Journal Logo

Systematic Review and Meta-Analysis

The predictive value of quantitative sensory testing: a systematic review on chronic postoperative pain and the analgesic effect of pharmacological therapies in patients with chronic pain

Petersen, Kristian Kjæra,b,*; Vaegter, Henrik B.c,d; Stubhaug, Audune,f; Wolff, Andrég; Scammell, Brigitte E.h,i; Arendt-Nielsen, Larsa,b; Larsen, Dennis B.a,b

Author Information
doi: 10.1097/j.pain.0000000000002019
  • Free
  • Global Year 2021

Abstract

1. Introduction

Chronic pain is a major problem in the adult population,18 and treatment is difficult because of the limited amount of available efficient drugs and the undesired side effects. It is evident that chronic postoperative pain is present in 10% to 50% of patients after different surgical treatments,49,87 and the effect of available pharmacological treatments remains low.29,86 Identifying patients responding to standard pain treatments is warranted as a start to implement individualized pain treatment.

Quantitative sensory testing (QST) can be combined using different sensory stimuli, and multiple QST protocols have been developed to probe the activity in specific nerve fibre populations and as proxies for spinal and pain modulatory functions.11,83 Most QST protocols allow the assessment of thermal, electrical, tactile, or pressure pain modalities.12,23,35,83,102 Reduced pain thresholds or increased sensory intensity ratings assessed at a local painful site mainly reflect modality-specific peripheral hyperalgesia, whereas assessments distant to a painful site may reflect widespread hyperalgesia as a surrogate measure of central hypersensitivity.10,34 Studies have found reduced pressure pain thresholds at distal sites to the knee in patients with knee osteoarthritis,8,37 increased sensitivity to thermal and mechanical stimuli in a subgroup of patients with neuropathies,82,99 and reduced electrical pain thresholds at the dorsal pancreatic referred dermatomes and at distant dermatomes in patients with chronic pancreatitis55 compared with pain-free subjects. These studies indicate that patients with chronic pain of different aetiologies show generalised widespread hyperalgesia.7

Dynamic QST protocols have been developed to explore the central wind-up process using the proxy temporal summation of pain (TSP).34 Facilitated TSP has been reported across multiple chronic pain disorders such as osteoarthritis,10 fibromyalgia,36 irritable bowel syndrome,7 and in subgroups of patients with neuropathic pain.62 Aspects of descending pain inhibitory control can most likely be mechanistically evaluated using the human proxies of conditioned pain modulation (CPM),107 exercise-induced hypoalgesia,70 or offset analgesia44 protocols. Impaired CPM has been reported in several chronic painful conditions when compared with pain-free subjects.7,107 Exercise-induced hypoalgesia seems functional in asymptomatic subjects93–96 and impaired in different pain populations,28,57,67,92 although the current literature is inconclusive.19,28,92 Similarly, offset analgesia seems functional in asymptomatic subjects44 and impaired in patients with chronic pain.90

Studies have suggested a possible association between preoperative QST parameters and chronic postoperative pain47,56,74 and that pretreatment QST may predict the analgesic effect of pharmacological interventions.75,110 Previous systematic reviews2,38,46,84,101 have investigated the predictive value of QST on postoperative pain, the most recent in 2017,84 and the predictive value of QST on the analgesic effect, the most recent in 2013.38 The most recent review from 2017 on surgical studies84 indicated that preoperative QST mainly predicts chronic postoperative pain but not acute postoperative pain. As multiple studies have been published since 2017, this review systematically summarizes the current literature on the possible predictive role of QST on (1) chronic postoperative pain and (2) the analgesic effect of pharmacological interventions in patients with chronic pain.

2. Methods

In accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, this systematic review investigated the predictive role of QST on chronic postoperative pain and the analgesic effect of pharmacological interventions in patients with chronic pain. The systematic review has been registered on the Open Science Framework web site (OSF.IO, registration citation: Ref. 58, link to protocol: DOI: 10.17605/OSF.IO/HSVYK).

2.1. Outcomes

The primary outcomes for chronic postoperative pain were postoperative pain intensity, postoperative pain relief, presence of moderate-to-severe postoperative pain, or validated questionnaires on pain and disability. The primary outcomes of pharmacological studies were pretreatment and posttreatment changes in pain scores, classification of responders to 30% or 50% pain relief, end of treatment pain intensity, or validated questionnaires on pain and disability.

2.2. Search strategy and selection of studies

A literature search was conducted in April 2020 in the databases PubMed and EMBASE. The search was limited to the literature published in the past 20 years (April 2000-April 2020). Only peer-reviewed studies published in English and with available full-text articles were considered eligible for the systematic review. Two searches were conducted to identify (1) the predictive value of QST for chronic postoperative pain outcomes and (2) the predictive value of QST on analgesic effect outcomes of pharmacological interventions. The MeSH terms and/or text word combinations are shown in Table 1.

Table 1 - Search strategy.
Focus Example key words (PubMed search)
1 Analgesic effect (“Analgesia”[tw] OR “Drugs”[tw] OR “Drug therapy”[tw])
2 Postoperative pain (“Postoperative pain”[MeSH] or “Postsurgical pain”[tw])
3 Quantitative sensory testing {(“Quantitative sensory testing”[tw] OR “QST”[tw]) OR (“Conditioned pain modulation”[tw] OR “CPM”[tw]) OR (“Temporal summation”[tw] OR “TSP”[tw])}
4 Surgery (“Surgery”[tw])
5 Limits “2000/01/01”[PDat], “English”[lang]
The MeSH and text word (tw) strings were permuted dependent on database.

All citations were exported to EndNote X4 (Thomson Reuters, Philadelphia, PA), and duplicates were removed. Owing to the large number of potential studies, the initial screening was conducted on title and abstract to remove citations that did not meet the scope of the systematic review. The screening process was independently performed by 2 reviewers (K.K.P. and D.B.L.) after the initial systematic database search. Disagreements in relevancy were solved by consensus. In case consensus was not reached, a third reviewer (H.B.V.) was consulted who made the final decision. After the screening, all full-text articles were obtained. Cross-referencing the included studies and the authors' own article collection was conducted for additional relevant literature. The inclusion and exclusion of the relevant literature are shown in Figure 1 (PRISMA flowchart).

Figure 1.
Figure 1.:
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram. QST, quantitative sensory testing.

2.3. Eligibility criteria

The studies included had to report the predictive model of a postoperative or postpharmacological outcome of at least 1 QST modality, including electrical, thermal, mechanical or pressure, pain, tolerance and suprathreshold stimuli, TSP, allodynia, exercise-induced hypoalgesia, offset analgesia, CPM, or the full German Research Network on Neuropathic Pain (DFNS) protocol. Studies were included if they reported associations between QST and pain-related outcome after surgery or pharmacological intervention using correlations (Spearman or Pearson correlations), regression models, or other predictive models such as support vector machine or reduction in numbers needed to treat. Studies on animals and healthy subjects were excluded. Surgical studies were included if they assessed follow-up pain at least 3 months after surgery. Acute postoperative pain and studies in the subacute phase were not considered. Pharmacological studies were included if they studied the long-term effect (weeks of treatment) of pharmacological interventions. Therefore, studies investigating the acute effect (minutes or hours after administration) were excluded. All studies were required to have a detailed description of the used QST paradigm.

2.4. Data extraction, synthesis, and assessment

The data extraction was conducted by K.K. Peterson. For studies on chronic postoperative pain, the data included type of surgery, patient cohort, QST parameters tested, follow-up time after surgery, the dependent outcomes of the prediction model, and preoperative predictors and their associated predictive value for the dependent variable. For the pharmacological studies, the data involved type of pharmacological intervention, patient cohort, QST parameters tested, treatment period, dependent variable in the prediction model, and predictive value of the QST parameters for the dependent variable. All outcomes were narratively synthesized to provide an overview of each QST paradigm assessed and presented within each domain of QST in its prediction of chronic postoperative pain or analgesic effect of pharmacological interventions. This was conducted because of the large heterogeneity in both the QST paradigms used and the clinical cohorts included in the studies, which meant that a meta-analysis would not be appropriate. The methodological and the overall quality of the included studies were assessed by 2 reviewers (K.K.P. and D.B.L.) using the Quality In Prognostic Studies (QUIPS) tool.43 Following the guidelines, each study was assessed on methodological quality in the major categories of “Study participation,” “Study attrition,” “Prognostic factor measurement,” “Outcome measurement,” “Study confounding,” and “Statistical analysis and reporting.” Disagreements in the risk of bias analysis were solved by consensus, and if no consensus was reached, a third reviewer (H.B.V.) was consulted.

3. Results

The literature searches identified 1811 surgical and 2689 pharmacological studies, of which 25 surgical and 11 pharmacological studies were included (Fig. 1). The outcome parameters of the surgical studies were postoperative pain intensity (12 studies1,4,39,48,56,61,66,74,76,98,103,109), postoperative pain relief (3 studies9,78,92), presence of moderate-to-severe postoperative pain (6 studies16,41,69,71,81,89), or validated questionnaires on pain and disability (3 studies40,105,106) (Table 2). The outcome parameters in the pharmacological studies were pretreatment and posttreatment changes in pain scores (4 studies25,75,77,110), migraine scores (1 study51), and classification of responders to 30% (3 studies5,63,72) or 50% (2 studies20,21) pain relief, and end of treatment pain intensity24 (Table 3).

Table 2 - Surgical studies assessing quantitative sensory testing (QST) as predictors of chronic postoperative pain.
Reference Year Type of surgery Patients QST parameters Follow-up time Postoperative dependent parameter Models and preoperative predictors
Joint-related surgeries
 Martinez et al.66 2007 Total knee arthroplasty Knee osteoarthritis: N = 20 ALL, MPT, HPT, CPT 4 mo Pain intensity Correlations (U):
No prediction
 Lundblad et al.61 2008 Total knee arthroplasty Knee osteoarthritis: N = 69 EDT, EPT 18 mo Pain intensity Regression (M):
Pre-op pain (OR: 6.48) and EPT (OR: 9.19)
 Gwilym et al.40 2011 Arthroscopic subacromial decompression Shoulder impingement syndrome: N = 17 MPT 3 mo Oxford shoulder score Correlation (U):
No prediction
 Wylde et al.105 2013 Total knee arthroplasty Knee osteoarthritis: N = 51 PPTs, HPT 13 mo WOMAC Correlations (U):
PPT: R = 0.37
 Valencia et al.98 2014 Arthroscopic shoulder surgery Rotator cuff tendinopathy: N = 73 HPT, hTSP, hCPM* 3 and 6 mo Pain intensity Regressions (M):
No prediction
 Noiseux et al.71 2014 Total knee arthroplasty Knee osteoarthritis: N = 193 MPT, HPT, PPT 6 mo Presence of moderate-to-severe postoperative pain Regression (M):
No prediction
 Petersen et al.74 2015 Total knee arthroplasty Knee osteoarthritis: N = 78 PPTs, mTSP, pCPM† 12 mo Pain intensity Regression (M):§
mTSP and Pre-op VAS: R2 = 0.13
 Wylde et al.106 2015 Total knee and hip arthroplasty Knee osteoarthritis: N = 239, hip osteoarthritis: N = 254 PPT 12 mo WOMAC Regression (M):
THA: PPT: β = 0.091 (WOMAC) − 0.114 (movement pain)
TKA: No prediction
 Petersen et al.78 2016 Total knee arthroplasty Knee osteoarthritis: N = 103 cPPT, cPTT, cTSP, cCPM, PPTs 12 mo Pain relief Regression (M):§
R2 = 0.379, using the cPPT and VAS
 Bossmann et al.17 2017 Total knee arthroplasty Knee osteoarthritis: N = 47 mTSP, pCPM‡ 6 mo WOMAC pain subscale Regression (M):
pCPM, heart rate variability, and gender: R2 = 0.09
 Vaegter et al.92 2017 Total knee arthroplasty Knee osteoarthritis: N = 14 cPPT, cPTT, PPT, pCPM†, EIH 6 mo Pain relief Correlations (U):
pCPM: R = 0.57
EIH: R = 0.53
 Arendt-Nielsen et al.9 2018 Total knee arthroplasty Knee osteoarthritis: N = 70 PPT 12 mo Pain relief Regression (M):
PPT: R2 = 0.09–0.110
 Petersen et al.76 2018 Total knee arthroplasty Knee osteoarthritis: N = 130 CDT, CPT, WDT, HPT, mTSP 12 mo Pain intensity Regression (M):§
Pre-op mTSP, WDT, HPT, and KL: R2 = 0.119
 Rice et al.81 2018 Total knee arthroplasty Knee osteoarthritis: N = 288 PPT, mTSP, pCPM† 6 and 12 mo WOMAC pain >30/100 Regression (M):
WOMAC pain, mTSP, Trait anxiety, and expected pain:
AUC: 0.70
Sensitivity: 0.72
Specificity: 0.64
Correctly classified: 65.67%
 Kurien et al.56 2018 Total knee arthroplasty Knee osteoarthritis: N = 50 PPTs, cPPT, cPTT, cTSP, cCPM, mTSP, Pain intensity Correlations (U):
mTSP: R = 0.343
PPT: R = −0.262
 Müller et al.69 2019 Segment spinal surgery Chronic low back pain: N = 141 PPT, PTT, HPT, CPT, CPM, EPT 12 mo Persistent pain or persistent disability Regression (M):
No prediction
Thoracic-related surgeries
 Yarnitsky et al.109 2008 Thoracic surgery Posterolateral and muscle-sparing lateral thoracotomy: N = 62 hCPM*, HPT, STHS 29 wk Pain intensity Regression (M):
hCPM* (OR: 0.55)
 Grosen et al.39 2014 Surgical correction of funnel chest Patients scheduled for surgical correction of funnel chest: N = 41 ALL, MDT, PPT, pCPM† 6 mo Pain intensity Regressions (M):
No prediction
 Bayman et al.16 2017 Thoracic surgery Patients scheduled for thoracic surgery: N = 99 CPT, STHS 6 mo Presence of postoperative pain Regression (M):
No prediction
Abdominal and gynecology-related surgery
 Aasvang et al.1 2010 Groin hernia repair (open and laparoscopic) Primary unilateral hernia: N = 442 WDT, HPT, STHS 6 mo Pain intensity Regression (M):
Activity (OR: 1.16-7.37), STHS (OR: 1.34)
Surgical type (OR: 0.45)
 Wilder-Smith et al.103 2010 Major abdominal surgery Lower and upper gastrointestinal or genitourinary tract issues: N = 20 EPTT, PTT, eCPM†, pCPM† 6 mo Pain intensity Regression (M):
Pre-op eCPM: R2 = 0.46
 Jarrell et al.48 2014 Gynecologic laparoscopy Patients with gynecological pain: N = 77 MDT, ALL 6 mo Pain intensity Regressions (M):
The presence of pre-op ALL (R2 = 0.590)
 Sng et al.89 2018 Elective abdominal or laparoscopic hysterectomy for benign conditions Patients with benign conditions scheduled for surgery: N = 159 mTSP 6 mo VAS >3/10 Regression (M):
mTSP: OR = 1.078
Pain during sexual intercourse: OR = 5.312
Morphine consumption (24-48 hours post-op): OR: 1.172
Breast cancer surgeries
 Andersen et al.4 2017 Breast cancer surgery Patients with breast cancer: N = 290 MDT, MPT, WDT, CDT, HPT, PPT, size of the hypoesthesia area using: ALL and cold (25°C) and warm (40°C) rolls. 12 mo Pain intensity Regression (M):
Size of the hypoesthesia area: OR: 1.003-1.006
 Habib et al.41 2019 Breast cancer surgery Patients scheduled for breast cancer surgery: N = 124 mTSP, hCPM* 6 and 12 mo VAS >3/10 and impact of pain on daily living Regression (M):
No prediction
QST modalities: ALL, dynamic mechanical allodynia; ANOVA, analysis of variance; CDT, cold detection threshold; CPT, cold pain threshold; cPPT, cuff-induced pressure pain threshold; cPTT, cuff-induced pressure tolerance threshold; CPM, conditioning pain modulation (c, cuff pressure test and condition stimuli; e, electrical test stimulus; heat, heat test stimulus; p, pressure test stimulus; *, hot water condition stimulus; †, cold-pressor tests as condition stimulus; ‡, pinching conditioning stimulus); EDT, electrical detection threshold; EIH, exercise-induced hypoalgesia; EPT, electrical pain threshold; EPTT, electrical pain tolerance threshold; HPT, heat pain threshold; KOOS, Knee Injury and Osteoarthritis Outcome Score (assesses pain, stiffness and function, and daily living of patients with knee osteoarthritis); (M), indicates multivariate analysis; MDT, mechanical detection threshold; MPT, mechanical pain threshold; N, number of patients participating in the study; OR, odds ratio; PPT, pressure pain threshold; PTT, pressure tolerance threshold; R, coefficient of determination; STCS, suprathreshold cold stimulus; STHS, suprathreshold heat stimulus; TSP, temporal summation of pain (c, using cuff stimuli; e, using electrical stimuli; h, using heat stimuli; m, using mechanical pinprick stimuli); (U), indicates univariate analysis; VAS, visual analog scale; WDT, warm detection threshold; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index (assesses pain, as well as stiffness and function of the knee and hip).
§Calculated for this review alone and not presented in the original article.

Table 3 - Pharmacological studies using quantitative sensory testing (QST) to predict analgesic effects.
Reference Year Type of intervention(s) Patients QST parameters Treatment period Dependent parameter Model and pretreatment predictor
Edwards et al.25 2006 Opioids: N = 16
TCAs: N = 14
Placebo: N = 14
Postherpetic neuralgia HPT 8 wk Change in pain intensity Regression (M):
Opioids: R2 = 0.35 using HPT
TCA: no prediction
Placebo: no prediction
Yarnitsky et al.110 2012 Duloxetine: N = 30 Painful diabetic neuropathy CDT, WDT, HPT, MDT, MPT, mTSP, hCPM* 5 wk Change in pain intensity Regression (M):
CPM: R2 = 0.673
Olesen et al.72 2013 Pregabalin: N = 31
Placebo: N = 29
Painful chronic pancreatitis EPT, PPT, differences in EPT and PPT in the affected and the unaffected area (EPT/PPT ratio) 3 wk Classify responders (>30% pain reduction) and nonresponders Support vector machine (M):
EPT ratio:
Sensitivity: 87.5%
Specificity: 80.0%
Demant et al.21 2014 Oxcarbazepine: N = 48
Placebo: N = 35
Peripheral neuropathic pain ALL, WDT, CDT, HPT, CPT, PHS MDT, MPT, MPS, mTSP, VDT, PPT (DFNS protocol) 6 wk Patients classified into irritable nociceptor (IN) or nonirritable nociceptor (NIN). Numbers needed to treat (NNTs) to obtain 50% pain relief for the IN and NIN. NNT for the IN: 3.9
NNT for the NIN: 13 (U)
Demant et al.20 2015 Lidocaine 5% patch: N = 40
Placebo: N = 40
Peripheral neuropathic pain ALL, WDT, CDT, HPT, CPT, PHS MDT, MPT, MPS, mTSP, VDT, PPT (DFNS protocol) 4 wk Patients classified into irritable nociceptor (IN) or nonirritable nociceptor (NIN). Numbers needed to treat (NNTs) to obtain 50% pain relief for the IN and NIN. NNT for the IN: 7.5
NNT for the NIN: not possible to determine due to recruitment issues (U)
Arendt-Nielsen et al.5 2016 COX-2 inhibitor: N = 37
Placebo: N = 37
Knee osteoarthritis PPT, pTSP, pCPM 4 wk Change in pain intensity for nonresponders (<30% and <50% reduction) Correlation (U):
pTSP predicting a nonresponse: R = 0.421-0.639
Edwards et al.24 2016 NSAID (topical gel): N = 35 Knee osteoarthritis PPT, mTSP, pCPM†, pCPM 4 wk Change in pain average daily pain intensity (ADP) and KOOS pain Regressions (M):
ADP: CPM: R = −0.38 CPM
Mainka et al.63 2016 Capsaicin (topical, 8%): N = 20 Peripheral neuropathic pain ALL, WDT, CDT, HPT, CPT, PHS MDT, MPT, MPS, mTSP, VDT, PPT (DFNS protocol) 8 wk Classifying responders or nonresponders; responders were defined as +30% reduction in pain or 2 points on a 0-10 NRS Regression (M):
Sensitivity: 70%, specificity: 100% for patients with CPT and MPT >0.8 compared with the z value from DFNS.
Petersen et al.75 2019 NSAID and paracetamol (oral): N = 132 Knee osteoarthritis cPPT, cPTT, cTSP 3 wk Change in pain intensity Regression (M):
R2 = 0.269 using the VAS and cTSP
Petersen et al.77 2019 NSAID and paracetamol (oral): N = 42 Knee osteoarthritis OA, cCPM 3 wk Change in pain intensity Correlation (U):
R2 = 0.186 using the cCPM and VAS
Kisler et al.51 2019 Duloxetine: N = 27
Placebo: N = 28
Migraine Tonic heat pain (47°C), mTSP, OA, TSalone, TSconditioned, hCPM† 8 wk Migraine improvement (pain reduction) Regression (M):
TSalone (R = 0.47), TSconditioned (R = 0.49)
QST modalities: ALL, dynamic mechanical allodynia; CDT, cold detection threshold; CPT, cold pain threshold; cPPT, cuff-induced pressure pain threshold; cPTT, cuff-induced pressure tolerance threshold; CPM, conditioning pain modulation (c, cuff pressure test and condition stimuli; e, electrical test stimulus; heat, heat test stimulus; p, pressure test stimulus; *, hot water condition stimulus; †, cold-pressor tests as condition stimulus); EDT, electrical detection threshold; EIH, exercise-induced hypoalgesia; EPT, electrical pain threshold; EPTT, electrical pain tolerance threshold; HPT, heat pain threshold; (M), indicates multivariate analysis; MDT, mechanical detection threshold; MPT, mechanical pain threshold; N, number of patients participating in the study; s, offset analgesia; OR, odds ratio; PPT, pressure pain threshold; PTT, pressure tolerance threshold; R, coefficient of determination; STCS, suprathreshold cold stimulus; STHS, suprathreshold heat stimulus; TSP, temporal summation of pain (c, using cuff stimuli; e, using electrical stimuli; h, using heat stimuli; m, using mechanical pinprick stimuli; p, using pressure stimuli); (U), indicates univariate analysis; VAS, visual analog scale; WDT, warm detection threshold.

Most studies (27 studies) reported using multivariate statistical models, and univariate analyses were reported in 9 studies.

3.1. Quality assessment

The quality assessment is presented in Table 4 (surgical) and Table 5 (pharmaceutical). The reviewers (K.K.P. and D.B.L.) initially agreed on 89% of the ratings. Consensus was reached on all ratings after discussion.

Table 4 - Risk of bias (RoB) for studies investigating the prognostic value of QST parameters on chronic postoperative pain.
Study participation Study attrition Prognostic factor measurement Outcome measurement Study confounding Statistical analysis and reporting
Martinez et al.66 M L M L M L
Lundblad et al.61 M H M M H L
Yarnitsky et al.109 M L L M L M
Aasvang et al.1 L L L M H L
Wilder-Smith et al.103 L L L M M M
Gwilym et al.40 M L L L H M
Wylde et al.105 M L L L M M
Grosen et al.39 L M L L L L
Jarrell et al.48 L H M M H L
Valencia et al.98 M M L L L L
Noiseux et al.71 M L L L M L
Petersen et al.74 M L L M M L
Wylde et al.106 L M L M L L
Petersen et al.78 L M L M M L
Bayman et al.16 M M L L L L
Bossmann et al.17 M M L L L L
Andersen et al.4 L L M M L L
Vaegter et al.92 M L L L M L
Arendt-Nielsen et al.9 L M L M L M
Petersen et al.76 L M M L M L
Rice et al.81 L L L L M L
Sng et al.89 L M L M M L
Kurien et al.56 M L L M M L
Habib et al.41 L L L L M H
Müller et al.69 L M L L L L
Using the Quality in Prognostic Studies (QUIPS) tool, the RoB assessment was based on study participation, study attrition, prognostic factor measurement, outcome measurement, study confounding, and statistical analysis. In general, low-to-moderate risk of bias was observed in most studies distributed across all factors. Risk of bias was evaluated as low (L), medium (M) or high (H).
QST, quantitative sensory testing.

Table 5 - Risk of bias (RoB) for studies investigating the prognostic value of QST parameters on the analgesic effect after nonsurgical treatments.
Study participation Study attrition Prognostic factor measurement Outcome measurement Study confounding Statistical analysis and reporting
Edwards et al.25 M M M L M L
Yarnitsky et al.110 M M L L M L
Olesen et al.72 L L L L M M
Demant et al.21 L L L M M L
Demant et al.20 L L M L M L
Arendt-Nielsen et al.5 L M L L M L
Edwards et al.24 M M L L M L
Mainka et al.63 L L L L M M
Petersen et al.75 L M L L M L
Petersen et al.77 L L L L M L
Kisler et al.51 M M L M M L
Using the Quality in Prognostic Studies (QUIPS) tool, the RoB assessment was based on study participation, study attrition, prognostic factor measurement, outcome measurement, study confounding, and statistical analysis. In general, low-to-moderate risk of bias was observed in most studies distributed across all factors.
QST, quantitative sensory testing.

3.2. Quantitative sensory testing variables

3.2.1. Electrical stimuli

Electrical stimuli were reported as electrical detection threshold, electrical pain threshold, or electrical pain tolerance threshold. Electrical stimuli were reported in 4 studies and found predictive in 2 studies (2/4 = 50%).61,72

3.2.1.1. Electrical detection threshold

Electrical detection threshold was reported in 1 surgical study and no pharmacological studies. Preoperative electrical pain threshold was not found significantly associated with the chronic postoperative pain intensity after total knee arthroplasty.61

3.2.1.2. Electrical pain threshold

Electrical pain threshold was reported in 2 surgical studies and 1 pharmacological study. Low preoperative electrical pain threshold and high pain at rest predicted the chronic postoperative pain intensity after total knee arthroplasty.61 Electrical pain threshold was not significantly associated with persistent chronic postoperative pain or disability after segment spinal surgery of patients with chronic low back pain.69

In the pharmacological studies, using a support vector machine analysis, the pretreatment ratio between electrical pain threshold at a pancreatic referred dermatome vs a nonaffected dermatome predicted the response to pregabalin (reduction in the clinical pain score of 30% or more after 3 weeks of treatment compared with placebo) with a sensitivity of 87.5% and a specificity of 80.0%.72

3.2.1.3. Electrical pain tolerance threshold

Electrical pain tolerance threshold was reported in 1 surgical study in which the preoperative electrical pain tolerance threshold was not associated with the chronic postoperative pain intensity after major abdominal surgery.103 No pharmacological studies reported electrical pain tolerance threshold.

3.2.2. Thermal stimuli

Thermal stimuli were reported as cold and warm detection threshold, cold and heat pain threshold, and suprathreshold heat and cold stimuli. Thermal stimuli were reported in 11 studies and found predictive in 5 studies (5/11 ≈ 45%).1,25,51,63,76

3.2.2.1. Warm detection threshold

Warm detection thresholds were reported in 3 surgical studies and 1 pharmacological study. One surgical study reported a linear regression model demonstrating that low preoperative warm detection threshold, low heat pain threshold, low degree of radiologically assessed osteoarthritis, and high TSP predicted a high chronic postoperative pain intensity after total knee arthroplasty in patients with knee osteoarthritis.76 Two surgical studies did not find an association between preoperative warm detection thresholds and chronic postoperative pain intensity (groin hernia repair surgery1 and breast cancer surgery4). One pharmacological (duloxetine for the treatment of diabetic peripheral neuropathy110) study did not find associations between the preoperative warm detection threshold and analgesic effect.

3.2.2.2. Heat pain threshold

Heat pain thresholds were reported in 9 surgical and 3 pharmacological studies. One surgical study reported a linear regression model including low preoperative heat pain threshold, low warm detection threshold, low degree of radiologically assessed osteoarthritis, and high TSP, which predicted a high chronic postoperative pain intensity after total knee arthroplasty in patients with knee osteoarthritis.76 Eight surgical studies did not find associations between preoperative heat pain thresholds and chronic postoperative pain intensity (thoracic surgery,109 groin hernia repair surgery,1 arthroscopic surgery of the shoulder,98 and breast cancer surgery4), postoperative Western Ontario and McMaster Universities Osteoarthritis Index scores (total knee arthroplasty105), and the presence of moderate-to-severe chronic postoperative pain (total knee arthroplasty71 and segmental spinal surgery69).

In the pharmacological studies, a hierarchical regression model demonstrated that low heat pain threshold was associated with a small analgesic effect of opioids in patients with postherpetic neuralgia.25 Furthermore, using multivariate regression models, a great analgesic response to duloxetine in patients with migraine51 and painful diabetic neuropathy110 was not significantly associated with pretreatment heat pain thresholds.

3.2.2.3. Cold detection threshold

Cold detection thresholds were reported in 2 surgical studies and 1 pharmacological study. Preoperative or pretreatment cold detection thresholds were not statistically significantly associated with the chronic postoperative pain intensity (total knee arthroplasty76 and breast cancer surgery4) or the analgesic effect (duloxetine for diabetic peripheral neuropathy110), respectively.

3.2.2.4. Cold pain threshold

Cold pain thresholds were reported in 4 surgical and no pharmacological studies. Preoperative cold pain thresholds were not significantly associated with the chronic postoperative pain intensity (total knee arthroplasty66,76) and the presence of moderate-to-severe chronic postoperative pain (thoracic surgery16 and segmental spinal surgery69).

3.2.2.5. Suprathreshold heat and cold stimuli

Suprathreshold heat and cold stimuli were reported in 3 surgical and no pharmacological studies. In a logistic regression model, preoperative high pain intensities to suprathreshold heat stimuli along with lowered warm detection thresholds and pain-related impairment of activity were predictive of the presence of moderate-to-severe chronic postoperative pain after hernia repair.1 Preoperative suprathreshold heat stimuli and suprathreshold cold stimuli were not significantly associated with postoperative pain intensity (thoracic surgery109) and the presence of moderate-to-severe chronic postoperative pain (thoracic surgery16).

3.2.3. Cutaneous mechanical stimuli

Cutaneous mechanical stimuli were reported as mechanical detection and pain threshold. Cutaneous mechanical stimuli were reported in 7 and predictive in no studies (0/7 = 0%).

3.2.3.1. Mechanical detection threshold

Mechanical detection thresholds were reported in 2 surgical studies and 1 pharmacological study. Preoperative and pretreatment mechanical detection thresholds were not statistically significantly associated with the chronic postoperative pain intensity (surgical correction of funnel chest39 and breast cancer surgery4) or analgesic effect (duloxetine for the treatment of diabetic peripheral neuropathy110).

3.2.3.2. Mechanical pain threshold

Mechanical pain thresholds were reported in 4 surgical studies and 1 pharmacological study. Preoperative and pretreatment mechanical pain thresholds were not associated with the chronic postoperative pain intensity (total knee arthroplasty66 and breast cancer surgery4), presence of moderate-to-severe postoperative pain (total knee arthroplasty71), postoperative Oxford shoulder score (arthroscopic subacromial decompression40), or analgesic effect (duloxetine for the treatment of diabetic peripheral neuropathy110).

3.2.4. Pressure stimuli

Pressure stimuli were studied as pressure pain and tolerance threshold as well as cuff-induced pressure pain and tolerance thresholds. Deep pressure stimuli were reported in 17 studies and were predictive in 5 studies (5/17 ≈ 29%).9,56,78,105,106

3.2.4.1. Pressure pain threshold

Pressure pain thresholds were reported in 11 surgical and 3 pharmacological studies. Low pressure pain thresholds assessed at the osteoarthritic affected knee56 were associated with postoperative pain, low pressure pain thresholds at the nonaffected knee9 were associated with postoperative pain relief, and low pressure pain thresholds at the forearm105 were associated with postoperative high Western Ontario and McMaster Universities Osteoarthritis Index scores after total knee arthroplasty. A study found that low pressure pain thresholds assessed at the forearm in patients with hip osteoarthritis were associated with high postoperative Western Ontario and McMaster Universities Osteoarthritis Index scores after total hip arthroplasty but not in patients with knee osteoarthritis after total knee arthroplasty.106 Eight studies did not find a statistically significant association between preoperative pressure pain thresholds and chronic postoperative pain intensity (total knee arthroplasty,56 surgical correction of funnel chest,39 and breast cancer surgery4), postoperative pain relief (total knee arthroplasty78,92), or the presence of moderate-to-severe postoperative pain (total knee arthroplasty71,81 and segmental spinal surgery69).

For the pharmacological studies, 3 studies found no association between pretreatment pressure pain thresholds and analgesic effect (COX-25 inhibitors and NSAID gels24 for the treatment of painful knee osteoarthritis and pregabalin for the treatment of painful chronic pancreatitis72).

3.2.4.2. Pressure tolerance threshold

Pressure tolerance thresholds were reported in 2 surgical and no pharmacological studies. Preoperative pressure tolerance threshold was not associated with the chronic postoperative pain intensity (major abdominal surgery103) and presence of persistent pain and disability (segmental spinal surgery69).

3.2.4.3. Cuff-induced pain detection threshold

Cuff-induced pressure pain thresholds were reported in 3 surgical studies and 1 pharmacological study. Low cuff-induced pressure pain threshold assessed at the lower leg in patients with knee osteoarthritis was associated with chronic postoperative pain relief after total knee arthroplasty.78 Two studies did not find a significant association between preoperative cuff-induced pressure pain thresholds and the chronic postoperative pain intensity (total knee arthroplasty56) or postoperative pain relief (total knee arthroplasty92). One pharmacological study demonstrated no association between the pretreatment cuff-induced pressure pain threshold and analgesic effect (oral NSAIDs and paracetamol for the treatment of pain in knee osteoarthritis75).

3.2.4.4. Cuff-induced pressure tolerance threshold

Cuff-induced pain tolerance thresholds were reported in 3 surgical studies and 1 pharmacological study. Three surgical studies did not find a significant association between the preoperative cuff-induced pain tolerance threshold and the chronic postoperative pain intensity (total knee arthroplasty56) or postoperative pain relief (total knee arthroplasty78,92). The pharmacological study did not find a significant association between the pretreatment cuff-induced pain tolerance threshold and analgesic effect (oral NSAIDs and paracetamol for the treatment of pain in knee osteoarthritis75).

3.2.5. Dynamic mechanical allodynia

Dynamic mechanical allodynia was reported in 3 surgical and no pharmacological studies and was reported predictive in 1 study (1/3 ≈ 33%). The presence of preoperative dynamic mechanical allodynia (yes/no) was associated with the chronic postoperative pain intensity in females undergoing gynecologic laparoscopy.48 Two studies did not find associations between preoperative dynamic mechanical allodynia and the postoperative pain intensity (surgical correction of funnel chest39 and total knee arthroplasty66).

3.2.6. Temporal summation of pain

Temporal summation of pain was reported in 14 studies (9 surgical and 5 pharmacological studies) and predictive in 7 studies (7/14 = 50%).5,56,74–76,81,89

In the surgical studies, TSP was assessed using mechanical stimuli,56,74,76,81,89 heat stimuli,98 and cuff stimuli.56,78 In 5 surgical studies, high preoperative TSP was associated with the chronic postoperative pain intensity after total knee arthroplasty56,74,76 and the presence of moderate-to-severe chronic postoperative pain (total knee arthroplasty81 and abdominal or laparoscopic hysterectomy89). Four studies reported that preoperative TSP was not associated with the chronic postoperative pain intensity (total knee arthroplasty78 and arthroscopic shoulder surgery98), postoperative WOMAC (total knee arthroplasty17), and the presence of moderate-to-severe chronic postoperative pain (breast cancer surgery41).

In the pharmacological studies, TSP was assessed using mechanical stimuli,24,51,110 computer-controlled pressure stimuli,5 and manual cuff stimuli.75 In 2 studies, high TSP was reported to be associated with poor analgesic effect after 4 weeks of treatment with COX-2 inhibitors5 and 3 weeks of treatment with NSAIDs and paracetamol75 in patients with knee osteoarthritis. Three studies did not find an association between pretreatment TSP and the analgesic effect of duloxetine for diabetic peripheral neuropathy,110migraine,51 or topical NSAIDs for painful knee osteoarthritis.24

3.2.7. Conditioned pain modulation

Conditioned pain modulation was reported in 17 studies (12 surgical and 5 pharmacological studies) and predictive in 7 studies (7/17 ≈ 41%).24,77,92,103,109,110

In the surgical studies, the test stimulus was assessed using pressure,17,39,69,74,81,92,103 cuff,56,78 heat,41,69,98,109 and electrical69,103 stimuli, with the conditioning stimuli being hot water,41,98,109 cold water,39,69,74,81,92,103 tonic cuff pressure,56,78 or pinching.17 In 4 studies, preoperative impaired CPM was associated with chronic postoperative pain intensity (thoracic surgery109 and major abdominal surgery103), high postoperative WOMAC scores (total knee arthroplasty17), and a reduction in postoperative pain relief (total knee arthroplasty92). Eight studies did not find an association between preoperative CPM and the chronic postoperative pain intensity (total knee arthroplasty,56,74,78,81 surgical correction of funnel chest,39 and arthroscopic shoulder surgery98), postoperative pain relief (total knee arthroplasty78), and the presence of moderate-to-severe chronic postoperative pain (total knee arthroplasty,81 breast cancer surgery,41 and segmental spinal surgery69).

In the pharmacological studies, the test stimulus was assessed using pressure,5,24 cuff,77 and heat51,110 stimuli, with the conditioning stimuli being hot water,110 cold water,51 or tonic cuff pressure.5,24,77 Impaired CPM was associated with a great analgesic effect of duloxetine in patients with diabetic neuropathy.110 This was also the case with topical NSAIDs24 or NSAIDs and paracetamol77 in patients with knee osteoarthritis. Pretreatment CPM was not associated with the analgesic effect of COX-2 inhibitors in patients with knee osteoarthritis5 or duloxetine in patients with migraine.51 Of note, Kisler et al.51 found that the pretreatment test stimulus and the conditioned test stimulus in the CPM paradigm predicted the analgesic effect of duloxetine in patients with migraine but not the calculated CPM effect itself.

3.2.7.1. Offset analgesia

Offset analgesia was reported in 1 pharmacological study in which the pretreatment offset analgesia showed no association with the analgesic effect of NSAID and paracetamol for patients with knee osteoarthritis.77

3.2.7.2. Exercise-induced hypoalgesia

Exercise-induced hypoalgesia was reported in 1 surgical study in which low preoperative exercise-induced hypoalgesia was associated with low postoperative pain relief after total knee arthroplasty.92

3.2.8. Hypoesthesia area

One study assessed the preoperative size of the hypoesthesia area using warm (40°C) and cold (25°C) rolls and found an association between the size of the hypoesthesia and the chronic postoperative pain intensity after breast cancer surgery.4

3.2.9. The German Research Network on Neuropathic Pain (DFNS) protocol

The German Research Network on Neuropathic Pain (DFNS) protocol consists of a wide range of QST modalities including allodynia, thermal detection and pain thresholds, paradoxical heat sensations, mechanical detection, pain thresholds, mechanical suprathreshold, TSP (wind-up), vibration detection threshold, and pressure pain threshold. The DFNS protocol was assessed in 3 pharmacological studies. In 2 pharmacological studies, the DFNS protocol was used to define the irritable nociceptor (IN) or non-IN (NIN) in patients with peripheral neuropathic pain. One study found that the number needed to treat for 50% pain relief of oxcarbazepine was 3.9 for the IN and 13 for the NIN.21 A study found that the number needed to treat for 50% pain relief of lidocaine 5% patch was 7.5 for the IN and not definable for the NIN due to recruitment issues. Finally, the DFNS protocol was used to predict the responders (+30% pain alleviation) and nonresponders to capsaicin patch treatment in patients with peripheral neuropathy and found a sensitivity of 70% and a specificity of 100% for patients with cold pain thresholds and mechanical pain thresholds >0.8 compared with z values from DFNS.63

3.2.10. Prediction of specific surgical procedures

In this review, 16 studies addressed joint-related surgeries, 3 studies addressed thoracic-related surgeries, 4 studies addressed abdominal and gynecology-related surgeries, and 2 studies addressed breast cancer surgeries. A significant preoperative prediction was demonstrated for 11 studies (69%) in the joint-related surgeries, 1 study (33%) in the thoracic-related surgeries, 4 studies (100%) in the abdominal and gynecology-related surgeries, and 1 study (50%) related to breast cancer surgery.

4. Discussion

The current systematic review describes the predictive role of QST on pain after surgical and pharmacological interventions. Twenty-five surgical (10 new studies since the latest review84) and 11 pharmacological (8 studies since the latest review38) studies published since 2000 were identified. Seventeen studies demonstrated an association between preoperative QST and chronic postoperative pain, and 11 studies demonstrated an association between pretreatment QST and the analgesic effect of pharmacological interventions but with a large heterogeneity in the QST paradigms used. Significant preoperative predictions were most often presented for joint-related surgeries and abdominal and gynecology-related surgeries. Temporal summation of pain, CPM, and different variations of pressure thresholds were the most frequently reported methods, and TSP and CPM were most frequently found as predictors of the chronic postoperative pain intensity, the presence of moderate-to-severe chronic postoperative pain, postoperative pain relief, and the analgesic effect to pharmacological interventions.

4.1. The predictive value of quantitative sensory testing

This review suggests a possible association between the selected QST parameters and chronic postoperative pain and the analgesic response to pharmacological interventions, but the results are not consistent. Overall, the most used QST paradigms were mechanical and pressure stimuli, TSP, and CPM, which were also most frequently associated with chronic postoperative pain or analgesic effect. In addition, the strength of the predictive value of QST varied with R2 values ranging from 0.13 to 0.673, which further underlines that variance explanation remains suboptimal at best. Finally, it is important to acknowledge that the literature on the predictive role of QST for chronic postoperative pain and analgesic response to pharmacological interventions is conflicting, and therefore, QST might not be appropriate as clinical guiding tool yet.

Decades of research has focused on the difference in QST parameters comparing pain-free subjects and patients with chronic pain; yet, the differences have not been established to be specific for the pain diagnosis.7,34,82,99,104 It is evident that some patients with chronic pain are generally more pain sensitive than others,6,91,99 but the underlying factors driving the increased pain sensitivity are still largely unclear. Studies have suggested that the pain sensitivity can be increased in patients with chronic pain because of, for example, sleep impairment,88 increased pain catastrophizing,68 or comorbidities such as diabetes.27 These factors are often observed in patients in chronic pain populations and warrant consideration when addressing the predictive value of QST in future studies.

A previous review84 suggested a link between different QST modalities and certain pain disorders. To exemplify this, cutaneous stimuli have been found to activate cutaneous fibres,3 and pressure stimuli using algometers have been found to target muscles30 or fascia.64 This could indicate that, for example, cutaneous activation would be suited for dermatological disorders, and pressure stimuli would be suited for patients with muscle or joint pain. The most studied population of the current review was patients with osteoarthritis, and the hypothesis that pressure stimuli should adhere better to these patients is not supported by the current review because pressure stimuli were rarely (approx. 29%) predictive of chronic postoperative pain or analgesic effect. In addition, joint-related surgeries and abdominal and gynaecology-related surgeries were most often studied, and preoperative QSTs were most often associated with chronic postoperative pain, which could indicate that preoperative QSTs are more frequently predictive of certain surgical procedures. Future studies should pursue this hypothesis to clarify whether there is an interaction between certain QST modalities and certain pain disorders.

This review investigated the predictive role of QST for chronic postoperative pain and analgesic response to pharmacological interventions, although these 2 outcomes are different. In a pain mechanistic context, surgical procedures aim to remove the peripheral pain driver, and it has been demonstrated that a pain-free recovery after removal of such peripheral drives by, for example, total knee37 or hip arthroplasty54 does normalize the responses to CPM and TSP when comparing preoperative and 6-month postoperative assessments. The advantage of certain pharmacological interventions is the possibility to target central pain mechanisms. As an example, preclinical trials have established that serotonin and noradrenalin are important for descending pain inhibition,13,14,60 and human administration of duloxetine (a serotonin and noradrenaline reuptake inhibitor) does improve CPM in patients with painful diabetic neuropathy.110 Likewise, the N-methyl-d-aspartate receptors are important for dorsal horn neuron excitability, and human administration of ketamine (an N-methyl-d-aspartate antagonist) reduces TSP in patients with fibromyalgia.36 Therefore, specific QST modalities might be better predictors for a certain treatment if the pain mechanistic profile is matched with the intervention.

The most consistent predictive QST paradigms were TSP and CPM, but these studies used multiple different protocols (assessment parameters and modalities). The reliability of CPM has been questioned,50 and studies comparing CPM protocols have highlighted that different test and conditioning stimuli combinations will yield different reliability results.45,97 Some CPM protocols operate under the premise that the conditioning stimulus should be painful (visual analog scale > 3) and that the CPM effect increases with increased conditioning stimulus intensity.35 Conversely, other protocols are based on the notion that the intensity of the conditioning stimulus is independent of the CPM effect.33 The 2015 recommendation for CPM108 called for standardization of CPM testing across laboratories, which would greatly increase the generalizability of CPM in future studies. However, the current review highlights that the variability in different QST modalities has the potential to be used as a predictor of pain after surgery and pharmacological interventions. It is well known that dynamic QST modalities such as TSP and CPM are influenced by multiple factors such as sleep deprivation,26,59 use of opioids,65 or high clinical pain intensities.10 Therefore, standardized protocols based on pain-free subjects might not represent the clinical use of QST.

4.2. Quality assessment

In this systematic review, the included studies overall demonstrated a low-to-moderate bias distributed among the 4 categories for bias assessment. For study participation, missing information on sampling frame and recruitment and place for assessment were the main causes for bias. For study attrition, lack of description of loss to follow-up and differences in participants completing and not completing the study were the main reasons for bias. In the category of prognostic factor measurement, validity and reliability of the used methods together with the uncertainty with respect to the cutoffs chosen were the main reasons for bias. Confounding was another category that revealed substantial bias, in that confounders in either study design or analysis were rarely accounted for and that multivariate analyses were not performed consistently.

4.3. Future directions

The current review highlights inconsistencies in QST as predictors of outcomes after surgical treatments and pharmacological interventions. The most consistent findings include studies assessing central pain mechanisms, and therefore, future studies are encouraged to include parameters such as TSP or CPM and to understand the variability in the assessment of TSP and CPM. Fjeld et al., 2019,31 studied patients with acute admission due to sciatica and found that the assessment of CPM 6 weeks after discharge did not predict pain at 12 months of follow-up, which does highlight that the timing of assessment might be crucial. In addition, other non-QST parameters such as pain catastrophizing or epigenetic markers have been associated with chronic postoperative pain15,22,32,79,80 and analgesic effects.39 Including these parameters will likely increase the strength of the predictive models in future studies. Finally, because of the large heterogeneity in choice of QST parameters, standardization is needed and assessment of both painful and nonpainful anatomic regions should be considered because this could influence the predictive value of QST.105,106

In 2016, a task force appointed by the International Association for the Study of Pain53 suggested “nociplastic” as a new pain descriptor to describe a state in which the nociceptive functions are changed. Several studies have now identified that specific subgroups of patients with chronic pain exist6,91 and that these patients respond poorly to standard care treatments.42,73–75 Understanding whether the modulation of specific central pain pathways before standard care treatments does increase the likelihood of positive analgesic responses would move this field forward. In this regard, acknowledging that a large variety of sensitization manifestations may represent a large variety of natural or acquired sensorimotor function modalities might be important.

Studies have found that, for example, duloxetine can improve CPM110 and ketamine can reduce TSP.36 A recent study52 on patients scheduled for total knee arthroplasty recruited “pain sensitive” patients using the Central Sensitization Inventory Questionnaire and found that preoperative and 6-week postoperative administration of duloxetine reduced the postoperative pain intensity at 2 to 12 weeks of follow-up compared with placebo. Studies combining preoperative predictors with pharmacological interventions targeting these predictions might pave the way for the future of personalized pain medicine.

4.4. Limitations

The current review is limited to studies assessing chronic postoperative pain defined as pain reported at least 3 months after surgery. Several studies on the predictive role of QST for acute postoperative pain have been conducted (see review from 201784). Of note, studies such as Lunn et al., 2013, and Izumi et al., 2017, studied the postoperative period from 30 days to 6 weeks after surgery, and these studies are not included in the current review.

Furthermore, the current review is limited by not addressing studies using QST to predict the acute effect of pharmacological interventions such as Wasner et al.100 and Schliessbach et al.85

5. Conclusion

This systematic review identified 17 surgical and 11 pharmacological studies reporting a predictive role of QST modalities for chronic postoperative pain or the analgesic effect of pharmacological interventions. The strengths of the predictive models vary, and no consistency was found for a single QST parameter. Pressure stimuli and dynamic QST parameters such as TSP and CPM were the most frequently assessed, and thermal stimuli, TSP, and CPM were most frequently associated with chronic postoperative pain or analgesic effects.

Conflict of interest statement

The authors have no conflicts of interest to declare.

Acknowledgements

The authors thank the Aalborg University Talent Management Programme (j.no. 771126) for providing the funding to initiate this work. The Center for Neuroplasticity and Pain (CNAP) is supported by the Danish National Research Foundation (DNRF121).

References

[1]. Aasvang EK, Gmaehle E, Hansen JB, Gmaehle B, Forman JL, Schwarz J, Bittner R, Kehlet H. Predictive risk factors for persistent postherniotomy pain. Anesthesiology 2010;112:957–69.
[2]. Abrishami A, Chan J, Chung F, Wong J. Preoperative pain sensitivity and its correlation with postoperative pain and analgesic consumption. Anesthesiology 2011;114:445–57.
[3]. Andersen HH, Elberling J, Sølvsten H, Yosipovitch G, Arendt-Nielsen L. Nonhistaminergic and mechanical itch sensitization in atopic dermatitis. PAIN 2017;158:1780–91.
[4]. Andersen KG, Duriaud HM, Kehlet H, Aasvang EK. The relationship between sensory loss and persistent pain 1 year after breast cancer surgery. J Pain 2017;18:1129–38.
[5]. Arendt-Nielsen L, Egsgaard LL, Petersen KK. Evidence for a central mode of action for etoricoxib (COX-2 inhibitor) in patients with painful knee osteoarthritis. PAIN 2016;157:1634–44.
[6]. Arendt-Nielsen L, Egsgaard LL, Petersen KK, Eskehave TN, Graven- Nielsen T, Hoeck HC, Simonsen O. A mechanism-based pain sensitivity index to characterize knee osteoarthritis patients with different disease stages and pain levels. Eur J Pain 2015;19:1406–17.
[7]. Arendt-Nielsen L, Morlion B, Perrot S, Dahan A, Dickenson A, Kress HG, Wells C, Bouhassira D, Mohr Drewes A. Assessment and manifestation of central sensitisation across different chronic pain conditions. Eur J Pain 2018;22:216–41.
[8]. Arendt-Nielsen L, Nie H, Laursen MB, Laursen BS, Madeleine P, Simonsen OH, Graven-Nielsen T. Sensitization in patients with painful knee osteoarthritis. PAIN 2010;149:573–81.
[9]. Arendt-Nielsen L, Simonsen O, Laursen MB, Roos EM, Rathleff MS, Rasmussen S, Skou ST. Pain and sensitization after total knee replacement or nonsurgical treatment in patients with knee osteoarthritis: identifying potential predictors of outcome at 12 months. Eur J Pain 2018;22:1088–102.
[10]. Arendt-Nielsen L, Skou STST, Nielsen TA, Petersen KK. Altered central sensitization and pain modulation in the CNS in chronic joint pain. Curr Osteoporos Rep 2015;13:225–34.
[11]. Arendt-Nielsen L, Yarnitsky D. Experimental and clinical applications of quantitative sensory testing applied to skin, muscles and viscera. J Pain 2009;10:556–72.
[12]. Arezzo JC, Bolton CF, Boulton A, Zimmerman I, Casey KL, Dyck PJ, CornblatH D, FieldS HL, HansoN P, GreenspaN J. Quantitative sensory testing-a consensus report from the peripheral neuropathy association. Neurology 1993;43:1050–2.
[13]. Bannister K, Lockwood S, Goncalves L, Patel R, Dickenson AH. An investigation into the inhibitory function of serotonin in diffuse noxious inhibitory controls in the neuropathic rat. Eur J Pain 2017;21:750–60.
[14]. Bannister K, Patel R, Goncalves L, Townson L, Dickenson AH. Diffuse noxious inhibitory controls and nerve injury: restoring an imbalance between descending monoamine inhibitions and facilitations. PAIN 2015;156:1803–11.
[15]. Baranoff J, Hanrahan SJ, Connor JP. The roles of acceptance and catastrophizing in rehabilitation following anterior cruciate ligament reconstruction. J Sci Med Sport 2015;18:250–4.
[16]. Bayman EO, Parekh KR, Keech J, Selte A, Brennan TJ. A prospective study of chronic pain after thoracic surgery. Anesthesiology 2017;126:938–51.
[17]. Bossmann T, Brauner T, Wearing S, Horstmann T. Predictors of chronic pain following total knee replacement in females and males: an exploratory study. Pain Manag 2017;7:391–403.
[18]. Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain 2006;10:287–333.
[19]. Burrows NJ, Booth J, Sturnieks DL, Barry BK. Acute resistance exercise and pressure pain sensitivity in knee osteoarthritis: a randomised crossover trial. Osteoarthr Cartil 2014;22:407–14.
[20]. Demant DT, Lund K, Finnerup NB, Vollert J, Maier C, Segerdahl MS, Jensen TS, Sindrup SH. Pain relief with lidocaine 5% patch in localized peripheral neuropathic pain in relation to pain phenotype: a randomised, double-blind, and placebo-controlled, phenotype panel study. PAIN 2015;156:2234–44.
[21]. Demant DT, Lund K, Vollert J, Maier C, Segerdahl M, Finnerup NB, Jensen TS, Sindrup SH. The effect of oxcarbazepine in peripheral neuropathic pain depends on pain phenotype: a randomised, double-blind, placebo-controlled phenotype-stratified study. PAIN 2014;155:2263–73.
[22]. Dworkin RH, Peirce-Sandner S, Turk DC, McDermott MP, Gibofsky A, Simon LS, Farrar JT, Katz NP. Outcome measures in placebo-controlled trials of osteoarthritis: responsiveness to treatment effects in the REPORT database. Osteoarthr Cartil 2011;19:483–92.
[23]. Dyck PJ, Zimmerman I, Gillen DA, Johnson D, Karnes JL, O'Brien PC. Cool, warm, and heat-pain detection thresholds: testing methods and inferences about anatomic distribution of receptors. Neurology 1993;43:1500–8.
[24]. Edwards RR, Dolman AJ, Martel MO, Finan PH, Lazaridou A, Cornelius M, Wasan AD. Variability in conditioned pain modulation predicts response to NSAID treatment in patients with knee osteoarthritis. BMC Musculoskelet Disord 2016;17:284.
[25]. Edwards RR, Haythornthwaite Ja, Tella P, Max MB, Raja S. Basal heat pain thresholds predict opioid analgesia in patients with postherpetic neuralgia. Anesthesiology 2006;104:1243–8.
[26]. Eichhorn N, Treede R-D, Schuh-Hofer S. The role of sex in sleep deprivation related changes of nociception and conditioned pain modulation. Neuroscience 2018;1:191–200.
[27]. Eitner A, Pester J, Vogel F, Marintschev I, Lehmann T, Hofmann GO, Schaible H-G. Pain sensation in human osteoarthritic knee joints is strongly enhanced by diabetes mellitus. PAIN 2017;158:1743–53.
[28]. Fingleton C, Smart KM, Doody CM. Exercise-induced hypoalgesia in people with knee osteoarthritis with normal and abnormal conditioned pain modulation. Clin J Pain 2017;33:395–404.
[29]. Finnerup NB, Attal N, Haroutounian S, McNicol E, Baron R, Dworkin RH, Gilron I, Haanpää M, Hansson P, Jensen TS, Kamerman PR, Lund K, Moore A, Raja SN, Rice ASC, Rowbotham M, Sena E, Siddall P, Smith BH, Wallace M. Pharmacotherapy for neuropathic pain in adults: a systematic review and meta-analysis. Lancet Neurol 2015;14:162–73.
[30]. Finocchietti S, Takahashi K, Okada K, Watanabe Y, Graven-Nielsen T, Mizumura K. Deformation and pressure propagation in deep tissue during mechanical painful pressure stimulation. Med Biol Eng Comput 2013;51:113–22.
[31]. Fjeld OR, Grotle M, Matre D, Pedersen LM, Lie MU, Småstuen MC, Storheim K, Heuch I, Stubhaug A, Zwart J-A, Nilsen KB. Predicting the outcome of persistent sciatica using conditioned pain modulation: 1-year results from a prospective cohort study. Scand J Pain 2019;20:69–75.
[32]. Giordano R, Petersen KK, Andersen HH, Lichota J, Simonsen OH, Arendt-Nielsen L. MicroRNAs expression as circulating genomic biomarkers in patients with chronic postoperative pain after total knee replacement. In: MicroRNAs expression as circulating genomic biomarkers in patients with chronic postoperative pain after total knee replacement. Boston: 17th World Congress on Pain, IASP 2018, 2018.
[33]. Granot M, Weissman-Fogel I, Crispel Y, Pud D, Granovsky Y, Sprecher E, Yarnitsky D. Determinants of endogenous analgesia magnitude in a diffuse noxious inhibitory control (DNIC) paradigm: do conditioning stimulus painfulness, gender and personality variables matter? PAIN 2008;136:142–9.
[34]. Graven-Nielsen T, Arendt-Nielsen L. Assessment of mechanisms in localized and widespread musculoskeletal pain. Nat Rev Rheumatol 2010;6:599–606.
[35]. Graven-Nielsen T, Izumi M, Petersen KK, Arendt-Nielsen L. User-independent assessment of conditioning pain modulation by cuff pressure algometry. Eur J Pain 2017;21:552–61.
[36]. Graven-Nielsen T, Kendall SA, Henriksson KG, Bengtsson M, Sörensen J, Johnson A, Gerdle B, Arendt-Nielsen L. Ketamine reduces muscle pain, temporal summation, and referred pain in fibromyalgia patients. PAIN 2000;85:483–91.
[37]. Graven-Nielsen T, Wodehouse T, Langford RM, Arendt-Nielsen L, Kidd BL. Normalization of widespread hyperesthesia and facilitated spatial summation of deep-tissue pain in knee osteoarthritis patients after knee replacement. Arthritis Rheum 2012;64:2907–16.
[38]. Grosen K, Fischer IWD, Olesen AE, Drewes AM. Can quantitative sensory testing predict responses to analgesic treatment? Eur J Pain 2013;17:1267–80.
[39]. Grosen K, Vase L, Pilegaard HK, Pfeiffer-Jensen M, Drewes AM. Conditioned pain modulation and situational pain catastrophizing as preoperative predictors of pain following chest wall surgery: a prospective observational cohort study. PLoS One 2014;9:e90185.
[40]. Gwilym SE, Oag HCL, Tracey I, Carr AJ. Evidence that central sensitisation is present in patients with shoulder impingement syndrome and influences the outcome after surgery. J Bone Joint Surg Br 2011;93-B:498–502.
[41]. Habib AS, Kertai MD, Cooter M, Greenup RA, Hwang S. Risk factors for severe acute pain and persistent pain after surgery for breast cancer: a prospective observational study. Reg Anesth Pain Med 2019;44:192–9.
[42]. Hansen S, Vaegter HB, Petersen KK. Pretreatment exercise-induced hypoalgesia is associated with change in pain and function after standardized exercise therapy in painful knee osteoarthritis. Clin J Pain 2020;36:16–24.
[43]. Hayden JA, van der Windt DA, Cartwright JL, Côté P, Bombardier C. Assessing bias in studies of prognostic factors. Ann Intern Med 2013;158:280–6.
[44]. Hermans L, Calders P, Van Oosterwijck J, Verschelde E, Bertel E, Meeus M. An overview of offset analgesia and the comparison with conditioned pain modulation: a systematic literature review. Pain Physician 2016;19:307–26.
[45]. Imai Y, Petersen KK, Mørch CD, Arendt Nielsen L. Comparing test–retest reliability and magnitude of conditioned pain modulation using different combinations of test and conditioning stimuli. Somatosens Mot Res 2016;33:169–77.
[46]. Ip HYV, Abrishami A, Peng PWH, Wong J, Chung F. Predictors of postoperative pain and analgesic consumption. Anesthesiology 2009;111:657–77.
[47]. Izumi M, Petersen KK, Laursen MB, Arendt-Nielsen L, Graven-Nielsen T. Facilitated temporal summation of pain correlates with clinical pain intensity after hip arthroplasty. PAIN 2017;158:323–32.
[48]. Jarrell J, Ross S, Robert M, Wood S, Tang S, Stephanson K, Giamberardino MA. Prediction of postoperative pain after gynecologic laparoscopy for nonacute pelvic pain. Am J Obstet Gynecol 2014;211:360.e1–8.
[49]. Kehlet H, Jensen TS, Woolf CJ, Centre M. Persistent postsurgical pain : risk factors and prevention. Lancet 2006;367:1618–25.
[50]. Kennedy DL, Kemp HI, Ridout D, Yarnitsky D, Rice AS. Reliability of conditioned pain modulation. PAIN 2016;157:2410–19.
[51]. Kisler LB, Weissman-Fogel I, Coghill RC, Sprecher E, Yarnitsky D, Granovsky Y. Individualization of migraine prevention. Clin J Pain 2019;35:753–65.
[52]. Koh IJ, Kim MS, Sohn S, Song KY, Choi NY, In Y. Duloxetine reduces pain and improves quality of recovery following total knee arthroplasty in centrally sensitized patients. J Bone Joint Surg 2019;101:64–73.
[53]. Kosek E, Cohen M, Baron R, Gebhart GF, Mico JA, Rice ASC, Rief W, Sluka AK. Do we need a third mechanistic descriptor for chronic pain states? PAIN 2016;157:1382–6.
[54]. Kosek E, Ordeberg G. Lack of pressure pain modulation by heterotopic noxious conditioning stimulation in patients with painful osteoarthritis before, but not following, surgical pain relief. PAIN 2000;88:69–78.
[55]. Kuhlmann L, Olesen SS, Grønlund D, Olesen AE, Phillips AE, Faghih M, Drewes AM. Patient and disease characteristics associate with sensory testing results in chronic pancreatitis. Clin J Pain 2019;35:786–93.
[56]. Kurien T, Arendt-Nielsen L, Petersen KK, Graven-Nielsen T, Scammell BE. Preoperative neuropathic pain-like symptoms and central pain mechanisms in knee osteoarthritis predicts poor outcome 6 months after total knee replacement surgery. J Pain 2018;19:1329–41.
[57]. Lannersten L, Kosek E. Dysfunction of endogenous pain inhibition during exercise with painful muscles in patients with shoulder myalgia and fibromyalgia. PAIN 2010;151:77–86.
[58]. Larsen DB. The Predictive Value of Quantitative Sensory Testing—A Systematic Review on Chronic Postoperative Pain and the Analgesic Effect of Pharmaceutical Therapies for Chronic Pain Patients. 2020:4. doi: 10.17605/OSF.IO/HSVYK.
[59]. Lee YC, Lu B, Edwards RR, Wasan AD, Nassikas NJ, Clauw DJ, Solomon DH, Karlson EW. The role of sleep problems in central pain processing in rheumatoid arthritis. Arthritis Rheum 2013;65:59–68.
[60]. Lockwood SM, Bannister K, Dickenson AH. An investigation into the noradrenergic and serotonergic contributions of diffuse noxious inhibitory controls in a monoiodoacetate model of osteoarthritis. J Neurophysiol 2019;121:96–104.
[61]. Lundblad H, Kreicbergs A, Jansson K-Å. Prediction of persistent pain after total knee replacement for osteoarthritis. J Bone Joint Surg 2008;90-B:166–71.
[62]. Maier C, Baron R, Tölle TR, Binder A, Birbaumer N, Birklein F, Gierthmühlen J, Flor H, Geber C, Huge V, Krumova EK, Landwehrmeyer GB, Magerl W, Maihöfner C, Richter H, Rolke R, Scherens A, Schwarz A, Sommer C, Tronnier V, Üeyler N, Valet M, Wasner G, Treede RD. Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): somatosensory abnormalities in 1236 patients with different neuropathic pain syndromes. PAIN 2010;150:439–50.
[63]. Mainka T, Malewicz NM, Baron R, Enax-Krumova EK, Treede R-D, Maier C. Presence of hyperalgesia predicts analgesic efficacy of topically applied capsaicin 8% in patients with peripheral neuropathic pain. Eur J Pain 2016;20:116–29.
[64]. Manafi Khanian B, Arendt-Nielsen L, Kjær Petersen K, Samani A, Graven-Nielsen T. Interface pressure behavior during painful cuff algometry. Pain Med 2016;17:915–23.
[65]. Martel MO, Petersen K, Cornelius M, Arendt-Nielsen L, Edwards R. Endogenous pain modulation profiles among individuals with chronic pain: relation to opioid use. J Pain 2019;20:462–71.
[66]. Martinez V, Fletcher D, Bouhassira D, Sessler DI, Chauvin M. The evolution of primary hyperalgesia in orthopedic surgery: quantitative sensory testing and clinical evaluation before and after total knee arthroplasty. Anesth Analg 2007;105:815–21.
[67]. Meeus M, Roussel NA, Truijen S, Nijs J. Reduced pressure pain thresholds in response to exercise in chronic fatigue syndrome but not in chronic low back pain: an experimental study. J Rehabil Med 2010;42:884–90.
[68]. Meints SM, Mawla I, Napadow V, Kong J, Gerber J, Chan S-T, Wasan AD, Kaptchuk TJ, McDonnell C, Carriere J, Rosen B, Gollub RL, Edwards RR. The relationship between catastrophizing and altered pain sensitivity in patients with chronic low-back pain. PAIN 2019;160:833–43.
[69]. Müller M, Limacher A, Agten CA, Treichel F, Heini P, Seidel U, Andersen OK, Arendt-Nielsen L, Jüni P, Curatolo M. Can quantitative sensory tests predict failed back surgery? Eur J Anaesthesiol 2019;36:695–704.
[70]. Naugle KM, Fillingim RB, Riley JL III. A meta-analytic review of the hypoalgesic effects of exercise. J Pain 2012;13:1139–50.
[71]. Noiseux NO, Callaghan JJ, Clark CR, Zimmerman MB, Sluka KA, Rakel BA. Preoperative predictors of pain following total knee arthroplasty. J Arthroplasty 2014;29:1383–7.
[72]. Olesen SS, Graversen C, Bouwense SAW, van Goor H, Wilder-Smith OHG, Drewes AM. Quantitative sensory testing predicts pregabalin efficacy in painful chronic pancreatitis. PLoS One 2013;8:e57963.
[73]. OʼLeary H, Smart KM, Moloney NA, Blake C, Doody CM. Pain sensitization associated with nonresponse after physiotherapy in people with knee osteoarthritis. PAIN 2018;159:1877–86.
[74]. Petersen KK, Arendt-Nielsen L, Simonsen O, Wilder-Smith O, Laursen MB. Presurgical assessment of temporal summation of pain predicts the development of chronic postoperative pain 12 months after total knee replacement. PAIN 2015;156:55–61.
[75]. Petersen KK, Olesen AE, Simonsen O, Arendt-Nielsen L. Mechanistic pain profiling as a tool to predict the efficacy of 3-week nonsteroidal anti-inflammatory drugs plus paracetamol in patients with painful knee osteoarthritis. PAIN 2019;160:486–92.
[76]. Petersen KK, Simonsen O, Laursen MB, Arendt-Nielsen L. The role of preoperative radiologic severity, sensory testing, and temporal summation on chronic postoperative pain following total knee arthroplasty. Clin J Pain 2018;34:193–7.
[77]. Petersen KK, Simonsen O, Olesen AE, Mørch CD, Arendt-Nielsen L. Pain inhibitory mechanisms and response to weak analgesics in patients with knee osteoarthritis. Eur J Pain 2019;23:1904–12.
[78]. Petersen KKKK, Graven-Nielsen T, Simonsen O, Laursen MBMB, Arendt-Nielsen L. Preoperative pain mechanisms assessed by cuff algometry are associated with chronic postoperative pain relief after total knee replacement. PAIN 2016;157:1400–6.
[79]. Pinto PR, McIntyre T, Araújo-Soares V, Almeida A, Costa P. Psychological factors predict an unfavorable pain trajectory after hysterectomy. PAIN 2018;159:956–67.
[80]. Pinto PR, McIntyre T, Araújo-Soares V, Costa P, Almeida A. Differential predictors of acute post-surgical pain intensity after abdominal hysterectomy and major joint arthroplasty. Ann Behav Med 2015;49:384–97.
[81]. Rice DA, Kluger MT, McNair PJ, Lewis GN, Somogyi AA, Borotkanics R, Barratt DT, Walker M. Persistent postoperative pain after total knee arthroplasty: a prospective cohort study of potential risk factors. Br J Anaesth 2018;121:804–12.
[82]. Rolke R, Baron R, Maier C, Tolle TR, Treede RD, Beyer A, Binder A, Birbaumer N, Birklein F, Botefur IC, Braune S, Flor H, Huge V, Klug R, Landwehrmeyer GB, Magerl W, Maihofner C, Rolko C, Schaub C, Scherens A, Sprenger T, Valet M, Wasserka B. Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): standardized protocol and reference values.[Erratum appears in Pain. 2006 Nov;125(1-2):197]. PAIN 2006;123:231–43.
[83]. Rolke R, Magerl W, Campbell KA, Schalber C, Caspari S, Birklein F, Treede RD. Quantitative sensory testing: a comprehensive protocol for clinical trials. Eur J Pain 2006;10:77–88.
[84]. Sangesland A, Støren C, Vaegter HB. Are preoperative experimental pain assessments correlated with clinical pain outcomes after surgery? A systematic review. Scand J Pain 2017;15:44–52.
[85]. Schliessbach J, Siegenthaler A, Bütikofer L, Vuilleumier P, Jüni P, Stamer U, Arendt-Nielsen L, Curatolo M. Predicting drug efficacy in chronic low back pain by quantitative sensory tests. Eur J Pain 2018;22:973–88.
[86]. Schnitzer TJ, Marks JA. A systematic review of the efficacy and general safety of antibodies to NGF in the treatment of OA of the hip or knee. Osteoarthr Cartil 2015;23:S8–S17.
[87]. Schug SA, Pogatzki-Zahn EM. Chronic pain after surgery or injury. Pain Clin Updat 2011;19:19.
[88]. Sivertsen B, Lallukka T, Petrie KJ, Steingrimsdottir OA, Stubhaug A, Nielsen CS. Sleep and pain sensitivity in adults. PAIN 2015;156:1433–9.
[89]. Sng BL, Ching YY, Han NLR, Ithnin FB, Sultana R, Assam PN, Sia ATH. Incidence and association factors for the development of chronic post-hysterectomy pain at 4- and 6-month follow-up: a prospective cohort study. J Pain Res 2018;11:629–36.
[90]. Szikszay TM, Adamczyk WM, Luedtke K. The magnitude of offset analgesia as a measure of endogenous pain modulation in healthy participants and patients with chronic pain. Clin J Pain 2019;35:189–204.
[91]. Vaegter HB, Graven-Nielsen T. Pain modulatory phenotypes differentiate subgroups with different clinical and experimental pain sensitivity. PAIN 2016;157:1480–8.
[92]. Vaegter HB, Handberg G, Emmeluth C, Graven-Nielsen T. Preoperative hypoalgesia after cold pressor test and aerobic exercise is associated with pain relief 6 months after total knee replacement. Clin J Pain 2017;33:475–84.
[93]. Vaegter HB, Handberg G, Graven-Nielsen T. Isometric exercises reduce temporal summation of pressure pain in humans. Eur J Pain 2015;19:973–83.
[94]. Vaegter HB, Handberg G, Graven-Nielsen T. Similarities between exercise-induced hypoalgesia and conditioned pain modulation in humans. PAIN 2014;155:158–67.
[95]. Vaegter HB, Handberg G, Jørgensen MN, Kinly A, Graven-Nielsen T. Aerobic exercise and cold pressor test induce hypoalgesia in active and inactive men and women. Pain Med 2015;16:923–33.
[96]. Vaegter HB, Lyng KD, Yttereng FW, Christensen MH, Sørensen MB, Graven-Nielsen T. Exercise-induced hypoalgesia after isometric wall squat exercise: a test-retest reliability study. Pain Med 2019;20:129–37.
[97]. Vaegter HB, Petersen KK, Mørch CD, Imai Y, Arendt-Nielsen L. Assessment of CPM reliability: quantification of the within-subject reliability of 10 different protocols. Scand J Pain 2018;18:729–37.
[98]. Valencia C, Fillingim RB, Bishop M, Wu SS, Wright TW, Moser M, Farmer K, George SZ. Investigation of central pain processing in postoperative shoulder pain and disability. Clin J Pain 2014;30:775–86.
[99]. Vollert J, Attal N, Baron R, Freynhagen R, Haanpää M, Hansson P, Jensen TS, Rice ASC, Segerdahl M, Serra J, Sindrup SH, Tölle TR, Treede R-D, Maier C. Quantitative sensory testing using DFNS protocol in Europe: an evaluation of heterogeneity across multiple centers in patients with peripheral neuropathic pain and healthy subjects. PAIN 2016;157:750–8.
[100]. Wasner G, Naleschinski D, Binder A, Schattschneider J, McLachlan EM, Baron R. The effect of menthol on cold allodynia in patients with neuropathic pain. Pain Med 2008;9:354–8.
[101]. Werner MU, Mjöbo HN, Nielsen PR, Rudin Å. Prediction of postoperative pain. Anesthesiology 2010;112:1494–502.
[102]. Wilder-Smith OH. A paradigm-shift in pain medicine. Aalborg: Center for Sensory-Motor Interaction (SMI), Department of Health Science and Technology, Aalborg University, 2013.
[103]. Wilder-Smith OH, Schreyer T, Scheffer GJ, Arendt-Nielsen L. Patients with chronic pain after abdominal surgery show less preoperative endogenous pain inhibition and more postoperative hyperalgesia: a pilot study. J Pain Palliat Care Pharmacother 2010;24:119–28.
[104]. Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science 2000;288:1765–9.
[105]. Wylde V, Palmer S, Learmonth ID, Dieppe P. The association between pre-operative pain sensitisation and chronic pain after knee replacement: an exploratory study. Osteoarthr Cartil 2013;21:1253–6.
[106]. Wylde V, Sayers A, Lenguerrand E, Gooberman-Hill R, Pyke M, Beswick AD, Dieppe P, Blom AW. Preoperative widespread pain sensitization and chronic pain after hip and knee replacement. PAIN 2015;156:47–54.
[107]. Yarnitsky D. Conditioned pain modulation (the diffuse noxious inhibitory control-like effect): its relevance for acute and chronic pain states. Curr Opin Anaesthesiol 2010;23:611–15.
[108]. Yarnitsky D, Bouhassira D, Drewes AM, Fillingim RB, Granot M, Hansson P, Landau R, Marchand S, Matre D, Nilsen KB, Stubhaug A, Treede RD, Wilder-Smith OHG. Recommendations on practice of conditioned pain modulation (CPM) testing. Eur J Pain 2015;19:805–6.
[109]. Yarnitsky D, Crispel Y, Eisenberg E, Granovsky Y, Ben-Nun A, Sprecher E, Best LA, Granot M. Prediction of chronic post-operative pain: pre-operative DNIC testing identifies patients at risk. PAIN 2008;138:22–8.
[110]. Yarnitsky D, Granot M, Nahman-Averbuch H, Khamaisi M, Granovsky Y. Conditioned pain modulation predicts duloxetine efficacy in painful diabetic neuropathy. PAIN 2012;153:1193–8.
Keywords:

Quantitative sensory testing; Chronic pain; Chronic postoperative pain; Analgesic effect

© 2020 International Association for the Study of Pain