British Journal of Anaesthesia 1996; 76: 116–121
Effects of 3 MAC of halothane, enflurane and isoflurane on cilia beat
frequency of human nasal epithelium in vitro
J. H. RAPHAEL, D. A. SELWYN, S. D. MOTTRAM, J. A. LANGTON AND C. O’CALLAGHAN
Summary
We have measured the effects of three times the
minimum alveolar concentration (MAC) of halothane, enflurane and isoflurane on cilia beat frequency of human nasal epithelial brushings from 18
healthy adult patients. Using the transmitted light
technique and paired perfusion chambers, the cilia
were exposed to 2.25 % halothane, 5 % enflurane or
3.6 % isoflurane in air, or air alone, in a controlled
and blinded manner. Over a 4-h observation period,
cilia beat frequency of the samples exposed to
inhalation anaesthetic agents demonstrated a significant reduction in frequency compared with
controls exposed to air alone. Mean cilia beat
frequency for the samples exposed to halothane
was 9.3 (SEM 1.3) compared with its controls of
11.4 (1.0); for the samples exposed to enflurane,
10.9 (1.3) compared with its controls of 11.6 (1.2);
and for the samples exposed to isoflurane, 10.8
(1.1) compared with its controls of 11.6 (1.2).
There was a statistically significant difference
between the samples exposed to all three volatile
agents and their associated controls (halothane,
P : 0.01;
enflurane,
P : 0.03;
isoflurane,
P : 0.01; nested repeated measures analysis of
variance utilizing polynomial contrasts). (Br. J.
Anaesth. 1996; 76: 116–121)
Key words
Anaesthetics volatile, halothane. Anaesthetics volatile, enflurane.
Anaesthetics volatile, isoflurane. Lung, trachea.
Respiratory infections are a major cause of morbidity
after surgery and are particularly common following
upper abdominal surgery and in smokers [1]. Despite
different anaesthetic techniques, the incidence of
postoperative chest infections has remained unchanged over the last 50 yr, ranging from 14 to 21 %
for upper abdominal surgery [1–4]. A recent study
also demonstrated an increase in mean hospital stay
from 7.8 to 10.7 days in those patients developing a
postoperative chest infection [1].
There is a predisposition to development of
postoperative chest infections in atelectatic regions
of the lung where secretions are retained [5]. In
clinical practice most emphasis has been placed on
promoting mucus clearance by encouraging
coughing through improved analgesic regimens;
however, the most important natural defence against
chest infection is the mucociliary transport system.
Mucus transport depends on the volume and
physical properties of the mucus and on the function
of the beating cilia. The depressant effects of
anaesthetic agents on mucus transport are well
established [6–8]; however, the effects of anaesthetic
agents on ciliary function have received little attention. In a pilot study, we have recently documented the depressant effect of halothane on human
respiratory cilia beat frequency [9].
We have developed a system to expose human
ciliated tissue to known concentrations of volatile
anaesthetic agents and measure cilia beat frequency
in a controlled and blinded manner [10]. With this
system we have investigated the effects of 3 MAC
of halothane, enflurane and isoflurane on human
respiratory cilia beat frequency.
Patients and methods
After obtaining Ethics Committee approval and
informed patient consent, we obtained samples of
ciliated epithelium from 18 non-smoking healthy
patients, eight male, mean age 34.3 (range 22–78)
yr, who were not receiving any medications and
had not suffered an upper respiratory tract infection
within the previous 4 weeks. The patients were
unpremedicated.
In a separate study we demonstrated that a single
induction dose of propofol had no effect on cilia beat
frequency measured subsequently in vitro [10]. After
an induction dose of propofol 2–3.3 mg kg91, samples
of ciliated respiratory epithelium were obtained by
passing a bronchoscopy brush over the inferior nasal
turbinates. The brush was agitated in Hanks buffered
salt solution (HBSS) to remove the tissue from the
brush.
To investigate the effects of the volatile anaesthetic
agents on the cilia we designed and built a perfusion
system [10]. This comprised two perfusion chambers
to house the ciliated tissue from one individual, each
consisting of an aluminium block with integral
channels at two sides and with reversibly sealed
coverslips at the top and bottom. Each chamber was
perfused using metallic connecting tubing passing
from a separate delivery bottle of HBSS into the
J. H. RAPHAEL, FRCA, D. A. SELWYN, FRCA, S. D. MOTTRAM, BSC,
J. A. LANGTON, MD, FRCA (University Department of Anaesthesia); C. O’CALLAGHAN, BMEDSCI, MRCP, DM (University
Department of Child Health); Leicester Royal Infirmary,
Leicester LE1 5WW. Accepted for publication: August 25, 1995.
Effects of volatile anaesthetics on cilia beat frequency
Figure 1
117
Perfusion system for the in vitro measurement of cilia beat frequency.
entry port of the chamber and out from its exit port
through plastic connecting tubing into a collecting
beaker. The chambers were perfused with HBSS
under the effect of gravity at 0.5 ml min91. Air
1000 ml min91 was either passed directly into one of
the bottles of HBSS or passed through a Tec 3
vaporizer before passing into the other bottle. This
divergence was achieved in a manner by which the
observer was blinded as both metallic delivery tubes
were wound around one another within a countercurrent water jacket. The outputs from the three Tec
3 vaporizers for halothane, enflurane and isoflurane
with an air flow of 1 litre min91 were calibrated using
an analyser (Capnomac, Datex).
The bottles of HBSS were immersed in a water
bath at 37 ⬚C that also flowed through the countercurrent heat exchanger surrounding the metallic
delivery tubing. The perfusion chambers were
heated to 37 (<0.1) ⬚C by means of a thermostatically controlled heating element mounted on the
underside of the chamber (fig. 1).
To confirm the ability of the system to successfully
deliver volatile agents to the chamber, we used
halothane as a test agent. We set a calibrated Fluotec
3 vaporizer to 4 % and allowed 15 min for equilibration. We sampled the perfusate from the
reservoir bottle by aspirating with airtight syringes
and obtained samples downstream of the perfusion
chamber by attaching the syringes to a T-piece. The
samples were extracted into known quantities of nheptane. These samples were then analysed by gas
chromatography (Perkins Elmer 8410) with a DB-17
column, using helium as the carrier gas and detection
by flame ionization standardized for halothane [11].
Analysis of cilia beat frequency was performed
using a modification of the transmitted light technique that most closely resembles that described
previously by Teichtahl, Wright and Kirsner [12].
This method detects ciliary movement by their
interference with a light beam which is transduced
from the voltage changes of a photodiode and
processed mathematically to give a power spectrum
of the frequencies produced by ciliary beating.
Samples of ciliated epithelium from a patient were
mounted between the coverslips of two paired
chambers. These were connected to the bottles
containing the HBSS perfusate with one bottle
receiving the volatile agent in air and the other air
alone. When the chambers had equilibrated to 37 ⬚C
as indicated by the temperature plate on the chamber
base, measurements of cilia beat frequency were
commenced. Acceptable ciliated edges for measurement were defined as those devoid of mucus and at
least 60 ␮m long. Measurements were not taken from
individual cells or disrupted edges.
Six patient samples were exposed to each of the
volatile agents studied and readings were taken from
both chambers before exposure to the volatile agent
and then at 1, 2 and 3 h after one chamber was
exposed to the volatile agent and the other continued
to be exposed to air alone. The vaporizer was set at
a concentration of 3 MAC for unpremedicated young
adults at 37 ⬚C. This represented 2.25 % for halothane, 5 % for enflurane and 3.6 % for isoflurane. We
analysed as many acceptable ciliated edges as possible
from a minimum of six to a maximum of 10 from
each chamber in each defined time band.
STATISTICAL ANALYSIS
The mean of the peak cilia beat frequencies from the
power spectrum was computed for each sample at
each time. The nested data were analysed by repeated
measures analysis of variance (SPSS for windows
v5.01). As the effects of volatile agents over time on
cilia beat frequency are unknown, we used polynomial contrasts. Significance was taken at P : 0.05.
Results
HALOTHANE GAS CHROMATOGRAPHY
Measurements from the gas chromatograph
indicated that there was no significant loss of the test
volatile agent during perfusion through the system
(table 1). From the measured aqueous concentrations
of halothane in the delivery bottle and downstream
of the perfusion chamber, the equivalent concentrations (in % atm) were calculated and the estimated
delivered concentrations were derived from a standard curve and were equivalent to 3.62 % (95 %
confidence intervals (CI) 3.02 to 4.22) for the
118
British Journal of Anaesthesia
Table 1 Mean (SEM) halothane concentrations in the delivery
bottle and downstream of the perfusion chamber during
exposure to 4 % halothane
Aqueous concn
(␮mol litre91)
Equivalent concn (% atm)
Estimated delivered concn
(%) atm
Delivery
bottle
(n : 4)
Downstream
of chamber
(n : 7)
785 (56.2)
933.4 (120.5)
2.79 (0.2)
3.62 (0.3)
3.32 (0.43)
4.3 (0.55)
Figure 3 Mean (SEM) nasal cilia beat frequency (CBF) after
exposure to 3 MAC of enflurane (!) or air (■) for 3 h.
Figure 2 Mean (SEM) nasal cilia beat frequency (CBF) after
exposure to 3 MAC of halothane (!) or air (■) for 3 h.
3-h observation period when the cilia were exposed
to enflurane, mean cilia beat frequency in the group
exposed to 3 MAC of enflurane in air was 10.9 (1.3)
compared with 11.6 (1.2) in the group exposed to air
alone. The univariate F test was not significant
(F : 2.3, P : 0.10); however, polynomial contrasts
were significant for the second polynomial (F : 6.2,
P : 0.03). Thus there was a statistically significant
difference between the group exposed to enflurane
and the group exposed to air (fig. 3).
ISOFLURANE
delivery bottle samples and 4.3 % (95 % CI 3.2 to
5.4) for the samples downstream of the perfusion
chamber.
HALOTHANE
There were a total of 349 measurements of cilia beat
frequency from the six paired samples exposed to
halothane and observed over a 4-h period. Over the
3-h observation period when the cilia were exposed
to halothane, mean cilia beat frequency in the group
exposed to 3 MAC of halothane in air was 9.3
(SEM 1.3) compared with 11.4 (1.0) in the group
exposed to air alone. There was a statistically
significant difference between the groups exposed to
halothane or air (univariate F test, F : 9.76,
P : 0.011) (fig. 2). The variances at each time for the
two treatment groups were not significantly different
(Levene’s test for equality of variances).
Multiple t tests with Bonferroni correction showed
a significant difference between the treatment groups
after exposure to halothane for 1, 2 and 3 h (time 0,
P : 1.46; time 1, P : 0.028; time 2, P : 0.004; time
3, P : 0.036).
ENFLURANE
There were a total of 438 measurements of cilia beat
frequency from the six paired samples exposed to
enflurane and observed over a 4-h period. Over the
There were a total of 402 measurements of cilia beat
frequency from the six paired samples exposed to
isoflurane and observed over a 4-h period. Over the
3-h observation period when the cilia were exposed
to isoflurane, mean cilia beat frequency in the group
exposed to 3 MAC of isoflurane in air was 10.8 (1.1)
compared with 11.6 (1.2) in the group exposed to air
alone. The univariate F test was not significant
(F : 3.96, P : 0.075); however, polynomials were
significant for the second polynomial (F : 4.58,
P : 0.01). Thus there was a statistically significant
difference between the group exposed to isoflurane
and the group exposed to air (fig. 4).
Discussion
In this in vitro study, we found that 3 MAC of
halothane, enflurane and isoflurane significantly
reduced cilia beat frequency of human nasal epithelium over a 4-h observation period. There was an
unexpected increase in cilia beat frequency in the
control group paired with samples exposed to
enflurane. The increase in cilia beat frequency in the
control group may represent movements of
exfoliated collections of cells that became more
apparent as the perfusate separated the tissue into
smaller segments over time. In our experimental
design we controlled for all external factors known to
affect cilia beat frequency (temperature, flow rate,
Effects of volatile anaesthetics on cilia beat frequency
Figure 4 Mean (SEM) nasal cilia beat frequency (CBF) after
exposure to 3 MAC of isoflurane (!) or air (■) for 3 h.
pH and osmolality), and when assessing the chamber
we found cilia beat frequency to remain stable for
many hours [10]. By conducting controlled and
blinded experiments we would anticipate that the
increase in cilia beat frequency in the control group
may have similarly occurred in the paired chamber
countered by the depressant effects on cilia beat
frequency of the volatile agent.
In this study we used a supraclinical concentration
of each volatile agent so as to determine the presence
of any effects. This restriction to one concentration
makes it impossible to quantitatively compare the
agents.
The depressant effects of anaesthetic agents on
respiratory mucus clearance are well established
[6–8]; however, there has been limited study of the
effects of anaesthetics on ciliary function, information relevant to understanding the mechanisms of
the depression of mucus transport. The ciliodepressant effects of anaesthetic agents were
recognized by Nunn and colleagues 20 yr ago,
investigating changes in the swimming velocity of
the protozoan, Tetrahymena pyriformis, exposed to
anaesthetic agents [13]. Manawadu, Mostow and
LaForce [14] investigated the effects of halothane on
ferret tracheal cilia, but only in a semiquantitative
way, by noting the presence or absence of ciliary
activity at different sites. Lee and Park demonstrated
suppression of cilia beat frequency of rabbit tracheal
specimens with halothane and enflurane [15]. They
found a 20–25 % reduction in cilia beat frequency
with 3 MAC of halothane and enflurane measured
after 10 min exposure. This is comparable with the
10 % reduction with enflurane and 25 % reduction
with halothane at the earliest time we measured,
after exposure for 1 h.
The limited investigations of inhalation anaesthetic agents on cilia beat frequency have been
restricted to non-human tissue. Although ciliary
structure is similar in different species, ciliary
function and in particular mechanisms of control are
not uniform. The use of human, rather than non-
119
human, tissue for studying the effects of pharmacological agents in clinical use is therefore
preferable. Previous work from our laboratory using
a different methodology investigated the effects of
halothane on human ciliated tissue [9]. In this study
with 1.8 % halothane, there was depression of cilia
beat frequency of 25 % after 1 h and 20 % after 2 h
exposure and with 5.7% halothane, reductions of
20 % and 40 % after 1 and 2 h exposure, respectively.
These are comparable with our findings of depression of 25 % at 1 h and 40 % after 2 h of
exposure to 2.25 % halothane.
The concentration of halothane (measured by gas
chromatographic analysis) varied because of loss of
some of the agent during sampling as a result of its
volatility. The concentration of the volatile agent in
the delivery bottle was smaller than downstream of
the perfusion chamber, because of the greater
difficulties in achieving an airtight seal when sampling from the bottle. The difference in the concentration of halothane in the delivery bottle and
downstream of the perfusion chamber was not
statistically significant. The derived delivered concentration of halothane downstream of the perfusion
chamber with the vaporizer set to 4 % was 4.3 %
(95 % CI 3.2 to 5.4) and suggests that this system
delivered volatile agents to the ciliated tissue in the
sample chamber without significant loss. In an early
pilot study, we noted loss of the volatile agent when
using plastic connecting tubing and therefore
replaced this with metallic tubing.
The ciliated epithelium of the nose sampled in this
study has been shown previously to have the same
beat frequency as cilia located distally in the
bronchial tree [16] and the similarity of our results
with those of Lee and Park using rabbit trachea
suggests that nasal cells are representative of cilia
from more distal parts of the respiratory tract [15].
We have validated our method of sample collection
and shown that beat frequency from awake
volunteers is the same as that after a sleep dose of
propofol [10].
The transmitted light technique for determination
of cilia beat frequency first described by Dalmann
and Rylander in 1962 is the most widely used
method for measurement of cilia beat frequency that
is reproducible, convenient to use and requires
minimal subjective assessment [17]. We refined
Teichtahl’s modification [12] by collecting the
signals produced by beating cilia over a 15-s period
and dividing this into three sequential 5-s intervals
for analysis. We computed the peak frequency of one
power spectrum obtained by calculating the mean of
the power spectra for each epoch. A mean of the
dominant frequency of each of the three 5-s epochs
was then computed. Although shortening the capture
time reduced the resolution of measurement to
0.2 Hz, this was not a limiting factor within the
experimental conditions used and it increased the
confidence interval of the calculated mean beat
frequency.
Postoperative chest infections are a common and
serious form of postoperative morbidity and mortality, and a recent study indicated an incidence of
21.5 % after upper abdominal procedures [1]. These
120
infections are associated with smoking [1], vertical
surgical incisions [1–4] and prolonged surgery [4].
There is no satisfactory evidence to associate
different anaesthetic techniques with the incidence
of chest infections. One of the important defences to
respiratory infection is mucociliary clearance and
this is known to be impaired by some anaesthetic
agents [6]; however, the mechanisms involved have
not been elucidated. Mucociliary clearance depends
on the volume and physical properties of the mucus
and the co-ordination and beating frequency of the
cilia. It appears that anaesthetic agents may not alter
the amount of mucus or its rheological properties
[18] and thus it is likely that they alter ciliary
function.
We found a 10–25 % reduction in cilia beat
frequency with these supraclinical concentrations of
anaesthetics after 1 h of exposure, yet other workers
have demonstrated 80 % reduction in the mucus
transport rate with similar concentrations of anaesthetic agents [6]. This raises questions about the
importance of ciliary beat frequency in determining
mucus transport rate.
Forty years ago, Hill investigated the movement of
carborundum particles across frog oesophagus and
rat trachea [19]. She described a hyperbolic relationship between mucus transport rate and particle
velocity such that there were disproportionately
larger decreases in mucus transport rates at lower
particle velocities. It was hypothesized that this
relationship arose because the transfer of power from
the beating cilia to the mucus was most effective
within a narrow range of ciliary beat frequencies.
Puchelle and Zahm studied the relationship between the transport rate of sputum from bronchitic
patients across mucus-depleted frog palate and the
rheological properties of sputum [20]. Not
surprisingly, the rheological properties of increased
viscosity, increased elasticity and reduced spinnability were associated with reduced transport
rates; however, the most important determinant of
mucus transport rate found by step-by-step multiple
regression was ciliary beat frequency.
The exact relationship between cilia beat frequency and the rate of mucus transport is complicated because of the interplay of multiple factors.
In a study of 20 patients, Katz and colleagues could
not find a direct relationship between cilia beat
frequency of human nasal mucosa and tracheal
clearance rates using a radioisotope technique.
However, in this study the range of mucus transport
rates and cilia beat frequencies in the patients
studied was small, limiting the chances of finding
any relationship if it should exist [21]. Duchateau
and co-workers investigated the relationship between
nasal cilia beat frequency measured photometrically
from biopsies in vitro with nasal mucociliary clearance of dye and saccharine in vivo in 31 healthy
volunteers [22]. The range of mucus transport rates
was twice that of Katz’s study and a good correlation
was found between the logarithm of mucus clearance
rate and cilia beat frequency. Hee and Guillerm
similarly described a non-linear relationship between
cilia beat frequency and mucus transport rates in
sheep [23]. These studies suggest that cilia beat
British Journal of Anaesthesia
frequency may have an important role in determining
the transport rate of mucus and that modest
reductions in cilia beat frequency may be associated
with large reductions in mucus transport rates.
In a separate study [10] we did not find any effect
on cilia beat frequency measured 1 h after in vivo
bolus administration of an induction dose of propofol. This raises the possibility of a difference between
this drug and the inhalation agents on respiratory
cilia; however, studies of the effects of propofol on
cilia beat frequency in vitro with on line measurements are required to compare this drug with
inhalation agents.
References
1. Dilworth JP, White RJ. Postoperative chest infection after
upper abdominal surgery: an important problem for smokers.
Respiratory Medicine 1992; 86: 205–210.
2. King DS. Postoperative pulmonary complications. Surgery
Gynecology and Obstetrics 1933; 56: 43–50.
3. Wightman JAK. A prospective study of the incidence of
postoperative pulmonary complications. British Journal of
Surgery 1968; 55: 85–91.
4. Garibaldi RA, Britt MR, Coleman ML, Reading JC, Pace
NL. Risk factors for postoperative pneumonia. American
Journal of Medicine 1981; 70: 677–680.
5. Gamsu G, Singer MM, Vincent HH, Berry S, Nadel JA.
Postoperative impairment of mucus transport in the lung.
American Review of Respiratory Diseases 1976; 114: 673–679.
6. Forbes AR. Halothane depresses mucociliary flow in the
trachea. Anesthesiology 1976; 45: 59–63.
7. Forbes AR, Horrigan RW. Mucociliary flow in the trachea
during anesthesia with enflurane, ether, nitrous oxide and
morphine. Anesthesiology 1977; 46: 319–321.
8. Forbes AR, Gamsu G. Mucociliary clearance in the canine
lung during and after general anesthesia. Anesthesiology 1979;
50: 26–29.
9. Gyi A, O’Callaghan C, Langton JA. A pilot study of the effect
of halothane on human ciliary beat frequency in vitro. British
Journal of Anaesthesia 1993; 71: 757P–758P.
10. Selwyn DA, Gyi A, Raphael JH, Key A, Langton JA. A
perfusion system for the in vitro measurement of human cilia
beat frequency. British Journal of Anaesthesia 1996; 76:
111–115.
11. Rutledge CO, Siefen E, Alper MH, Flake W. Analysis
of halothane in gas and blood by gas chromatography.
Anesthesiology 1963; 24: 862–867.
12. Teichtahl H, Wright PL, Kirsner RLG. Measurement of in
vitro ciliary beat frequency: a television–video modification of
the transmitted light technique. Medical and Biological
Engineering and Computing 1986; 24: 193–196.
13. Nunn JF, Sturrock JE, Wills EJ, Richmond JE, McPherson
CK. The effect of inhalational anaesthetics on the swimming
velocity of Tetrahymena pyriformis. Journal of Cell Science
1974; 15: 537–554.
14. Manawadu BR, Mostow SR, LaForce FM. Impairment of
tracheal ring ciliary activity by halothane. Anesthesia and
Analgesia 1979; 58: 500–504.
15. Lee KS, Park SS. Effect of halothane, enflurane and nitrous
oxide on tracheal ciliary activity in vitro. Anesthesia and
Analgesia 1980; 59: 426–430.
16. Low P-MP, Luk CK, Dulfano MJ, Finch PJP. Ciliary beat
frequency of human respiratory tract by different sampling
techniques. American Review of Respiratory Diseases 1984;
130: 497–498.
17. Dalmann T, Rylander R. Frequency ciliary beat measured
with a photosensitive cell. Nature (London) 1962; 196:
592–593.
18. Rubin BR, Finegan B, Ramirez O, King M. General
anesthesia does not alter the viscoelastic properties of human
respiratory mucus. Chest 1990; 98: 101–104.
19. Hill JR. The influence of drugs on ciliary activity. Journal of
Physiology 1957; 139: 157–166.
20. Puchelle E, Zahm JM. Influence of rheological properties of
Effects of volatile anaesthetics on cilia beat frequency
human bronchial secretions on the ciliary beat frequency.
Biorheology 1984; 21: 265–272.
21. Katz I, Zwas T, Baum GL, Aharonson E, Belfer B. Ciliary
beat frequency and mucociliary clearance: what is the
relationship? Chest 1987; 92: 491–493.
22. Duchateau GSMJE, Graamans K, Zuidema J, Merkus
121
FWHM. Correlation between nasal cilia beat frequency and
mucus transport rate in volunteers. Laryngoscope 1985; 95:
854–859.
23. Hee J, Guillerm R. Discussion on smoke and mucociliary
transport. European Journal of Respiratory Diseases 1985; 66
(Suppl. 139): 86–88.