Food and fluid texture consumption in a population-based cohort of preschool children with cerebral palsy: relationship to dietary intake

Cerebral palsy (CP) is a group of non-progressive motor disabilities1 that can affect the range, strength, and coordination of motor control, including that required for eating, drinking, and controlling saliva. Oropharyngeal dysphagia (OPD) is common in approximately 85% of preschool children with CP,2 and is used here to describe difficulty with oral movements necessary for efficient preparation and transport of the bolus, or pharyngeal movements which are important to swallow safely. OPD may limit the range of food/fluid textures children can safely and efficiently consume, which can lead to reduced dietary intake affecting nutritional status. It has been widely documented that children with CP are shorter and lighter compared to their peers.3

Food and fluid textures are classified using standardized terminology according to similarities in their physical properties (e.g. firmness and flow rate).4 Modifications to food/fluid textures may be recommended to address mealtime safety or efficiency, or to encourage development of oral sensorimotor skills.5 There is a general consensus that risk of aspiration (food or fluid entering the trachea below the vocal folds) is increased when consuming thin fluids.6 Chewable foods may present a choking hazard if poorly masticated, and they are less efficiently consumed compared to purées in children with CP and those with typical development.7 Foods typically introduced to children at a younger age, such as purées, are processed on the midline with a suckle feeding pattern, while more complex textures, such as chewable foods, require tongue lateralization and separation of movement (of lip, tongue, and jaw function).8

There is a paucity of literature investigating the food and fluid textures in diets of preschool children with CP across the spectrum of motor severity. Studies have been skewed to school-aged children9-11 and those with moderate to severe CP.10-12 To our knowledge, no study has looked at habitual food/fluid texture consumption in children with CP. The Gross Motor Function Classification System (GMFCS) level and age of children varied between studies, as well as food texture terminology. This has resulted in variability in prevalence estimates, with the literature documenting modifications to solids in 25% to 64% of children with CP,9-12 and fluid modification in 16%.13 Earlier studies have had limited exploration of three major factors related to feeding ability on food/fluid textures, including gross motor function, OPD severity, and children's individual energy requirements. Children with CP with poorer gross motor function were reported to have poorer ability consuming textures,14 and poorer gross motor function has also been associated with OPD.2, 9, 11, 12, 15 Three studies have found an increased likelihood of texture modification in children with CP who have OPD.9-11 Children with CP have also been found to have lower dietary energy intake than children with typical development,16 which may be partially attributed to OPD11 but may also reflect their lower energy requirements.17

This study aimed to explore the food/fluid textures consumed by preschool children with CP and the relationship with OPD severity and gross motor function. Understanding these factors will support better management decisions for feeding and nutrition in young children with CP, particularly when considering the safety and efficiency of textures consumed.

Method

This is a cross-sectional population-based study of young children with CP, conducted in Queensland, Australia from April 2009 to August 2013. It forms part of two longitudinal studies, Queensland CP Child: Motor Function and Brain Development (NHMRC 465128),18 and Queensland CP Child: Growth, Nutrition and Physical Activity (NHMRC 569605).19, 20 The data reported from the Pediatric Evaluation of Disability Inventory (PEDI) represents a subset of the data reported by Weir et al.,14 but includes only initial assessments and Queensland-born participants. It serves to determine the convergent validity of the PEDI with direct OPD assessment.

Results

There were 178 eligible children referred, of which 132 children consented to participate in the Growth, Nutrition and Physical Activity study, with 99 completing three full days of the weighed food record. Of the children who declined participation, 18 participated in only the concurrent motor study (finding the burden of two studies too great), and 28 declined both studies (eight because of study burden, 13 because of family circumstances, two were non-English speaking, four resided interstate, and one passed away). Participants’ age ranged from 17 to 37 months corrected age 2 years 3 months. Participant characteristics, and a comparison to non-participants and the Australian CP population,30 are presented in Table 1.

The proportion of children whose parents identified modified diets or limited ability on textures is shown in Table 2. There were 39% of children with modified diets (on the CPFQ), and this proportion increased as gross motor function declined: levels I to II=19%, III=50% (adjusted OR=3.5, p=0.06), IV to V=78% (adjusted OR=31.9, p<0.001). Beyond the exclusion of tough chewables, only 23% had a modified diet. Based on parent report on the PEDI, 41% had limited ability in eating a full range of table foods. The number of children with limited ability increased as gross motor function declined (I–II=26%, III=36% [adjusted OR=1.3, p=0.69], IV–V=78% [adjusted OR=16.6, p<0.001]). There was evidence for an effect of age on the modification of the child's diet (OR=0.9, p=0.006) but no evidence for an effect on inability to consume a full range of table foods (OR 0.9, p=0.08). Four children (4%) were restricted to thickened fluids, all of whom were classified in GMFCS level V. Concordance between the CPFQ and PEDI was high for purées (98.0%), semi-solids (95.0%), chewables (88.9%), and tough chewables/table foods (86.7%).

The average daily habitual intake (taken orally or via tube, and not including quantities lost due to spillage) of quantity of food/fluid (g), energy from food (kJ), and water content (g) is presented in Table 3. This table also presents the proportion of children from combined GMFCS levels meeting estimated average requirements. Children's energy intake decreased with increasing GMFCS level, although the amount of food/fluid consumed and water intake was not significantly related to GMFCS level. Energy intake had a strong positive association with the proportion of chewables in children's diets (kJ: β=11.2, p=0.01; grams: β=16.7, p=0.03) and the proportion of fluids consumed (g: β=14.7, p<0.01). Water intake was strongly associated with consuming a lower proportion of chewable foods (kJ: β=−3.6, p<0.01; grams: β=−7.5, p<0.01).

The proportion of total average habitual energy intake from food/fluid textures by GMFCS level is shown in Figure 1 (see also Fig. S1, online supporting information, for proportion by weight in g). Children classified in GMFCS level III had a significantly lower proportion of overall energy from chewables compared to children in GMFCS levels I to II. Children in GMFCS levels IV to V and those who were tube fed had a significantly lower proportion of chewables and greater proportion of fluids in their diet compared to children in GMFCS levels I to II. There was evidence for an effect of age on the proportion of purées and chewables in children's diets, by GMFCS level. Children's score on the DDS significantly influenced the proportion of textures in children's diets for chewables and fluids, as shown in Figure S1. There was evidence for an effect of age on the percentage of purées and fluids habitually consumed when considering the influence of the DDS.

Children's OPD severity and safety on each texture, and the frequency of texture consumption, are presented in Table SI (online supporting information). Parents were found to include or report ability on food/fluid textures in their child's diet for which their child had OPD in 41% to 64% of children (see Table SI). However, the agreement between parent-reported inclusion/ability and clinician's Swallowing Safety Recommendation (continue or supervision only) was much higher (72–79% children, Table SI). Only three children who had food/fluid orally were assessed to be unsafe on all textures (based on the single clinical feeding assessment), and therefore recommended to have all textures excluded. The proportion of children who were recommended for exclusion of one or more textures was 7% of children on purées, 15% of those on chewables, and 9% for those drinking thin fluids. The recommendation for referral to instrumental assessment was made in 13% for purées, 3% for chewables, and 13% for thin fluids.

Discussion

Modified food/fluid textures are a common feature of the diets of children with CP aged 18 to 36 months, reported by just under half of the parents. Modifications to diets also corresponded to parents reporting limitations in their child's ability on food textures, indicating parents are generally excluding foods/fluids for which they perceive their child has difficulty. For about half of the children, the modification was only exclusion of tough chewable foods. This may reflect milder OPD or typical family preferences for children of this age. Only children with moderate to severe CP were restricted to purées/semi-solids, representing a small percentage of the group overall (14%), although this constituted one-third of children classified in GMFCS levels III to V. This was consistent with two previous studies, finding about one-third of children (32–35%) were restricted to purées/semi-solids, with these studies only recruiting from GMFCS III to V.10, 31 In the present study, 10 children did not consume thin fluids (four drank thickened fluids, six via tube), which was less than the 16% consuming thickened fluids reported by Wilson et al.13 Their study did not include data on children's motor severity, so differences in the motor severity between samples cannot be accounted for. The present study and our team's previous work on the same larger study14 found that poorer gross motor function is associated with increased difficulty with food/fluid textures, and consequently increased modification to food/fluid textures. This specific relationship has not been reported on previously, but is consistent with the association reported between increased OPD severity and poorer gross motor function.2, 9, 11, 12, 15

The amount of food/fluid consumed by children with CP was on average consistent across the sample; however, the energy intake and texture constitution between GMFCS levels differed markedly. Regardless of GMFCS level, children tended to consume about a kg of food/fluid daily, but energy intake ranged from an average of 3084kJ for children fed by tubes to 4310kJ for children classified in GMFCS levels I to II. Children in GMFCS levels III to V and those with more severe OPD were found to consume a lower proportion of chewable foods and more fluids. About half of all children's diets were made up of fluids, increasing to almost two-thirds for children in GMFCS levels IV to V. Purées and fluids are likely to be more efficiently eaten by children with lower gross motor function, so despite these children receiving less energy from equivalent amounts of foods, these textures may still be a more efficient means of achieving adequate energy intake. Children consuming diets with less chewable foods may require an increased amount of food/fluid or have parents modify the energy density of easier to manage textures in order to achieve adequate energy intake.

Parents have been reported to under-detect OPD in their child with CP compared to direct assessment.¹⁵ Our study found there was generally poor agreement (40–60%) between the parent-reported and direct OPD assessment (DDS); however, there was better agreement when considering the child's swallowing safety (70–80%). This indicates that parents may not consider specific oral phase impairments (e.g. limitations to biting or drinking from a cup) as ‘limited ability’ on that texture, but that they are accurately identifying many food/fluid textures for which their child may need referral or have excluded from their diet.

The relationship between the severity of OPD and swallowing safety on food/fluid textures, and the frequency of that mealtime risk (i.e. proportion of the texture in the child's diet) has not been discussed in the literature. Our results found that children with better gross motor function tended to have ‘no to mild’ OPD, and this translated into children generally being recommended as safe to continue with the texture, or possibly requiring supervision (mostly with chewable foods). The only exception was 10% of children from this group were recommended for referral to instrumental assessment for thin fluids, which reflects the greater complexity in managing this texture, particularly for young children. Children with moderate limitations in their gross motor function had only mild OPD with purées and fluids; however, they had moderate OPD on chewable foods. This group was mostly managing purées safely, but about half were recommended to receive supervision on thin fluids and chewables, and a small group were recommended for instrumental assessment. About one-quarter of their diets consisted of chewable foods and half of thin fluids, thus representing a group who may be a priority for feeding intervention. Finally, those with the poorest gross motor function had more severe OPD on all textures, with many recommended for either referral or exclusion (chewable foods most commonly).