Conditioned pain modulation (CPM) as a biomarker of fibromyalgia and pain severity: effect of the CPM paradigm used

Abstract

Fibromyalgia (FM) is a widespread chronic pain syndrome, possibly associated with the presence of central dysfunction in descending pain inhibition pathways. Conditioned Pain Modulation (CPM) has been proposed as a biomarker of FM. Nonetheless, the wide variety of methods used to measure CPM has hampered robust conclusions being reached.
To clarify the validity of CPM as a biomarker of FM, we tested two CPM paradigms (parallel and sequential) in a sample of 23 female patients and 23 healthy women by applying test (mechanical) stimuli and conditioning (pressure cuff) stimuli. We evaluated whether CPM indices could correctly classify patients and controls, and we also determined the correlations between the indices and clinical variables such as symptomatology, impact, depression, quality of life, pain intensity, pain interference, fatigue and numbness. In addition, we compared the clinical status of CPM responders (efficient pain inhibitory mechanism) and non-responders.
We observed that only parallel CPM testing correctly classified about 70% of patients with FM. In addition, more than 80% of healthy participants were found to be responders, while the rate was about 50% in the FM patients. The sequential CPM test was not as sensitive, with a decrease of up to 40% in the response rate for both groups. On the other hand, although we did not observe any correlation between CPM measures and clinical symptoms, we found that patients with impaired CPM response displayed more severe symptoms, though less generalized pain. In summary, our findings demonstrate the influence of the CPM paradigm used and confirm that CPM may be a useful marker to complement FM diagnosis. However, the findings also cast doubts on the sensitivity of CPM as a marker of pain severity in FM.

Full text

Page 1/19
Conditioned pain modulation (CPM) as a biomarker
of bromyalgia and pain severity: effect of the CPM
paradigm used
Antonio Gil-Ugidos (  antonio.gil.ugidos@usc.es )
University of Santiago de Compostela
A. Vázquez-Millán
University of Santiago de Compostela
N. Samartin-Veiga
University of Santiago de Compostela
M. T. Carrillo-de-la-Peña
University of Santiago de Compostela
Research Article
Keywords: conditioned pain modulation, bromyalgia, chronic pain
Posted Date: January 18th, 2024
DOI: https://doi.org/10.21203/rs.3.rs-3866485/v1
License:   This work is licensed under a Creative Commons Attribution 4.0 International License. 
Read Full License
Additional Declarations: No competing interests reported.

Page 2/19
Abstract
Fibromyalgia (FM) is a widespread chronic pain syndrome, possibly associated with the presence of
central dysfunction in descending pain inhibition pathways. Conditioned Pain Modulation (CPM) has
been proposed as a biomarker of FM. Nonetheless, the wide variety of methods used to measure CPM
has hampered robust conclusions being reached.
To clarify the validity of CPM as a biomarker of FM, we tested two CPM paradigms (parallel and
sequential) in a sample of 23 female patients and 23 healthy women by applying test (mechanical)
stimuli and conditioning (pressure cuff) stimuli. We evaluated whether CPM indices could correctly
classify patients and controls, and we also determined the correlations between the indices and clinical
variables such as symptomatology, impact, depression, quality of life, pain intensity, pain interference,
fatigue and numbness. In addition, we compared the clinical status of CPM responders (ecient pain
inhibitory mechanism) and non-responders.
We observed that only parallel CPM testing correctly classied about 70% of patients with FM. In
addition, more than 80% of healthy participants were found to be responders, while the rate was about
50% in the FM patients. The sequential CPM test was not as sensitive, with a decrease of up to 40% in the
response rate for both groups. On the other hand, although we did not observe any correlation between
CPM measures and clinical symptoms, we found that patients with impaired CPM response displayed
more severe symptoms, though less generalized pain. In summary, our ndings demonstrate the inuence
of the CPM paradigm used and conrm that CPM may be a useful marker to complement FM diagnosis.
However, the ndings also cast doubts on the sensitivity of CPM as a marker of pain severity in FM.
Background
Fibromyalgia (FM) is dened as a widespread chronic pain syndrome associated with fatigue, sleep
disturbance, cognitive impairment, affective disturbance and somatic symptoms (1). Although the
aetiology of FM remains unknown, the presence of central dysfunction in descending pain inhibition
pathways is currently the best-supported hypothesis regarding the pathophysiology of the condition (2).
One of these abnormalities refers to a deciency in conditioned pain modulation (CPM; 3). CPM occurs
when the presence of a second noxious stimulus (i.e. a conditioning stimulus; CS) produces a decrease in
the perceived pain evoked by a given stimulus (i.e. test stimulus; TS), applied in a contralateral area (4).
This phenomenon implies that the processing of pain signals inhibits the nociceptive input originating
from a heterotopic region; decits in the CPM mechanism could thus lead to an enhanced sensation of
pain. Considered a spinal-level mediated process involving cortical regions and brainstem structures such
as the anterior cingulate cortex and the periacueductal gray (5), CPM seems to be partly mediated by
diffuse noxious inhibitory controls (DNIC; 6), as demonstrated in animal studies (7; 8). Thus, descending
pain-inhibitory networks play a modulatory role in CPM (9), and disruptions in this circuitry may be
associated with increased pain sensitivity (10).

Page 3/19
Impairments in CPM are present in several chronic pain states (11; 12), and they are particularly relevant
and consistent in patients with FM (13; 14; 15). Thus, CPM has been suggested as a potential biomarker
to distinguish clinical proles with different disease prognosis and symptom severity in chronic pain
diseases (16). For instance, Gerhardt et al. (2017) reported a signicant negative correlation between
CPM measures and pain intensity during the previous month in a sample of FM patients; other authors
found that FM patients with defective CPM reported poorer sleep quality and showed greater impairment
in sustained attention (17; 18). However, a recent review brings into question the validity of CPM as a
biomarker of clinical pain (19), given that the decits in CPM are not always correlated with clinical
manifestations of pain (pain intensity, disability due to pain, pain duration or number of painful areas).
The disparity of results may be due to the heterogeneity of CPM protocols: the characteristics of the
various stimuli used (e.g. CS such as cold water, tourniquets and hot water, and TS such as pressure,
thermal, mechanical and electrical stimuli); the body area stimulated; the nature of stimulation (both
painful and non-painful); and the mode of presentation of CS, which can be applied simultaneously or
subsequently to the TS (19). Previous research along this line has suggested that some methodological
factors could inuence the magnitude of the observed CPM response (20), although the data are variable.
For instance, the role of the moment of application of the CS stimulus on CPM is unclear: in some
studies, the CPM response was better when the TS and CS were presented simultaneously than when
they were presented sequentially (20), while others did not nd any signicant differences in relation to
the presentation (sequential or parallel) (21).
Bearing this in mind, the aim of the present study was to test the validity of CPM as a diagnostic
biomarker of FM, using two paradigms which differed in CS was presented either during (parallel CPM) or
after the TS (sequential CPM). To this end, we compared the ability of both CPM paradigms to classify
the participants as FM patients or healthy controls. As a second objective, we investigated whether CPM
could be used as a biomarker of pain severity in FM, correlating CPM indices with clinical variables and
comparing the clinical status of CPM responders (ecient pain inhibitory mechanism) and of non-
responders.
Methods
PARTICIPANTS
The sample comprised 23 female FM patients and 23healthyfemalecontrols. The inclusion criteria for
the group of FM patients wereas follows: (1)age more than 18 years; (2)diagnosis of FM by a physician
or rheumatologist;and(4)fullmentof the FM diagnosis criteria (1), i.e.aWidespread Pain Index (WPI)
score of 7 or higher and a Symptom Severity Scale (SSS) score of 5 or higher or a WPI score between 3
and 6 and a SSS score of 9 or higher. Exclusion criteria were history of drug abuse and psychiatric
disorders (other than depression and anxiety).
CPM stimuli and procedure

Page 4/19
Test Stimulus (TS):thePressure Pain Threshold (PPT) was obtained using an algometer (Wagner Force
One, Model FDI).PPTwas selected as the TS because pressure stimuliare considered more reliable
thanother stimulisuch asheat pain thresholds(22). Moreover, the CPM effectisbest interpreted using
pain thresholdsandstimuli of predened intensityas TS(23).Theareaof stimulation, applied by a
previously trained researcher, was a 1 cm2 patch on the dominant forearm, over the extensor carpi
radialis longus, and the pressure velocity (rate) was 35 kPa/s. The stimulation was delivered 3 times,
separated by 20 seconds, and the mean value was considered the PPT.
Conditioningstimulus (CS)was delivered by a pressure cuff on the opposite arm, with a constant
pressure of 240 mm hg (30 kPa), applied for 120 seconds. Previous data supports that using the
contralateral dermatome provides the most reliable results (24).
Two paradigms were delivered for all the participants, in a counterbalanced order (see Figure 1): 
Sequential paradigm: The TS (pressure algometer) was delivered before the CS (pressure cuff) and after
the CS (mean of 3 PPT measurements separated by 20 seconds).
Parallel paradigm: The TS was assessed before and during the application of the CS (pressure cuff). 
Calculation of the CPM effect:We calculated the CPM effect for each experimental paradigm by
subtracting the mean PPT for TS before and after/during presentation of the CS (either sequential or
parallel). Negative values (CPM < 0) indicated an elevation of the threshold, i.e. an inhibitory response
due to application of the CS. 
QUESTIONAIRE MEASUREMENT
Both FM patients and healthy participants completed different validated Spanish versions of tests and
questionnaires of interest to the study:
The Fibromyalgia Survey Questionnaire (FSQ; 25) was used to assess FM symptoms, to check the
inclusion criteria for the patients and characterize the groups. The FSQ is based on the diagnostic criteria
proposed by Wolfe et al. (2010) and includes the Symptom Severity Scale (SSS) and the Widespread Pain
Index(WPI). The SSS considers three key symptoms (fatigue; cognitive problems in attention,
concentration or memory; and non-restorative sleep), assessed on a scale of 0–3 (0 = not present to 3 =
extreme). In addition, the SSS assesses abdominal pain, depression and headache, determined as
present (1) or not present (0). The SSS score ranges from 0 to 12. The WPI score indicates the number of
body areas with pain reported by the patient (from 0 to 19).
The Beck Depression Inventory (BDI-1A; 26) was used to evaluate the severity of depressive symptoms.
The BDI-1A is composed by 21 items representative of symptoms such as sadness, feelings of failure,
pessimism, suicidal desire, etc. Each item is answered by the participants on a 4-point scale, ranging from
0 to 3. The total score (range from 0 to 63) was recorded. Higher scores are associated with greater

Page 5/19
severity of depressive symptoms, and patients can be classied as having nodepression (0–13), mild
depression (14–19), moderate depression (20–28) or severe depression (29–63).
The Revised Fibromyalgia Impact Questionnaire (FIQ-R; 27) was administered to measure the functional
disability and health status of patients with FM. The FIQ-R is a self-reporting questionnaire including21
items scored on an 11-point numerical rating scale of 0–10, with 10 being “worst.” The scores are
calculated for three domains: function (from 0 to 30), general impact (from 0 to 20) and symptoms (from
0 to 50). A total score (range from 0 to 100) was also considered. Higher scores are associated with
greater disease severity and functional impact, and patients are classied as havingmild (0–38),
moderate (39–58) or severe (59–100) symptoms.
The 36-item Short Form Health Survey (SF-36; 28) assesses Quality of Life (QoL) and provides a prole of
health status and function. It is composed of 36 items distributed across 8 scales: physical function,
physical role, body pain, general health, vitality, social function, emotional role and mental health (i.e. the
most relevant health concepts included in the Medical Outcomes Study (MOS)). The scores on each
subscale range from 0 to 100 (0 represents the worst possible health level and 100, the best). In this
study, we calculated the score for the eight subscales and a mean score for the SF-36.
In addition, participants completed different ad hoc Visual Analogue Scales (VAS)related to their status
in the last week (pain intensity, numbness, fatigue), in the last month (pain intensity, interference due to
pain, depressive state) and the level of deterioration in their health.
Data analysis
The pretest clinical status of the clinical and control groups wasrstcompared. As none of the variables
studied (except age) met the criteria for normality in both groups (based on Kolmogorov-Smirnov and
Shapiro-Wilk test), non-parametric tests wereused to calculate differences in the meanvalues(Man
Whitney U test).
Repeated-measures ANOVA was then used to compare the CPM results obtained by the groups.
Paradigm (levels: sequential and parallel) was used as an intra-subject factor and Group (levels: FM
patients and healthy controls) as an inter-subject factor.
The patients were then classied as responders or non-responders by both paradigms. For a patient to be
considered a CPM responder, the difference between the PPT to the TS before and after the CS had to be
greater than the standard deviation for the mean of the group (CPM > SD). The number of responders for
each group in each of the paradigms was then compared using chi-square tests.
Binary logistic regression analysis was used to assess the sensitivity and specicity of the parallel CPM
and sequential CPM paradigms to correctly classify the participants as FM patients or healthy controls.
Finally, considering allofthe clinical variables, we performed non-parametric tests (Man Whitney U)to
check possible differences in the clinical prole of FM patients who were CPM respondersand those who

Page 6/19
wereCPM non-responders.
Data were analyzed using the SPSS statistical package (v.24.0; IBM Corporation, Armonk, NY, United
States).
RESULTS
Sample characteristics and baseline scores
Participants in both groups were matched by age (patients: mean 49.78 years (±8.67); healthy controls:
mean 50.22 years (±12.57);
t
=0.137 (44);
p
=.892) and menstrual phase (
X2
=1.575;
p
=.665).
The
Man Whitney
U test results showed that patients with FM obtained lower scores than the healthy
controls for all the clinical variables. Thus, patients with FM had a higher rate of depressive symptoms,
poorer quality of life, fatigue state and numbness, greater interference in their daily life due to pain and
greater pain intensity, both in the previous week and in the previous month (see Table 1). The PPTs were
similar in both groups.
Table 1: Comparison between groups in the variablesassessed.

Page 7/19
 FM
patients
Mean (SD)
Healthy controls
Mean (SD)
Man
Whitney U
Sig.
BDI 15.52 (±
6.74) 4.22 (±6.05) -4.889
p
<
.001
FIQ-R 47.71 (±
24.14) 5.95 (±5.17) -5.636
p
<
.001
SSS 7.57 (±
1.97) 2.87 (± 1.46) -5.621
p
<
.001
WPI 9.26 (4.63) 2.35 (± 1.64) -4.816
p
<
.001
SF-36 48.07
(±15.37) 75.73 (±9.06) -5.042
p
<
.001
Intensity of fatigue-previous week
(VAS) 6.7 (±1.94) 2.26 (±1.42) -5.310
p
<
.001
Intensity of numbness - previous
week (VAS) 6.22
(±3.72) 0.35 (±0.57) -5.150
p
<
.001
Intensity of pain- previous week
(VAS) 5.43 (±
2.84) 1.87 (± 1,01) -4.344
p
<
.001
Intensity of pain- previous month
(VAS) 5.65
(±2.69) 2.22 (±1.09) -4.413
p
<
.001
Pain pressure threshold (PPT) 1.27
(±0.68) 1.27 (±0.26) -0.616
p
=
.538
BDI: Beck Depression Inventory; FIQ-R: Fibromyalgia Impact Questionnaire -revised; SSS:
Symptom
Severity Scale; WPI: Widespread Pain Index; SF36: Short Form Health Survey; VAS: Visual Analogue Scale

CPM effect: modulation byparadigm
The repeated measures ANOVA showed a signicant main effect of Paradigm (F(1,44)=29.513; p<.001),
as well as a Paradigm x Group interaction (F(1,44)=6.343; p=.015). Post-hoc tests (with Bonferroni
correction) showed that the parallel CPM produced more effective inhibitory responses (i.e. higher
thresholds to the second application of the TS; Parallel CPM=-0,58; Sequential CPM=-0,01; p<.001), while
the differences between the FM patients and healthy controls were only signicant for the parallel CPM
(FM patients=-0.15; Healthy controls: -0.45; p=.004). 
Distribution of CPM responders and non-responders in each group
The participants were classied as CPM responders and CPM non-responders (see Table 2). Application
of the sequential paradigm yielded a similar number of responders in the patients and control group

Page 8/19
(according to the chi-square test results). However, application of the parallel paradigm yielded a
signicantly greater number of healthy controls with an adequate CPM response than of FM patients.
Table 2: Distribution of responders and non-responders in the different paradigms and groups
Sequential CPM Parallel CPM
CPM
Responders CPM non-
responders CPM
Responders CPM Non -
responders
FM patients
(N=23)
7
(30.43%)
16
(69.67%)
13
(56.52%)
10
(43.48%)
Healthy
controls
(N=23)
9
(39.13%)
14
(61.87%)
20
(86.96%)
3
(13.04%)
Total 16 30 33 13

X2
= 0.096 (
p
=0.757)
X2
= 3.860 (
p
=0.049*)
CPM: Conditioned Pain Modulation. Participants were classied as responders if the magnitude of the
CPM (difference of PPT before vs. after the CS) was greater than the standard deviation for the mean of
the group (CPM > SD).
Accuracy of CPMforclassifiyinghealthy controlsandFM patients.

In the binary logistic regression model including the effect of the parallel CPM paradigm as the predictor
variable, the Omnibus test yielded
X2=
9.569 (
p
=0.002) so thatinclusionof this variablecontributesto
explaining the group to which each participant belongs. Given this effect, the model correctly classied
69.6% of the participants as patients or controls(seeTable 3).TheNagelkerkeR2index suggests a
moderate relationship between the predictor and the outcome. The sequential CPM did notproduce a
signicanteffect(
p
=0.707).
Table 3:

Binary logistic regression model results using the parallel CPM paradigm for predicting and
classifying participants.
Predicted
FM patients Healthy
controls
Correctly classied

Observed
FM patients 16 7 69.6%
Healthy controls 7 16 69.6%

Page 9/19
Exp(B) Sig. Cox and Snell R2Nagelkerke R2
0.203
p
= 0.006** 0,188 0.250
Differences in the clinical prole of CPM responder and non-responder patients.
For the sample of patients, non-parametric tests revealed signicant differences between CPM responder
and CPM non-responder patients in indices of disease severity, assessed by the FIQ-R: patients who
exhibited an ecient CPM response had less severe symptoms (only using data from the parallel
paradigm). A signicant difference in the WPI index (body extent of pain) was also found, although in the
opposite direction: CPM responder patients showed more generalized pain throughout the body than their
non-responsive counterparts (see Table 4). There were no signicant differences in the other clinical
variables.
Table 4: Clinical state differences between parallel-CPM respondersandnon-responders (only data for the
parallel paradigmare shown).
 CPM
Responder
(N=13)
CPM
Non responder
(N=10)
Mann-Whitney U
test
Sig.
BDI 13.85
(±6.19) 17.70 (±7.12) -1.123 .261
FIQ-R 38.12
(±18.03) 60.18 (±26.18) -1.985 .047
SSS 6.85
(±1.41) 8.50 (±2.32) -1.766 .077
WPI 11.08
(±4.15) 6.90 (±4.31) -2.030 .042
SF-36 52.26
(±13.21) 42.64 (±16.93) -1.488 .137
Intensity of fatigue-previous week
(VAS) 6.23
(±1.69) 7.30 (±2.16) -1.923 .055
Intensity of numbness - previous
week (VAS) 6.08
(±3.57) 6.40 (±4.09) -0.317 .751
Intensity of pain- previous week
(VAS) 4.77
(±2.49) 6.30 (±3.16) -1.467 .142
Intensity of pain- previous month
(VAS) 5.15
(±1.19) 6.30 (±3.23) -1.407 .159
Pain pressure threshold (PPT) 1.22
(±0.86) 1.33 (±0.38) -1.430 .153

Page 10/19
BDI: Beck Depression Inventory; FIQ-R: Fibromyalgia Impact Questionnaire -revised; SSS:
Symptom
Severity Scale; WPI: Widespread Pain Index; SF36: Short Form Health Survey; VAS: Visual Analogue Scale

Correlation between CPM effect and clinical variables
Analysis of the correlation between the clinical variables evaluated and the sequential or parallel CPM
indices did not indicate the presence of any signicant correlation (see Table 5).
Table 5: Correlation between clinical variables and CPM results in the clinical group (n=23).
 Sequential CPM Parallel CPM
Pearson r
Coecient Sig. Pearson r
Coecient Sig.
BDI -0.294 0.173 0.176 0.423
FIQ-R 0.076 0.730 0.217 0.320
SSS -0.135 0.539 0.161 0.462
WPI 0.137 0.534 -0.290 0.179
SF36 -0.102 0.644 0.069 0.753
Intensity of fatigue-previous week (VAS) -0.360 0.091 0.336 0.117
Intensity of numbness - previous week
(VAS) -0.234 0.282 0.007 0.977
Intensity of pain- previous week (VAS) -0.152 0.488 0.250 0.251
Intensity of pain- previous month (VAS) -0.227 0.297 0.114 0.605
Pain pressure threshold (PPT) 0.249 0.252 0.092 0.677
BDI: Beck Depression Inventory; FIQ-R: Fibromyalgia Impact Questionnaire-revised; SSS:
Symptom Severity
Scale; WPI: Widespread Pain Index; SF36: Short Form Health Survey; VAS: Visual Analogue Scale
Discussion
Application of a heterotopic noxious stimulus (Conditioning Stimuli-CS) is known to provoke a reduction
in pain sensitivity (known as the CPM effect) and this mechanism of pain modulation is known to be
impaired in chronic pain populations (29). Thus, the use of a CPM index (3) has been suggested as a
potential biomarker to distinguish clinical proles in different chronic pain syndromes, such as
Fibromyalgia (FM) (16; 15).
Although conceptually clear, the wide variety of the procedures used in CPM testing makes it dicult to
compare results across studies and to obtain solid conclusions about the utility of CPM as a biomarker
of chronic pain (19). To date, the stimuli most frequently used as Test Stimuli (TS) have been mechanical

Page 11/19
stimuli such as pressure pain produced by an algometer (30) or tourniquet (31), electrical stimuli (32) or
thermal stimuli delivered with thermodes (33). Similar variability can be found for the CS, with immersion
in cold water (34) and tourniquet pressure (35) being the most common stimuli. Previous CPM studies
have used a combination of stimuli, with different duration, location, temporality and intensity, and with
the experimental design being adapted to their interests. Consequently, there is a lack of a standardized
method for CPM testing (36).
In this study, we assessed the validity of two CPM paradigms (parallel and sequential) as diagnostic
biomarkers of FM, by comparing patients and healthy controls. Considering that the reliability of the CPM
paradigm is highly dependent on the chosen stimulus combination and characteristics (37), we used
mechanical stimuli, delivered by an algometer (TS) and a pressure cuff (CS), as these produced good
results in a pilot study. Participants in both groups were classied as CPM responders and non-
responders.
When the parallel paradigm was used, more than 80% of healthy participants were classied as
responders, while in the FM patients the rate was about 50%. This difference seems to reinforce the
widely reported presence of a CPM impairment in patients with FM and other chronic pain conditions (12;
13; 14; 15), which may be due to impaired descending inhibitory pain modulation (38). We also found
that CPM testing correctly classied almost 70% of patients with FM. On the other hand, when the
sequential paradigm was used, the rate of responders in both patients and healthy participants was less
than 40%, with no difference in the between-group analysis. Thus, the parallel protocol appears to
generate a more consistent CPM effect. Likewise, different studies have found that the sequential
paradigm was less ecient in different pain conditions like migraine (39). Thus, our ndings conrm the
utility of CPM for classifying FM patients and healthy controls, although only the parallel CPM paradigm
proved useful for classication. This demonstrates that some methodological factors can inuence the
magnitude of the CPM effect and its clinical application (20; 37; 40), and it reinforces the need for a
standardized protocol.
Our results are consistent with those of previous studies conducted in patients with FM, highlighting the
existing dysfunction in CPM in these patients. However, most previous FM studies have obtained similar
results with sequential paradigms. For example, in a recent study, Knezevic and cols (2023) reported
lower pain thresholds and clear impairment of CPM in FM patients. Previous studies also observed lower
CPM inhibitory ecacy in FM, although it is not clear whether the impaired endogenous pain inhibition is
a cause or consequence of prolonged pain (41). Given the effectiveness of the sequential paradigm in
these studies and as the differences between the sequential and parallel protocols may not be signicant
(21), the fact that we observed that the parallel paradigm was much superior is of interest. Considering
that different chronic pain pathologies like knee osteoarthritis (42; 43), migraine (20) and lumbosacral
radiculopathy (44) cause alterations in the CPM effect, it would be interesting to further investigate
several pain pathologies using different protocols and combinations of stimuli to identify the typical
patterns or proles for each pathology. The nal goal is to develop a standardized CPM protocol, which
could be adapted to each of the most prevalent chronic pain pathologies.

Page 12/19
There is a consistent body of literature supporting the utility of CPM for classifying patients with chronic
pain problems and as a marker for diagnosis, prognosis and treatment of different diseases. CPM tests
can improve the determination of different pain modulation proles (45). Furthermore, considering
whether a given person has a pronociceptive or antinociceptive prole makes it easier to predict the future
occurrence of pain and to make decisions about treatment (46). In this vein, there is evidence for the
utility of CPM to predict surgery outcomes: patients who exhibited a dysfunctional endogenous pain
modulatory system prior to surgery were found to be at greater risk of developing chronic pain
postoperatively (35). This study shows that the parallel CMP can correctly classify 70% of the patients,
and thus supports its use as a complementary method for diagnosing FM.
Diagnosis, understanding of pathophysiological pathways and treatment of FM have been hampered by
the lack of blood biomarkers of the disease and the subjective nature of the pain experience. The
incorporation of potential biomarkers based on the objective measurement of pain thresholds and
endogenous central mechanisms of analgesia is crucial to advance the understanding of FM (47; 48).
They allow for the generation or validation of classication criteria, and even help to improve the design
of protocols to develop other treatment modes. As an example, pharmacological trials to test new drugs
for pain treatment can be improved if CPM is used for enriched enrolment. CPM identies individuals with
faulty pain modulation mechanisms, and if enrolled selectively, can increase the chances of better results,
and optimise the clinical trial.
Nevertheless, some inconsistent results were obtained regarding the clinical relevance of CPM and its
usefulness as a marker of disease severity. We found that patients with impaired CPM (non-responders)
had more severe symptoms with a greater functional impact (higher scores in the FIQ-R), but less
generalized pain (fewer painful areas throughout the body). Nevertheless, the CPM responses and clinical
symptoms were not signicantly correlated. Similarly, previous studies yielded mixed results regarding
the relationship between CPM responses and the clinical characteristics of FM patients. For example,
some authors found that patients with FM and defective endogenous pain modulation had poorer sleep
quality and greater impairment in cognitive functioning (i.e. sustained attention) (17; 18), while others
reported that the spatial extent of pain was not associated with CPM in FM patients (14). A recent
systematic review also showed no signicant correlations between clinical pain (intensity, interference
due to pain) and CPM (19). In summary, the above results cast doubts about the sensitivity of CPM as a
marker of disease severity, at least for now.
Several factors may have affected our results. First, it has been shown that the CPM effect is induced by
the use of different combinations and types of stimuli (pressure stimuli, caloric stimuli, cold stimuli,
among others). In this study, we used a combination of two pressure stimuli, but testing other
combinations of stimuli and including quantitative sensory testing (QST) measures complementary to
CPM, such as temporal summation (TS) or exercise-induced analgesia, could be used to better
characterize both pronociceptive and antinociceptive mechanisms in patients with FM.

Page 13/19
On the other hand, we did not control the effects of inter- and intra-individual variability on CPM. Previous
studies have reported that individual characteristics such as age (49), sex (49), alcoholism (50), sleep
(50), socioeconomic status (49), exercise (49), ovulatory phase (51), education level (52), psychological
factors (21) and chronic stress (53), among other, can affect the CPM response. Situational factors such
as attentional focus on the CS (53) or the patients’ expectations (21) can also inuence CPM. We used a
sample composed exclusively of women, to better control gender differences. Although FM is generally
considered much more prevalent in women than in men (54), this assumption is now under debate (55).
In addition, there was no control over the medication taken by the participants or the menstrual phase,
which could inuence the variables evaluated. Given the complexity of the FM syndrome, evaluation of
larger samples and subgroup analysis are desirable. For all these reasons, the ndings must be
extrapolated with caution and assumptions considered carefully.
Conclusions
FM is a widespread chronic pain syndrome of unknown etiology but often related to a central dysfunction
in descending inhibitory pathways. In this study, we investigated whether CPM could be used as a
sensitive biomarker of FM diagnosis and pain severity. We found that the CPM paradigm used (parallel or
sequential) strongly affected the results obtained. The parallel CPM showed 70% accuracy in classifying
participants as patients or controls, and it produced a higher proportion of impaired pain modulation
patterns in the FM group. This result reinforces the widely reported presence of CPM impairment in FM
patients but suggests the need to use a standardized method of assessment.
Concerning the sensitivity of CPM as a biomarker of pain severity in FM, we did not nd any signicant
correlation between CPM and clinical symptoms, although patients with an impaired CPM response
reported more severe symptoms and functional impact of the disease.
Overall, our ndings conrm the inuence of some methodological factors on the magnitude of the CPM
effect and suggest the usefulness of CPM (parallel paradigm) as a diagnostic tool for FM; however, they
cast doubts on the sensitivity of CPM as a biomarker of disease severity.
Declarations
Ethics statement
The study involves human participants and was reviewed and approved by the Research Ethics
Committee of Galicia (CEIC-SERGAS; code: 2021/021). The participants provided their written informed
consent to participate in this study.
Availability of data and materials
The raw data supporting the conclusions of this article will be made available by the authors, without
undue reservation.

Page 14/19
Competing interests
The authors declare that they have no competing interests.
Consent for publication
Not applicable.
Funding
This study was funded by the Spanish State Research Agency (Call: Retos 2019. Project reference:
PID2019-107986RB-I00). A. Gil-Ugidos was supported by a grant from the Spanish Government
(Ministerio de Ciencia e Innovación; grant number PRE2020-091845).
Authors’ contributions
AV-M, NS-V, and MC designed the study. AG-U recruited the participants and made the assessments. AG-U,
AV-M, and MC analyzed and interpreted data. AG-U wrote the rst draft of the manuscript, and AV-M and
NS-V corrected it. AG-U and MC wrote the last version of the manuscript. All authors approved the
submitted version.
Acknowledgements
Our thanks to all those who volunteered to participate in this study, and to Dr Carina Fernandes who read
the rst draft and gave us her feedback.
References
1. Wolfe F, Clauw DJ, Fitzcharles M, Goldenberg DL, Katz RS, Mease P, et al. The American College of
Rheumatology Preliminary Diagnostic Criteria for Fibromyalgia and Measurement of Symptom
Severity. Arthritis Care & Research. 2010 May;62(5):600–10.
2. Häuser W, Ablin J, Fitzcharles MA, Littlejohn G, Luciano JV, Usui C, et al. Fibromyalgia. Nat Rev Dis
Primers. 2015 Aug 13;1(1):15022.
3. Yarnitsky D. Conditioned pain modulation (the diffuse noxious inhibitory control-like effect): its
relevance for acute and chronic pain states. Current Opinion in Anaesthesiology. 2010
Oct;23(5):611–5.
4. Nir RR, Yarnitsky D. Conditioned pain modulation. Current Opinion in Supportive & Palliative Care.
2015 Jun;9(2):131–7.
5. Bogdanov VB, Viganò A, Noirhomme Q, Bogdanova OV, Guy N, Laureys S, et al. Cerebral responses
and role of the prefrontal cortex in conditioned pain modulation: an fMRI study in healthy subjects.
Behavioural Brain Research. 2015 Mar; 281:187–98.

Page 15/19
. Le Bars D, Dickenson AH, Besson JM. Diffuse noxious inhibitory controls (DNIC). I. Effects on dorsal
horn convergent neurones in the rat: Pain. 1979 Jun;6(3):283–304.
7. Bouhassira D, Bing Z, Le Bars D. Studies of the brain structures involved in diffuse noxious inhibitory
controls: the mesencephalon. Journal of Neurophysiology. 1990 Dec 1;64(6):1712–23.
. Bouhassira D, Villanueva L, Bing Z, Le Bars D. Involvement of the subnucleus reticularis dorsalis in
diffuse noxious inhibitory controls in the rat. Brain Research. 1992 Nov;595(2):353–7.
9. Bouhassira, D., Villanueva, L., Le Bars, D.Intracerebroventricular morphine decreases descending
inhibitions acting on lumbar dorsal horn neuronal activities related to pain in the rat. Journal of
Pharmacology and Experimental Therapeutics. 1988;247(1), 332-342.
10. Harper DE, Ichesco E, Schrepf A, Hampson JP, Clauw DJ, Schmidt-Wilcke T, et al. Resting Functional
Connectivity of the Periaqueductal Gray Is Associated With Normal Inhibition and Pathological
Facilitation in Conditioned Pain Modulation. The Journal of Pain. 2018 Jun;19(6): 635.e1-635.e15.
11. Pielsticker A, Haag G, Zaudig M, Lautenbacher S. Impairment of pain inhibition in chronic tension-
type headache. Pain. 2005 Nov;118(1):215–23.
12. Lewis GN, Rice DA, McNair PJ. Conditioned Pain Modulation in Populations With Chronic Pain: A
Systematic Review and Meta-Analysis. The Journal of Pain. 2012 Oct;13(10):936–44.
13. Meeus M. Does Acetaminophen Activate Endogenous Pain Inhibition in Chronic Fatigue
Syndrome/Fibromyalgia and Rheumatoid Arthritis? A Double Blind Randomized Controlled Cross-
over Trial. Pain Phys. 2013 Mar 14;2;16(2;3): E61–70.
14. Gerhardt A, Eich W, Treede RD, Tesarz J. Conditioned pain modulation in patients with nonspecic
chronic back pain with chronic local pain, chronic widespread pain, and bromyalgia. Pain. 2017
Mar;158(3):430–9.
15. O’Brien AT, Deitos A, Triñanes Pego Y, Fregni F, Carrillo-de-la-Peña MT.Defective Endogenous Pain
Modulation in Fibromyalgia: A Meta-Analysis of Temporal Summation and Conditioned Pain
Modulation Paradigms. The Journal of Pain. 2018 Aug;19(8):819–36.
1. Ramaswamy S, Wodehouse T. Conditioned pain modulation—A comprehensive review.
Neurophysiologie Clinique. 2021 Jun;51(3):197–208.
17. Paul-Savoie E. Is the Decit in Pain Inhibition in Fibromyalgia Inuenced by Sleep Impairments?
TORJ. 2012 Oct 4;6(1):296–302.
1. Coppieters I, Ickmans K, Cagnie B, Nijs J, De Pauw R, Noten S, et al. Cognitive performance is related
to central sensitization and health-related quality of life in patients with chronic whiplash-associated
disorders and bromyalgia. PAIN PHYSICIAN. 2015;18(3): E389–401
19. Fernandes C, Pidal-Miranda M, Samartin-Veiga N, Carrillo-de-la-Peña MT.Conditioned pain
modulation as a biomarker of chronic pain: a systematic review of its concurrent validity. Pain. 2019
Dec;160(12):2679–90.
20. Nahman-Averbuch H, Yarnitsky D, Granovsky Y, Gerber E, Dagul P, Granot M. The role of stimulation
parameters on the conditioned pain modulation response. Scandinavian Journal of Pain. 2013 Jan
1;4(1):10–4.

Page 16/19
21. Ibancos-Losada MDR, Osuna-Pérez MC, Castellote-Caballero MY, Díaz-Fernández Á. Conditioned Pain
Modulation Effectiveness: An Experimental Study Comparing Test Paradigms and Analyzing
Potential Predictors in a Healthy Population. Brain Sciences. 2020 Aug 30;10(9):599.
22. Kovacevic M, Klicov L, Vuklis D, Neblett R, Knezevic A. Test-retest reliability of pressure pain threshold
and heat pain threshold as test stimuli for evaluation of conditioned pain modulation.
Neurophysiologie Clinique. 2021 Oct;51(5):433–42.
23. Klyne DM, Schmid AB, Moseley GL, Sterling M, Hodges PW. Effect of Types and Anatomic
Arrangement of Painful Stimuli on Conditioned Pain Modulation. The Journal of Pain. 2015
Feb;16(2):176–85.
24. Nuwailati R, Curatolo M, LeResche L, Ramsay DS, Spiekerman C, Drangsholt M. Reliability of the
conditioned pain modulation paradigm across three anatomical sites. Scandinavian Journal of Pain.
2020 Apr 28;20(2):283–96.
25. Carrillo-de-la-Peña MT, Triñanes Y, González-Villar A, Romero-Yuste S, Gómez-Perretta C, Arias M, et
al.Convergence between the 1990 and 2010 ACR diagnostic criteria and validation of the Spanish
version of the Fibromyalgia Survey Questionnaire (FSQ). Rheumatol Int. 2015 Jan;35(1):141–51.
2. Sanz J, Vázquez C. FIABILIDAD, VALIDEZ Y DATOS NORMATIVOS DEL INVENTARIO PARA LA
DEPRESIÓN DE BECK. Psicothema. 1998 dic; 10(2):303-18.
27. Salgueiro M, García-Leiva JM, Ballesteros J, Hidalgo J, Molina R, Calandre EP.Validation of a
Spanish version of the Revised Fibromyalgia Impact Questionnaire (FIQR).Health Qual Life
Outcomes. 2013 Dec;11(1):132.
2. Alonso J, Prieto L, Antó JM. La versión española del SF-36 Health Survey (Cuestionario de Salud SF-
36): un instrumento para la medida de los resultados clínicos. Medicina clínica. 1995 Jan
1;0(104):771-776.
29. Arendt‐Nielsen L, Morlion B, Perrot S, Dahan A, Dickenson A, Kress HG, et al. Assessment and
manifestation of central sensitisation across different chronic pain conditions. European Journal of
Pain. 2018 Feb;22(2):216–41.
30. Heredia-Rizo AM, Petersen KK, Madeleine P, Arendt-Nielsen L. Clinical Outcomes and Central Pain
Mechanisms are Improved After Upper Trapezius Eccentric Training in Female Computer Users With
Chronic Neck/Shoulder Pain.The Clinical Journal of Pain. 2019 Jan;35(1):65–76.
31. Oono Y, Nie H, Matos RL, Wang K, Arendt-Nielsen L. The inter- and intra-individual variance in
descending pain modulation evoked by different conditioning stimuli in healthy men.Scandinavian
Journal of Pain. 2011 Oct 1;2(4):162–9.
32. Bouwense SA, Ahmed Ali U, Ten Broek RP, Issa Y, Van Eijck CH, Wilder-Smith OH, et al. Altered central
pain processing after pancreatic surgery for chronic pancreatitis. British Journal of Surgery. 2013
Nov 14;100(13):1797–804.
33. Pickering G, Pereira B, Dufour E, Soule S, Dubray C. Impaired Modulation of Pain in Patients with
Postherpetic Neuralgia.Pain Research and Management. 2014;19(1): e19–23.

Page 17/19
34. Jarrett ME, Shulman RJ, Cain KC, Deechakawan W, Smith LT, Richebé P, et al. Conditioned Pain
Modulation in Women With Irritable Bowel Syndrome. Biological Research For Nursing. 2014
Oct;16(4):368–77.
35. Vaegter HB, Palsson TS, Graven-Nielsen T. Facilitated Pronociceptive Pain Mechanisms in Radiating
Back Pain Compared With Localized Back Pain.The Journal of Pain. 2017 Aug;18(8):973–83.
3. Nuwailati R, Bobos P, Drangsholt M, Curatolo M. Reliability of conditioned pain modulation in healthy
individuals and chronic pain patients: a systematic review and meta-analysis.Scandinavian Journal
of Pain. 2022 Apr 26;22(2):262–78.
37. Kennedy DL, Kemp HI, Ridout D, Yarnitsky D, Rice ASC. Reliability of conditioned pain modulation: a
systematic review.Pain. 2016 Nov;157(11):2410–9.
3. Staud R. Abnormal endogenous pain modulation is a shared characteristic of many chronic pain
conditions.Expert Review of Neurotherapeutics. 2012 May;12(5):577–85.
39. Nahman-Averbuch H, Nir RR, Sprecher E, Yarnitsky D. Psychological Factors and Conditioned Pain
Modulation: A Meta-Analysis.The Clinical Journal of Pain. 2016 Jun;32(6):541–54.
40. Petersen KK, McPhee ME, Hoegh MS, Graven-Nielsen T. Assessment of conditioned pain modulation
in healthy participants and patients with chronic pain: manifestations and implications for pain
progression.Current Opinion in Supportive & Palliative Care. 2019 Jun;13(2):99–106.
41. Potvin S, Marchand S. Pain facilitation and pain inhibition during conditioned pain modulation in
bromyalgia and in healthy controls.Pain. 2016 Aug;157(8):1704–10.
42. Graven-Nielsen T, Arendt-Nielsen L. Assessment of mechanisms in localized and widespread
musculoskeletal pain.Nat Rev Rheumatol. 2010 oct;6(10):599–606.
43. Arendt‐Nielsen L, Egsgaard LL, Petersen KK, Eskehave TN, Graven‐ Nielsen T, Hoeck HC, et al. A
mechanism‐based pain sensitivity index to characterize knee osteoarthritis patients with different
disease stages and pain levels.European Journal of Pain. 2015 Nov;19(10):1406–17.
44. Knezevic A, Kovacevic M, Jeremic-Knezevic M, Nikolasevic Z, Tomasevic-Todorovic S, Zivanovic Z, et
al.Patients with neuropathic pain from lumbosacral radiculopathy demonstrate similar pressure
pain thresholds and conditioned pain modulation to those with bromyalgia.Neurophysiologie
Clinique. 2023 Aug;53(4):102841.
45. Yarnitsky D, Granot M, Granovsky Y. Pain modulation prole and pain therapy: Between pro- and
antinociception.Pain. 2014 Apr;155(4):663–5.
4. Granovsky YG, Yarnitsky D. Personalized Pain Medicine: The Clinical Value of Psychophysical
Assessment of Pain Modulation Prole. Rambam Maimonides Med J [Internet]. 2013 Oct 29 [cited
2024 Jan 9];4(4). Available from: https://www.rmmj.org.il/issues/18/Articles/336
47. Dadabhoy D, Crofford LJ, Spaeth M, Russell IJ, Clauw DJ. Biology and therapy of bromyalgia.
Evidence-based biomarkers for bromyalgia syndrome. Arthritis Res Ther. 2008;10(4):211.
4. Mease P, Arnold LM, Choy EH, Clauw DJ, Crofford LJ, Glass JM, et al. Fibromyalgia Syndrome
Module at OMERACT 9: Domain Construct. J Rheumatol. 2009 Oct;36(10):2318–29.

Page 18/19
49. Hermans L, Van Oosterwijck J, Goubert D, Goudman L, Crombez G, Calders P, et al. Inventory of
Personal Factors Inuencing Conditioned Pain Modulation in Healthy People: A Systematic Literature
Review. Pain Practice. 2016 Jul;16(6):758–69.
50. Klyne DM, Moseley GL, Sterling M, Barbe MF, Hodges PW. Individual Variation in Pain Sensitivity and
Conditioned Pain Modulation in Acute Low Back Pain: Effect of Stimulus Type, Sleep, and
Psychological and Lifestyle Factors. The Journal of Pain. 2018 Aug;19(8): 942.e1-942.e18.
51. Rezaii T, Hirschberg AL, Carlström K, Ernberg M. The Inuence of Menstrual Phases on Pain
Modulation in Healthy Women. The Journal of Pain. 2012 Jul;13(7):646–55.
52. Skovbjerg S, Jørgensen T, Arendt-Nielsen L, Ebstrup JF, Carstensen T, Graven-Nielsen T. Conditioned
Pain Modulation and Pressure Pain Sensitivity in the Adult Danish General Population: The DanFunD
Study. The Journal of Pain. 2017 Mar;18(3):274–84.
53. Mertens MG, Hermans L, Crombez G, Goudman L, Calders P, Van Oosterwijck J, et al. Comparison of
ve conditioned pain modulation paradigms and inuencing personal factors in healthy adults.
European Journal of Pain. 2021 Jan;25(1):243–56.
54. Mas AJ, Carmona L, Valverde M, Ribas B, EPISER Study and Group. Prevalence and impact of
bromyalgia on function and quality of life in individuals from the general population: results from a
nationwide study in Spain. Clin. Exp. Rheumatol. 2008; 26: 519–526.
55. Segura-Jiménez V, Estévez-López F, Soriano-Maldonado A, Álvarez-Gallardo IC, Delgado-Fernández
M, Ruiz JR, et al. Gender Differences in Symptoms, Health-Related Quality of Life, Sleep Quality,
Mental Health, Cognitive Performance, Pain-Cognition, and Positive Health in Spanish Fibromyalgia
Individuals: The Al-Ándalus Project. Pain Research and Management. 2016; 2016:1–14.
Figures

Page 19/19
Figure 1
CPM procedures used in the study.