Respiratory Physiology & Neurobiology 150 (2006) 82–86

Short communication

Laryngeal water receptors are insensitive to body

temperature in neonatal piglets
L. Xia a , J.C. Leiter a,b,∗ , D. Bartlett Jr. a
a Department of Physiology, Dartmouth Medical School, Lebanon, NH 03756, USA
b Department of Medicine, Dartmouth Medical School, Lebanon, NH 03756, USA

Accepted 14 May 2005

Abstract

Heat stress and the laryngeal chemoreflex (LCR) have both been implicated as possible contributors to the sudden infant
death syndrome (SIDS). We recently reported that moderate hyperthermia, induced in decerebrate piglets by external heating,
substantially prolonged the LCR elicited by injecting 0.1 ml of water into the larynx through a prepositioned transnasal catheter.
To examine the question of whether hyperthermia influences the responses of laryngeal water receptors, we recorded single fiber
action potentials in fine strands of the superior laryngeal nerve (SLN) in decerebrate piglets while the larynx was filled with
water or isotonic saline. Water receptors, identified by their much brisker response to water than to saline, were studied with
body temperature at 37.9 ± 0.2 ◦ C, after warming the animal to 40.6 ± 0.2 ◦ C and after cooling back to 37.7 ± 0.3 ◦ C. The results
show no effect of body temperature change, in this range, on the responses of the laryngeal water receptors and thus suggest that
the potentiation of the LCR by hyperthermia is mediated by a central action.
© 2005 Elsevier B.V. All rights reserved.

Keywords: Hyperthermia; Laryngeal chemoreflex (LCR); Laryngeal water receptors; Piglet; Sudden infant death syndrome (SIDS); Superior
laryngeal nerve (SLN)

1. Introduction Velde et al., 2003). This array of responses, collectively

In infant mammals of many species, introduction of
water or other fluids into the larynx elicits a complex set
of protective respiratory responses, usually including
apnea and swallowing and sometimes accompanied by
cough (Boggs and Bartlett, 1982; Thach, 2001; van der
∗ Corresponding author. Tel.: +1 603 650 8533;
fax: +1 603 650 6130.
E-mail address: james.c.leiter@dartmouth.edu (J.C. Leiter).
known as the laryngeal chemoreflex (LCR), depends
on laryngeal mucosal chemoreceptors with afferents in
the superior laryngeal nerves (SLNs) (Harding et al.,
1978; Boggs and Bartlett, 1982). It is most prominent
in neonates and is modified with maturation to include
less apnea and more coughing (Boggs and Bartlett,
1982; Thach, 2001). As the apnea can be prolonged
in young animals, the LCR has been suspected to play
a role in the sudden infant death syndrome (SIDS)
(Downing and Lee, 1975; van der Velde et al., 2003).

1569-9048/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.resp.2005.05.021
 L. Xia et al. / Respiratory Physiology & Neurobiology 150 (2006) 82–86
Because thermal stress has been implicated as a risk
factor for SIDS (Stanton, 1984; Harper et al., 2000;
Gunteroth and Spiers, 2001), we recently conducted
a study of the influence of elevated body temperature
on the LCR in decerebrate piglets, aged 3–16 days.
The reflex was elicited by injecting 0.1 ml of distilled
water into the larynx through a prepositioned transnasal
catheter. The resulting respiratory inhibition, however
measured, was substantially more prolonged when the
animal’s core temperature was raised to 39–40 ◦ C com-
pared with trials at 37–38 ◦ C (Curran et al., 2005). This
finding is of interest with respect to SIDS and also raises
questions about its mechanism: the influence of temper-
ature could occur at any point along the reflex pathway,
from the receptors in the laryngeal epithelium through
the synaptic connections in the brainstem that deter-
mine the integrated respiratory response. To explore
the first possibility, we have examined the influence of
body temperature on the behavior of water-responsive
laryngeal receptors by recording their afferent activi-
ties in the SLNs of piglets under conditions similar to
those in our study of the LCR.
2. Methods
2.1. Preparation
Experiments were performed on a total of 15 piglets
aged 3–19 days with body weights ranging from 2.4 to
5.4 kg. Piglets were housed with the sow in our animal
care facility before they were studied. The Institutional
Animal Care and Use Committee of Dartmouth College
approved all protocols in these studies.
Each animal was anesthetized with 2–4% halothane
in O2 . A rectal temperature probe was inserted, and
body temperature was maintained at 37–38.5 ◦ C with a
heating pad. Femoral arterial and venous catheters were
inserted for measurement of blood pressure and admin-
istration of drugs, respectively. A tracheal cannula was
placed low in the neck, and artificial ventilation was
instituted to maintain end-tidal CO2 , monitored with
an infrared analyzer, at approximately 5%. A second,
rostrally directed tracheal cannula was placed 1–2 cm
below the cricoid cartilage for introduction of test solu-
tions into the larynx. The carotid sinus regions were
exposed bilaterally, the internal and external carotid
arteries were ligated to facilitate decerebration, and the
83
vagus nerves were sectioned. The animal was placed
prone, with the head positioned in a stereotaxic appa-
ratus. The skull was opened, the brain was transected
at the level of the superior colliculi, and all brain tis-
sue rostral to the transection was removed. Following
decerebration, halothane anesthesia was discontinued,
and the animal was returned to the supine position
and paralyzed using pancuronium bromide (1 mg/kg,
i.v.). Supplemental doses of pancuronium were given
as required, usually at a rate of 0.5 mg/kg/h.
The SLNs were sectioned bilaterally as far from the
larynx as possible. Laryngeal receptor activities were
recorded from fine strands of the peripheral cut end
of an SLN, dissected with fine scissors and watch-
maker’s forceps and placed on a bipolar wire electrode
under mineral oil (Bartlett et al., 1976; Bartlett and
Knuth, 1992). Receptor activities were amplified, fil-
tered (10–500 Hz), sampled at 20 kHz and recorded
digitally on a laboratory computer.
2.2. Protocol
We searched for activities of water receptors by fill-
ing the larynx with 3–5 ml of distilled water while
recording from a fine strand of the SLN. When this
elicited single receptor activity, we then filled the lar-
ynx with 3–5 ml of isotonic saline. Receptors that
responded to water with an initial discharge frequency
at least five times that elicited by saline were judged
to be water receptors and were examined further, as
described below. Receptors with comparable responses
to water and saline were discarded as they did not
exhibit chemical specificity for water and presum-
ably were responding to some other aspect – perhaps
mechanical or thermal – of the stimulus. The focus on
water receptors was based on evidence that receptors of
this kind are primarily responsible for the LCR (Boggs
and Bartlett, 1982).
Once a water receptor was identified, we recorded
its response to three to five trials each with water and
saline, separated by at least 5 min, with the animal’s
body temperature at 36.6–38.8 ◦ C. We then warmed
the piglet by external heating to a core temperature of
39.2–41.4 ◦ C and repeated the trials. The animal was
then cooled back to 36.2–38.5 ◦ C, and a final set of
trials was run.
Earlier experience (Boggs and Bartlett, 1982) and
preliminary studies with three receptors in the present
84 L. Xia et al. / Respiratory Physiology & Neurobiology 150 (2006) 82–86
series indicated that the receptor responses are not
appreciably influenced by the temperature of the water
or saline (20–38 ◦ C) placed in the larynx. We there-
fore used room temperature fluids for the trials. In five
piglets, we conducted the protocol described above
while recording from the whole SLN, at a sampling
rate of 1 kHz, rather than from a single fiber. The pur-
pose of these whole nerve recordings was to include
activity from fragile fibers such as unmyelinated ones,
which might be destroyed in the course of single fiber
dissection.
2.3. Analysis
The recorded data were played back off-line for
computer-assisted analysis. Single fiber action poten-
tials were identified by wave form and amplitude. The
frequency response of each receptor was determined
for each trial by measuring the number of action poten-
tials during a 3-s period in the unstimulated condition
(usually zero), and for 0.5 s intervals for 2 s and for
1 s intervals at 3, 5, 8, 10, and 15 s after the introduc-
tion of water or saline into the larynx. The number of
impulses per second in each bin was averaged for each
body temperature (control, hyperthermia or recovery).
The whole nerve responses were analyzed by a simi-
lar approach, but using moving time averages of nerve
activity (time constant = 100 ms) rather than receptor
discharge frequencies. To combine the responses of
the different receptors, weighting them equally, we
scaled each receptor’s frequency data to its maxi-
mum value as 100% and averaged the results over all
receptors.
Statistical analysis to determine whether body tem-
perature influenced receptor behavior was done by
analysis of variance for repeated measures, comparing
control, hyperthermic, and recovery responses. Spe-
cific preplanned comparisons were made after P-values
were adjusted by the Bonferroni method. A P-value of
less than 0.05 was considered statistically significant.
3. Results
Recording single unit afferent activities from piglet
SLNs was much more difficult than expected based on
our previous experience with single fiber nerve record-
ings (Bartlett et al., 1976; Bartlett and St. John, 1979;
Boggs and Bartlett, 1982; Bartlett and Knuth, 1992).
Even nerves that had robust whole nerve responses to
intralaryngeal water were often silent after dissection
into fine strands. We ultimately succeeded in recording
from 13 water receptors in 10 animals, however, and
the responses of the receptors to the protocol were very
consistent.
We identified two types of water-sensitive recep-
tor responses as described previously (Harding et al.,
1978). Injection of water caused a sudden increase
in receptor activity, which underwent adaptation over
the course of several seconds in 10/13 receptors, and
in the remaining three receptors, there was slightly
delayed activation after injecting water, and little adap-
tation in two of the three receptors. The average peak
activity occurred ∼9 s after water stimulation in two
of these receptors, and the activity had not fallen to
50% of the peak value even 20 s after the stimulus was
delivered. In the remaining receptor, the peak activ-
ity occurred ∼2 s after the water stimulus and adapted
relatively slowly so that ∼6 s after the water stimu-
lus, activity had fallen 50% from the peak value. The
injection of saline caused a similar pattern of recep-
tor responses, but the level of activity and the duration
of the response were markedly attenuated compared
to the response to water. There was no discernable
effect of body temperature on the responses of the
receptors.
The initial difficulty recording single recep-
tor activity prompted us to analyze the inte-
grated responses of the whole SLN after water or
saline injection under normothermic conditions (rec-
tal temperature = 38.3 ± 0.l ◦ C), hyperthermic condi-
tions (40.6 ± 0.2 ◦ C), and in a final normothermic state
(38.3 ± 0.3 ◦ C). As can be seen in the top panel of
Fig. 1, water injected into the larynx elicited a prompt
increase in integrated SLN activity. The response to
saline was significantly less (P < 0.05). There was no
effect of elevating body temperature on the response of
the whole nerve to either water or saline.
The lower panel of Fig. 1 shows the average
responses of the 13 receptors during water or saline
stimulation under baseline, hyperthermic and recov-
ery conditions. The response to water was more robust
and longer lasting than the response to saline. How-
ever, body temperature had no effect on the timing or
intensity of the response to either water or saline stimu-
lation. The same pattern was found when the responses
 L. Xia et al. / Respiratory Physiology & Neurobiology 150 (2006) 82–86
Fig. 1. In the top panel, the average integrated SLN activity is shown
as a function of time from 3 s before to 10 s after filling the larynx with
water (above) or saline (below) during each of the three temperature
conditions. Similarly, in the lower panel, the average frequency of
individual water receptor impulses is shown as a function of time
from 3 s before through the 15th second after filling the larynx. There
was no effect of body temperature on the responses to water or saline.
The receptor response to water was significantly more vigorous than
the response to saline as described previously (Harding et al., 1978).
were considered as percent of the maximum frequency.
Thus, the data provide no support for the hypothesis that
hyperthermia would accentuate or prolong the recep-
tors’ responses to intralaryngeal water.
4. Discussion
The two different receptor types, both responding to
intralaryngeal water, but with different temporal pat-
terns, have been described previously by Harding et
al. (1978) and by Anderson et al. (1990). The findings
in this study indicate that over a rather narrow range
(36.3–41.4 ◦ C), body temperature does not influence
the discharge of either type of laryngeal water receptor
in neonatal piglets. This temperature range is of interest
85
because of our finding that the duration and intensity
of the LCR are markedly increased by warming decer-
ebrate piglets to approximately this extent (Curran et
al., 2005).
The responses of laryngeal water receptors do not
appear to have been studied at different temperatures
previously. Schoener and Frankel (1972), studying
slowly adapting pulmonary stretch receptors in rats,
found a progressive increase in discharge frequency
with body temperature in 13 of 30 units tested, but
noted that most of these units showed very little modu-
lation by body temperature in the 37–39 ◦ C range. The
discharge frequency of pulmonary stretch receptors
in reptiles, examined over a much wider temperature
range, has been reported to exhibit a Q10 of approx-
imately 1.5 (Furilla and Bartlett, 1988). This degree
of sensitivity to temperature would only result in a
change of receptor discharge frequency of about 8.5%
under the conditions of our study and could have been
missed. However, an 8.5% change in receptor activity
alone is unlikely to have caused the extent of prolon-
gation of the LCR we observed in decerebrated ani-
mals after body temperature was elevated; the average
change in LCR duration exceeded 700% (Curran et al.,
2005).
Our results indicate that the striking influence of
body temperature on the LCR in piglets (Curran et al.,
2005) is not attributable to effects of body temperature
on laryngeal water receptor behavior, but is probably
due to a central effect of temperature on the brain
stem. This interpretation is consistent with a report by
Haraguchi et al. (1983), who stimulated the SLN elec-
trically and demonstrated that the latency and thresh-
old of thyroarytenoid muscle activation decreased as
body temperature was changed from 34 to 41 ◦ C in
anesthetized puppies and adult dogs. The reduction
in threshold as body temperature increased was much
greater in neonatal animals compared to adult dogs.
These findings, and our own, strongly implicate a cen-
tral action of elevated body temperature to augment the
duration and intensity of the LCR.
Acknowledgement
This study was supported by grant HD 042707
from the National Institute of Child Health and Human
Development.
86 L. Xia et al. / Respiratory Physiology & Neurobiology 150 (2006) 82–86
References
Anderson, J.W., Sant’Ambrosio, F.B., Mathew, O.P.,
Sant’Ambrogio, G., 1990. Water-responsive laryngeal recep-
tors in the dog are not specialized endings. Respir. Physiol. 79,
33–43.
Bartlett Jr., D., Jeffery, P., Sant’Ambrogio, G., Wise, J.C.M., 1976.
Location of stretch receptors in the trachea and bronchi of the
dog. J. Physiol. (Lond.) 258, 409–420.
Bartlett Jr., D., Knuth, S.K., 1992. Responses of laryngeal receptors
to intralaryngeal CO2 in the cat. J. Physiol. 457, 187–193.
Bartlett Jr., D., St. John, W.M., 1979. Adaptation of pulmonary
stretch receptors in different mammalian species. Respir. Physiol.
37, 303–312.
Boggs, D.F., Bartlett Jr., D., 1982. Chemical specificity of a laryngeal
apneic reflex in puppies. J. Appl. Physiol. 53, 455–462.
Curran, A.K., Xia, L., Leiter, J.C., Bartlett Jr., D., 2005. Elevated
body temperature enhances the laryngeal chemoreflex in decer-
ebrate piglets. J. Appl. Physiol. 98, 780–786.
Downing, S.E., Lee, J.C., 1975. Laryngeal chemosensitivity: a pos-
sible mechanism of sudden infant death. Pediatrics 55, 640–649.
Furilla, R.A., Bartlett Jr., D., 1988. Intrapulmonary receptors in the
garter snake (Thamnophis sirtalis). Respir. Physiol. 74, 311–
322.
Gunteroth, W.G., Spiers, P.S., 2001. Thermal stress in sudden infant
death: is there an ambiguity with the rebreathing hypothesis?
Pediatrics 107, 693–698.
Haraguchi, S., Fung, R.Q., Sasaki, R., 1983. Effect of hyperthermia
on the laryngeal closure reflex. Implications in the sudden infant
death syndrome. Ann. Otol. Rhinol. Laryngol. 92, 24–28.
Harding, R., Johnson, P., McClelland, M.E., 1978. Liquid-sensitive
laryngeal receptors in the developing sheep, cat and monkey. J.
Physiol. (Lond.) 277, 409–422.
Harper, R.M., Kinney, H.C., Flemming, P.J., Thach, B.T., 2000.
Sleep influences on homeostatic functions: implications for sud-
den infant death syndrome. Respir. Physiol. 119, 123–132.
Schoener, E.P., Frankel, H.M., 1972. Effect of hyperthermia and
PaCO2 on the slowly adapting pulmonary stretch receptor. Am.
J. Physiol. 222, 68–72.
Stanton, A.N., 1984. Overheating and cot death. Lancet 2,
1199–1201.
Thach, B.T., 2001. Maturation and transformation of reflexes that
protect the laryngeal airway from liquid aspiration from fetal to
adult life. Am. J. Med. 111 (8a), 69S–77S.
van der Velde, L., Curran, A., Filiano, J.J., Darnall, R.A., Bartlett Jr.,
D., Leiter, J.C., 2003. Prolongation of the laryngeal chemoreflex
prolongation after inhibition of the rostroventral medulla: a role
in SIDS? J. Appl. Physiol. 94, 1883–1895.