Journal of Human Nutrition and Dietetics

CLINICAL NUTRITION
Feeding difculties in children with inherited metabolic disorders: a pilot study
S. Evans,* N. Alroqaiba,* A. Daly,* C. Neville,* P. Davies & A. MacDonald*
*Dietetic Department, Birmingham Childrens Hospital, Birmingham, UK Institute of Child Health, University of Birmingham, Birmingham, UK

Keywords childhood eating behaviour, eating patterns, feeding difculties, metabolic disorders, video recording. Correspondence S. Evans, Dietetic Department, Birmingham Childrens Hospital, Steelhouse Lane, Birmingham B4 6NH, UK. Tel.: +44 (0) 121 333 8020 Fax: +44 (0) 121 333 8021 E-mail: [email protected] How to cite this article Evans S., Alroqaiba N., Daly A., Neville C., Davies P. & MacDonald A. (2012) Feeding difculties in children with inherited metabolic disorders (IMD): a pilot study. J Hum Nutr Diet. 25, 209216 doi:10.1111/j.1365-277X.2012.01229.x

Abstract Background: In children with inherited metabolic disorders (IMD), feeding difculties are often assumed to be inherent, although there is little evidence describing their frequency or severity. The present study aimed to describe feeding patterns/difculties among children with IMD on protein-restricted diets from one centre. Methods: Data from an observational, pilot study of 20 IMD children, nine females (median age, 2.7 years; range, 16 years) were compared with data obtained from a retrospective historical group of 15 healthy children (HC), 12 females, aged 15 years (median 3.0 years). Caregivers completed a feeding assessment questionnaire, and three separate video recordings were taken of each child eating at home. Results: The main feeding problems identied by the caregivers questionnaire in the IMD group (compared to HC) were: poor appetite (55% versus 7%; P = 0.004), limited food variety (55% versus 27%; P = 0.04) and lengthy mealtimes (70% versus 20%; P = 0.006). During mealtimes, children from the IMD group were more likely to vomit, exhibit negative behaviour, get distracted and self-feed less often. From video recordings of meals, although the median meal duration was similar for the two groups (18 min IMD versus 16 min HC), the HC ate twice the quantity of food (3.4 mouthfuls min)1 versus 1.5 mouthfuls min)1; P < 0.001). During mealtimes, IMD caregivers were less likely to talk to their children (median parent to child communications: IMD group, seven in 10 min; HC, 17 in 10 min). Eighty-three percent of IMD children regularly ate alone. Conclusions: In children with IMD on protein restrictions, severe feeding difculties were common. Caregivers need to focus more attention on the social aspects of feeding. Further larger scale studies are required.

Introduction Many children with inherited metabolic disorders (IMD) of protein metabolism are dependent on complex dietary treatments. Although the priority is to attain metabolic stability, with optimal growth and nutritional status, there are many barriers that impede the attainment of such goals. Feeding difculties, in particular, can restrict progress, and are anecdotally described in IMD children. They
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

have only been systematically reported in phenylketonuria (PKU) (MacDonald et al., 1997) and one type of long chain fatty acid oxidation disorder (Lund & Leonard, 2001) but not in other disorders. The aetiology of feeding problems is potentially extensive. Neurological impairment is common in IMD, particularly in urea cycle disorders (UCD) and organic acidaemias (OA). There may be difculties with chewing and swallowing, gastro-oesophageal reux, vomiting, slow
209

Feeding difficulties in children with inherited metabolic disorders

S. Evans et al.

feeding and anorexia. Poor appetite may be associated with elevated levels of toxic metabolites (i.e. glutamine and ammonia in UCD) (Feillet & Leonard, 1998). In addition, gagging, vomiting and gastro-oesophageal reux, as a result of varying aetiology, are common in methylmalonic acidaemia (MMA) and propionic acidaemia (PA). A dislike of protein containing foods is often seen in UCD and OA (Feillet & Leonard, 1998; Dixon, 2007). Similarly, there may be a preference for certain foods such as salty avours in MMA, possibly associated with increased urinary losses of sodium. Medications, given to utilise alternative pathways for nitrogen excretion (e.g. sodium benzoate and sodium phenylbutyrate) may contribute to anorexia, nausea and vomiting (Feillet & Leonard, 1998) or gastritis. Problems with administering strong tasting protein substitutes or essential amino acids may create additional anxiety for some families. Tube feeding, particularly overnight feeding, is commonly used in PA to prevent the production of odd-chain fatty acids associated with fasting (Touati et al., 2006) or in other conditions to supplement a poor oral intake. Tube feeding may further impede appetite or cause gastro-intestinal intolerance which, in turn, impacts on appetite. However, the extent and severity of feeding problems has not been reported in inherited disorders of protein metabolism requiring diet therapy. Therefore, the present pilot study aimed to describe the type and incidence of feeding patterns and feeding difculties among young children requiring special diet therapy for OA, UCD and amino acid disorders (except PKU) and to compare these with historical data obtained from healthy children (HC) of a similar age. Materials and methods Subjects Inclusion criteria comprised children: (i) aged 16 years; (ii) diagnosed with a UCD, OA or amino acid disorder (not PKU) requiring diet therapy; (iii) having only one clinical diagnosis; and (iv) monitored by one IMD treatment centre. Exclusion criteria comprised children: (i) <12 months and >6 years of age; (ii) more than one diagnosis; and (iii) diagnosed with classical PKU; and (iv) not on diet therapy. Thirty children met the inclusion criteria. Caregivers were sent a letter of invitation with an information sheet about the study and a reply slip to return if they were interested in the study. Informed written consent was obtained from the childs main caregiver(s). Although the children were too young to give assent, a series of pictures was used to explain the study. A favourable ethical opinion was obtained from the South Birmingham Research Ethics Committee.
210

Historical data from healthy children Historical data from a previous study (MacDonald et al., 1997) on a group of 15 HC aged 15 years (median 3.0 years) (12 female, three male) with no medical problems or dietary restrictions were used for comparison purposes. The caregivers of these children had completed the same feeding assessment questionnaire and the children were lmed once during a routine mealtime and videos were analysed using the same criteria. Because this comprised historical data from HC, they were not age- or gender-matched to the present study population and the video analysis was performed by a different observer. Data were validated and cross-checked by a research psychologist. Similarly, the video analysis for this study was cross-checked independently by two observers. There was a high degree of inter-observer correlation for individual variables, with a minimum correlation coefcient of r = 0.82 and a maximum of r = 0.97 (MacDonald et al., 1997). The HC could not be considered as a control group, although they were of similar age to the study group and their data demonstrate normal dietary patterns and so serve as a useful comparison. Given that there is currently no data on the comparison of HC and children with IMD in terms of feeding difculties, these data provide a rare opportunity to highlight the frequency and severity of problems in IMD. Study design In this observational study, data were collected over a 3week period for each child with IMD. Demographic data were obtained from patients medical and dietetic notes, including age at diagnosis, medications, diet plan, symptoms, feeding history, and speech and language therapist (SALT) referrals. Study procedures/investigations Feeding assessment questionnaire A validated feeding assessment questionnaire (consisting of multiple choice and open questions) was completed by the main caregiver with the assistance of a research dietitian. The questionnaire, adapted from Harris & Booth (1992), had previously been used by MacDonald et al. (1997) to assess feeding behaviour in children with PKU compared to an age- and gender-matched control group of HC. The questionnaire examined maternal perception of feeding problems, including appetite, feeding environment, gastro-intestinal symptoms, feeding behaviour, feeding patterns, feeding skills, parental management and food preferences. An interpreter was used for non-English speaking caregivers (n = 3).
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

S. Evans et al.

Feeding difficulties in children with inherited metabolic disorders

Video-recorded meals Each child with an IMD was lmed eating their midday or evening meal at home by the same research dietitian. Both parent and child were aware of the presence of the camera. Three meals (at the same time of day) were lmed for each child within a 3-week time period to maximise consistency, reliability and repeatability, as well as to minimise the effect of the researchers presence. The recorded meals were then analysed using a scored category system based on the observed behaviour of both mother and child during the mealtime (reported in MacDonald et al., 1997). This included an analysis of the number of times the child was distracted from the meal; the manner in which the childs attention was redirected back to the meal; the number of mouthfuls eaten during the meal; child and parental vocalisations; child behaviour; and parental intervention during the mealtime. Meal duration to the nearest minute was documented. A mean score was calculated from the three videos for each child. The video analyses were carried out by one observer. Five of the videos were analysed independently by a second investigator. There was a good degree of inter-observer correlation for individual variables with a minimum correlation coefcient of r = 0.63 and a maximum of r = 0.96. Statistical analysis Descriptive statistics including percentages, medians and ranges for all variables were recorded. Data were analysed using unpaired t-tests, MannWhitney U-tests and Fishers exact test as appropriate for comparisons between subgroups of the IMD group and comparisons with HC. Results Subjects Twenty children with UCD, OA and amino acid disorders (except PKU) (11 males; median age, 2.7 years; range: 0.96 4 years) from one UK IMD centre were recruited. Ten potential participants were not recruited as a result of the travel distance from the study centre (n = 7), the caregiver declining to perform study (n = 2) or a return to normal diet (n = 1). Children had a range of disorders: UCD (n = 4), glutaric aciduria type 1 (n = 4), propionic acidaemia (n = 3), isovaleric acidaemia (n = 3), tyrosinaemia (n = 3), methylmalonic acidaemia (n = 2) and other (n = 1). Fifty-ve percent of children (n = 11) were diagnosed within the rst month of life; and 20% (n = 4) were over the age of 1 year. Children were from a variety of different ethnic groups, including; Pakistani (n = 9), white British
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

(n = 6), black African (n = 3), Indian (n = 1) and Bangladeshi (n = 1). Most children were an only child (n = 7) or had one sibling (n = 8). Four children had a single caregiver only. Dietary treatment All subjects had dietary protein restrictions of not more than 1 g kg)1 per day of intact protein. Fifty-ve percent (n = 11) were taking a protein substitute of l-amino acids that incorporated carbohydrates, vitamins and minerals; 30% (n = 6) orally and 25% (n = 5) via a tube. In total, 50% (n = 10) received tube feeds in addition to oral diet, although this included four children who only had tastes of food. Medications Thirty-ve per cent (n = 7) of children were taking sodium benzoate and/or phenylbutyrate; 60% (n = 12) lcarnitine (for OA); 15% (n = 3) l-arginine (for UCD) and 25% (n = 5) were prescribed anti-reux medications for vomiting. Sixty percent (n = 12) were on two or more medications. Speech and language therapy Thirty per cent had seen a SALT. Results from feeding assessment questionnaire Feeding problems as perceived by primary caregiver Children with IMD were perceived by their caregivers to have many feeding difculties. The main feeding problems identied in the IMD group compared to the retrospective HC were poor appetite, limited variety of foods, lengthy mealtimes and vomiting (Table 1). Seventy percent of the IMD children had more than one feeding difculty. Conversely, 67% (n = 10) of the HC had no feeding problems at all. Similar results were demonstrated when IMD children on oral diet only (no tube feeding; n = 10) were compared with the HC, although the results were less signicant (Table 1). Specically, children with UCD, MMA and PA and tube fed children were more likely to have feeding problems, including poor appetite, limited variety and vomiting than other diagnoses or than those on oral diet only. Younger children (<3 years) were more likely to have feeding complications than older children, although this was not statistically signicant (Table 2). Gastro-intestinal symptoms These were more frequent in the IMD group compared to HC. Vomiting was reported by 30% (n = 6) of caregivers of IMD children compared to no vomiting in the HC (P = 0.03). Diarrhoea, constipation and abdominal pain
211

Feeding difficulties in children with inherited metabolic disorders

S. Evans et al. Table 1 Caregiver perception of feeding problems, mealtime behaviour, feeding skills and food texture

Feeding problem Poor appetite Limited variety Slow to feed Cannot chew No solid food No feeding problems Mealtime behaviour Throws food Spits Chews not swallows Turns head Closes mouth Pushes spoon away Cries Dribbles from mouth Feeding skills Fed by caregiver Has self fed Uses a spoon Uses a fork Drinks from bottle Food texture Mashed food Pureed food Food with lumps

IMD group (%) (n = 20) 55 (11) 65 (13) 70 (14) 20 (4) 20 (4) 5 (1) 45 (9) 50 (10) 55 (11) 55 (11) 55 (11) 50 (10) 35 (7) 35 (7) 80 (16) 90 (18) 65 (13) 40 (8) 50 (10) 80 (16) 30 (6) 65 (13)

HC (%) (n = 15) 7 (1) 27 (4) 20 (3) 0 (0) 0 (0) 67 (10) 53 27 40 7 33 20 13 0 (8) (4) (6) (1) (5) (3) (2) (0)

P* 0.004 0.04 0.006 0.12 0.12 <0.001 0.74 0.30 0.50 0.004 0.31 0.09 0.12 0.01 <0.001 0.12 0.01 0.002 0.09 <0.001 0.03 <0.001

IMD oral diet only (%) (n = 10) 20 50 40 10 10 10 30 30 70 40 40 40 0 20 80 90 70 40 60 (2) (5) (4) (1) (1) (1) (3) (3) (7) (4) (4) (4) (0) (2) (8) (9) (7) (4) (6)

P 0.54 0.40 0.38 0.40 0.40 0.01 0.41 0.99 0.23 0.12 0.99 0.38 0.50 0.15 0.002 0.40 0.18 0.01 0.48 <0.001 0.15 0.05

13 (2) 100 (15) 100 (15) 93 (14) 20 (3) 7 (1) 0 (0) 100 (15)

80 (8) 20 (2) 70 (7)

*Fishers exact test [inherited metabolic disorders (IMD) versus healthy children (HC)]. Fishers exact test (IMD oral versus HC). Table 2 Eating problems as perceived by caregiver by diagnosis, tube fed versus oral diet and age Eating problem Poor appetite Limited variety Slow to eat Vomiting Gastro-intestinal problems Total number eating problems UCD/MMA/PA (%) (n = 9) 89 78 89 56 33 (8) (7) (8) (5) (3) Other (%) (n = 11) 27 55 55 9 18 (3) (6) (6) (1) (2) P* 0.01 0.37 0.15 0.05 0.45 0.03 Tube fed (%) (n = 10) 90 80 100 40 40 (9) (8) (10) (4) (4) Oral diet (%) (n = 10) 20 50 40 20 10 (2) (5) (4) (2) (1) P* 0.005 0.35 0.01 0.63 0.16 0.002 <3 years (%) (n = 11) 64 73 82 36 27 (7) (8) (9) (4) (3) >3 years (%) (n = 9) 44 56 56 22 11 (4) (5) (5) (2) (1) P* 0.65 0.64 0.33 0.64 0.80 0.23

Median 3

Median 2

Median 3.5

Median 1.5

Median 3

Median 2

*Fishers exact test. MMA, methylmalonic acidaemia; PA, propionic acidaemia; UCD, urea cycle disorder.

were also more common in the IMD group but not signicantly so. Mealtime behaviour IMD children demonstrated more negative behaviours than the HC (Table 1). These included turning their head away or closing their mouth when offered food, knocking the spoon away, crying at the beginning of mealtimes, and dribbling from the mouth. These results were similar but less signicant when IMD children on oral diet alone
212

were compared with the HC, although chewing but not swallowing food was seen to be more common in the oral IMD group (Table 1). Caregivers rating of mealtime IMD children were reported to be more difcult to feed than the HC. On a scale of 15 (one being not difcult), IMD caregivers reported a median score of 3 (range 15), whereas the HC had a median score of 1 (range 13) (P = 0.007).
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

S. Evans et al.

Feeding difficulties in children with inherited metabolic disorders

Self-feeding skills IMD children as a whole group and when separated into those on oral diet alone demonstrated fewer self-feeding skills than the HC (Table 1). Although most IMD children (90%) could nger feed and 65% could use a spoon, 80% were still being spoon fed by caregivers. In comparison, all children in the HC could self-feed and only few were spoon fed by caregivers. Similarly, half of the IMD children still drank from a bottle compared to only three HC. Food texture Children with IMD in general required softer and more processed food textures than the HC. The majority of IMD children regularly used mashed foods, almost onethird required pureed food and one-third could not eat food with lumps. Similar results were seen for IMD children on oral diet alone. In comparison, no HC required pureed food, one had mashed food and all could eat food with lumps (Table 1). There was no difference with HC in terms of liking for sour, sweet or savoury foods. However, IMD children were more likely than HC to like highly avoured foods (60% versus 27%; P = 0.09). Results from video-recorded meals Two of the 20 children had no oral diet so videorecordings were collected on 18 children. Self-feeding Although the results of the caregiver feeding assessment questionnaire stated that 80% (n = 16) of the IMD children could self-feed, only 44% (n = 8) were observed to self-feed. IMD children were offered food by spoon a median of 14 times in 10 min, whereas HC were spoon fed a median of less than once in 10 min (P = 0.02). Family mealtime Eighty-three per cent (n = 15/18) of children regularly ate alone rather than with the rest of the family, even though siblings were often home at the time. Three of these chilTable 3 Observed child-feeding behaviour Behaviour Turns head Closes mouth Distracted and redirected Positive vocalisations Negative vocalisations

dren did on occasion eat with siblings or parents. Only three children regularly ate with siblings. Child-feeding behaviour The IMD group both as a whole group and when separated into those on oral diet only, more commonly turned their head away or closed their mouth when offered food. They also had fewer positive vocalisations relating to food than the HC and were more often distracted and redirected to the task of eating, although the latter was not statistically signicant (Table 3). Mealtime duration Although median meal duration was similar for the two groups (18 min IMD versus 16 min), the HC ate at least twice the quantity of food (3.4 mouthfuls min)1 versus 1.5 mouthfuls min)1) (Fig. 1). Those IMD children on oral diet alone (n = 10) had the same median meal duration as the IMD group as a whole, although the number of mouthfuls per minute was greater (2.1) but still less than the HC group (Fig. 1). Sixty-one per cent (n = 11) of the IMD children needed prompting by their caregiver at the start of the meal compared to 7% (n = 1) of the HC (P = 0.003). Similarly, 40% (n = 4) of those IMD children on oral diet only required prompting, although this was not statistically signicant. Mealtime communication between caregiver and child For every 10 min of time, there was less parental feeding directed vocalisation (e.g. is it nice? or is it hot?) in the IMD group (median 6.8 times; 6.3 in the oral diet only group) compared to HC (median 16.5 times). However, this was not statistically signicant. Parents also used more prompting vocalisations (e.g. eat some more please or are you going to nish that?) in the IMD group (median 5.3 times) compared to HC (median 2.7 times) (P = 0.05). There was little difference between the two groups for the amount of food related praising. Caregivers were less likely to talk to their IMD children during mealtimes but, when they did, it was more often to prompt them to eat more.
IMD group (n = 18) 2.3 0.2 3.1 0 0.4 (07.6) (08.6) (033.3) (02.5) (07.9) IMD oral diet only (n = 10) 0.5 0 2.6 0 0.1 (07.6) (07.0) (0.833.3) (00.8) (04.6)

HC (n = 15) 0 0 1.8 0.8 0.7 (03.3) (00.7) (06) (010) (05)

P* 0.004 0.006 0.10 0.06 0.75

P 0.07 0.04 0.18 0.01 0.48

Data are given as the median/10 min (range). *MannWhitney test [inherited metabolic disorders (IMD) versus healthy children (HC)]. MannWhitney test (IMD oral versus HC).
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

213

Feeding difficulties in children with inherited metabolic disorders

S. Evans et al.

Figure 1 Number of mouthfuls of food eaten per minute P < 0.001; MannWhitney test [health children (HC) versus inherited metabolic disorders (IMD) all; and HC versus IMD oral].

Mealtime strategies Caregivers used different strategies to persuade children to eat or complete their food, including taking food away, offering another food/meal, rewards or distractions. Parents in the IMD group were more likely to use three or more strategies, particularly distraction (television or toys) compared to the retrospective HC (Table 4). Indeed, twice as many IMD children regularly watched television during mealtimes compared to HC (Table 4). Similar but less signicant results were seen with IMD children on oral diet alone, although bribery played a greater role in this group compared to HC (P = 0.03) (Table 4). Fifty percent (n = 9) of caregivers of IMD children were frequently observed to focus attention on messy eating (wiping the childs face or the table). Discussion Children with IMD of protein metabolism such as UCD, OA and other amino acid disorders commonly experienced a number of feeding difculties, including poor appetite, limited variety of foods and inadequate attention to the social aspects of eating. This compares with similar studies in disorders such as PKU and cystic brosis (Mac-

Table 4 Observed strategies used to make children eat IMD group (%) (n = 18) 94 (17) 78 (14) 28 (5) 11 44 33 61 67 44 (2) (8) (6) (11) (12) (8) HC (%) (n = 15) 60 (9) 80 (12) 13 (2) 0 20 0 40 33 7 (0) (3) (0) (6) (5) (1) IMD oral diet only (n = 10) 90 (9) 80 (8) 30 (3) 10 70 40 50 50 40 (1) (7) (4) (5) (5) (4)

Strategy Take food away Make eat Offer next course Rewards Bribery Threats Distraction Television 3 strategies

P* 0.006 0.99 0.40 0.49 0.18 0.02 0.004 0.08 0.009

P 0.18 0.99 0.36 0.40 0.03 0.02 0.70 0.44 0.12

*Fishers exact test [inherited metabolic disorders (IMD) versus healthy children (HC)]. Fishers exact test (IMD oral versus HC).

Donald et al., 1997; Blissett et al., 2001; Powers et al., 2002). Problems were more pronounced in those with UCD, MMA or PA. Even allowing for the more complex feeding issues associated with tube feeding, when the IMD children on oral diet alone were compared with the HC group, the results were similar (although less signicant, as a result in part to smaller subject numbers) to that for the group as a whole. Many of the feeding difculties observed in the present study appeared to be associated with management of feeding behaviour rather than something intrinsic associated within the clinical condition. Mealtimes were very functional and frequently given separately from the rest of the family, so there was less opportunity for young children to observe and model others eating. Children were easily distracted from the task of eating as a result of a lack of focus associated with eating alone, a lack of interaction from caregivers and the added distractions of television and toys. Observation of caregiver management identied much negative feeding interaction that was designed to cajole children into eating. Caregivers of children with IMD require additional education to understand the principles of normal feeding and the importance of encouraging independent feeding skills within the restrictions of the low protein diet. Helping parents with the practical management of mealtimes should be an integral part of care. This may require the collective assistance of specialist dietitians, SALT and psychologists experienced in IMD and feeding difculties. From observations, children with IMD did not spend longer eating than HC, although they only consumed half the quantity of food. Even those with IMD on oral diets only still ate signicantly lower quantities of food than HC. Previous studies in young children have demonstrated that longer meals were associated with a lower food intake (Parkinson & Drewett, 2001). Many of the children on tube feeds had daytime bolus and overnight tube feeds and, generally, there was less pressure to encourage oral food and hence less incentive to ensure that all food was eaten. Subsequently, mealtimes were shorter than expected. Unlike in fatty acid oxidation disorders (Lund & Leonard, 2001), tube feeding did independently appear to negatively impact on appetite, although there was also a cumulative effect of age and
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

214

S. Evans et al.

Feeding difficulties in children with inherited metabolic disorders

severity of diagnosis. Although tube feeding may be an inevitable part of management for many of these children, where possible, it should be carried out in conjunction with normal feeding and mealtime management strategies. Control of symptoms such as reux, vomiting and anorexia is likely to reduce feeding difculties. Just over one-third of the study children were on nitrogen-scavenging drugs, which are prescribed to utilise alternative pathways for nitrogen excretion. Over half of these children had frequent vomiting and all but one had poor appetite. There is evidence that sodium benzoate and phenylbutyrate have been associated with nausea, vomiting, anorexia, gastritis, oral mucositis and taste disturbances (Feillet & Leonard, 1998; Batshaw et al., 2001; Niemi & Enns, 2006). Therefore, care is needed in prescribing these medications. Spreading their administration over several doses helps to minimise toxicity and they should be administered with protein-containing meals (Batshaw et al., 2001). Anti-emetic medication is also helpful. It must be acknowledged that this was a single centre study and the sample size was small as a result of the rarity of these conditions and the time taken to collect and analyse the dataset on this detailed review of feeding. Therefore, the results do not necessarily reect what would be seen in another population of similar children. However, the data collected are very comprehensive and reect those reported by parents in clinics and observed by health professionals about the feeding issues in this group of children. In addition, any comparison data with the HC were not age- and gender-matched (HC predominantly girls), although they were of similar age and lived within the same geographical area. In addition, the videos for the HC were analysed by a different observer, and all the data were historical. This may have impacted on the comparison data. However, comparison analysis of data for just the girls did not show any major disparity with that of the entire group. This is the only study to describe the frequency and severity of feeding problems in IMD children and the ndings strongly suggest that these problems are worse than for HC. It is acknowledged that further multicentre studies with concurrent groups of age-/gender-matched controls are required. In addition, multicentre studies investigating feeding behaviour of tube fed and orally fed IMD children from different diagnostic groups on diet therapy are required. They should include an analysis of the inuence of drug type and administration on feeding behaviour. Also, longitudinal studies analysing the development of feeding behaviour in accordance with the type and severity of IMD disorder, diet therapy and drug treatments would be invaluable. Video-recording a mealtime enables examination of all aspects of patient and carer communication and interaction, including how and where the child is fed, feeding
2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

skills, suitability of feeding equipment and the parenting skills of the caregiver. Previous studies have demonstrated that videotaping provides a useful technique for characterising behaviour associated with disordered eating (Tappe et al., 1998; Blissett et al., 2001; Powers et al., 2002). However, the presence of an unfamiliar person with a camera at the mealtime is likely to have had an effect on the mealtime procedure for both parent and child. It is possible that families may have eaten together more often but did not want to be lmed, and so chose to eat their meals at an alternative time. In addition, it is likely that dialogue between caregiver and child was more restrained as a result of the presence of the camera. However, for comparison purposes, HC experienced the same limitations, although the data were interpreted by different observers.Children with IMD of protein metabolism (except PKU) generally have less frequent mealtimes, with a limited variety and smaller quantity of food, as well as insufcient mealtime encouragement, and the social aspect of mealtimes is largely neglected. Normal weaning progression is delayed in terms of the food texture, exploration and independent feeding skills. Caregivers would benet from access to a specialist feeding clinic with professionals trained in IMD (including dietitians, SALT and psychologists) to address these issues, including group sessions on the normal weaning process, self-feeding, positive reinforcement and appropriate mealtime strategies, with an emphasis on the social aspects of eating.

Acknowledgments The authors would like to thank all the children and their families for participating in the study and, in particular, for enabling us to enter their homes to lm the children eating. Conict of interest, source of funding and authorship The authors declare that they have no conict of interests. The present study was performed as part of routine clinical care. The principal authors research time was funded by Nutricia Clinical Care. SE performed data collection, data analysis and preparation of the manuscript. NA was involved in the study design, data collection, data analysis and preparation of the manuscript. AD performed data collection and preparation of the manuscript. CN performed data collection and preparation of the manuscript. PD performed data analysis and preparation of the manuscript. AMcD performed study design, data analysis and preparation of the manuscript. All authors critically reviewed the manuscript and approved the nal version submitted for publication.
215

Feeding difficulties in children with inherited metabolic disorders

S. Evans et al.

References
Batshaw, M., MacArthur, R. & Tuchman, M. (2001) Alternative pathway therapy for urea cycle disorders: twenty years later. J. Pediatr. 138, S46S55. Blissett, J., Harris, G. & Kirk, J. (2001) Feeding problems in Silver-Russell syndrome. Dev. Med. Child Neurol. 43, 3944. Dixon, M. (2007) Organic acidaemias and urea cycle disorders. In Clinical Paediatric Dietetics, 3rd edn. eds V. Shaw & M. Lawson, pp. 357374. Oxford: Blackwell Publishing. Feillet, F. & Leonard, J. (1998) Alternative pathway therapy for urea cycle disorders. J. Inher. Metab. Dis. 21(Suppl. 1), 101 111. Harris, G. & Booth, I.W. (1992) The nature and management of eating problems in preschool children. In Feeding Problems and Eating Disorders in Children and Adolescents. eds P. Cooper & A. Stein, Vol. 5, pp. 2861. Monographs in Clinical Paediatrics, Chur, Switzerland: Harwood Academic Press. Lund, A.M. & Leonard, J.V. (2001) Feeding difculties in long-chain 3-hydroxyacyl-CoA dehydrogenase deciency. Arch. Dis. Child. 85, 487488. MacDonald, A., Rylance, G., Asplin, D., Harris, G. & Booth, I.W. (1997) Abnormal feeding behaviours in phenylketonuria. J. Hum. Nut. Diets 10, 163170.

Niemi, A.K. & Enns, G.M. (2006) Pharmacology review: sodium phenylacetate and sodium benzoate in the treatment of neonatal hyperammonaemia. Am. Acad. Paediatrics 7, e486. Parkinson, K.N. & Drewett, R.F. (2001) Feeding behaviour in the weaning period. J. Child Psychol. Psychiatry 42, 971 978. Powers, S.W., Patton, S.R., Byars, K.C., Mitchell, M.J., Jelalian, E., Mulvihill, M.M., Hovell, M.F. & Stark, L.J. (2002) Caloric intake and eating behavior in infants and toddlers with cystic brosis. Pediatrics 109, E755. Tappe, K.A., Gerberg, S., Shide, D., Anderson, A. & Rolls, B. (1998) Videotape assessment of changes in aberrant mealtime behaviors in anorexia nervosa after treatment. Appetite 30, 171184. Touati, G., Valayannopoulos, V., Mention, K., de Londay, P., Jouvet, P., Depondt, E., Assoun, M., Souberbielle, J., Rabier, D., Ogier de Baulny, H. & Saudubray, J. (2006) Methylmalonic and propionic acidurias: management without or with a few supplements of specic amino acid mixture. J. Inherit. Metab. Dis. 29, 288298.

216

2012 The Authors Journal of Human Nutrition and Dietetics 2012 The British Dietetic Association Ltd.

Copyright of Journal of Human Nutrition & Dietetics is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.