Aerobic Training Increases Pain Tolerance

in Healthy Individuals
MATTHEW D. JONES1, JOHN BOOTH1, JANET L. TAYLOR1,2, and BENJAMIN K. BARRY1,2
1
School of Medical Sciences, University of New South Wales, Sydney, AUSTRALIA; and 2Neuroscience Research Australia,
Sydney, AUSTRALIA

ABSTRACT
JONES, M. D., J. BOOTH, J. L. TAYLOR, and B. K. BARRY. Aerobic Training Increases Pain Tolerance in Healthy Individuals. Med.
Sci. Sports Exerc., Vol. 46, No. 8, pp. 1640–1647, 2014. The hypoalgesic effects of acute exercise are well documented. However, the
effect of chronic exercise training on pain sensitivity is largely unknown. Purpose: To examine the effect of aerobic exercise training on
pain sensitivity in healthy individuals. Methods: Pressure pain threshold, ischemic pain tolerance and pain ratings during ischemia were
assessed in 24 participants before and after 6 wk of structured aerobic exercise training (n = 12) or after 6 wk of usual physical activity
(n = 12). The exercise training regimen consisted of cycling three times per week for 30 min at 75% of maximal oxygen consumption
reserve. Results: Significant increases in aerobic fitness (P = 0.004) and ischemic pain tolerance (P = 0.036) were seen in the exercise
group after training, whereas pressure pain threshold and pain ratings during ischemia were unchanged (P 9 0.2). No change in aerobic
fitness (P 9 0.1) or pain sensitivity (P 9 0.1) was observed in the control group. Conclusion: Moderate- to vigorous-intensity aerobic
exercise training increases ischemic pain tolerance in healthy individuals. Key Words: PAIN THRESHOLD, PAIN RATING,
HYPOALGESIA, ISCHEMIC PAIN, CHRONIC EXERCISE

I
n healthy individuals, the hypoalgesic effect of acute Furthermore, the aerobic exercise group, despite tolerating
exercise is well documented (25). Chronic exercise train- higher intensities of mechanical pressure, reported a more se-
ing is also demonstrated to reduce pain sensitivity in pa- vere subjective appraisal of pain after that training period.
tients with persistent pain (16,40), and accordingly, exercise Resistance training had no influence on pain tolerance (5).
training has become an important part of treatment in these Although the study provides preliminary, but weak, evidence
patients (8,15). However, despite the growing evidence for that aerobic exercise training may lead to increased pain
a pain-relieving effect of exercise training in chronic disease tolerance, it has several limitations. The effect of exercise
populations, the effect of aerobic training on pain sensitivity training on pain threshold, and on the duration that the pain-
in healthy individuals is largely unknown. Hence, little is ful stimuli could be tolerated, were not quantified. The vol-
known of how exercise training may modulate pain inde- ume and intensity of exercise performed by participants
pendently of disease. was highly varied across the 12-wk period, making it diffi-
To our knowledge, only one study has examined the effect cult to identify the volume and intensity of exercise needed
of chronic aerobic exercise on pain sensitivity in healthy par- to elicit the hypoalgesic response. Lastly, maximal aerobic
ticipants (5). Anshel and Russell (5) examined the effect of capacity was not measured, so the influence of exercise
12 wk of aerobic or resistance or combined aerobic and re- training on V̇O2max, pain sensitivity, and endurance perfor-
sistance exercise on pressure pain tolerance in 48 unfit males. mance cannot be determined.
The group who completed aerobic exercise training (cycling) Numerous other studies have sought to find a relation be-
tolerated a greater magnitude of mechanical pressure applied tween fitness or sporting achievement and pain sensitivity
to the upper limb. A similar pattern of change was apparent based on the anecdotal observation that athletes are more
for the lower limb, but the results did not reach significance. stoical. Findings from these cross-sectional studies are equivo-
APPLIED SCIENCES

cal and have varied depending on the sport and even the
phase of the competitive season or the standard of competi-
tion from which athletes have been studied (31,33,34), as
Address for correspondence: Matthew D. Jones, BExPhys, MSc, School of
Medical Sciences, University of New South Wales, Kensington 2052; well as with the coping strategies used by different individuals
E-mail: [email protected]. (23,28). Frequently, the volume, the intensity, the duration,
Submitted for publication July 2013. and the type of exercise training performed by the athletes
Accepted for publication January 2014. have been poorly controlled or quantified. Furthermore,
0195-9131/14/4608-1640/0 findings have depended heavily on the modality and protocol
MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ for evoking pain, in particular whether pain thresholds or pain
Copyright Ó 2014 by the American College of Sports Medicine tolerance have been measured (31). In general, these cross-
DOI: 10.1249/MSS.0000000000000273 sectional investigations provide evidence that chronic exercise

1640

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
can increase pain tolerance, but not pain threshold (28,33,36). was used to assess six subscales of mood (tension, depression,
This notion is supported by Anshel and Russell (5), who confusion, anger, vigor, and fatigue) (29). The long form of
demonstrated an increase in pain tolerance after exercise the International Physical Activity Questionnaire was used to
training. To our knowledge, this is the only study examining evaluate physical activity levels (9). Height, body mass, and
the effect of exercise training on pain sensitivity in healthy, arm and thigh circumferences and skinfolds were recorded.
nonathlete individuals. Therefore, it is still largely unclear Pressure pain threshold, ischemic pain tolerance, pain ratings
whether chronic exercise can influence pain sensitivity in- during ischemia, and aerobic capacity (V̇O2peak) were then
dependently of athletic status. assessed as described in the following paragraphs.
The present study was designed to examine the effect of Pressure pain threshold was assessed for four muscular
moderate- to vigorous-intensity chronic aerobic exercise on sites (trapezius, biceps brachii, rectus femoris, and tibialis
pain sensitivity in healthy adults. Pressure and ischemic anterior). All measurements were made on the right side of
noxious stimuli were chosen as they are arguably the most the body and in the following rotational order: trapezius,
similar to the pain experienced during physical activity and biceps brachii, rectus femoris, and tibialis anterior. Three
chronic disease (33). It was hypothesized that, based on pre- practice trials were performed on the left trapezius muscle
vious studies, chronic aerobic exercise would increase pain before testing to familiarize the participant with the proce-
tolerance, but not affect pain threshold. dure. The rubber-tipped probe of the handheld algometer
(Wagner Force 10 FDX-25; Wagner Instruments, Greenwich,
CT) was applied perpendicularly to the participant’s skin, and
METHODS the force was increased gradually at a rate of approximately
Participants. Participants were recruited using adver- 1 kgIsj1. Participants were instructed to give a verbal com-
tisements placed around billboards on campus. Eligibility mand of ‘‘stop’’ when the sensation of pressure turned to pain.
criteria included 1) apparently healthy with no history of This procedure was repeated two more times, for a total of
chronic pain or chronic disease, 2) between the ages of 18 three measurements per site. Pressure pain threshold was
and 50 yr, and 3) absence of a current diagnosis of depres- recorded as the average of these three measurements. A pilot
sion. Twenty-seven participants (5 males and 22 females) study examining the reliability of this measure showed high
were recruited for this study. Throughout the 6-wk interven- within- and between-session intrarater reliability across all
tion, three participants withdrew because of injury unrelated four testing sites (ICC 9 0.9).
to the study, leaving a total of 24 participants who completed Ischemic pain tolerance was assessed via a modified sub-
the study (exercise: 1 male and 11 females, 24.4 T 4.3 yr; maximal ischemic tourniquet test. Participants grasped, with
control: 2 males and 10 females, 21.8 T 1.6 yr; P = 0.013). their dominant hand, a custom-built grip force device that
Procedures. This study was approved by the University was instrumented with a force transducer (Transducer Tech-
of New South Wales Human Research Ethics Committee. niques MLP-200). Force was sampled at 200 Hz with a 12-bit
Written informed consent was obtained from each participant analog-to-digital device (USB-6008; National Instruments,
before testing. Participants were recruited from the same Austin, TX) and stored in conjunction with the ratings of
university staff and student population and on the basis of pain and the target grip force profile. Custom software was
volunteering for either the exercise or the control group. That written (Labview version 9.0; National Instruments) to pro-
is, participants were not randomly assigned but were allo- vide visual feedback of the grip force and auditory tones
cated to either the exercise or control group based on their prompted the start and end of each contraction. After the
willingness to participate in either group. For the exercise determination of the participant’s maximal voluntary force,
group, the experiment consisted of 20 sessions, including an the sleeve of a standard sphygmomanometer was placed
initial assessment, 18 exercise sessions, and a final assess- around the participant’s upper arm, which was then exsan-
ment. Control subjects performed only the initial and final guinated by raising it above the level of the heart for 60 s.
assessment and were asked to maintain their regular level of The cuff was inflated to 200 mm Hg before the arm was
physical activity during the 6-wk period. Before the initial returned to horizontal. Prompted by the auditory tones and
and final assessments, participants were asked to abstain monitored by visual feedback, gripping exercise was per-
APPLIED SCIENCES
from vigorous exercise for 24 h and from caffeine for 4 h. formed at 30% maximal force for as long as tolerable (4-s
Compliance to these requests was confirmed verbally at the contraction and 4-s rest). Pain tolerance was the total time
start of the session. participants were able to sustain the handgrip exercise under
During their first and last visits and before assessments ischemic conditions. During testing, subjective ratings of
of pain sensitivity and aerobic capacity, participants com- pain were recorded using a 0–10 numeric pain rating scale
pleted several questionnaires to assess their psychological every 30 s (38). Participants were instructed to choose the
status and their physical activity levels. The Distress Risk number on the scale that corresponded to their level of pain,
and Assessment Method questionnaire, which is composed with 0 = ‘‘no pain’’ and 10 = ‘‘worst possible pain.’’ The
of the Zung Depression Index and the Modified Somatic experimenter was prepared to terminate the procedure if
Perceptions Questionnaire, was used to evaluate distress, de- the limit of pain tolerance was not reached by 10 min. This
pression, and somatization (22). The Profile of Mood States time limit was not made known to participants, who were

EXERCISE TRAINING AND PAIN TOLERANCE Medicine & Science in Sports & Exercised 1641

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 instructed only to continue the handgrip exercise for as RESULTS
long as tolerable. The results of a pilot study showed high
Maximal aerobic capacity (V̇O2peak). The V̇O2peak of
reliability for the pain tolerance measurement (ICC = 0.94).
participants is outlined in Table 1. A significant group–time
A V̇O2peak test was performed on a Monark 828e cycle
interaction was observed for V̇O2peak (F1,22 = 24.00, P G 0.001).
ergometer (Vansbro, Sweden) with use of an Ultima CPX
A significant difference in V̇O2peak between groups was
gas analysis system (Medgraphics, Minnesota USA). Par-
observed at baseline (t22 = 3.11, P = 0.02), but this differ-
ticipants were instructed to maintain a pedaling speed of
ence disappeared after the intervention (t22 = 0.24, P = 0.81).
70 rpm throughout the test. Exercise began with a 5-min
Exercise training caused a significant increase in V̇O2peak
warm-up at 35 W, after which the workload increased at a
(t11 = j5.39, P = 0.004, + 14.6%), whereas V̇O2peak was
rate of 35 W every 2 min until 105 W was reached. After
not significantly different at follow-up in the control group
this, workload increased at a rate of 35 W every 1 min until
(t11 = 1.45, P = 0.72, j2.8%).
V̇O2peak was obtained. Criteria used for the determination
Workload, HR, and RPE during V̇O2peak assessment. A
of V̇O2peak were as follows: no further increase in oxygen
significant group–time interaction was observed for work-
consumption despite an increase in workload, HR within
load (F1,22 = 11.5, P = 0.003) but not peak HR (F1,22 = 3.63,
T5 bpm of the participants age-predicted maximum HR, a
P = 0.07) or RPE (F1,22 = 0.77, P = 0.39) during V̇O2peak
respiratory exchange ratio 9 1.15, and volitional fatigue.
assessment. For participants in the exercise group, an increase
The exercise training consisted of cycle ergometer exer-
in peak workload was observed in the final compared with the
cise performed three times per week for 30 min at 75% of
initial V̇O2peak assessment (t11 = j2.66, P = 0.08, + 8.6%).
HR reserve. The age-predicted maximum HR was determined
Conversely, a decrease in peak workload was observed for
using the prediction equation HRmax = 207 j 0.7  age (12).
participants in the control group in the final V̇O2peak assessment
This workload was chosen to correspond to an intensity of
compared with the initial V̇O2peak assessment (t11 = 2.16, P =
75% V̇O2reserve, considered moderate–vigorous intensity
0.2, j6.7%). Peak HR and RPE were unchanged in both
(27,35). Each session began with a 5-min warm-up at 35 W
groups between each V̇O2peak assessment (P 9 0.5; Table 1).
and concluded with a 5-min cool down at 35 W. After the
Exercise training. Six of the exercise participants
warm-up, workload was adjusted to correspond to the in-
completed all 18 exercise sessions, whereas the other six
tensity that elicited an HR equivalent to 75% HR reserve.
completed 17 of the 18 sessions. For the exercise group,
Participants were then required to maintain this intensity for
there was a significant increase in the average exercise
30 min. Measurements of workload, HR, and RPE were re-
workload between the first and the last exercise session of
corded every 5 min throughout the exercise sessions. During
the intervention (t11 = j2.67, P = 0.02, + 9.4%), whereas
the exercise intervention, workload was adjusted as neces-
the average RPE during these sessions remained unchanged
sary to ensure participants maintained their target HR. Par-
(t11 = 1.46, P = 0.17) (Table 2). The average RPE across all
ticipants were required to complete a minimum of 17 exercise
exercise sessions was 15, which equates to a subjective rat-
sessions to be included in the study. A minimum of 2 d sep-
ing of ‘‘hard.’’
arated the final exercise session and the final assessment.
Ischemic pain tolerance. The duration of ischemic
Data processing and analysis. Pressure pain thresh-
contractions, and pain ratings during ischemia, for partici-
olds for the trapezius and biceps brachii sites and for the
pants in each group are shown in Figure 1. At follow-up, the
rectus femoris and tibialis anterior sites were combined to
give an average value for the upper and lower body, re-
spectively. Pain tolerance was the total time that participants
were able to sustain the handgrip exercise under ischemic TABLE 1. Duration and peak workload, HRmax, RPE, and RER during the maximal
conditions. Pain ratings during ischemia were analyzed in aerobic test before and after the intervention.

two ways: 1) the slope of the regression line was used to Before After
provide the rate of increase in pain rating (i.e., pain ratings Exercise
V̇O2peak (mLIminj1Ikgj1) 36.3 T 5.5*,** 41.6 T 6.4*
per second), and 2) the peak pain rating value was also used Duration (min) 6.4 T 1.4*** 7 T 1.5***
(peak pain rating). Linear regression analysis revealed that
APPLIED SCIENCES

Workload (W) 200.5 T 41.7 217.7 T 41.9

HR (bpm) 173.6 T 11.3 176.1 T 8.7
increases in pain rating were sufficiently linear so that the RPE 18.8 T 1.2 18.6 T 1.4
slope of the regression line was suitable to quantify the in- RER 1.43 T 0.13 1.33 T 0.8
crease in pain rating (mean r2 for each group, exercise: r2 = Control
V̇O2peak (mLIminj1Ikgj1) 42.9 T 4.9** 41.7 T 6.1
0.89, SD = 0.13; control: r2 = 0.91, SD = 0.08). Data were Duration (min) 7.2 T 1 7 T 1.3
analyzed using the Statistical Package for the Social Sciences Workload (W) 217.9 T 32.2 203.3 T 37.3
HR (bpm) 180.9 T 12.8 178.1 T 12.2
(version 20; SPSS Inc., Chicago, IL). A repeated-measures RPE 18.6 T 1.4 18.5 T 1.2
ANOVA was used to examine differences between groups RER 1.37 T 0.8 1.3 T 0.7
and across time. Bonferroni-adjusted paired-sample t-tests Data are presented as mean T SD. The duration of the maximal aerobic test does not
and independent-sample t-tests were also used post hoc to include the warm-up or cooldown.
*Significant increase across time within the exercise group, P = 0.004.
examine any within and between group differences, respec- **Significant difference between groups at baseline, P = 0.02.
tively. Significance was set at the > = 0.05 level. ***Significant increase across time within the exercise group, P = 0.04.

1642 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
TABLE 2. Workload, HR, and RPE during the first and last exercise training session for
participants in the exercise group.
First Session Final Session
Workload (W) 96.8 T 20.7* 105.9 T 23.9*
HR (bpm) 160.7 T 6.9 158.2 T 7.8
RPE 15.2 T 1.8 14.6 T 1.3
Data are presented as mean T SD.
*Significant difference between the first and last exercise session, P = 0.02.

duration of ischemic contractions was increased in 10 of

12 exercise participants and 5 of 12 control participants.
There was a significant group–time effect observed for is-
chemic pain tolerance (F1,22 = 8.4, P = 0.008). For the ex-
ercise group, there was a significant increase in ischemic
pain tolerance after training (t11 = j3.15, P = 0.036,

FIGURE 2—Group mean T SEM data for pain tolerance (s) and pain
ratings during ischemia for the exercise and control groups before and
after the intervention. A. Duration of ischemic contractions (s). B. The
rate of increase in pain rating during ischemia. C. Peak pain rating
during ischemia. *Significant difference, P G 0.05.

+20.3%), whereas ischemic pain tolerance was unchanged in

APPLIED SCIENCES
the control group across time (t11 = 1.77, P = 0.44, j3.7%)
(Fig. 2A). Ischemic pain tolerance was not significantly
different between groups at baseline (t22 = j0.92, P = 1).
No relationship was observed between the change in dura-
tion of ischemic contractions and the change in V̇O2peak for
either group (Fig. 3). However, when both groups were
FIGURE 1—Pain tolerance (s) and pain ratings during ischemia for
each participant in the exercise group before and after the interven- combined, there was a significant positive relationship be-
tion. The duration of ischemic handgrip exercise performed by each tween the change in V̇O2peak and the change in ischemic
participant is represented by lengths of bars. The dashed lines repre- pain tolerance (r2 = 0.21, P = 0.02; Fig. 3). Presumably, the
sent the average duration of ischemic contractions for all participants.
Ratings of pain throughout the exercise are represented by shading significant correlation for the combined data occurred mainly
(scale at top right). because of the way the groups were clustered. That is, an

EXERCISE TRAINING AND PAIN TOLERANCE Medicine & Science in Sports & Exercised 1643

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 FIGURE 3—The relationship between the change in V̇O2peak (mLIminj1Ikgj1) and the change in pain tolerance (s) for the exercise and control groups
and both groups combined. Correlation coefficients are reported separately for the exercise and control groups as well as both groups combined.

exercise group with significant changes in V̇O2peak and the exercise group only (t11 = j2.5, P = 0.12, + 48.5%). No
pain tolerance and a control group with minimal change in relationship was observed between the change in vigorous
these variables. physical activity level and the change in ischemic pain tol-
Pain ratings during ischemia. Pain ratings increased erance for either the exercise (r2 = 0.005, P = 0.82) or
progressively for all participants during the ischemic pain control group (r2 = 0.19, P = 0.15).
tolerance task (Fig. 1). There was no group effect (F1,22 = 0.14, Mood. Regarding the mood of participants on the days
P = 0.7) or group–time interaction (F1,22 = 1.67, P = 0.21) for of testing before and after the intervention, the Profile of
the rate of increase in pain during ischemia (Fig. 2B). There Mood States (POMS) showed no significant difference be-
was no group effect (F1,22 = 3.3, P = 0.08) or group–time tween the groups for any mood state (F1,22 = 3.4, P 9 0.08). A
interaction (F1,22 = 0, P = 1) on peak pain rating during is- significant effect of time on vigor was observed (F1,22 = 2.96,
chemia (Fig. 2C). P = 0.04). Vigor increased significantly in the control group at
Pressure pain threshold. There was no significant follow-up compared with baseline (t11 = j3.4, P = 0.024), but
group effect on pressure pain threshold for either the upper all other mood states remained unchanged across time in both
(F1,22 = 0.6, P = 0.45) or lower body (F1,22 = 2.9, P = 0.1).
There was no significant group–time effect on pressure
pain threshold for either the upper (F1,22 = 1.9, P = 0.18) or
lower body (F1,22 = 0.61, P = 0.44; see Fig. 4).
Physical activity. Aside from the exercise group com-
pleting the aerobic training intervention, participants were
instructed to maintain their usual levels of physical activ-
ity. Self-reported physical activity questionnaires identi-
fied that all participants adhered to this instruction. There
was a significant difference between groups in self-reported
walking before the intervention (exercise: 653.8 T 431
METIminIwkj1; control: 2477.8 T 2043.6 METIminIwkj1;
t11.98 = 3.02; P = 0.044). Self-reported moderate physical
activity was not different between the groups at baseline
(exercise: 986.5 T 1033 METIminIwkj1; control: 1718.7 T
APPLIED SCIENCES

1529 METIminIwkj1; t22 = 1.37; P = 0.73), or after the

intervention (exercise: 527 T 541 METIminIwkj1; control:
1792.6 T 1463 METIminIwkj1; t13.96 = 2.81; P = 0.056).
Similarly, self-reported vigorous physical activity was not
different between the groups at baseline (exercise: 939.5 T
848 METIminIwkj1; control: 1708.2 T 1642 METIminIwkj1;
t22 = 1.44; P = 0.66) or after the intervention (exercise: 1398.3 T
772 METIminIwkj1; control: 2028.2 T 1705 METIminIwkj1;
t22 = 1.17; P = 1). Vigorous physical activity level increased,
although not significantly, in the final week of the inter- FIGURE 4—Mean T SEM pressure pain threshold (PPT) for each
vention compared with baseline (F1,22 = 6.59, P = 0.02) in group before and after the intervention.

1644 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
groups (F1,22 = 3.94, P 9 0.24). A significant difference be- during exercise training. Notably, the subjective rating of the
tween groups was observed for the Modified Somatic Percep- ischemic pain was preserved, implying that a similar degree
tions Questionnaire at baseline (PRE—exercise: 4.08 T 2.64; of discomfort was tolerated for a longer duration.
control: 1.17 T 1.75; t19.08 = j3.19; P = 0.02) and after the The impact of chronic exercise on ischemic pain toler-
intervention (POST—exercise: 2.83 T 2.29; control: 0.83 T ance, but not pressure pain thresholds, was perhaps surprising
0.94; t14.59 = 2.8; P = 0.04). in light of many previous demonstrations that acute bouts of
exercise consistently raise pressure pain thresholds (18). The
mechanisms underlying hypoalgesia after exercise are unclear.
DISCUSSION
Several theories have been proposed to explain how acute
The results of the present study indicate that in healthy exercise reduces pain sensitivity. These include increases in
adults, 6 wk of moderate- to vigorous-intensity chronic aero- endogenous opioids, cannabinoids and stress hormones, con-
bic exercise increases tolerance to noxious ischemic stimuli. ditioned pain modulation (a form of endogenous pain inhibi-
In contrast, pressure pain threshold and pain ratings during tion in which pain in one location may inhibit pain in another),
ischemia were unchanged after training. The results suggest and changes in the attentional modulation of pain (i.e., dis-
that increases in pain tolerance may be one psychological traction) (10,30,39). However, these remain equivocal for
aspect of exercise adaptation, which has not been demon- acute exercise and their importance with chronic exercise
strated previously. This is the first study to clearly demon- remains largely unknown. Indeed, the lack of change of pres-
strate an effect of chronic exercise on pain sensitivity in sure pain thresholds in the current study suggests that acute
healthy, nonathlete adults. Previous cross-sectional studies and chronic exercise influence pain sensitivity through dif-
of athletes have alluded to this adaptation, but findings have ferent mechanisms.
been mixed and depended heavily on the context in which The aspect, rather than the modality, of pain sensitivity
pain was assessed. that was assessed may explain the different findings for pres-
The specific site and mechanisms of the observed training sure and ischemic pain in the current study. Pressure pain was
adaptations are difficult to ascertain and several possibilities measured only as the point at which the mechanical stim-
exist. In general terms, physiological adaptations may have ulus became painful, whereas the assessment of ischemic pain
occurred that resulted in diminished signaling in response to concerned the capacity of the individual to tolerate a stim-
the noxious stimulus. Alternatively, psychological adapta- ulus that was above the threshold for pain. Pain threshold is
tions may have occurred that simply permitted a greater thought to predominantly reflect muscle nociception (32),
tolerance of a similar level of discomfort, with activity pre- whereas pain tolerance additionally involves a strong psy-
served through the pain pathways. Pressure and ischemic chosocial and behavioral component (6). Although the cur-
pain are conveyed by different afferents (mechanosensitive rent study reports an increase in tolerance to ischemic pain,
and chemosensitive afferents, respectively) (13). The current Anshel and Russell (5) reported that tolerance to pressure
study did not directly test the activity of either the mechano- pain increased in healthy adults after aerobic training. How-
sensitive or chemosensitive nociceptors. Pressure pain thresh- ever, they assessed pain tolerance as the peak pressure that
olds were unchanged in both the upper and the lower body, could be endured rather than the duration for which pain
although these two areas would have experienced different could be tolerated. In addition, cross-sectional comparisons
peripheral adaptations to cycle exercise training. Thus, it is of pain sensitivity between athletes and nonathletes indicate
unlikely that mechanoreceptor firing was altered. that athletes tolerate more pain when exposed to a range of
We also propose that during the tourniquet test, nocicep- noxious stimuli, whereas pain threshold usually does not
tor activity in the arm was the same pre- and posttraining, as differ between the groups (36). This effect is mediated by
this limb was untrained. Moreover, the arm was occluded personality traits, coping strategies and a higher level of pain
from central circulatory influence during testing so that any self-efficacy (17,28). An early study that compared thresh-
cardiovascular training adaptations (e.g., increased cardiac olds and tolerances for different pain modalities in athletes
output and associated delivery of nutrients to exercising mus- and nonathletes found strong correlations between tolerance
cle) were unlikely to improve during performance of the task. across modalities (31). On the other hand, in healthy individ-
APPLIED SCIENCES
Therefore, changes in muscle nociceptor afferent signals uals, when fitness is not taken into account, threshold and
originating at the periphery are unlikely to account for the tolerance measures within pain modalities are more closely re-
increased ischemic pain tolerance or the different response lated than threshold measures or tolerance measures across mo-
of participants to ischemic but not pressure pain after train- dalities including ischemic, pressure and thermal pain (7,14).
ing. This result provides evidence for a central mechanism There were several limitations to the present study. Although
as the primary modulator of the increased pain tolerance recruitment of participants was from the same university pop-
and suggests a new psychological adaptation to training. The ulation of staff and students and conducted concurrently, the
afferents activated during the ischemic task are similar to allocation of participants to the groups was not randomized.
those activated during exercise (19).Therefore, it is possible Furthermore, participants in the exercise group received more
that prolonged performance of the ischemic task was facili- attention than participants in the control group as they were
tated by repeated exposure to the noxious ischemic stimulus supervised for all 18 sessions, whereas control participants

EXERCISE TRAINING AND PAIN TOLERANCE Medicine & Science in Sports & Exercised 1645

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 only met with the investigator for the initial and final assess- (3). However, these same muscle afferents can also minimize
ment. Therefore, behavioral artifacts cannot be discounted. locomotor muscle fatigue by stimulating ventilator and car-
Second, there was a significant difference in self-reported diovascular response to rhythmic exercise (2). Therefore, ex-
physical activity and V̇O2peak between the groups at baseline, ercise training may facilitate the development of brain function
which may have influenced the results. For example, higher that increases tolerance of these signals and associated sen-
levels of physical activity, particularly vigorous activity, are sations, and this increase in tolerance may contribute to im-
associated with reduced pain sensitivity (11,26). However, proved endurance performance.
pain sensitivity was not different between the groups at base- Our results also provide evidence of a systemic hypoalgesia
line. Moreover, there was no correlation between V̇O2peak after exercise training, whereby pain tolerance increased in the
and pain sensitivity for either group. Therefore, it is unlikely arm after 6 wk of training with the legs. This is consistent with
that the initial difference in aerobic capacity between the reductions in pain sensitivity in nonexercising limbs in healthy
groups at baseline influenced the results. adults and patients with peripheral arterial disease (5,40). This
Lastly, based on the Modified Somatic Perceptions Ques- finding may have important clinical applications for exer-
tionnaire, the exercise group reported higher somatization cise prescription in patients with persistent pain. For instance,
compared with the control group both before and after the patients with persistent pain may gain a pain relieving benefit
intervention. Somatization is the tendency to experience and of exercise by training with unaffected or pain free limbs. This
communicate somatic symptoms in response to psychological would serve to improve their functional capacity and clinical
stress (e.g., an increased HR and feeling hot all over) (20). outcomes, without the risk of exacerbating their symptoms. A
This difference could mean that the exercise group had greater transfer of endurance training to untrained limbs has previ-
attention to the painful stimuli before training and greater ously been shown after exercise training. That is, exercise
scope to change with training. However, the scores for both training with the lower body can improve V̇O2peak and other
groups were very low before and after the intervention (ex- cardiovascular parameters when subsequent exercise is per-
ercise group: 4.1 T 2.6 and 2.8 T 2.3, respectively; control formed solely with the upper body (37). This same transfer
group: 1.1 T 0.8 and 0.8 T 0.9, respectively); out of a possible effect may also apply to pain sensitivity.
score of 39 and from a clinical perspective, a score of 12 or To conclude, the results from this study demonstrated that
more is considered necessary to influence pain sensitivity 6 wk of moderate- to vigorous-intensity aerobic exercise
(22). Moreover, Main (21) found no relation between heightened training increased pain tolerance in healthy individuals. This
somatic awareness (i.e., Modified Somatic Perceptions Ques- demonstration that exercise may influence pain sensitivity
tionnaire score) and ischemic pain tolerance. Therefore, the be- independently of disease provides new insight into how some
tween group difference is unlikely to have influenced the results. clinical populations with low exercise tolerance and capacity
One implication of our results is that increasing pain tol- may benefit from aerobic training. That is, increasing pain
erance may contribute to enhanced endurance performance tolerance in these patients through regular aerobic training
via a greater tolerance of afferent feedback associated with may facilitate more exercise as well as exercise at a higher
metabolic disturbance in muscles. Despite the conjecture that intensity, which may provide greater clinical benefits.
surrounds the influence of signals from muscle afferents on
endurance performance (4,24), it is generally accepted that
they are important (1). Discharge of small-diameter muscle
afferents, including nociceptors, increases in the presence of The results of this study do not constitute endorsement by the
American College of Sports Medicine.
metabolites associated with muscle fatigue and this feedback There was no funding received for this study. There were no
inhibits central neural drive and subsequently performance conflicts of interest during this study for any of the authors.

REFERENCES
1. Amann M. Significance of group III and IV muscle afferents for 5. Anshel MH, Russell KG. Effect of aerobic and strength
the endurance exercising human. Clin Exp Pharmacol Physiol. training on pain tolerance, pain appraisal and mood of unfit
APPLIED SCIENCES

2012;39(9):831–5. males as a function of pain location. J Sports Sci. 1994;12(6):

2. Amann M, Blain GM, Proctor LT, Sebranek JJ, Pegelow DF, 535–47.
Dempsey JA. Group III and IV muscle afferents contribute to 6. Baker SL, Kirsch I. Cognitive mediators of pain perception and
ventilatory and cardiovascular response to rhythmic exercise in tolerance. J Pers Soc Psychol. 1991;61(3):504–10.
humans. J Appl Physiol. 2010;109(4):966–76. 7. Bhalang K, Sigurdsson A, Slade GD, Maixner W. Associations
3. Amann M, Proctor LT, Sebranek JJ, Eldridge MW, Pegelow DF, among four modalities of experimental pain in women. J Pain.
Dempsey JA. Somatosensory feedback from the limbs exerts in- 2005;6(9):604–11.
hibitory influences on central neural drive during whole body en- 8. Busch AJ, Barber KA, Overend TJ, Peloso PM, Schacter CL.
durance exercise. J Appl Physiol. 2008;105(6):1714–24. Exercise for treating fibromyalgia syndrome. Cochrane Database
4. Amann M, Secher NH. Point: afferent feedback from fatigued lo- Syst Rev. 2007;17(4):CD003786.
comotor muscles is an important determinant of endurance exer- 9. Craig CL, Marshall AL, Sjostrom M, et al. International physical
cise performance. J Appl Physiol. 2010;108(2):452–4; discussion activity questionnaire: 12-country reliability and validity. Med Sci
7; author reply 70. Sports Exerc. 2003;35(8):1381–95.

1646 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
10. Ellingson L, Cook D. Exercise induces hypoalgesia through con- 25. Naugle KM, Fillingim RB, Riley JL 3rd. A meta-analytic review of
ditioned pain modulation. J Pain. 2011;12(4 Suppl):37. the hypoalgesic effects of exercise. J Pain. 2012;13(12):1139–50.
11. Ellingson LD, Colbert LH, Cook DB. Physical activity is related 26. Naugle KM, Riley JL 3rd. Self-reported physical activity pre-
to pain sensitivity in healthy women. Med Sci Sports Exerc. 2012; dicts pain inhibitory and facilitatory function. Med Sci Sports Exerc.
44(7):1401–6. 2014;46(3):622–9.
12. Gellish RL, Goslin BR, Olson RE, McDonald A, Russi GD, 27. Norton K, Norton L, Sadgrove D. Position statement on physical
Moudgil VK. Longitudinal modeling of the relationship between activity and exercise intensity terminology. J Sci Med Sport. 2010;13(5):
age and maximal heart rate. Med Sci Sports Exerc. 2007;39(5): 496–502.
822–9. 28. Ord P, Gijsbers K. Pain thresholds and tolerances of competitive
13. Giordano J. The neurobiology of nociceptive and anti-nociceptive rowers and their use of spontaneous self-generated pain-coping
systems. Pain Physician. 2005;8:277–90. strategies. Percept Mot Skills. 2003;97(3 Pt 2):1219–22.
14. Hastie BA, Riley JL 3rd, Robinson ME, et al. Cluster analysis of 29. Pollock V, Cho DW, Reker D, Volavka J. Profile of Mood States:
multiple experimental pain modalities. Pain. 2005;116(3):227–37. the factors and their physiological correlates. J Nerv Ment Dis.
15. Hayden JA, van Tulder MW, Malmivaara A, Koes BW. Exercise 1979;167(10):612–4.
therapy for treatment of non-specific low back pain. Cochrane 30. Ruble SB, Hoffman MD, Shepanski MA, Valic Z, Buckwalter JB,
Database Syst Rev. 2005;(3):CD000335. doi: 10.1002/14651858. Clifford PS. Thermal pain perception after aerobic exercise. Arch
CD00035. Phys Med Rehabil. 2005;86(5):1019–23.
16. Hoffman MD, Shepanski MA, Ruble SB, Valic Z, Buckwalter JB, 31. Ryan DE, Kovacic CR. Pain tolerance and athletic participation.
Clifford PS. Intensity and duration threshold for aerobic exercise- Percept Mot Skills. 1966;22(2):383–90.
induced analgesia to pressure pain. Arch Phys Med Rehabil. 32. Scholz J, Woolf CJ. Can we conquer pain? Nat Neurosci.
2004;85(7):1183–7. 2002;(5 suppl):1062–7.
17. Johnson MH, Stewart J, Humphries SA, Chamove AS. Marathon 33. Scott V, Gijsbers K. Pain perception in competitive swimmers.
runners’ reaction to potassium iontophoretic experimental pain: pain Br Med J (Clin Res Ed). 1981;283(6284):91–3.
tolerance, pain threshold, coping and self-efficacy. Eur J Pain. 2012; 34. Sternberg WF, Bailin D, Grant M, Gracely RH. Competition alters
16(5):767–74. the perception of noxious stimuli in male and female athletes.
18. Koltyn KF. Exercise-induced hypoalgesia and intensity of exer- Pain. 1998;76(1-2):231–8.
cise. Sports Med. 2002;32(8):477–87. 35. Swain DP, Leutholtz BC. Heart rate reserve is equivalent to
19. Light AR, Hughen RW, Zhang J, Rainier J, Liu Z, Lee J. Dorsal %V̇O2 reserve, not to %V̇O2max. Med Sci Sports Exerc. 1997;29(3):
root ganglion neurons innervating skeletal muscle respond to physi- 410–4.
ological combinations of protons, ATP, and lactate mediated by 36. Tesarz J, Schuster AK, Hartmann M, Gerhardt A, Eich W. Pain
ASIC, P2X, and TRPV1. J Neurophysiol. 2008;100(3):1184–201. perception in athletes compared to normally active controls: a
20. Lipowski ZJ. Somatization: the concept and its clinical application. systematic review with meta-analysis. Pain. 2012;153(6):1253–62.
Am J Psychiatry. 1988;145(11):1358–68. 37. Tordi N, Belli A, Mougin F, Rouillon JD, Gimenez M. Specific
21. Main CJ. The Modified Somatic Perception Questionnaire (MSPQ). and transfer effects induced by arm or leg training. Int J Sports
J Psychosom Res. 1983;27(6):503–14. Med. 2001;22(7):517–24.
22. Main CJ, Wood PL, Hollis S, Spanswick CC, Waddell G. The 38. Williamson A, Hoggart B. Pain: a review of three commonly used
Distress and Risk Assessment Method. A simple patient classifi- pain rating scales. J Clin Nurs. 2005;14(7):798–804.
cation to identify distress and evaluate the risk of poor outcome. 39. Yarnitsky D, Arendt-Nielsen L, Bouhassira D, et al. Recommen-
Spine (Phila Pa 1976). 1992;17(1):42–52. dations on terminology and practice of psychophysical DNIC
23. Manning EL, Fillingim RB. The influence of athletic status and testing. Eur J Pain. 2010;14(4):339.
gender on experimental pain responses. J Pain. 2002;3(6):421–8. 40. Zwierska I, Walker RD, Choksy SA, Male JS, Pockley AG, Saxton
24. Marcora S. Counterpoint: afferent feedback from fatigued locomotor JM. Upper- vs lower-limb aerobic exercise rehabilitation in pa-
muscles is not an important determinant of endurance exercise per- tients with symptomatic peripheral arterial disease: a randomized
formance. J Appl Physiol. 2010;108(2):454–6; discussion 6–7. controlled trial. J Vasc Surg. 2005;42(6):1122–30.

APPLIED SCIENCES

EXERCISE TRAINING AND PAIN TOLERANCE Medicine & Science in Sports & Exercised 1647

Copyright © 2014 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.