![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Vol. 282, Issue 2, 802-811, 1997
Pain Research Group, Department of Anesthesia Research Laboratories, Harvard Medical School, Brigham and Women's Hospital, 75 Francis Street, Boston, Massachusetts
 |
Abstract |
|---|
 Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Mechanosensitive A
-fibers (n = 29) and nociceptive
A
- (n = 6) and C-fibers (n = 10) of
the rat sciatic nerve were superfused with lidocaine (LID, 0.1-1.4 mM)
in vivo. The [LID] to abolish single electrically
stimulated impulses (tonic blockade) in axons was 0.2 to 0.8 mM for
A-, 0.1 to 0.6 mM for A- and 0.1 to 1.4 mM for C-fibers. Within
each of the fiber groups there was no dependence of blocking [LID] on
conduction velocity; slower fibers were no more susceptible than faster
ones. Mean blocking concentrations differed between groups, with
C-fibers having an IC50 = 0.80 ± 0.32 mM
(± S.E.), significantly higher (P < .05, ANOVA) than A-fibers (IC50 = 0.41 ± 0.15 mM) and A-fibers
(IC50 = 0.32 ± 0.18 mM). The [LID]
causing 50% impulse failure in A-fibers during a 200-Hz, 10-stimulus train (phasic blockade) ranged from 0.2 mM to 0.7 mM; the
mean IC50 equaled 0.28 mM (n = 17). Stimulation of nociceptive A-fibers (n = 4) and
C-fibers (n = 5) at 5 or 10 Hz for 10 pulses produced
no phasic block at [LID]s (0.1-0.5 mM) below those required for
tonic blockade. Uptake of 14C-lidocaine by the
nerve, measured in vivo under conditions identical with
those for electrophysiology, showed that: a) little drug was in the
segments of nerve beyond the superfusion chamber, b) lidocaine was
uniformly distributed in the nerve within the chamber, c) the
intraneural lidocaine content was identical with that in nerves
equilibrated in vitro. The results show a lack of monotonic dependence of sensitivity to local anesthetic on fiber diameter, but do
suggest that mean susceptibility to nerve block by lidocaine differs
for fibers grouped by, and perhaps according to, function.
| |
Introduction |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
It
is a widespread belief in anesthesia and pharmacology that impulses in
nerve fibers of smaller diameter are blocked by lower concentrations of
local anesthetics (LA) than those of larger diameter (Catterall and
Mackie, 1996
). This historical hypothesis is based on studies of CAPs
(Gasser and Erlanger, 1929; Heinbecker et al., 1934) that
measured summed electrical elevations from axons of smaller diameter
(slow CV) and showed that these elevations were decreased more by LA
than those of larger diameter (fast CV) axons. This hypothesis,
however, has been contradicted by results from single unit studies on
mammalian fibers in vivo (Franz and Perry, 1974) and
in vitro (Fink and Cairns, 1984, 1987). Some CAP studies
have also found that fibers grouped by CV did not show a susceptibility
to LA that always correlated with conduction speed (Heavner and de
Jong, 1974; Gissen et al., 1980). On the other hand, a great
deal of interfiber variation in LA sensitivity has been observed in
axons of similar conduction velocity and, by inference, similar
diameter (Fink and Cairns, 1987; Raymond, 1992). It has been suggested
that the observed variations in LA sensitivity might be linked to
differences in functional modalities of peripheral axons (Raymond and
Gissen, 1987); this report is the first to investigate that hypothesis.
Local anesthetics have three important effects on axonal excitability.
The first is a tonic reduction of excitability (measured at very low
firing frequencies, e.g., < 0.1 Hz) which culminates in
conduction failure. In vitro studies have shown that
increase of LA at sub-blocking concentrations causes the conduction
velocity of axons to decrease progressively until abolition of
conduction (Fink and Cairns, 1984; Raymond, 1992). The second
well-known effect of LA is its use-dependent action. During trains of
stimuli in nerves exposed to LA in vitro, CAPs undergo
progressive diminution, pulse by pulse, that increases with the
frequency and number of impulses in the trains (Trubatch, 1972;
Courtney et al., 1978). Voltage-clamp studies have revealed
that this "use-dependent" action is caused by the increased
affinity of LA for those configurations of Na+
channels that predominate during the membrane depolarization associated
with the impulse, thus increasing the population of LA-bound channels
and decreasing the capability of the axon to generate depolarizing
currents needed to sustain propagation (see review, Butterworth and
Strichartz, 1990). The third effect is the suppression of the
activity-dependent aftereffects of axonal impulse conduction, which has
been demonstrated in rat (Raymond et al., 1991) and frog
(Raymond, 1992) sciatic nerve in vitro. To our knowledge, no
in vivo single unit study on mammalian fibers has ever
investigated these three effects of LA simultaneously.
The primary aim of the present work was to examine LA blockade of
functionally characterized sensory axons. We focused on primary
afferents of rat sciatic nerve conducting in the A-, A- and
C-fiber ranges. Lidocaine, a well-studied and widely used clinical LA,
was used. We measured the range of lidocaine concentrations that was
required to abolish single impulses induced by infrequent electrical
stimulation (tonic blockade). For low threshold mechanosensitive (LTM)
A-fibers, we also studied the use-dependent action of lidocaine as
well as its effects on activity-dependent conduction latency changes
with trains of high-frequency electrical stimulation. Additionally, we
determined the net uptake of radiolabeled lidocaine to assess the
concentrations and spatial distribution of lidocaine along the nerve.
An abstract reporting part of this work has been published (Huang
et al., 1995).
| |
Materials and Methods |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Animal preparation.
Male Long-Evans rats weighing 500 to
550 g (Charles River Laboratories, Wilmington, MA) were initially
anesthetized with an intraperitoneal injection of 50 mg
kg
1 sodium pentobarbital (Abbott
Laboratories, North Chicago, IL). The jugular vein was cannulated for
intravenous administration of sodium pentobarbital (5-6 mg
kg1 h 1),
which maintained general anesthesia throughout the experiment; supplementary boli were administered as necessary, with use of absence
of both the corneal reflex and the increase of heart rate during
electrical stimulation as the endpoint for adequate anesthesia. The
heart rate was monitored with a Tektronix 408 EKG monitor (Tektronix,
Beaverton, Oregon). Core body temperature was maintained at 38.0 ± 0.5°C with a water-circulated heating pad. At the end of an
experiment rats were sacrificed by an overdose of sodium pentobarbital
(100 mg kg1). Animal treatment for these
experiments was approved by the Harvard Medical Area Standing Committee
on Animals.
Solutions. The lidocaine solutions were made by adding lidocaine HCl powder (Sigma Chemical Co., St. Louis, MO) to modified Liley solution (NaCl, 118 mM; KCl, 5 mM; CaCl2·2H2O, 2 mM; NaHCO3, 25 mM; NaH2PO4, 1.2 mM; MgCl2·6H2O, 1 mM; Na2HPO4, 3.6 mM; glucose, 10 mM). The solution was warmed to 35 ± 0.5°C and bubbled with a mixture of 95% O2/5% CO2 to maintain the pH between 7.4 and 7.5 before perfusing it through the bathing chamber.
Identification of single units.
Single units were isolated
and characterized by natural stimulation, applying innocuous and
noxious mechanical stimuli to various sides of the glabrous skin of the
foot. The receptive field, modality and location were identified
conventionally (Sanders and Zimmerman, 1986; Leem et al.,
1993). A unit was characterized as nociceptive when it responded to
either noxious mechanical pinching with a serrated forcep tip or the
hard squeezing of a skin fold by the fingers of the experimenter. A
unit was characterized as low-threshold mechanoreceptive when it
responded to light touch by the tip of a glass probe. Three subtypes of
low-threshold mechanoreceptive afferents were sought in this study:
slowly adapting mechanoreceptive (SA), rapidly adapting
mechanoreceptive (RA) and muscle (M) afferents. Rapidly adapting units
reduce discharge rates quickly after stimulus onset (<4 s) and also
fire when the stimulus is released. Slowly adapting units respond for a
longer time (>4 s) and, as the name implies, take longer to reduce
firing rates under maintained stimulation. Muscle afferents responded
to any mechanical stimulus that altered muscle tension
(e.g., stretching by lateral displacement) and continued to
discharge with little or no adaptation.
1, between
2.5 and 15 m s1 and above 15 m
s1 were considered C-, A- and
A-fibers, respectively.
The "normalized latency increase" was calculated according to the
following formula, to account for the variations in conduction length
of fibers:
|
|
) and cross-coupled physiological
and electrical latency changes (Thalhammer et al., 1994)
were used to ensure that an electrically activated unit was identical
with the unit activated by natural stimulation of its receptive field.
Measurement of tonic and phasic block.
Tonic block was
measured by increasing the concentration of lidocaine in the nerve
chamber by steps of 0.1 mM until impulse conduction was fully
abolished. All latency measurements were obtained at steady state after
20 min of lidocaine application. At steady state, each unit was
stimulated by infrequent stimuli (1 Hz for A fibers and 0.2 Hz for
A and C-fibers) to determine the tonic action of lidocaine.
-fibers with use of a preset stimulation profile
that was developed to reveal the various phases of changes in latency
that usually occur during repetitive stimulation. The profile consists
of three discrete phases: 1) a 1-Hz stimulus period, lasting 30 s,
to assess base-line conduction latency and latency fluctuations under
conditions when no cumulative effects were noticeable; 2) a period of
repeated short bursts (20 trains, presented at 1.3 Hz, of 10 stimuli at
200 Hz), for measuring cumulative changes during each short burst and
during the entire period of repeated bursting; 3) a 1-Hz recovery
period lasting 60 s. The activity-dependent measurements consisted
of burst supernormality and depression, which were derived from the
response latencies to the above profile, as will be explained below.
Phasic block was quantified as the percent of impulses that failed to
conduct during the 200-Hz stimulation in the above profile.
Phasic block was examined in A- and C-fibers by stimulating these
fibers with 10 to 20 pulses at 5 and 10 Hz, frequencies well within
their natural discharge pattern (Burgess and Perl, 1967).
The response signals were amplified by a differential electrometer
preamplifier (MetaMetrics, Cambridge, MA) and filtered with an
adjustable, high-frequency single pole filter (1-5 kHz). Filtered
signals were further amplified and visualized on a Tektronix 5103N
oscilloscope (Tektronix, Inc., Beaverton, OR) and relayed to a
differential amplitude discriminator designed and built in our
laboratory. Response latency data were obtained through our "Hunter" system, which times both stimuli and responses (±1 µs resolution) and stores each event to disk. The data were then graphed
and analyzed by MicroCal Origin (Microcal Software, Northampton, MA)
and Quattro Pro (Borland International, Scotts Valley, CA).
Stability of preparation. The whole procedure required observation of a given unit for periods as long as 6 h. Control experiments showed that resting conduction latency of three individual tactile units remained quite stable, increasing by less than 3% when continuously monitored for as long as 6 h. In consideration, the background increase of the conduction latency was not taken into account in the measurements reported in this study. All electrophysiological measurements in this study were done within 6 h after a unit was isolated.
In vivo uptake of lidocaine.
Lidocaine uptake
was studied using the 14C-lidocaine solution
prepared by adding 10 µl 14C-lidocaine
(DuPont-NEN, Boston, MA; 0.1 mCi/ml ethanol) to each 5 ml of 0.4 mM
lidocaine solution that was superfused through the chamber. After
in vivo exposure times of 7 min, 20 min or 60 min, four
pieces of the nerve were excised: one distal to the chamber (7 mm), one
within the bathing chamber (22 mm) and two adjacent pieces proximal to
the chamber (7 mm each). The nerve pieces were quickly frozen on a flat
surface of dry ice. The nerve segment from within the bathing chamber
was further cut into three segments of equal length (7.3 mm each) using
a surgical blade. All excised nerve segments were then desheathed under
a dissecting microscope. Each desheathed piece of nerve was weighed (± 0.1 mg) and then digested at 50°C for 2 h in a 5-ml
scintillation vial containing a mixture of 0.5 ml of tissue solubilizer
(Solvable, DuPont-NEN) plus 0.1 ml of deionized water. Five milliliters
of Aquasol 2 (DuPont-NEN) liquid scintillation cocktail was added and
radioactivity was assayed by liquid scintillation counting for 5 min.
The specific radioactivity of the lidocaine solution was determined by
adding 10 µl of the 14C-lidocaine solution to
Solvable (0.5 ml), deionized H2O (0.1 ml) and
Aquasol 2 (5.0 ml) mixture and counting identically by liquid
scintillation counting. Derived counts per minute were divided by moles
of lidocaine added to define the specific radioactivity. Efficiency of
counting 14C was the same in all conditions and
the radioactivity was assumed to represent lidocaine HCl. Tissue drug
was expressed as nanomoles lidocaine per mg wet weight of nerve
(Popitz-Bergez et al., 1995).
Statistics. The results are reported here as means ± S.E.; the numbers of independent observations are also included. One-way ANOVA test and unpaired two-tail Student's t test (on normally distributed data) were used to evaluate the difference of mean tonic blocking concentrations of lidocaine among functional groups of axons, and the difference of mean normalized latency increases (see "Results") before block. P < .05 was considered to be statistically significant.
| |
Results |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Tonic blocking of conduction by lidocaine. From about 100 experiments (1 unit per rat per experiment), 45 units were selected that fulfilled the following criteria: 1) receptive fields were all in the glabrous skin; 2) no cutaneous injury had been made within or close to (<10 mm) the receptive field; 3) the shape and amplitude of the unit action potential remained relatively stable during the whole experiment; 4) after conduction block by lidocaine the response of a unit to both natural and electrical stimulation returned within 30 to 45 min after washout with Liley solution.
Axons of sensory afferents differed in their tonic sensitivity to lidocaine. Traces from a nociceptive afferent C-fiber (resting CV = 1.05 m s1) in which infrequently
stimulated impulse conduction (0.2 Hz) was blocked by 0.8 mM lidocaine
are shown in figure 1. At incrementally increasing concentrations of lidocaine the conduction latency increased
progressively, reaching steady state after 15 to 20 min at each
concentration until, at 0.8 mM lidocaine, conduction failure occurred.
|
- or A-fibers had a narrower range: 0.1 to 0.6 mM for
nociceptive A-fibers and 0.2 to 0.8 mM for LTM A-fibers. The
median blocking sensitivity for different functional groups differed
(table 1). Nociceptive C-fibers had the
highest median blocking concentration, 0.80 ± 0.32 mM, which is
significantly higher than that of nociceptive A-fibers (P < .0001) and LTM A-fibers (P < .02). Nociceptive A-fibers
tended to be blocked at a lower concentration (0.32 ± 0.18 mM)
than that for blocking the LTM A-fibers (0.41 ± 0.15 mM),
although the difference was not statistically significant (P = .21). Furthermore, nociceptive fibers as a group (including both A-
and C-fibers) showed a significantly higher median blocking concentration (0.62 ± 0.36 mM) than the median for LTM
A-fibers (P < .01).
|
|
-fibers and LTM A-fibers, respectively. Similarly, there was no
correlation between the normalized latency increase that occurred just
before conduction block and the resting CV (fig. 3B) for nociceptive
A-fibers and LTM A-fibers (correlation coefficients = 0.49 and 0.09, respectively). However, a significant correlation was found
between the normalized latency increase and the resting CV for
nociceptive C-fibers (correlation coefficient = 0.82).
Furthermore, on average, in these most LA-resistant axons, conduction
also slowed more before the impulse was abolished than in the
myelinated fibers (table 1).
|
-fiber group to see whether or not
the lidocaine susceptibility was related to the functional submodalities. Table 2 shows that there
was no significant difference among the mean tonic blocking
concentrations for the three functional subgroups of the
mechanosensitive A-fibers: RA, SA and M. Among the three subgroups,
M axons showed the smallest mean latency increase before conduction
block, but this was not significantly lower than those for axons of the
other two subgroups.
|
Use-dependent block by lidocaine.
Lidocaine induced a
use-dependent failure of conduction during high frequency stimulation
of mechanoreceptor afferents. At 0.1 mM lidocaine, no impulse
conduction failures were observed in any LTM A-units
(n = 17: 7 RA, 4 SA, 6 M) during the 10-pulse, 200-Hz
stimulation train used to induce phasic blockade. The phasic blocking
concentration of lidocaine that caused either 10% or 50% failure of
impulse conduction in fibers that were not already tonically blocked
ranged from 0.2 to 0.6 mM and 0.2 to 0.7 mM, respectively. The
use-dependent blocking effect, assessed as the ratio of phasic
block/tonic block, was greatest at lidocaine concentrations of 0.2 to
0.3 mM (fig. 4A). We calculated the
weighted failure (the number of failed impulses over total number of
stimulated impulses) during burst stimulation at different lidocaine
concentrations for 17 LTM A-units and plotted a
concentration-response curve for weighted failure (fig. 4B). Whereas
the IC50 of lidocaine for tonic block was 0.41 mM
(table 1), a 50% phasic block at 200 Hz could be produced by 0.28 mM
lidocaine (fig. 4B). This reflects an effective increase of 46% in
lidocaine's potency for blocking impulses in A afferents.
|
-fibers produced by lidocaine did
not correlate with resting CV (fig. 5).
The correlation coefficients for the 10% or 50% phasic blocking
concentrations against the resting CV were 0.16 and 0.24, respectively. Furthermore, the use-dependent block susceptibility was
not related to the functional submodalities of the LTM A-fibers
(data not shown).
|
-fibers and five C-nociceptors were studied for
use-dependent block using a stimulation profile consisting of 5-Hz and
10-Hz bursts (10 stimuli in each burst) separated by a 30-s, 0.5-Hz
recovery period. These are impulse frequencies consistent with
discharge to a natural noxious stimulus in small diameter nociceptive
fibers (Burgess and Perl, 1967; Perl, 1968; Raymond et al.,
1990). None of these fibers showed any phasic conduction failure during
the bursts at lidocaine concentrations that were below their tonic
blocking concentrations (0.1-0.7 mM). The results indicate that the
frequency or the number of burst stimuli used here do not induce a
phasic blocking action of the LA on nociceptive axons. Higher stimulus
frequencies on longer pulse trains were not used to minimize the
contribution of the intrinsic failure propensity previously recorded in
these axons. (Raymond et al., 1990).
Activity-dependent action of lidocaine.
The excitability of a
drug-free fiber is affected by preceding impulse activity. After the
refractory period some axons remain "subnormal" (CV is decreased),
whereas others enter a period of increased excitability and conduction
velocity called the "supernormal" period (Adrian and Lucas, 1912;
Gasser and Grundfest, 1936; Bowe et al., 1987). In each
burst of the 10-pulse stimulation profile used here, supernormality was
indicated by the slightly shorter latency of the response to the second
stimulus in the train (fig. 6A). The
"burst supernormality" was calculated as the average latency
difference between the first and second impulses of each of the 20 bursts, and was normalized as a percent of the resting latency. For
example, the RA unit characterized in figure 6 has a burst
supernormality of +0.65%, i.e., on average, the second impulse of each burst traveled 0.65% faster than the first impulse.
|
-units (7 RA, 4 SA and 6 M) with a preset
stimulation profile (see "Materials and Methods"). Under control
conditions, not all units have a burst supernormality. Lidocaine
changed burst supernormality in a dose-dependent manner in these axons:
at 0.1 mM lidocaine, the burst supernormality decreased by an average
of 198 ± 88% from the drug-free values; at 0.2 mM lidocaine, the
burst supernormality decreased by an average of 443 ± 127%
(n = 16: 6 RA, 4 SA, 6 M). As a result, at 0.2 mM
lidocaine positive supernormality was not observed in any of the units
that showed this behavior under control conditions (n = 10: 4 RA, 2 SA, 4 M); in fact, only further slowing of conduction was
present as a result of previous impulse activity during burst stimulation. As shown in figure 6B, lidocaine at 0.4 mM produced tonic
slowing of the resting conduction and an even greater slowing plus
occasional failures during the burst, and at 0.5 mM lidocaine (fig. 6C)
no impulses in the burst were conducted after the first one, reflecting
use-dependent block (see fig. 4A above). Supernormal behavior returned
after lidocaine washout, although some tonic conduction slowing
persisted (fig. 6D).
After prolonged repetitive impulse conduction in LA-free conditions,
axons show a long-lasting increase in conduction latency which is
termed "depression" (Raymond, 1979; Carley and Raymond, 1987). In
the present study most A-units (14 of 17) showed some degree of
depression after the 200-Hz burst stimulation (fig. 7A). Among these fibers, depression was
reduced by lidocaine in a concentration-dependent manner in 11 (4 RA, 4 SA, 3 M; fig. 7, B and C). In general, lidocaine induced a slowing of
conduction in A-fibers during a burst but suppressed the
endogenous slowing that otherwise follows prolonged bursts
in the absence of lidocaine, as shown by the altered patterns
of dots and hatching in figure 7.
|
In vivo uptake of
14C-lidocaine.
The time course of
lidocaine uptake by the nerve during superfusion in vivo had
a phase of rapid uptake from 0 to 7 min; a steady state was reached in
about 20 min (fig. 8, inset). Further exposure for 60 min did not result in a significantly greater uptake of
lidocaine by the nerve than that at 20 min (P > .10).
|
).
| |
Discussion |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
Lidocaine susceptibility and functional modalities.
With
regard to diameter, the in vivo results on rat sciatic nerve
reported here are consistent with prior in vitro findings in
frog sciatic nerve (Raymond, 1992) and rabbit vagus nerve (Fink and
Cairns, 1984), as well as in vivo work in cat sciatic nerve (Franz and Perry, 1974), all of which showed that nerve diameter, as
estimated by conduction velocity, does not predict the tonic anesthetic
susceptibility for conduction failure of individual fibers.
-fibers were blocked over a relatively narrow range of lidocaine concentrations (0.2-0.8 mM), whereas nociceptive C-fibers tended to be
blocked over a wider range (0.1-1.4 mM). When characterized by their
median blocking concentrations (IC50s),
nociceptive A-fibers were blocked at the lowest concentration, 0.32 mM, compared with 0.41 mM for LTM A-fibers and 0.80 mM for
nociceptive C-fibers.
The difference in tonic lidocaine susceptibility observed between
nociceptive C-fibers and A-fibers was not altogether surprising. These two groups of fibers subserve two distinct types of pain perception. Nociceptive C-fibers innervating the glabrous skin have
been studied extensively in rat, monkey and human, where they are
believed to be involved in the conduction of burning pain or "slow
pain" (LaMotte and Campbell, 1978; Fleischer et al., 1983;
Campbell et al., 1989), and are not easily sensitized to
heat (Campbell and Meyer, 1983; LaMotte et al., 1983).
Nociceptive A-fibers, also called high-threshold mechanoreceptive
A-afferents (Burgess and Perl, 1967; Perl, 1968), are probably
similar to the type I nociceptive A-fibers found in human and monkey,
fibers which are responsible for the primary hyperalgesia after injury to glabrous skin (Campbell and Meyer, 1986) and which have lower thresholds to mechanical stimuli than nociceptive C-fibers (Campbell et al., 1989). Pricking pain or "fast pain" is conducted
by A-fibers. The different functional roles and impulse codes for
nociceptive C-fibers and A-fibers could be associated with the
physiological attributes of axons, and in turn, result in a difference
in impulse conduction safety (Raymond et al., 1990) and
hence, susceptibility to lidocaine.
The data on tonic susceptibility of single units to lidocaine reported
here and in other studies have been understood to imply that intrinsic
differences in conduction safety exist among fibers (Raymond and
Gissen, 1987). Prior studies strongly suggest that such differences do
not originate solely on the basis of fiber size (Fink and Cairns, 1984;
Raymond, 1992). Conduction safety depends on several features of axonal
electrophysiology other than size, including the density and gating
behavior of particular ion channels, the locus and degree of
myelination, the density of
Na+/K+ or other
electrogenic ion pumps, etc. Grouping fibers by function in this study
has, to a certain extent, revealed such a difference, with the
assumption that lidocaine is equally intrinsically potent in inhibiting
the Na+ channels in these different primary
afferents.
It is important to point out that in this study we did not attempt to
differentiate functional subgroups within A- or C-nociceptive fibers
based on coding or firing pattern. It is possible, therefore, that the
wide range of tonic lidocaine-blocking concentrations observed in
nociceptive C-fibers reflects a variation in lidocaine sensitivity
among different subgroups of nociceptive axons.
The length of nerve segment exposed to LA.
Many studies of
differential block with local anesthetics, in which nerve trunks were
exposed to LA, were done with exposure lengths of 30 mm or less (see
review, Raymond and Gissen, 1987). Studies with longer nerve regions
exposed (Fink and Cairns, 1984; Raymond, 1992) do not support the
historically familiar finding that nerve axon diameter governs the
susceptibility to blockade. How does impulse conduction proceed in
regions exposed to near blocking anesthetic concentrations? Such
conduction is "decremental" (Lorente de Nó and Condouris,
1959; Raymond et al., 1989), decreasing in velocity
continuously through the exposed axons' length, a fact that
contradicts the traditional idea (Franz and Perry, 1974; Nathan and
Sears, 1961) that an exposure length sufficient to cover three
successive nodes in the largest fibers (Takeuchi and Tsaski, 1942)
ensures that length is not a factor in potency or in differential
blockade. Because the effect of a change in exposure length on tonic LA
susceptibility is smaller at long initial lengths, being minimal above
15 mm exposure in large myelinated fibers (Raymond et al.,
1989), in this study the bathing chamber was made as long as was
anatomically congruent (22 mm).
Triple actions of lidocaine: tonic, use-dependent and
activity-dependent effects.
The dynamic effects of tonically
sub-blocking concentrations of lidocaine during impulse discharge are
complex and will depend on the pattern of impulses contained within a
burst. For the LTM A-fiber population that was examined carefully
here, the overall slowing in CV and the increase in failure probability
during 200-Hz burst stimulation was large at concentrations (>0.4 mM)
where half the A-fibers were already tonically blocked (fig. 6). At lower lidocaine concentrations (0.2-0.3 mM), phasic block at 200 Hz
accounted for a relatively greater increase in failures, even though
the overall failure rate was less than at 0.4 mM. Lidocaine's phasic
impulse blocking action is determined by its intrinsic dissociation
rate from "activated" axonal sodium channels (Chernoff, 1990) as
well as by its concentration-dependent association rate. The effective
potency for impulse failure is thus frequency dependent (fig. 4A)
in a way that will be more pronounced for the more slowly dissociating
LAs, such as bupivacaine (Chernoff, 1990).
Single unit vs. CAP data.
The sensitivity of
A-fiber impulses to lidocaine can be compared with data from
Rosenberg and Heavner (1980) in which inhibition of the A-fiber CAP was
studied in vitro in rat sciatic nerve. The median value for
tonic block of A-fibers from the present study was 0.41 mM (table
1), substantially higher than the 0.19 mM IC50
value from CAP inhibition found by Rosenberg and Heavner (1980). The
difference suggests that the suppression of CAP amplitudes by low
concentrations of lidocaine (<0.2 mM) is not caused by the abolition
of impulse conduction in fibers so much as it is caused by the
differential slowing of impulse conduction and/or the decrease in the
impulse currents in single fibers. Neither of these two factors
presages a change in perception to the extent that abolition of
impulses does, although both it may contribute to altered burst
patterns.
The correlation of impulse blockade with behavioral
modifications.
Impulse blockade can be correlated with behavioral
changes assessed neurologically in the rat after injection of lidocaine near the sciatic nerve (Popitz-Bergez et al., 1995). Within
5 to 10 min after injection of 0.1 ml of 38 mM (1%) lidocaine (pH = 6.8), a steady state of intraneural lidocaine is reached. This plateau level, averaging about 4 nmol/mg wet nerve and maintaining a
constant longitudinal, Gaussian-like spread with 
10 mm, persists for 10 to 15 min and then declines to control levels over the
next 40 to 50 min. The plateau value of intraneural lidocaine from such
injections can also be reached by equilibrium in vitro incubation in bathing concentration of 0.8 mM lidocaine (Strichartz, G. R. and Leeson, S., unpublished observation), a concentration which in the present study blocked impulses in 100% of both LTM A-fibers and nociceptive A-fibers and in 65% of the nociceptive C-fibers. During this plateau period, neurological examination showed
that motor function and proprioception in regions innervated by the
sciatic nerve were also fully blocked and that nociception was greatly
reduced (Thalhammer et al., 1995).
) and might
be correlated with lidocaine's differential impulse blockade. Proprioception was the first activity to show impairment, followed by
motor function and nociception. This finding is consistent with the
dependence of proprioception on mechanosensitive A-fibers and the
greater mean vulnerability of such fibers to impulse blockade by
lidocaine (table 1). During regression of the block at 30 to 40 min
there was an early, graded recovery of nociception; withdrawal to deep
digit pinch, mediated by proximal thigh and hip muscles, returns at a
time when pinching only the skin produces no withdrawal response.
Intraneural lidocaine assessed at this stage corresponds to an
equilibrium bathing concentration of 0.2 to 0.3 mM, for which tonic
block of nociceptive A-fibers is 50 to 65%, of LTM A-fibers is
10 to 40% and block of nociceptive C-fibers is 10 to 15% (cf. fig.
2). This partial impulse blockade is thus consistent with the
observation of partial anesthesia in vivo.
Activation of A-fibers may be more prominent during skin pinch
compared with a relatively greater recruitment of C-fibers when deeper
tissue structures of the digit are squeezed, which could account for
the differential behavioral response on the basis of differential tonic
block. However, until we know the afferent discharge patterns in the
various fiber groups for these different stimuli and the overall phasic
failure properties under these conditions, we cannot predict the
integrated actions of LAs on peripheral afferent impulse propagation.
Given the current tools and techniques, an integrated approach should
be capable of resolving this correlation under the dynamic conditions
of in vivo local anesthesia.
One caution is in order. The small number of A and C-nociceptors in
this study may not be representative of the overall population of
nociceptors, because only robust axons with persistent conduction properties fit our selection criteria. Possibly, fibers with more marginal conduction properties are important contributors to
nocifensive input in intact animals and some care should be used in
extending these results to the behavior of all nociceptors in
conscious, responding animals.
In conclusion, the results reported here support prior single unit
studies showing the lack of correlation of LA sensitivity with fiber
diameter. The separation of nociceptors in A- and C-fibers on the
basis of CV shows that these two classes, despite serving the same
general sensory modality, are blocked at about 4-fold different
anesthetic concentrations. Therefore, function per se does
not determine susceptibility to block, assuming that pinching pain,
burning pain and aching pain are the same modality (Campbell and
LaMotte, 1983). On the basis of the apparent differential sensitivity
to lidocaine of A- over C-fibers, during peripheral nerve block an
anesthetist might presume from stimuli that selectively induce first
pain (e.g., pin prick) the existence of a degree of
analgesia that is in actuality insufficient to block second pain
(e.g., burning, aching). Which of these forms of pain
accompanies surgical incision and its sequelae is an important question
for establishing anesthesia adequate for surgery and sufficient to preempt postoperative hyperalgesia (Kissin and Raja, 1995).
| |
Acknowledgments |
|---|
We wish to thank Dr. S. Leeson and Dr. F. A. Popitz-Bergez for help with the lidocaine uptake experiments, D. S. Chang for collecting some of the data and S. S. Waikar for writing a computer program for data analysis. Excellent secretarial support by Ms. Ellen Jacobson is gratefully regarded.
| |
Footnotes |
|---|
Accepted for publication April 21, 1997.
Received for publication October 22, 1996.
1 This work was supported by US National Institutes of Health grant GM 35647 (to G.R.S.).
Send reprint requests to: Gary R. Strichartz, PhD, Department of Anesthesia Research Laboratories, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115.
| |
Abbreviations |
|---|
LID, lidocaine; CV, conduction velocity; LTM, light-touch mechanoreceptor; CAP, compound action potential; LA, local anesthetic; SA, slowly adapting mechanoreceptive afferents; RA, rapidly adapting mechanoreceptive afferents; M, muscle afferents.
| |
References |
|---|
|
Top
Abstract
Introduction
Materials & Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  V. Pinto, V. A. Derkach, and B. V. Safronov Role of TTX-Sensitive and TTX-Resistant Sodium Channels in A{delta}- and C-Fiber Conduction and Synaptic Transmission J Neurophysiol, February 1, 2008; 99(2): 617 - 628. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sakaguchi, Y. Kuroda, and M. Hirose The antiproliferative effect of lidocaine on human tongue cancer cells with inhibition of the activity of epidermal growth factor receptor. Anesth. Analg., April 1, 2006; 102(4): 1103 - 1107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 |
, nociceptive
C-fibers, n = 10; 
, LTM A
, nociceptive A
, RA,
n = 10; 
M, n = 8; 
, nociceptive A
