Algometry for the assessment of central sensitisation to pain in fibromyalgia patients: a systematic review

Abstract Introduction The pathophysiology of fibromyalgia (FM) is related to central sensitisation (CS) to pain. Algometry allows assessing CS based on dynamic evoked pain. However, current algometrýs protocols require optimising, unifying and updating. Objectives 1) identify the dynamic pain measures used most frequently to effectively assess CS processes in FM, and 2) consider the future of the algometry assessing CS in these patients. Methods Cochrane Collaboration guidelines and PRISMA statements were followed. The protocol was registered in PROSPERO database (ID: CRD42021270135). The selected articles were evaluated using the Cochrane risk of bias (ROB) assessment tool. The PubMed, Scopus, and Web of Science databases were searched. Results Thirty-four studies were selected, including measures such as temporal summation of pain (TSP), aftersensations (AS), spatial summation of pain (SSP), the noxious flexion reflex (NFR) threshold, conditioned pain modulation (CPM), cutaneous silent period (CuSP), and slowly repeated evoked pain (SREP); and evoked pain combined with neuroimaging. Each measure offered various advantages and limitations. According to ROB, 28 studies were of low quality, 3 of moderate quality, and 3 of high quality. Conclusions Several pain indicators have been demonstrated to successfully examine CS involvement in FM in the last years. Algometry, especially when it involves diverse body sites and tissues, might provide further insight into (1) the evaluation of psychological factors known to influence pain experience, (2) new dynamic pain indicators, and (3) the simultaneous use of certain neuroimaging techniques. Further research clarifying the mechanisms underlying some of these measures, and homogenisation and optimisation of the algometrýs protocols, are needed. KEY MESSAGES Algometry allows for assessing Central Sensitisation by applying dynamic evoked pain. The future of algometry could relapse in its combination with neuroimaging. Recently-emerged pain indicators should be considered for algometrýs new protocols.


Introduction
Fibromyalgia (FM) is a syndrome characterised by widespread, persistent and diffuse pain that is usually accompanied by a wide range of physical (e.g. fatigue, insomnia and stiffness) and psychological (e.g. anxiety, depression and cognitive alterations) symptoms [1][2][3], which are of relevance for its diagnosis [3]. FM seems to be more prevalent in females than males (9:1) [4]. Another hallmark of FM is pain-related hyperresponsivity [5], which underlies two of its characteristic symptoms: (1) allodynia, defined as increased sensitivity to innocuous stimulation; and (2) hyperalgesia or excessive response to painful stimulation, which may be primary or secondary [6]. Primary hyperalgesia refers to sensitisation of nociceptors (peripheral sensitisation), whereas secondary hyperalgesia involves increased responsiveness of the central nervous system (CNS) to painful stimulation (central sensitisation, CS) [7]. While the aetiology of FM remains unknown, its pathophysiology has been widely related to CS processes [8,9]. CS can be defined as an increase in CNS pain responsiveness as a result of altered nociception at the brain and spinal cord levels [7]. Three main CS processes have been distinguished: (1) long-term potentiation of synaptic transmission in the CNS, (2) activation of certain pain facilitatory pathways outside the receptive fields, and (3) dysfunction in descending inhibitory pain mechanisms [10].
Algometry has proven to be a suitable instrument to assess central allodynia and hyperalgesia in FM, as well as in other pain conditions [11,12]. Algometry is often indistinctly referred to as dolorimetry or quantitative sensory testing (QST), with both of those terms also being used in the pain literature [13,14]. While the tender point count (as used for diagnosing FM) is limited only to the evaluation of certain spontaneous pain points via the application of pressure [2], algometry evaluates and detects abnormal nociception using various types of pain stimulation in different body locations [15,16], providing a more sensitive, in-depth and exhaustive pain evaluation [17,18]. Besides, pain responses evaluated by algometry have shown strong associations with the clinical pain in FM [19], as well as with improved clinical outcomes after intervention [20].
Algometry examines responses to evoked painful stimulation in order to estimate/quantify the underlying mechanisms involved in different pain conditions. In this way, though algometry has the advantage of offering a reliable and controlled application of pain stimulation [13], it is not always able to discriminate between FM and other similar pain conditions [20,21], being necessary for its combination with tender point count for achieving a high diagnostic accuracy [22].
Several factors have to be considered in algometry, including the type of stimulation (heat, cold, pressure/ mechanical, electrical, etc.), algometer surface, application rate, anatomic location of stimulation, and type of tissue to be tested (skin, subcutaneous/deep muscle, bone, or peripheral nerves) [9,23,24]. Furthermore, the moderating role of psychological factors such as pain catastrophizing, fear of pain, anxiety and depression, all of which have a proven influence on pain responsiveness [25,26], should be considered and measured.
Since its inception, algometry has mostly been used as an evoked measure of static pain, evaluating pain response in a basal nociceptive state [27,28]. Threshold and tolerance, defined as the minimal stimulation to induce pain and maximum painful stimulation that can be endured, respectively, are the most common static evoked pain measures [29]. Under different pain modalities, both have been demonstrated to be significantly lower in FM patients than in healthy controls (HC) [18,30,31]. However, neither parameter relates specifically to CS [17], and both are influenced by individual factors such as somatosensory sensitivity (e.g. the higher the mechanosensitivity, the lower the pressure pain threshold), muscle strength, and the peripheral impulse input to the stimulated tissue [24,32,33]. In fact, the evaluation of CS requires an assessment of the increase in pain responsiveness to repeated painful stimulation [8], which may be determined via dynamic pain indicators [24,27,28,34,35]. Dynamic evoked pain measures facilitate the assessment of CS processes, such as the increased activity of ascending pain pathways and diminished activity of descending pain mechanisms, as measured by temporal summation of pain (TSP) [35][36][37], and conditioned pain modulation (CPM) [38,39], respectively. Therefore, the measures of dynamic pain seem suitable to evaluate FM as a CS-related chronic pain condition.
However, although the utility of algometry to examine pathophysiological mechanisms of FM pain is not in dispute, no recent systematic review has examined the algometry techniques used most widely to assess the specific involvement of CS processes in these patients. A systematic review exploring and synthetising the varied data obtained over the last few years on the use of evoked pain measures for assessing CS in FM is needed to optimise, unify and update current QST protocols. Therefore, to determine the potential of algometry as a tool to assess CS, a systematic review of studies published in the last decade using evoked pain measures to assess CS in FM was conducted. The present systematic review aimed to: (1) identify the dynamic evoked pain measures used most frequently to effectively assess CS processes in FM, by reviewing findings published in the past decade, and (2) assess the future potential of algometry as a method to assess CS involvement in these patients.

Search strategy
This systematic review was conducted according to the guidelines of the Cochrane Collaboration and reported based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [40]. The inclusion criteria and analyses were specified in advance, and the protocol was registered in the Prospective Register of Systematic Reviews (PROSPERO) international database (registration ID: CRD42021270135). The search string was as follows: central pain sensitisation AND fibromyalgia AND (algometry OR algometer OR dolorimetry OR dolorimeter OR evoked OR experimental OR elicited OR testing OR pressure OR mechanical OR thermal OR heat OR electrical OR cold).
The PubMed, Scopus, and Web of Science (WOS) databases were searched independently by two researchers. Discrepancies were resolved by consensus. Two reviewers (P.d.l.C. and C.M.G.S.) independently screened all articles and selected those that satisfied the inclusion criteria for full-text analysis. The titles and abstracts of the articles were screened to remove irrelevant studies (i.e. those clearly unrelated to the aims of this review); the remaining shortlisted articles were screened in depth for eligibility. In order to compile the final set of articles to be reviewed, the full texts of relevant articles were retrieved and screened based on the inclusion and exclusion criteria. Both reviewers decided whether the articles should be included or excluded; any discrepancies were reviewed by the senior author (G.A.R.d.P.), who was the final arbiter regarding the inclusion of a study. The article screening and inclusion processes are illustrated as a PRISMA flowchart ( Figure 1). Before data extraction and quality assessment, C.I.M. screened all articles in order to confirm their eligibility for this study. The search was restricted to articles published in the past 2011-2020 decade, and 2021 years. The last search was conducted on December 31, 2021.

Eligibility criteria
Studies were included if they were: (1) peer-reviewed original studies using algometry techniques to assess CS processes in FM patients; (2) full-text publications; (3) published in the period 2011-2021; and (4) published in English. The following studies were excluded: (1) reviews and meta-analyses; (2) comments, editorials, case reports, letters, and meeting abstracts; (3) non-human studies; and (4) non-quantitative studies.  ❑TSP/WU. ❑Activation of pain-related brain areas by fMRI during TSP.
Primary Outcomes: ᭜FM patients: TSP was not significantly higher than for HC, neither TSP group differences were related to brain activity.
❑NFR threshold. Primary Outcomes: ᭜FM patients: greater TSP than HC, and the same one that for chronic whiplash patients. ᭜Non-significant group differences in CPM.
(continued) (in a ensitizat order). ᭜Painful (or not) ¼ 5N above/below the threshold. ᭜Different specific location for each series. ᭜A 0-100 NRS.
❑O2Hb levels of pain-related brain areas implicated in pain processing to repeated painful pressure stimulation.
Primary Outcomes: ❑FM patients: augmented cerebral activation of pain-related brain areas (especially in the dorsolateral prefrontal cortex) in comparison to MD patients and HC, using identical (subjective) painful stimulation. Bosma et al. (2016), Canada [44]. (continued)

Data extraction and quality assessment
The study characteristics, methodologies and results were extracted independently by P.d.l.C. and C.M.G.S., and any discrepancies were reviewed by G.A.R.d.P. Data were extracted in the following sequence: first author (publication year), country, sample size, age (mean ± SD), sex (%), condition/s, pain modality, body area, device/tool used to evoke pain, details of the procedure, CS measure, primary and secondary outcomes. The study characteristics are shown in Table 1. The data were reviewed by C.I.M. to ensure the accuracy of the extraction thereof. In order to evaluate the quality of the selected articles, both C.M.G.S. and C.I.M. independently evaluated the risk of bias (ROB) in each study according to the Cochrane ROB assessment tool. Additionally, P.d.l.C. or G.A.R.d.P. reviewed articles listing C.M.G.S. or C.I.M. as authors. The Cochrane ROB tool contains seven evaluation items: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other bias. For each item, the ROB was graded as high, medium or low. Discrepancies were resolved by discussion with the senior author (G.A.R.d.P.), who made the final decision.

Data synthesis
In line with the stated aims of this review, whether the studies used algometry specifically to assess CS to pain in FM patients was analysed. This was done for each individual article identified by the search according to the PRISMA flow diagram (first: title; second: abstract; third: text). The included studies are detailed in Table 1, and evaluated in terms of reporting biases in the Risk of Bias section and Table 2.

Literature search and study characteristics
A total of 572 articles were identified by database searches, 376 of which were finally selected for screening after removing duplicates. Figure 1 displays the PRISMA flow chart, which details the number of studies excluded at each stage of the screening process. An analysis of 40 full-text articles was performed to determine their eligibility for the review. Among these articles, 34 fulfilled the inclusion criteria and were subjected to the data extraction (Table 1) and quality assessment processes (

Participants
The 34 selected studies on the use of algometry to measure CS to pain in FM included 2177 participants (average of 64 subjects per study; range: 14-236 subjects). Approximately half of the participants were FM patients (n ¼ 1134); the others were controls: HC (n ¼ 678) and individuals with other chronic medical conditions (n ¼ 365). The weighted mean age of all participants was 45.03 ± 7.46 years. In all studies, the age was similar between the patient and control groups, except in 3 studies that reported a significant group difference in age that was not fully controlled for [45,46,58], and 2 studies that did not report participant age [41,52]. Regarding participant sex, 21 studies enrolled only women, 12 other studies had a clear female predominance (59-96%), and 1 study did not report this data [52] (see Table 1).

Temporal summation of pain and indicators derived therefrom
TSP can be defined as an increase in pain perception due to repeated stimulation (> 0.33 Hz), as opposed to the magnitude of these stimuli. The TSP response is based on the wind-up (WU) effect of neurons of the dorsal horn of the spinal cord [75]. It should be noted that TSP usually refers to the temporal summation of second pain, which specifically involves C-fibers [76]. TSP is also referred to as WU [35], and multiple evoked pain indicators are derived from it. Examples include the WU function, which models the WU responses to several series of TSP [68], and the WU ratio, which is based on a comparison between pain ratings in response to a TSP procedure and pain ratings for single stimuli (of the same intensity and in the same body location) [43]. Thirteen articles that used TSP/WU, WU function or the WU ratio as indicators of CS of pain in FM patients were included in the review (see Table 1 for details) [43][44][45][50][51][52]55,59,60,63,66,68,69]. Eight of those studies reported a higher TSP/WU, WU ratio or WU function in FMS patients compared to HC [44,45,52,59,66,68] and patients with other medical conditions (chronic low back pain, chronic whiplash, or rheumatoid arthritis) [50,55,59]. However, in three studies, TSP and the WU ratio did not differ between FM, HC, and patients with other medical conditions (chronic fatigue, chronic low back pain, myofascial syndrome, or temporomandibular disorder) [43,45,60,69]. Secondary findings included an increase in TSP after relaxation and cognitive stress tasks in FM patients, a reduction in TSP after these procedures in HC and chronic whiplash patients [51]; and a lack of relationship between TSP and severity of FM symptoms [45], which was contrary to the higher TSP in FM patients reporting more severe clinical pain (although around 33% of these patients did not show a CS response according to the WU ratio) [63].

Repetition-induced summation of activity-related pain
Another TSP-related indicator, which probably also has the WU effect as its underlying mechanism, is the repetition-induced summation of activity-related pain (RISP) [61]. RISP can be defined as a progressive increase in pain ratings over successive weight lifts, even when the overall physical demands of the task remain constant [77].
One article using RISP as an indicator of CS in FM patients was included in our review [61] (Table 1). That study showed that RISP was significantly higher in FM patients than in chronic low back pain patients. Furthermore, RISP was associated with pain disability, but not with depression or catastrophizing in FM patients [61].

Aftersensations (as)
Aftersensations (AS) can be defined as an increase in pain ratings that occurs some seconds after each series of TSP. AS seems to depend on after-discharges of dorsal horn neurons following repetitive C-fiber activation [37].
Five articles that used AS as indicator of CS in FM patients were included in our review [44,60,66,68,69] ( Table 1). Among these articles, four reported higher AS in FMS patients compared to HC [44,66,68,69], while and did not show a difference in AS between TMD patients suffering versus not suffering from FM [60]. A secondary finding was that AS were positively associated with clinical pain in FM patients [66,68].

Spatial summation of pain
Spatial summation of pain (SSP) is defined as an increase in pain intensity due to an increase in the size of the stimulated body area, which also depends on the activation of C-fiber in the spinal cord [78,79]. SSP also seems to involve the central mechanism known as diffuse noxious inhibitory control. As well as a progressive increase in the size of the stimulated area, factors like pain intensity and surface size also are relevant to evoking SSP [80].
Two articles that used SSP as an indicator of CS in FM patients were included in our review [59,72] (Table  1); neither article reported significantly higher SSP in FM patients compared to HC and chronic low back patients [59]. However, Van Oosterwijck et al. [72] found lower SSP in FM patients who had received an educational intervention on the physiology of pain.

Conditioned pain modulation
CPM is defined as inhibition of the pain produced by a painful (test) stimulus due to the presentation of another (conditioning) stimulus. CPM reflects the level of pain reduction elicited by stimulus interference [81]. Although CPM is mainly used as an inhibitory pain modulation paradigm, its underlying processes involve both complex central facilitatory and inhibitory mechanisms of pain processing, such as diffuse noxious inhibitory control; thus the specific mechanisms involved in this pain measure have not yet been elucidated [81,82].
Eight articles that used CPM as an indicator of CS in FM patients were included in our review [45,46,48,50,51,59,67,73] (Table 1). Among these studies, three reported significantly lower CPM in FMS patients compared to HC and patients with other medical conditions (major depressive disorder or rheumatoid arthritis) [45,46,48]. Two studies did not report CPM differences among FM patients, HC and patients with other medical conditions (chronic low back pain or chronic whiplash) [50,59]. Moreover, a study in which CPM was used as a marker of clinical improvement in FM patients after pregabalin treatment reported a significant recovery in CPM after the treatment [73]. Lower CPM after relaxation and stressful tasks in FM, HC and chronic whiplash patients were observed in another study, but there were no significant differences in CPM among these groups [51]. As secondary findings, CPM was inversely associated with brain-derived neurotrophic factor (BDNF) in three studies [46,48,67], and with some clinical FM symptoms in one study [67]. Additionally, CPM seemed to be moderated by depression levels [48].

Noxious flexion reflex threshold
The noxious flexion reflex (NFR) threshold is defined as the intensity of electrical stimulation required to elicit the NFR. NFR is a polysynaptic spinal withdrawal reflex elicited by the activation of nociceptive A-delta afferents (monitored using electromyography) in response to electrocutaneous stimulation of varying intensity applied to the ipsilateral sural nerve of the biceps femoris muscle. Thus, the NFR threshold indicates the magnitude of the excitatory component of the spinal protective reflex [83]. However, there is no consensus definition of the NFR threshold, varying according to factors such as the electromyographic criteria and the required duration of the NFR response. This makes it difficult to compare this indicator between studies [84].
Six articles using the NFR threshold as an indicator of CS in FM patients were included in our review [42,49,57,64,71,74] (Table 1). Among these studies, two reported a lower NFR threshold in FM patients compared to HC [49,57], while three did not observe NFR threshold differences between FM patients and HC [64,71,74]. At this juncture, it should be noted that Ang et al. found that depression had a confounding effect on the inverse relationship between the NFR threshold and FM pain; this association was only observed in non-depressed FM patients [42]. As a secondary finding, a lower NFR threshold was associated with more severe clinical pain [57,71] and pain sensitivity [49], although there were no associations with other clinical FM symptoms [49]. Some studies showed that depression and emotional states modulate these relationships [57,64,71]. Additionally, and in accordance with the influence of depression on the NFR response, FM patients with lower NFR thresholds were more likely to possess the Met/Met genotype of the COMT Val158Met polymorphism (this was less common among individuals with depression) [57].

Cutaneous silent period
The cutaneous silent period (CuSP) can be defined as the brief interruption in voluntary muscle contraction that follows the painful electrical stimulation of a cutaneous nerve, i.e. the inhibitory component of the NFR [85]. The mechanisms underlying CuSP are related to sensorimotor integration processes occurring at the spinal and supraspinal levels [86].
Two articles using CuSP as an indicator of CS in FM patients were included in our review [41,65] (Table 1). Differences in CuSP onset latencies between FM and HC, but not in CuSP duration, were found by Sahin et al. [65]. Contrary, Al-Mahdawi et al. observed differences in CuSP duration between FM and HC, but not in CuSP onset latencies [41].

Slowly repeated evoked pain sensitisation
Slowly repeated evoked pain (SREP) sensitisation is reflected in an increase in pain ratings in response to repeated evoked pressure stimuli at a frequency of 0.03 Hz. The mechanisms underlying SREP sensitisation are still unknown. This dynamic pain indicator is currently in the latter stages of development. Although a lower stimulation frequency is used in the SREP protocol compared to TSP, WU is not considered a possible underlying mechanism of SREP sensitisation. Longterm potentiation of CS processes has been proposed [54]. SREP sensitisation showed ! 85% diagnostic accuracy for discriminating FM patients from HC and patients with other non-CS chronic pain conditions [53,55,56].
Four articles using SREP sensitisation as a CS indicator in FM were included in our review [53][54][55][56] (Table  1). All of these articles reported greater SREP sensitisation in FM patients compared to HC [53,54,56] and rheumatoid arthritis patients [55,56]. In fact, SREP sensitisation response was only found in FM patients in all of the studies. Regarding secondary findings, SREP sensitisation was positively associated with clinical pain levels [53,56] and other typical FM symptoms [56]. Furthermore, SREP sensitisation was inversely related to blood pressure [54] and proved to be a reliable measure [55].

Evoked pain combined with neuroimaging
Evoked pain measures have also been combined with neuroimaging techniques to assess CS in FM patients [87][88][89]. Eight such articles were included in our review; four of them used TSP plus functional magnetic resonance imaging (fMRI) [44,52,66,69], while two assessed pressure pain stimuli [70] and maintained cold stimulation [58] in combination with O 2 Hb brain levels using functional near-infrared (fNIR) spectroscopy, one combined secondary hyperalgesia and fMRI assessments [47], and another combined measures of pain perception to pressure stimulation and cerebral blow flow (CBF) using functional transcranial Doppler (fTCD) ultrasonography [62].
Regarding the first of the six articles introduced above, Burgmer et al. associated the more severe secondary hyperalgesia seen in FM patients compared to HC with impairment of central pain inhibition, in turn, related to the cerebral mid-brain mechanisms implicated in pain transmission [47]. Meanwhile, € Uceyler et al. found higher O 2 Hb levels in FM patients than HC in brain areas related to pain processing, such as the dorsolateral prefrontal cortex, in response to repeated painful stimuli [70]. In response to cold pain stimulation, Donadel et al. observed a lower change in O 2 Hb concentrations in the left prefrontal cortex of FM patients in comparison with HC, suggesting dysfunction in this inhibitory pain-related brain area [58]. This specific difference allows for discrimination greater than 80% between FM patients and HC [58]. Montoro et al. observed a complex pattern of temporal CBF changes in response to pressure pain stimulation, with alterations seen in most stages of the response in FM patients relative to HC; this suggested the presence of CS in FM [62]. Using fMRI combined with a TSP procedure, Bosma et al. observed a lower activation in dorsal horn and brain areas related to pain inhibition (the rostral ventromedial medulla and periaqueductal grey region) in FM patients than in HC [44]. Moreover, greater AS was associated with a greater influence of dorsal horn activity on pain [44]. Craggs et al. study was the only one that did not report differences between FM and HC in pain-related brain activity in response to a TSP procedure [52]. Schreiber et al. combined a TSP procedure with fMRI, which revealed greater brain activity of the medial temporal lobe in FM than in HC; this was also seen in the context of AS [66]. Lastly, Staud et al. observed a larger extent of spinal cord involvement and a greater activity of painrelated brain areas during TSP evocation in FM patients in comparison with HC [69]. More details are provided in Table 1.

Risk of bias
The Cochrane ROB evaluation was performed independently by C.M.G.S. and C.I.M., except for articles that they authored; in such cases, P.d.l.C. or G.A.R.d.P carried out the ROB evaluation. Thus, no article was evaluated by an author thereof. The initial agreement rate was 95%. The consensus was achieved either through discussing how the criteria were interpreted again or based on the input of the senior reviewer (G.A.R.d.P). The ROB evaluation revealed that 3 studies were of high quality [42,51,72], 3 were of moderate quality [46,54,59], and the reimaging 28 studies were of low quality. Details on the ROB assessments are shown in Table 2.

Discussion
In the present systematic review, the dynamic evoked pain measures used most frequently during the last few years to effectively assess CS processes in FM were identified. In addition, the future of algometry as a method to assess CS involvement in these patients was analysed. Studies published during the last decade including FM patients and dynamic evoked pain indicators, specifically designed to assess augmented responsiveness of the CNS, were reviewed.
Regarding the dynamic evoked pain measures used by the studies in this review, higher efficacy was seen for the TSP/WU, WU ratio, WU function, AS, and RISP (all of which are derived from or have potential as TSP-related measures) as indicators of CS in FM patients compared to other CS pain measures (e.g. SSP, CPM, and the NFR threshold). In general, the TSP-related pain measures demonstrated greater CSrelated pain responses in FM patients compared to HC [44,45,52,59,66,68,69] and other medical conditions [50,53,55]. In this regard, though TSP-related pain measures failed to reveal group differences in CS levels in four studies [43,45,60,69], one of them compared FM with a similar CS condition as chronic fatigue syndrome [45]. Strikingly, associations between most of those indicators (except AS) and the clinical pain levels of FM patients were rarely reported [68]. In spite of this, it is known that TSP has limited power to determine the extent of CS pain involvement in FM on an individual basis, i.e. it is most effective for assessing large samples of patients [90].
In contrast to the TSP-related indicators mentioned above, SSP was less effective for determining the levels of CS to pain in FM patients. Of the two studies using SPP, only one compared FM patients with controls (HC and a low back pain group); it reported no group differences [59]. The other study used SSP to quantify changes in CS levels after a physiologic education intervention in FM patients and demonstrated encouraging outcomes [72]. However, these findings should be treated with caution considering the small number of SSP studies included in the present systematic review.
Regarding CPM and the NFR threshold, although half of the studies reviewed reported greater CS in FM compared to HC and/or patients with other medical conditions [45,46,48,49,57], the other half did not observe group differences in CS levels [50,51,59,64,71,74]. Thus, the CPM and NFR findings may be considered equivocal. As a possible explanation for this, CPM and NFR share a limitation, namely wide variation in the protocols used to evoke NFR and CPM responses [84][85][86][87][88][89][90][91]. In comparison, TSP (based on the WU effect) is more easily evoked using protocols based on repeated stimulation at a frequency ! 0.33 Hz [8]. It should also be noted that, in the studies reporting lower CPM in FM patients, this was consistently associated with higher BDNF levels [46,48,67]; this could be interpreted as objective evidence that deficient CPM in FM patients can serve as a CS indicator. In line with this, recovery of CPM after treatment of FM patients has been demonstrated [73], as well as an inverse association between CPM and the severity of FM symptoms [67]. Nonetheless, Bourke et al. did not find associations between CPM and clinical pain, physical function, fatigue or anxiety in these patients, though inefficient CPM was related to enhanced TSP [45]. Regarding the NFR threshold, the reported modulatory effect of depressive mood and affective states on this pain indicator is also notable [42,57,64,71]. This confounding effect might preclude exploration of the associations among the NFR threshold, clinical pain [42], and pain sensitivity [49,74] in FM patients.
Despite the limited evidence regarding the efficacy of CuSP as a CS indicator, two studies reported differences between FM patients and HC in CuSP parameters [41,65]. Both studies explored the evocation of the spinal inhibitory response through the latency and duration of CuSP, and both revealed deficiencies in pain inhibition in FM. Given the above findings and scant evidence for the efficacy of CuSP as a CS pain indicator, further research is needed. Future studies should explore the suitability of including CuSP in QST protocols to evaluate CS.
On other hand, SREP has proven to be a dynamic evoked pain technique that is highly sensitive to the augmented responsiveness to pain that characterises FM. Although the underlying mechanisms of SREP remain unknown, its positive association with clinical pain [53], and higher reliability [55] and diagnostic accuracy compared to TSP for discriminating FM patients from HC [53,54,56] and those with other non-CS pain conditions like rheumatoid arthritis [55], support its use in clinical practice. Nevertheless, further research on the physiological basis of SREP is needed.
Finally, the reviewed studies combining algometry with neuroimaging suggest that imaging modalities such as fMRI allow for more objective quantification of CS. In line with this, seven of the eight reviewed studies using evoked pain indicators showed that the application of painful stimulation in FM patients was clearly associated with greater activation of central areas related to pain facilitation, and/or with lower activation of areas related to pain inhibition in comparison to HC [44,47,58,62,66,68,69]. In turn, the studies using dynamic pain indicators sensitive to CS together with fMRI reported a physiological correlation between the increasing pain responses to TSP and AS, manifesting as less activation of brain areas and spinal pathways related to pain inhibition, and greater activation of those related to pain facilitation [44,66,69]. No significant differences in the pain-related brain areas implicated in the TSP response were observed by Craggs et al. [52]. All these findings support the combined use of neuroimaging and evoked painful stimulation protocols, especially those based on dynamic evoked pain, to more comprehensively assess CS processes in FM patients. In addition, the high temporal resolution of neuroimaging techniques such as fNIR spectroscopy [58,70] and fTCD ultrasonography [62] allows almost real-time assessment of the CS processes in response to painful stimulation.
Each one of these dynamic evoked pain measures has different benefits and drawbacks, as indicated by the present systematic review. Despite their limitations, to some extent, they all have the ability to discriminate the CS processes involved in the pain pathophysiology of FM patients. Indeed, this review supports the recommendations of using evoked pain measures together with clinical pain reports [92] and tender point count [22] for CS assessments of FM patients. Also, combined use of algometry and simultaneous neuroimaging techniques is recommended to obtain deeper insight into the specific mechanisms underlying CS to pain. However, it must be emphasised that neuroimaging cannot replace pain reports, although it can improve understanding of the neural mechanisms underlying pain experience. Thus, its use is mainly limited to the research setting, also taking into account the high costs associated with applying the technique for screening purposes in clinical settings [93]. Another key recommendation for accurate interpretation of pain responses, based on a large proportion of the reviewed studies, is combining algometry with an assessment of psycho-affective and psychosocial factors known to influence the pain experience, especially: pain catastrophizing, anxiety, and depression.
This systematic review not only revealed the dynamic evoked pain indicators most useful to assess CS in FM but also identified some less well-known ones, such as CuSP, RISP and the recently developed SREP sensitisation. Further development and validation of the specific procedures for CS assessment are necessary to achieve consensus and optimise current QST methods. It is imperative to develop a unified algometry protocol by integrating current QST protocols, some new algometry measures, measurement of the psychological factors affecting pain responsiveness, and perhaps also some relatively accessible neuroimaging techniques to homogenise the assessment of CS processes in FM and other chronic pain conditions. Some limitations of the present systematic review should be considered. First, we could not control for the diverse pain modalities assessed by the reviewed studies due to insufficient data. Moreover, it should be noted that both skin and muscle (subcutaneous/deep) tissues were differentially stimulated in different body locations using heat, mechanical, and electrical stimulation, with the pain responses obtained suggesting CS involvement in most cases. Thus, these findings suggest that for a more global assessment of CS in FM, the current QST protocols should consider diverse types of stimulation, as well as different sites and tissues for stimulation. Second, the lack of methodological consistency among the studies may account for the differences in findings between the NFR threshold and CPM. Third, other limitations are the own indicated by the authors of reviewed studies (see Risk of Bias section), and the low quality of the majority of these studies in terms of bias assessment (see Table 2). Fourth, despite the well-known sex differences in pain perception in FM [94], all of the reviewed studies included only (or predominantly) female participants. Thus, studies including a greater proportion of males seem to be necessary. Lastly, 5 of the 30 reviewed studies were published by the authors of this review [53][54][55][56]62]. However, it is important to note that none of the authors of this review examined the bias quality in their own studies.
In conclusion, this review demonstrated the utility of algometry for examining the involvement of CS processes in FM by itself. Nevertheless, the future of algometry may be dependent on (1) consideration of the psychological factors conditioning the pain experience of FM patients, (2) new dynamic evoked pain indicators, (3) global assessments involving diverse body locations and tissues for stimulation, and (4) the simultaneous use of certain neuroimaging techniques (mainly for research purposes). All this could allow a more optimised analysis of CS-related pain responses using evoked pain protocols from research and clinical practice. However, greater consensus regarding CS measurement methods, and clarification of the underlying mechanisms (e.g. of RISP and SREP sensitisation) are still needed. In conclusion, our findings underline the benefits of algometry for assessing the contribution of CS to FM.

Disclosure statement
All authors declare that there is no conflict of interest.