Trials. 2013 May 7;14:126.

Tongue pressure profile training for dysphagia post stroke (TPPT): Study protocol for an exploratory randomized controlled trial.

Catriona M Steele1,2,3*, Mark A Bayley1,2, Melanie Péladeau-Pigeon1 and Shauna L Stokely1

1Swallowing Rehabilitation Research Laboratory, Toronto Rehabilitation Institute – University Health Network, 550 University Avenue #12-125, Toronto, ON M5G 2A2, Canada.

2University of Toronto, Toronto, Canada.

3Bloorview Research Institute, Holland Bloorview Kids Rehab, Toronto, Canada.

 

Abstract                  

Background: It is estimated that approximately 50% of stroke survivors will experience swallowing difficulty, or dysphagia. The associated sequelae of dysphagia include dehydration, malnutrition, and aspiration pneumonia, all of which have can have serious medical consequences. To improve swallowing safety and efficiency, alternative nutritional intake methods (for example, a feeding tube) or a modified diet texture (such as pureed foods or thickened liquids) may be recommended but these modifications may negatively affect quality of life. An alternative approach to treating dysphagia has emerged over the past few years, targeting stronger lingual muscles through maximal isometric pressure tasks. Although these studies have shown promising results, thin-liquid bolus control continues to be challenging for patients with dysphagia. Previous work investigating lingual pressures when healthy participants swallow has suggested that greater task specificity in lingual exercises may yield improved results with thin liquids.

Methods/design: This is a small, exploratory randomized clinical trial being conducted with post-stroke patients 4 to 20 weeks after onset of dysphagia secondary to impaired lingual control. At enrollment, participants are randomly assigned to one of two treatment protocols, either tongue pressure profile training (TPPT) or the control treatment, tongue pressure strength-and-accuracy training (TPSAT). Each treatment protocol consists of 24 sessions of treatment over 8 to 12 weeks with monitoring of tongue pressure as well as a baseline and outcome videofluoroscopic swallowing study. Tongue pressure measures, videofluoroscopic measures, and functional outcome measures will be obtained following training of 60 participants (30 in each condition), to determine whether TPPT yields better outcomes.

Discussion: This study will continue to explore options beyond tube feeding and modified diets for people with neurogenic dysphagia following stroke. Should the novel protocol, TPPT, prove to be more effective than the TPSAT protocol, this may influence standards of care and best practices for patients with dysphagia involving impaired thin-liquid control as a result of stroke.

Trial registration: Clinicaltrials.gov NCT01370083

Keywords: Deglutition, Dysphagia, Penetration-aspiration, Pressure, Tongue

PMID: 23782628

 

Supplement

The tongue plays a critical role in swallowing.  When we swallow liquids, we need adequate tongue strength to create a chamber around the bolus that keeps it in the mouth.  We then apply pressure to the bolus with our tongues to generate the forces that squeeze the bolus backwards into the pharynx.

Several research studies show that healthy older people lose strength in their tongues (1, 2).  This age-related decline in tongue pressure generation capacity is clearly seen on tasks requiring people to squeeze a pressure bulb between the tongue and the hard palate with maximum effort, but is not seen in swallowing tasks, for which people typically use only a portion of their tongue pressure reserve (3). Nevertheless, studies suggest that reduced tongue strength is a factor associated with aspiration of material into the airway during swallowing (4).  The association between reduced tongue strength and swallowing impairment (dysphagia) points to the possibility that treatment targeting increased tongue strength may also improve swallowing function.

Initial proof of concept for tongue strengthening comes from a study performed in healthy seniors (5).  An 8-week course of exercise was used, in which participants practiced squeezing a pressure bulb attached to a hand-held pressure meter, known as the Iowa Oral Performance Instrument.  Borrowing exercise design principles from sports medicine, Robbins and colleagues measure maximum strength and then establish targets in the 80% of maximum strength range for repeated practice.  Practice involves 60 task repetitions per day on alternating days of the week, with new targets set each week across the 8 weeks of treatment.  Following this protocol, Robbins et al. clearly established that tongue strength increases in healthy seniors. They then went on to use this protocol protocol in people with dysphagia following a stroke.  These individuals had reduced tongue strength at the beginning of therapy, and showed significant improvements in tongue strength over 8 weeks.  Measurements of their swallowing function showed improvements in aspiration after treatment, but there was no clear pattern of improvement seen in the presence of residues in the oropharynx after the swallow.

The apparent resistance of post-swallow residues to improving in association with greater tongue strength motivated us to explore variations in tongue pressure resistance training tasks, which might lead to improved outcomes.  We initially hypothesized that a protocol, which emphasized the ability to generate precise pressure targets selected across the range of 25-85% of a person’s maximum pressure capacity range might have better swallowing outcomes. This idea was motivated by the idea that different consistencies of food and liquid require a person to generate different tongue pressures to match the flow-resistance properties of the bolus. However, a preliminary case series in 6 patients with dysphagia following brain injury who completed this protocol (which we called Tongue Pressure Strength and Accuracy Training or TPSAT) showed similar results to the Robbins protocol, namely improvement in aspiration without a corresponding improvement in post-swallow residues (6).

In considering how tongue pressure training might be modified to achieve better outcomes, including improved residues, we returned to examining patterns of tongue pressure generation in healthy swallowing. We noted that healthy people show a very interesting variation in the way they release tongue pressures from the palate.  That is, when swallowing thin liquids, healthy people appear to control the flow of thin liquids into the pharynx by releasing tongue pressure more gradually. This prompted us to incorporate a focus on gradual tongue release pressure into our tongue pressure training protocol.

The randomized clinical trial described in this paper compares treatment outcomes of the TPSAT protocol to a novel protocol known as Tongue Pressure Profile Training (TPPT) in people with dysphagia post stroke.  We are specifically enrolling adults who show pre-swallow aspiration of thin liquids, or post-swallow residue in the valleculae with nectar-thick liquids.  Participants receive 24 sessions of therapy scheduled twice or three times a week and the Iowa Oral Performance instrument is used to provide biofeedback. Aspiration and residue outcomes are measured based on blinded review of a standard videofluoroscopy protocol, including 3 swallows each of thin and nectar-thick barium. Aspiration outcomes are measured using the 8-point Penetration-Aspiration Scale (7).  Vallecular residue is measured using the Normalized Residue Ratio Scale or NRRS-V (8), which is a pixel-based measure of residue taken in ImageJ software.  The measure calculates a ratio of the amount of residue to the area of the valleculae.  This ratio is then adjusted for differences in participant height using the length of the cervical spine. Each participant’s functional swallow status is determined based on the worst scores seen across the protocol of 3 thin and 3 nectar-thick swallows.

The approach to analysis in this study will involve both groupwise results and individual case results, using single subject design methods.  Two cases are selected to illustrate this purpose.

Example 1 is a patient who was assigned to the tongue pressure strength and accuracy protocol.  Pre-treatment, they had aspiration at a level 7 on the penetration-aspiration scale with thin liquids, and significant vallecular residue with nectar-thick liquids.  The valleculae were measured to be 95% full, corresponding to an NRRS-V score of 0.28 after adjustment for height.

Figure 1 illustrates the participant’s anterior tongue strength data across treatment.  We use a run chart to document change, by calculating the mean and standard deviations of the first three treatment sessions and using these values to establish a threshold for improvement, equivalent to a moderate effect size or Cohen’s d of 0.6 (9). When at least 3 sessions show values above this threshold, we conclude that the patient has achieved a clinically meaningful improvement in tongue strength.  For this participant, anterior tongue strength had more than doubled by the end of the treatment protocol and was approaching normal values at 40 kPa (1).

Figure 1

Figure 1. Measures of mean anterior maximum isometric tongue-palate pressure across 24 sessions of treatment.  A baseline measure is calculated using the values recorded in the first three treatment sessions and used to establish a confidence interval equivalent to a Cohen’s d effect size of 0.6 around that baseline mean, as shown in the shaded rectangle.  When subsequent session values exceed the upper confidence interval boundary for three successive measures, this is considered to reflect an improvement in anterior tongue strength.

The patient also showed a doubling of posterior tongue strength (Figure 2) but these values still fell substantially below the normal target of 40 kPa. In terms of their videofluoroscopic outcomes, this particular patient showed improvement in their penetration aspiration scale scores from 7 to 3.  However, residue remained unchanged on nectar-thick liquids.

Figure 2 copyFigure 2. Measures of mean posterior maximum isometric tongue-palate pressure across 24 sessions of treatment.  A baseline measure is calculated using the values recorded in the first three treatment sessions and used to establish a confidence interval equivalent to a Cohen’s d effect size of 0.6 around that baseline mean, as shown in the shaded rectangle.  When subsequent session values exceed the upper confidence interval boundary for three successive measures, this is considered to reflect an improvement in posterior tongue strength.

Example 2 is a participant who was assigned to the Tongue Pressure Profile Training protocol.  This individual began with penetration-aspiration scale scores in the normal range at a 2.  However, they had significant vallecular residue for nectar-thick liquids.  The valleculae were measured to be completely full, which translated to an NRRS-V score of 0.22 after adjustment for height.

Figure 3 copyFigure 3. Measures of mean posterior maximum isometric tongue-palate pressure across 24 sessions of treatment. In the TPPT protocol, this measure is taken during performance of the task emphasizing gradual release of pressure. A baseline measure is calculated using the values recorded in the first three treatment sessions and used to establish a confidence interval equivalent to a Cohen’s d effect size of 0.6 around that baseline mean, as shown in the shaded rectangle.  When subsequent session values exceed the upper confidence interval boundary for three successive measures, this is considered to reflect an improvement in posterior tongue strength.

 

Figure 3 illustrates posterior isometric pressures for this participant over the course of treatment.  Almost a doubling of strength was achieved. Interestingly, regular effort saliva swallow pressures remained unchanged in this participant across treatment (Figure 4).  Remember, however, that it is typical to use less than 50% of maximum tongue pressure capacity during swallowing (1).

Figure 4 copyFigure 4. Measures of mean tongue-palate pressure during regular effort saliva swallows across 24 sessions of treatment. In the TPPT protocol, this measure is taken during performance of the task emphasizing gradual release of pressure. A baseline measure is calculated using the values recorded in the first three treatment sessions and used to establish a confidence interval equivalent to a Cohen’s d effect size of 0.6 around that baseline mean, as shown in the shaded rectangle.  When subsequent session values exceed the upper confidence interval boundary for three successive measures, this is considered to reflect an improvement in saliva swallowing pressures.

 

Finally, Figure 5 shows data for effortful saliva swallow pressures for example 2.  These showed a significant increase over treatment, almost doubling in amplitude. In reflecting on the different pattern of change seen in these tasks across the course of treatment for this participant, we can note that at baseline, they were using almost all of their posterior tongue strength range during swallows.  Post treatment, they were performing more typically by using about half of their tongue pressure capacity for regular swallows, but had gained a cushion of functional reserve, enabling them to increase their tongue pressures with effort, should the demands of a particular swallowing task require them to do so.

Figure 5 copyFigure 5. Measures of mean tongue-palate pressure during effortful saliva swallows across 24 sessions of treatment. In the TPPT protocol, this measure is taken during performance of the task emphasizing gradual release of pressure. A baseline measure is calculated using the values recorded in the first three treatment sessions and used to establish a confidence interval equivalent to a Cohen’s d effect size of 0.6 around that baseline mean, as shown in the shaded rectangle.  When subsequent session values exceed the upper confidence interval boundary for three successive measures, this is considered to reflect an improvement in effortful saliva swallowing pressures.

In terms of their swallowing outcomes, this participant showed stable penetration-aspiration scale scores in the normal range.  Their post-swallow residue improved from the baseline measure where the valleculae were 100% full to a post treatment measure of 60% full.  While this does not show complete resolution, it is definitely a change in the right direction.

Enrollment is underway, and we hope to be able to draw clear conclusions regarding different patterns of treatment response by 2016.

 

References

1.            Nicosia MA, Hind JA, Roecker EB, Carnes M, Doyle J, Dengel GA, Robbins J 2000 Age effects on the temporal evolution of isometric and swallowing pressure. Journals of Gerontology Series A-Biological Sciences & Medical Sciences 55:M634-640.

2.            Stierwalt JA, Youmans SR 2007 Tongue measures in individuals with normal and impaired swallowing. American  Journal of Speech Language Pathology 16:148-156.

3.            Steele CM 2013 Optimal approaches to measuring tongue-pressure functional reserve.  Journal of Aging Research, Article ID 542909.

4.            Butler SG, Stuart A, Leng X, Wilhelm E, Rees C, Williamson J, Kritchevsky SB 2011 The relationship of aspiration status with tongue and handgrip strength in healthy older adults. Journals of Gerontology Series A – Biological Sciences & Medical Sciences 66:452-458.

5.            Robbins J, Gangnon RE, Theis SM, Kays SA, Hewitt AL, Hind JA 2005 The effects of lingual exercise on swallowing in older adults. Journal of the American Geriatric Society 53:1483-1489.

6.            Steele CM, Bailey GL, Polacco RE, Hori SF, Molfenter SM, Oshalla M, Yeates EM 2013 Outcomes of tongue-pressure strength and accuracy training for dysphagia following acquired brain injury. International Journal of Speech Language Pathology, 15(5)492-502.

7.            Rosenbek JC, Robbins JA, Roecker EB, Coyle JL, Wood JL 1996 A penetration-aspiration scale Dysphagia 11:93-98.

8.            Pearson WG, Jr., Molfenter SM, Smith ZM, Steele CM 2013 Image-based measurement of post-swallow residue: the normalized residue ratio scale Dysphagia 28:167-177.

9.            Kotrlik JW, Williams HA 2003 The incorporation of effect size in informaton technology, learning, and performance research. Information Technology, Learning, and Performance Journal 21:1-7.

 

Acknowledgments:  The authors gratefully acknowledge the Heart and Stroke Foundation for funding this project. Additional funding support was received through a Canadian Institutes of Health Research New Investigator grant to the author and from the Toronto Rehabilitation Institute University Health Network, which receives funding under the Provincial Rehabilitation Research Program from the Ministry of Health and Long-term Care in Ontario. The views expressed here do not necessarily reflect those of the ministry.

 

Contact:Figure 6
Catriona M. Steele, Ph.D., S-LP(C), CCC-SLP, BCS-S

Senior Scientist and Professor
Toronto Rehabilitation Institute – University Health Network
550 University Avenue, 12th floor,
Toronto, ON, Canada, M5G 2A2
catriona.steele@uhn.ca
www.steeleswallowinglab.ca

 

Multiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier SchönmannMultiselect Ultimate Query Plugin by InoPlugs Web Design Vienna | Webdesign Wien and Juwelier Schönmann