Vitamin A treatment for severe sepsis in humans; a prospective randomized double blind placebo-controlled clinical trial.

PURPOSE: To test the benefits of Vitamin A treatment in patients with sepsis on length of time in ICU, days on ventilator, days on intravenous blood pressure support and 28-day mortality. The trial was prospective, randomized and double-blind. As part of a larger sepsis trial, 63 patients with sepsis were randomized to receive either 100,000 IU of Vitamin A intramuscular or placebo over 7-days. Data analysis was by ANOVA with two tailed test and p < 0.05 as significant. RESULTS: The mean age was 51 ± 2 (mean ± SEM) with 54% female. Groups were well matched with regards to APACHE III score, WBC count, and incidence of bacteremia. In addition, all patients had an ACTH stimulation test using 250 mcg of ACTH IV and serum cortisol was measured at time 0, 30 and 60 min. Baseline cortisol of 24.6 ± 1 mg/dl increased to 41 ± 2 mg/dl at 30 min and 49 ± 2 at 60 min. There was no significant difference between the groups. All cortisol responses were greater than 11.9 mg/dl. Serum Vitamin A level was below normal in 54% of the patients. After randomization, 100,000 IU of Vitamin A daily was given to 32 patients and blinded placebo was given daily to 32 patients for seven days. This was administered as a 1 cc injection of either medication or placebo and was blinded from all but the research pharmacist. The number of days in the ICU was slightly, but not significantly reduced (p = 0.39) by approximately 2 days in the Vitamin A treated patients. The average number of days on blood pressure agents and the day on ventilator were similar. The 28-day mortality rates were similar between the two groups (28 vs 34% placebo vs Vitamin A group). Seven days of high dose intramuscular Vitamin A treatment in patients with sepsis where approximately 50% were vitamin A deficient had no benefit in adults with sepsis.

Prevalence of Malnutrition Risk and the Impact of Nutrition Risk on Hospital Outcomes: Results From nutritionDay in the U.S.

BACKGROUND: Malnutrition risk estimates vary greatly, and no robust data on the association between food intake and outcomes exist for hospitals in the United States (U.S.). This study aimed to determine the prevalence of malnutrition risk and to evaluate the impact of food intake on mortality using the nutritionDay in the U.S. dataset. METHODS: This study analyzed data from 2009 to 2015 for all adult patients from participating hospitals. Prevalence of malnutrition risk was determined by mapping self-reported nutritionDay survey questions to the Malnutrition Screening Tool (MST). Fine and Gray competing-risk analysis with clustering was used to evaluate the impact of nutrition risk and food intake on patients' 30-day in-hospital mortality, while controlling for age, mobility, and other disease-related factors. RESULTS: Analysis included data from 9959 adult patients from 601 wards. The overall prevalence of malnutrition risk (MST score ≥2) was 32.7%. On nutritionDay, 32.1% of patients ate a quarter of their meal or less. Hospital mortality hazard ratio was 3.24 (95% CI: [1.73, 6.07]; P-value < 0.001) for patients eating a quarter compared with those who ate all their meal and increased to 5.99 (95% CI: [3.03, 11.84]; P-value < 0.0001) for patients eating nothing despite being allowed to eat. CONCLUSION: This study provides the most robust estimate of malnutrition risk in U.S. hospitalized patients to date, finding that approximately 1 in 3 are at risk. Additionally, patients who have diminished meal intake experience increased mortality risk. These results highlight the ongoing issue of malnutrition in the hospital setting.