Publication:

Even preoperative carbohydrate loading is too hard? Why RCT’s should not be considered the gold standard for nutrition research in acute hip fracture: results of a feasibility study

Background Malnutrition is a high-risk co-morbidity in acute hip fracture patients. Pre-operative carbohydrate loading may improve nutritional status and therefore patient outcomes. The feasibility of nutrition focused randomised control trial designs in hip fracture is at best questionable. This study was designed to undertake efficacy testing of pre-operative carbohydrate loading and explore the broader feasibility of conducting randomised controlled trials in acute hip fracture.

Methods This two arm randomised controlled feasibility study recruited patients previously living in the community who had fractured their hip undergoing surgery at our institution. Patients in the intervention arm received a 400 mL (50g) carbohydrate load 2 hours prior to surgery.  Information was collected regarding the fidelity of pre-operative carbohydrate provision and consumption as well as patient demographic and admission details.

Results Thirty-two patients consented to participate, 60% of the eligible patient cohort. Results demonstrated evenly matched intervention and control groups in terms of demographic details and pre-surgical morbidity and mortality risk. However, of the 17 patients allocated to the intervention arm less than half (41%) completed the carbohydrate loading intervention and even fewer 23.5% (n=4) completed all follow up due to a number of patient and system related factors.

Conclusions Evaluating the clinical effectiveness of providing pre-operative carbohydrate loading in hip fracture and the associated outcomes is not feasible using a randomised control trial methodology. It is recommended that researchers consider a ‘silver standard’ of research and practice such as pragmatic, registry-based cluster randomised trials to ensure feasibility, relevancy and applicability when evaluating nutritional interventions in this cohort.