A modified Hargreaves' method for assessing threshold temperatures for heat nociception

Abstract

This study describes a modified Hargreaves' method for assessing paw withdrawal threshold temperatures for heat (PWT-H) nociception in the hind paws of rats. This method utilises radiant heat to maintain controlled lamp temperatures (CLTs) on a glass floor beneath the rat hind paw. An ascending series of CLTs were applied for 10s, with 5-10min intervals between applications, until characteristic withdrawal behaviour was observed or a cutoff CLT was reached. The average plantar epicutaneous temperatures measured from anaesthetised rats corresponding to CLTs and withdrawal latencies were used for determining PWT-H. The mean PWT-H in 2-month-old (mo) naïve Sprague-Dawley rats (n=38) was 47.6±0.2°C, which is comparable to the noxious threshold temperature for human glabrous skin (46.5±0.5°C). The PWT-H was consistent between trials and daily assessments over four consecutive days. No significant differences were observed between the PWT-H in 2-, 6- to 8-, and >24-mo F344 rats, but the PWT-H in 1-mo rats was significantly reduced. Three hours following plantar incision, the PWT-H decreased to 37.5±0.2°C, consistent with previous observations of C-fibre afferents from incised glabrous skin firing at 36.7±3.6°C. Parallel testing, using the current method and an electronic von Frey device, illustrated similar degrees of incision-induced hyperalgesia, gradual improvements in hyperalgesia, and reversals induced through morphine and gabapentin. In conclusion, the present method facilitates a comparison of PWT-H using electrophysiological and human psychophysical studies involving thermosensation, and as a behavioural assay identical to von Frey testing, this method also measures the threshold for nociception.

Keywords: Gabapentin; Morphine; Nociception; Radiant heat; Rat; Threshold temperature.

Figure 1

A closed feedback-controlled system maintains a fixed lamp temperature on the underside of a glass floor beneath the hind paw of an unrestrained rat. The parameters of temperature and stimulus duration are entered using a keypad and display. Once triggered, the proportional-integral-derivative controller unit of the stimulator signals the lamp to generate radiant heat onto a thermocouple sensor. The sensor provides real-time temperature input for the controller unit, which in turn, continuously modulates the current to the lamp to correct any error between the measured and user-defined temperatures.

Figure 2

Schematic diagram of the experimental set-up used for assessing paw withdrawal threshold temperatures for heat nociception (PWT-H) in unrestrained rats. The set up consists of a stimulator unit, thermocouple sensor connected to a transmitter unit for temperature feedback, and a 150W halogen projector lamp inside an aluminum cup.

Figure 3

Controlled lamp temperature (CLT) on the underside of glass floor produces reproducible epicutaneous thermocouple temperatures (ETT) on plantar hind paw of anesthetized rats. Representative digital oscilloscope traces (a) illustrate CLT (lower panel) and ETT (upper panel). The stimulator was set to 40.0°C for the first stimulation and increased in 2.5°C steps to a cut-off temperature of 70.0°C. Each temperature was applied 4 times per animal and 24 traces were obtained from 6 animals. Second-by-second ETT values were averaged from the 24 traces, and are shown in b. Each data point represents the average ETT ± SEM.

Figure 4

Characteristics of heat stimulus used for measuring paw withdrawal threshold for heat nociception (PWT-H). a) Correlation of the rates of changes in ETT and CLT in each second of 10-s heat applications (spearman r = 0.99, P<0.0001). (b) Latency to stable (peak±0.2°C) ETT (elapsed time from onset to peak temperature) varies with the set temperature. (c) The rate of ETT change is different between applications, but is consistently highest over the first 2-s of 10-s heat application. (d) One hundred thirty temperature values were obtained from 10-s stimulations at 13 set temperatures (13 × 10, 1-s bin). The PWT-H data ranges from 35.0°C to 54.9°C, with an average step size of 0.15°C.

Figure 5

Characteristics of PWT-H in normal rats. (a) Three trial testing sessions of paw withdrawal thresholds for heat (PWT-H) and mechanical nociception from pre-incision testing (n=18) (P>0.05, ANOVA followed by Newman-Keuls multiple comparison test) are shown. PWT-H measurements were consistent between multiple trials within the same testing session (a) or daily testing across four consecutive days (b) (n=8, P>0.05, ANOVA followed by Newman-Keuls test). PWT-H was observed in the static but not dynamic phase of ETT change (c, n=38 also see Fig. 4c). (d) Histogram of PWT-H distribution within naïve Sprague-Dawley rats (n=38). These data are normally distributed (P>0.1, Kolmogorov-Smirnov test). (e) PWT-H from F-344 rats of different age groups (*P<0.05, *** P<0.001, ANOVA followed by Newman-Keuls test). mo, months old.

Figure 6

PWT-H measure is comparable with PWT-M in detection of incision-induced hyperalgesia, and its recovery over time. Baseline PWT data used in Figure 5a. (a) Time courses of the mean PWT-H and PWT-M evaluated at baseline, 3-h, 1, 2, 3, 4, 7, and 9-d post-incision. PWT-H: sham vs. incision, F_1,16 =72.0, P<0.0001 at 3-h, 1, 2, 3, 4, and 7-d; PWT-M: sham vs. incision, F_1,16 =73.5, P<0.0001 at 3-h, 1, 2, 3, 4, and 7-d (2-way ANOVA followed by Bonferroni’s post-hoc tests). (b) Correlation between PWT-M and PWT-H throughout post-incision recovery period. The degree of hyperalgesia increases from left to right along the regression line. (Pearson r = 0.87; P<0.0001).

Figure 7

PWT-H and PWT-M measures are identical in illustrating effects of analgesic compounds on recovery of incision-induced hyperalgesia. Two hours after incision, and on the 1^st, 2^nd, 3^rd, and 4^th postoperative days, rats were tested for baseline, and 30 min following blinded administrations of saline or morphine 1 or 3-mg/kg (a, PWT-H; F_2,19=26.26 and PWT-M; F_2,19=15.99, 1-way ANOVA followed by Dunnett’s test), gabapentin (b, 50 mg/kg i.p.; *P<0.05, ***P<0.001, unpaired t-test; n=8), meloxicam (c, 10 mg/kg; i.p.; n=8), ibuprofen (d, 20 mg/kg i.p.; n=6), or diclofenac (d, 12 mg/kg; i.p.; n=6), on each respective day. Data are presented as mean±SEM of the percentage of maximum possible effect (%-MPE). See materials and methods for calculation of %-MPE values.