Article Text
PDF
Abstract
Objective Glucose polymer-based emergency feeds (EF), used during illness to prevent metabolic decompensation and encephalopathy in inherited metabolic disorders, should be produced accurately and safely.
Design In a randomised, prospective, controlled study, the aim was to investigate if when preparing age-appropriate EF, a pre-measured sachet of glucose polymer, compared with scoops and weighing (using digital scales), decreased carer errors.
Subjects 47 carers (3 men, 44 women) of 52 inherited metabolic disorders patients were recruited.
Setting and intervention The carers made EF using all three techniques (weighing, scoops and pre-measured sachets) under supervision in controlled and home conditions. A 100-ml aliquot of each EF was analysed for carbohydrate concentration.
Results Under controlled preparation conditions, with 1 litre EFs, the % median glucose polymer concentration closest to target amounts was (1) pre-measured sachets (105%), (2) weighing (107%) and (3) scoops (118%; p<0.001). Similarly, under home conditions, the closest method was (1) pre-measured sachets (111%), (2) weighing (112%) and (3) scoops (118%; p<0.05). Under home preparation conditions, with 200 ml EFs, the pre-measured sachets were more likely to be within 20% of target weight than weighing (p<0.05), but there was no difference with scoops. Common errors observed were inaccurate water measurements (40% controlled and home conditions), incorrect scoop measurements and difficulty using digital scales.
Conclusions Overall, using pre-measured sachets was more accurate in EF production. Pre-measured sachets are likely to decrease preparation error and, therefore, reduce the risk of feed intolerance, particularly osmotic diarrhoea and consequential metabolic decompensation and encephalopathy.
Statistics from Altmetric.com
Introduction
In order to minimise the effect of catabolism, children with many inherited metabolic disorders are advised to consume emergency feeds (EF) as soon as their carers recognise any minor symptoms of illness, poor appetite or signs of deteriorating metabolic control.1,–,6 EF are age-related, concentrated carbohydrate solutions, with glucose polymer being the carbohydrate of choice.7,8 It is mixed with water and administered throughout the day and night. It is essential that EF is prepared accurately, using a method associated with minimum error. The preparation of very dilute or over concentrated EF may exacerbate metabolic decompensation and even encephalopathy.
What is already known on this topic?
▶. In inherited metabolic disorders, severe metabolic decompensation is precipitated by catabolism; in combination with poor energy intake.
▶. Administration of carbohydrate (usually glucose polymer) based emergency feeds, given regularly throughout the entire 24 h is recommended during illness or trauma.
What this study adds?
▶. Preparation errors are common in emergency feed preparation under controlled and home conditions.
▶. Emergency feed preparation technique deteriorates under home conditions.
▶. The use of pre-measured sachets of glucose polymer reduced feed preparation mistakes but some error still occurred.
Carers of inherited metabolic disorders children are issued with age-appropriate EF instructions, glucose polymer and scoops or weighing scales to measure glucose polymer powder. No “ready to use” glucose polymer feeds are available. Considering that EF is usually prepared under stress during a child's illness, an EF requiring minimum preparation may be a safer option. Standardised age-appropriate sachets of glucose polymer (SOS range (Vitaflo International, Liverpool, UK)) have been developed for EFs. In this randomised study, the aim was to investigate if the use of age-appropriate pre-weighed sachets of glucose polymer decreased the number of carer errors in EF production when compared with two conventional methods of EF measurement.
Methods
Subjects
Forty-seven carers (3 men, 44 women) of 52 patients with inherited metabolic disorders from the West Midlands (UK) were recruited. In one case, both parents shared responsibility for EF production so both carers of one child took part in this study; therefore, 53 observations were completed. Of the 47 carers, 60% were of Asian origin (n=28), 32% white Caucasian (n=15) and 8% (n=4) African-Caribbean.
There were three inclusion criteria: patients were (1) aged 0–20 years; (2) inherited metabolic disorders diagnosis and (3) principal EF ingredient is glucose polymer. Exclusion criteria included carers of patients with clinical conditions where glucose polymer was contraindicated, for example, Citrin deficiency; and where a protein substitute was prescribed as part of the EF, for example, Maple Syrup Urine Disease.
Of 52 patients, 19 had organic acidaemias (eg, propionic acidaemia, methyl malonic acidaemia); 9 glycogen storage disease (GSD) types I and III; 9 long chain fatty acid oxidation disorders; 8 medium-chain acyl-CoA dehydrogenase deficiency (MCADD), 6 urea cycle disorders and 1 lysinuric protein intolerance.
Before the study, methods used to measure glucose polymer comprised: weighing by digital electronic scales, 39% (n=20); scoops, 58% (n=30); glucose polymer sachets, 2% (n=1). Only 2% (n=1) used commercial high carbohydrate drinks. Seventeen (33%) carers did not speak English as their first language. An Asian-speaking interpreter (Urdu and Punjabi) was available for non-English-speaking carers throughout the trial.
The study was approved by the Medical Ethics Committees of South Birmingham Health Authority. Informed consent was obtained from all carers.
Product composition
Two types of glucose polymer were used:
Super Soluble Maxijul (SHS International)
SOS (Vitaflo International) comprised of four different sizes of pre-weighed sachets of glucose polymer (SOS 10, 15, 20 and 25), each equivalent to 10%, 15%, 20% and 25% carbohydrate solutions, respectively, when made up to 200 ml with water (table 1).
Product specification (per 100 g powder) of glucose polymer powders
Study design
This was an 8-week, single-centre, two-stage, randomised, controlled, prospective study in which at each stage, three different methods for preparing EF using glucose polymer were compared. The three methods were: (1) weighing using digital scales, (2) scoops and (3) pre-measured age-appropriate sachets.
Part 1
The carer who usually prepared the EF was invited to attend a research centre to prepare age-appropriate EFs. They prepared 1 litre of the EF using each of the three methods. At the study start, all carers received the same verbal and written explanation by the same inherited metabolic disorders dietitian with individualised, age-appropriate EF recipes. All issued instructions were in the same detail as given in clinic. Carers were shown how to measure glucose polymer powder using:
Digital scales (1 g accuracy): use of the scales was demonstrated and digital measurement reading understanding was checked.
Large scoops (each weighing 25 g glucose polymer); instructions were given on correct filling of scoops, number of scoops and using the flat edge from the back of a knife to level each scoop.
The pre-measured sachets of glucose polymer; carers were shown how to open and empty each sachet.
For all three methods, carers were instructed to add water to 1000 ml indicated by the graduated jug. Three workstations were provided, each with hot water, 2 litres, graduated, plastic measuring jugs, appropriate scoops and digital electronic scales, whisks, metal spoons, 2.5-kg containers of glucose polymer and age-appropriate SOS sachets of glucose polymer.
One of two observers watched each carer prepare EFs, every carer always having the same observer for all three methods of feed preparation. The following criteria were assessed:
▶. Use of digital scales: the ability of carers to weigh powder within 1 g of target amount.
▶. The number of powder scoops added compared with the target amount.
▶. Using the level “back” of a knife to “even” scoop measures.
▶. The number of sachets added to water. Glucose polymer powder remaining in the pre-measured sachets was re-weighed and recorded by the observer.
▶. The water measurement volume. Observers noted if water was added as recommended or any variation. Observers also checked that measuring jugs were flat on the work surface when water volume was measured; the carer was eye-level with the jug.
Part 2
During week 8, all carers were visited at home by the same observer as part 1. Carers were requested to prepare EF, using the same three techniques (weighing, scoops and sachets) to estimate powder amount, but they were also asked to prepare 200 ml and 1 litre of EFs. They were given the same written instructions for preparing 1 litre of feeds as in part 1 but with additional written instructions for the 200 ml EF (using scoops estimated to measure 4.5 g of glucose polymer). No extra verbal instruction was given. Carers used their own equipment but were given suitable digital scales or appropriate scoops if unavailable. In both study parts, feed preparation order was randomised by computer-generated random number sequences. However, since all three methods are different in character, there was no realistic way that the use of one method could influence the results with a subsequent method. One hundred millilitres of aliquots of each feed prepared were collected for carbohydrate analysis.
The distribution of glucose polymer concentrations prepared in parts 1 and 2 by carers was: carbohydrate 10%, n=4; carbohydrate 15%, n=14; carbohydrate 20%, n=28 and carbohydrate 25%, n=6.
Carbohydrate analysis of glucose polymer solutions
The carbohydrate analysis was performed by measuring sample osmolality and using a calibration curve to determine concentration. The calibration curve was constructed by preparing 50% stock solutions for SOS and Super Soluble Maxijul glucose polymer and then preparing 5–45% solutions from the stock solution. Their osmolalities were measured, and a calibration curve of osmolality against concentration was constructed. The calibration curve was then used to determine the carbohydrate concentrations of the samples from the osmolality. The samples were stored at −20°C until analysed.
Statistical analysis
As the error rate is unknown, sample-size consideration was based on finding clinically worthwhile improvement in administered carbohydrate concentrations if achieved by the sachets. Assuming n=40, there is 90% power for identifying, using a p=0.05 significance level with McNemar's test, that 10% or less of carers when using sachets will make an error of 10% or more in the amount of carbohydrate administration compared with 30% or more of carers using another method. Summary statistics quoted are medians, ranges or percentages. As there were extreme outlier values and general non-normality of distribution, group comparisons (eg, preparation methods) with results expressed as percentages of target values were made using non-parametric tests (Friedman, Wilcoxon Signed rank, Kruskal–Wallis), and for categorical data, McNemar's, χ2 or Fisher's exact test as appropriate. Tests for a time period effect in successive testing of the methods was also made using the above approach. These methods were also used to compare outcome differences between controlled and home conditions.
Results
Parts 1 and 2: accuracy of feed preparation
When 1 litre EFs were prepared, the accuracy of production deteriorated significantly by median amounts between 2% and at most 8% between controlled and home conditions for all three methods (p<0.001 in all cases; table 2). Under controlled conditions, the median percentage of glucose polymer concentration closest to the target amount was (1) pre-measured sachets (105%), (2) weighing of powder (107%) and (3) scoops (117%). Similarly, under home conditions, the closest method to the actual estimated amount of percentage carbohydrate was the pre-measured sachets (111%), followed by weighing (112%) and finally scoops (118%). When only 200 ml of EF was produced at home, the methods gave similar median results with sachets (112%), scoops (112%) and weighing (114%; figure 1), though weighing gave a greater spread of high values which is the reason for a marginally significant difference (p=0.05) between the distributions. There was no significant evidence of a period effect in successively applying the methods in either part 1 or 2.
Percentage carbohydrate concentration estimated for the weighed, scoops and sachets (n=53) methods in EF production under controlled and home conditions. p Value refers to distributional comparison between the three methods. Asterisks denote outliers. EF, emergency feeds.
Percentage carbohydrate concentration measured using weighed, scoops and sachets methods of estimating glucose polymer powder in EF production (n=53)
When preparing 1 litre EFs, for weighing and the pre-measured sachets, the sample percentage within 10% of target weight deteriorated between controlled and home conditions (p<0.01 in all cases). Under controlled conditions using scoops, only 34% (n=18) of samples were within 10% of target weight, and this deteriorated to as low as 15% (n=8) under home conditions (p<0.05). Pre-measured sachets had a significantly higher percentage of samples within 20% of estimated weight for the controlled (p<0.05) and home conditions (p<0.01) when compared with scoops. When 200 ml EF was produced under home conditions, the pre-measured sachets were more likely to be within 20% of target weight than the weighing method (p<0.05), but no noteworthy differences were observed when compared with the scoops (table 3).
Percentage of samples within 10% and 20% of target concentrations using weighed, scoops and sachet methods of estimating glucose polymer powder in EF production (n=53)
In all three methods and under both study conditions, the EF carbohydrate concentration was more concentrated than estimated when prepared by the majority of carers (% mean number of samples over estimated weight: weighed method, 81%; scoops, 91%; sachets, 79%). There were no statistically significant differences in EF accuracy between English- and non-English-speaking carers.
Parts 1 and 2: observed errors in emergency feed production
The most commonly observed feed preparation error was inability to measure water accurately, which was consistently measured wrongly by up to 40% of the carers under all study conditions in parts 1 and 2 of the trial. There were generally more errors in EF production under home rather than controlled conditions. Under home conditions, almost 20% of carers were unable to use the digital scales, approximately one quarter were not using standard procedures to measure the powder with scoops (ie, not levelling off scoop with back of knife), and 15% were not completely emptying the sachets (table 4).
Percentage of carer preparation errors in EF production using the three the methods under controlled and home conditions
The amount of powder left in the individual sachets after preparation of 1 litre of EF under controlled and home conditions was similar (controlled: median 0.24 g (range 0.07–2.15 g); home: median 0.24 g (range 0.04–1.84 g)). When 200 ml EF was made, a median of 0.12 g (range 0.03–3.83 g) remained in the sachets.
Discussion
This is the first prospective, randomised, controlled trial in a group of carers of patients with inherited metabolic disorders comparing three methods (weighing, scoops and pre-measured sachets) of glucose polymer powder measurement for EFs under controlled and home conditions. EF preparation errors were common. Accuracy of EF production significantly deteriorated using all three methods between controlled and home conditions with wide extremes of carbohydrate concentrations prepared (10–224%). EFs were likely to be prepared more concentrated than targeted with as many as 40% of carers measuring water inaccurately for all preparation techniques. Incorrect water measurement was a consistent problem with all three methods, particularly when 1 litre of feed was produced. Some carers failed to measure fluid volume when jugs were rested on a flat surface; instead, they lifted measuring jugs to eye level.
Accuracy of EF preparation is important. Over concentrated solutions of glucose polymer are likely to be hyperosmolar, causing diarrhoea, potentially exacerbating metabolic decompensation, leading to hospital admission and intravenous glucose therapy. In clinical practice, the frequency of over concentrated EFs is unknown. A case study showed that a boy, aged 3 years with GA1 and given glucose polymer, developed severe hypernatreamic dehydration.9 In our own clinic, one child with GSD type 1b, whose carer did not speak English as her first language, was admitted to hospital with ongoing diarrhoea after prolonged use of an over-concentrated EF. In our study, some non-English-speaking carers were innumerate, and they struggled identifying any numbers beyond 10 on digital scales. Although the pre-measured sachets of EF were the most accurate method, there were still errors in adding water and emptying sachets thoroughly.
In reality, little attention is given to instructing carers about EF preparation, and their understanding is rarely checked. EF guidelines are now directly available from the British Inherited Metabolic Disease Group website for patients with MCADD. They advocate using scoops or weighing to measure glucose polymer for 1 litre and 200 ml EFs, but this may be without any verbal health professional instruction.10 In our study, the amount of verbal instruction given to carers was equivalent to the teaching given in a clinic. Although there were inaccuracies in feed production, carers were still better at preparing the feeds immediately following the verbal instruction than 8 weeks later when they had written instruction only.
There are some limitations with this study. It is possible that the presence of a dietetic observer may have reduced EF preparation error, even in the home, because carers paid more attention to feed preparation. However, it was considered important to study carer technique in detail and observe feed preparation to gain some objective estimate of common issues. It is clear from this study that technique deteriorated between controlled and home conditions, but it is still unknown if accuracy would have deteriorated further if carers had been studied over a longer period.
This study demonstrates that ER written guidelines are not always followed according to instruction, and because EF is prepared irregularly, this increases the possibility of further error. Pre-measured sachets were most likely to lead to an EF within 20% of target carbohydrate amount in almost 90% of study conditions, and although their preparation was not without mistakes, there was still a significant improvement compared with other methods. With wider neonatal screening programs and increasing numbers of patients diagnosed with inherited metabolic disorders, pre-measured ER sachets should decrease risk of over concentrated glucose polymer solutions.
Footnotes
-
Funding This study was funded by Vitaflo International.
-
Competing interests None.
-
Patient consent Obtained
-
Ethics approval The study was approved by the Medical Ethics Committees of South Birmingham Health Authority.
-
Provenance and peer review Not commissioned; externally peer reviewed.