A Randomized, Double Blind, Placebo-Controlled Trial of Nutritional Supplementation During Acute Illness

Background:  Recent studies that have found protein-energy undernutrition to be a strong predictor of inpatient and post discharge clinical outcomes.  The elderly are at a high risk for undernutrition, and previous studies indicate that those with poor nutritional status during the first month after hospitalization are at a much higher risk for non-elective readmission. 

 

Objective:  The purpose of this study was to assess whether nutritional support of older patients during acute illness leads to improvement in clinical outcomes. 

 

Methods:  This was a randomized, double blind, placebo-controlled, parallel group study.  The inclusion criteria consisted of patients 65 years or older, able to swallow, and able to sign an informed written consent form.  Patients were excluded if they had undergone gastric surgery, were diagnosed with malabsorption or morbid obesity, in a coma, diagnosed with severe dementia or malignancy, living in an institution, or already taking supplements.  Of the 445 patients randomized, 223 were allocated to the supplement group and 222 were allocated to the placebo group.  Those in the supplement group received 2 bottles (200 ml each) of nutritional supplement daily at 8 am and 12 pm in addition to the standard hospital diet.  The supplement consisted of 995 kcal of energy and 100% of the Reference Nutrient Intakes of vitamins for healthy older people.  The treatment duration was for 6 weeks.  The clinical outcome measures were 6 months of disability, non-elective readmission, length of hospital stay, discharge destination, morbidity, and mortality.  Nutritional outcomes were assessed through body mass index, body weight, mid upper arm circumference (MUAC), triceps skin fold (TSF), serum albumin, serum transferrin, red-cell folate, and plasma vitamin B12 concentrations.  An intent-to-treat analysis was used, and patients had assessments at baseline, 6 weeks, and 6 months post randomization.  A Cox proportional hazards model was used to examine the 6-month risk of non-elective readmission and mortality between the 2 groups after controlling for a number of non-nutritional factors.  An 80% power with a 5% 2-sided significance was used. 

 

Results: No statistically significant differences existed between groups in body weight, body mass index, MUAC, TSF, or serum transferrin.  The supplement group had significant increases in serum albumin concentration at 6 months (40.5 g/L with placebo and 42 g/L with supplement; p = 0.04).  The supplement group also had increased concentrations at the end of 6 weeks in red-cell folate (p = 0.001) and vitamin B12 (p = 0.002) compared to placebo.  Over the 6-month period, 65 patients in the supplement group and 89 patients in the placebo group were readmitted to the hospital (adjusted hazard ratio 0.68; 95% CI 0.49-0.94; p = 0.02).  No other statistically important differences between the supplement and placebo group were found in length of hospital stay, infections, disability, discharge destination, or mortality.  Nausea was the most common side effect, reported by 22% of the patients with no difference between the supplement and placebo group.  Approximately 15% of the placebo group and 10% of the supplement group were adherent to taking 64-84 supplement drinks.  The authors concluded that nutritional supplementation of hospitalized elderly people led to improved nutritional status and a significant reduction in the number of non-elective readmissions.

Strengths:  The study design was optimal, since randomized, double blind, and placebo controlled are the gold standards for efficacy studies.  All patients and investigators were blinded, and identical treatments were used. The success of blinding was assessed by a pilot trial of the placebo and active treatments among volunteers, which resulted in no one distinguishing between the placebo and actual supplement.  All outcome measures were clearly defined for either nutritional or clinical outcome, and researches independently verified the collected data throughout the study.

 

Weaknesses:  The inclusion/exclusion criteria did not accurately represent the population of interest.  The authors enrolled a patient with no regard to concomitant medications and included patients that had very different admission diagnosis.  It was difficult to determine how this particular combination of chronic conditions, fractures, or surgeries is impacted by nutritional status.  The amount of kcal and nutrients in the supplement treatment might not have been appropriate for every patient, and the amount of nutrients in each supplement was not specified.  The follow up that occurred at the 6-month point was about 18 weeks after the patients had discontinued the treatments, which was an inappropriate duration to determine the effects from the supplements.  It was unclear what happened when a patient was discharged before the end of the 6-week treatment period, and the authors did not mention if the changes in diet before and after discharge were accounted for.  Even though food diaries were given out, nowhere in the study was the dietary intake of the patients discussed and how nutritional status might have been affected.  It was unclear how many patients were in each analysis.  The authors stated that an intent-to-treat analysis was used, but the nutritional outcomes did not involve all patients randomized, and the authors never mentioned why all the patients were not included.  The statistically significant results did not appear to be clinically significant, and the authors never reported any of the reasons for patient readmissions.  Compliance was very poor and could have greatly impacted the results. 

 

Conclusion:  While the rationale behind the study’s objective provides reason for exploring the benefits of supplements in the acutely ill elderly, the methods and data handling of this study weakens its findings.  Given all the study’s limitations, there is not enough evidence to support the routine use of nutritional supplements in acutely ill patients.  Additional support is needed from other controlled trials that will have a stronger design and factor in more of what this study lacked.  Perhaps future trials could be more specific with their target population, focusing on supplementation in a certain disease state or physical condition in the elderly that has been linked with nutritional status as a prognostic indicator.

 

Gariballa S, Forster S, Walters S, Powers H.  A randomized, double-blind, placebo-controlled trial of nutritional supplementation during acute illness.  Am J Med 2006;119:693-699. 

 

Lucy A. Dalzot, Pharm.D. Candidate