at the study’s six‐month mark.
Results: There was no clinically significant differencein BP between the two groups throughout the treatment period. Mean BP on day 7 was 147/82 mm Hg in the candesartan group and 152/84 mm Hg in the placebo group. Investigators did not find a statistically significant difference between the3 candesartan and placebo groups in either of the coprimary endpoints at the end of follow‐up: the composite vascular endpoint hazard ratio = 1.09 (95% CI 0.84 – 1.41), and the functional status endpoint odds ratio = 1.13 (95% CI 0.97 – 1.32).
Conclusion(s): There was no evidence of a beneficial effect associated with the antihypertensive
candesartan in acute stroke.
Comment: This study was underpowered to detect a difference, leading the authors to conduct a post hoc meta-analysis that corroborated the study’s findings. The findings presented here are unlikely to have any effect on clinical practices within the emergency or critical care setting; current AAN guidelines recommend against the acute treatment of hypertension unless it is elevated to a SBP of > 220 mm Hg or DBP > 120 mm Hg or if the patient has a SBP > 180 mm Hg and is otherwise a candidate for thrombolysis. This trial provides further evidence to
support AAN’s recommendation.
DALTEPARIN VERSUS UNFRACTIONATED HEPARIN IN CRITICALLY ILL PATIENTS.
Cook D, Meade M, Guyatt G, et al. N Engl J M. 2011;1-10.
Study Question: Is dalteparin superior to unfractionated heparin (UFH) for the prevention of
proximal leg deep vein thrombosis (DVT) in critically ill patients?
Study Description: ICU patients are at high risk of VTE as a result of a confluence of risk factors. Adult patients who were expected to stay in the ICU for at least 3 days were randomized to receive dalteparin (5,000 units SQ daily; n = 1,873) or UFH (5,000 units twice daily; n = 1,873 UFH) in this prospective, multicenter study. Outcomes included the incidence of: proximal DVT (primary); PE; VTE; bleeding; heparin‐induced thrombocytopenia (HIT); and death.
Results: Proximal DVT occurred at an incidence of 5.1% and 5.8% in the dalteparin and UFH groups, respectively (p = 0.57), but PE developed in significantly fewer patients receiving dalteparin (p = 0.01). Rates of other VTE, death, catheter‐related thrombosis, major bleeding, and HIT did not differ significantly between the two groups.
Conclusion(s): The authors concluded that dalteparin is not superior to UFH in decreasing the
incidence of proximal DVT, but they note that the results might have been different if the study
enrollment had been larger or if they had used different drugs or doses.
Comment: To a limited extent, one may extrapolate these findings to medically ill patients, as 76% of the admissions in the study were medical, with the reminder being surgical admissions excluding major trauma, neurosurgery, and orthopedic surgery patients. Patients in the dalteparin group did have significantly fewer PEs. However, this study’s generalizeable is limited by the fact that much of it took place before October 2009 when there was an update in United States Pharmacopeia standards resulting in up to a 10% reduction in UFH potency. Even more importantly, many institutions dose UFH q8h in the majority of their medical patients, preventing these health systems from applying the study’s findings to their own populations.
COMPARISON OF NOREPINEPHRINEDOBUTAMINE TO EPINEPHRINE FOR HEMODYNAMICS, LACTATE METABOLISM, AND ORGAN FUNCTION VARIABLES IN
CARDIOGENIC SHOCK. A PROSPECTIVE, RANDOMIZED PILOT STUDY.
Levy B, Perez P, Perny J, et al. Crit Care Med. 2011;39:450-5.
Study Question: In non-ACS cardiogenic shock, which vasopressors are most suitable in optimizing systemic and regional hemodynamics?
Study Description: Patients were enrolled in this open‐label trial if they met all of the following
inclusion criteria: EF < 30%; CI < 2.2 L/min/m2; absence of hypovolemia; SBP < 90 mmHg or MAP < 60 mmHg; UOP < 0.5 mL/kg/h and resistant to diuretics; lactate > 2 mmol/L; signs of hypoperfusion; and no evidence of acute cardiac ischemia. Prior to randomization, patients were started on dobutamine for hypotension associated with low cardiac output and signs of hypoperfusion, with dopamine added for persistent hypotension. For patients who remained hypotensive, dopamine was stopped and patients were then randomized to either norepinephrine plus dobutamine (NE‐DOB) or epinephrine (EPI) alone.
Results: Thirty patients were treated according to protocol, with fifteen in each treatment arm. There were no differences in baseline values between the two groups. Patients in both treatment arms saw significant elevation in MAP, CI, and mixed venous oxygen saturation when compared to baseline; HR was significantly higher in the EPI group (p < 0.05). EPI was associated with new supraventricular arrhythmia in two patients and with sustained ventricular tachycardia in one patient. Oliguria was reversed in ten patients in the EPI group and thirteen patients in the NE‐DOB group. Arterial lactate concentrations increased in the EPI group and decreased in the NE‐DOB group, but the lactic acidosis observed in the EPI group was transient and resolved within twelve hours.
Conclusion(s): Both EPI and NE‐DOB improved arterial pressure, oxygen delivery, and renal
perfusion in patients with cardiogenic shock after failure of dopamine.
Comment: In light of the recent norepinephrine shortage, this trial demonstrates that EPI may be an alternative in non‐ACS cardiogenic shock.
CONTINUATION OF STATIN THERAPY IN PATIENTS WITH PRESUMED INFECTION: A RANDOMIZED CONTROLLED TRIAL.
Kruger PS, Harward ML, Jones MA, et al. Am J Respir Crit Care Med. 2011;183:774-81.
Study Question: Does administration of a statin (atorvastatin) influence organ dysfunction or
inflammation in patients previously on statins admitted to the hospital with presumed infection?
Study Description: This study was a randomized, double‐blind clinical trial. Patients who: had a
proven or suspected infection; were receiving antibiotics; met two SIRS criteria; and were receiving a statin prior to admission were included. Atorvastatin 20 mg or matched placebo was given at the earliest possible opportunity and continued for 28 days or until discharge or death. The primary endpoint was proportion of patients in each group with severe sepsis at prespecified time points.
Results: Baseline characteristics and need for intravenous antibiotics were similar in the 75
patients receiving placebo and 75 patients receiving atorvastatin. Severe sepsis was present at baseline in 32% of patients in both treatment arms, and there proved to be no significant difference between groups at any time during follow‐up. Likewise, SOFA scores were similar at baseline and decreased over time with no significant difference between the groups.
Conclusion(s): This study does not suggest any beneficial immunological effects when continuing statin therapy in the setting of acute infection.
Comment: Statins have been shown to have beneficial effects beyond lipid lowering within
several subgroups, including: those with ACS; acute stroke; and major non‐cardiac surgery. While this study failed to measure a benefit in the short term treatment of an infectious episode in medically ill patients previously taking a statin, it does not rule out the possibility of a beneficial effect in more seriously ill patients or those not previously taking a statin.
RELATIONSHIP BETWEEN TIME TO TARGET TEMPERATURE AND OUTCOME IN PATIENTS TREATED WITH THERAPEUTIC HYPOTHERMIA AFTER CARDIAC ARREST.
Haugk M, Testori C, Sterz F, et al. Crit Care. 2011;15:R101.
Study Question: In therapeutic hypothermia after cardiac arrest, does time to target temperature
correlate with neurologic outcome?
Study Description: This article describes a single center, retrospective chart review of cardiac arrest patients who had received therapeutic hypothermia. Time between the return of spontaneous circulation (ROSC) and the attainment of a target temperature of < 34°C by esophageal probe was compared to Pittsburgh cerebral performance category (CPC) data. Because the study was meant to assess long term neurological outcome, patients who died
during the study’s acute phase were excluded.
Results: A total of 588 patients were studied. Time from ROSC to target temperature was divided into groups: < 120 minutes, 120‐220 minutes, > 220 minutes. Thirty‐seven, 50, and 56% of patients had favorable neurologic outcomes (CPC 1‐2) in the < 120 min, 120‐220 min, > 220 min groups, respectively (p = 0.01). The median time to reach target temperature was 209 minutes in patients with favorable neurologic outcomes versus 158 minutes in patients with unfavorable neurologic outcomes. Logistic regression showed longer time to target temperature and lower age to be associated with a favorable neurologic outcome.
Conclusion(s): The study’s authors concluded that a faster decline in body temperature predicts a less favorable neurologic outcome, although they acknowledge that finding may be indicative of more severe injuries and a resultant greater loss of thermoregulation.
Comment: The exact mechanism behind the difference in neurologic outcomes remains unclear,
as there were multiple potential confounders (e.g., the difference in time between first attempt at
resuscitation and ROSC, time to attainment of target temperature, age, and severity of neurologic injury). Nonetheless, the study does suggest a need for more research to determine the optimal rate of cooling in the hypothermic treatment of cardiac arrest patients.
RANDOMIZED TRIAL OF INITIAL TROPHIC VERSUS FULL-ENERGY ENTERAL NUTRITION IN MECHANICALLY VENTILATED PATIENTS WITH ACUTE RESPIRATORY FAILURE.
Rice TW, Mogan S, Hays MA, et al. Crit Care Med. 2011;39:967-74.
Study Question: Do MV patients initially receiving trophic EN have improved clinical outcomes
compared to those who receive full-energy nutrition?
Study Description: This study was a single‐center, open‐label trial of medical ICU patients (n = 200) expected to be MV ≥ 72 hours. Exclusion criteria were: history of chronic lung disease; > 48 hours since inclusion criteria had been met; refractory shock; moribund state; end‐stage liver disease; PN use; malnutrition; and refusal of consent. Patients were randomized to initially receive either trophic EN (10 mL/hr) for 6 days and were then advanced to goal caloric intake or to full‐energy nutrition (25‐30 non‐protein kcal/kg based on IBW and 1.2‐1.6 g/kg
protein). MV management and weaning were standardized.
Results: The primary outcome of days alive and free of MV in 28 days was similar between groups (23 versus 23 days; p = 0.9). Similarly, there were no differences in secondary outcomes, including: allcause hospital mortality; 28‐day all‐cause mortality, organ‐failure free days; ICU‐free days; hospital‐free days; new infections; need for prokinetic agents; or GI intolerance (e.g., diarrhea or gastric residuals > 300 mL).
Conclusion(s): Patients receiving initial trophic EN have similar clinical outcomes as those who receive full-energy EN with trend toward improved GI tolerance
Comment: Support for the SCCM and ASPEN consensus guideline recommendation to advance to goal caloric intake within 48‐72 hours in critically ill patients is limited. This study does suggest that initial, short term trophic feeding leads to similar outcomes as rapid advancement to dietary goals in medical ICU patients. Practitioners should treat the initiation of EN within 24‐48 hours as a priority of care. However, extrapolation of these results to surgical or trauma patients is not appropriate on the basis of this trial.
HYDROCORTISONE THERAPY FOR PATIENTS WITH MULTIPLE TRAUMA.
Roquilly A, Mahe PJ, Seguin P, et al. JAMA. 2011;305:1201-9.
Study Question: Does stress‐dose hydrocortisone (HC) decrease the prevalence of HAP in trauma patients?
Study Description: This study was a multicenter, double‐blind trial that enrolled multi‐trauma
patients expected to require MV for more than 48 hours. Patients were randomized to receive either an intravenous continuous infusion of HC or placebo. The treatment was discontinued if patients were not deemed to have critical illness‐related corticosteroid insufficiency (CIRCI) based on a corticotropin test. The rate of HAP within 28 days of study inclusion
was the primary endpoint. Secondary outcomes included: MV‐free days; ICU LOS; vasopressor
support; non‐HAP infections; organ failure; mortality; and adverse drug events.
Results: Of the 149 randomized patients included in the analysis, 76% had CIRCI (HC, n = 56; placebo, n = 57). There was a significantly different rates of HAP at day 28 with placebo in patients overall (HC 35.6% versus placebo 51.3%; p = 0.007) and in those with
CIRCI (HC 35.7% versus placebo 54.4%; p = 0.01). When comparing patients overall, those treated with HC experienced significantly: more MV‐free days (16 versus 12 with placebo; p = 0.001); fewer ICU days (18 versus 24 with placebo; p = 0.03); less ARDS or ALI (4.3% versus 14.5% with placebo; p = 0.04); and less hyponatremia (0% versus 9.2% with placebo; p =
0.01) in the intervention arm. In the subset of patients with CIRCI, those treated with HC
experienced significantly: more ventilator‐free days (16 versus 10 with placebo; p < 0.001); fewer ICU days (17 versus 25 with placebo; p = 0.002); fewer days on vasoactive drugs (2.5 versus 3 with placebo; p = 0.04); and less hyponatremia (0% versus 12.3% with placebo; p = 0.008).
Conclusion(s): The authors conclude that early stress‐dose steroid treatment in multi‐trauma
patients reduces the rate of HAP at 28 days, time on MV, and ICU LOS.
Comment: Although early steroids were shown to be beneficial in this study, it would be reassuring to see the results replicated in other trials before application in practice.
DIURETIC STRATEGIES IN PATIENTS WITH ACUTE DECOMPENSATED HEART FAILURE.
Felker GM, Lee KL, Bull DA, et al. N Engl J Med. 2011;364:797-805.
Study Question: What is the optimal diuretic strategy for patients with acute decompensated
heart failure (ADCHF)?
Study Description: This study was a randomized,double‐blind trial that enrolled patients within 24 hours of initial presentation with: ≥ one symptom and ≥ one sign of heart failure; a history of chronic heart failure; and prior receipt of diuretics for ≥ 1 month. Patients were excluded for: renal
insufficiency; hypotension; or need for vasopressors or inotropes. Patients were randomized to one of four treatment arms: low dose (total daily IV furosemide equal to total daily PO loop diuretic dose at home in furosemide equivalents) by IV bolus; low dose by continuous infusion; high dose (total daily IV furosemide 2.5 times the total daily PO furosemide dose) by IV bolus; or high dose by continuous infusion. Patients were maintained on their assigned dosage regimen for 48 h, at which time the dose could be increased by 50% if necessary.
Results: Bolus versus continuous infusion: There was no significant difference in the primary efficacy endpoint (patient‐reported global assessment of symptoms) or the primary safety endpoint (change in SCr from baseline to 72 hours). Low dose versus high dose: There was a trend toward greater improvement in patient‐reported global assessment in the high dose group (NS). There was no difference between groups of the change in SCr at 72 hours;
however, more patients in the high dose group met the prespecified secondary safety endpoint of
worsening renal function, which was defined as a rise in SCr of at least 0.3 mg/dL. Patients in the highdose group had greater fluid loss, weight loss, and relief from dyspnea. More patients in the low dose group had a serious adverse event (38% versus 50%, p = 0.03), including renal failure (based on investigator report).
Conclusion(s): There were no significant differences in patients’ global assessment of symptoms or in changes from baseline renal function with either bolus versus continuous infusion, or low versus high dose IV furosemide.
Comment: These results may not be applicable to patients with low outpatient loop diuretic
requirements or those with newly diagnosed heart failure.
BENEFICIAL ASSOCIATION OF BETABLOCKER THERAPY ON RECOVERY FROM SEVERE ACUTE HEART FAILURE TREATMENT: DATA FROM THE SURVIVAL OF PATIENTS WITH
ACUTE HEART FAILURE IN NEED OF INTRAVENOUS INOTROPIC SUPPORT TRIAL.
Böhm M, Link A, Cai D, et al. Crit Care Med. 2011;39:940-4.
Study Question: Does maintaining or initiating betablockers (BBs) during hospitalization affect postdischarge survival in the short or long term?
Study Description: This is a post hoc analysis of the SURVIVE trial, which examined 180 day survival in patients receiving dobutamine or levosimendan for ADCHF. The present analysis evaluated survival at 31 and 181 days and divided patients into groups on the basis of whether they were taking BBs at admission and at discharge, making it possible for
them to fall into one of four groups: Yes/Yes, Yes/No, No/No, and No/Yes.
Results: There were 1,104 patients included, distributed as follows: Yes/Yes (n = 549), Yes/No (n = 40), No/No (n = 259), No/Yes (n = 256). Significant baseline differences, including age and reason for hospitalization, were taken into account in regression analyses. Kaplan‐Meier survival curves demonstrated a significant increase in survival between the Yes/Yes and No/No groups (p < 0.001), and between the Yes/Yes and Yes/No groups (p =0.028).
Conclusion(s): For patients with ADCHF, BB use at hospital admission and discharge was associated with improvement in both short and long term outcomes. Data suggested cessation of BB therapy at discharge in those admitted on a BB isdetrimental.
Comment: These results appear to confirm the beneficial effects of BB therapy on survival in
patients with ADCHF. In the context of controversy over continuation of BB therapy during
hospitalization for ADCHF, this study suggests that judicious use of these agents may be favorable.
THE IMPORTANCE OF EARLY TREATMENT WITH TRANEXAMIC ACID IN BLEEDING TRAUMA PATIENTS: AN EXPLORATORY ANALYSIS OF THE CRASH-2 RANDOMIZED CONTROLLED TRIAL.
The CRASH-2 Investigators. Lancet. 2011;377:1096-101.
Study Question: In trauma patients, what is the impact of tranexamic acid on mortality associated with bleeding?
Study Description: This article describes a post hoc analysis of the CRASH‐2 trial, which was a
prospective, multinational, randomized, doubleblind trial of adult trauma patients at risk of
significant bleeding who were treated with the antifibrinolytic tranexamic acid or placebo within 8
hours of injury. The primary endpoint of mortality attributable to bleeding was stratified according to time from and type of injury, severity of hemorrhage, and Glasgow Coma Scale score (GCS).
Results: The study included 20,118 trauma patients approximately 35 years of age, the majority of whom were hemodynamically stable, had GCS 13‐15, and a third of whom each presented: ≤ 1 hour, >1-3 hours, and ≥ 3 hours from injury. Tranexamic acid reduced the risk of bleeding death (RR 0.85, 95% CI 0.76‐0.96; p = 0.0077) but not non‐bleeding death (RR 0.94,
95% CI 0.86‐1.02; p = 0.13). Time to treatment with tranexamic acid was significantly associated with the risk of bleeding‐related death, with therapy in the first 3 hours protective (NNT 42 for ≤ 1 hour from injury, p < 0.0001; 77 for >1‐3 hours; p < 0.03). In contrast, in patients who were > 3 hours from injury, tranexamic acid was associated with an increased risk of bleeding‐related (NNH 77; p = 0.004) but not overall mortality (RR 1, 95% CI 0.9‐1.13).
Conclusion(s): Early tranexamic acid administered to bleeding trauma patients may improve bleedrelated mortality, whereas late administration may increase death from bleeding.
Comment: Cautions about the merits of post hoc analyses aside, selection of appropriate patients who would benefit from this therapy remains challenging.However, but time from injury may be a reasonable exclusion for late presentation traumas.
THROMBOMODULIN ALFA IN THE TREATMENT OF INFECTIOUS PATIENTS COMPLICATED BY DISSEMINATED INTRAVASCULAR COAGULATION (DIC): SUBANALYSIS FROM THE PHASE 3 TRIAL.
Aikawa N, Shimazaki S, Yamamoto Y, et al. Shock. 2011;35:349-54.
Study Question: Is thrombomodulin α (TM‐α) effective for the treatment of infection‐induced DIC?
Study Description: This article describes a retrospective analysis from a phase 3 multicenter,
double‐blind, randomized trial in patients with DIC caused by hematologic malignancy or nfection.
Results suggested that TM‐α was more effective than heparin in DIC resolution and resulted in less bleeding‐related adverse events. The present analysis excluded patients with non‐infection
induced DIC. TM‐α 0.06 mg/kg IV over 30 min daily was compared to heparin 8 units/kg/hour for six consecutive days. Outcomes included DIC resolution, defined by JMHW criteria, at 7 days and
28‐day all‐cause mortality.
Results: Of the 227 patients in the original study, 80 patients were identified with infection‐induced DIC. The only difference in baseline demographics was age, which trended higher in the TM‐α group. Comparing TM‐α to heparin, respectively, the 28-day mortality rates were 21.4% and 31.6% (p = NS), and the DIC resolution rate was 73.2% versus 63.2% (p = NS). Two patients in the heparin group and none inthe TM‐α group died due to bleeding.
Conclusions: The authors were unable to confirm the effectiveness of either group secondary to
limitations in study design, although a prospective, randomized trial of 750 patients with sepsis‐induced DIC is presently in progress.
Comment: It is uncertain as to whether TM‐α will be a viable option in the treatment of sepsis-induced DIC in future. The results of the aforementioned, larger trial are anticipated.
IMPACT OF VANCOMYCIN EXPOSURE ON OUTCOMES IN PATIENTS WITH
METHICILLIN‐RESISTANT STAPHYLOCOCCUS AUREUS BACTEREMIA: SUPPORT FOR
CONSENSUS GUIDELINES SUGGESTED TARGETS.
Kullar R, Davis SL, Levine DP, et al. Clin Inf Dis. 2011;52:975‐81.
Study Question: How is the extent of exposure to vancomycin associated with outcomes in patients with MRSA bacteremia?
Study Description: This was a retrospective cohort study of 320 adult patients who received
vancomycin as initial therapy for a documented MRSA bloodstream infection for at least 72 h. Only first episodes of bacteremia were included in the analysis, and patients were excluded if they received < 3 days of vancomycin therapy. Vancomycin treatment failure was defined as: 1) mortality within 30 days; 2) persistent signs and symptoms of infection at the end of vancomycin therapy; or 3) persistent bacteremia, defined as ≥ 7 days. Postinfectionhospital LOS was calculated from the first blood culture positive for S. aureus until discharge ordeath.
Results: A total of 168 patients (52.5%) experienced treatment failure with vancomycin. Several patients had treatment failure for more than one reason, and the results are as follows: 21% due to 30‐day mortality; 55.7% due to persistent signs or symptoms of infection at the end of therapy; and 76% due to ≥ 7 days of bacteremia. Of the deaths that occurred, 74.3% were attributed to MRSA bacteremia (infective endocarditis and pneumonia were the most common sites), although eight of the nine other patients also had persistent infection orprolonged bacteremia. Patients in whom a first vancomycin trough returned at 15 to 20 mg/L saw significantly lower failure rates. Patients more likelyto experience therapeutic failure had lower AUC24h:MIC ratios and higher MIC values. Unsurprisingly, nephrotoxicity was most common in patients with trough levels > 20 mg/L.
Conclusions: Vancomycin has been the empiric antimicrobial of choice for MRSA infections in the recent past, but resistance is increasingly limiting its utility. This study showed a > 50% failure rate of vancomycin in patients treated for MRSA bacteremia. Its results support current guidelines in that: 1) lower failure rates were seen with AUC:MIC ratios > 400; 2) lower failure rates were seen with vancomycin trough levels of 15 to 20 mg/L; and 3) higher MICs result in increasing vancomycin failure rates.
Comment: Current guidelines recommend trough concentrations of 15 to 20 mg/L in patients with
complicated infections (e.g., meningitis, osteomyelitis, and pneumonia). Based on this study,
those patients with complicated bacteremia may also benefit from empiric target troughs of 15 to 20 mg/L.
RISK OF UPPER GASTROINTESTINAL BLEEDING WITH LOW-DOSE ACETYLSALICYLIC ACID ALONE AND IN COMBINATION WITH CLOPIDOGREL AND OTHERMEDICATIONS.
Rodriguez L, Lin K, Hernandez‐Diaz S, et al. Circulation. 2011;123:1108‐15.
Study Question: Is the risk of upper gastrointestinal bleeding (UGIB) higher with low dose ASA (75 to 300 mg/d) or clopidogrel alone, in combination, or in coadministration
with select other gastrotoxic medications (e.g., NSAIDs, oral anticoagulants, oral
corticosteroids, SSRIs, and statins)?
Study Description: This study was a nested casecontrol analysis of information from a UK primary care database where 2,049 patients with a UGIB diagnosis were compared to a control group of 20,000. For inclusion, patients had to be aged 40 to 84 years and have an on‐file prescription history of at least one year. The study end points were: diagnosis of UGIB; reaching 85 years of age; death; or the study’s end at the close of 2007.
Results: Of the confirmed UGIB cases, 30.8% involved low dose ASA, 3.3% involved clopidogrel, and 1.6% involved both agents. The risk of UGIB was higher in current users of low dose ASA (RR 1.80) or clopidogrel (RR 1.67) compared with nonusers, regardless of length of therapy. Compared with low dose ASA monotherapy, the risk of UGIB was significantly increased when coadministered with clopidogrel (RR 2.08), oral anticoagulants (RR 2.00), NSAIDs (RR 2.63 or 2.66, depending on dose range), or high‐dose oral corticosteroids (> 10 mg/day of prednisone equivalents; RR 4.43). An increased risk was not apparent when ASA was coadministered with statins, low dose corticosteroids, or SSRIs.
Conclusions: The use of low dose ASA is associated with an almost 2‐fold increase in the risk of UGIB compared with nonuse, and that risk may be increased when it is coadministered with other, seelct drugs. Neither ASA nor clopidogrel monotherapy appears to be associated with a
significantly higher risk of UGIB than the other.
Comment: Because the database only captured recorded prescription medications, over-the-counter NSAID could not be taken into account.
A RANDOMIZED TRIAL FOR THE TREATMENT OF REFRACTORY STATUS EPILEPTICUS
Rossetti AO, Milligan TA, Vulliémoz S, et al. Neurocrit Care. 2011;14:4-10.
Study Question: Is propofol as effective as barbiturates (thiopental or pentobarbital) in achieving control of refractory status epilepticus (RSE)?
Study Description: This article described a multicenter, prospective, randomized, single-blind trial of adult patients with RSE, defined as ongoing clinical or EEG seizure activity or repetitive seizures without return to baseline for 30 minutes despite treatment with benzodiazepines or an antiepileptic drug (AED). Exclusion criteria were seizure due to cerebral anoxia, pregnancy, study drug intolerance, mitochondrial disorders, egg allergy, serum triglyceride levels > 470 mg/dL or serum creatine kinase concentrations > 1500 units/L. Each study drug was administered as a bolus followed by continuous infusion and was titrated toward burst-suppression. Concomitant benzodiazepines and AEDs were administered as well. The primary outcome was RSE control after 36 to 48 hours of stable burst-suppression, defined as patient alive without clinical indication for titrating study drug higher during the first 7 days following the initial weaning and ≤ 1 discrete seizure per hour during the 2 hours after attainment of continuous, stable activity on electroencephalogram (EEG). Secondary outcomes included functional outcome at day 21 and after 3 months, length of mechanical ventilation, and safety outcomes.
Results: Due to low enrollment (n = 23), the study was terminated early. No differences in baseline characteristics were noted. The primary outcome of RSE control was not statistically different between the propofol and barbiturate arms (43% vs. 22%, p = 0.4). The duration of mechanical ventilation in survivors was significantly longer in the barbiturate arm (4 days vs. 13.5 days, p = 0.03) despite a similar duration of study-drug administration duration (2 days). No other differences in secondary outcomes were observed.
Conclusions: The authors concluded that use of barbiturates for RSE may be associated with longer duration of mechanical ventilation.
Comment: This study was significantly underpowered to develop conclusions about the relative efficacy of propofol and barbiturates for RSE. However, as it is the first prospective head-to-head trial comparing these agents for RSE, it does offer some insight into and guidance in the management of this infrequent neurological disorder.
A RANDOMIZED ACTIVE-CONTROLLED STUDY COMPARING THE EFFICACY AND SAFETY OF VERNAKALANT TO AMIODARONE IN RECENT-ONSET ATRIAL FIBRILLATION
John Camm KA, Capucci A, Hohnloser SH, et al. JACC 2011; 57:313-321.
Study Question: Is intravenous vernakalant superior to amiodarone for the acute conversion of recent-onset atrial fibrillation (AF)?
Study Description: This was a phase III, multicenter, randomized, double-blind, double-dummy, comparator-controlled study of 232 patients with symptomatic, recent-onset AF (duration of 3 to 48 h), enrolled from April 2008 until November 2009. The study population included men and women between 18 and 85 years who were eligible for cardioversion, hemodynamically stable, and were taking adequate anticoagulation therapy. Patients were excluded if they had an uncorrected QT
interval > 440 ms; familial long QT syndrome; previous torsades de pointes, ventricular fibrillation, or sustained ventricular tachycardia; symptomatic bradycardia, known sick sinus syndrome, or ventricular rate < 50 beats/min; or QRS interval > 140 ms. Vernakalant group received a 10-minute infusion of 3mg/kg, followed by 15 minutes of observation and an additional 10-minute infusion of 2mg/kg if still in AF. The amiodarone group received 60 minutes of a 5 mg/kg infusion, followed by an additional 50 mg infusion over 60 minutes. Both groups received placebo injection in a separate line.
Results: A little under 52% of vernakalant patients converted from AF to sinus rhythm (SR) within 90 minutes of first exposure to the drug as compared to 5.2% of amiodarone patients (p < 0.0001). Around 53% of vernakalant patients were symptom-free at 90 minutes compared with 32.8% of amiodarone patients (p = 0.0012). SR was maintained for 98.3% of the vernakalant patients converted through 4 hours compared to 100% of amiodarone patients. There was a higher incidence of atrial flutter in vernakalant patients (8.6%) compared with amiodarone patients (0.9%) within 4 hours post-dose, although none of the atrial flutter events were considered to be serious.
Conclusions: Vernakalant was superior to amiodarone for the conversion of recent-onset AF. The low arrhythmic potential of vernakalant provides a rapid-acting therapeutic alternative to amiodarone for conversion of atrial fibrillation.
Comment: Limitations of this study include a short follow-up period for efficacy and a short infusion period for amiodarone.
IS EARLY VENOUS THROMBOEMBOLISM PROPHYLAXIS SAFE IN TRAUMA PATIENTS WITH INTRACRANIAL HEMORRHAGE
Koehler DM, Shipman J, Davidson MA, et al. J Trauma. 2011;70:324-9.
Study Question: In traumatic brain injury (TBI) patients with evidence of intracranial hemorrhage injury (IHI), are rates of IHI progression different in patients who receive early (< 72 hours) vs. late (> 72 hours) venous thromboembolism prophylaxis (VTE px)?
Study Description: This article described a retrospective cohort study in a Level 1 trauma center from July 1, 2004, to June 30, 2008, comparing two years before and after a VTE px practice management change that emphasized VTE px beginning 48 hours post admission with enoxaparin 30 mg SQ q 12 hours. The study looked primarily at IHI progression events, which included subarachnoid hemorrhages, subdural hematomas, epidural hematomas, axonal shear injuries, intraparenchymal contusions, and intraventricular hemorrhages. The study also reported VTE outcomes.
Results: Six hundred sixty-nine patients were enrolled. The time from admission to first dose of enoxaparin was 2.77 + 0.49 days and 5.31 + 1.97 days in the early (n = 268, 40%) vs. late (n = 401, 60%) groups, respectively. At baseline, patients in the late group had more severe head and neck abbreviated injury scores (p = 0.003), indicating an elevated risk of VTE. IHI progression events before VTE px were 9.38% vs. 17.41% in the early vs. late groups, respectively (p < 0.001). IHI progression events after VTE px were 1.46% vs. 1.54% (p = 0.912). There was no statistically significant difference in rates of deep venous thrombosis (1.5% vs. 3.5%, p = 0.117) or pulmonary embolism (1.5% vs. 2.2%, p = 0.49) between early and late groups,
respectively, although there were some trends toward higher VTE rates in the late group.
Conclusions: Early VTE px does not increase rates of IHI progression in hemodynamically stable patients with TBI.
Comment: This study was limited by lack of power and possible selection bias as trauma physicians could delay VTE px if they felt uncomfortable with earlier initiation of therapy. It is worth noting that the IHI progression rates observed in this study were consistent with those of previous studies. This study suggests that it is safe to start VTE px at 48 hours postadmission in selected patients with TBI.
TICAGRELOR VERSUS CLOPIDOGREL IN PATIENTS WITH ACUTE CORONARY SYNDROMES UNDERGOING CORONARY ARTERY BYPASS SURGERY: RESULTS FROM THE PLATO (PLATELET INHIBITION AND PATIENT OUTCOMES) TRIAL
Held C, Åsenblad N, Bassand JP, et al. J Am Coll Cardiol. 2011;57;672-84.
Study Question: What is the efficacy and safety of ticagrelor compared to clopidogrel in the subset of patients in the PLATO trial who underwent a coronary artery bypass graft procedure (CABG)?
Study Description: This article describes a post-hoc analysis of the PLATO trial, which was a multicenter, international, prospective, randomized, double-blind, clinical trial comparing ticagrelor to clopidogrel within 24 hours of ischemic symptom onset in addition to aspirin. For patients with acute coronary syndromes (ACS) being considered for CABG, clopidogrel and ticagrelor were held pre-operatively for 5 and 1 to 3 days, respectively.
Included patients underwent a CABG with last intake of study drug within 7 days preceding surgery. The primary efficacy endpoint was the time from CABG to first occurrence of the composite of death from vascular causes, and the primary safety measure was time from CABG to the first occurrence of any major bleeding event.
Results: The analysis included 1,261 patients from the PLATO trial having undergone CABG within 7 days of last study drug intake (6.8% of original study’s population). The ticagrelor patients discontinued their study drug 2 (30.1%), 3 to 5 (43.8%) and > 5 (26.1%) days prior to the CABG, whereas the corresponding numbers for patients taking clopidogrel were 27.7%, 37.9%, and 34.5%. The primary end point of composite death from vascular causes was 10.6% in ticagrelor and 13.1% in clopidogrel (p = 0.29) with a reduction in cardiovascular death favoring ticagrelor (p < 0.01). There were no statistically significant differences in bleeding events among groups.
Conclusions: In ACS patients requiring CABG, ticagrelor with aspirin was associated with reductions in cardiovascular deaths without increased risk of bleeding when compared to aspirin and clopidogrel.
Comment: Primary endpoint was a composite outcome in which a subgroup suggested benefit; ticagrelor is not currently FDA approved.
FIDAXOMICIN VERSUS VANCOMYCIN FOR CLOSTRIDIUM DIFFICILE INFECTION
Louie TJ, Miller MA, Mullane KM, et al. N Engl J Med. 2011;364:422-31.
Study Question: Is fidaxomicin non-inferior to vancomycin for the treatment of patients with C. difficile infection?
Study Description: This prospective, double-blind, randomized controlled trial included patients with a positive stool sample for C. difficile obtained within 48 hours of randomization and the presence of diarrhea. Patients were excluded if they received more than 4 doses of metronidazole or vancomycin or any other treatment for C. difficile within the 24 hours preceding randomization or if they had life-threatening or fulminant disease or a concurrent diagnosis of ulcerative colitis or Crohn’s disease. Patients were randomized to either fidaxomicin 200 mg PO BID or vancomycin 125 mg PO QID, each for 10 days. The primary endpoint was clinical cure (< 4 unformed stools for 2 consecutive days maintained for the remainder of the treatment course and without requirement of further therapy for the 2 days following the end of treatment).
Results: Six hundred twenty-nine patients were included; groups were similar at baseline with respect to previous treatment failure and C. difficile infection and incidence of NAP1/BI/027 strain. The rate of clinical cure was non-inferior with fidaxomicin at 88.2% vs. 85.8% with vancomycin in the modified intention-to-treat analysis. Fewer patients in the fidaxomicin group had a recurrence of infection (15.4% vs. 25.3%; p = 0.005), although this difference was only observed for non-NAP1/BI/027 strains.
Conclusions: Fidaxomicin is not inferior to vancomycin in achieving clinical cure of C. difficile
and may lead to lower rates of recurrence in non-North American Pulsed Field type 1 strains.
Comment: The results of this study are only applicable in patients with mild or moderate disease, as fulminant disease was excluded.
TRANSDERMAL NICOTINE REPLACEMENT THERAPY IN CIGARETTE SMOKERS WITH ACUTE SUBARACHNOID HEMORRHAGE
Seder DB, Schmidt JM, Fernandez L, et al. Neurocrit Care. 2011;14:77-83.
Study Question: In active smokers who develop aneurysmal subarachnoid hemorrhage (SAH), does nicotine replacement therapy (NRT) increase mortality or worsen other outcomes like cerebral ischemia due to vasospasm, imaging evidence of vasospasm, or delirium?
Study Description: This article described a retrospective study that utilized a prospectively collected database from patients admitted to a single-center ICU after a non-traumatic SAH. Patients who were active smokers prior to admission and who survived for > 3 days were included. Patients were started on a 21-mg transdermal nicotine patch daily at the discretion of the neurointensivist. Standard medical management of all patients was directed by one of four neurointensivists.
Results: Six hundred five patients were included in the database, of which 569 patients met inclusion criteria and 55% received NRT. Baseline demographics were similar between the two groups, with the exceptions of a higher percent of patients in the NRT group being identified as: heavy smokers (> 10 cigarettes/day); patients with diabetes; heavy drinkers; and those with cerebral edema present on admission CT. Multivariate analysis revealed that, in smokers, NRT was strongly associated with 3 month survival. There were no significant differences in
most other medical complications or vascular events, including vasospasm and myocardial infarction. However, delirium was more frequent in the NRT group. Also, as compared to the NRT group and non-smokers, smokers who did not receive NRT had lower rates of pneumonia, pulmonary edema, fever, and seizures.
Conclusions: NRT is safe in active smokers who develop aneurismal SAH.
Comment: As NRT was prescribed in a non-randomized method, significant selection bias may in part account for the results. Furthermore, although there was an increase in survival in patients who received NRT compared to both non-NRT smokers and non-smokers, the number of deaths was small. This difference could therefore be due to chance, a confounder not taken into account, or the authors’ hypothesized neuroprotective effects of nicotine.
EFFECT OF ORAL BETA-BLOCKER ON SHORT AND LONG-TERM MORTALITY IN PATIENTS WITH ACUTE RESPIRATORY FAILURE: RESULTS FROM THE BASEL II-ICU STUDY
Noveanu M, Breidthardt T, Reichlin T, et al. Crit Care. 2010;14:R198.
Study Question: Do beta-blockers affect in-hospital and one-year mortality rates in patients with acute respiratory failure?
Study Description: A post-hoc analysis was conducted on data from a prospective, randomized, controlled, single-blinded multicenter study (B-type natriuretic peptide for Acute Shortness of Breath Evaluation, or BASEL II-ICU trial), performed between 12/04 and 3/07 in seven intensive care units in Switzerland. Outcomes included in-hospital mortality and one-year mortality.
Results: Data from 314 patients with acute respiratory failure were reviewed. At baseline, beta-blocker use was more frequent in both the in-hospital (p = 0.001) and one-year (p<0.0001) survivors vs. non-survivors. A multivariate analysis found that in-hospital (HR 0.33, p = 0.007) and one-year (HR 0.29, p = 0.0003) mortality rates independently favored patients who were taking beta-blockers on admission as compared to those who were not.
Conclusions: The authors conclude that beta-blocker therapy is associated with decreased mortality rates in patients with acute respiratory failure.
Comment: The data presented were observational in nature and based on a post-hoc analysis of a randomized controlled trial. It is unclear whether a compliance bias, rather than beta-blocker use itself, was responsible for the differences observed between survivors and non-survivors. Based on a univariate analysis, survivors were more likely to take beta-blockers, in addition to statins, aspirin, clopidogrel, and ACE inhibitors or ARBs at admission. Compliance with the medication management of comorbidities, such as heart failure, coronary artery disease, and hypertension, likely also influenced the outcome of mortality in a compliant vs. non-compliant group.
COMPARISON OF BIVALIRUDIN AND ARGATROBAN FOR THE MANAGEMENT OF HEPARIN-INDUCED THROMBOCYTOPENIA
Skrupky LP, Smith JR, Deal EN, et al. Pharmacotherapy. 2010;3:1229-38.
Study Question: Are there any differences between the direct thrombin inhibitors (DTIs) (argatroban and bivali rudin) in the achievement of therapeutic anticoagulation or in clinical outcomes in patients with heparin-induced thrombocytopenia (HIT).
Study Description: Although DTIs are the drugs of choice for treatment of HIT, only lepirudin and argatroban are FDA-approved for this indication. This study compared the achievement of therapeutic anticoagulation and clinical outcomes in patients who received argatroban or bivalirudin for HIT. Adult patients receiving either argatroban or bivalirudin for at least 24 hours between January 2007 and July 2008 for known or suspected HIT were enrolled in this retrospective, single-center study (N=138; n=46 argatroban; n=92 bivalirudin).
Results: The median initial doses for argatroban and bivalirudin were 1 mcg/kg/min and 0.06 mg/kg/hr, respectively. The achievement and maintenance of therapeutic activated partial thromboplastin times (aPTTs) was similar in both groups, but the argatroban group had a higher number of supratherapeutic aPTTs overall (18% vs 7.9%, p=0.046). Rates of thromboembolic events during or after DTI therapy, clinically significant bleeding, and death were not significantly different between groups.
Conclusions: The dosing information used to achieve therapeutic targets in this study is useful for clinical practice; however, prospective, randomized trials need to be conducted to confirm the safety and efficacy of bivalirudin in HIT management.
Comment: Statistically significant baseline differences between groups as a result of service provider preferences were a major limitation of this study in addition to the retrospective study design.
FAST POINT-OF-CARE COAGULOMETER GUIDED REVERSAL OF ORAL ANTICOAGULATION AT THE BEDSIDE HASTENS MANAGEMENT OF ACUTE SUBDURAL HEMORRHAGE
Rizos T, Jenetzky E, Herweh C, et al. Neurocrit Care. 2010;13:321-5.
Study Question: Is point-of-care (POC), international normalized ratio (INR)-based, stepwise reversal of oral anticoagulants faster than central laboratory (CL) INR-based reversal in patients with nontraumatic subdural hematoma (SDH)?
Study Description: Patients taking phenprocoumon, an oral coumarin anticoagulant with a mechanism similar to warfarin, who were admitted with a nontraumatic SDH were included in this prospective study. Upon arrival to the emergency department, blood samples were sent to the CL for INR testing and an INR was obtained by POC measurement. Stepwise reversal of anticoagulation was performed using prothrombin complex concentrates (PCC) until INR was < 1.4.
Results: Ten patients were treated according to protocol. Agreement between initial POC and CL INR values was excellent, with a mean deviation of paired differences of 0.013 (SD: 0.32); limits of agreement were -0.61 to +0.64. During INR reversal, mean deviation of paired differences between POC and CL INR values was 0.081 (SD: 0.14); limits of agreement were +0.36 to -0.19. Median total time for INR reversal was shorter for POC testing versus CL testing (27 min vs 70 min). An estimated median reduction in PCC dose of 25% could be obtained with POC testing.
Conclusions: POC measurement of INR is a fast and economical way to reverse oral anticoagulants in patients with acute phenprocoumon-associated SDH.
Comment: Given the small sample size, further investigation should be conducted to determine the full extent of possible benefits and consequences of POC testing in this patient population. Additionally, inherent differences in turn-around time for CL INR results and PCC delivery and administration will grossly affect the utility of this regimen at various institutions. Additionally, selection of an appropriate POC device and quality control is critical.
DAPTOMYCIN PHARMACOKINETICS IN CRITICALLY ILL PATIENTS RECEIVING CONTINUOUS VENOVENOUS HEMODIALYSIS
Vilay AM, Grio M, Sowinski KM, et al. Crit Care Med. 2011;39:19-25.
Study Question: To determine the pharmacokinetics (PK) of daptomycin in patients receiving continuous venovenous hemodialysis (CVVHD)
Study description: Daptomycin, a concentration- dependent antimicrobial for non-pulmonary resistant gram-positive infections, is increasingly being used in critically ill patients. Researchers conducted a prospective, open-label study of eight adults with a known or suspected gram-positive infection and were receiving CVVHD. Patients were excluded if: their treatment was intended for osteomyelitis, meningitis, or pneumonia; they were pregnant; or they were receiving additional extracorporeal therapies.
Results: Daptomycin was infused at a dose of 8 mg/kg q48h to attain a goal Cmax of 100 mcg/mL and an AUC0-24 of 500 mcg*hr/mL to approximate the estimated target goals for patients with bacteremias. Patients in this analysis had mean ± SD BMI of 28.3 ± 6.6 kg/m2, received an average dose of 7.7 ± 0.6 mg/kg, and underwent a mean dialysate flow rate of 26 ± 4 mL/kg/h. The observed Cmax was 81.2 ± 19 mcg/mL and calculated PK parameters included a median apparent steady-state Vd 0.17 L/kg, with a fraction unbound of 17.5 ± 5% (both higher than in healthy individuals), and a median T1/2 15.9 hours. A simulation comparing 4 mg/kg q24h to 8 mg/kg q48h demonstrated a higher Cmax at steady-state with the 8 mg/kg dose, with comparable AUC96-144.
Conclusion: This study demonstrates that 8 mg/kg q48h achieves approximate therapeutic goals for bacteremic CVVHD patients and is well tolerated and safe.
Comment: Intra-patient variability within the context of a small sample size may have had an affect on PK modeling, and the results should not be extrapolated to patients at extremes of weight.
TELAVANCIN VERSUS VANCOMYCIN FOR HOSPITAL-ACQUIRED PNEUMONIA DUE TO GRAM-POSITIVE PATHOGENS
Rubinstein E, Lalani T, Corey GR, et al. Clin Inf Dis. 2011;52:31-40.
Study Question: Is telavancin noninferior to vancomycin for the treatment of hospital-acquired pneumonia (HAP) due to Gram-positive pathogens?
Study Description: Two identical multicenter, randomized, double-blinded, comparator-controlled phase III trials, termed ATTAIN (Assessment of Telavancin for Treatment of Hospital-Acquired Pneumonia) were conducted from January 2005 to June 2007. Adult patients with a diagnosis of HAP were randomized to receive either telavancin (10 mg/kg IV q24h) or vancomycin (1 g IV q12h) for seven to 21 days. Site-specific vancomycin monitoring and dose adjustments and concomitant therapy for Gram-negative coverage with aztreonam or piperacillin-tazobactam were permitted.
Results: Just over 1,500 patients received at least one dose of study medication. Of these, a total of 480 patients were considered microbiologically evaluable, defined as having a Gram-positive respiratory pathogen cultured from baseline respiratory or blood cultures. S. aureus was the most common pathogen cultured, with MRSA making up 56% of S. aureus isolates in the telavancin group and 65% of isolates in the vancomycin group. Cure rates were similar between groups for all patients treated and for the microbiologically evaluable subgroup. The incidences of serious adverse events (AEs) and treatment-emergent AEs resulting in discontinuation of study medication were both higher in the telavancin group, although specific types of AEs were not clearly stated. The telavancin group had a higher rate of renal impairment (10% vs 8%) and serum creatinine increase (16% vs 10%) than the vancomycin group.
Conclusions: The authors conclude that telavancin is as effective as vancomycin in the treatment of HAP caused by Gram-positive bacteria and possesses an acceptable safety profile.
Comment: While telavancin was found to be noninferior to vancomycin, the use of vancomycin trough monitoring was inconsistent among sites. When used, the majority of vancomycin troughs were in the range of 5–15 µg/mL, which is inadequate for HAP and not surprising given the low, flat-dosing strategy. Consistent with past literature, this study strongly supports the close monitoring of renal function during telavancin therapy.
Intensive Insulin Therapy in the Neurocritical Care Setting is Associated with Poor Clinical Outcome
Graffagnino C, Gurram AR, Kolls B, et al. Neurocrit Care. 2010;13:307-12.
Study Question: Is there any benefit from Intensive Insulin Therapy (ITT; blood glucose [BG] goal 80 – 120 mg/dL) on morbidity and mortality in neurologically injured patients compared to Standard Infusion Therapy (SIT; BG goal < 150 mg/dL)?
Study Description: Retrospective, before-(Feb 2005 to Aug 2006, SIT)-and-after (Sept 2006 – Mar 2008, IIT), single center study of 3709 patients in a neuro ICU. Endpoints included hospital mortality, length of stay (LOS), and hypoglycemia.
Results: Hospital mortality was higher in the IIT group (11% vs 7.9%, p = 0.0013). Hypoglycemia in any group was associated with increased mortality (BG < 70 mg/dL: OR 3.26, 95% CI 2.52-4.22; BG < 40 mg/dL: OR 3.65, 95% CI 2.21-6.02; BG < 20 mg/dL: OR 6.25, 95% CI 2.41-16.23). The use of IIT was associated with more hypoglycemia (BG < 70 mg/dL, OR 1.8, 95% CI 1.5-2.3). LOS was longer in the IIT group (mean of 9.5 vs 8.7 days, p=0.046). Hyperglycemia was also associated with an increased mortality (OR 1.99, 95% CI 1.46-2.71).
Conclusions: Both hyperglycemia and hypoglycemia results in increased mortality. The IIT protocol in this study resulted in more hypoglycemia and hyperglycemia.
Comment: Although limited by being a single center study, assessing outcomes during different time periods, and not assessing metrics for neurological and critical illness, it does underscore the importance of avoiding both hypo- and hyperglycemia in any glycemic protocol used to treat patients in the neuro ICU population.
INTENSIVE VERSUS CONVENTIONAL INSULIN THERAPY IN CRITICALLY ILL NEUROLOGIC PATIENTS
Green DM, O’Phelan KH, Bassin SL, et al. Neurocrit Care. 2010;13:299-306.
Study Question: Is there any benefit from Intensive Insulin Therapy (ITT; BG goal 80 – 110 mg/dL) on mortality and neurological function in neurologically injured patients compared to Conventional Therapy (CT; BG goal < 151 mg/dL)?
Study Description: Prospective, randomized, single center study of 81 patients expected to have an ICU stay of at least 3 days in a neuro ICU from September 2004 to July 2008. An insulin infusion guided by the Glucommander software was used to achieve BG goal for the IIT. For the CT protocol, BG was measured q6h and an algorithm for SQ regular insulin was used to achieve BG goal. If SQ insulin therapy did not adequately result in in-range glucose values, patients could be switched to an IV insulin infusion. Endpoints included 90-day death from any cause, a 90-day modified Rankin score, ICU LOS, hospital LOS, number of days of mechanical ventilation, number of hypoglycemia episodes, bloodstream infections, pneumonia, and PRBC transfusions.
Results: Investigators were not able to meet enrollment necessary (540 patients) to meet power. There were no differences in any of the endpoints except that there were more hypoglycemic events (BG < 60 mg/dL) in IIT (47% vs 11%, p = 0.0006).
Conclusions: There was no benefit of IIT in this small study.
Comment: Although a well designed study, and the first to assess neurologic outcome, there is still a need for a well-powered study to draw any conclusions. Hopefully, continuous monitoring of BG will provide more optimal BG control while avoiding hypoglycemia.
EARLY JEJUNAL FEEDING INITIATION AND CLINICAL OUTCOMES IN PATIENTS WITH SEVERE ACUTE PANCREATITIS
Hegazi R, Raina A, Graham T, et al. JPEN 2011;35:91-6.
Study Question: In patients with severe acute pancreatitis (SAP), is there an association between the time to initiation of enteral tube feedings and achievement of goal rates and clinical outcomes?
Study Description: This was a retrospective chart review of patients with SAP receiving jejunal feeding. Patients were included if they had one or more major local complications such as pseudocysts, necrosis, or abscesses. Exclusion criteria were: chronic liver, renal, or respiratory failure; HIV/AIDS; and acute-on-chronic pancreatitis. Caloric needs were calculated using a range of 20-25 kcal/kg of ideal body weight (IBW). A semi-elemental formula was started at a rate of 20-25 mL/hr and advanced based on tolerance and bowel function q12-24h until the target rate was reached.
Results: Seventeen patients total were included in the study. Patients who never reached their goal enteral feeding rate had a significantly longer ICU stay than those patients who did. Patients in whom it took >12 days from the onset of abdominal pain to reach their goal enteral feeding rate also had a significantly longer ICU stay (19 vs. 9 days, p<0.001). Patients who did not survive were significantly more likely to experience a delay in tube feeding initiation as compared to survivors (17 vs. 8 days, p<0.05).
Comment: This study indicates that earlier initiation of jejunal feeding in SAP and more rapid
achievement of goal enteral feeding rates are associated with shorter ICU lengths of stay and
lower mortality rates. The small number of patients and retrospective nature of the study, however, make it difficult to establish a causal link between the intervention and clinical outcomes.
LACTOBACILLUS GG AS TREATMENT FOR DIARRHEA DURING ENTERAL FEEDING IN CRITICAL ILLNESS: RANDOMIZED CONTROLLED TRIAL
Ferrie S, Daley M. J Parent Enteral Nutr. 2011;35:43-9.
Study Question: In critically ill patients experiencing diarrhea during enteral tube feeds, does the administration of the probiotic Lactobacillus GG (LGG) reduce duration and severity of loose stools?
Study Description: A randomized, double-blind, placebo-controlled trial that enrolled 36 consecutive ICU patients who were receiving tube feeds and had persistent diarrhea > 48 hours after discontinuation of laxatives. Diarrhea was defined as 3 or more loose stools or >200 mL of liquid stool in a 24 hr period with no suspected/known malabsorption. Patients received either LGG 1 capsule via feeding tube q12h for seven days or a matched placebo.
Results: There were no differences in baseline characteristics between groups. Investigators found no significant differences between groups in duration or severity of diarrhea, adverse effects, or six-month mortality. In the fourteen-day study period, there were also no differences in mean number of loose stools per day (1.58 [SD 0.88] in LGG group versus 1.10 [SD 0.79] in placebo; P=0.15) or duration of diarrhea from day 1 of study (3.83 d [SD 2.39] in LGG group versus 2.56 [SD 1.85]; P=0.096). Results were similar in the per-protocol and intent-to-treat analysis.
Conclusion: Results of this study do not demonstrate a benefit in treatment of diarrhea in critically ill patients with LGG.
Comment: Although there was no difference in duration of diarrhea, the study was only powered to demonstrate a decrease in duration of at least 2.7 days, warranting a larger study to evaluate a more conservative decrease in duration
INTERRUPTED PHARMOCOLOGIC THROMBOPROPHYLAXIS INCREASES VENOUS THROMBOEMBOLISM IN TRAUMATIC BRAIN INJURY
Salottolo K, Offner P, Levy AS, et al. J Trauma. 2011;70: 19-26.
Study Question: What is the optimal administration of pharmacologic thromboprophylaxis (PTP) in patients with stable traumatic intracranial hemorrhage (tICH)?
Study Description: This was a retrospective cohort study of adult (age ≥ 18 years) trauma patients consecutively admitted to two Level I trauma centers in Denver, Colorado. Patients enrolled (n = 480) were admitted with blunt traumatic brain injury (TBI) to Swedish Medical Center from September 1, 2007 to August 31, 2009 and to St. Anthony Central Hospital from May 1, 2007 to November 30, 2008. TBI was identified based on International Classification of Diseases-9th Revision diagnosis. The primary outcome was the development of a VTE during hospitalization, while secondary outcomes included in-hospital mortality and hemorrhage progression. Data on time to initiation of PTP and interruption of PTP were also captured.
Results: There were 255 patients (53.1%) who received PTP during hospitalization. Those who did receive PTP had multiple baseline characteristics that were significantly different than those patients who did not received PTP. Similarly, those patients who received continuous PTP (73.7%) differed significantly from those whose PTP was interrupted. Roughly 95% of all patients enrolled received mechanical prophylaxis with sequential compression devices (SCD). A total of 15 patients (3.13%) developed VTE. Neither administration of PTP nor timing of PTP was an independent predictor of developing VTE (PTP vs. none: odds ratio [OR] = 0.36, p = 0.18; early PTP vs. late PTP: OR = 2.00, p = 0.41). Patients with interrupted PTP had significantly increased odds of developing VTE compared with patients with continuous PTP (OR = 7.07, p = 0.04). Walking before discharge significantly decreased the odds of developing a VTE (OR = 0.19, p = 0.02).
Conclusions: Interrupted administration of PTP in patients with TBI is associated with significantly increased risk of VTE. These findings underscore the importance of continuous PTP administration, and every effort should be made to avoid interruption if possible.
Comment: This article discussed the possibility of a statistical type 2 error in the detection of differences in the primary outcome due to the relatively small number of patients. Also, potentially confounding the results is the fact that many patients had therapy interrupted due to the need for surgical intervention, which could independently lead to increased rates of VTE.
THROMBOEMBOLIC PROPHYLAXIS WITH LOW-MOLECULAR-WEIGHT HEPARIN IN PATIENT WITH BLUNT SOLID ABDOMINAL ORGAN INJURIES UNDERGOING NON-OPERATIVE MANAGEMENT: CURRENT PRACTICE AND OUTCOMES
Eberle BM, Schnüriger B, Inaba K, et al. J Trauma. 2011;70: 141-7.
Study Question: At a Level I trauma center, does the current practice of utilizing low-molecular-weight heparin (LMWH) for prophylaxis of venous thromboembolic complications (VTE) in patients with blunt splenic, liver, and/or kidney injuries impact the failure rate of nonoperative management and packed red blood cell (PRBC) transfusions?
Study Description: This was a 4 year retrospective study including all adult patients (≥15 years) who had an available abdominal CT scan and “an initial attempt” at nonoperative management. Those who underwent an exploratory laparotomy within six hours of admission were defined as cases undergoing operative management, while exploratory la pa rotomies after six hours signaled failure of non-operative management. Patients were further divided into three groups: (1) use of LMWH within three days of admission (“early”), (2) use of LMWH three days after admission (“late”), or (3) no LMWH during entire hospital stay. Endpoints included failure of nonoperative management, amount of PRBC transfusions and rates of VTE.
Results: Of the 565 patient who were admitted with blunt abdominal organ injuries, 312 patients underwent an initial attempt at nonoperative management. Overall, 35.6% of patients received LMWH during their hospital stay: 13.2% early and 22.4% late. The severity of solid organ injuries at baseline was similar. However, severe pelvic and lower extremity fractures were significantly more common in the early and late groups as compared to those who did not receive LMWH. Differences in transfusion requirements within the first 24 hours were not statistically significant between the early and late groups, but were statistically different when compared to those not receiving LMWH (p=0.005). The overall transfusion requirements were lower in the early LMWH group compared to the late group (p=0.027), while the adjusted failure rate of non- operative management was no different between LMWH groups. The overall rate of VTE was 1.3%, and all events occurred prior to LMWH administration.
Conclusions: There were no differences between either early, late, or no LMWH administration with respect to failure rates of non-operative management or rates of VTE.
Comment: In addition to the finding that early use of LMWH did not appear to increase non-operative failure rates or blood transfusion requirements, it is important to recognize other risk factors for failure rates of non-operative management (contrast extravasation, pseudoaneurysm, AV-fistulas, large hemoperitoneum volume) that were not described by authors. As expected, the study showed that current practice of LMWH administration was guided by injury type and VTE risk factors.
Critical Care Pharmacotherapy Literature Update Contributors: Marcus Costner, PharmD, BCPS (VA); Erin Frazee, PharmD (Mayo); Haley Goodwin, PharmD (Johns Hopkins); Deanna McMahon Horner, PharmD, BCPS (UCSF); Emily Hutchison, PharmD, BCPS (Clarian Methodist); Shawn Kram, PharmD, BCPS (Via Christi); Jessica Mercer, PharmD, BCPS (MUSC); Erin Nystrom, PharmD, BCNSP (Mayo); Heather Personett, PharmD (Mayo); Angela Plewa, PharmD, BCPS (Stroger); Bridgette Therriault, PharmD (Wesley); Charles J Turck, PharmD, BCPS (UMass); Peter Herout, PharmD, BCPS (EPI-Q, Inc.).
Materials on this page © 2011 by Scientia Perpetuam, LLCSM Follow us on facebook at Critical Care Pharmacotherapy