Method to enhance swallowing safety and appeal of beverages for treating dysphagia based on rheological and sensory parameters
10568831 ยท 2020-02-25
Assignee
- Wisconsin Alumni Research Foundation (Madison, WI)
- The Regents Of The University Of Minnesota (Minneapolis, MN)
Inventors
- Heather N. Mendenhall (Madison, WI, US)
- Richard W. Hartel (Madison, WI, US)
- JoAnne Robbins (Madison, WI)
- Jacqueline A. Hind (Madison, WI, US)
- Zata M. Vickers (St. Paul, MN, US)
Cpc classification
A61K47/00
HUMAN NECESSITIES
A23L33/40
HUMAN NECESSITIES
A61K9/0053
HUMAN NECESSITIES
A61K9/0056
HUMAN NECESSITIES
A23L33/30
HUMAN NECESSITIES
International classification
A61B5/03
HUMAN NECESSITIES
A61B5/00
HUMAN NECESSITIES
A23L33/00
HUMAN NECESSITIES
A61K9/00
HUMAN NECESSITIES
Abstract
An edible composition of matter having an apparent viscosity of from about 150 cP to about 2000 cP at about 30 s1; a yield stress of from 0 Pa to about 20 Pa at 1 s.sup.1; and a flow index of from about 0.2 to about 0.6. The composition may include an imaging agent. The compositions are useful for providing sustenance to dysphagic subjects and for evalualuating dysphagia.
Claims
1. A method of providing sustenance to a dysphagic subject comprising feeding to the subject one or more edible compositions of matter having an apparent viscosity of from about 150 cP to about 2000 cP at about 30 s1; a yield stress of from 0 Pa to about 20 Pa at 1 s.sup.1; and a flow index of from about 0.2 to about 0.6.
2. The method of claim 1, wherein the apparent viscosity of the edible composition is from about 200 cP to about 2000 cP, the yield stress is from 0 to about 15 pA, and the flow index is from about 0.3 to about 0.5.
3. The method of claim 1, wherein the apparent viscosity of the edible composition is from about 250 cP to about 1800 cP, the yield stress is from 5 to about 14 pA, and the flow index is from about 0.3 to about 0.5.
4. The method of claim 1, wherein the apparent viscosity of the edible composition is from about 250 cP to about 1800 cP, the yield stress is from 0 to about 2 pA, and the flow index is from about 0.3 to about 0.5.
5. The method of claim 1, wherein the apparent viscosity of the edible composition is from about 200 cP to about 500 cP, the yield stress is from 0 to about 2 Pa, and the flow index is from about 0.3 to about 0.5.
6. The method of claim 5, wherein the apparent viscosity of the edible composition is from about 250 cP to about 400 cP.
7. The method of claim 5, wherein the apparent viscosity of the edible composition is from about 270 cP to about 330 cP.
8. The method of claim 1, wherein the apparent viscosity of the edible composition is from about 1200 cP to about 1800 cP, the yield stress is from 5 to about 14 Pa, and the flow index is from about 0.3 to about 0.5.
9. The method of claim 8, wherein the apparent viscosity of the edible composition is from about 1250 cP to about 1700 cP.
10. The method of claim 8, wherein the apparent viscosity of the edible composition is from about 1400 cP to about 1600 cP.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
DETAILED DESCRIPTION
(15) As noted earlier, apparent viscosity is the standard measure against which thickened fluids are typically measured. However, commercially-available thickened beverages, which are supposedly based on the standards of viscosity set by the commercially available diagnostic fluids, vary greatly from the apparent viscosity of, for example, the commercially available VARIBARbrand fluids.
(16) The present inventors have found that characterizing a fluid by its apparent viscosity is, by itself, insufficient to describe accurately and adequately the functionality of a thickened fluid. It has been found that whether a thickened fluid is perceived as satisfying and refreshing, with little to no mouth-coating, depends on other rheological parameters in addition to apparent viscosity. Thus, the crux of the present method is to identify these additional rheological parameters and to define the suitable ranges of these parameters that are key to developing thickened fluids that are safe to drink and satisfying to the subject. In other words, the conventional approach relies entirely upon apparent viscosity as the sole criterion by which thickened edible fluids are measured. The present inventors, however, have established other, results-effective rheologic variables that yield thickened fluids that are both safer for dysphagic subjects to consume and satisfying to the palate.
(17) There are many ways to characterize fluid rheology. The following approach was chosen based on two primary stages of swallowing. First, as an individual holds a liquid bolus in the oral cavity, very low shear rate is applied to the boluses he/she prepares to swallow. The viscosity at very low shear rates (<1 s.sup.1) is similar to a yield stress (.sub.o), or the force required to initiate motion of the fluid from rest. A certain amount of yield stress helps to prevent uncontrolled flow of the fluid toward the airway, until the oral mechanisms, including the tongue, provide sufficient force to initiate controlled flow. Secondly, once a swallow is initiated, the range of shear rates important in the mouth is from about 1 s.sup.1 to about 100 s.sup.1. It is within this rough range where many thickened fluids exhibit shear-thinning behavior. In this range of shear rates, the apparent viscosity of shear thinning fluids decreases with increasing shear rate. This shear thinning under shear flow plays a role in reducing the apparent viscosity of the fluid during the swallow, making it easier to completely clear the fluid from the mouth during the swallow.
(18) A Hershel Bulkley rheological model can be used to characterize fluids in this range:
=k.sup.n+.sub.0
Here, is shear stress, is shear rate, k is the consistency coefficient and n the flow index, and .sub.0 the yield stress. According to the present method and composition, four parameters are needed to define a biophysically-designed fluid with ideal swallowing characteristics. That is, the fluid must be refreshing, leave minimal mouth coating, yet be sufficiently cohesive, with advantageous flow rate precluding aspiration into the airway. The critical rheological parameters are apparent viscosity (measured at about 30 s.sup.1 to be consistent with the VARIBAR-brand standards), yield stress (or viscosity at very low shear rates), consistency coefficient, and flow index. Each of these parameters must fall within specific ranges to create a satisfactory thickened fluid for use with dysphagic subjects.
(19) Apparent Viscosity: Apparent viscosity is preferably between about 150 cP to 2000 cP, more preferably between about 200 cP and 2000 cP, and most preferably between about 250 cP and 1800 cP, all measured at about 30 s.sup.1. Nectar-type compositions have a preferred apparent viscosity of about 200 to 500 cP, more preferably about 250 cP to about 400 cP, and with a most preferred range of about 300 cP30 cP (all at 30 s.sup.1). (That is, about 270 cP to 330 cP at 30 s.sup.1.) Thin honey-type compositions have a preferred apparent viscosity of about 1200 to 1800 cP, more preferably about 1250 cP to about 1700 cP, and with a most preferred range of about 1500 cP (100 cP) all at 30 s.sup.1. (That is, about 1400 cP to about 1600 cP at 30 s.sup.1.)
(20) Yield Stress (.sub.0) or Low Shear (1 s.sup.1) Viscosity: The data presented in the Examples indicates that yield stress is an important criterion in terms of flow behavior. A higher yield stress requires greater force to initiate flow. Thin honey fluids with yield stress values in excess of about 20 Pa were observed to have a gelled consistency that did not pour. A fluid that does not easily pour influences perception of thickness. More importantly, if a thickened fluid does not flow readily, it will require that the subject tilt his or her head back in order to consume the fluid, which is contraindicated in a dysphagia characterized by incomplete airway closure. Fluids with higher yield stresses also pose difficulty when consumed through a straw because greater intraoral pressure is required to move the liquid through the straw. The ability to generate sufficient intraoral pressure is frequently reduced in dysphagic patients.
(21) Thin honey fluids with yield stress value less than about 10 Pa seem to flow readily. However, fluids with no yield stress are not easily formed into a bolus on the tongue in preparation for swallowing. Thus, thin honey fluids with a low but measurable yield stress (between 5 and 14 Ps at 1 s.sup.1) are facilitative for safer swallowing of thickened fluids. For nectar fluids, yield stress is less of a concern because of the lower concentrations of thickeners. Still, a finite value is needed to allow formation of the bolus on the tongue. Yield values between 0 and 2 Pa at 0.1 s.sup.1 are appropriate for nectar-thick fluids.
(22) Flow index (n): Flow index (n) is an indication of how a fluid responds at different shear rates. Water is a Newtonian fluid with a flow index of n=1.0, meaning its apparent viscosity does not change with shear rate. Lower n-values indicate more shear thinning behavior: as more force (shear) is applied, the fluid appears thinner, with lower apparent viscosity. Flow index was found to have a significant effect on perceived thickness of thickened fluids. The selection of an ideal target flow index (n) is a key criterion to improving the sensory attributes of thickened fluids. According to the present disclosure, fluids having a flow index falling between about 0.2 to about 0.6, more preferably between about 0.3 and about 0.5 produce lower, more desirable, scores for perceived thickness. Fluids with higher n-values are consistently rated as thicker, less desirable and more mouth-coating. Fluids with lower flow indices (less than 0.2) are highly shear thinning and typically exhibit higher values for yield stress and gel-like structures that are sensorially undesirable and physiologically problematic for dysphagic patients.
(23) Consistency coefficient (k): Of the four rheological parameters that have been defined, only the three listed previously (apparent viscosity, yield stress, and flow index) are independently set. The consistency coefficient is a dependent variable. The k-value of a thickened fluid is automatically determined by the target apparent viscosity required to match a desired endpoint, such as the apparent viscosity of a diagnostic standard.
(24) In short, the present method uses rheological parameters beyond apparent viscosity in order to achieve a beverage that is appropriately designed for therapeutic use, as well as patient enjoyment and appeal. That is, the present method yields compositions that are adequately viscous for use with dysphagic individuals, but are perceived as less thick and more refreshing than conventional thickened liquids that evaluated solely by their apparent viscosity. More appealing thickened fluids for the treatment of dysphagia will result in improved compliance among patients, thus reducing the health-related consequences associated with aspiration due to non-compliance.
(25) It is preferred that the edible solutions comprise a tasty and familiar base liquid vehicle, such as a non-pulpy fruit juice, or a liquid that has been treated to have an identifiable food flavor. Water will also suffice. The vehicle is adjusted to the required apparent viscosity, yield stress, and flow index by adding a suitable thickening agent thereto. See the Examples for a non-limiting list of suitable thickening agents. Any food-grade thickening agent, now known or developed in the future, may be used. Apple juice, for example, is a suitable vehicle. In the case where a contrast agent is included in the formulation, a vehicle having a familiar taste generally yields more useful information regarding the swallowing defects exhibited by the patient because the patient tends to swallow more naturally. This is not the case when a patient is offered a contrast solution that has an offensive taste, odor, or consistency.
(26) The utilities of the subject compositions and methods are several-fold. A primary utility is that by using standardized solutions, consistency in diagnosing and treating dysphagia is promoted. Rather than supplying patients an arbitrarily thickened food or X-ray imaging product, the patient is supplied a solution of known rheological properties. The patient's ability to swallow the solution properly (e.g., without aspiration, retention, regurgitation, and the like) is greatly improved. The subject's condition can also be evaluated consistently over time, either by a gross physical exam or radiographic means or other visualization means, including X-ray, magnetic resonance imaging, and the like. Using two or more standardized solutions as described herein allows the results of two (or more) distinct swallowing studies (e.g., one using the first solution, the other using the second solution) to be compared and contrasted. Moreover, it allows the results from different patients to be compared directly, without variations in the rheological properties of the edible solutions introducing uncontrolled variables into the comparison.
(27) The solutions described herein are useful in radiographic imaging of the mouth because they taste more food-like than conventional barium-containing imaging agents and are therefore more palatable. The rheological properties recited herein for the solutions also promote the proper amount of adhesion between the solutions and the mucus membranes lining the mouth and throat. Consequently, the solutions tend to deposit a sufficient amount of imaging material on the mucus membranes to generate a radiographic image, but do not deposit so much imaging material as to change the swallowing dynamics of the patient under study, nor to leave an artificial coating after swallowing is complete. This is a distinct improvement over conventional barium agents, whose thick, chalky consistency is neither palatable, nor conducive to the generation of good radiographic images of the throat and mouth.
(28) As used herein the term imaging agent means any imaging agent, now known or developed in the future, that functions to image the swallowing process or functions to facilitate imaging of the swallowing process. Barium-containing imaging agents are preferred for use with videofluoroscopic imaging of swallowing. Barium sulfate is most commonly used.
(29) Numerical ranges as used herein are intended to include every number and subset of numbers contained within that range, whether specifically disclosed or not. Further, these numerical ranges should be construed as providing support for a claim directed to any number or subset of numbers in that range. For example, a disclosure of from 1 to 10 should be construed as supporting a range of from 2 to 8, from 3 to 7, 5, 6, from 1 to 9, from 3.6 to 4.6, from 3.5 to 9.9, and so forth.
(30) All references to singular characteristics or limitations of the present disclosure shall include the corresponding plural characteristic or limitation, and vice-versa, unless otherwise specified or clearly implied to the contrary by the context in which the reference is made.
(31) All combinations of method or process steps as used herein can be performed in any order, unless otherwise specified or clearly implied to the contrary by the context in which the referenced combination is made.
(32) The methods of the present invention can comprise, consist of, or consist essentially of the essential elements and limitations of the method described and claimed herein, as well as any additional or optional ingredients, components, or limitations described herein or otherwise useful in formulating edible compositions of specifically defined rheological properties.
(33) Liquids may be described by several rheological properties, as discussed above. One of those properties, apparent viscosity, is measured with a viscometer. In simple terms, apparent viscosity is the torque required to rotate a spindle immersed in the solution being analyzed. As discussed in detail by Li et al (1992) Dysphagia 7:17-30, viscous liquids are classified as either Newtonian or non-Newtonian. A Newtonian liquid flows at a rate that is directly proportional to the applied shear force while a liquid not obeying this proportionality is deemed to be non-Newtonian.
(34) The complexity and cost of viscometers makes them impractical for dietitians and speech pathologists to utilize in most clinical settings. Therefore, clinicians often resort to simpler means of generating a rough estimate of viscosity, such as the line spread test (LST), as a practical alternative to characterizing the viscosity of thickened liquids.
(35) The LST was originally developed in the 1940's as a simple tool to measure food consistency. See 8. Grawemeyer & Pfund (1943) Food Research 8:105-108, and Adams & Birdsall (1946) Food Industries 78-80. The LST measures the flow of a food/liquid by placing a standard amount of food/liquid in a cylinder, lifting the cylinder and allowing the food/liquid to flow on a horizontal surface for 1 minute and then measuring the distance it has spread. The simplicity of this tool is extremely appealing in clinical practice to measure objectively the general consistency of thickened liquids. As described in the Examples, the LST can roughly distinguish between the viscosity of pre-thickened liquids, and may be a first step toward establishing the utility of this tool in clinical practice.
(36) However, the LST is not overly discriminating as a tool to measure apparent viscosity. Thus, one of the objectives of the methods and compositions described herein is to provide a set or kit of solutions of known and standardized rheological properties, one of which is apparent viscosity.
(37) Viscosity of the compositions is measured at room temperature, generally about 23 C., and can be determined using any number of conventional and commercially available spindle-type viscometers, such as those manufactured by Brookfield Engineering Laboratories, Middleboro, Mass. Brookfield's instruments use a rotating spindle immersed in the fluid to measure viscosity. Brookfield's instruments are of the cone-and-plate rheometer-type design and are ideal for measuring the viscosity of non-Newtonian fluids at low shear rates. The preferred instrument from among Brookfield's offerings is Model LVDV1+, an 18-speed model with digital readout. Viscometers and their operation are widely known and will not be described in great detail herein. For detailed information regarding Brookfield's Model LVDV1+ viscometer, see Brookfield's Manual No. M/92-021-M0101, available from Brookfield. (It is also available on-line at www.brookfieldengineering.com.) For detailed information regarding measuring viscosity, see Li et al, supra, and Brookfield's technical manual entitled More Solutions to Sticky Problems (also available at www.brookfieldengineering.com).
(38) For radiographic imaging of the mouth and throat, it is preferred that compositions having the above-noted rheological properties are formulated using a liquid base vehicle having an identifiable food flavor, such as a juice flavor or honey, chocolate, vanilla, etc. Apple juice is particularly preferred a base vehicle. Apple juice is very advantageous for this purpose because it is widely available, relatively inexpensive, and pulp-free. It is quite palatable and familiar to virtually everyone, and can be stored and transported as a concentrate. The liquid base can be naturally or artificially flavored, and can also contain additional components such as colorants, preservatives, and the like. To the liquid vehicle is added a thickening agent and/or a radio-opaque imaging agent. Because a suspension of radio-opaque material will, by itself, alter the rheological properties of the base solution, depending upon the nature of the suspension used, a thickening agent may not be required to arrive at solutions having the desired rheological characteristics.
(39) Any other type of non-pulpy juice, liquid, or water may be used as the vehicle. A fruit juice is much preferred as the vehicle, however, because of its familiar taste and aroma. An ultimate goal of the method being an accurate evaluation of the patient'[s true swallowing dynamics, presenting an imaging composition which is as closely simulative as possible to a food the patient would normally ingest and enjoy is highly desirable. If the formulation is presented without an imaging agent, the ultimate goal is to provide a satisfying foodstuff that the patient can ingest safely, with a reduced risk of aspirating the formulation or otherwise having the formulation enter the airway rather than the alimentary canal.
(40) If the formulation is to include an imaging agent, the preferred radio-opaque imaging agent is a suspension of barium sulfate. Suitable barium sulfate and barium sulfate suspensions are available commercially from numerous sources. Preferred commercially available barium sulfate and edible suspensions thereof can be obtained from Bracco Diagnostics Inc., Princeton, N.J.
(41) For radiographic imaging purposes, the patient is positioned laterally before a suitable fluoroscopic device and asked to swallow one or more of the three compositions. A videofluoroscope and suitable recording equipment are then used to visualize and record the passage of the composition through the mouth and throat during and after swallowing. If desired, the study can be performed using any combination or both of the two edible solutions.
EXAMPLES
(42) The following examples are submitted to describe in greater detail the methods and compositions described herein. The examples are not intended to limit the scope of the claims in any fashion.
Example 1
Analysis of Thickeners with Dysphagic and Healthy Subjects
(43) Subjects:
(44) Twenty-three (23) healthy patients (22-72 years, mean=40 yr) and fifteen (15) dysphagic control subjects (23-92 years, mean=67 yr) participated in this study, which was approved by the University of Wisconsin Health Sciences Institutional Review Board.
(45) Methods:
(46) 1. Videofluoroscopic Oropharyngeal Evaluation: All subjects swallowed two 5 ml trials from a spoon of VARIBAR-brand Nectar, VARIBAR-brand Thin Honey, and each of the six prototype thickened fluids containing 3% barium to make them radiopaque (see Table 3) for a total of 16 swallows. Imaging was completed laterally and pressure data were simultaneously recorded using the Digital Swallowing Workstation. Simultaneous, time-linked oral pressures were measured using the three-sensor array depicted in
(47) Data Reduction: (a) Residue: Post-swallow barium residue was judged using a three-point scale (0=no residue, 1=coating of residue, 2=pooling of residue) for each swallow at four locations (oral cavity, valleculae, posterior pharyngeal wall, and upper esophageal sphincter). (b) Timing: Durations of critical swallowing events were assessed via a frame-by-frame analysis of the videofluoroscopic images. (c) Pressure: Maximum pressure during every swallow was recorded at each sensor location (anterior, middle and posterior). To ensure that the peak pressure was associated with bolus transit the measurement was taken between the time of beginning of posterior bolus movement and when the bolus tail passed into the upper esophageal sphincter.
(48) 2. Sensory Assessment: Subjects completed a sensory written evaluation after tasting each prototype fluid using a questionnaire with labeled affective magnitude and visual analog scales assessing overall like/dislike, extent of flavor lingering, amount of mouth-coating and degree of thirst-quenching.
(49) 3. Attribute Assessment: Dysphagic subjects also completed an online questionnaire regarding types of thickeners, if any, they had used and their preference for various attributes of thickeners.
(50) Results:
(51) Residue: Residue data are presented in
(52)
(53) Maximum pressures generated during swallowing were ranked within subjects for fluids within the same viscosity group (300 cPs and 1500 cPs). The results are shown in
(54) Sensory:
(55) Dysphagic and healthy subjects rated agar highest in Overall Liking compared to other 300 cP fluids. See
(56) Attributes:
(57) Dysphagic subjects were provided a list of 35 potential attributes and asked which ones they would prefer to have in thickened liquids they purchase. The attributes ranked as most preferred were high protein, thirst quenching, high fiber, clear (non-milky), low sweet and chocolate flavored.
(58) Conclusions:
(59) Overall, the fluids most similar rheologically to the barium standards (agar in the 300 cP group and tara gum in the 1500 cP group) were observed to result in the least amount of post-swallow residue, meaning that patients would be less at risk for aspirating the fluid after the swallow. Additionally, agar and tara gum required the lowest lingual pressures to move the bolus through the oral cavity and were cleared faster from the oral cavity indicating that weaker, more disabled patients would be able to swallow these beverages more safely than the other thickened fluids. When patients were questioned about their preferences, agar was rated highest in overall liking and both agar and tara gum were rated highest in thirst quenching-both of which are critical characteristics for compliance with drinking thickened liquids.
Example 2
Assessing Thickened Beverages Using a Variety of Thickeners
(60) Samples:
(61) The 15 thickeners listed below in Table 1 were used to prepare 30 thickened beverage samples at nectar and thin honey viscosities. These beverages were formulated to match the viscosities of VARIBAR-brand diagnostic fluids: .sub.30 s1=30030 cP for nectars and .sub.30 s1=1500100 cP for thin honeys. (In the records of the U.S. Patent & Trademark Office, VARIBAR is a recorded as registered trademark of E-Z-EM, Inc., Lake Success, N.Y., USA. However, E-Z-EM was acquired by Bracco Diagnostics Inc. in April of 2008.) Individual samples of 28 ml were removed from the refrigerator an hour before serving and served at room temperature in 60 ml cups labeled with random, 3-digit codes. New codes were generated for each test session.
(62) All beverage samples were prepared using a water-based base liquid. The base liquid was sweetened with 3% (w/w) sucrose, and pH was adjusted to 4.450.03, using citric acid (Tate & Lyle, Decatur, Ill., USA) and buffered with sodium citrate (Tate & Lyle). Prepared samples were flavored using lemonade-type flavor 645081 (Guividan Flavors Corp., Cincinnati, Ohio), at a rate of 0.24% (w/w).
(63) Thickeners were mixed into base liquid using a rotational mixer (Controller: Master Servodyne Controller, Cole Palmer Instrument Co., Chicago, Ill., USA; Rotational head: Model E650, 1/16- in CAP, -24 THD, Robbins Meyers Electro-Craft, Hopkins, Minn., USA) equipped with a 12 5/16 in diameter stainless steel impeller shaft (Model A712, Caframo, Wiarton, ON, Canada) and a 1.5 in stainless steel blade (Caframo).
(64) Rheological Profiling:
(65) Apparent viscosity of fluid samples at 30 s.sup.1 was initially determined using Brookfield DVIII-Ultra Programmable Rheometers (Brookfield Engineering, Middelboro, Mass., USA), models HA and RV, with a concentric cylinder configuration. All experiments were conducted under isothermal conditions at 22.00.1 C. A stepped shear rate test (ascending and descending) was used, ranging from 3.75 s.sup.1 to 60 s.sup.1 and allowing for ten revolutions at each step. The shear rate and shear stress data were determined by Rheocalc-brand software (Brookfield Engineering).
(66) Further rheological profiling was performed on the same samples using a Universal Dynamic Spectrometer (Paar Physica UDS 200 Controlled Stress Rheometer, Physica Messtechnik GmbH, Stuttgart, Germany). Flow curves were determined using logarithmically stepped shear rates from 1000 s.sup.1 to 1 s.sup.1. Apparent viscosity (30 s.sup.1, 22 C.), flow index (n), consistency coefficient (K), and yield stress (.sub.0) were measured for all beverage samples. Data were recorded and fit to mathematical models using Physica
(67) RheoPlus-brand software (Physica Messtechnik). Yield point values for the fluids were determined by the stress required to initiate flow at constant shear. One shear rate did not work well for both nectar and thin honey consistencies, so different shear rates were used for each; for nectars, =0.1 s.sup.1, and for thin honeys, =1.0 s.sup.1.
(68) Subjects:
(69) Twenty-four subjects (6 males, 18 females) aged 19- 58 with a median age of 26, were recruited from the University of Minnesota using an email screener. Subjects having a swallowing disorder and/or any food allergies were excluded. Subjects were paid a cash incentive for their participation. The University of Minnesota Institutional Review Board approved this study.
(70) TABLE-US-00001 TABLE 1 Thickeners Used for Beverages Usage Level Thickener Trade Name Supplier Consistency (% w/w) Agar Quick Gelagar - Setexam Thin Honey 0.87 QT30 (Kenitra, Morocco) Nectar 1.72 Sodium alginate Satialgine S 1100 Cargill Thin Honey 1 (Minnetonka, MN) Nectar 1.7 Iota (.Math.) carrageenan Viscarin SD 389 FMC BioPolymer Thin Honey 1.09 (Philadelphia, PA) Nectar 1.77 Cellulose gum TIC Pretested TIC Gums Thin Honey 2.93 Ticaloid EZ-1900 (White Marsh, MD) Nectar 5.36 Powder Methylcellulose Ticacel LV TIC Gums Thin Honey 1.84 Powder Nectar 2.94 Microcrystalline Avicel AC 4125 FMC BioPolymer Thin Honey 4.78 cellulose Nectar 6.59 Guar gum Guar Gum FG6070 P.L. Thomas & Co., Inc. Thin Honey 0.56 (Morristown, NJ) Nectar 1.04 Konjac gum Ticagel Konjac TIC Gums Thin Honey 0.32 HV Nectar 0.59 Tara gum TIC Pretested Tara TIC Gums Thin Honey 0.55 Gum 100 Nectar 0.97 Xanthan gum Ticaxan Rapid-3 TIC Gums Thin Honey 0.53 Powder Nectar 2.16 Calcium caseinate Ca. caseinate Spray DMV International Thin Honey 11.51 (Veghel, Netherlands) Nectar 13.1 High methoxyl Unipectine AYD Cargill Thin Honey 3.25 pectin 258 Nectar 4.96 Low methoxyl Coyote Brand Pectin Gum Technology Thin Honey 6.35 pectin LM 0929 Corporation Nectar 9.77 (Tucson, AZ) Waxy rice starch Remyline XS-B - A&B Ingredients, Inc. Thin Honey 3.61 Extra stable waxy (Fairfield, NJ) Nectar 5.26 rice starch Acetylated distarch Stabitex - Instant Cargill Thin Honey 4.33 phosphate 12625 Nectar 5.43
(71) Training Session:
(72) Panelists attended one training session during the first week of the study. At the beginning of the session, they were provided with a lexicon based on the texture lexicon described by Hootman (1992). The lexicon was modified using panelist input to describe relevant differences among the 30 products sampled (Table 2).
(73) TABLE-US-00002 TABLE 2 Thickened Beverage Texture Lexicon Attribute Description Thickness The force required to push the liquid against the roof of the mouth. Adhesiveness The amount of work needed to remove the liquid from the palate. Stickiness The amount of force needed to pull the tongue away from the roof of the mouth. Touch the tongue to the roof of the mouth, then pull it directly down. Slipperiness The enhanced lubricating or friction-reducing qualities at the back of the throat as you swallow Mouth-coating The degree to which the product coats the inside of the mouth after swallowing
(74) During the training session, each panelist was provided with a subset of ten different thickened beverages chosen as they best represented the entire set of 30 beverages in terms of thickness and mouth-coating, and included all chemical categories of thickeners. The different sensory attributes were illustrated by selecting samples from this set that varied perceivably in each sensory attribute. Panelists were asked to plug their noses using nose clips while rating tastes in order to avoid aromas. The remaining sensory attributes were rated without nose clips. All flavors besides lemon flavor were grouped together and rated on an intensity scale as Other Flavor. Panelists were also instructed how to rate the different textural attributes: thickness, adhesiveness, stickiness, slipperiness and mouth-coating (Table 2).
(75) The panelists also learned a standardized sampling method to minimize variability in delivery as well as perception of the sensory attributes. The standardized sampling method ensured that each panelist evaluated the same amount of each sample. The standardized method required that they first gently stir the sample 30 times with a plastic spoon (Bakers and Chefs Plastic Spoons, Bentonville, Ark., USA) in order to mimic the stirring done before instrumental measurements were taken. Panelists were prompted to do this for each beverage during the training and test sessions by computerized instructions. Panelists then had to fill the spoon full with the sample, level it off if necessary with a plastic coffee stirrer (Propak Coffee/Bar Stirrers), place the nose clips on, put the spoonful of sample in their mouth and evaluate the sample for sweet, sour and bitter taste, remove the nose clips and evaluate the sample for lemon and other flavor. (These attributes were rated by the panelists to prevent them from placing these sensations into ratings of the texture attributes.) Panelists were allowed to expectorate this first spoonful of sample after evaluating it. Then, they had to consume a second spoonful of the same sample and evaluate thickness, adhesiveness, stickiness and slipperiness before swallowing. Mouth-coating and number of swallows required to cleanse the palate were evaluated after the second spoonful was swallowed.
(76) Data Collection Sessions:
(77) The seven data collection sessions were divided into four sensory evaluation sessions, one lingual pressure measurement session and two residual mouth-coating measurement sessions.
(78) Sensory Evaluation Sessions:
(79) Panelists typically attended four sensory evaluation sessions during two consecutive weeks. The set of 30 thickened beverages was randomly divided into two groups of 15 beverages. The same 15 beverages were served to all panelists at each session. The position in which each of these 15 beverages was served at each session was determined by a design for 15 treatments in two blocks balanced for the effect of order of presentation and first-order carry-over effects (MacFie et al., 1989). Panelists sampled the complete set of 30 beverages in the first week (i.e. 15 beverages during session 1 and the remaining 15 during session 2). Similarly, all 30 beverages were sampled again through the second week.
(80) The panelists tasted each beverage using the standardized sampling method described in the training session. At each session, the panelists rated each beverage for tastesweet, sour and bitter; flavorlemon and other; and texturethickness, adhesiveness, stickiness, slipperiness, mouth-coating, and the number of swallows required to clear the palate. Ratings for all these sensory attributes were collected on a 15-point numerical category scale, with end-points of 0 and 15. The scale had tick marks labeled with numbers at regular intervals of 1 unit. Panelists could choose any point on the scale from 0 to 15, with end anchors none and intense. Panelists also noted the number of swallows required to cleanse the palate after the thickened beverage was swallowed. The first swallow was not included in this rating. Twenty-two (22) panelists completed all four sensory evaluation sessions.
(81) Lingual Pressure Measurement Session:
(82) Panelists attended one swallowing pressure measurement session. A three-sensor array was affixed to the hard palate using Stomahesive-brand adhesive (ConvaTec, Skillman, N.J., USA) with the anterior sensor at the alveolar ridge and the posterior sensor at the approximate junction of the hard and soft palates. See
(83) Residual Mouth-Coating Measurement Session:
(84) Panelists attended two, hour-long residual mouth-coating measurement sessions during two consecutive weeks. Riboflavin (1 g/L; Sigma-Aldrich Co., St. Louis, Mo.) was incorporated into the base liquid of all the thickened beverages during preparation. The complete set of 30 thickened beverages was randomly divided into two groups of 15 beverages. At each session, all panelists were served the same set of 15 thickened beverages. The positions in which these 15 beverages were served was determined by a design for 15 treatments in two balanced blocks of 30 consumers each so as to balance the effect of order of presentation and first-order carry-over effects (MacFie et al., 1989).
(85) A rinse-and-spit protocol was developed using the following steps: panelists gently stirred the sample 30 times with a plastic spoon (Bakers and Chefs Plastic Spoons, Bentonville, Ark.), filled the spoon full with the sample, leveled it off if necessary with a plastic coffee stirrer (Propak Coffee/Bar Stirrers), and swallowed this spoonful. Panelists then rinsed their mouth with 10 ml of water, swishing it three times inside the mouth, and spit this rinse water into a 16 oz spit cup (Dart Foam Cups, Mason, Mich., USA) labeled with the same code number as the sample. This rinse-and-spit protocol was repeated five times after swallowing each sample. (At the start of each session, panelists swallowed a spoonful of water instead of the thickened beverage sample, to familiarize them with the protocol as well as to remove food particles from their mouth.)
(86) The riboflavin concentration in the rinse water from each judge for each sample was determined using a chemical analysis and fluorescent spectroscopy method (Scott et al, 1946). The total riboflavin content in the rinse water was calculated by multiplying this riboflavin concentration by the volume of rinse water. We considered the total riboflavin content to be a measure of residual mouth-coating.
(87) Data Analysis:
(88) Raw data from the tests above were compiled on Microsoft Excel-brand sheets. Statistical analyses were carried out using SAS computer software (The SAS System for Windows, Version 9.2, The SAS Institute Inc., Cary, N.C., USA) and XLSTAT 2010 (Addinsoft SARL, New York, N.Y., USA). Unless otherwise noted, statistical significance was set at 0.05. Although the taste and flavor measurements varied among the beverages, they were unrelated to any of the rheological, sensory texture or swallowing pressure measurements and are not included in any data analyses reported here.
(89) Linear regression fit by restricted maximum likelihood (REML) was used to relate sensory and swallowing pressure measurements to the rheological measurements. Similar analyses were run separately for thin honey beverages and nectar beverages in order to visualize relationships within each of the two viscosity ranges.
(90) Principal components analysis with varimax rotation (XLSTAT) was used to summarize the relationships among the three groups of measurements (sensory, rheological and swallowing pressure) for the nectar and the thin honey beverages separately. The main data set for each principal components analysis contained the 15 beverages as observations and the rheological measurements as variables. Sensory texture measurements and swallowing pressures for each of the 15 beverages were added as supplementary variables.
(91) The relationship between the sensory perception of mouth-coating and instrumental measurement for predicting mouth-coating was examined using linear regression. The dependent variable was the sensory perception of mouth-coating; the instrumental measurement of residual riboflavin served as a predictor. Each judge was included in the model as a dummy variable.
(92) Selection of Thickeners to be Used in Videofluoroscopy:
(93) Six thickeners were chosen from the fifteen thickeners initially selected for formulation to move forward to videofluoroscopy studies. Three different thickeners were used to produce fluids at Nectar consistency and three others at Thin Honey consistency. See Table 3. Thickener selection for two of the fluids was based on rheological similarities (apparent viscosity, flow index, and consistency) to the diagnostic fluids (VARIBAR-brand Nectar and Thin Honey). The remaining four thickeners were selected based on sensory attributes. Those thickeners that were rated at the furthest sensory extremes (i.e., minimum or maximum thickness, mouth coating, adhesiveness, slipperiness) were chosen for Nectar or Thin Honey consistency. The thickeners selected for Nectars were xanthan gum, methylcellulose gum, and agar. For Thin Honeys, the thickeners selected were high methoxyl pectin, iota carrageenan, and tara gum.
(94) TABLE-US-00003 TABLE 3 Rheology of Thickeners Selected for Videofluoroscopic and Sensory Analysis Apparent Yield Viscosity Flow Consistency Stress (.sub.0), Thickened Fluid (.sub.30s1), cP Index (n) (K) Pa .Math. s Diagnostic Standards VARIBAR Thin Honey 1440 0.53 8104 10250 VARIBAR Nectar 231 0.34 2468 1510 Rheology-based Choices Tara Gum Thin Honey 1413 0.38 9970 6918 Tara Gum Thin Honey + 3% Barium 1405 0.38 9879 7210 Agar Nectar 282 0.33 2898 2410 Agar Nectar + 3% Barium 287 0.3236 3145 2225 Sensory- based choices HM Pectin Thin Honey 1498 0.74 3125 2295 HM Pectin Thin Honey + 3% Barium 1470 0.74 3048 2530 .Math.-CRG Thin Honey 1248 0.28 15168 35925 .Math.-CRG Thin Honey + 3% Barium 1395 0.25 18710 35400 Xanthan Gum Nectar 247 0.22 3756 2853 Xanthan Gum Nectar + 3% Barium 242 0.20 4020 3045 Methylcellulose Nectar 263 0.56 343 38 Methylcellulose Nectar + 3% Barium 276 0.59 361 44
Results:
(95) The relatively huge differences in viscosity between the two thickness levels (nectar and thin honey) overwhelmed comparisons among the thickeners, so the observations were separated into those for the nectar beverages and those for the thin honey beverages. The object was for the beverages within these subsets to have very similar apparent viscosities, thus shifting the focus towards relationships of the other rheological properties with the sensory and swallowing measurements.
(96) Nectar Beverages:
(97) The sensory properties thickness, stickiness, adhesiveness, mouth-coating and number of swallows were highly negatively correlated with principal component 1, and highly positively correlated with the flow behavior index (n-value). See
(98)
(99) TABLE-US-00004 TABLE 3 Best linear regression equations, fit by restricted maximum likelihood (REML), relating sensory and swallowing pressure measurements to the rheological measurements for nectar fluids only. Values are beta coefficients (the change in the sensory or swallowing pressure measurement produced by a one unit increase in the rheological measurement.) Empty cells indicate no significant relationship between the sensory or swallowing pressure measurement and the rheological measurement. Sensory/ Rheological measurements Swallowing Apparent Adjusted Attribute Intercept viscosity n-value k-value R.sup.2 Thickness 2.6 7.6 0.00084 0.86 Adhesiveness 4.7 10.2 0.0001 0.80 Stickiness 2.2 6.3 0.00051 0.69 Slipperiness 13.8 9.7 0.0012 0.74 Mouth coating 3.6 11.8 0.0013 0.72 No. of swallows 1.15 5.48 0.00056 0.80 Bulb 1 154 Bulb 2 154 Bulb 3 170
(100) Thin Honey Beverages:
(101) The sensory properties adhesiveness, stickiness, mouthcoating and number of swallows were positively correlated (r>0.6) with component 1 and with the n-value. See
(102)
(103) TABLE-US-00005 TABLE 4 Best linear regression equations, fit by restricted maximum likelihood (REML), relating sensory and swallowing pressure measurements to the rheological measurements for thin honey fluids only. Values are beta coefficients (the change in the sensory or swallowing pressure measurement produced by a one unit increase in the rheological measurement.) Empty cells indicate no significant relationship between the sensory or swallowing pressure measurement and the rheological measurement. Sensory/ Rheological measurement Swallowing Apparent n- Yield Adjusted Attribute Intercept viscosity value k-value stress R.sup.2 Thickness 3.8 5.8 0.082 0.34 Adhesive- 1.7 12 0.6 ness Stickiness 2.1 6.2 0.54 Slipperiness 5.4 2.5 0.34 Mouth 2.4 7.9 0.53 coating No. of 2.6 3.6 0.54 swallows Bulb 1 177 0.65 0.25 Bulb 2 176 0.001 0.36 Bulb 3 188
(104) Sensory Attributes and Swallowing Pressures:
(105) Relationships between the swallowing pressures and the sensory attributes were generally weak, and even in the cases of significance, linear relationships had low R.sup.2 values. Swallowing pressures for nectar beverages were negatively related to the number of swallows and positively related to thickness. See Table 5. Swallowing pressures for thin honey beverages were not consistent among the three bulbs. See Table 6.
(106) TABLE-US-00006 TABLE 5 Best linear regression equations, fit by restricted maximum likelihood (REML), relating swallowing pressures and sensory attributes. Data for nectar thick samples Swallowing Mouth No. of Adjusted pressure Intercept Thick Adhesive Sticky Slippery coating swallows R.sup.2 Bulb 1 146 13.7 11.3 0.02 Bulb 2 181 9.8 0.02 Bulb 3 163 9.5 7.3 0.01
(107) TABLE-US-00007 TABLE 6 Best linear regression equations, fit by restricted maximum likelihood (REML), relating swallowing pressures and sensory attributes. Data for thin honey samples Swallowing Mouth- No. of Adjusted pressure Intercept Thick Adhesive Sticky Slippery coating swallows R.sup.2 Bulb 1 133 8.0 0.19 Bulb 2 161 12 15.8 0.16 Bulb 3 183 7.7 7.1 0.30
(108) The amount of riboflavin left in the mouth was positively correlated with the sensory mouth-coating scores. A 1 mg increase in riboflavin corresponded to a 4.1 unit increase in the sensory perception of mouth-coating (p<0.001;