Impact of Low-Dose Ketamine on the Usage of Continuous Opioid Infusion for the Treatment of Pain in Adult Mechanically Ventilated Patients in Surgical Intensive Care Units. Buchheit JL, Yeh DD, Eikermann M, Lin H. J Intensive Care Med. 2019 Aug;34(8):646-651.
Ketamine is emerging as a useful analgesic and sedative in the surgical intensive care unit (SICU). This study is a retrospective review of mechanically ventilated adult patients in two SICUs at a tertiary care academic medical center who met a safety screen for spontaneous breathing trial and received both low-dose ketamine (1-5ug/kg/min) and opioid continuous infusions. The authors’ primary endpoint was the slope of change of morphine equivalents (MEs) pre- and post-ketamine infusion based on a standard practice of decreasing opioid and sedation infusions by 25% based on the Critical Care Pain Observation Tool (CPOT) and Richmond Agitation Sedation Score (RASS). They also observed vasopressor and sedative use, vital signs, need for glycopyrrolate for ketamine-associated increased salivation, and Critical Care Observation of Motor Activity (COMA) strength assessment.
From January 2012 to August 2015, 267 patients received low-dose ketamine continuous infusion; the authors excluded 115 for not receiving mechanical ventilation, 64 for lack of opioid continuous infusion and 48 for not passing the safety screen, resulting in 40 eligible patients for analysis (median 58.5yo, 77.5% male, median APACHE II 14.5 and median ketamine maintenance dose 5 ug/kg/min [IQR 3.5-5]). The authors observed a significant decline in ME rates during the 24-hours preketamine to 24-hours postketamine (0mg/hr to -0.256mg/hr, p<0.001). Other significant findings from 1 hour preketamine to 6 hours postketamine include: decreased median ME infusion rate (6.66mg/hr to 5mg/hr, p<0.004); decreased median phenylephrine equivalent (70 to 40 mg/hr, p = 0.019); decreased median propofol infusion (180 to 150 mg/hr, p=0.014) and decreased frequency of RASS <-1 and increased frequency of RASS >0 (both p<0.001). Significant findings from 24-hours preketamine to 24-hours postketamine include: decreased median ME infusion rate (6.66 to 0mg/hr, p<0.001) and decreased median propofol infusion (150 to 32.5mg/hr, p=0.002). Vital signs were similar between groups as were COMA scores. No patients required glycopyrrolate for excess salivation.
This interesting study uses a narrow perspective to address a specific question about the utility of ketamine for SICU patients during the period of potential extubation. Although burdened by a low n and the limits of a retrospective design without a comparison group who did not receive ketamine, the study adds more fuel to the idea that ketamine may be a useful drug for analgesia and sedation in the SICU and without the same clinical challenges associated with narcotics. The study suggests MEs may be reduced with the addition of ketamine, an important consideration amid the growing opioid epidemic. It would have been interesting to see if rates of successful extubation (and other ventilation criteria) were similar with and without ketamine. Larger and comparative studies using ketamine in the SICU would help clarify optimal surgical critical care analgesia.
Association Between Ratio of Fresh Frozen Plasma to Red Blood Cells During Massive Transfusion and Survival Among Patients Without Traumatic Injury. Mesar T, Larentzakis A, Dzfik W, Chang Y, Velmahos G, Yeh DD. JAMA Surg. 2017 Jun 1;152(6):574-580.
In light of the somewhat conflictual PROMMTT and PROPPR studies, this study aims to address whether massive transfusion concepts for trauma patients apply to patients who have not suffered a traumatic injury. The authors performed a large retrospective review of all massive transfusions (MTs) provided in an urban academic hospital from January 1, 2009 through December 31, 2012. All patients who received MT (≥10U RBCs in the first 24-hours of admission) and survived >30 minutes after arrival were assessed, and FFP:RBC ratios and survival data were observed.
767 of 865 MT events were for patients without trauma. The majority of these events were (from most to less): cardiac surgery, liver transplant surgery, general surgery, vascular surgery, medicine, orthopedic surgery, and other surgeries. Most MT events (62.9%) occurred for intraoperative bleeding, followed by prehospital bleeding (21.6%). FFP:RBC ratios were similar overall between survivors and nonsurvivors; whereas, overall amount of blood and blood products as well as age were higher among nonsurvivors. Examining trends of mortality rates, the authors divided the study population into those receiving median [IQR] FFP:RBC ratios that were high (1:0.9 [1:0.4-1:1.1]), medium (1:1.4 [1:1.2-1:1.7]) and low (1:3.0 [1:1.7-1:21]). Across these subgroups, mortality rates were similar overall and among cardiac and liver transplant surgeries. However, medical and general and orthopedic surgeries showed an association of mortality with higher FFP:RBC ratios. Conversely, vascular surgeries showed the opposite association. The respective trends for medical, general surgery and vascular surgery subgroups remained significant after regression analysis; trauma surgery MT events showed a similar trend as those for vascular surgery.
While the study suffers from the well-known limitations of single-center retrospective studies, its novelty challenges the belief that higher FFP:RBC ratios may be beneficial compared to other MT ratios. Data on FFP:RBC ratios for nontrauma patients are limited, and hemostatic features of patients who are injured may be very different from those of patients without traumatic injury. While the vascular surgery trend of higher mortality with lower FFP:RBC ratios has been observed previously, the dramatic association of mortality with higher FFP:RBC ratios among general surgery and medical patients (adjusted OR 4.27 & 8.48, respectively) begs questions about more nuanced and precise allocation of blood and blood products among patient subgroups. Overall, there did not seem to be any survival benefit from different FFP:RBC ratios among nontrauma patients, further complicating clinical translation of these data. At minimum, this study encourages closer inspection of use of FFP:RBC ratios among varieties of patients.
Daily propranolol administration reduces persistent injury-associated anemia after severe trauma and chronic stress. Alamo IG, Kannan KB, Bible LE, Loftus TJ, Ramos H, Efron PA, Mohr AM. J Trauma Acute Care Surg. 2017 Apr;82(4):714-721.
This is an animal study, examining norepinephrine-mediated persistent, injury-associated anemia. This anemia is associated with suppression of bone marrow (BM) erythroid colony growth, decreased iron levels and elevated erythropoietin (EPO) levels and possibly altered ferroportin, transferrin and transferrin receptor-1 (TFR-1). The authors used a clinically relevant rodent model of lung contusion (LC), hemorrhagic shock (HS) and chronic stress (CS) to observe if nonselective B-blocker propranolol (BB, 10mg/kg/day) restores BM function and improves iron homeostasis. The authors created five groups of study subjects (naïve control, LCHS±BB and LCHS/CS±BB) and compared liver, BM and peripheral blood samples after sacrifice on day 7; n=8/group.
Results showed that propranolol restored erythroid function: BM functions significantly decreased with LCHS-BB and LCHS/CS-BB compared to naive, while these same decreases were not noted for LCHS+BB and LCHS/CS+BB; BM function was significantly higher in the +BB groups compared to the –BB groups. Results also showed that propranolol improved anemia by demonstrating a similar pattern for hemoglobin concentration as BM cellularity and BM components. Ferroportin, transferrin and BM TFR-1 demonstrated a similar pattern, suggesting improved iron delivery with +BB. Propranolol also seemed to have an effect on the EPO-hepcidin axis with the +BB groups showing improved iron homeostasis with increased levels of hepcidin and decreased levels of plasma EPO.
Through consistent, empiric findings, this elegant animal study stresses the importance of the sympathetic system on cellular function: an overabundance of sympathetic response leads to end organ dysfunction. While most studies on this autonomic phenomenon examine cardiac dysfunction, this study demonstrates such dysfunction at the liver and bone marrow. The authors implicate norepinephrine as the primary agent of sympathetic response in their injury model and suggest a model of resultant iron homeostasis dysfunction that results in anemia that cannot be managed by exogenous iron. As increasing data demonstrate worse outcomes for patients receiving transfusions, the use of a simple (and relatively inexpensive) drug like propranolol or other techniques used to modulate sympathetic response may improve rates of anemia without the dangers associated with transfusion. Clearly, these data need to be translated to humans before widespread application.
Early, goal-directed mobilisation in the surgical intensive care unit: a randomised controlled trial. Schaller SJ, Anstey M, Blobner M, Edrich T, Grabitz SD, Gradwohl-Matis I, Heim M, Houle T, Kurth T, Latronico N, Lee J, Meyer MJ, Peponis T, Talmor D, Velmahos GC, Waak K, Walz JM, Zafonte R, Eikermann M. Lancet. 2016 Oct 1;388(10052):1377-1388.
This is an international (U.S.A., Austria & Germany), multicenter, randomized controlled trial, using an early, goal-directed mobilisation model (SICU Optimal Mobilisation Score, SOMS). The authors enrolled adult SICU patients who required mechanical ventilation (MV) for less than 48 hours, were expected to require MV for at least another 24 hours and who were functionally independent at baseline. They excluded patients who had been admitted >5 days before screening; a motor component of GCS <5; an irreversible disorder with a 6-month mortality >50%; high intracranial pressure; a cardiopulmonary arrest; unstable fractures contributing to probable immobility; been simultaneously included in another trial; an acute MI; a lower part of their legs missing; a rapidly developing neuromuscular disease; a ruptured or leaking aortic aneurysm or were pregnant. Randomisation was a block design stratified by GCS and APACHE II. SOMS intervention began no later than 1 day after enrollment, and mean SOMS during SICU stay was the primary outcome. SICU length of stay (LOS) and mini-modified functional independence measure scores (mmFIM) were key secondary outcomes. Functional capacity, muscle strength (MRC), quality of life (QOL), hospital LOS, mortality, discharge disposition, ICU delirium-free days, ventilator-free days and medication usages as well as laboratory results were also observed. Power analysis indicated n=100 per group was sufficient.
200 eligible patients were randomly assigned to standard treatment (96 [48%]) or early, goal-directed mobilisation therapy (104 [52%]). Intention-to-treat analyses revealed the intervention group had higher mean SOMS (p<0.0001); shorter ICU LOS (mean decrease of 3 days, p=0.0054); higher mmFIM scores (p=0.002 at hospital discharge, p=0.009 at ICU discharge); shorter hospital LOS (p=0.011); and higher rate of discharge home (61% vs. 30%, p<0.0001). There was no difference in mortality rate, QOL, MRC or mobility unrelated outcomes except more ICU delirium-free days in the intervention group. These main outcomes were similar between groups in the per-protocol analysis. A higher percent of mobilisation-related adverse events (e.g. hypotension) were noted in the intervention group compared to the control group.
This complex, highly-detailed RCT demonstrates the benefit of early application of goal-directed mobilisation in the SICU. Although the population is rather narrowed by the inclusion and exclusion criteria (therefore limiting generalizability), the study creates a clear clinical picture of the patient for whom this study applies. Unlike previous work, this study did not demonstrate a difference in ventilator-free days; however, this study did demonstrate a similar decrease in rate of delirium seen in other studies. Importantly, the authors draw distinctions between their findings among trauma and major surgical patients and other data looking at ARDS, stroke and other patient subtypes, suggesting that mobilisation resource allocation should take into account a patient’s specific disease entity. Clearly, these data show a potential benefit in the era of value-based health care in the SICU and should bolster efforts to increase early-mobilization in the SICU.
Intraoperative Infusion of Dexmedetomidine for Prevention of Postoperative Delirium and Cognitive Dysfunction in Elderly Patients Undergoing Major Elective Noncardiac Surgery: A Randomized Clinical Trial. Deiner S, Luo X, Lin HM, Sessler DI, Saager L, Sieber FE, Lee HB, Sano M. JAMA Surg. 2017 Aug 16;152(8):e171505.
Post-operative delirium is a common complication in elderly patients who undergo major surgical interventions. Unfortunately, this disease process is associated with extended durations of hospital stay, increased costs, skilled nursing facility placement, and a greater rate of death at 1 year. Dexmedetomidine has been shown to decrease the rate of delirium when used as a sedative in the ICU likely due to its decreased reliance on opioids and neuroprotective properties resulting from α2-adrenergic agonist effect. The aim of this study was to determine whether the intra-operative use of dexmedetomidine ameliorates post-operative delirium, their primary outcome.
The authors performed double-blind, randomized, parallel-group placebo controlled clinical trial between 2008 and 2014 at major academic institutions throughout the United States. Written informed consent was obtained pre-operatively. Subjects were included if they were 68 years of age or older who were planning to undergo major elective non-cardiac operations including spine, thoracic, orthopedic, urologic, and general surgery under general anesthesia. Major surgery was defined as a planned minimum 2-day hospital stay. Patients were excluded if they had a history of dementia or scored a 20 or less on a Mini-Mental State Examination. Additional exclusion criteria included emergency surgery, cardiac or intra-cranial surgery, planned post-operative intubation, baseline sever handicap condition, illiteracy, Parkinson’s disease, life expectancy less than 6 months, dialysis, sick sinus syndrome, 2nd/3rd degree heart block, bradycardia, α2-adrenergic agonist contra-indication, ASA IV/V, or hepatic dysfunction. Patients were randomized to dexmedetomidine at 0.5 μg/kg/hour or placebo using a 1:1 block of 6 patients per center. The placebo and drug were blinded to the patient and clinicians. Only the pharmacy was aware of the randomization result. Benzodiazepines and nitrous oxide were avoided but all other anesthetic medications were permitted. General anesthesia was provided with propofol and/or sevoflurane. A 1-hour battery of cognitive and delirium tests was performed prior to surgical intervention via telephone to serve as a baseline. An intent-to-treat analysis was performed. Power was based on a 15% incidence of delirium with a 50% reduction in delirium in the drug group compared to placebo. 706 patients provided 80% power with a 0.05 α rate.
404 patients were randomized in total with 197 in the study group and 202 in the placebo group. The study was stopped early due to rules as set forth by the study DSMB. The reason for early stopping was due to futility based on the interim analysis; there was no mathematical possibility of showing benefit if the study completed enrollment. There was an 11.8% incidence of post-op delirium; 23 (12.2%) in the study group and 23 (11.4%) in the placebo group (OR 1.06; 95% CI 0.79-1.41, p=.77). The baseline characteristics associated with delirium, no matter the treatment group, included lower educational level, baseline mile cognitive impairment, type of surgical procedure (orthopedic and spine greatest) and surgical times greater than 4 hours. 330 patients completed the cognitive battery at 3 months follow-up and 204 at 6-months. There was no difference in cognitive scores at follow-up among treatment groups; scores increased equally over time.
As a well conducted randomized controlled trial, there were few limitations. There were relatively few drop-outs and they had a reasonable follow-up rate. They did not evaluate the duration of delirium or the consequences of delirium such as dismissal to a skilled facility. Additionally, the authors did not report on intra-operative and post-operative morphine equivalents or other pain control and sedative use differences (including post-operative dexmedetomidine use) between groups.
The authors concluded that the intra-operative use of dexmedetomidine did not affect post-operative rates of delirium nor longer-term post-operative cognition. This is in contradiction to prior trials studying the use of dexmedetomidine post-operatively in the ICU. Perhaps intra-operative use of the drug in combination to post-operative use in higher-risk patients may provide more benefit.