Comparison of National Dysphagia Diet and International Dysphasia Diet Standardization Initiative Levels for Thickened Drinks Prepared with a Commercial Xanthan Gum-Based Thickener Used for Patients with Dysphagia

Abstract

The thickness levels of fluids for patients with dysphagia are established according to the guidelines of the National Dysphagia Diet (NDD) and International Dysphagia Diet Standardization Initiative (IDDSI). The nectar- (level 2), honey- (level 3), and pudding-like (level 4) fluids in NDD are consistent with the mildly (level 2), moderately (level 3), and extremely (level 4) thick fluids in IDDSI, respectively. In this study, NDD levels were compared with IDDSI levels by estimating both the apparent viscosity (η_a,50) and the residual volume (mL) in the IDDSI syringe flow test of thickened drinks prepared with a commercial xanthan gum-based thickener at different concentrations (0.1∼3.1%, w/w). The concentration range of the thickener in thickened drinks at each IDDSI and NDD level increased in the following order: water> orange juice> milk. A small difference was noted in the range of thickener concentration in the same NDD and IDDSI levels for thickened milk when compared with other thickened drinks. These results indicate that the thickener concentration ranges of thickened drinks for the classification of NDD levels differed from those of IDDSI levels, and they appeared to be greatly influenced by the type of drink. These findings could provide useful information for practically indi-cating the reliable thickness levels by the IDDSI flow test in clinical practice.

INTRODUCTION

When there are problems with the food’s easy passage from the mouth to the esophagus, dysphagia occurs, which makes it challenging to swallow food or liquids (Kim and Yoo, 2015). One of the main therapies for dysphagia is to thicken fluids to promote adequate oral intake (Seo and Yoo, 2013). The National Dysphagia Diet (NDD) Task Force suggested using a sophisticated rheometer or viscometer to accurately determine viscosity values for each NDD thickness level (National Dysphagia Diet Task Force, 2002). However, it is not provided in most clinical settings because this instrument is often inaccessible and its use impractical because of cost and the need for training (Kim et al., 2014). The International Dysphagia Diet Standardization Initiative (IDDSI) also developed a standard framework for classifying dysphagia diets and provided thickness levels for dysphagia foods or liquids (IDDSI, 2019).

Table 1 lists NDD and IDDSI levels for thickened fluids. The IDDSI board suggested categorizing thickened fluids by levels using the syringe flow test (SFT) or the fork drip test (FDT) as the IDDSI testing method (Kim et al., 2018). In most cases, thickened fluids are described into five different levels (0∼4) set up by an SFT for a pragmatic measurement of flow: levels 0 (thin), 1 (slightly thick), 2 (mildly thick), 3 (moderately thick), and 4 (extremely thick) (Cichero et al., 2017). For the NDD framework, thickened fluids are categorized into four “consistency” levels established by the NDD task force: thin (1∼50 mPa･s), nectar-like (51∼350 mPa･s), honey-like (351∼1,750 mPa･s), and pudding-like (>1,750 mPa･s). These NDD levels were classified by the apparent viscosity (η_a,50) of a thickened fluid at a shear rate of 50 s⁻¹, which corresponds to the shear rate in oral cavities made during swallowing (Lotong et al., 2003; Adeleye and Rachal, 2007; Sopade et al., 2007; Payne et al., 2011). Recently, nectar- (NDD level 2), honey- (NDD level 3), and pudding-like (NDD level 4) fluids are recognized to correspond to mildly (IDDSI level 2), moderately (IDDSI level 3), and extremely (IDDSI level 4) thick fluids, respectively (Fig. 1) (IDDSI, 2019). Although many studies have conducted the SFT for IDDSI level classification or viscosity (η_a,50) measurement for NDD level classification (Kim et al., 2018; Jeong et al., 2021; Kwong and Tse, 2021; Bolivar-Prados et al., 2022; Brooks et al., 2022), no study has been attempted to directly compare NDD and IDDSI levels of various thickened drinks at different thickener concentrations. Therefore, for thickened drinks prepared with a commercial xanthan gum (XG)-based thickener at different concentrations, which is used for patients with dysphagia, the NDD levels were compared and investigated with IDDSI levels by measuring the η_a,50 and the remaining volume (mL) in the syringe by the SFT and by observing the shape of the thickened drinks on the fork by the FDT. The findings of this study could provide valuable information for predicting the reliable viscosity of the SFT and FDT practically.

Table 1.

NDD and IDDSI levels for thickened fluids (NDD Task Force, 2002; IDDSI, 2019)

Test and parameters	Thickness level	Description	Reference values
Apparent viscosity (ηa,50) in NDD	1	Thin-like	1∼50 mPa･s
	2	Nectar-like	51∼350 mPa･s
	3	Honey-like	351∼1,750 mPa･s
	4	Pudding-like	>1,750 mPa･s
IDDSI flow test	0	Thin	<1 mL
	1	Slightly thick	1∼4 mL
	2	Mildly thick	4∼8 mL
	3	Moderately thick	>8 mL
	4	Extremely thick	10 mL1)

¹⁾The fork drip test was performed to determine an “extremely thick level” in IDDSI.

NDD, National Dysphagia Diet; IDDSI, International Dysphagia Diet Standardization Initiative.

Fig. 1.

Comparison of National Dysphagia Diet (NDD) and International Dysphagia Diet Standardization Initiative (IDDSI) levels for drinks. Adapted from the framework of International Dysphagia Diet Standardisation Initiative (2019).

MATERIALS AND METHODS

Materials and sample preparation

Three commercially accessible cold drinks in Korea such as bottled water (Jeju Samdasoo), orange juice (The Coca-Cola Co.), and whole milk (Seoul Dairy Co.) were used as the dispersion medium to prepare thickened drinks. Table 2 shows the main ingredients and nutritional information of the cold drink products. Refrigerated cold drinks were purchased from a local store. A commercial XG-based food thickener powder, including XG, guar gum, and dextrin (Visco-up^Ⓡ, Rheosfood Inc.), was used to thicken all dispersion media with different thickener concentrations (0.1∼3.1%, w/w). Based on the manufacturer’s guidelines, the thickened drinks were prepared by mixing the weighted thickener powder with cold drinks for 60 s at room temperature. The thickened sample was stabilized and chilled for 4 h in a refrigerator before experimental use.

Table 2.

Composition of the drinks used in this study

Type of drink	Main ingredient	Nutrition (per 100 mL)
Water	Water	Carbohydrate (0 g) Protein (0 g) Fat (0 g)
Orange juice	Concentrated orange juice, high-fructose corn syrup, calcium, vitamin, and citric acid	Carbohydrate (9.0 g) Protein (0.6 g) Fat (0 g)
Milk	Raw milk	Carbohydrate (4.5 g) Protein (3.0 g) Fat (4.0 g)

Viscosity measurement

The viscosity of thickened drinks was measured with a rheometer (RheoStress 1, Haake GmbH) through plate geometry. The sample was applied between parallel plates at 8°C. The diameter of the disc and gap were 35 mm and 500 mm, respectively. After a 5 min stabilization time at 8±0.1°C, the flow behaviors were measured over shear rate boundaries of 0.1∼100 s⁻¹. In the flow curves (shear stress versus shear rate) of the samples, the power law model (Eq. 1) was employed to obtain the apparent viscosity (η_a,50) values of the thickened samples.

$$\text{σ=K}\cdot \stackrel{\text{.n}}{\text{γ}}$$ (Eq. 1)

where σ is the shear stress (Pa), K is the consistency index (Pa･sⁿ), $\stackrel{·}{\gamma }$ is the shear rate (s⁻¹), and n is the flow behavior index. η_a,50 was determined at 50 s⁻¹ from the values of the power law equation factors (K and n).

SFT

The experiment was conducted using a 10-mL Luer-Lok Tip syringe (Becton Dickinson Medical Pte., Ltd.) in the same way as in the IDDSI testing methods. For the measurement, the syringe was filled with 10 mL of the sample, and the nozzle was sealed with a finger. For the sample to flow out of the nozzle, the finger was removed for 10 s. After 10 s, the nozzle was closed by the finger, and the amount of sample left in the syringe was measured.

FDT

The FDT was performed to determine an extremely thick level (IDDSI level 4) in IDDSI. The FDT was performed with a disposable plastic fork purchased from a local store. The sample velocity of the plunge and the positioning, placement, and formation of the sample remaining on the fork were observed by lifting the fork out of a beaker containing 100 mL of the samples at room temperature, and the level was selected according to the IDDSI framework and detailed level definitions (version 2.0) (IDDSI, 2019).

Statistical analysis

Three repeated measurements were made, and the experimental results are presented as mean±standard deviation. Statistical Analysis System (SAS) version 9.4 (SAS Institute) was used to compare the results using the analysis of variance and Duncan’ multiple range test. Statistical significance was confirmed at P<0.05.

RESULTS AND DISCUSSION

Table 3 shows the apparent viscosity (η_a,50) and SFT values of the thickened drinks prepared with different thickener concentrations. The η_a,50 and SFT values were statistically significant depending on the thickener concentrations (P<0.05), indicating that a small difference in the concentration of the thickener makes a significant difference in the η_a,50 and remaining volume (mL) of each thickened drink (Jeong et al., 2021). As the thickener concentration increased, the η_a,50 and SFT values of each thickened drink increased. Among the thickened drinks, orange juice and milk needed a lower thickener concentration to reach thickness levels 3 and 4 when compared with water. This study revealed that drink composition is greatly influenced the results of the IDDSI flow test and NDD viscosity, showing that the η_a,50 and SFT values within each thickness level varied depending on the type of drink. Sopade et al. (2008) also reported that an increase in solids content, which is caused by an increase in the number of molecules, leads to intermolecular interactions, resulting in increased viscosity. Specifically, for thickened milk, when the protein as a long-chain polymer is mixed with water, the protein molecules interact with the thickener in the water, increasing fluid viscosity (Hadde et al., 2015). In thickened orange juice, in addition to the solids content, its viscosity might be increased by the friction created between the liquid and the syringe surface because, unlike other drinks, orange juice contains many particles with irregular shapes such as pulp. Moreover, the η_a,50 values of thickened water were higher than those of other thickened drinks at lower thickener concentrations (<0.8%). This can be explained by the difference in time to reach the equilibrium viscosity of thickened drinks. Previous studies have shown that the time to reach equilibrium viscosity is longer in thickened orange juice and milk samples than in thickened water, indicating that it was greatly dependent on the type of drink (Garcia et al., 2005; Cho and Yoo, 2015; Kim and Yoo, 2015).

Table 3.

Apparent viscosity (η_a,50) values in NDD and SFT values in IDDSI of thickened drinks at different thickener concentrations

Concentration (%)	Water			Orange juice			Milk
	ηa,50 (mPa･s)	SFT (mL)		ηa,50 (mPa･s)	SFT (mL)		ηa,50 (mPa･s)	SFT (mL)
0.1	16.21±0.25a	0.65±0.06a		−	−		−	−
0.2	50.80±0.65b	1.25±0.06b		26.73±0.34a	0.93±0.12a		28.25±0.97a	1.50±0.10a
0.3	63.30±0.08c	1.65±0.06c		33.61±0.19b	1.20±0.20a		42.43±0.45b	3.00±0.00b
0.4	97.05±0.29d	2.10±0.10d		79.36±0.13c	2.97±0.06b		74.11±1.75c	4.80±0.00c
0.5	116.0±0.11e	2.95±0.06e		96.68±0.83d	3.07±0.12b		92.20±0.40d	5.93±0.12d
0.6	181.2±1.80f	3.87±0.12f		164.2±0.90e	5.50±0.10c		144.7±0.21e	7.40±0.12e
0.7	219.5±0.60g	4.77±0.06g		199.0±0.18f	6.45±0.10d		219.4±0.19f	8.43±0.06f
0.8	274.5±0.49h	5.47±0.12h		286.6±0.43g	8.10±0.17e		334.0±1.41g	9.80±0.06g
0.9	304.4±2.20i	6.17±0.06i		303.6±0.22h	8.37±0.06f		474.8±0.00h	101)
1.0	340.4±4.72j	7.00±0.00j		382.4±0.74i	9.40±0.00g		575.0±3.90i	101)
1.1	377.2±4.34k	7.53±0.12k		525.6±2.42j	9.63±0.06h		697.6±6.71j	101)
1.2	411.3±12.1tl	7.83±0.06l		556.4±5.28k	9.73±0.12h		809.6±6.03k	101)
1.3	501.3±4.75m	8.87±0.12m		761.8±5.16l	101)		870.6±9.36l	101)
1.4	574.1±5.14n	8.97±0.06m		799.6±13.3m	101)		1,047±6.91m	101)
1.5	609.6±7.97°	9.27±0.12n		967.3±3.54n	101)		1,107±7.68n	101)
1.6	665.4±5.06p	9.47±0.06°		1,024±7.39°	101)		1,199±0.78°	101)
1.7	770.0±13.8q	9.80±0.00p		1,101±10.3p	101)		1,492±4.89p	101)
1.8	798.9±8.52r	9.83±0.06q		1,210±7.33q	101)		1,779±6.54q	101)
1.9	848.5±12.6s	9.90±0.00r		1,399±5.26r	101)
2.0	897.5±0.36t	9.97±0.06s		1,594±11.7s	101)
2.1	995.5±7.53u	101)		1,695±7.22t	101)
2.2	1,024±18.1v	101)		1,715±3.68u	101)
2.3	1,033±10.6v	101)		1,844±7.71v	101)
2.4	1,153±14.3w	101)
2.5	1,246±3.04x	101)
2.6	1,282±7.59y	101)
2.7	1,346±13.5z	101)
2.8	1,431±20.3α	101)
2.9	1,456±20.4α	101)
3.0	1,597±10.6β	101)
3.1	1,835±5.64γ	101)

Data are expressed as mean of three measurements±standard deviation.

Values in columns with different letters were significantly different (P0.05).

¹⁾The fork drip test was performed to determine an “extremely thick level” in IDDSI.

NDD, National Dysphagia Diet; SFT, syringe flow test; IDDSI, International Dysphagia Diet Standardization Initiative; −, not detected.

The thickener concentration ranges corresponding to NDD and IDDSI levels at the same thickness level were different (Table 4). For instance, the concentration range values of NDD and IDDSI level 2 thickeners were 0.2∼ 1.0% and 0.7∼1.2%, respectively. This result suggested that the IDDSI flow tests might not be sensitive enough to measure changes in the viscosity of thickened drinks. Nevertheless, the SFT is a simple, inexpensive, and practical clinical measurement tool to determine the IDDSI levels of thickened drinks (Park et al., 2014; Cichero et al., 2017; Kim et al., 2018; Jeong et al., 2021). The thickener concentration range for thickened drinks in each range of NDD and IDDSI levels increased in the following order: water> orange juice> milk (Table 4). Specifically, the thickener concentration ranges for thickened water in NDD and IDDSI levels 2 and 3 were greatly wider than those of other thickened drinks, and the thickened water needed more amounts of thickener to reach NDD and IDDSI level 4. These differences in the concentration of thickeners added to drinks can be attributed to the different compositions of the dispersing media that interact with the thickener-containing gums (Cho and Yoo, 2015). In the comparison of thickened milk containing solids, the thickened orange juice showed larger differences in the thickener concentration between all NDD and IDDSI levels. This means that orange juice, unlike other drinks, has many irregularly shaped particles, such as pulp; thus, its flow properties can be greatly affected by friction created by the flow of fluid through a syringe surface or fork prong. Thus, the difference between NDD and IDDSI levels according to the thickener concentration added may vary depending on the type of drink.

Table 4.

Thickener concentrations corresponding to the NDD and IDDSI levels for different thickened drinks (water, orange juice, and milk)

Drink	Thickness level	NDD	Concentration (%)	IDDSI	Concentration (%)
Water	2	Nectar-like	0.2∼1.0	Mildly thick	0.7∼1.2
	3	Honey-like	1.1∼3.0	Moderately thick	1.3∼2.9
	4	Pudding-like	>3.0	Extremely thick	>2.9
Orange juice	2	Nectar-like	0.4∼0.9	Mildly thick	0.6∼0.7
	3	Honey-like	1.0∼2.2	Moderately thick	0.8∼2.1
	4	Pudding-like	>2.2	Extremely thick	>2.1
Milk	2	Nectar-like	0.4∼0.8	Mildly thick	0.4∼0.6
	3	Honey-like	0.9∼1.7	Moderately thick	0.7∼1.7
	4	Pudding-like	>1.7	Extremely thick	>1.7

NDD, National Dysphagia Diet; IDDSI, International Dysphagia Diet Standardization Initiative.

Fig. 2 shows the FDT results of the extremely thick level for each thickened drink tested according to the definitions of IDDSI levels (version 2.0) (IDDSI, 2019). The thickener concentrations for IDDSI level 4 of thickened water, orange juice, and milk samples were 3.0%, 2.2%, and 1.8%, respectively, showing that the most of the thickened sample is above the fork and the small amount of sample forms a short tail below the fork prongs. The results of the FDT used to determine the extremely thick level in IDDSI were almost consistent with the pudding-like thickness obtained from the viscosity values in NDD, showing that there were not much differences in thickener concentration values between NDD level 4 (>3.0%, water; >2.2%, orange juice; >1.7%, milk) and IDDSI level 4 (>2.9%, water; >2.1%, orange juice; >1.7%, milk). From these observations, for some thickened fluid samples, the FDT may be a simple, practical, and suitable method for determining the IDDSI level 4 of some thickened drinks, even though FDT is a subjective and qualitative method (Karsten Hadde et al., 2016). Therefore, further studies on various thickened drinks with different components are needed to improve the accuracy of FDT.

Fig. 2.

Fork drip test results of the extremely thick level (level 4) in the International Dysphagia Diet Standardization Initiative for thickened drinks. (A) Water at 3.0% concentration, (B) orange juice at 2.2% concentration, and (C) milk at 1.8% concentration.

In conclusion, the thickener concentration ranges of thickened drinks for the classification of NDD levels differed from those of IDDSI levels, and they were influenced by the type of drink. Moreover, awareness of the differences in thickener concentration for classifying thickness levels between the NDD and IDDSI appears valuable for IDDSI classification using IDDSI flow tests (SFT or FDT) in clinical practice. The findings of this study may provide useful information for predicting the viscosity or thickness of thickened drinks based on the IDDSI flow tests and may indicate the need for improved experimental guidelines regarding the IDDSI level settings of thickened drinks. Knowledge of the flow properties of thickened drinks using the IDDSI flow tests would be useful for providing valuable information for determining the reliable thickness levels in IDDSI.