Contact burn injury methodology.

Heating paradigm: The most common heating paradigm was 420 s at 47°C (40/64 [63%] studies; Fig 2) [1,2,5,6,10,12–16,18,20,21,27,29,32–34,36–39,41–44,46,48–50,53,55–62,64]. One study applied a longer exposure time in a control group than in the intervention group (420 s vs. 360 s both at 50°C) [7], while another study applied a higher temperature during a preliminary trial compared to the intervention trial (300 s at 49°C vs. 300 s at 47°C) [11].

Contact thermode surface area: The most commonly applied active thermode surface area were 12.5 cm² (41/64 [64%] studies) [1,5,6,10,12,13,16,18–21,29,32–34,36–39,41–44,46–62,64], 3.75 cm² (10/64 [16%] studies) [7,22–26,30,31,35,40], and 9 cm² (5/64 [8%] studies) [2,3,14,15,27]. The remaining 8/64 (13%) studies used active areas ranging from 0.3–10.2 cm² [4,8,9,11,17,28,45,63].

Induction site of CBI: Induction sites were the calf (46/64 [72%] studies) [1,2,5–7,10,12,13,16,20–26,29–44,46,47,49,50,52–60,64], volar forearm (12/64 [19%] studies) [3,4,8,9,11,17,19,28,48,61–63], thigh (4/64 [6%] studies) [14,15,18,27], abdomen (1/64 [2%] studies) [51], or dorsum of the hand (1/64 [2%] studies) [45].

Contact thermode application pressure: Applied contact thermode pressures were reported in 34/64 (53%) studies [4–7,10–13,16,23,28–34,39,40,45–47,51–55,57–63], with a median (range) of 5 (1.3 to 26) kPa.

Pain ratings: In 41/64 (64%) studies, pain ratings were reported during CBI-induction. Pain intensity assessments were: in 36 studies with the visual analog scale (VAS) [1,3,6,10–16,18,19,21,27,29,32–34,36–39,41–44,50,51,56–62,64], 15 of which were an electronic visual analog scale [6,11–15,18,19,21,27,39,50,51,61,62]; in three studies a numeric rating scale [2,8,28]; one study a verbal rank scale [31]; and in one study a ‘magnitude estimation’ method [17].

Post-burn changes.

This section describes the changes induced by the CBI as a conditioning stimulus.

Primary hyperalgesia area, thermal stimuli: Warmth detection thresholds were generally not altered [1,5,41,49,54,55,58–60], but in five studies [16,32,33,42,60] increased post-CBI. In these studies, the duration of the change was less than 1 h [16,33,60]. Only one study observed a significant numerical increase in cool detection threshold following the CBI [33], while ten studies observed no post-CBI changes in cool detection threshold [16,41–43,49,54,58–60,64]. A total of 37 studies reported primary hyperalgesia to heat stimulation after the CBI, assessed as reduced heat pain threshold (HPT) or increased suprathreshold heat pain response [1,5–13,16,17,20,21,23–26,28,32–35,37,41,42,49–60,64], while two studies did not report any sensory changes [4,43]. Primary hyperalgesia lasted between 24–48 h post-CBI in one study (300 s at 49°C, 3.75 cm²) [23].

Primary hyperalgesia area, mechanical stimuli: Three studies investigated tactile detection thresholds and found no post-CBI sensory changes [53–55]. Post-CBI primary hyperalgesia to punctate mechanical stimuli was a consistent finding [1,4,10,11,16,23,24,26,29,32–35,41,42,47–49,53–60,62,64].

Secondary hyperalgesia, thermal/electrical stimuli: Secondary hyperalgesia to heat was observed in two studies [34,62], but this finding was not corroborated by four other studies [7,49,54,55]. Further, Aδ-fibers, but not C-fibers, were found to be sensitized to short radiant heat stimuli in the primary hyperalgesia area 24 h post-CBI, while no sensitization was present in the SHA [50]. Hyperalgesia to electrical stimulations was induced in the SHA in one study [62], but was not corroborated by another [29].

Secondary hyperalgesia, mechanical stimuli: A total of 15 studies [1,4,10,11,16,26,32–34,37,38,41–43,52,54,64] assessed SHAs to punctate mechanical stimulation during baseline conditions while 28 studies [5–7,12,13,20,21,23,24,28,29,35,36,44,46–51,53,55–60,62] assessed post-CBI areas without baseline assessments. In all studies the CBI induced robust SHA to punctate mechanical stimuli, with few studies reporting sustained SHA at 24 h (300 s at 49°C, 3.75 cm²) [23,26]. Areas of allodynia, during or after the CBI, were assessed in 25 studies [4,6,7,10,12,13,16,19–21,23,28,32–34,36,37,39,49,54,55,58,60–62]. Allodynia, tested by brush, developed shortly after the start of the CBI induction, but was generally more evanescent than SHA to punctate mechanical stimuli.

Temporal summation: Studies evaluated the following stimulation modalities: electrical [29,62], thermal [62], punctate mechanical [47,48,53–55,62], and mechanical impact [62]. Temporal summation was generally demonstrated post-CBI. Further, temporal summation to thermal stimuli did not increase post-CBI [62].

Objective inflammatory variables: Erythema was measured in eleven studies and was generally long-lasting (up to 48 h) [1,23,24,26,33,35,36,41,58,59,64]. Blood flow was increased in the primary CBI-area [4,46]. Skin temperatures were increased in four studies [46,54,55,59], although, unaltered in one study [5]. Skin temperatures returned to baseline values 60 min post-CBI in the primary hyperalgesia area, while temperatures concomitantly were increased outside the CBI-area and then returned to baseline after 30 min [46]. An area of flare was reported in six studies [4,7,22,28,46,62]. Flare was not present 160 min post-CBI [7]. Dermal thickness, measured by a high-resolution ultrasound scanner, was found to increase post-CBI for up to 360 min [1,64].

Adverse effects: In 19/64 (30%) studies, blistering was reported [7,22–26,29–36,39,42,46,49,60]. In 21/64 (33%) studies the presence of blistering was examined and did not occur [5,6,9–13,16–18,38,41,47,48,50–55,62], while in 23/64 (36%) studies blistering was not examined, nor reported [1–4,8,14,15,19–21,27,28,37,40,43–45,56–59,61,63,64]. Three studies reported slight localized skin color changes in 25% of subjects persisting for three weeks [32–34], while hyperpigmentation occurred in one subject 23 days post-CBI [38].

Intervention studies.

The 37 studies [1,5–7,10–13,16,17,20–22,24–26,30–32,35,36,39,41,43,47–49,51–55,58–60,63,64] evaluated ten drug groups (antiarrhythmic agents [49], gabapentinoids [60], glucocorticoids [35,51,58], glutamate receptor antagonists [10], local anesthetics [7,11,22,30], melatonin [1], N-methyl-D-aspartate (NMDA) receptor antagonists [12,13,20,21,32,47,53–55], non-steroidal anti-inflammatory drugs (NSAIDs) [17,24,26,39,51,52], opioids [5,16,25,43,47,48,54,55,63], and opioid antagonists [6]), along with hyperbaric oxygen [41,64], local cooling [59], and nerve blocks [31,36].

The total number of per-protocol subjects was 641, with a median of 17 (12 to 22) subjects per study. All but one study mentioned the ratio of males to females [17], and another study only described the intention-to-treat group’s male to female ratio [36]. For the remaining studies, a gender ratio (males/females) of 6.4 (532/83) was present. A total of eleven studies reported the bodyweight of subjects [1,5,22,24–26,35,41,43,51,64], while four of these reported BMI [1,41,43,64]. Nine studies included an a priori sample size estimate [1,41,43,47,51,58–60,64]. Four studies [6,22,26,52] performed post hoc power-analyses, while an additional four studies estimated post hoc minimal detectable differences for all outcomes [30,32,36,60]. After an incorrect a priori sample size estimate, one study presented a post hoc power analysis [1].

All studies were described as randomized. A total of 29/37 studies were double-blind [1,6,10–13,16,17,20,21,24–26,30,32,35,39,43,47–49,51–55,58,60,63], and all were, along with two single-blinded studies [36,64], placebo-controlled. Two of the placebo-controlled studies used both active and inactive placebos [48,53]. The non-pharmacodynamic studies used control-conditions (i.e., ambient atmospheric pressure [41,64] and ambient temperature of contact thermode [59]).

Among the 34 studies involving drug administration, seven studies evaluated the effect of several drugs [22,43,47,48,51,54,55], and three studies administered naloxone along with other drugs to examine any opioid-mediated effects [16,20,63]. Multiple drug dosing was evaluated in seven studies [1,6,12,13,21,43,48].

Methodological quality was assessed by the Oxford Quality Scoring System [75]. The median (IQR) score for the 37 intervention studies was 2 (2 to 3). Fifteen studies had a high-quality score ≥ 3 (Table 3) [1,6,11,17,20,21,41,43,48,51,52,54,55,58,64].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Intervention study results.

https://doi.org/10.1371/journal.pone.0254790.t003

Methodological studies. The seven studies [4,23,27,33,44,61,62] focused on characterizing the CBI as a model of tonic heat pain [27]; validating reproducibility [33]; evaluating the vascular and sensory effects of heat conditioning [61]; determining time-courses of hyperalgesia [23]; characterizing SHA assessments [44], and comparing the CBI-model to other models of evoked hyperalgesia [4,62].

The total number of per-protocol subjects was 135, with a median of 12 (11 to 18) subjects per study. The gender ratio (males/females) was 1.7 (85/50). Only one study included data on subjects’ height, weight, or BMI [44]. Two studies reported a priori sample size estimates [33,44].

Physiological studies. The fourteen studies [2,8,9,14,15,28,29,34,37,38,40,45,46,50] focused on comparing SHAs to areas of flare [46], the adrenergic system [8,9], the endogenous opioid system [37,38,45], heritability of pain responses [28], imaging [2,14,15], latent sensitization [37,38], local inflammatory mediators [40], nociceptive fibers [50], secondary hyperalgesia to heat [34], and temporal summation [29].

The total number of per-protocol subjects was 482, with a median of 21.5 (20 to 24) subjects per study. One study did not report the gender of subjects [8], but for the remaining studies, a gender ratio (males/females) of 0.6 (167/290) was obtained. Three studies provided height and weight of subjects [2,37,38], and one also reported BMI [37]. Four studies reported a priori sample size estimates [2,28,37,38], and one of these included a post hoc sample size estimate [37]. One study calculated post hoc minimal detectable differences for all outcomes [29].

Predictive studies. Three of the four predictive studies focused on patients undergoing knee surgery [18,56,57]: Two of these studies investigated the predictive potential of the CBI model [18,56] while the third investigated how surgery modulates hyperalgesic responses [57]. The fourth study evaluated the predictability of pain during a CBI, based on QST assessments [42].

The three knee surgery studies involved patients [18,56,57], while the fourth study included healthy subjects [42]. Accounting for duplicate subjects, the total number of per-protocol subjects was 232. The ratio between male and female subjects was 1.1 (124/108). One study provided height, weight, and BMI of subjects [42]. Two of the four studies reported a priori sample size estimates [18,57].

Intervention studies.

This subsection focuses on the efficacy of interventions in attenuating hyperalgesia, pain, and other inflammatory responses following a CBI. An overview of the intervention studies is presented in Table 3. In general, no intervention showed clear efficacy in attenuating CBI-induced hyperalgesia.

NMDA receptor antagonists: Ketamine reduced the SHA in four studies [13,53–55], while two studies found transient to no effects on hyperalgesia [21,32]. The effects of ketamine on SHAs were observed regardless of whether it was administered preemptively [13,53,55] or post-CBI [54]. Furthermore, ketamine reduced the SHA regardless of a preceding naloxone infusion, indicating that the effect of ketamine was not mediated by the endogenous opioid system [20]. Ketamine was claimed to act synergistically with morphine in reducing temporal summation [47]. Dextromethorphan, another NMDA receptor antagonist, marginally reduced the SHA [12].

Opioids: A preemptive infusion [55], but not a post-CBI infusion [47,48,54], of morphine, attenuated SHAs, and temporal summation. Epidural morphine injection reduced the SHA regardless of administration timing [5]. No difference was found between systemic buprenorphine or morphine administration on hyperalgesia, nor between the effect on high- and low pain-sensitizer subjects [43].

Glucocorticoids and NSAIDs: One comparative study found an effect of methylprednisolone and ketorolac, respectively, on reducing the SHA, with no statistical difference between the two drugs [51]. No other effects of glucocorticoids on hyperalgesia and inflammation were reported [35,58]. The effects of NSAIDs on heat pain perception were weak and observed exclusively within the range of 46°C to 49°C, but interestingly not at higher temperatures (50°C to 51°C) [17]. Pain intensity evoked by motor brush stimulation in the SHA was lower following ibuprofen compared to placebo, while the magnitude of SHA was unaltered [39]. No effect of topical NSAID on hyperalgesia and inflammatory variables has been observed [24,26,52].

Antiarrhythmics: Adenosine infusion reduced the SHA without reducing primary hyperalgesia to mechanical or thermal stimuli [49].

Nerve blocks: A saphenous nerve block was found to reduce primary hyperalgesia and SHAs [31], while a sympathetic nerve block did not alter pain or hyperalgesia [36]. These studies were not double-blinded.

Local anesthetics: Two studies evaluated the effect of administering intravenous [11] and subcutaneous [7] lidocaine pre- vs. post-CBI and found no clear superiority of one over the other [7,11]. EMLA-cream [22,30] and bupivacaine [22] had no clear effect on thermal hyperalgesia [30] nor any inflammatory variables [22,30].

Miscellaneous: Gabapentin increased mechanical pain thresholds but did not affect mechanical pain intensity nor other sensory variables [60]. Two studies evaluated the effects of hyperbaric oxygen treatment on secondary hyperalgesia immediately post-CBI using almost identical experimental designs [41,64]. Both studies found that hyperbaric oxygen treatment reduced the SHA when compared to ambient pressure [41,64]. Interestingly, the sequences starting with hyperbaric oxygen, conditioned the response in the subsequent ambient pressure test day (serving as a control), generating a similar reduction in SHAs. Riluzole (glutamate receptor antagonist) [10], melatonin [1], μ-opioid receptor antagonists [6], and local cooling of the CBI [59] had no clear effect on mechanical or thermal hyperalgesia [1,6,10,59,60], nor on other inflammatory variables [1,59].

Methodological studies. Repeatability: Within-day comparisons using the CBI-model (420 s at 47°C, 12.5 cm²) revealed that a 20% difference could be detected for all variables in a crossover design with twelve subjects (α = 0.05, β = 0.20), except for cool detection threshold and heat pain response at 43°C (see S1 Table for statistical definitions) [33]. On each day, the CBI produced robust hyperalgesia throughout the 6 h study period. Between-day comparisons revealed that the intra-individual, between-days coefficients of variation ranged from 9 to 36% for all sensory assessments along with erythema, except for the heat pain response to 43°C, and the brush-evoked allodynia, both showing a low reproducibility. These two variables were recommended to omit in between-day comparison studies [33]. A study using a comparable CBI-model (420 s at 47°C, 9 cm²) observed a correlation between test-retest of pain intensity scores following CBIs (r² = 0.44; P < 0.05) [27]. A study categorized as an intervention study, tested reliability (intraclass correlation coefficient [ICC]), and variability of areas of flare, secondary hyperalgesia, and allodynia using a relatively milder CBI-model (330 s at 45°C, 0.3 cm²) [28]. The authors concluded that the model was reliable in an investigator independent manner, although primary hyperalgesia outcomes were not addressed. The testing was performed 15 min post-CBI, consequently only addressing the reproducibility of the hyperalgesic response at a limited time span after the CBI.

Validation: The induction of a CBI produced long-lasting hyperalgesia, with primary hyperalgesia lasting between 24–48 h and secondary hyperalgesia remaining until 24 h post-CBI (300 s at 49°C, 3.75 cm²) [23]. One test-retest study evaluated the performance of different punctate stimulators for the assessment of SHAs [44]. The delineation of SHAs with nylon filaments and weighted-pin instruments were highly correlated with the application pressure. The weighted-pin instrument applying a high pressure (10,424 kPa) showed the greatest inter-observer reliability following a CBI [44].

A comparison of the CBI-model to other cutaneous heat pain models can be found in S3 File.

Physiological studies. A functional magnetic resonance imaging (fMRI) study found that high-sensitizers, i.e., subjects with large SHAs, had less activation of the default mode network (precuneus, posterior cingulate cortex) during noxious punctate mechanical stimulation post-CBI compared to low-sensitizers, i.e., subjects with smaller SHAs. Further, an inverse relationship between the magnitude of SHA and the volume of the caudate nucleus was found [2]. A twin-study including 51 monozygotic and 47 dizygotic twin pairs found that the SHA, area of brush-evoked allodynia, HPT, and pain during the CBI all had statistically significant heritable components. However, neither shared genetics nor environmental factors could explain the extent of CBI-induced heat hypersensitivity (ΔHPT [pre- vs. post-CBI]) [28].

Predictive studies. In patients undergoing elective arthroscopic anterior cruciate ligament repair, pain during a preoperative CBI significantly correlated with self-reported dynamic pain on postoperative days 0 to 2 (r = 0.65; P < 0.01) and days 3 to 10 (r = 0.57; P < 0.01) and resting pain ratings on postoperative days 0 to 2 (r = 0.60; P < 0.01) and days 3 to 10 (r = 0.59; P < 0.01) [56]. Other QST-variables such as thermal thresholds, mechanical thresholds, and SHA displayed a relatively weaker predictive value. A later high-powered study investigated the ability of a phasic (5 s) vis-á-vis a tonic (420 s) heat stimulus to predict postoperative pain, using identical stimulus areas (12.5 cm²) and temperatures (47°C) after total knee arthroplasty [18]. The authors found a significant, but very weak, correlation between postoperative VAS during walking on days 1 to 7, and, the phasic (r = 0.25; P = 0.02) and tonic (r = 0.27; P = 0.01) heating paradigms, respectively, deeming these variables clinically irrelevant predictors. Further, arthroscopic knee surgery did not alter postoperative hyperalgesic responses to CBIs when compared to the baseline preoperative responses [57].

Paradigms and thermode handling.

The included studies reveal substantial variations in the applied heating paradigms (Table 2). Although the relationship between exposure time and temperature of a burn injury is established (Fig 2) [72], the time-course of hyperalgesia has only been investigated in a few studies, and thus different heating paradigms have not been compared. The heating paradigm of 420 s at 47°C is, however, the most commonly applied [1,2,5,6,10,12–16,18,20,21,27,29,32–34,36–39,41–44,46,48–50,53,55–62]. Reproducibility has been investigated for this heating paradigm [33], hyperalgesia is long-lasting [33] and adverse events are minor and infrequent (reported in 10/39 studies) [29,32–34,36,39,42,46,49,60].

None of the included studies have investigated the effects of changing the active thermode areas. The SHA has been observed to increase proportionally with the CBI-area (preliminary observations by the authors). Interestingly, a study attempting to replicate the heat/capsaicin model with a smaller thermode than previously reported (9 cm² vs. 12.5–15.7 cm²) [70,73] found that the smaller thermode was not able to produce stable SHAs, and, further, could not show an effect of gabapentin, as previously reported [91]. Similarly, another study could not find an effect of pre- and post-conditioning on heat sensory outcomes in the heat/capsaicin model [61]. These authors also used a smaller thermode during rekindling (3.75 cm²) compared to previous studies [70,73]. The findings indicate that changes in the contact thermode area probably may affect the degree of hyperalgesia and the pharmacological sensitivity of the CBI-method.

Contact thermode application pressures were reported in 34/64 of the included studies, utilizing a wide range of pressures. Interestingly, one study investigated the effect on thermal thresholds of varying application pressures (0.32–10 N) using an active thermode area of 9 cm² [92]. The authors concluded that altering the application pressure did neither affect the thermal thresholds nor affect the intra- or inter-subject reproducibility. However, these data only comprised phasic stimuli, not the tonic CBI-stimulus.

Induction site.

Differences in sensitivity may occur between different skin sites as normal skin on the volar forearm was significantly more sensitive than the medial calf based on thermal detection thresholds [33] and mechanical pain thresholds [93], although conflicting results were found for heat pain thresholds [33,93].

Repeated CBIs on an ipsilateral, homologous skin site have been observed to induce habituation effects across sessions, e.g., decreased SHAs and increased heat pain thresholds in the second session [33,38,42,64]. This problem may be mitigated by using block-randomization and cross-over designs, thereby evenly dividing active drugs and placebo between sessions.

Pathophysiological changes.

As previously reported, Moritz and Henriques described first-degree burn injuries ranging from transient hyperemia to prolonged erythema with the formation of small vesicles [72]. Erythema reflects vascular changes in the burn area, and the degree of redness is measured by skin reflectance spectrophotometry [94]. The duration of erythema has varied substantially across CBI-studies. Other indices of inflammation, i.e., increased dermal blood flow, a rise in skin temperature, and flare, are only short-lasting phenomena post-CBI [46,61,95–97]. Increased dermal thickness as an index of skin edema, measured by high-frequency ultrasound technique, has only been applied in a limited number of studies, but seem to persist several hours post-CBI [1,64]. Interestingly, none of the included studies reported any significant effect on the classic inflammatory variables (edema, erythema, or temperature increase) by the drugs tested.

Assessing the central anti-hyperalgesic effect, by evaluating changes in SHA, is an essential outcome of many analgesic studies. The SHA is delineated via punctate stimulation with a nylon filament or a weighted-pin instrument. The CBI induces rather consistent secondary hyperalgesia, with a duration of 24 h. Interestingly, two studies have indicated that 12/100 (420 s at 47°C, 12.5 cm²) [42], and 10/64 (300 s at 46°C, 12.5 cm²) [19] of the subjects do not develop measurable SHAs. Paradoxically, QST assessments by themselves may induce secondary hyperalgesia [33,98].

Temporal summation, another measure of central sensitization, is generally facilitated in the CBI-model [47,48,53–55,62]. The temporal summation response appears modality-dependent due to the stimulation of different receptor subtypes [29,62]. Temporal summation has primarily been investigated in studies using ketamine and morphine, with the former usually providing a significant mitigating effect. However, other analgesics, e.g., dextromethorphan [99], imipramine [100], venlafaxine [101], and gabapentin [102], have been found to mitigate temporal summation, and could thus also be applied in the CBI-model.

General issues.

Only a few studies have investigated test-retest reproducibility of the CBI-model variables, and when tested, the applied statistical methods have been inconsistent [27,28,33,44]. When calculating ICCs as a measurement of reliability, the between-subject variance and within-subject variance should preferably be reported, since the former may increase as a result of measurement heterogeneity on a group level, thereby artificially increasing ICCs [103]. However, the included studies did not provide this information [28,44]. An agreement test, presented as a Bland-Altman plot, has been recommended as a valid test-retest statistic mainly because of its detailed, transparent, and honest presentation of data distribution [103]. However, only one of the included studies presented this statistic, and the authors mainly focused on the inter-observer agreement [44]. One study claimed that there was a good test-retest reproducibility of pain scores during a CBI [27], but this was only based on a correlation coefficient between test-retest scores which is a suboptimal measure of reproducibility [103–105]. A review providing a post hoc analysis of test-retest data from a CBI-study found that SHAs were reproducible across multiple sessions. These calculations were based on ICCs, suggesting that subjects could be phenotyped according to the pattern of sensitization [42,90].

When assessing quantitative sensory thresholds, the subject’s reaction time, depending on nerve conduction velocity and cognitive-executive abilities, may influence the results. However, only a few of the included studies reported reaction time tests [1,5,43]. Sensory tests applying the ‘method of limits’ are especially subjected to the influence of reaction time. The effect of analgesic drugs with potential sedative properties may increase response latency, erroneously leading to higher pain thresholds (‘pseudoanalgesia’) [43]. This is an underreported source of systematic bias.

The heating area as an important variable.

While the exposure time and the contact thermode temperature are well-known critical variables in the thermal energy transfer, the size of the heating area has not been systematically examined. The active thermode area governs the intensity of pain during CBI-induction and the primary hyperalgesia area, but also likely influences the magnitude of the SHA, the duration of secondary hyperalgesia, and sensory perception within the hyperalgesia areas. Hypothetically, these spatial characteristics may affect the pharmacodynamic effects of relevant analgesics.

The contact burn injury as an inflammatory model.

The CBI is a surrogate, experimental model of inflammatory pain, and therefore demonstrates certain limitations compared to clinical pain. Although the model involves the cardinal signs of inflammation, i.e., edema, erythema, local hyperthermia, and evoked pain, the pharmacological sensitivity to anti-inflammatory analgesics, NSAIDs, and glucocorticoids, is ambiguous. The model lacks the spontaneous pain component usually present in acute clinical inflammatory states. A hypothetical explanation is that the area of the skin injury, corresponding to 0.06% of the body surface area, only activates a limited part of the mesencephalic-subcortical pain network. The limited nociceptive drive likely affects the response to relevant analgesics. Obviously, augmenting the extent of thermal injury by increasing the exposure time and/or the contact thermode area/temperature would lead to a second-degree CBI at the undesirable cost of irreversible skin damages. However, the inflammatory response could instead be augmented synergistically by topical administration of pro-inflammatory agents such as mustard oil [116], sodium lauryl sulfate [117], or the well-investigated capsaicin [73], enhancing the primary and secondary hyperalgesia components.

Characterization of phenotypes.

As previously mentioned, phenotyping based on the SHA is possible across individuals [90]. These phenotypes are characterized by differences in brain structure and function [2], with ‘high-sensitizers’ presenting similar brain activation patterns to chronic pain patients [2]. Phenotyping based on a subject’s ‘sensitization’ pattern assessed by QST provides a potential for predicting the pharmacodynamic response to analgesics [118–120].