Effect of prolonged experimental neck pain on exercise-induced hypoalgesia

1 . Introduction

Neck pain is one of the most considerable musculoskeletal problems,18,40 with many developing pain lasting for >3 months.8,9 Despite a multitude of possible treatment strategies,23,24 even comprehensive interventions seem to be no more effective than advice.22,35 Although exercise training (multiple sessions), including strength and endurance exercises, has shown promise as treatment modalities,23 the influence on central pain mechanisms remains unclear.22,35

Increased local and widespread pain sensitivity is often observed in patients with neck pain compared with asymptomatic individuals.12,48,50,64 Hyperalgesia, especially in distant areas away from the painful site, is suggested to be a sign of altered central pain processing.4,64 This difference in central nociceptive processing between those with pain and asymptomatic subjects can also be observed in response to exercise.39 In healthy participants, exercise (single session) including resistance and aerobic exercises, lasting approximately 3 to 40 minutes, has shown decreased pain sensitivity, also known as exercise-induced hypoalgesia (EIH).30,39 The underlying mechanism of EIH remains unclear, but it has been suggested that exercise engages the endogenous pain modulatory system to decrease pain sensitivity, although this mechanism may not be working efficiently in patients with persistent pain.39,55 Interestingly, other studies have found increased pain sensitivity after exercises consisting of 3 to 6 series of repeated arm movements or submaximal cycling for approximately 2 to 5 minutes in populations with persistent neck pain,12,59 which is contrasted by studies showing positive effects on pain sensitivity after exercise training 3 to 5 times per week consisting of resistance exercise with/without 30 minutes of aerobic exercises over 2 to 6 months.3,28 Understanding this relationship between exercise and altered pain sensitivity is crucial to optimize exercise rehabilitation of neck pain.12,22,59 Although studies have examined EIH in persistent neck pain,12,26,45,46,59 there is a paucity of studies investigating this in the early stages after the onset of neck pain. This lack of research may be explained by the difficulties of recruiting participants immediately before or after the onset of neck pain. Therefore, experimental pain models lasting several days provide an opportunity for investigating sensory changes during the early stage of neck pain.1,11 Previous experimental neck pain studies have been limited by pain models lasting only minutes.11,13 This short-lasting pain may not be sufficient to illuminate how pain may affect EIH over time. One study has used eccentric exercise to induce neck muscle pain lasting for a few days,1 whereas intramuscular injections of nerve growth factor (NGF) have shown promise as a model inducing muscle pain lasting several days.42,47 The advantage of an NGF model compared with eccentric exercises is the ability to extend the duration of pain with repeated injections.37,52

This study aimed to investigate the effect of NGF-induced experimental neck pain over 16 days in healthy participants and assess pain sensitivity and EIH, compared with control injections of isotonic saline. It was hypothesized that (1) the NGF group would display increased pain sensitivity and decreased EIH after neck pain onset compared with both baseline and the control group; and (2) these changes would normalize once the pain had subsided.

2 . Methods 2.1 . Participants

One hundred forty-three participants reported interest in the study after advertisements on social media and at local educational facilities (Fig. 1). Forty-one did not report back after receiving further information regarding the study, and 4 were no longer interested. Of those remaining, 47 fulfilled the inclusion criteria: healthy pain-free participants between 18 and 50 years of age and able to speak and read Danish or English. Exclusion criteria were any neck or shoulder pain during the past 6 months, any previous neck or shoulder surgery, any current or previous recurring painful conditions, or any neurologic, musculoskeletal, or mental illnesses that could influence the results. Furthermore, regular use of analgesics, drug addiction, heavy exercise during the days leading up to the test sessions, left-hand dominance, or pregnancy was also a cause of exclusion. Participants were asked to refrain from starting new exercise regimes, whereas any ongoing regular exercise training (eg, running, walking, cycling, and fitness) was allowed because there is no consensus that physical activity levels influence EIH in a young population.7,36 Participants were given written and verbal information about the study before providing their written informed consent. The study was approved by the local ethics committee (N-20180063). This study focusing on the effect of exercise on pain sensitivity is part of a larger project (Clinicaltrial.gov: NCT03848247) investigating potential effects of prolonged experimental neck pain on pain sensitivity, cortical excitability, and alterations in human movement.

F1Figure 1.:

Flowchart showing the inclusion and exclusion of participants for the study. DOMS, delayed-onset muscle soreness; PPT, pressure pain threshold.

As no previous study has used a similar experimental NGF neck pain model, the sample size calculations were based on pressure pain threshold (PPT) data from the neck area in clinical neck pain.12 Using G*power v3.1.9.4 (Heinrich-Heine-Universität, Düsseldorf, Germany), a sample size calculation was performed for a repeated measure analysis of variance (RM-ANOVA) with 2 groups, 4 days, 90% power, α 0.05, and Cohen's f for a medium effect size of f = 0.27. The correlation between repeated measures was set at 0.3 to account for interindividual differences in response to experimental neck pain, giving an estimated sample size of 36 needed. Therefore, 40 participants (20 per group) were recruited for this study to be conservative.

2.2 . Protocol

This study used a parallel-group design and was conducted in a laboratory setting at Aalborg University in 2019 with pilot testing starting in January, and the final session was completed in May. Participants were randomized (balanced) to either a neck pain condition (NGF group) or a control condition (control group; Fig. 2). Participants had to participate on 4 separate days; day 0, day 2, day 4, and day 15. Each session lasted approximately 2 hours, except for day 2 lasting approximately 30 minutes. All sessions started with participants reporting the sleep duration in the previous night, filling out the Neck Disability Index (NDI), and describing any painful experience (see Experimental Pain section below). Day 0, day 4, and day 15 followed the same protocol (Fig. 2). From a seated position on a chair without a backrest, PPTs were assessed with participants leaning forward with their upper body and forehead resting on a table in front of them. Pressure pain thresholds were assessed before and immediately after an upper limb exercise session, followed by reassessing any potential painful experience. Participants were informed that the objective of the study was to investigate any potential relationship between neck pain, pain sensitivity, and exercise but remained naive to both the study hypothesis and how exercise may influence EIH responses.

F2Figure 2.:

Study overview: Each session started with participants answering the NDI and recording any potential pain (NRS, body chart, and MPQ) before PPTs were assessed. On day 0, day 4, and day 15, before participants completed a hand-bike exercise, PPT and pain experience were recorded once more. On day 2, the hand-bike exercise was not performed. On day 0 and day 2, participants received injections of NGF or isotonic saline. MPQ, McGill Pain Questionnaire; NDI, Neck Disability Index; NGF, nerve growth factor; NRS, numerical rating scale; PPT, pressure pain threshold.

The protocol for day 2 included questionnaires and describing any potential painful experience before assessing PPTs. At the end of day 0 and day 2, participants received an injection of either NGF or isotonic saline as a control injection depending on group allocation. Participants picked an opaque envelope containing information on group allocation, which was only revealed to the person giving the injection.

2.3 . Experimental neck pain

A solution of sterile recombinant human NGF (0.5 mL, 5 μg, Skanderborg pharmacy, Skanderborg, Denmark) was injected into the right (side of hand dominance) splenius capitis muscle at the C3 level between the borders of the upper trapezius and the sternocleidomastoid muscles following a previously described protocol.11,13 An intramuscular NGF injection causes muscle pain/soreness during muscle contractions and increased sensitivity to pressure stimulation lasting a few days.6,25,42,53 Therefore, a protocol including 2 repeated NGF injections separated by 2 days was implemented to prolong the painful sensation.15,42,51 For the control group, injections of 0.5 mL isotonic saline (0.9%) were used, which should not cause any discomfort during muscle contraction or increased mechanical sensitivity.6,11,13,21 Both participants and assessors were blinded to group allocation.

Neck pain intensity was scored on an 11-point numerical rating scale (NRS; 0 = no pain, 10 = worst imaginable pain)31 during nonstandardized head movements where participants were asked to rotate their heads to look over their shoulders. The area of perceived pain was marked on a body chart.11–13 The marked areas on body charts were extracted using Adobe Acrobat Pro (San Jose, CA; v.2019.012.20034) in arbitrary units (a.u.), and the combined value (the area from anterior, posterior, and side views) was used for further analysis. The NDI questionnaire,60 expressed as a percentage of scored items, was used to assess changes regarding disability associated with neck pain. Quality of pain was described by choosing words from the McGill Pain Questionnaire.16,34 On day 4, all participants were asked which injection type, NGF or saline, they believed they received.

2.4 . Pressure pain sensitivity

Pressure pain sensitivity was assessed by PPTs using a handheld pressure algometer (Somedic, Hörby, Sweden) mounted with a 1 cm2 probe protected by a disposable cover. The pressure was steadily increased by 30 kPa/s. The threshold was defined as the instant that the pressure went from a pressure to first becoming painful, at which time point the participant would push a button wired to the algometer that recorded the pressure. Pressure pain thresholds were recorded at 3 bilateral locations (6 in total): (1) A neck site over the splenius capitis muscle (neck) just superior to the injection site (C3 level).11–13 (2) A segmental head site over the temporal muscle (head) above the base of the ear.11–13,29 (3) A distal leg site over the upper one-third of the tibialis anterior muscle (leg).10,32,42,44Pressure pain thresholds were always assessed first at the neck, then the head, followed by the leg site on one side of the body, followed by the contralateral side, and this was repeated 3 times. The start side (right/left) for PPT was randomized in a balanced way for both groups (NGF/control) and was the same between sessions for each participant. The average of the 3 measurements for each site was used for further analysis.

2.5 . Exercise-induced hypoalgesia

Upper limb exercise was performed using an adjustable hand-bike (SCI FIT Pro II; SCIFIT, Tulsa, Oklahoma) without a backrest. A progressive exercise protocol was adapted from previous studies.41,43 Participants were asked to maintain 70 rounds per minute (RPM) during the entire hand-bike exercise, which was ensured by visually observing the display showing RPMs. After 3 minutes with a bike resistance of 3.0 a.u., the resistance was increased by 0.5 a.u. every minute. The exercise session was stopped when the participants could no longer maintain 70 RPM or at a 16-minute time limit. The time of the last completed minute was noted and used as a target for the following sessions. Every minute during the hand-bike exercise, participants were asked to rate any neck pain on the NRS. Exercise-induced hypoalgesia was calculated by subtracting pre-exercise PPT values from postexercise values.

2.6 . Statistics

The statistical analyses were conducted using SPSS v.25 (IBM, Chicago, IL). Demographic and sleep data were compared between groups using a Mann–Whitney U test. Completed minutes of exercise, NRS scores of neck pain, area of perceived pain, and NDI were assessed over days for each group (NGF, control) using a Friedman's ANOVA. A Bonferroni-corrected Wilcoxon test was used as a post hoc test when indicated. Group comparisons at each time point were conducted using a Mann–Whitney U test, and likewise, a Bonferroni correction was implemented to correct for multiple comparisons. For the Friedman's ANOVA, the Kendall's W was reported, whereas η2 was reported for the Wilcoxon and the Mann–Whitney U test.

Pressure pain threshold data were analysed for normality using the Shapiro–Wilk test, after which the appropriate statistical approach was selected accordingly. First, PPTs at day 0 were analysed for potential group differences and to confirm an EIH response using a RM-ANOVA with Group (NGF, control) as between-subject factor and Side (right, left) and Time (pre-exercise and postexercise), after which pre-exercise PPTs at day 2, day 4, and day 15 were normalized (percentages) to baseline at day 0, thereby showing the change from baseline. Normalised PPT data were analysed with a RM-ANOVA with Group (NGF, control) as between-subject factor and Side (right, left) and Days (day 2, day 4, and day 15) as within-subject factors. This was performed separately for each site (neck, head, and leg), and Bonferroni-corrected pairwise comparisons were used for post hoc analysis.

Changes in EIH were analysed using a RM-ANOVA with Group (NGF, control) as between-subject factor and Side (right, left) and Days (day 0, day 4, and day 15) as within-subject factors. This was performed separately for each site (neck, head, and leg), and Bonferroni-corrected pairwise comparisons were used for post hoc analysis and reported along with 95% confidence interval (CI) and Cohen's d. For the RM-ANOVAs, the partial eta squared (ηp2) and 95% CI were reported. A significance level of 0.05 was accepted. All data in text and figures/tables are presented as the mean and SEM or median and interquartile range (25th and 75th percentile).

3 . Results

Of the 47 recruited participants, a total of 40 (20 in each group) completed the study (Fig. 1). Of those completing the study, 3 participants were unable to guess the type of injection they received when asked on day 4. No significant difference between groups was seen for age, height, weight, or hours of sleep (Table 1).

Table 1 - Median and interquartile range (25th and 75th percentile) for age, height, weight, hours of sleep (all days), and Neck Disability Index (all days). NGF group Control group Age (y) 26.5 (23.0-28.0) 26.0 (23.8-28.0) Height (cm) 175.0 (168.0-183.3) 172.5 (167.3-179.5) Weight (kg) 72.0 (60.3-85.0) 69.5 (64.0-78.3) Sleep (hr)  Day 0 7.5 (7.0-8.0) 7.3 (6.8-8.0)  Day 2 7.0 (6.9-8.0) 7.3 (6.9-8.0)  Day 4 7.0 (6.0-7.6) 7.0 (6.4-7.1)  Day 15 7.0 (6.5-8.0) 7.0 (6.4-7.6) NDI  Day 0 2.0 (0.0-2.5) 0.0 (0.0-2.1)  Day 2 10.0 (5.8-14.5)* 0.0 (0.0-2.0)  Day 4 10.0 (5.5-14.9)* 0.0 (0.0-0.5)  Day 15 1.0 (0.0-2.9) 0.0 (0.0-2.0)

The NDI was expressed as a percentage of scored items.

*Significantly different compared to day 0 and 15 and compared with the control group (P < 0.001).

NDI, Neck Disability Index; NGF, nerve growth factor.


3.1 . Perceived pain intensity, area, quality, and disability

For pre-exercise neck pain, the Friedman's ANOVA indicated a difference between days in the NRS scores during head movements (Table 2) for the NGF group (χ2(3) = 51.503, P < 0.001, W = 0.858). On day 2 and day 4, the NGF group had higher pain NRS scores compared with day 0 and day 15 (Wilcoxon: P < 0.002, η2 > 0.700). Similarly, for the NGF group, a difference between days was also found for neck pain during head movement after exercise (χ2(2) = 36.273, P < 0.001, W = 0.906) with significantly higher NRS scores at day 4 compared with day 0 and day 15 (Wilcoxon: P < 0.001, η2 > 0.742) and at day 15 compared with day 0 (Wilcoxon: P = 0.034, η2 = 0.32). No within-group difference was observed when comparing pre-exercise and postexercise pain NRS scores for the 2 groups. The NGF group displayed higher pain NRS scores compared with the control group on day 2, day 4, and day 15 at both pre-exercise and postexercise (Mann–Whitney: P < 0.03, η2 > 0.09).

Table 2 - Median (interquartile range: 25th and 75th percentile) pain scores using a numerical rating scale (0 = no pain, 10 = worst imaginable pain) and area of perceived pain (a.u.) for both groups (nerve growth factor, control) pre-exercise and postexercise on all days. NGF group Control group NRS score  Day 0 (pre-exercise) 0.0 (0.0-0.0) 0.0 (0.0-0.0)  Day 0 (postexercise) 0.0 (0.0-0.0) 0.0 (0.0-0.0)  Day 2 3.0 (2.0-4.0)*† 0.0 (0.0-0.0)  Day 4 (pre-exercise) 3.0 (2.0-4.0)*† 0.0 (0.0-0.0)  Day 4 (postexercise) 3.0 (2.0-4.0)*† 0.0 (0.0-0.0)  Day 15 (pre-exercise) 0.0 (0.0-1.0)† 0.0 (0.0-0.0)  Day 15 (postexercise) 0.0 (0.0-1.0)*† 0.0 (0.0-0.0) Area of perceived pain  Day 0 (pre-exercise) 0.0 (0.0-0.0) 0.0 (0.0-0.0)  Day 0 (postexercise) 0.0 (0.0-0.0) 0.0 (0.0-0.0)  Day 2 26.0 (12.5-52.0)*† 0.0 (0.0-0.0)  Day 4 (pre-exercise) 28.5 (9.0-50.5)*† 0.0 (0.0-0.0)  Day 4 (postexercise) 16.0 (8.25-38.0)*† 0.0 (0.0-0.0)  Day 15 (pre-exercise) 0.0 (0.0-7.75)† 0.0 (0.0-0.0)  Day 15 (postexercise) 0.0 (0.0-6.75)† 0.0 (0.0-0.0)

*Significant within-group difference compared with day 0 and day 15 (P < 0.04).

†Significant compared with the control group (P < 0.03).

NGF, nerve growth factor; NRS, numerical rating scale.

For the perceived area of pre-exercise pain (Table 2 and Fig. 3), the Friedman's ANOVA indicated a difference over days for the NGF group (χ2(3) = 49.281, P < 0.001, W = 0.821), with the post hoc test revealing larger areas of pain on day 2 and day 4 compared with day 0 and day 15 (Wilcoxon: P < 0.001, η2 > 0.730). Similarly, for the postexercise pain assessment (χ2(2) = 31.433, P < 0.001, W = 0.785), larger pain areas were seen for the NGF group on day 4 compared with day 0 and day 15 (Wilcoxon: P < 0.006, η2 > 0.480), whereas no differences were found when comparing areas of perceived pain before and after exercise. Between-group differences were observed with the NGF group displaying larger areas of pre-exercise pain on day 2, day 4, and day 15, as well as postexercise pain on day 4 and day 15 (Mann–Whitney: P < 0.03, η2 > 0.066).

F3Figure 3.:

Superimposed pre-exercise body chart drawings (posterior and right/injection side) for all test days (day 0, day 2, day 4, and day 15) and groups (NGF, Control). Darker markings indicate that these were more frequently chosen areas. NGF, nerve growth factor. § Significant within-group difference compared to Day 0 and Day 15 (P<0.001). *Significant difference compared to the control-group (P<0.03).

A difference over days was indicated for the NDI in the NGF group by the Friedman's ANOVA (χ2(3) = 44.103, P < 0.001, W = 0.735), with the post hoc test showing higher scores at day 2 and day 4 compared with day 0 and day 15 (Wilcoxon: P < 0.002, η2 > 0.658), and the Mann–Whitney U test confirmed that these days were also higher compared with the control group (Table 1; P < 0.001, η2 > 0.557). On day 2, the words chosen from the McGill Pain Questionnaire by >25% of the NGF group were “Annoying” (65%), “Sore” (45%), “tiring” (35%), and “tight” (35%). On day 4, the words chosen by >25% of the NGF group were “Annoying” (70%), “Sore” (55%), “pressing” (40%), “hot” (30%), and “tight” (30%). For the control group, no words were chosen by >25% on any day.

3.2 . Pressure pain sensitivity at baseline

No significant baseline (day 0) group difference in PPTs was found for any site (Table 3). For all sites, a main effect of time was observed (neck: F(1,38) = 52.7, P < 0.001, ηp2 = 0.581, 95% CI [−26.7 to −15.1]; head: F(1,38) = 21.0, P < 0.001, ηp2 = 0.357, 95% CI [−36.6 to −14.2]; leg: F(1,38) = 8.1, P < 0.007, ηp2 = 0.177, 95% CI [−51.2 to −8.8]), with postexercise PPT being significantly higher than prevalues. In addition, the analysis revealed a side difference for the neck and leg sites (neck: F(1,38) = 7.0, P < 0.012, ηp2 = 0.157, 95% CI [−18.7 to −2.5]; leg: F(1,38) = 4.6, P < 0.039, ηp2 = 0.108, 95% CI [1.3-45.2]), with the right side compared with the left having lower PPTs at the neck while the opposite was true for the leg. No side-by-time interactions were observed for any site.

Table 3 - Mean (kPa ± SEM, N = 40) pressure pain threshold values for the nerve growth factor (n = 20) and the control (n = 20) groups at the neck, head, and leg sites bilaterally (right, left) pre-exercise and postexercise. NGF group Control group Pre Post Pre Post Neck (right) 166.0 ± 12.7*† 184.9 ± 15.0 217.0 ± 23.0*† 237.1 ± 20.4 Neck (left) 182.7 ± 14.7* 194.0 ± 15.8 218.7 ± 19.1* 251.9 ± 22.2 Head (right) 251.0 ± 15.4* 280.9 ± 17.2 310.0 ± 24.9* 331.2 ± 26.5 Head (left) 262.4 ± 19.7* 283.8 ± 18.8 316.8 ± 30.0* 345.8 ± 28.9 Leg (right) 533.7 ± 61.5*† 561.5 ± 65.9 611.0 ± 46.0*† 639.3 ± 52.3 Leg (left) 495.0 ± 52.1* 526.7 ± 58.9 599.3 ± 42.7* 631.5 ± 52.2

*Significant main effect of time (P < 0.007).

†Significant main effect of side (P < 0.039).

NGF, nerve growth factor.


3.3 . Pre-exercise pressure pain sensitivity

For normalised PPTs at the neck site, the RM-ANOVA showed a Group × Time × Side interaction (Fig. 4; F(2,76) = 14.0, P < 0.001, ηp2 = 0.270). Post hoc pairwise comparisons revealed reduced PPTs for the NGF group compared with the control group at day 2 on both sides (right: 95% CI [26.0-54.0], P < 0.001, d = 0.914; left: 95% CI [6.8-26.9], P = 0.002, d = 0.538), day 4 (right: 95% CI [40.5-67.9], P < 0.001, d = 1.268; left: 95% CI [6.9-28.2], P = 0.002, d = 0.527), and day 15 (right: 95% CI [5.6-37.2], P = 0.009, d = 0.433; left: 95% CI [6.9-34.8], P = 0.005, d = 0.477). Furthermore, the NGF group displayed reduced PPTs on the right side on day 4 compared with day 2 (95% CI [−28.5 to −4.6], P = 0.004, d = 0.776) and day 15 (95% CI [−52.0 to −26.5], P < 0.001, d = 1.722) as well as on day 2 compared with day 15 (95% CI [−35.1 to −10.4], P < 0.001, d = 1.029). Moreover, the NGF group displayed reduced PPTs on the right side compared with the left side on day 2 (95% CI [−32.6 to −15.4], P < 0.001, d = 1.260) and day 4 (95% CI [−46.9 to −27.5], P < 0.001, d = 1.737).

F4Figure 4.:

Mean (+SEM, N = 40) normalised pressure pain thresholds for the NGF (n = 20, solid bars) and the control (n = 20, gray bars) groups at the neck, head, and leg sites bilaterally (right, left) on day 2, day 4, and day 15. Significantly different compared with the control group (*P < 0.03). Significantly different compared with other days (#P < 0.05). Significantly different between sides (¤P < 0.02). NGF, nerve growth factor; PPT, pressure pain threshold.

For normalised PPTs at the head site, a significant Group × Time interaction was found (F(2,76) = 4.58, P = 0.013, ηp2 = 0.108). The post hoc comparison showed that the NGF group displayed reduced PPT values compared with the control group on day 2 (95% CI [−21.0 to −3.3], P = 0.009, d = 0.440) and day 4 (95% CI [−27.4 to −8.5], P < 0.001, d = 0.608). In addition, the NGF group had lower PPTs on day 4 compared with day 2 (95% CI [−16.1 to −0.2], P = 0.042, d = 0.577) and day 15 (95% CI [−17.9 to −2.5], P = 0.006, d = 0.738).

For the leg site, a main group effect was seen with the NGF group overall displaying reduced PPTs compared with the control group (F(1,38) = 5.49, P = 0.024, ηp2 = 0.126, 95% CI [−21.2 to −1.5]). In addition, a Side × Time (F(2,76) = 5.71, P = 0.005, ηp2 = 0.131) interaction was found. On day 2, lower values were found for the PPTs on the right compared with the left side (P = 0.015, 95% CI [−11.8 to −1.3], d = 0.401). Furthermore, right-sided PPTs were higher on day 15 compared with day 2 (P = 0.027, 95% CI [0.8-17.0], d = 0.435) and day 4 (P = 0.009, 95% CI [1.9-16.1], d = 0.503), whereas left-sided PPTs were lower on day 4 compared with day 2 (P = 0.003, 95% CI [−16.3 to −2.8], d = 0.562) and day 15 (P = 0.001, 95% CI [−21.6 to −5.3], d = 0.652).

3.4 . Exercise-induced hypoalgesia

No significant within- or between-group difference was observed for the number of completed minutes during hand-bike exercise, with the overall number of minutes being 7.0 [5.0-10.0] for the NGF group and 6.5 [5.0-8.0] for the control group.

For the neck sites (Fig. 5), the EIH analysis revealed a group effect showing a smaller EIH effect for the NGF group compared with the control group (F(1,38) = 22.1, P < 0.001, ηp2 = 0.367, 95% CI [−34.5 to −13.7]).

F5Figure 5.:

Mean (+SEM, N = 40) exercise-induced hypoalgesia (EIH) effects (pre-exercise PPT subtracted from postexercise PPT) for the NGF (n = 20, solid bars) and the control (n = 20, gray bars) groups at the neck, head, and leg sites on day 0, day 4, and day 15. Significant difference compared with the NGF group (*P < 0.02). Significant difference compared with day 0 (§P < 0.05). NGF, nerve growth factor; PPT, pressure pain threshold.

For the head site, a Group × Time interaction was found (F(2,76) = 5.9, P = 0.004, ηp2 = 0.135). The post hoc comparison showed that the NGF group displayed a lower EIH effect compared with the control group at day 4 (95% CI [−61.4 to −22.9], P < 0.001, d = 0.728) and day 15 (95% CI [−54.3 to −7.6], P = 0.011, d = 0.425). In addition, the control group displayed an increased EIH effect at day 4 compared with day 0 (95% CI [9.6-49.2], P = 0.002, d = 0.831).

For the leg site, a Group × Time interaction was seen (F(2,76) = 7.1, P = 0.001, ηp2 = 0.158). The NGF group had a lower EIH effect compared with the control group at day 4 (P < 0.001, 95% CI [−154.7 to −72.4], d = 0.883) and day 15 (P < 0.001, 95% CI [−122.7 to −34.4], d = 0.569). In addition, at day 4 (P = 0.014, 95% CI [11.6-126.4], d = 0.673) and day 15 (P = 0.045, 95% CI [0.8-103.8], d = 0.569), an increased EIH was observed for the control group compared with day 0.

4 . Discussion

This study demonstrated that intramuscular NGF injections into healthy pain-free participants induced neck pain lasting for days and diminished EIH response compared with a control group. The NGF group reported larger areas of pain and higher pain intensity and displayed local and widespread hypersensitivity to pressure compared with the control group without neck pain.

4.1 . Prolonged pain and disability

Neck pain intensity during head movements and area of perceived pain in the current study were comparable with previous findings for clinical neck pain.12,44 Surprisingly, unlike previous observations based on clinical neck pain, this study did not find increased pain intensity or area of perceived pain after exercise. This discrepancy between clinical populations,12 where pain may arise from several cervical structures,8,14 and the current NGF model could be due to the model itself. NGF injections increase mechanical sensitivity that in turn can cause pain during muscle contraction.6 However, the injected muscle, splenius capitis, in the current study, was not directly involved in the exercise, which could explain why perceived pain did not change. An alternative explanation for the lack of increased intensity or area of pain after exercise could be that the NGF model did not sufficiently sensitize central mechanisms. Another factor to consider is how pain develops over time20 where the current model resembles acute–subacute pain, which may not be comparable with persistent neck pain. However, when considering some of the words most chosen by the participants to describe their experienced pain in the current study, such as “annoying,” “tiring,” or “nagging,” these represent evaluative and affective aspects of pain,33 which are believed to be increasingly present in persistent pain compared with more acute conditions.38 This is supported by previous studies where words describing affective aspects were commonly reported by those with persistent neck pain12,27 compared with healthy populations experiencing a short-lasting experimental neck pain.11,19

Taken together, the pain intensity, the area of perceived pain, and the words chosen to describe pain, the current NGF model may be suitable for mimicking the early phases of neck pain. This is supported by the NDI ratings, where the NGF group had a median score of 10% on day 2 and day 4, which would be interpreted as mild disability.49,61

4.2 . Local and widespread hyperalgesia

This study found decreased PPTs locally over the neck site, ipsilateral to the injection, which was most pronounced on days with the highest pain intensity and could be considered a clinically meaningful reduction.

留言 (0)

沒有登入
gif