Dysphagia 9:83-87 (1994)
Dysphagia
Temperature Acceleration in Cold Oral Stimulation
Marilyn Selinger, PhD,~ Thomas E. Prescott, PhD,t and Irwin Hoffman, PhD 2
Denver VAMC/126, Denver, Colorado, and Greenwood Village, Colorado, USA
Abstract. This study measured the temperature acceler-
ation of a cold probe as it contacts human tissue. Both the
effects of touching a cold probe to the oral cavity and
maintaining contact of the cold probe with the oral cavity
were investigated. The results indicated a rapid warming
pattern. This warming is effected first by temperature
changes resulting from the probe being moved from ice
into room temperature and second by the contact to oral
mucosa. In fact, in some cases, the probe had reached
minimal cold sensation levels by the time it reached the
oral cavity. Results also indicated that 6 sec after the
probe is lifted from the ice, the temperature closely
aproximates temperatures perceived as warm or at least
neutral, but not cold.
Key words: Thermal stimulation - - Oral mucosa - -
Deglutition - - Deglutition disorders.
Thermal stimulation is a technique used with many dys-
phagic patients for swallowing treatment. The sensitivity
and perception of the oral cavity is not completely under-
stood. Both the responsivity of the oral mucosa and the
study of cold thermal application need more quantifica-
tion. One of the first aspects of this research is under-
standing the nature of the stimuli used in cold application
in the oral cavity. In order to begin quantifying the re-
sponse to cold application we looked at acceleration of
temperature in a cold probe (not a mirror) as it contacts
the human oral mucosa.
Stimulation of the oral cavity, in order to examine
the origins of the deglutition reflex, was reported by
Pommerenke [1] in 1928. His findings indicated a large
This investigation was supported in part by Veterans Administration
Merit Review (Research Advisory Group) grant #331.
Address offprint requests to: Marilyn Selinger, Ph.D., Denver VAMC/
126, 1055 Clermont St., Denver, CO 80220, USA
9 Springer-u NewYorkInc. 1994
variation in different individuals. Using 126 subjects, no
single area consistently elicited a swallow. The anterior
faucial pillars, however, were reported to be the most
sensitive to light tactile stimulation with a blunt glass rod;
this sensitivity was present 77% of the time. Pomme-
renke concluded that the anterior pillars "bear a more
important relationship to the swallowing mechanism."
Somewhat more recently, Doty and Bosma [2]
examined electromyographic patterns of deglutition.
They concluded that there were three anatomical regions
comprising a "leading complex" of muscles which fire
concurrently to initiate the act of swallowing. This com-
plex contains suprahyoid suspensory muscles, muscles
involved in the closing of the palatopharyngeal isthmus,
and the muscles surrounding the faucial pillars.
The relationship between temperature stimulation
and deglutition has also been discussed in the literature
[3-9]. Early studies reported the importance of punctate
stimulation and temperature sensitivity in the oral mu-
cosa [10,11]. These findings suggested that there were
warm and cold spots responsible for thermal sensitivity in
the mouth. The results also indicated that the warm re-
ceptors were less dense than those of the cold receptors.
Subsequently, cold was concluded to be a stronger stim-
ulus for the swallowing reflex. Grossman and Hattis [3]
reported that the medial portion of the anterior faucial
pillars exhibited high sensitivity to cold, that sensitivity
was reported to decrease as the stimulation moved later-
ally on the pillars.
The origins of cold thermal stimulation treatment
for swallowing disorders probably come from these early
investigations concerning tactile stimulation and cold re-
ceptors. More recent studies, however, have stressed the
importance of spatial summation over larger areas of
tissue as the origins of temperature responsivity in the
oral cavity. Green [7] reported that even though there
appears to be a greater amount of innervation for the cold
sense, the oral cavity was actually more sensitive to
warm temperatures. Utilizing a series of experiments
84
concerned with stimulation through larger spatial sum-
mation with warm and cool liquids, it was found that
thermal sensitivity was greater to higher temperatures
than cool ones. Therefore, it was concluded that small
areas of stimulation are poor predictors of oral respon-
siveness to stimulation of larger areas regardless of the
amount of innervation for temperature sensitivity. The
more recent data on oral sensitivity and perception raises
interesting issues and opens new avenues for research in
the area of temperature responsivity in the human oral
cavity.
The technique referred to as "thermal stimula-
tion" in reference to swallowing treatment is often rec-
ommended to heighten sensation and therefore improve
the speed at which the pharyngeal swallow is triggered in
dysphagic patients. Logemann (1983, 1986) recom-
mends the training of patients with thermal stimulation.
Thermal stimulation is believed to enhance the sensitivity
of the faucial arches, thereby triggering the pharyngeal
swallow more rapidly. Logemann [12,13] recommends
using cold stimulation with a laryngeal mirror for height-
ened sensitivity.
Lazarra et al. [14] have reported one experiment
using cold thermal stimulation as a means of improving
swallowing in 25 neurologically impaired patients. The
cold stimulation technique described by the authors in-
volves "lightly touching the area with the back of an iced
size 00 laryngeal mirror." Changes in transit time were
measured using videofluoroscopic recording of the swal-
lows. These authors report that the thermal sensitization
improved immediate transit times in 23 of 25 patients.
Measured improvement in transit times were concluded
to be large enough to reduce the risk of aspiration.
Rosenbek et al. [15] followed the recommended
treatment for cold thermal stimulation by icing a "00"
laryngeal mirror and subsequently stroking each faucial
pillar three times. These authors report results of limited
change from thermal stimulation treatment in patients
with multiple strokes.
Selinger, et al. [16] examined the effects of cold
thermal stimulation on a patient exhibiting dysfunctional
swallow with laryngeal penetration. It was concluded
that cold thermal stimulation with a 00 laryngeal mirror
was not efficacious for the patient. Several issues were
raised from this attempt to follow the actual technique
reported by Logemann [12,13]. In 1983, Logemann
states that "To stimulate the reflex, the mirror is held in
ice water for approximately 10 seconds and lightly
touched to the base of the anterior faucial arch . . . . Light
contact is repeated 5 to 10 times." Logemann [13] stated
that "The mirror is placed in ice in order to make it cold,
and it is then put in contact in a stroking motion with the
anterior faucial arch. After five or six contacts on each
side of the oral cavity at the faucial arches . . . asked to
M. Selingeret al.: TemperatureAccelerationin Cold Oral Stimulation
swallow." It was felt that the actual thermal stimulation
procedure as it is described is somewhat imprecise. It was
possible to have missed an important step that would
have made the difference between efficacious and nonef-
ficacious treatment. Naturally, this concern is generaliz-
able to all patients and all clinicians.
As a first step in examining issues related to ther-
mal stimulation, this study was designed to measure tem-
perature acceleration in a cold probe as it contacts human
tissue. The following questions were asked:
1. What is the temperature acceleration of a cold probe
upon contact with the faucial pillars?
2. What is the temperature of the cold probe over time
with repeated contact to the pillars?
Methods
The measurementof temperature accelerationwas accomplishedutiliz-
ing the IBM Personal Science Lab (PSL) temperature probe. This
program allows quantificationof temperaturechange using a computer
probe peripheral. The PSL probe is an active metal device from the
AD950 familysimilarto a transistorand is encasedin an insulatormade
of ceramic. The probe is sensitivefrom -40~ to 105~ The thermal
time constant of the probe is 0.14 sec. The changes at the probe are
measured by samplingtemperature 10 timesper sec. This unit is manu-
factured by IBM EducationalSystems (Atlanta, GA). The heating and
cooling curves follow the laws of exponential models and conse-
quently, ambientchangesin temperatureare quite rapid.
The measurement sequence was designed to approximate the
techniquedescribedby Logemann[12] so that it wouldbe similarto the
technique used in this treatmentfor dysphagia. Temperaturesampling
was carried on for 25 sec for each of 10 trials. This resulted in 250
temperature measurementsbased on the samplingrate of 10 times per
sec. The probewas placedagainstan ice cube in a glass of ice; the probe
was left in the ice for the first five sec. AlthoughLogemann[12] has
recommended "approximatelyten seconds" our probe temperaturere-
mained consistently stable, that is, with no significant temperature
decrement over 5 sec at a temperature just below freezing (31.4-
31.9~ Based on this measurement,it was felt that 5 sec was adequate
for reaching and maintaininga level at freezing. At 5 sec the probe was
lifted from the ice and moved into the experimeutor's mouth; it was
subsequentlyplacedon the right faucialpillar. The mostobviousdiffer-
ence betweenthe techniquereported here and that of Logemannis that
only one pillar was stimulatedand the proberemainedstationaryon that
pillar rather than being used to stroke the pillar. Specific temperatures
from the 250 temperaturemeasureswere sampledat 0 sec, 5 sec, 8 sec,
15 sec, and 25 sec for use in the analysis of variance (ANOVA)
procedure. Measurementsat 0 and 5 sec were of the probe in the ice, at
8 sec, the probe had been out of the ice for 3 sec and had been placed
againstthe faucialpillar, and the next two temperaturesamples(15 and
25 sec) representthe temperatureof the probe in the mouth.
Results
In order to determine if there were any systematic tem-
perature differences across the sampling period, a one-
way ANOVA for repeated measures [17] was performed.
M. Selinger et al.: TemperatureAccelerationin Cold Oral Stimulation
Table 1. A N O V A for repeated measures
Source SS DF
Blocks/Subjects 135.345 9
Temperature 32197.638 4 8049.409
Error 133.816 36
Total 32466.800 49
SS = sum of squares; DF = degrees of freedom; MS = mean square.
40.00
31.96
I 23.92
15,88
7.8<-
-0.200
5.05 8.02 15.0
SECONDS
The analysis yielded a significant [F(4,36)= 3.717,
p < 0.001) effect. Table 1 summarizes the ANOVA for
the sampled temperatures.
To more precisely determine where the signifi-
cant effects occurred, a post hoc Scheffe test [18] was
performed on the temperature means. This analysis indi-
cated that the significant effect occurred between the
probe in the ice (X = 0 ~ F) and when it was placed in the
mouth (X = 32.15 ~ F), and between the initial mouth
placement and the subsequent measure at 15 sec
(X = 85.55 ~ F). This measure represents 10 sec out of
the ice. The temperature of a closed mouth approximates
98-99~
In order to further define the relationship between
temperature acceleration and time, a first-degree (linear)
equation was used [19]. Results from this analysis indi-
cated a significant (F(1,36) = 76.37, p < 0.01) linear
trend. Subsequent analysis for a second degree (qua-
dratic) equation indicated no significant (0.05) trend. The
relationship between temperature acceleration and time
was then defined as linear. This finding is consistent with
the results of the Scheffe test which indicated that the
significant effects occurred between the probe in the ice
and the probe into the mouth. In other words, the two
temperature measurements at each end of the data were
not different from the measurement that preceded it.
Subsequently, the "line of best fit" [20] was com-
puted (Fig. 1). The data indicated the slope of the regres-
85
MS F P
2165.498 < 0.001
3.717
Fig. 1. The graph represents the "line of best fit"
for Y given X at 5 sec and at 15 sec. Utilizing the
line, it can be seen that an increase in the time that
the probe was withdrawn from the ice resulted in a
linear increase (slope = 2.59) in the temperature of
24,94 the probe. Correlation coefficient of r = 0.82 was
obtained.
sion line to be 2.59108. Utilizing that line it can be seen
that an increase in the time that the probe was withdrawn
from the ice resulted in a linear increase in the tempera-
ture of the probe. In other words, if one wishes to predict
the temperature of the probe it can be done by knowing
the length of time of removal of the probe from the ice
(r = 0.82). The data indicated that removal of the cold
probe resulted in a neutral temperature sensation of the
probe in 6-7 sec. Thermal neutrality is defined as temper-
atures at which subjects report only mild or no sensation
of hot or cold in the oral cavity [6] (Green, 1985).
Discussion
These results indicate that the temperature changes did
not happen by chance. Careful examination of the tem-
perature responses to 2 of the 10 temperature acceleration
measures indicates that between 1.15 and 2.8 sec after
the probe is lifted from the ice its temperature has accel-
erated to one of only moderate sensation of cold (20~
[8]. In addition, between 6 and 7 sec after the probe is
lifted from the ice the temperature has accelerated to one
of warmth sensation (30~ [8].
The initial question addressed in this investiga-
tion was, What is the temperature acceleration of a cold
probe upon contact with the faucial pillars? The results
indicated a rapid warming pattern. This warming is ef-
86
fected first by temperature changes resulting from the
probe being moved from ice into room temperature and
second by the contact to oral mucosa. In fact, in some
cases, the probe had reached minimal cold sensation lev-
els by the time it reached the oral cavity.
The second question addressed in this investiga-
tion was, What is the temperature of the cold probe over
time with repeated contact to the pillars? The results
indicated that 6 sec after the probe is lifted from the ice,
the temperature closely aproximates temperatures per-
ceived as mildly warm or at least neutral, but not cold.
At this time, actual quantification of the differ-
ences between the PSL probe and a 00 laryngeal mirror
have not been made. It is possible, however, to make
some comparative statements regarding the nature of the
two instruments and thus hypothesize on the ability of
each to retain cold temperatures.
First, the probe is lower in thermal mass than the
mirror. Lower thermal mass will result in a more rapid
temperature acceleration because it does not store as
much energy as the larger mirror. The probe is insulated
by ceramic material which acts like metal but does not
store as much energy. However, the probe handle has
additional insulation to prevent additional temperature
acceleration as a result of body heat.
Alternatively, the mirror may be more affected by
body heat when it is handled and by room temperature
because it is not insulated. In addition, the mirror has two
temperature time constants involved in temperature
changes: metal and glass. The temperature response of
both the metal and glass may need to be taken into ac-
count separately when quantifying the thermal properties
of the mirror.
Neither the mirror nor the probe can be frozen and
kept frozen while they are in ambient temperature atmo-
spheres. Taking all of these differences into account the
PSL probe may be more sensitive to temperature acceler-
ation, implying that the subsequent measured tempera-
ture changes may be more rapid.
The differences in the tools used to apply cold
stimulation is important in that it points out the lack of
control over the general parameters involved in thermal
stimulation. However, the application of the PSL probe
is a first step in controlling and quantifying the treatment
techniques that are used with dysphagic patients. Various
tools are used for this treatment (laryngeal mirrors,
Q-Tips) resulting in probable variance in temperature
acceleration curves and varying results. All of the vary-
ing techniques and tools need to be known if one wishes
to attribute changes, or lack thereof, to any treatment.
In general, these findings indicated that thermal
stimulation of the faucial pillars with a cold probe may
not be what it is believed to be. By interfacing the time
M. Selinger et al.: Temperature Acceleration in Cold Oral Stimulation
and temperature measurements from this study with a
thermal stimulation procedure, it can be seen that by the
time the probe comes in contact with the first faucial
pillar it has reached a temperature of minimally cold or
neutral in relation to the temperature of the mouth. Re-
peated contact or stroking of the faucial pillars beyond a
maximum of 4 sec amounted to probable light tactile
stimulation and possibly minimal warm stimulation but
did not necessarily result in cold thermal stimulation.
Results from studies reporting on the efficacy of
thermal stimulation [14-16] have provided conflicting
conclusions. It is possible that the reported differences in
the success of thermal stimulation result from the impre-
cise methodology of the thermal stimulation procedure.
Multiple stroking of each faucial pillar, when done care-
fully, probably results in increased amounts of time in the
mouth, with actual cold stimulation being present in only
the first few sec. Changes in the temperature of the probe
over time and in relation to the oral cavity temperature
probably cause the thermal effects to average out to a
neutral point. Thus, the original idea of cold stimulation
for heightening oral sensation may not be an actuality in
technique. The only potentially consistent stimulation to
the oral cavity is tactile and this too is not controlled for
in terms of time or amount.
References
1. Pommerenke WT: A study of sensory areas eliciting the swal-
lowing reflex. Am J Physio184(1): 36-41, 1928
2. Doty RW, Bosma JF: An electromyographic analysis of reflex
deglutition. J Neurophysiology 19:44-60, 1956
3. Grossman RC, Hattis BF: Oral mucosal sensory innervation and
sensory experience: a review. In: Bosma JF (ed.): Oral Sensa-
tion and Perception. Springfield, IL: Thomas, 1967, pp 5-62
4. Miller AJ: Deglutition. PhysiolRev 62:129-184, 1982
5. Green RG: Thermal perception on lingual and labial skin. Per-
cept Psychophys 36 (3):209-220, 1984
6. Green BG: Heat pain thresholds in the oral-facial region. Per-
cept Psychophys 38 (2): 110-114, 1985
7. Green BG: Oral perception of the temperature of liquids. Per-
cept Psychophys 39 (1):19-24, 1986
8. Green BG, Gelhard B: Perception of temperature on oral and
facial skin. Somatosens Res 4(3): 191-200, 1987
9. Kennedy JG, Kent RD: Physiological substrates of normal
deglutition. Dysphagia 3:24-37, 1988
10. Zotterman Y: Specific action potentials in the lingual nerve of
cat. ScandArch Physiol 75:105-119, 1936
11. Yamada M, Maruhashi J, Miyake N: The distribution of sensory
spots on the oral mucous membrane. Jpn J Physiol 2: 328-332,
1952
12. Logemann J: Evaluation and Treatment of Swallowing Disor-
ders. San Diego: College Hill Press, 1983
13. Logemann JA: Treatment for aspiration related to dysphagia: an
overview. Dysphagia 1:34-38, 1986
14. de Lama Lazzara G, Lazarus C, Logemann JA: Impact of ther-
M. Selinger et al.: Temperature Acceleration in Cold Oral Stimulation 87

mal stimulation on the triggering of the swallowing reflex. Dys- 17. Madigan S, Lawrence V: Anova H. Northridge: Human Systems
phagia 1:73-77, 1986 Dynamics, 1982
15. Rosenbek JC, Robbins J, Fishback B, Levine RL: Effects of 18. Roscoe JT: Fundamental Research Statistics for the Behavioral
thermal application on dysphagia after stroke. J Speech Hearing Sciences. New York: Holt, Rinehart, & Winston, 1975
Res 34:1257-1268, 1991 19. Winer BJ: Statistical Principles in Experimental Design. New
16. Selinger M, Prescott TE, McKinley R: The efficacy of thermal York: McGraw-Hill, 1971
stimulation: a case study. Rocky Mountain J Commun Dis 6:21- 20. Edwards AL: Statistical Methods of the Behavioral Sciences.
23, 1990 New York: Holt, Rinehart, & Winston, 1963