Pressure Pain Threshold of the Upper Trapezius Trigger Point: A Systematic Review with Meta-Analysis of Baseline Values and Their Modification after Physical Therapy

Background: Myofascial trigger points (TrP) are diagnosed upon the presence of clinical signs among which hypersensitivity is considered one of the most important. The detection of the pressure pain threshold (PPT) is used to quantify the degree of hypersensitivity. However, there is a lack of normative data about how hypersensitive a TrP is. Therefore, the objective was to quantify the PPT for myofascial TrP in the upper trapezius muscle and its modification after manual or instrumental physical therapy interventions. Methods: A systematic review and meta-analysis were conducted among three databases (MEDLINE, Cochrane Library, and PEDro). Two independent reviewers conducted the electronic search and assessed the methodological quality of the included studies. Results: Eleven studies with a high-risk bias indicated that the PPT at TrP sites was 105.11 kPa lower (95% CI: −148.93; −61.28) at active TrP sites (Chi-squared = 1.07, df = 1 (p = 0.30), I2 = 7%) compared to the PPT of the upper trapezius muscles of healthy subjects. In addition, the PPT of TrP was also lower than the reference values coming from the pain-free population. Moreover, the PPT increased after both manual and instrumental treatment by 28.36 kPa (95% CI: 10.75; 45.96) and 75.49 kPa (95% CI: 18.02; 132.95), respectively. Conclusions: The results of the present study show that TrP has a decreased PPT when compared to healthy muscles and that physical therapy may increase the PPT. However, the clinical relevance of this decreased PPT needs to be further elucidated. Further, the high risk of bias in all the retrieved studies undermines the validity of the results.


Introduction
A trigger point (TrP) is defined as a hypersensitive spot within a contracted muscle fiber, that is painful to compress, can induce referred pain, and can generate autonomic phenomena [1]. Other symptoms usually reported are muscle stiffness, spasms, and limitations in movement of adjacent joints [2]. Several musculoskeletal pain syndromes are thought to be associated with TrP and are considered under the umbrella term "myofascial pain syndromes" [1], whose prevalence is generally reputed to be high [3,4]. An active TrP 2 of 25 is defined when its palpation can reproduce the familiar pain or referred pain pattern of the patient, either present or past. In contrast, a latent TrP is defined when the somatosensory sensations evoked during palpation are not related to the patient's symptoms [5]. The nociceptive afference arising from both latent and active TrPs is thought to increase the central excitability of the nervous system, causing peripheral [6] or central sensitization [7,8], in which the alteration of the dorsal root ganglion and an expansion or new formation of the receptive fields are considered responsible for the referred sensation evoked during palpation [9].
Manual muscle palpation constitutes an important procedure for the clinical assessment of the TrP, whose diagnosis lacks a proper gold standard. According to experts' opinion, the three main clinical findings to diagnose a TrP are the detection of a taut band, the detection of a hypersensitive spot inside, and the elicitation of referred pain [5]. The presence of a local twitch response was an additional criterion reported in a systematic review that also found disagreement on which are the most important criteria to be used [10]. For this reason, studies on the reliability of this palpatory examination showing poor to moderate intra and inter-reliability undermine its use in clinical practice [11,12]. The limits in quantifying the extent of hypersensitivity intrinsic to palpatory examination are overcome with the use of pressure algometry, a device that, by applying increasing force over a limited constant surface, allows the quantification of the minimum pressure, reported in kg/cm 2 or kilopascal (kPa), and is able to induce pain or discomfort, indicated as the Pain Pressure Threshold (PPT). A reduction in PPTs is merely interpreted as the increased sensitization of the painful body part, or of body parts far from the painful area that reflect, respectively, the degree of peripheral and/or central sensitization of the pain pathways. Indeed, it is important to remember that a lowered PPT may be a proxy of central sensitization if it is also found in healthy, pain-free areas [7,13]. Otherwise, if PPTs are lowered only in the symptomatic area, they are considered as the expression of peripheral sensitization [6].
The detection of a PPT specific to TrPshas been shown as a reliable procedure for the diagnosis of the TrPs themselves [14,15], and normative values for healthy subjects have been provided [16]. However, for muscles with TrP, there are a lack of normative PPT values that may inform decisions (more than manual muscle palpation) on the presence of a hypersensitive spot when diagnosing a TrP. For example, the measurement of the PPT is recommended to establish the extent of increased pain sensitivity in patients with headaches [15,17]. The PPT is considered clinically meaningful when its value is around 20% less than the PPT of healthy subjects for the same muscle [16]. Although no differences between dominant and non-dominant arms are reported [15], lowered thresholds are usually found in women compared to men, in older adults compared to younger adults and in lower limb/trunk muscles compared to upper quadrant muscles [16]. In clinical practice the PPT of the affected muscle is usually compared with the contralateral healthy side or with the lower limb muscles for patients with a pain condition of the upper quadrant. However, it is worth noting that when central sensitization is suspected, such as when multiple TrPs are found, lowered PPTs are also retrieved in the contralateral healthy side [7,18,19]. Neziri et al. [20] and Waller et al. [21] have suggested that values at the 5th and 95th percentile of the PPT distribution in a pain-free population indicate hyper and hyposensitivity thresholds, respectively. Referring to these absolute reference values would be helpful when dealing with patients with central sensitization.
An increase in the PPT has been proposed as a suitable parameter for the efficacy of treatment targeting TrPs, indicating less mechanical sensitivity over the TrP region [15,17]. Indeed, several studies on treatment efficacy have been conducted that measure the PPT at the TrP site before and after an intervention. Among these are dry needling [22,23], botulinum toxin [24], ischemic compression therapy [25,26], Kinesio taping [27], as well as lidocaine patches [28], exercises, and massage [29]. Although any of these treatments may be claimed to be effective in managing TrPs, a normative value of a clinically meaningful amount of pre-post difference in the PPT following whatever intervention has not been established yet. This kind of value may inform the clinical effectiveness of the intervention beyond its statistical significance. Among the numerous studies on the PPT, many have been conducted on the upper trapezius muscle, which therefore represents a suitable model given its high involvement in many musculoskeletal pain syndromes of the upper quadrant and its anatomical position that allows accessibility to both manual and instrumental assessment and treatment. The purpose of this systematic review and meta-analysis is to evaluate whether upper trapezius muscles with TrPs have a different PPT when compared to healthy muscles and whether they resemble the suggested hypersensitivity threshold [20,21]. Furthermore, it will be analyzed whether physical therapy treatment of a TrP is able to influence the PPT of the treated muscle.

Protocol and Registration
The protocol of this systematic revision was prospectively registered at PROSPERO (https:// www.crd.york.ac.uk/prospero/ accessed on 3 November 2022) with the number CRD42020152611.

Eligibility Criteria
Type of studies: the study types considered were non-randomized and randomized controlled trials (RCT) based on manual or instrumental physical therapy treatments, except for case reports and case series.
Type of participants: studies were included when participants were older than 18 and presented an active or latent TrP in the upper trapezius muscle or no TrP in the same muscle in the case of healthy subjects recruited as controls. Studies were excluded when participants had any of the following conditions: the presence of comorbidities due to medical disease (neurological, rheumatic, oncology, cardiac, or metabolic dysfunctions) or previous surgical interventions in the examined area.
Type of interventions: the measurement of the PPT was made with both electrical and manual pressure algometry using kg/cm 2 or kPa as the unit of measure or providing values allowing transformation into kPa.
Type of comparators: acceptable comparators were the same muscle in a healthy group recruited in the same study.
Type of outcomes: the primary outcome was the PPT difference between an active or latent TrP and healthy control muscles. Another primary outcome was the post-treatment PPT values between the intervention and placebo control group. The PPT was converted into kPa according to the unit of measure used in the study and the dimension of the probe used.

Search Strategy
The search was performed using the databases MEDLINE (through the search engine PubMed), Cochrane, and PEDro, looking for online publications until 31 August 2020. The search terms used were myofascial pain, trigger point or trigger points, pressure pain threshold, algometry or algometer. Relevant articles were screened for additional RCTs to consider. The full strategy is reported in Appendix A.

Study Selection
Two authors (A.B., S.D.A.) independently searched the databases to identify appropriate records to screen, applying the eligibility criteria. When the screening process ended, the full text of the identified records was retrieved and assessed for eligibility in the qualitative/quantitative synthesis.
Any disagreement was resolved by consensus; if no consensus was reached, a third reviewer (T.G.) made the final decision. The inter-rater agreement of the screening and of the eligibility processes before consensus were expressed using a percentage agreement and Cohen's kappa [30].

Data Collection
Two independent authors (A.B., S.D.A.) manually extracted data from the included studies, filling a pre-formatted table that included data about population samples, type of myofascial disorder, nature of TrPs (active or latent), analyzed muscles, type of algometer used and, for RCTs, treatment conducted. Since PPT values have already been shown to be different between the sexes, sex-disaggregated data were not calculated in the present review. Any disagreement was resolved by consensus; if no consensus was reached, a third reviewer (T.G.) made the final decision.

Risk of Bias (RoB) Assessment
The Cochrane Risk of Bias tool version 2.0 (RoB2.0) [30] was used to assess the internal validity of the included RCTs. In addition, non-randomized clinical trials were assessed using the Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool [31]. For RoB2.0, the domains randomization process, deviations from the intended intervention, missing outcome data, measurement of the outcome, and results reporting were evaluated to obtain for each study an overall risk of bias judgement that ranged from low-when all domains have a low risk of bias-to high-when the study has at least one domain with high-risk bias or multiple domains showing biasing concerns. For ROBINS-I, the domains evaluated were confounding bias, selection bias, classification of intervention bias, missing data, measurement of outcome, and selection of reported result. Two independent authors (A.B., L.P.) assessed the included studies, and a third reviewer (TG) made the final decision when consensus could not resolve the disagreement. The inter-rater agreement of the assessment of the risk of bias before consensus was calculated using percentage agreement and Cohen's kappa [30].

Analysis and Synthesis of Results
The PPT was analyzed using the pooled mean difference (MD). The variance was expressed with 95% confidence intervals (95% CI). As our interest was understanding the treatment effect against a placebo control, the PPT values derived from different intervention arms of the same RCT study [32][33][34] were merged using well-established methods [35]. A global PPT value for active and latent TrPs was obtained by calculating the weighted mean and SD using the values reported in individual studies. The obtained values were compared with the weighted mean of values from two studies reporting a PPT on the upper trapezius muscle in the general population [20,21]. The comparison was made with a one-sample t-test. Alpha was 0.05.
The outcome measures from the individual trials were combined through metaanalysis where possible using the random-effects models described by DerSimonian and Laird [36] as some heterogeneity of population and treatments would be expected among interventions.
Heterogeneity was analyzed by means of the I 2 statistic and the Chi 2 test. A p-value lower than 0.1 indicated the presence of a statistically significant heterogeneity for the Chi 2 test [36]. The degree of heterogeneity was expressed with the percentage of I 2 . Percentage values of 25, 50, and 75% indicated a low, moderate, and high degree of heterogeneity, respectively [36]. If a study did not provide usable summary measures for an outcome, it was included in the review but excluded from the meta-analysis, e.g., Gemmell et al. [37] and Kavadar et al. [38]. For the included studies, the numbers lost to follow-up in each group and the reasons for attrition were recorded. For missing data, the similarity of the group was evaluated, then the corresponding authors of the included studies were contacted (e.g., by emailing or writing to the corresponding author), and if no information was provided we conducted analyses using only the available data (e.g., we did not impute missing data). Cohen's kappa and percentage agreement (PA) were judged as acceptable when higher than 0.6 and 80%, respectively [31]. Analysis was performed using Revman 5.0 [39] and R software [40] with Hmisc package v 4.4-0 [41].
The reporting of this study has been performed in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement [42].

Level of Evidence
The overall quality of evidence was evaluated using the grading of recommendations assessment, development, and evaluation approach (GRADE) for the main outcome based on the methodological quality of the included trials [43]. In addition, the quality of the evidence was evaluated using the software GRADEPro GDT [44], which assessed the risk of bias, inconsistency, indirectness, imprecision, and other considerations about publication bias, degree of the effect, presence of confounding factors, and dose-response gradient.

Study Selection
The search strategy identified 868 records, of which 276 were duplicates. Following the screening process, 542 records were excluded. Of the 50 eligible articles, 39 were excluded after careful reading of the full-text, resulting in 11 articles included in the review [32][33][34]37,38,[45][46][47][48][49][50]. The list of excluded articles and the reason for their exclusions are reported in Appendix B. Two additional articles [37,38] were discarded from the analysis of global PPT values for TrPs for reasons explained below. However, they were included in the comparisons regarding treatment effectiveness as the device used was the same for both the experimental and control group. The flowchart is reported in Figure 1. The inter-rater reliability was acceptable for the screening (k = 0.62, PA = 95.1%) and eligibility processes (k = 0.72, PA = 90%).

Study Characteristics
The majority of studies had an RCT design investigating the effect of various instrumental [34,38,[48][49][50] or manual [32][33][34]46] physical therapy treatments on various clinical outcomes comprising the PPT ( Table 1). The study by Abbaszadeh-Amirdehi et al. [45] was a prospective clinical trial and one study did not clearly report on the research design [47].
A total of 574 subjects (at least 309 female, as Abu Taleb et al. [32] and Öztürk et al. [50] did not report gender distribution), with a mean age of (29.64 ± 12.59 SD) years constituted the pooled population of subjects with TrPs. Two studies [34,46] also recruited a sample of 68 healthy controls (years, 36.1 ± 14.9 SD) to compare the PPT of the upper trapezius muscle without a TrP with the PPT of an upper trapezius hosting an active TrP of 104 subjects (years, 39.7 ± 12.3 SD).
Among the intervention studies, six studies investigating active TrPs recruited subjects with acute [32,37,38,49] or chronic [46,50] neck pain, while one study recruited subjects with myofascial pain syndrome [34] and one with shoulder pain [48]. One study recruited a pain-free population to investigate the PPT on latent TrPs [46], using parts of the clinical criteria, such as palpable taut band, hypersensitive tender spot, elicitation of local twitch response, and reproduction of referred pain pattern typical of the investigated muscle, used for the identification of active TrPs [1].
The studies comparing physical therapy modalities with the sham procedures used extracorporeal shockwave [48], ultrasound [34,38,49], low-level laser therapy [49], and kinesiotape [50] as interventions, while the same interventions with the device turned off or depowered were used in the placebo control groups of all these studies. The pooled population of this subgroup was constituted of 192 participants (years, 31.18 ± 11.92 SD). J. Clin. Med. 2022, 11, x FOR PEER REVIEW 6 of 23

Study Characteristics
The majority of studies had an RCT design investigating the effect of various instrumental [34,38,[48][49][50] or manual [32][33][34]46] physical therapy treatments on various clinical outcomes comprising the PPT ( Table 1). The study by Abbaszadeh-Amirdehi et al. [45] was a prospective clinical trial and one study did not clearly report on the research design [47]. Several types of algometers were employed across studies, of which only four precisely reported the instrument used [34,[48][49][50]. Two studies [37,38] that neither described the algometer used nor reported the PPT measurement procedure (Table 2)were not included in calculating global PPT values for the PPT. Most studies used PPT values coming from the average of three repetitions at different time intervals lasting no longer than 60 s. The application rate of the algometer pressure was heterogeneous across studies. The PPT was measured after one treatment session in all the studies using a manual intervention, except for one study that measured the PPT after 24 h [46] and after three treatment sessions [33]. In the studies using instrumental physical therapy, the PPT was registered at the end of the treatments provided, which varied in number and frequency of sessions and lasted for 1-3 weeks ( Table 2).

Risk of Bias within Studies
The risk-of-bias summary is reported in Figure 2A,B. All the studies were deemed as having a high risk of bias, as they presented at least one domain with a high risk of bias.

Study
Examiner Characteristic TrP Type Specified Location Specified Crit to Simon Abbaszadeh-Amirdehi et al. [45] NR Active X X Abu Taleb et al. [32] NR Active X Ganesh et al. [46] More than 10 years of clinical experience in diagnosingTrP Latent X Gemmell et al. [37] NR Active X X Hong et al. [34] NR Active X X Ji et al. [48] Medical doctor Active X Kavadar et al. [38] Physician Active X

Risk of Bias within Studies
The risk-of-bias summary is reported in Figure 2A,B. All the studies we having a high risk of bias, as they presented at least one domain with a high Among all the studies, one [34] had four domains with a high risk of b randomized clinical trial [45] evaluated with ROBINS-I resulted in two do Among all the studies, one [34] had four domains with a high risk of bias. One nonrandomized clinical trial [45] evaluated with ROBINS-I resulted in two domains with a moderate risk of bias. Three studies [47,48,50] had two domains with a high risk of bias, and six studies [32,33,37,38,46,49] had one domain with a high risk of bias. Among all domains, the one with a high risk of bias in the majority of the studies was the domain regarding results reporting [32,34,37,38,49,50], followed by the domain regarding the randomization process [34,45,47,48], measurement of the outcome [33,34,45,50], and deviation from the intended intervention [34,45]. The only domain without a high risk of bias in any study was the one regarding missing outcome data, that, on the other hand, showed concerns about its biasing in seven studies [32,34,37,38,47,48,50]. The inter-rater reliability of the risk of bias assessment was acceptable (k = 0.89, PA = 92.7%).
The weighted mean of the baseline PPT values coming from patients with active or latent TrPs and from the general healthy population gave overall values for these conditions ( Table 4  Six studies (356 subjects) were meta-analyzed for the comparison between manual intervention and minimal active treatment with separate analyses for subgroups having active or latent TrPs. In general, the manual treatment effectively increased the PPT with an MD of 28.36 kPa (95% CI: 10.75; 45.96, χ 2 = 19.73, df = 6 (p = 0.003), I 2 = 70%). However, this positive result was biased by the subgroup with latent TrPs, while the separated analysis for the subgroup with active TrPs showed a large confidence interval with no effectiveness (PPT = 104.43 kPa; 95% CI: −23.97; 232.83) (Figure 4). The comparison between physical therapy modalities and minimal active treatment included six studies investigating active TrPs. The treatment effect was positive, with a PPT increase of 75.49 kPa (95% CI: 18.02; 132.95, χ 2 = 43.16, df = 5 (p < 0.001), I 2 = 88%) ( Figure 5).
The summary of findings for each comparison and the quality of assessments are reported in Table 5.

Risk of Bias across Studies
A publication bias was observed for the comparison between manual treatment and minimal active intervention regarding the active TrP subgroup analysis (Figure 6b). Indeed, the distribution of studies on active TrPs was uneven across the pooled values, with two studies biasing the positive results of the meta-analysis. All the other comparisons did not show evidence of a publication bias as studies were evenly distributed among the pooled values ( Figure 6).

Risk of Bias across Studies
A publication bias was observed for the comparison between manual treatment and minimal active intervention regarding the active TrP subgroup analysis (Figure 6b). Indeed, the distribution of studies on active TrPs was uneven across the pooled values, with two studies biasing the positive results of the meta-analysis. All the other comparisons did not show evidence of a publication bias as studies were evenly distributed among the pooled values ( Figure 6).

Discussion
The main finding of this systematic review with meta-analysis was that the PPT was lower at active TrP sites of the upper trapezius when compared to the upper trapezius without TrPs of healthy subjects. The quality of the evidence was moderate, according to the GRADE tool. In all the retrieved studies [32][33][34]37,38,[45][46][47][48][49][50], the TrP was first identified through manual palpation, and then the PPT over the TrP site was measured. Considering that the measurement of the PPT for active and latent TrPs comes from a young

Discussion
The main finding of this systematic review with meta-analysis was that the PPT was lower at active TrP sites of the upper trapezius when compared to the upper trapezius without TrPs of healthy subjects. The quality of the evidence was moderate, according to the GRADE tool. In all the retrieved studies [32][33][34]37,38,[45][46][47][48][49][50], the TrP was first identified through manual palpation, and then the PPT over the TrP site was measured. Considering that the measurement of the PPT for active and latent TrPs comes from a young population with no difference in gender distribution, the results of the present review point out that the PPT values for TrPs were lower than the reference PPT values of the upper trapezius measured in two studies on pain-free populations with similar demographic characteristics [20,21]. Both these studies suggested the fifth percentile of the distribution to label hypersensitivity that can be roughly indicated as lower than 110 kPa [20] and 134 kPa [21]. For active TrPs, the only studies that yielded these criteria were those from Abbaszadeh-Amirdehi et al. [45] and Abu Taleb et al. [32] for active TrPs and Ruiz-Sáez et al. [47] for latent TrPs. As hypersensitivity of the trigger spot is considered a cardinal clinical criterion for the diagnosis of TrP [5], the use of different thresholds may affect the clinical assessment. When a patient's complaints are driven by nociceptive pain and with signs that can be interpreted according to a peripheral sensitization of the pain system, then clinicians may refer to the normative PPT values reported in this review when comparing the affected upper trapezius with the contralateral one. Otherwise, when a patient's complaints are compatible with a central sensitization syndrome, the comparison with the contralateral side may be inappropriate, with the risk of missing the decrease in PPT. In this particular case, the clinician should refer to the thresholds coming from studies on general pain-free populations [20,21]. Future studies may change the normative values reported in this review as only studies with a high risk of bias were included.
Another main finding was that either manual or physical therapy modalities and interventions are likely to increase the PPT values in subjects with active TrPs; however, it should be considered that the duration of the increase in PPT after an intervention has not been reported. The quality of the evidence was very low, according to the GRADE tool. The high heterogeneity observed in the comparisons between manual or instrumental treatment and minimal active intervention was expected as our main interest was in estimating the extent to which the PPT may vary following a treatment, rather than the effectiveness of the treatment itself, therefore studies with different clinical presentations, instruments and therapies used and their dosage, and the time to follow-up were inserted in the metaanalysis. The only criteria shared across all the retrieved studies was the use of established criteria [1] to diagnose a latent or an active TrP and a clear reporting of the algometer used. Both the analyzed interventions (manual or instrumental physical therapy treatment) showed an increase in PPT after physical or manual treatment; however, this result is likely to change as all the included studies were judged as having a high risk of bias. Furthermore, as the effectiveness of an intervention should consider thoroughly the patient's health status instead of a mere modification of the PPT, we suggest that future studies should link the PPT change with change scores obtained through patient-reported outcome measures (e.g., disability, satisfaction) using a suitable analysis such as ROC curves.
Several hypotheses have been advanced to explain the somatosensory alterations caused by TrPs [53][54][55]. Despite controversies in the identification of the nociceptive locus, all point towards an increased afference to the motoneuron due to increased activity of the dorsal horn that may also explain the typical hypersensitivity found at TrP sites.
The fact that treatments were not as effective at restoring a normal PPT found in healthy subjects may partially explain why people do not recover fully after the first episode of neck pain [56]. It further supported the idea that once a TrP is treated (with manual or instrumental physical therapy treatment), a multimodal approach that also integrates exercises [57] is needed to favor the recovery of muscle that has developed a TrP. However, most of the studies included in this systematic review with meta-analysis measured the outcome of a single treatment approach in the short term (immediately after a one-session intervention). This fact should encourage clinicians and researchers to extend the follow-up to understand the influence of physical therapy interventions on the PPT of active TrPs in the long-term. In conclusion, measuring PPT values may constitute a valid procedure in supporting the diagnosis of myofascial TrPs in the upper trapezius muscle and monitoring a patient's clinical improvement [15]. However, it should not be considered the first outcome measure as it relates only to a specific aspect of a multidimensional construct, such as a painful experience.
This study has some strengths and limitations that need consideration. The main strength of the study is that it is the first that has meta-analyzed PPT values in patients with active TrPs. This value, which had a moderate quality of evidence, may be used as a reference in clinical practice as well as in research in calculating the sample size for studies on the myofascial trigger point. Among the limitations, first, despite the search strategy encompassing three databases, some articles potentially affecting the result of the present review may have been retrieved with more databases included. Furthermore, only articles published in English were included in the review and this, as well as the lack of searching in the grey literature, may limit the generalizability of the result. Furthermore, all the analyzed studies had a high risk of bias, and a publication bias was found in the comparisonof manual and minimal active interventions. Moreover, the comparison of the PPT value between patients with an active TrP and healthy controls emerged from studies with different designs (one non-randomized clinical trial [46] and one RCT [35]). However, we only used the baseline values of the PPT; therefore, the different study designs should not affect the datum. Finally, the studies included in the meta-analysis were very heterogeneous in terms of the interventions and modalities; this may represent another limitation, as well as the heterogeneity of the duration of the treatment (some just one session, some multiple sessions). Therefore, future reviews on a similar topic based on more rigorous studies and also including studies with negative results may alter the reported findings.

Conclusions
Our findings showed that TrP has a decreased PPT when compared to an upper trapezius without TrP and healthy subjects and that either manual or physical therapy interventions may increase the PPT. However, the high risk of bias in all the included studies undermines the validity of the results.

Conflicts of Interest:
The authors declare no conflict of interest, and the funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results. Table A1. Search strategy.

Electronic Database Search String
Cochrane Library (((("pressure pain threshold")) OR (algometer)) OR (algometry))) AND (((((("trigger points")) OR ("trigger point")) OR ("myofascial pain")) Appendix B  The study is about the difference in diagnostic criteria (LTR present or not) inthe effectiveness of DN  The study did not provide data at post-intervention but only graphs. The study did not measure PPT for the healthy controls. The study is on healthy population but active TrP are detected