Trial Review

Registering a new trial?

To achieve prospective registration, we recommend submitting your trial for registration at the same time as ethics submission.

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been endorsed by the ANZCTR. Before participating in a study, talk to your health care provider and refer to this information for consumers
Trial registered on ANZCTR


Registration number
ACTRN12621000300875
Ethics application status
Approved
Date submitted
21/12/2020
Date registered
18/03/2021
Date last updated
18/03/2021
Date data sharing statement initially provided
18/03/2021
Date results information initially provided
18/03/2021
Type of registration
Retrospectively registered

Titles & IDs
Public title
Carbohydrate intake and refeeding syndrome in children and adolescents with anorexia nervosa
Scientific title
Comparison of a low carbohydrate intake and standard carbohydrate intake on the risk of refeeding syndrome (hypophosphatemia) in children and adolescents with anorexia nervosa
Secondary ID [1] 303070 0
Nil known
Universal Trial Number (UTN)
Trial acronym
Linked study record

Health condition
Health condition(s) or problem(s) studied:
Eating disorder - anorexia nervosa 320140 0
Condition category
Condition code
Mental Health 318082 318082 0 0
Eating disorders

Intervention/exposure
Study type
Interventional
Description of intervention(s) / exposure
A standard carbohydrate (CHO) mealplan was used which provides 50-60% of total energy from carbohydrate as per the Australian Guide to Healthy Eating recommendations. The starting calorie intake of this mealplan is matched to the control group with increments to calories over the first 7 days as per usual care.
The starting caloric prescription of the meal plan was assessed by the dietitian following a comprehensive nutrition assessment including anthropometric measures, malnutrition diagnosis and recent food and fluid intake. Participants were commenced on a meal plan of oral food and fluid providing a minimum of 2000kcal (8400kJ). Participants deemed at high risk of refeeding syndrome were those who were less than 70% of their expected body weight on admission, or who had minimal carbohydrate intake for 7-10 days. These participants were commenced on a meal plan of 1500kcal/day (6300kJ).
The meal plan includes 3 main meals (breakfast, lunch and dinner) and 3 snacks (morning tea, afternoon tea and supper). The meal plans were increased incrementally by approximately 400kcal twice weekly, until the participant reached a meal plan of 3000kcal (12600kJ). This was usually achieved by day 7 of admission (and by day 10 for those starting on 1500kcal/day). Following this, increases to meal plans were dependent on adequacy of weight gain with the expectation of 1-1.5kg weight gain per week as per local hospital guidelines. Once the participants reached the 3000kcal meal plan there was no further difference in the carbohydrate content of the meal plans, with carbohydrates providing 50-60% total energy in each meal plan.
The dietitian was responsible for choosing the appropriate starting mealplan (as stated above) and ordering the prescribed meals and snacks from the hospital food service department. The nursing staff were responsible for setting the participants up at meal times with the prescribed mealplan. All meals and snacks were supervised and supported by nursing staff. If participants were unable to consume the entirety of their prescribed meal or snack they were required to have a nutritionally equivalent supplement drink or “bolus”. This was initially offered orally, however if the participant was unable to consume it orally it was administered via a nasogastric tube.
As per local hospital guidelines, nursing staff recorded oral intake after each meal/snack on Cerner under fluid balance tab. The dietitian would review this information twice weekly. Participants were on supervised bed rest following meal and snack times and bathroom visits were supervised. Locker searches were performed by nursing staff if there was suspicion of food hiding.
Patients were recruited within 24hrs of admission. The intervention was commenced immediately following recruitment (i.e at the next meal or snack). Participants were reviewed by the dietitian twice weekly until discharge.
Participants were medically reviewed daily to monitor for clinical features of refeeding syndrome (RFS) including signs of congestive cardiac failure, confusion, and seizures. Participants were monitored closely for biochemical markers of RFS, with analysis of electrolytes, calcium, magnesium, phosphate, glucose daily for 7 days and then twice weekly thereafter. In those patients deemed at high risk for RFS, biochemical markers were evaluated daily for 10 days and twice weekly thereafter during the admission. Prophylactic phosphate was not routinely administered. Phosphate was prescribed in the form of Sandoz Phosphate 500mg twice per day with titration as indicated if refeeding hypophosphatemia (RH) occurred (<0.95mmol/L in patients under 16 years and <0.87mmol/L in patients >16 years as per local hospital guideline). A general multivitamin was not provided to participants.



Intervention code [1] 319358 0
Treatment: Other
Comparator / control treatment
A low carbohydrate mealplan was used which provides <40% of total energy from carbohydrate as per standard care which is based on literature recommendations.

The starting caloric value of both the standard CHO (treatment) and low CHO (control) meal plan was assessed by the Dietitian following a comprehensive nutrition assessment including anthropometric measures, malnutrition diagnosis and recent food and fluid intake.

The control mealplan was commenced on admission for all patients. Following randomisation to the control arm, participants continued on the control mealplan if allocated. The mealplan was made up of oral food and fluid and provided a minimum of 2000 calories (8400kJ).

Patients deemed at high risk of refeeding syndrome included those who were less than 70% of their expected body weight on admission, or who had very low carbohydrate intake for an extended period of time. These high-risk patients were commenced on a meal plan of 1500 calories/day (6300kJ).

Meal plans were increased incrementally by ~400 calories twice weekly, until the patient reached a meal plan of 3000 calories (12600kJ) usually by day seven of admission (but day 10 for those starting on 1500 calories/day). Following this, increases in meal plans were dependent on adequacy of weight gain with the expectation of a minimum of 1kg weight gain per week as per local guidelines. Once the patients reached 3000 calories there was no further difference in the carbohydrate content of the meal plans between the two groups with carbohydrates providing 50-60% total energy.

All meals and snacks were supervised by nursing staff. If patients were unable to consume the entirety of their prescribed food they were required to have a nutritionally equivalent supplement drink or “bolus”, either orally or via nasogastric tube. Nutritional intake was documented on the fluid balance chart and reviewed by the dietitian twice weekly.

Patients were also on supervised bed rest following meal and snack times and bathroom visits were supervised as per standard care.

Participants were reviewed by the dietitian twice weekly until patient discharge.
Control group
Active

Outcomes
Primary outcome [1] 326070 0
Incidence of hypophosphatemia.
Blood sample taken as per standard practice by pathology nurse.
Serum phosphate result recorded on Cerner and transferred to data collection sheet by dietitian.
Timepoint [1] 326070 0
Daily during first 7 days of admission.
Secondary outcome [1] 390017 0
Comparison of weight changes in week one and over course of admission.
Weight measured using digital scale. Participants weighed backwards by nursing staff so as not to see weight.
Timepoint [1] 390017 0
Weight taken at baseline and twice weekly thereafter until discharge (this is usual care).
Secondary outcome [2] 390018 0
Composite secondary outcome - changes to other biochemical markers associated with refeeding syndrome (i.e potassium and magnesium).
Timepoint [2] 390018 0
Bloods measured at baseline, then daily for first 7 days of admission, and then twice weekly thereafter until discharge (this is usual care).
Secondary outcome [3] 390987 0
Changes to blood glucose levels associated with refeeding syndrome.
Timepoint [3] 390987 0
Blood glucose level (BGL) measured four times a day for first week of admission.

Eligibility
Key inclusion criteria
- diagnosis of anorexia nervosa (restrictive or atypical) as per DSM-V
- patients admitted to the Paediatric and Adolescent medical ward
- 18 years of age or less
- expected to be inpatient for a minimum of 7 days
- managed according to the standard eating disorders protocol on medical ward
Minimum age
No limit
Maximum age
18 Years
Sex
Both males and females
Can healthy volunteers participate?
No
Key exclusion criteria
- >19 years of age
- patients not meeting criteria for diagnosis of anorexia nervosa as per DSM-V
- patients transferred from another hospital where nutrition rehabilitation has already started
- patients with low phosphate on admission
- patients not managed according to the usual eating disorder protocol
- patient or parent not willing to consent to the study

Study design
Purpose of the study
Treatment
Allocation to intervention
Randomised controlled trial
Procedure for enrolling a subject and allocating the treatment (allocation concealment procedures)
Patients were randomly allocated to treatment group (standard CHO) or control group (low CHO) through the use of a computer-generated block randomisation schedule (in permuted blocks of 4). The sequence of study entry was determined before the beginning of the study by an independent person not involved with recruitment. The group allocation was placed inside sealed opaque envelopes and numbered with participant numbers. The envelope was opened at the time of allocation of each participant. Researchers involved with participant recruitment will be blinded to the allocation sequence but will be un-blinded to the group allocations. Once allocated to either low or standard CHO diet groups, patients were no longer blinded from the researchers, in order to enable the researchers to provide the appropriate diet items for the participants. Participants were not informed of which diet group they were allocated to, although given they were required to orally consume the food items, they could be able to notice differences in the food provided to co-patients.
Methods used to generate the sequence in which subjects will be randomised (sequence generation)
Computerised sequence generation - block randomisation schedule
Masking / blinding
Blinded (masking used)
Who is / are masked / blinded?
The people receiving the treatment/s


The people analysing the results/data
Intervention assignment
Parallel
Other design features
Phase
Not Applicable
Type of endpoint/s
Safety/efficacy
Statistical methods / analysis
It was estimated that over a twelve month period, 60 patients would be recruited from the Austin Hospital Paediatric and Adolescent Inpatient Unit.. This estimate was based on ward statistics for admissions over the 2 previous years. A convenience sample was to be used. This estimate assumed 100% consent rate, when in reality consent rate proved to be low (26%).

Data is reported as the mean and standard deviation for quantitative factors. The incidence of hypophosphatemia and relationship between malnutrition status and hypophosphatemia was assessed using two-way repeated measures, ANOVA. Changes in body weight over the time of admission was completed by unpaired t test, except for the comparison of percentage point change in % EBW from admission to discharge which was also assessed by two-way repeated measures ANOVA to assess the effect of both time and intervention. Study findings were in assessed in terms of statistical significant via P values. Statistical analysis was completed using software. The feeding plans were analysed with Foodworks Professional.

Recruitment
Recruitment status
Stopped early
Data analysis
Data analysis is complete
Reason for early stopping/withdrawal
Participant recruitment difficulties
Date of first participant enrolment
Anticipated
Actual
28/09/2017
Date of last participant enrolment
Anticipated
Actual
18/09/2019
Date of last data collection
Anticipated
Actual
22/10/2019
Sample size
Target
60
Accrual to date
Final
24
Recruitment in Australia
Recruitment state(s)
VIC
Recruitment hospital [1] 18244 0
Austin Health - Austin Hospital - Heidelberg
Recruitment postcode(s) [1] 32308 0
3084 - Heidelberg

Funding & Sponsors
Funding source category [1] 307477 0
Charities/Societies/Foundations
Name [1] 307477 0
Austin Medical Research Foundation
Country [1] 307477 0
Australia
Primary sponsor type
Hospital
Name
Austin Health
Address
Austin Health
Studley Road
Heidelberg, 3084
VICTORIA
Country
Australia
Secondary sponsor category [1] 308149 0
None
Name [1] 308149 0
N/A
Address [1] 308149 0
N/A
Country [1] 308149 0

Ethics approval
Ethics application status
Approved
Ethics committee name [1] 307553 0
Austin Health Human Research Ethics Commitee
Ethics committee address [1] 307553 0
Austin Health
Studley Road
Heidelberg 3084
VICTORIA
Ethics committee country [1] 307553 0
Australia
Date submitted for ethics approval [1] 307553 0
21/11/2016
Approval date [1] 307553 0
10/03/2017
Ethics approval number [1] 307553 0
HREC/16/Austin/533

Summary
Brief summary
Nutritional rehabilitation is integral in the treatment of anorexia nervosa (AN) . During nutritional rehabilitation, the provision of adequate nutrition to achieve weight restoration and medical stability must be balanced against managing the risks of refeeding syndrome (RS). Refeeding hypophosphatemia (RH) is commonly used to indicate the risk for the development of RS. The risk of developing RS is the greatest during the initial 72 hours following commencing nutritional rehabilitation. Current guidelines support an aggressive feeding model, however suggest restricting calories from carbohydrates and including foods rich in phosphate during nutritional rehabilitation in order to avoid RS. However, despite this suggestion, there have been few studies that have investigated the effect of nutrition composition, specifically carbohydrate intake, on the incidence of RH in patients with AN.
The aim of this study was to compare a standard carbohydrate calorie matched aggressive feeding protocol with a low carbohydrate feeding protocol on the risk of refeeding syndrome (hypophosphatemia) in patients admitted to a child and adolescent eating disorder program. This study was a single centre randomised controlled trial. Consent was obtained from both the child or adolescent and their parent or guardian prior to participation in the study. The Paediatric and Adolescent Inpatient Unit at the Austin Hospital in Melbourne, Australia, provides tertiary level inpatient care for children and adolescents with eating disorders who require medical stabilisation. Whilst the preference is for outpatient treatment of the eating disorder, criteria for admission includes postural instability, bradycardia, dehydration, food refusal, rapid weight loss or being severely underweight (<75% expected body weight). Patients were randomly assigned via concealed allocation to either a low carbohydrate feeding plan which provided <40% of total energy from carbohydrate, as per current practice, or standard carbohydrate feeding plan which provided 50-60% of total energy from carbohydrate as per the Australian Guide to Healthy Eating recommendations. Calorie intake for both feeding plans was matched. Oral feeding was encouraged with oral bolus or enteral nutrition available if required. Patients were not prescribed any prophylactic nutrition supplementation during the admission, including multivitamin supplements. If phosphate levels were low, oral phosphate in the form of Sandoz phosphate was to be administered. The medical team monitored daily for any signs of RS and this was documented in the medical file. To our knowledge, there are limited studies that have investigated the link between carbohydrate intake and RH in orally fed children and adolescents. If carbohydrate intake doesn't need to be restricted in order to minimise the risks of RS, this may promote both an earlier and more normalised approach to eating and weight restoration.
Trial website
Trial related presentations / publications
Public notes

Contacts
Principal investigator
Name 107650 0
Mrs Kellie Draffin
Address 107650 0
Nutrition and Dietetics Department
Austin Health
Studley Rd
Heidelberg 3084
VICTORIA
Country 107650 0
Australia
Phone 107650 0
+61 03 9496 5011
Fax 107650 0
Email 107650 0
[email protected]
Contact person for public queries
Name 107651 0
Kellie Draffin
Address 107651 0
Nutrition and Dietetics Department
Austin Health
Studley Rd
Heidelberg 3084
VICTORIA
Country 107651 0
Australia
Phone 107651 0
+61 03 9496 5011
Fax 107651 0
Email 107651 0
[email protected]
Contact person for scientific queries
Name 107652 0
Kellie Draffin
Address 107652 0
Nutrition and Dietetics Department
Austin Health
Studley Rd
Heidelberg 3084
VICTORIA
Country 107652 0
Australia
Phone 107652 0
+61 03 9496 5011
Fax 107652 0
Email 107652 0
[email protected]

Data sharing statement
Will individual participant data (IPD) for this trial be available (including data dictionaries)?
No
No/undecided IPD sharing reason/comment
No individuals will be able to be identified in the data collection. All data (paper and digital) is stored in a locked, research draw in the Nutrition and Dietetics department or stored within a password protected file on the Nutrition and Dietetics drive. All information collected for this research project has identifying information removed and kept private, confidential and secure. Information is in coded form. The completed study outcomes will report group data only; no individual data will be presented.


What supporting documents are/will be available?

No Supporting Document Provided



Results publications and other study-related documents

Documents added manually
No documents have been uploaded by study researchers.

Documents added automatically
No additional documents have been identified.