|
Article 1
A Multicenter, Randomized Clinical Trial of IV Iron Supplementation for Anemia of Traumatic Critical Illness. Pieracci FM, Stovall RT, Jaouen B, Rodil M, Cappa A, Burlew CC, Holena DN, Maier R, Berry S, Jurkovich J, Moore EE. Crit Care Med. 2014 Sep;42(9):2048-57.
Anemia is a common problem in critically ill patients and its etiology is multifactorial. While chronic blood loss from medical therapies and hemodilution contribute to anemia, inflammation can disrupt normal erythropoiesis and iron metabolism. Since most critically ill surgical patients were iron deficient in a previous study, the authors designed a multicenter randomized trial testing whether IV iron therapy would reverse iron deficiency and decrease transfusion requirements.
Inclusion criteria were adult trauma patients at four Level 1 trauma centers with anemia (hemoglobin < 12 mg/dl) and an expected intensive care unit length of stay > 5 days. Exclusion criteria were designed to identify patients with chronic disorders involving iron deficiency, disorders of erythropoiesis, or inflammation. Subjects were randomized to receive placebo or 100 mg iron sucrose three times per week until either ICU discharge or 6 total doses. Follow up was performed for 42 days or until hospital discharge.
Baseline demographic variables were similar for the 150 randomized subjects except the iron group had more male patients. Both groups had evidence of functional iron deficiency but iron supplementation did not change hemoglobin values or decrease transfusions. No increased risk of infection was associated with iron supplementation and mortality between groups was similar. The authors conclude that routine iron supplementation cannot be recommended for anemic critically ill surgical patients. The short duration of therapy, small group size of subjects requiring prolonged (>14 days) ICU stay, and relatively low dose of iron supplementation could have contributed to these results.
Article 2
High-Protein Enteral Nutrition Enriched With Immune-Modulating Nutrients Vs Standard High-Protein Enteral Nutrition And Nosocomial Infections In The ICU: A Randomized Clinical Trial. van Zanten AR, Sztark F, Kaisers UX, Zielmann S, Felbinger TW, Sablotzki AR, De Waele JJ, Timsit JF, Honing ML, Keh D, Vincent JL, Zazzo JF, Fijn HB, Petit L, Preiser JC, van Horssen PJ, Hofman Z. JAMA. 2014 Aug 6;312(5):514-24.
This trial is a double blind, randomized controlled, multi-institutional trial conducted in Europe (Germany, Netherlands and Belgium) from 2010 to 2012 that examined the potential outcome benefit of “immunonutrition” in the critically ill patient when compared to high protein controls. The trial involved 301 patients in total of which 109 patients were critically ill as a result of trauma. The majority of the nutrition support was delivered through the enteral route (13-15% received supplemental parenteral nutrition) and the immunonutrition arm consisted of glutamine, omega-3 fatty acids, selenium and anti-oxidants. The control diet was iso-nitrogenous and iso-caloric. There was no reported difference between either arm with regard to morbidity or mortality. At 6 months, patients in the immunonutrition arm had a higher adjusted mortality.
Immunomodulating formulas are commonly utilized in the setting of the critically injured trauma patient based largely on small randomized trials and theoretical benefit. This is the largest study addressing the question to date in a heterogeneous ICU patient population. Limitations of the study include the lack of additive arginine which is routinely included in most available immunonutrient formulas in the United States. This study echoes the findings of the SIGNET (1), EDEN-OMEGA (2) and REDOX (3) trials demonstrating exogenous administration of anti-oxidants, glutamine, and omega-3 alone or in combination without arginine fails to improve outcome. In summary, enteral formulas fortified with glutamine, omega-3 fatty acids, and anti-oxidants demonstrated no outcome benefit in trauma patients and potential harm in medical patients.
Article 3
Lower Versus Higher Hemoglobin Threshold For Transfusion In Septic Shock (TRISS Trial). Holst LB et al, TRISS Trial Group. N Engl J Med. 2014 Oct 9;371(15):1381-91.
This was a multicenter, non-blinded, randomized control trial designed to evaluate the effects on mortality of leuko-reduced blood transfusion at a lower versus a higher hemoglobin threshold among patients with septic shock in the ICU. The original transfusion threshold was set forth in the treatment protocol outlined in the Emanuel Rivers study. That early goal directed therapy protocol called for blood transfusion to a hematocrit ≥ 30% within the first six hours of ongoing hypoperfusion. The authors set to evaluate it within this specific septic shock patient cohort.
Inclusion criteria were patients 18 years or older in the ICU, fulfilling criteria for septic shock, who had a blood concentration of hemoglobin of 9 g per deciliter or less. The study included 1005 patients. Patients were excluded if they declined transfusion, had previous adverse transfusion reactions, had already received blood in the ICU, had acute coronary syndrome, and if they were experiencing life-threatening bleeding. The groups were then randomly assigned to receive single units of leuko-reduced red cells in the lower threshold group (Hgb ≤ 7g/dL) versus the higher threshold group (Hgb ≤ 9g/dL). The intervention period was the entire ICU stay up to a maximum of 90 days after randomization.
The authors concluded that patients in septic shock who received transfusion based on the lower threshold as compared to the higher threshold, had received fewer transfusions (median of 1U PRBC’s vs 4U) and had similar 90-day mortality rates (43% vs 45%), use of life support, and numbers of days alive and out of the hospital (30% vs 31%). Furthermore, the numbers of patients with ischemic events (7.2% vs 8%) and severe adverse reactions to blood (0% vs 0.2%) were also similar between the two intervention groups.
Strengths of the study include the adequate sample size, the low risk of bias given randomization and blinding of the statisticians involved, and the generalizability of the findings, given the study’s conduction in university and non-university hospitals as well as its pragmatic design. Limitations include its non-blinded nature of treating physicians and their patients, the wide confidence interval for the point estimate of mortality, and the limited power to detect differences in the secondary outcomes studied.
Article 4
Crystalloid Administration During Trauma Resuscitation: Does Less Really Equal More? Sharpe JP, Magnotti LJ, Croce MA, Paulus EM, Schroeppel TJ, Fabian TC, Weinberg JA. J Trauma Acute Surg. 2014 Dec;77(6):828-32; discussion 832.
This study was a prospective design over a 6.5 year time period evaluating the effect of crystalloid resuscitation during initial resuscitation of the trauma patient. Inclusion criteria included all patients who received at least 1 unit of PRBC’s in the resuscitation room. 383 patients were included over the time period of the study were grouped according to crystalloid/PRBC ratio (C/PRBC). The C/PRBC ratio was defined at the ratio of crystalloid infused in liters to the units of PRBC’s transfused in the resuscitation room, and two groups were identified, high ratio (≥.75) and low ratio (≤.75). Outcomes included 24 hour mortality, in hospital mortality, ventilator free days, and Pa02/Fi02 (P/F) ratios at 48 hours of hospitalization. ARDS was defined at P/F ratio at 48 hours less than 200.
No difference was found between the two groups in respect to age, sex, mechanism of injury, admission systolic blood pressure, admission base excess, and Glasgow Coma Score (GCS). Patients in the high group did have a higher mean ISS (32 vs 27), admission heart rate (118 vs112, p=0.02), and spent more time in the resuscitation room (50 vs 40 minutes, p≤0.001). The high ratio group also received fewer PRBC’s (mean 2.7 U vs 4.0 U p=0.001), fewer platelets (0.05 U vs 0.70 U, p=0.04), and more crystalloid (mean, 4.0L vs 1.3L, p≤0.001). The volume of crystalloid in the high-ratio group ranged from 1.0 L to 11.6 L and in the low group 0.5L to 8.2L.
On univariable analysis, the low-ratio group did have a lower incidence of ARDS (24% vs. 35%, p=0.03), but no other differences between groups were found for other study endpoints. However, after regression analysis C/PRBC was not found to be significantly associated with ARDS (OR 1.43; 95% CI 0.77-2.73), 24 hour mortality (OR 0.89; 95% CI, 0.49-1.63), or in-hospital mortality (OR 0.89; 95% CI, 0.52-1.53). Regression analysis identified ISS, admission base excess and, time in the resuscitation room as independent predictors of in hospital mortality. ISS alone was an independent predictor of 24-hour mortality and ARDS.
The authors conclude that previous associations of early crystalloid administration with mortality and the development of ARDS could likely be secondary to intention bias, with the use of crystalloid serving as a surrogate for severely injured patients, rather than true deleterious effects of crystalloid resuscitation early in the resuscitation room. Further, they conclude that the concern regarding the morbidity associated with high-volume crystalloid may be excessive and that use of crystalloid during early resuscitation cannot be condemned.
Article 5
A Clinical Trial of Progesterone for Severe Traumatic Brain Injury. Skolnick BE, Maas Al, Narayan RK, van der Hoop RG, Macallister T, Ward JD, Nelson NR, Stocchetti N. N Engl J Med. 2014 Dec 25;371(26):2467-76.
This trial is one of two phase 3 multi-center trials conducted to further evaluate the effects of progesterone in severe traumatic brain injury (GCS <= 8). More than 180 previous pharmacologic studies have demonstrated a neuro-protective benefit with progesterone administration. There have also been two prospective, controlled, single center trials that demonstrated a clinical benefit. This trial was conducted as a randomized, prospective, multi-national (21 participating countries), placebo controlled trial. 1195 patients were enrolled in the trial aged 16-70. Progesterone dosing began within 8 hours of injury and was continued for 120 hours. The primary efficacy endpoints were the Glasgow Outcome Scale (GOS) score at 6 months after injury and mortality. Statistical analysis demonstrated no treatment effect of progesterone compared to placebo, with similar proportions of favorable GOS scores (50.4% vs. 50.5%). There was no difference seen in mortality.
The authors conclude that progesterone administration was not clinically effective in the treatment of severe TBI. They postulate that the failure of this trial and many other TBI trials is likely multi-factorial. Some proposed issues include the complexity and variability of the injuries, the uni-dimensional characterization of TBI (often based on single scores such as GCS), and limitations in translating experimental data to humans.
|