Bittencourt et al.

Archives of Physiotherapy (2023) 13:17 Archives of Physiotherapy

https://doi.org/10.1186/s40945-023-00171-8

RESEARCH ARTICLE Open Access

Does the painDETECT questionnaire

identify impaired conditioned pain modulation
in people with musculoskeletal pain? –
a diagnostic accuracy study
Juliana Valentim Bittencourt1* , Eduardo Gallas Leivas1 , Arthur de Sá Ferreira1    and
Leandro Alberto Calazans Nogueira1,2   

Abstract
Background People with neuropathic-like symptoms had more unfavourable pain features than people with noci-
ceptive. Moreover, deficient conditioned pain modulation is common in people with neuropathic-like symp-
toms. PainDETECT questionnaire have been used to assess the central sensitisation sign and symptoms. However,
whether the painDETECT questionnaire can identify the conditioned pain modulation’s impairment is still unknown.
Therefore, the current study aimed to evaluate the diagnostic accuracy of the painDETECT questionnaire in detecting
the impairment of conditioned pain modulation in people with musculoskeletal pain.
Methods We conducted a diagnostic accuracy comparing the painDETECT questionnaire (index method)
with the cold pressor test, the psychophysical test used to assess the conditioned pain modulation (reference stand-
ard). We determined diagnostic accuracy by calculating sensitivity, specificity, predictive values, and likely hood ratios.
Results We retrospectively enrolled 308 people with musculoskeletal pain in outpatient departments. Most partici-
pants were female (n 20 = 220, 71.4%) and had a mean age of 52.2 (± 15.0) years. One hundred seventy-three (56.1%)
participants were classified as nociceptive pain, 69 (22.4%) as unclear, and 66 (21.4%) as neuropathic-like symptoms.
According to the cold pressor test, 60 (19.4%) participants presented impairment of conditioned pain modulation.
The cutoff point of 12 of the painDETECT questionnaire showed values of diagnostic accuracy below 70% compared
to the cold pressor test, except for a negative predictive value [76.9 95% Confidence Interval (CI) 71.7 to 81.5]. The
cutoff point 19 showed high specificity (78.6%, 95% CI 73.0 to 83.5), high negative predictive value (80.5%, 95% CI 78.1
to 82.7), and accuracy of 67.5% compared to the cold pressor test.
Conclusion The painDETECT questionnaire seems valuable for ruling out people with musculoskeletal pain
and impairment of conditioned pain modulation.
Keywords Musculoskeletal pain, Neuropathic pain, Pain mechanisms, Central nervous system sensitization, Diffuse
noxious inhibitory control

*Correspondence:
Juliana Valentim Bittencourt
[email protected]
Full list of author information is available at the end of the article

© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​
mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 2 of 9

What's already known about this topic? nociceptive pain have demonstrated CS-related signs and
• PainDETECT is a screening tool to identify neuropathic-like symptoms. Lluch et al. showed that 28 to 34% of people
symptoms.
• People with neuropathic-like symptoms have impairment of pain
with osteoarthritis knee pain had CS-related signs and
modulation. symptoms considering different aspects of CS-related
signs and symptoms (i.e., clinical manifestations of CS,
quantitative sensory testing results, dysfunctional endog-
What does this study add?
• The cutoff point of 19 had high specificity and negative predictive enous nociceptive inhibition, and neuroimaging) [17].
value compared to the cold pressor test. Additionally, the interaction between osteoarthritis knee
• Our data suggest that the painDETECT questionnaire is a valuable and CS increased nocturnal discomfort and disability
instrument for ruling out people with musculoskeletal pain and
impairment of conditioned pain modulation [18]. Physiotherapy approaches (e.g. education, exercise,
manual therapy and transcutaneous electrical nerve stim-
ulation) can target particular pain phenotypes and indi-
vidualise care [19]. Therefore, detecting the impairment
Background of conditioned pain modulation in people presenting
Neuropathic pain leads to unfavourable outcomes and musculoskeletal pain (i.e., nociceptive pain and neuro-
remains a major clinical challenge. People with neuro- pathic-like symptoms) may assist clinicians in offering
pathic-like symptoms showed unfavourable pain features appropriate therapeutic strategies for these groups.
(i.e., pain intensity and functional limitation) compared The use of the painDETECT questionnaire in assess-
to their counterparts [1]. Previous studies showed that ing CS-related signs and symptoms has been advocated.
neuropathic pain interferes with several aspects of an PainDETECT was designed as a screening tool, but it
individual’s life, such as poor sleep quality [2], physi- may also function as a measure of characteristics that
cal impairment [3], and a large psychosocial burden point to enhanced central pain processing [20]. Gwylim
[4]. Moreover, there is a high prevalence of depres- et al. revealed that people with higher painDETECT
sion among people with neuropathic pain which harms questionnaire scores had more CS signs [21]. Like-
the quality of life [5]. There are screening tools capable wise, the modified painDETECT questionnaire may
of identifying key neuropathic-like symptoms. PainDE- assist identification of CS in adults with knee osteoar-
TECT questionnaire is a reliable, simple, and validated thritis since higher modified painDETECT question-
screening tool for identifying neuropathic-like symptoms naire scores (> 12) were 5.6 times more likely to have
in people with chronic low back pain [6]. PainDETECT CS-related signs and symptoms [22]. Of note, the prior
questionnaire has been validated for various illnesses, study screened people with knee osteoarthritis using
including rheumatoid arthritis, osteoarthritis, fibromy- the modified painDETECT, which targets symptoms
algia, cancer pain and lumbar spondylolisthesis [7]. Fur- ‘in or around’ each knee rather than their “main area
thermore, compared to other available instruments, the of pain”, pain spreading up or down from the knee, and
painDETECT questionnaire is one of the best options for a figure with gender-neutral [22]. It is unclear whether
screening neuropathic-like symptoms (presenting 85% of the painDETECT questionnaire results may detect peo-
sensitivity, 80% of specificity, and 83% of positive predic- ple with impairment of pain modulation. Therefore, the
tive accuracy) [8]. Thus, using the painDETECT ques- current study aimed to evaluate the diagnostic accu-
tionnaire is popular among researchers and clinicians to racy of the painDETECT questionnaire in detecting the
identify neuropathic-like symptoms in people with mus- impairment of conditioned pain modulation in people
culoskeletal pain. with musculoskeletal pain.
Neuropathic pain is involved with peripheral and cen-
tral sensitisation (CS). Several instruments are available
to identify the clinical features of CS in the musculoskele- Methods
tal population [9]. The cold pressor test is one of the most Study design and ethical considerations
appropriate conditioned pain modulation paradigms to We conducted and reported a diagnostic accuracy study
assess descending nociceptive modulatory pathways [10]. following the Standards for Reporting of Diagnostic
People with neuropathic-like symptoms have impair- Accuracy Studies (STARD) guidelines [23] (Additional
ment of pain modulation, which is considered indicative file 1). The Research Ethics Committee of the Federal
of CS-related signs and symptoms [11, 12]. For instance, Institute of Rio de Janeiro approved this study (number:
carpal tunnel syndrome [13], painful diabetic neuropathy 02228818.0.3001.5258) following the Helsinki Declaration
[14], painful peripheral neuropathy [15], and complex- for research in humans. All people with musculoskeletal
regional pain syndrome [16] have a deficient conditioned pain who met the eligibility criteria signed the informed
pain modulation. Likewise, people with presumably consent form before undergoing the study procedures.
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 3 of 9

Study population Index method

We recruited retrospectively people with musculoskel- The PainDETECT questionnaire is a self-administered
etal pain from two public physiotherapy departments tool used to assess neuropathic-like symptoms. It com-
(i.e., Gaffrée and Guinle University Hospital and Cabo prises four domains with the following components:
Frio Rehabilitation Centre) and three private physiother- pain intensity (three questions), pain course pattern
apy departments (i.e., Augusto Motta University Centre, (four graphs), areas of pain and presence of radiating
Saúde Clin Physiotherapy Clinic, and Fisiofit Physio- pain (body chart drawing), and sensory descriptor items
therapy Clinic) in Rio de Janeiro State and Minas Gerais, of pain (seven questions). The first domain consists
Brazil between March and September 2019. In public of three questions assessing pain intensity, including
physiotherapy departments, orthopaedists, general prac- the strongest and average pain levels over the past four
titioners, or other health professionals have often weeks. The final score is calculated based on a nine-item
referred people with musculoskeletal pain. People in all representation of the last three domains (pain course
private physiotherapy departments reported seeking care pattern, radiating pain, and gradation of pain). The sec-
for their musculoskeletal condition primarily due to pain. ond domain (pain course pattern) has answer options
The study involved people with acute pain (less than of Persistent pain with slight fluctuations = 0, Persis-
three months) and chronic pain (pain duration greater tent pain with pain attacks = -1, Pain attacks without
than three months). We defined musculoskeletal pain pain between them =  + 1, and Pain attacks with pain
as pain originating from muscles, ligaments, bones, or between them =  + 1. The score for this domain varies
joints in a specific body region [24]. We excluded peo- between 0 and + 1. The third domain (radiating pain)
ple who had undergone spinal surgery, pregnant women, includes a dichotomous question, "Does your pain radi-
people in the acute inflammatory phase of rheumato- ate to other regions of your body?" with answer options
logic diagnoses, people with tumours, those who were of yes or no, corresponding to scores of + 2 or 0, respec-
illiterate, or those unable to complete the self-reported tively. The fourth domain (gradation of pain) comprises
questionnaires. seven questions, each with six possible answers scored
from 0 (never) to 5 (very strongly). The scores given in
Procedures each domain are summed up to achieve a final score
The evaluation included the clinical history, physical ranging from -1 to 38. The PainDETECT questionnaire
examination, and cold pressor test performed on the is validated for neuropathic pain conditions [30–32] and
same day for people with musculoskeletal pain. We col- has also been validated for mixed pain conditions such
lected sociodemographic and clinical information using as rheumatoid arthritis, osteoarthritis, cancer pain, and
an instrument that included demographic data (age, sex, lumbar spondylolisthesis [33]. The original question-
weight, height, education level, and income) as well as naire’s cut-off points indicate that scores ≤ 12 suggest a
characteristics of musculoskeletal pain (pain intensity neuropathic component is unlikely, scores between 13
and duration). We measured pain intensity using the and 18 show an unclear neuropathic component, while
Numeric Pain Rating Scale, ranging from 0 to 10 (0 rep- scores ≥ 19 suggest a probable neuropathic component
resents no pain, and 10 illustrates the worst pain possi- [33]. For screening purposes, we considered scores ≤ 12
ble). This scale is commonly used in musculoskeletal pain indicative of nociceptive pain, scores between 13 and 18
studies and has demonstrated good reproducibility levels as unclear, and scores ≥ 19 indicative of neuropathic-like
[25]. Pain duration was recorded in months, with chronic symptoms. The PainDETECT questionnaire was cross-
pain defined as lasting over three months and acute pain culturally adapted to the Brazilian context [27].
lasting less than three months [26]. Neuropathic-like
symptoms were measured using the painDETECT ques- Reference method
tionnaire, with the Brazilian version proving helpful in The psychophysical measure of the descending nociceptive
identifying such symptoms [27]. The cold pressor test inhibitory system
assessed the Conditioned Pain Modulation (CPM), which We used the cold pressor test as a psychophysical meas-
evaluates the descending nociceptive inhibitory system ure to evaluate the descending nociceptive inhibitory sys-
[28, 29]. An examiner supervised the completion of all tem [34] and assess conditioned pain modulation [35]. In
questionnaires, providing clarification when needed, and this test, the conditioning stimulus was the immersion of
the process took approximately 10 min per person with the people’s hands in a bucket of temperature-controlled
musculoskeletal pain. After completing the question- cold water (1ºC – 4ºC) for up to one minute. We moni-
naires, musculoskeletal pain patients were referred for tored the water temperature using a thermometer (5130
CPM evaluation. model, Incoterm). People with musculoskeletal pain were
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 4 of 9

instructed to keep their hands immersed in the water for within and between-group differences (nociceptive
without making muscle contractions or changing posi- pain, unclear, or neuropathic-like symptoms) for the out-
tions. They could withdraw their hand from the water come measures with continuous variables (i.e., pressure
when they could no longer tolerate the painful stimulus. pain threshold values for the dorsal region of the fore-
We maintained constant room temperature, humidity, arm and anterior tibial of the participants). The diagnos-
lighting, and noise throughout the procedure. tic accuracy of the painDETECT questionnaire (index
method) was compared with the psychophysical measure
Pressure pain threshold of the descending nociceptive inhibitory system (refer-
We used a digital pressure algometer (model Force Ten ence standard). We calculated sensitivity, specificity,
FDX, Wagner Instruments, Greenwich, USA) to meas- likelihood ratio, positive likelihood ratio, negative likeli-
ure the pain threshold. We performed the pressure pain hood ratio, disease prevalence, positive predictive value,
threshold assessment before and after one minute of negative predictive value, and accuracy with correspond-
the cold pressor test. The evaluation was conducted on ing exact 95% binomial confidence intervals (CIs) for two
the distal part of the dorsal forearm and tibialis ante- predefined cutoff points (12 and 19). For diagnostic accu-
rior muscle, which had not been immersed in water and racy tests (i.e., sensitivity, specificity, predictive values,
were unrelated to people’s musculoskeletal complaints. and accuracy), values < 50% were interpreted as low, 50%
The assessment was done in the same order for all peo- to 70% as moderate, and > 70% to 100% as high. A signifi-
ple with musculoskeletal pain. Before the evaluation, we cance level of less than 5% (p < 0.05) was considered for
explained the operation of the pressure algometer and all analyses. The statistical analysis was performed by Jef-
how the pressure pain threshold would be measured. We freys’s Amazing Statistics Program (JASP), version 0.16.3,
also conducted a familiarisation procedure by applying and Prism for Macintosh, Version 8 (GraphPad Software
pressure to the dominant forearm, ensuring the people Inc., San Diego, CA).
with musculoskeletal pain understood the test. The force
on the algometer was gradually increased (1 kg-force/s) Sample size calculation
until the primary subject felt a change from pressure The sample calculation was based on the values obtained
to pain. The pressure pain threshold was recorded in from the study by Gervais-Hupé et al. [37]. The authors
kilograms-force (Kgf ) when the people with musculo- observed a sensitivity of 61.5% and specificity of 77.6% in
skeletal pain verbally indicated experiencing pain. We identifying impaired conditioned pain modulation using
classified the efficiency of the conditioned pain modula- the cutoff point of 12 in the painDETECT questionnaire
tion based on the following strategy: evidence of impair- in people with knee osteoarthritis. The estimate was cal-
ment of pain modulation in both evaluated sites. Only culated considering the prevalence of central sensitisa-
people with musculoskeletal pain showing impaired tion of 21.43% in people with musculoskeletal pain [38],
conditioned pain modulation in both the anterior tibi- the alpha value of 5%, and the precision of the estimate
alis muscle and the distal part of the dorsal forearm were of 12%. Thus, it was necessary to include 295 people with
classified as having impaired conditioned pain modula- musculoskeletal pain.
tion. Using upper and lower limb sites aimed to avoid
including people with peripheral sensitisation, follow- Results
ing recommendations for conditioned pain modulation Characteristics of the participants
[36]. The efficiency of the conditioned pain modula- A total of 308 people with musculoskeletal pain were
tion was assessed by calculating the difference between enrolled. Most participants were female (n 20 = 220,
the pressure pain threshold values obtained during the 71.4%), had a mean age of 52.2 (± 15.0) years, and had a
cold pressor test (the difference between final and ini- mean of moderate pain intensity (Table 1). Two-hundred
tial values). Negative values indicated an inefficiency of sixty-six (86.3%) participants had chronic pain, and 42
the conditioned pain modulation, while null or positive (13.6%) had acute pain. Overall, 43 (13.9%) people with
values were considered a typical response of the condi- musculoskeletal pain reported a previous diagnosis of
tioned pain modulation. fibromyalgia, 48 (15.5%) people with musculoskeletal
pain described migraine, 80 (25.9%) people with muscu-
Statistical analysis loskeletal pain had anxiety and 73 (23.7%) people with
Demographic and clinical variables of the study popula- musculoskeletal pain had a prior history of depressive
tion were presented as the mean and standard deviation disorder. Low back pain (n = 166, 53.8%) was the leading
for continuous variables. Categorical variables were pre- complaint, followed by upper back (n = 136, 44.1%), right
sented as absolute values and frequencies. Independent shoulder (n = 131, 42.5%), neck (n = 123, 39.9%) and left
one-way analysis of variance (ANOVA) was used to test shoulder (n = 116, 37.6%).
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 5 of 9

Table 1 Characteristics of the study people with musculoskeletal high negative predictive value. The cutoff point 19 of the
pain (n = 308) painDETECT questionnaire showed low sensitivity, high
Characteristics Values (n = 308) specificity, and high negative predictive value, despite the
accuracy below 70% compared to the cold pressor test.
Sex (female), n (%) 220 (71.4%) Measures of sensitivity, specificity, likelihood ratios, dis-
Age (years), mean (SD) 52.2 (± 15.0) ease prevalence, predictive values, and accuracy regard-
Weight (kg), mean (SD) 73.3 (± 16.6) ing the predefined cutoff point of the painDETECT
Height (meters), mean (SD) 1.6 (± 0.1) questionnaire for the detection of impairment of the con-
Body Mass Index (Kg/m2), mean (SD) 27.8 (± 13.1) ditioned pain modulation are shown in Table 3.
Private Health Insurance, yes, n (%) 71 (23.0%)
Physical Activity, yes, n (%) 159 (51.6%) Discussion
Pain characteristics This study investigated the diagnostic accuracy of the
Pain intensity at the moment, mean (SD) 5.8 (± 2.4) painDETECT questionnaire in identifying the impair-
Strongest pain level in the last 4 weeks, mean (SD) 8.0 (± 2.0) ment of conditioned pain modulation in people with
Pain level on average in the last 4 weeks, mean (SD 6.6 (± 2.2) musculoskeletal pain. Our findings revealed that the
Pain duration (months), mean (SD) 66.7 (± 100.7) painDETECT questionnaire exhibited low sensibility and
Pain intensity, mean (SD) high specificity for the cutoff point 19 only compared to
Final painDETECT score, mean (SD) 11.9 (± 7.7) the cold pressor test. Our data showed high negative pre-
Nociceptive pain (0–12), n (%) 173 (56.1%) dictive values for both cutoffs of the painDETECT ques-
Unclear (13-18), n (%) 69 (22.4%) tionnaire, which suggests that a negative test can reliably
Neuropathic-like symptoms (≥ 19), n (%) 66 (21.4%) exclude the impairment of the conditioned pain modu-
Cold pressor test, impaired, n (%) 60 (19.4%) lation in this population of musculoskeletal pain people.
Data are presented as mean (SD) for continuous variables and as frequency The low prevalence of the impairment of the conditioned
counts (%) for categorical variables pain modulation in the study sample likely increases the
negative predictive value. Accordingly, many people with
Sixty-six (21.4%) people with musculoskeletal pain had shared values on the painDETECT questionnaire were
painDETECT questionnaire scores ≥ 19, being 5 (7.5%) diagnosed with preserved conditioned pain modulation.
classified as acute pain, and 69 (22.4%) people with Our findings showed that values below 19 points in the
musculoskeletal pain had painDETECT questionnaire painDETECT questionnaire correctly detect preserved
scores between 13–18 points, being 12 (17.3%) classi- conditioned pain modulation in most people with mus-
fied as acute pain. All people with musculoskeletal pain culoskeletal pain. Likewise, a previous study considered
completed the painDETECT questionnaire and the cold scores above 18 in the painDETECT questionnaire as
pressor test. Then, there were no missing values for the CS-related signs and symptoms [39]. The painDETECT
painDETECT questionnaire and the cold pressor test questionnaire has been used for the neurobiological con-
results. No adverse events were associated with the pain- firmation of central sensitisation in people with features of
DETECT questionnaire and the cold pressor test Fig. 1. neuropathic pain [40]. However, a definitive cutoff is not a
Table 2 presents pressure pain threshold values ​​for consensus. The cutoff of 12 in the modified painDETECT
people with musculoskeletal pain in the dorsal region of questionnaire presents a sensitivity of 61.5% and specific-
the forearm and anterior tibial. The pressure pain thresh- ity of 77.6% in identifying CS-related signs and symptoms
old at the anterior tibial before the cold pressor test was in people with knee osteoarthritis [37]. The exact cutoff
reduced in people with unclear classification and neu- has been advised to indicate CS-related signs and symp-
ropathic-like symptoms compared to people with noci- toms in people with chronic painful knee osteoarthritis
ceptive pain. The pressure pain threshold at the dorsal [22]. Nonetheless, considering the relatively low sensitiv-
forearm pressure after the cold pressor test was reduced ity and specificity measures, the authors did not recom-
in the people classified as unclear and with neuropathic- mend this cutoff [22]. Similarly, our data suggested that
like symptoms compared to people with nociceptive the cutoff of 12 points had insufficient accuracy in identi-
pain. There is no significant difference in within-group fying the impairment of the conditioned pain modulation
comparison in the dorsal region of the forearm and ante- in a heterogeneous sample of people with musculoskel-
rior tibial of the people with musculoskeletal pain. etal pain. Moreover, the low values of likelihood ratios for
both cutoff points and the low accuracy for cutoff point 12
Diagnostic accuracy of the painDETECT questionnaire limit the applicability of the painDETECT questionnaire
The cutoff point 12 of the painDETECT questionnaire to identify the impairment of the conditioned pain modu-
showed sensitivity, specificity, accuracy below 70%, and a lation in people with musculoskeletal pain.
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 6 of 9

Fig. 1 Flowchart of the study

Neuropathic-like symptoms are linked with periph- to a decrease in conditioned pain modulation over time
eral and CS-related signs and symptoms. Some genetic [45]. Carpal tunnel syndrome [13] and painful periph-
variants could be an essential modulator in develop- eral neuropathy [15] are examples of impairment of
ing CS-related signs and symptoms in neuropathic conditioned pain modulation. On the other hand, peo-
pain [41]. Still, CS-related signs and symptoms mani- ple with painful diabetic neuropathy had a preserved
fest most in painful conditions with the neuropathic conditioned pain modulation despite pain duration
component [42, 43]. Many strategies, such as con- [46]. Hence, the relationship between conditioned pain
ditioned pain modulation, could assess clinical fea- modulation and neuropathic pain may be a particular
tures of CS-related signs and symptoms. Conditioned feature of this population.
pain modulation is a predictor in developing and Our study provides new insight for implementation
treating neuropathic pain [44] but may perform dis- in clinical use and further studies. PainDETECT ques-
similarly in neuropathic pain conditions. Gagné et al. tionnaire can be used as an initial screening strategy by
suggested that the presence of neuropathic pain leads physiotherapists and other health professionals to screen
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 7 of 9

Table 2 Comparison of pain threshold values between people with neuropathic-like symptoms, nociceptive pain, and unclear
classification
Characteristics Nociceptive pain Unclear Neuropathic- ANOVA Comparison Groups p-value
n = 173 n = 69 like symptoms
n = 66

Baseline
Dorsal forearm algometry (kgf ) 4.1 (1.5) 3.6 (1.2) 3.6 (1.2) 5.290 Nociceptive versus Unclear .029
Nociceptive versus Neuropathic .024
Tibialis anterior algometry (kgf ) 4.6 (1.7) 4.2 (1.7) 3.5 (1.4) 8.673 Nociceptive versus Neuropathic < .001
After Cold Pressor Test
Nociceptive versus Unclear .042
Dorsal forearm algometry (kgf ) 4.4 (1.6) 3.9 (1.6) 3.6 (1.3) 8.417 Nociceptive versus Neuropathic < .001
Tibialis anterior algometry (kgf ) 4.9 (1.9) 4.5 (2.0) 4.0 (1.5) 3.961 Nociceptive versus Neuropathic .003
Within-group change
Dorsal forearm algometry (kgf ) 0.3 (1.2) 0.3 (1.1) -0.0 (1.1) 1.829 - .162
Tibialis anterior algometry (kgf ) 0.3 (1.3) 0.3 (1.1) 0.4 (1.1) 0.221 - .802
Data are presented as mean (SD) for continuous variables

Table 3 Sensitivity, specificity, positive predictive value, Strengths and limitations of the study
negative predictive value, and accuracy are the two predefined We acknowledge the strengths and limitations of the
cutoff points of the painDETECT questionnaire for detecting of present study. First, this study’s originality verified the
impairment of the conditioned pain modulation diagnostic accuracy of the painDETECT questionnaire
painDETECT painDETECT to detect impairment of conditioned pain modulation.
12 19 Second, we used conditioned pain modulation, a reli-
Sensitivity %, (95% CI) 46.6% (33.6 – 60.0) 21.6% (12.0 – 34.2) able measure [47] through a psychophysical test (cold
Specificity %, (95% CI) 43.1% (36.8 – 49.5) 78.6% (73.0 – 83.5) pressor test), to detect the impairment of the condi-
Positive Likelihood Ratio (95% CI) 0.8 (0.6 – 1.1) 1.0 (0.5 – 1.7) tioned pain modulation using two different anatomical
Negative Likelihood Ratio (95% CI) 1.2 (0.9 – 1.6) 1.0 (0.8 – 1.1) regions to ensure the appropriate classification of the
Impaired CPM Prevalence %, (95% CI) 19.4% (15.2 – 24.3) 19.4% (15.2 – 24.3) participants. Finally, the large sample size can be con-
Positive Predictive Value (95% CI) 16.5% (12.9 – 21.0) 19.7 (12.5 – 29.5) sidered a strength of this study. The main limitation of
Negative Predictive Value (95% CI) 76.9% (71.7 – 81.5) 80.5% (78.1 – 82.7) the study is that the cold pressor test and the painDE-
Accuracy (95% CI) 43.8% (38.2 – 49.5) 67.5% (61.9 – 72. 7) TECT questionnaire are not gold-standard for diagnos-
Abbreviation: CPM Conditioned pain modulation ing the impairment of conditioned pain modulation and
neuropathic pain, respectively. Treede suggested that
an experiment with secondary hyperalgesia induced by
neuropathic-like symptoms in people with musculoskeletal intradermal capsaicin injection is the only documented
pain. Similar pain characteristics are present in musculo- occurrence of central sensitisation that meets its for-
skeletal pain conditions. Using the painDETECT question- mal definition [48]. Nevertheless, the cold pressor test
naire, the physiotherapist can classify the people according is the most common method used [35] and has good to
to the pain phenotype. Accordingly, the physiotherapist can excellent intra-session reliability in healthy and people
offer adequate treatment strategies to a given person. with chronic pain [49] for conditioned pain modulation
Researchers should use instruments with high accuracy assessment. Also, the painDETECT questionnaire can
to assess the presence of CS-related signs and symptoms identify neuropathic-like symptoms, but positive neuro-
and neuropathic-like symptoms to confirm the present pathic classification in the painDETECT is insufficient to
study’s findings. Moreover, future studies should concen- classify neuropathy [50].
trate on methods to pragmatically characterise people with
impairment of conditioned pain modulation to facilitate the
decision-making of physiotherapists. Finally, the diagnostic Conclusion
accuracy of the painDETECT is only one of the considera- The painDETECT questionnaire seems valuable for rul-
tions when determining a screening tool for musculoskele- ing out people with musculoskeletal pain and impair-
tal pain. Therefore, additional aspects should be considered. ment of conditioned pain modulation.
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 8 of 9

Abbreviations References
ANOVA Analysis of variance 1. Bittencourt JV, Bezerra MC, Pina MR, Reis FJJ, de Sá FA, Nogueira LAC. Use
CI Confidence Interval of the painDETECT to discriminate musculoskeletal pain phenotypes.
CS Central Sensitisation Arch Physiother. 2022;12(1):1–8.
CPM Conditioned Pain Modulation 2. Guntel M, Huzmeli ED, Melek I. Patients With Neuropathic Pain Have Poor
JASP Jeffreys’s Amazing Statistics Program Sleep Quality. J Nerv Ment Dis. 2021;209(7):505–9.
Kg Kilograms 3. Verriotis M, Peters J, Sorger C, Walker SM. Phenotyping peripheral neuro-
Kgf Kilograms-force pathic pain in male and female adolescents: pain descriptors, somatosen-
Kg/m2 Kilograms-force per square meter sory profiles, conditioned pain modulation, and child-parent reported
SD Standard Deviation disability. Pain. 2021;162(6):1732–48.
STARD Standards for Reporting of Diagnostic Accuracy Studies 4. Melek LN, Smith JG, Karamat A, Renton T. Comparison of the neuropathic pain
symptoms and psychosocial impact of Trigeminal Neuralgia and Painful Post-
Traumatic Trigeminal Neuropathy. J Oral Facial Pain Headache. 2019;33(1):77–88.
Supplementary Information 5. Cherif F, Zouari HG, Cherif W, Hadded M, Cheour M, Damak R. Depression
The online version contains supplementary material available at https://​doi.​ prevalence in neuropathic pain and its impact on the quality of life. Pain
org/​10.​1186/​s40945-​023-​00171-8. Res Manag. 2020;2020:1–8.
6. Freynhagen R, Baron R, Gockel U, Tölle TR. Pain DETECT: a new screening
Additional file 1. STARD checklist. questionnaire to identify neuropathic components in patients with back
pain. Curr Med Res Opin. 2006;22(10):1911–20.
7. Freynhagen R, Tölle TR, Gockel U, Baron R. The painDETECT project - Far
Acknowledgements more than a screening tool on neuropathic pain. Curr Med Res Opin.
Not applicable. 2016;32(6):1033–57.
8. Hiyama A, Katoh H, Sakai D, et al. Clinical impact of JOABPEQ mental
Authors’ contributions health scores in patients with low back pain: analysis using the neuro-
Conceptualisation, J.V.B., E.G.L., and L.A.C.N.; Methodology, J.V.B., E.G.L., A.S.F., pathic pain screening tool painDETECT. J Orthop Sci. 2017;22:1009–14.
and L.A.C.N.; Investigation, J.V.B., E.G.L., and L.A.C.N.; Writing – Original Draft, 9. Middlebrook N, Rushton AB, Abichandani D, Kuithan P, Heneghan NR, Falla
J.V.B., E.G.L., A.S.F., and L.A.C.N.; Writing – Review & Editing, J.V.B., E.G.L., A.S.F., D. Measures of central sensitization and their measurement properties in
and L.A.C.N. musculoskeletal trauma: A systematic review. Eur J Pain. 2021;25(1):71–87.
10. Mertens MG, Hermans L, Crombez G, Goudman L, Calders P, Van Oosterwijck
Funding J, et al. Comparison of five conditioned pain modulation paradigms and
This study was financed in part by the Fundação Carlos Chagas Filho de influencing personal factors in healthy adults. Eur J Pain. 2021;25(1):243–56.
Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) [Grant number: 11. Bannister K, Dickenson AH. What the brain tells the spinal cord. Pain.
E-26/211.104/2021] and Coordenação de Aperfeiçoamento de Pessoal 2016;157(10):2148–51.
de Nível Superior—Brasil (CAPES) [Finance Code 001; Grant number: 12. Ossipov MH, Morimura K, Porreca F. Descending pain modulation and
88881.708719/2022–01, grant number: 88887.708718/2022–00, and grant chronification of pain. Curr Opin Support Palliat Care. 2014;8(2):143–51.
number 88887.466981/2019–00]. 13. Arias-Buría JL, Ortega-Santiago R, De-la-Llave-Rincón AI. Understand-
ing central sensitization for advances in management of carpal tunnel
Availability of data and materials syndrome. 2020;9:1–7.
Not applicable. 14. Niesters M, Proto PL, Aarts L, Sarton EY, Drewes AM, Dahan A. Tapentadol
potentiates descending pain inhibition in chronic pain patients with
diabetic polyneuropathy. Br J Anaesth. 2014;113(1):148–56.
Declarations 15. Niesters M, Aarts L, Sarton E, Dahan A. Influence of ketamine and
morphine on descending pain modulation in chronic pain patients: A
Ethics approval and consent to participate randomized placebo-controlled cross-over proof-of-concept study. Br J
Research involving human subjects complied with all relevant national regula- Anaesth. 2013;110(6):1010–6.
tions, institutional policies and is in accordance with the tenets of the Helsinki 16. Seifert F, Kiefer G, Decol R, Schmelz M, Maihöfner C. Differential endog-
Declaration (as amended in 2013), and has been approved by equivalent enous pain modulation in complex-regional pain syndrome. Brain.
research ethical committee (number: 02228818.0.3001.5258). Informed con- 2009;132(3):788–800.
sent has been obtained from all individuals included in this study. 17. Lluch E, Torres R, Nijs J, Van Oosterwijck J. Evidence for central sensitiza-
tion in patients with osteoarthritis pain: a systematic literature review. Eur
Consent for publication J pain. 2014;18(10):1367–75.
Not applicable. 18. Sasaki E, Ota S, Chiba D, Kimura Y, Sasaki S, Ando M, et al. Association
between central sensitization and increasing prevalence of nocturnal
Competing interests knee pain in the general population with osteoarthritis from the Iwaki
Each author certifies that he or she, or a member of his or her immediate fam- Cohort Study. J Pain Res. 2021;14:2449.
ily, has no commercial association (i.e., consultancies, stock ownership, equity 19. Chimenti RL, Frey-Law LA, Sluka KA. A mechanism-based approach to
interest, patent/licensing arrangements, etc.) that might pose a conflict of physical therapist management of pain. Phys Ther. 2018;98(5):302–14.
interest in connection with the submitted manuscript. 20. Moreton BJ, Tew V, Das Nair R, Wheeler M, Walsh DA, Lincoln NB. Pain
phenotype in patients with knee osteoarthritis: Classification and meas-
Author details urement properties of painDETECT and self-report leeds assessment of
1
Rehabilitation Science Postgraduate Program at Augusto Motta Uni- neuropathic symptoms and signs scale in a cross-sectional study. Arthritis
versity Centre (UNISUAM), Avenida Paris, 84, Bonsucesso, CEP, Rio de Care Res. 2015;67(4):519–28.
Janeiro 21041‑020, RJ, Brasil. 2 Physiotherapy Department at Federal Institute 21. Gwilym SE, Keltner JR, Warnaby CE, Carr AJ, Chizh B, Chessell I, et al. Psy-
of Rio de Janeiro (IFRJ), Rio de Janeiro, Brazil. chophysical and functional imaging evidence supporting the presence
of central sensitization in a cohort of osteoarthritis patients. Arthritis Care
Received: 30 January 2023 Accepted: 17 August 2023 Res. 2009;61(9):1226–34.
22. Hochman JR, Davis AM, Elkayam J, Gagliese L, Hawker GA. Neuropathic
pain symptoms on the modified painDETECT correlate with signs of cen-
tral sensitization in knee osteoarthritis. Osteoarthr Cartil. 2013;21(9):1236–
42. https://​doi.​org/​10.​1016/j.​joca.​2013.​06.​023.
Bittencourt et al. Archives of Physiotherapy (2023) 13:17 Page 9 of 9

23. Bossuyt PM, Reitsma JB, Bruns DE, Bruns DE, Glasziou PP, Irwig L, et al. 42. Freynhagen R, Baron R. The evaluation of neuropathic components in
STARD 2015: An updated list of essential items for reporting diagnostic low back pain. Curr Pain Headache Rep. 2009;13(3):185–90.
accuracy studies1. Radiology. 2015;277(3):826–32. 43. Woolf CJ. Central sensitization: implications for the diagnosis and treat-
24. Murray CCJL, Abraham J, Ali MK, Alvarado M, Atkinson C, Baddour LM, ment of pain. Pain. 2011;152(3):S2-15.
et al. The state of US health, 1990–2010: burden of diseases, injuries, and 44. Granovsky Y. Conditioned pain modulation: a predictor for develop-
risk factors. JAMA - J Am Med Assoc. 2013;310(6):591–608. Available from: ment and treatment of neuropathic pain. Curr Pain Headache Rep.
https://​jama.​jaman​etwork.​com/​data/​Journ​als/​JAMA/​927436/​joi13​0037.​ 2013;17(9):1–7.
pdf%​5Cn http://​ovidsp.​ovid.​com/​ovidw​eb.​cgi?T=​JS&​PAGE=​refer​ence&D=​ 45. Gagné M, Côté I, Boulet M, Jutzeler CR, Kramer JLK, Mercier C. Condi-
emed1​1&​NEWS=​N&​AN=​20135​03627%​5Cn http://​ovidsp.​ovid.​com/​ovidw​ tioned pain modulation decreases over time in patients with neuro-
eb.​cgi?T=​JS&​PAGE=​refer​ence&D=​medl&​NEWS=​N&​AN=​23842​577. pathic pain following a spinal cord injury. Neurorehabil Neural Repair.
25. Hawker GA, Mian S, Kendzerska T, French M. Measures of adult pain: Visual 2020;34(11):997–1008.
analog scale for pain (vas pain), numeric rating scale for pain (nrs pain), 46. Granovsky Y, Nahman-Averbuch H, Khamaisi M, Granot M. Efficient
mcgill pain questionnaire (mpq), short-form mcgill pain questionnaire conditioned pain modulation despite pain persistence in painful diabetic
(sf-mpq), chronic pain grade scale (cpgs), short form-36 bodily pain scale neuropathy. Pain Rep. 2017;2(3):1–7.
(sf. Arthritis Care Res (Hoboken). 2011;63(S11):S240–52. 47. Kennedy DL, Kemp HI, Ridout D, Yarnitsky D, Rice ASC. Reliability of condi-
26. Merskey H, Bogduk N. International association for the study of pain. Task tioned pain modulation: A systematic review. Pain. 2016;157(11):2410–9.
force on taxonomy. Classification of chronic pain: descriptions of chronic 48. Treede RD. Gain control mechanisms in the nociceptive system. Pain.
pain syndromes and definitions of pain terms. 1994. 2016;157(6):1199–204.
27. do Rio JPM, Bittencourt JV, Corrêa LA, Freynhagen R, dos Reis FJJ, de Melo 49. Nuwailati R, Bobos P, Drangsholt M, Curatolo M. Reliability of conditioned
TB, et al. Cross-cultural adaptation of the painDETECT questionnaire into pain modulation in healthy individuals and chronic pain patients: a
Brazilian Portuguese. Braz J Anesthesiol. 2022;72:44–48. systematic review and meta-analysis. Scand J Pain. 2022;22(2):262–78.
28. Vaegter HB, Handberg G, Kent P. (345) Brief psychological screening ques- 50. Hasvik E, Haugen AJ. Call for Caution in Using the Pain DETECT Ques-
tions can be useful for ruling out psychological conditions in patients tionnaire for Patient Stratifi cation Without Additional Clinical Assess-
with chronic pain. J Pain. 2017;18(4):S61. ments: Comment on the Article by Soni et al. Arthritis Rheumatol.
29. Kent P, Mirkhil S, Keating J, Buchbinder R, Manniche C, Albert HB. The 2019;71(7):1201–2.
concurrent validity of brief screening questions for anxiety, depression,
social isolation, catastrophization, and fear of movement in people with
low back pain. Clin J Pain. 2014;30(6):479–89 (2013/11/28). Publisher’s Note
30. Cappelleri JC, Koduru V, Bienen EJ, Sadosky A. Characterizing neuropathic Springer Nature remains neutral with regard to jurisdictional claims in pub-
pain profiles: enriching interpretation of painDETECT. Patient Relat Out- lished maps and institutional affiliations.
come Meas. 2016;7:93–9 (2016/07/28).
31. Packham TL, Cappelleri JC, Sadosky A, MacDermid JC, Brunner F.
Measurement properties of painDETECT: Rasch analysis of responses
from community-dwelling adults with neuropathic pain. BMC Neurol.
2017;17(1):1–9.
32. Abu-Shaheen A, Yousef S, Riaz M, Nofal A, AlFayyad I, Khan S, et al.
Testing the validity and reliability of the Arabic version of the painDE-
TECT questionnaire in the assessment of neuropathic pain. PLoS ONE.
2018;13(4):1–13.
33. Freynhagen R, Tölle TR, Gockel U, et al. The painDETECT project–far
more than a screening tool on neuropathic pain. Curr Med Res Opin.
2016;32:1033–57.
34. Lewis GN, Heales L, Rice DA, Rome K, McNair PJ. Reliability of the condi-
tioned pain modulation paradigm to assess endogenous inhibitory pain
pathways. Pain Res Manag. 2012;17(2):98–102.
35. Lewis GN, Rice DA, McNair PJ. Conditioned pain modulation in popula-
tions with chronic pain: a systematic review and meta-analysis. J pain.
2012;13(10):936–44 (10.1016/j.jpain.2012.07.005).
36. Yarnitsky D, Bouhassira D, Drewes AM, Fillingim RB, Granot M, Hansson
P, et al. Recommendations on practice of conditioned pain modulation
(CPM) testing. Eur J Pain (United Kingdom). 2015;19(6):805–6.
37. Gervais-Hupé J, Pollice J, Sadi J, Carlesso LC. Validity of the central sen-
sitization inventory with measures of sensitization in people with knee
osteoarthritis. Clin Rheumatol. 2018;37(11):3125–32.
38. Nogueira LAC, Chaves ADO, Wendt ADS, De SRLS, Reis FJJ, De AFG, et al.
Central sensitization patients present different characteristics compared
with other musculoskeletal patients: A case–control study. Eur J Physi-
Ready to submit your research ? Choose BMC and benefit from:
other. 2016;18(3):147–53.
39. Rifbjerg-Madsen S, Christensen AW, Boesen M, Christensen R, Danneski-
• fast, convenient online submission
old-Samsøe B, Bliddal H, et al. Can the painDETECT Questionnaire score
andMRI help predict treatment outcome in rheumatoid arthritis: Protocol • thorough peer review by experienced researchers in your field
for the Frederiksberg hospital’s Rheumatoid Arthritis, pain assessment • rapid publication on acceptance
and Medical Evaluation (FRAME-cohort) study. BMJ Open. 2014;4(11):1–8.
• support for research data, including large and complex data types
40. Soni A, Wanigasekera V, Mezue M, Cooper C, Javaid MK, Price AJJ,
et al. Central Sensitization in Knee Osteoarthritis: Relating Presurgical • gold Open Access which fosters wider collaboration and increased citations
Brainstem Neuroimaging and PainDETECT-Based Patient Stratification to • maximum visibility for your research: over 100M website views per year
Arthroplasty Outcome. Arthritis Rheumatol. 2019;71(4):550–60.
41. Sachau J, Bruckmueller H, Gierthmühlen J, Magerl W, May D, Binder A, At BMC, research is always in progress.
et al. The serotonin receptor 2A (HTR2A) rs6313 variant is associated with
higher ongoing pain and signs of central sensitization in neuropathic Learn more biomedcentral.com/submissions
pain patients. Eur J Pain. 2021;25(3):595–611.