Nociceptive and Cognitive Changes in a Murine Model of Polytrauma

Abstract

Polytrauma commonly involves concussion (mild traumatic brain injury [mTBI]) and peripheral trauma including limb fractures. Interactions between mTBI and peripheral injuries are poorly understood, both leading to chronic pain and neurobehavioral impairments. To elucidate these interactions, a murine polytrauma model was developed. mTBI alone resulted in similar increased mechanical allodynia in male and female mice. Female fracture and polytrauma groups displayed greater increases in hind paw tactile hypersensitivity for weeks after injury than did the respective male groups. Capsaicin-evoked spontaneous pain behaviors were greater in fracture and polytrauma female mice compared with male mice. The mTBI and polytrauma male mice displayed significant deficits in spatial working memory. All fracture, mTBI, or polytrauma groups had deficits in object recognition memory. Only male mTBI or polytrauma mice showed greater agitation and increased risk-taking behavior in open field testing as well as zero maze tests. Additionally, impaired diffuse noxious inhibitory control was observed in all mTBI and polytrauma mice. The model presented offers clinically relevant features useful for studying persistent pain as well as cognitive and other behavioral changes after TBI including polytrauma. A better understanding of nervous system dysfunction after TBI and polytrauma might help prevent or reduce persistent pain and disability in these patients.

Perspective:

The polytrauma model presented has relevant features of chronic pain and neurobehavioral impairments useful for studying mechanisms involved in their development. This model may have special value in understanding altered descending pain modulation after TBI and polytrauma.

With a global incidence estimated to be about 106 per 100,000 individuals, traumatic brain injury (TBI) is a leading cause of trauma-related disability.¹⁵ In the United States, more than 1.7 million injuries occur per year, resulting in 250,000 hospitalizations and 52,000 deaths ⁹. The majority (75–80%) of all TBI cases are, however, mild, and are accompanied by the rapid resolution of the immediate symptoms including disorientation, dizziness, tinnitus, nausea, and balance problems.²⁴ The incidence of TBI is higher overall in men than in women, and is particularly common in the very young, adolescents, and elderly.³ Women have a higher rate of mortality after TBI, but scant attention has been given to many other outcomes.^8,31 The economic costs of TBI are immense in terms of medical care and indirect costs including lost wages and productivity. In 2010, total costs in the United States were estimated to be $76.5 billion.⁷ TBI is therefore a significant public health problem.

While the immediate symptoms of mild TBI (mTBI) dissipate rapidly, various long-term sequelae are common. For example, chronic pain is experienced by many with histories of TBI. One recent meta-analysis involving data from more than 4,200 patients determined the average rate of post-TBI pain to be >50%.²⁸ Surprisingly, the same study showed mTBI to be associated with relatively high (75.3%) rates of pain whereas moderate to severe injuries associated with much lower (32.1%) rates. Although headache is common after TBI,²⁰ the resulting pain can be widespread and may include the back and extremities.¹¹ Quantitative sensory testing of the painful limbs of TBI patients has demonstrated mechanical allodynia and other sensory abnormalities consistent with neuroplastic changes in pain processing within the brain and spinal cord.²⁹ Recent studies in human subjects suggest that disrupted descending inhibition of nociceptive signaling pathways may contribute to pain after TBI.⁶

TBI can occur in isolation, but in many cases TBI is accompanied by additional extracranial injuries such as fractures, soft tissue damage, and visceral injuries. Motor vehicle accidents, sports-related injuries, and military trauma are all settings in which TBI may be a component of the resulting polytrauma. The prevalence of pain in the setting of polytrauma is exceptionally high and can exceed 80%.¹⁹ In a recent study involving patients with concomitant TBI and peripheral injury, it was noted that worsened functional outcomes were experienced by those with TBI plus a peripheral injury compared with those having TBI alone, but the interaction was more likely in the setting of less severe TBI.²⁶ A recent study using a mouse TBI plus fracture multitrauma model identified elevated brain levels of interleukin-1β a month after injuries leading the investigators to speculate that exacerbated neuroinflammation supported the increased risk taking observed in the animals.⁴¹

In this report, we describe studies using a closed-head TBI model with or without tibia fracture to study post-TBI nociceptive sensitization in male and female mice. We further evaluate the role of descending inhibition in these mice.

Methods

Animals, Experimental Groups, and Conditions

Male and female C57BL/6J mice 11 to 12 weeks old obtained from the Jackson Laboratory (Bar Harbor, MA) were kept in our facility a minimum of 1 week before initiating the experiments. All mice were kept under standard conditions with a 12-hour light/dark cycle and an ambient temperature of 22 ± 1°C. Food and water were available ad libitum. Mice were habituated to handling by the experimenters for a few minutes each day for 3 days before initiation of experiments. Figure 1 provides an outline of the experimental timeline. The following 4 experimental groups were used referring to the use of TBI/limb fracture: sham/sham, TBI/sham, sham/fracture, or TBI/fracture (polytrauma). For these experiments, we have calculated group sizes (8 to 10 mice) to have approximately 80% power to detect 25% changes at the P < .05 level. For behavioral experiments, the experimenters were blind to the identity of treatments but not their experimental condition as fracture and polytrauma mice had obvious gait differences after injury. Data analysis was carried out by the experimenters blind to the experimental condition of the animals and identity of treatments.

Figure 1.

Overview of the experimental timeline. Abbreviations: mTBI, mild traumatic brain injury; B, baseline mechanical assay; Fx, fracture; DNIC, diffuse noxious inhibitory control; OLM, object location memory; ORM, object recognition memory; CAP, capsaicin.

Ethics Approval

All experiments were approved by the Veterans Affairs Palo Alto Health Care System Institutional Animal Care and Use Committee (Palo Alto, CA) and followed the animal subjects’ guidelines of the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Closed-Head Model of mTBI

The closed head mTBI and sham procedures were based on previously described protocols.^21,50 Briefly, a benchmark stereotaxic impactor (MyNeurolab, St. Louis, MO) actuator was mounted on a stereotaxic frame (David Kopf Instruments, Tujunga, CA) at a 40° angle with a 5-mm impactor tip. After isoflurane anesthesia induction, mice were placed in a foam mold held in prone position on the stereotaxic frame and maintained under anesthesia for the duration of the procedure. The stereotaxic arm was adjusted so that the head impact was at a fixed point relative to the right eye and ear, corresponding to the S1 somatosensory cortex. The impact delivered by the device to the head was 5.8 to 6.0 m/s with a dwell time of .2 seconds and impact depth of 5 mm. After impact, the mice were recovered from anesthesia on a warming pad before returning to their home cages. Skull fractures were not observed in our study similar to previously reported studies using comparable head impact forces.^21,50 For sham groups, the previous procedure was performed except that the impact device was discharged in the air.

Tibial Fracture

Mice underwent right-leg distal tibia fracture or sham procedures a day after TBI, as previously described.^12,46 Briefly, mice were anesthetized with isoflurane and a closed fracture of the right tibia just distal to the midpoint was done using a hemostat. Next, the limb was wrapped in casting tape (Scotchcast Plus, 3M Corporation, St. Paul, MN.) so the hip, knee, and ankle were all fixed, forming a spica around the abdomen. The model results in a well characterized transverse bone fracture and soft-tissue injury and cast immobilization is essential for the extended pain-related phenotype.¹³ To prevent postinjury edema constriction, a window was made in the cast over the dorsal paw and ankle. Sham groups underwent the same duration of anesthesia as the fracture mice. Mice were recovered from anesthesia on a warming pad and given 1.5-mL normal saline for hydration. Both injured and sham groups received subcutaneous buprenorphine for 2 days. Food and water gel-packs were provided on the floor for 72 hours post procedure. Casts were removed 3 weeks after fracture.

Mechanical Nociceptive Assay

Mechanical sensitivity was assessed using nylon von Frey filaments (Stoelting Co, Wood Dale, IL) according to the “up-down” algorithm developed by Chaplan et al.⁵ We have applied this technique previously to detect 50% withdrawal threshold in mice after injury.^38,46 After acclimating mice on the wire mesh platform inside plastic enclosures (10-cm radius), sequential fibers with increasing stiffness ranging from .004 to 1.7 g were applied to the plantar surface of hind limb and left in place 5 seconds. When 4 fibers had been applied after the first response, the testing terminated. Withdrawal of hind paw from the fiber was considered a response. If a response occurred after application of a fiber then a less stiff fiber was applied, if no response was observed the next stiffest fiber was applied. Mechanical withdrawal threshold was determined by a data-fitting algorithm for significance analysis.³⁵

Capsaicin-Evoked Behaviors

This assay was performed in separate cohorts of mice to determine if any differences in latent sensitization exist between the treatment groups after recovery.⁴⁵ The total duration of behaviors (shaking, biting, and licking) for 5 minutes after subcutaneous (s.c.) hind paw capsaicin injection (2.0 μg/5 μL; Sigma-Aldrich, St. Louis, MO) was measured. Control experiments used vehicle (.25% dimethyl sulfoxide, .25% ethanol, .125% Tween-80 in saline) plantar injections. The mice were gently restrained for the plantar injections. The assay was done at 20 to 22 weeks post-TBI and fracture trauma when all groups had recovered to baseline mechanical threshold values.

Open Field, Zero Maze, and Object Memory Tests

The open field (OF) arena (40 cm × 40 cm × 40 cm) was used to assess locomotion, object location (OLM), and object recognition (ORM) working memory tests. In OF test, total locomotor activity (distance traveled) and time spent in the center 20% of the arena for 10 minutes was determined. Increased time spent in the peripheral zones of a novel environment indicative of thigmotactic behavior was used as an index of anxiety, while time spent in the center 20% used as an index of risk-taking behavior.

The elevated zero maze (ZM) was used as well, to measure anxiety and risk-taking behavior according to previously published methods.⁴⁵ The maze is 24 inches above the floor, has an outer diameter of 24 inches and inner diameter of 20 inches, and 2 closed (6-inch-tall walls) and open quadrants. Mice were randomly placed facing one of the closed quadrants at the beginning of the 5-minute test. Total time spent in open and closed quadrants were recorded. Time spent in the open quadrants was compared across groups to measure anxiety and risk-taking behavior.

For OLM and ORM experiments, mice were initially habituated to 2 identical objects after being placed in the middle of the OF arena, as previously described.^45,51 Next, during a 5-minute trial, one of the objects was moved to a novel location and exploratory behavior (investigation time) was recorded. Subsequent to a 5-minute period in home cages, mice were returned to the arena with one of the previous identical objects being replaced with a novel one. The new object had a distinct shape and size difference from the identical object sets, and 5-minute recordings were done for exploratory behavior toward objects. As mice explore novel locations or objects more than familiar ones, time spent exploring the novel compared with familiar was used to assess spatial and nonspatial working memory.

All recordings from the above experiments were analyzed in real time by TopScan software (Version 2.0, CleverSys, Reston, VA).

Diffuse Noxious Inhibitory Control Assessment

Diffuse noxious inhibitory control (DNIC) assessment of post-trauma mechanical hypersensitivity was done using a noxious stimulation–induced analgesia protocol modified for mice.^18,33 DNIC was induced by s.c. application of capsaicin (7.5 μg/5 μL; Sigma-Aldrich) to the right forepaw after brief isoflurane (3%) anesthesia. Control mice received 5 μL of vehicle (.25% dimethyl sulfoxide, .25% ethanol and .125% Tween 80 in saline). The amount of DNIC induced by capsaicin injection was evaluated by hind paw mechanical threshold testing. Separate groups of mice in the sham/fracture or TBI/fracture groups were assessed for DNIC at 4 weeks postinjury, at which time mechanical hypersensitivity was the same between the 2 groups.

To assess DNIC in the TBI alone group, the mice were allowed to recover to baseline mechanical thresholds after TBI (3 weeks). Next, right hind paw prostaglandin E2 (PGE2) injections were given to produce brief hypersensitivity.⁴⁴ Separate groups of sham/sham or TBI/sham groups were given ipsilateral or contralateral hind paw injection (s.c.) of 100 ng/15 μL PGE2 (Cayman Chemicals, Ann Arbor, MI) and tested after 1 hour for mechanical hypersensitivity. Stock PGE2 solutions were made in 100% ethanol and further diluted (1:1,000) in .9% saline before use. Vehicle (15 μL) injections contained the same dilution of ethanol in saline. Subsequent to mechanical threshold evaluations, mice underwent capsaicin injection in the ipsilateral forepaw, and changes in mechanical withdrawal thresholds were followed.

Assessment of Analgesics on mTBI-Induced Mechanical Hypersensitivity

The effects of anti-inflammatory and antineuropathic analgesic agents on TBI induced mechanical hypersensitivity were assessed at the 72-hour time point. Systemic (intraperitoneal) injections of carprofen (5 mg/kg; Sigma-Aldrich) or gabapentin (50 mg/kg; Sigma-Aldrich) were given to sham or TBI mice and mechanical withdrawal thresholds assessed after 90 or 60 minutes, respectively. The doses of the both drugs were in the high range used by other investigators examining effects in separate models.^39,48 Vehicle control animals were employed.

Data Analysis

The data for mechanical sensitivity were analyzed by multiple t tests with the Holm-Sidak method to correct for multiple comparisons (no assumption of same scatter). For these experiments, groups of mice were randomly assigned to treatment groups and followed across days weeks till recovery. Therefore, measurements for each subject were done for 10 time points for TBI alone experiments and 13 to 14 time points for polytrauma groups. To assess overall sex differences after trauma, repeated 2-way analysis of variance (ANOVA) was used. The data from capsaicin induced spontaneous pain behaviors, OF, and ZM experiments were analyzed by 1-way ANOVA followed by Tukey’s post hoc test for multiple comparisons. Two-way ANOVA followed by Tukey’s post hoc tests (95% confidence interval of differences plus significance: α = .05) were used to compare groups in the DNIC, analgesic, and object memory tests. D’Agostino and Pearson omnibus test showed the data obtained were normally distributed.

All data are presented as mean ± SEM, and for all analyses p < .05 were taken to be significant.

Results

Mechanical Hypersensitivity After mTBI, Tibial Fracture, or Combined Injury

Male and female mice displayed increased contralateral and ipsilateral (relative to injury side) hind paw mechanical sensitivity after mTBI, with peak increases lasting for 72 hours and gradual recovery to baseline values by 14 days after injury. TBI generally resulted in increased contralateral versus ipsilateral limb mechanical hypersensitivity, though significant limb differences were only seen at some time points in female mice between days 1 to 7 postinjury (Fig. 2A and 2B). No significant male versus female differences for respective contralateral (F_9,162 = 1.64, P = .11) or ipsilateral (F_9,162 = 1.82, P = .07) hind paw thresholds after TBI was seen on further analysis.

Figure 2.

Mechanical hypersensitivity after mild traumatic brain injury (TBI). (A) Male and (B) female mice displayed increased hind paw mechanical sensitivity after TBI. Data were analyzed by multiple t tests using the Holm-Sidak method to correct for multiple comparisons. The asterisk indicates a significant difference in comparison with respective limb measurements in the sham group. The number sign indicates a significant difference between ipsilateral and contralateral limb to the side of injury. Error bars = SEM. n = 8 to 10 per group.

The mice in the fracture or combined injury polytrauma groups of both sexes displayed increased mechanical hypersensitivity for weeks after injury with recovery to baseline values at 17 to 19 weeks in male cohorts and 19 to 21 weeks in female ones (Fig. 3A and 3B). Within-group analysis revealed that the polytrauma group had significantly greater mechanical hypersensitivity than fracture alone in male mice across several weeks, no significant differences between the female mice of either injury group were seen. Further analysis showed significant male versus female mice mechanical hypersensitivity differences after fracture (F_12,204 = 7.14, P < .001) and polytrauma (F_12,216 = 8.07, P < .001), with the female groups demonstrating greater nociceptive sensitivity.

Figure 3.

Mechanical hypersensitivity after peripheral trauma or combined traumatic brain injury (TBI) and peripheral trauma. (A) Male and (B) female mice in the fracture (Fx) or polytrauma groups displayed mechanical hypersensitivity for weeks after injury, with recovery to baseline values at 17 to 19 weeks in male cohorts and 19 to 21 weeks in female cohorts. Male mice but not the female mice showed significant differences between Fx- and polytrauma-induced mechanical hypersensitivity. The asterisk indicates a significant difference in comparison with the respective sex-combined injury group. Error bars = SEM. n = 8 to 10 per group. Data were analyzed by multiple t tests using the Holm-Sidak method to correct for multiple comparisons.

Capsaicin-Induced Spontaneous Pain Behaviors After mTBI, Tibial Fracture, or Combined Injury

To detect differences in latent sensitization in the experimental groups, capsaicin evoked behaviors were measured after recovery from mechanical hypersensitivity (20 to 22 weeks) in male and female mice. Both fracture (male mice: P < .05; female mice: P < .001) and polytrauma groups (male mice: P < .001; female mice: P < .001) displayed increased spontaneous pain behavior after hind paw capsaicin application compared with respective sham groups. For both sexes, combined TBI and tibial fracture groups displayed greater capsaicin-evoked behaviors than did mice with either injury alone (Fig. 4A and 4B). Female mice displayed greater duration of capsaicin-evoked behaviors than male mice for all injury (TBI: male mice = 37.60 ± 2.50 vs female mice = 60.85 ± 2.80, P < .001; fracture: male mice = 47.86 ± 3.61 vs. female mice = 85.24 ± 3.11, P < .001; and TBI/fracture : male mice = 76.34 ± 7.10 vs. female mice = 107.90 ± 4.08, P < .001) conditions.

Figure 4.

Capsaicin-evoked behaviors after mild traumatic brain injury (TBI), peripheral trauma, or combined injury. The total duration of behaviors (shaking, biting, and licking) in (A) male and (B) female mice for 5 minutes after hind paw capsaicin (2.0 μg/5 μL) or vehicle injection was measured. The assay was done at 20 to 22 weeks post-trauma period when all groups had recovered to baseline mechanical threshold values. Error bars = SEM. n = 8 to 10 per group. *P < .05. ***P < .001 for comparison to respective sham groups. ^###P < .001 for comparison within groups. Data were analyzed by 1-way analysis of variance followed by Tukey’s post hoc test for multiple comparisons. Abbreviation: Fx, fracture.

Working Memory Deficits After mTBI, Tibial Fracture, or Combined Injury

Next, we sought to determine if differences in spatial and nonspatial working memory deficits existed between the experimental groups. Previously, we had shown that memory deficits accompany pain sensitization after fracture.⁴⁵ Similar to our previous findings, male fracture mice had deficits in nonspatial (ORM, P = .53 novel vs same) but not in spatial (OLM, P < .01 novel vs same) working memory. TBI alone and polytrauma male groups displayed deficits in OLM as seen by insignificant differences in the time spent on novel versus familiar location (Fig 5A) (TBI: P = .98; polytrauma: P = .96, novel vs same). Neither the sham nor the other injury female groups showed significant preference for novel location in the OLM test, and therefore it was not possible to assess spatial working memory differences in these groups (Fig 5B). Male TBI and polytrauma groups displayed deficits in the ORM test, as seen by insignificant differences in time spent on novel versus familiar object (Fig 5C) (TBI: P = .53; polytrauma: P = .95, novel vs same). Contrary to the OLM test performance, the female sham mice displayed significant preference for the novel object (ORM, P < .05, novel vs same). Neither fracture nor TBI nor polytrauma female mice demonstrated a significant preference for novel object (Fig 5D) (fracture: P = .80; TBI: P = .12; polytrauma: P = .91, novel vs same). As the sham female mice showed more total combined object exploratory activity in either test than respective male mice did (OLM: male mice = 11.47 ± 2.71 vs. female mice = 43.14 ± 6.19, P < .001; ORM: male mice = 17.29 ± 2.71 vs. female mice = 34.57 ± 6.16. P < .05), comparing injury type effects between sexes would be challenging.

Figure 5.

Working memory evaluations after mild traumatic brain injury (TBI), peripheral trauma, or combined injury. Object location (OLM) and object recognition (ORM) working memory tests were employed. The interval between acquisition and retrieval was 5 minutes in either test. (A–C) Male fracture (Fx) mice displayed no deficits in spatial working memory (OLM) but had deficits in nonspatial (ORM) working memory. The TBI alone and polytrauma male groups displayed significant deficits in both OLM and ORM tests. (B–D) Female mice in either experiential group displayed no significant preference for novel location in the OLM test, but only the female sham mice displayed significant preference for the novel object (ORM). All recordings were analyzed in real time by automated software and data obtained were analyzed by 2-way analysis of variance followed by Tukey’s post hoc tests. Error bars = SEM. n = 8 to 10 per group. *P < .05 or ***P < .001 for comparison between novel and familiar location/object.

Evaluation of Locomotor Activity, Anxiety, and Risk-Taking Behavior After mTBI, Tibial Fracture, or Combined Injury

Determination of OF locomotor activity in the male mice showed increased distance traveled after TBI alone (P < .001) or when combined with tibial fracture (P < .05) compared with control mice (Fig 6A), indicative of greater agitation in these groups. Female mice after fracture or TBI did not show significant differences in locomotor activity, though the polytrauma mice displayed decreases (P < .01) in distance traveled compared with control mice (Fig 6B). Overall activity in the female groups was higher than in males. Male polytrauma mice spent significantly more time (P < .01) in the center 20% of the OF, suggestive of increased risk-taking behavior and decreased anxiety (Fig 6C). None of the female groups significantly differed in time spent at the center compartment (Fig 6D). Next, we went on to investigate anxiety and risk-taking behavior in the same groups using the elevated ZM test.³⁴ Male polytrauma mice spent significantly more time (P < .01) in the ZM open arms than did other male groups (Fig 7A), providing corroborating evidence for decreased anxiety and increased risk-taking behavior. Fracture, TBI and the combination did not alter open arm times in the female groups (Fig 7B).

Figure 6.

Locomotor activity and anxiety evaluations after mild traumatic brain injury (TBI), peripheral trauma, or combined Injury. In the open field test the (A) male mice showed increased distance traveled after TBI or polytrauma compared with control mice. (B) Female mice after fracture (Fx) or TBI did not show significant differences in locomotor activity, though the female polytrauma mice displayed decreases in distance traveled compared with control mice. (C) Male polytrauma mice but not (D) female polytrauma mice spent more time in the center 20% compartment of the arena, which was indicative of decreased anxiety and increased risk-taking behavior. All recordings were analyzed in real time by automated software and data obtained were analyzed by 1-way analysis of variance followed by Tukey’s post hoc tests. Error bars = SEM. n = 8 to 10 per group. *P < .05, **P < .01, or ***P < .001 for comparison with sham.

Figure 7.

Anxiety evaluations after mild traumatic brain injury (TBI), peripheral trauma, or combined Injury. In the zero maze the (A) male polytrauma mice displayed reduced anxiety, as seen by increased time spent in the open arms, indicative of increased risk-taking behavior. (B) None of the female groups showed differences in anxiety levels as measured by time spent in the open arms. All recordings were analyzed in real time by automated software and data obtained were analyzed by 1-way analysis of variance followed by Tukey’s post hoc tests. Error bars: SEM, n = 8 to 10 per group, **P < .01 for comparison with sham. Abbreviation: Fx, fracture.

DNIC Deficits After mTBI, Tibial Fracture, or Combined Injury

DNIC has been suggested to worsen some types of pain in TBI patients. In experiments directed at detecting possibly deficient DNIC in TBI and sham injured mice, we first allowed TBI animals to recover for 3 weeks after injuries, such that hind paw nociceptive thresholds had returned to baseline (Fig 2). DNIC was then tested by sensitizing hind paws with PGE2 followed an hour later by ipsilateral forepaw capsaicin injection to stimulate descending inhibition.³³ Figures 8A and 8B show that capsaicin robustly reverses ipsilateral hind paw sensitization in sham injured male and female mice, but has almost no effect in animals 3 weeks after TBI, suggesting profound disruption of DNIC by TBI in either sex. Further analysis showed differences in response to capsaicin between male and female TBI mice were not statistically different (F_1,16 = .60, P = .45). Separate groups of male and female mice were tested for DNIC of the contralateral hind paw. After ipsilateral forepaw capsaicin application, male and female TBI groups exhibited reduced DNIC effects on contralateral paw mechanical sensitivity compared with sham groups (Fig. 8C and 8D).

Figure 8.

Descending noxious inhibitory control deficits after mild traumatic brain injury (TBI). To determine differences in descending pain control inputs, capsaicin (CAP) (arrow) noxious stimulation–induced analgesia was used. The mTBI (A, C) male and (B, D) female groups show descending pain control deficits in either hind limb compared with control mice. Hind paw mechanical sensitivity data were analyzed by 2-way analysis of variance followed by Tukey’s post hoc tests. Error bars = SEM. n = 8 to 10 per group. ***P < .001 for comparison between mTBI and sham. Abbreviation: prostaglandin E2.

Additional experiments were undertaken 4 weeks after tibial fracture in mice with or without TBI, when mechanical sensitivity was same between the groups. In these experiments, we observed that forepaw capsaicin injection transiently reversed hind paw sensitization in fracture alone groups, but that this form of analgesia was very limited in mice that had both tibial fracture and TBI (Fig. 9A and 9B). No significant differences in response to capsaicin were seen between male and female mice fracture-alone groups (F_1,16 = .24, P = .63) or combined injury groups (F_1,16 = .32, P = .58). For both groups, DNIC assessments were done in fractured (ipsilateral) side. As the contralateral paw in either group is not sensitized at 4 weeks, sensitization with PGE2 would be needed for DNIC assessments making interpretation of results difficult.

Figure 9.

Descending noxious inhibitory control deficits after peripheral trauma or polytrauma. To determine differences in descending pain control inputs, capsaicin (CAP) noxious stimulation–induced analgesia was used. Polytrauma (A) male and (B) female groups displayed deficits in descending pain control compared with fracture (Fx) mice alone. Hind paw mechanical sensitivity data were analyzed by 2-way analysis of variance followed by Tukey’s post hoc tests. Error bars = SEM. n = 8 to 10 per group. ***p < .001 for comparison between polytrauma and Fx. Abbreviation: TBI, traumatic brain injury.

Assessment of Anti-Inflammatory and Antineuropathic Analgesic Agents on TBI-Induced Mechanical Hypersensitivity

Systemic injections of carprofen (5 mg/kg) or gabapentin (50 mg/kg) were given to sham or TBI mice and mechanical withdrawal thresholds assessed after 90 or 60 minutes, respectively. On day 3 following injury, carprofen administration did not attenuate TBI induced hind paw sensitization in male or female mice (Fig. 10A and 10B). Similarly, gabapentin failed to attenuate TBI-induced hind paw sensitization in both sexes (Fig. 10C and 10D). Control experiments using vehicle in sham and TBI groups did not show significant changes in paw withdrawal threshold in either sex (data not shown). These data suggest the increased mechanical hypersensitivity seen after TBI is centrally modulated.

Figure 10.

Effects of anti-inflammatory and nonopioid analgesic agents on traumatic brain injury (TBI)–induced mechanical hypersensitivity. After baseline mechanical sensitivity measurements, caprofen (5 mg/kg, intraperitoneal) or gabapentin (50 mg/kg, intraperitoneal) were given to sham or TBI mice 3 days after injury and mechanical withdrawal thresholds assessed after 90 or 60 minutes, respectively. Neither (A, B) carprofen nor (C, D) gabapentin administration (arrows) resulted in significant changes in TBI-induced hind paw sensitization in male or female mice. Hind paw mechanical sensitivity data were analyzed by 2-way analysis of variance followed by Tukey’s post hoc tests. Error bars = SEM. n = 6 to 8 per group.

Discussion

Our understanding of the long-term sequalae of TBI is presently improving. Cognitive, emotional, and sensory symptoms are increasingly recognized in this patient population. Only recently have human and animal studies begun to explore the mechanisms supporting high pain prevalence after TBI with very little having been written about the interaction of TBI with peripheral injuries or the impact of sex on pain-related outcomes. In this regard the studies contained in this report are highly novel. Our major findings are: 1) mTBI causes nociceptive sensitization lasting about 2 weeks, and the resulting sensitization is refractory to anti-inflammatory and antineuropathic analgesics, being more consistent with central pain syndromes; 2) mTBI worsens nociceptive sensitization after limb fracture; 3) sex does not impact post-TBI nociceptive sensitization, but does affect the sensitization caused by limb fracture; 4) descending regulation of nociception is profoundly affected by TBI in both sexes; and 5) both concomitant limb fracture and sex modulate anxiety, risk-taking behavior, and memory changes after TBI. Together, these observations provide a basis for exploring the molecular, cellular, and circuit-level alterations after TBI, supporting pain and other long-term sequelae.

One aspect of the experimental design adding to the face validity of our mTBI mouse model is the use of closed-head injury in generating mTBI, the predominant form of TBI. Though there are many methods used to model brain injury in animals, they most often involve opening or penetration of the skull. Conversely, sports-related concussions or acceleration/deceleration injuries and combat-related exposures to blast waves do not typically cause damage to the skull itself. Previously described TBI injuries used in pain studies including those by our own group have involved the application of a percussive pressure wave to the exposed dura of the animal, or the direct impact of an object against the brain.^10,22,37 The present model first described by Shitaka et al⁴⁰ somewhat more realistically involves a non-penetrating blow against the skull. Variants of the model have been used to study repetitive brain injury such as can occur to those playing contact sports.¹⁴ In fact, neither we nor others using this closed-head TBI model identified gross motor or behavioral changes in the TBI mice, suggesting that more subtle forms of injury may support pain-related changes in the brain after TBI.

Similar to the other models involving disruption of the skull, we did observe sensitization to noxious mechanical stimuli. For example, Macolino et al²² demonstrated sensitization of the tissues overlying the skull and the forepaw after cortical impact-induced injury. Our group recently focused on hind limb sensitization after TBI induced by lateral fluid percussion.¹⁰ The more distal site was assessed to better examine the sensitization of areas well outside of the head region, similar to areas affected in patients with TBI including the back and limbs. Different mechanisms may underlie pain and sensitization in the head versus back and limbs as altered expression of the neurokinin 1 receptor has been demonstrated in the trigeminal nucleus after TBI,²⁷ while the enhanced expression of pronociceptive brain-derived neurotrophic factor occurs in lumbar spinal cord tissue after TBI and associated hind limb sensitization.¹⁰

It is not uncommon for TBI to occur with other injuries on the battlefield in a motor vehicle accident, and so on. Others have combined TBI with injuries such as long bone fracture to model “polytrauma.”³² An excellent review has been written detailing the known interactions of TBI and various forms of peripheral injury in clinical populations and laboratory models.²⁶ These combined injury models have been used to study bone healing and other consequences of injury, but not persistent pain.^23,47 Most studies have investigated consequences of multiple injuries on serum levels of injury markers and the extent of brain injury within the first several days. One exception was a report by Shultz et al⁴¹ demonstrating in a closed-head TBI plus tibial fracture multitrauma model not only enhanced neuroinflammation and exacerbated changes in ventricular volume, but also increased risk-taking behavior in male mice about a month after injury. Our own studies add significantly to this literature by showing that polytrauma mice display enhanced peripheral nociceptive sensitization for months and have latent sensitization after recovery. These data are consistent with reports of chronic pain at sites distant from the head in multitrauma TBI patients.

A major mechanism for the control of pain-related signaling flowing into the brain from periphery is descending regulation. Fibers originating in brainstem send axons to the dorsal horn of the spinal cord where they may positively or negatively regulate signal transmission.³⁰ Such regulation can be demonstrated by applying a noxious stimulus to a site in the periphery while following a pain or nociceptive threshold in another area. In animals, this type of regulation is referred to as DNIC while in humans it is referred to as conditioned pain modulation.⁴⁹ This name emphasizes the general or “diffuse” nature of the regulation. Laboratory and clinical studies have linked the efficiency of descending regulation to chronic pain syndromes such as fibromyalgia, tension headache, neuropathic and chronic postoperative pain.^33,43,49 Importantly, brainstem imaging has recently shown that the extent of injury in the periaqueductal gray matter is correlated with pain levels in TBI patients.¹⁷ Our observations showed that weeks after TBI, mice of either sex had severely reduced DNIC providing a possible basis for the enhanced nociception seen after limb fracture. It should be noted that TBI patients with chronic post-traumatic headache display diminished conditioned pain modulation also.⁶ Existing therapies such as serotonin and norepinephrine reuptake inhibitor drugs might be used to augment DNIC in patients with TBI just as they seem to be useful in controlling pain in patients with other forms of chronic pain characterized by dysfunctional DNIC.³⁰ The brainstem, which contains major descending regulatory centers including the periaqueductal gray matter, locus coeruleus, and rostral ventromedial medulla, has been noted to be an area prone to TBI injury.²⁵ Studies directed at understanding the extent of injury to these centers after TBI are warranted.

Additionally, examining similarities or differences in pain behavior after TBI over other cortical regions besides the somatosensory cortex would complement the findings of our study.¹⁶

The present study has been focused on characterizing a novel polytrauma paradigm and is limited to evaluating evoked tactile sensitivity and pain-related behaviors. Further studies examining the extent of spontaneous pain, as well as, additional pharmacological and behavioral testing to distinguish decreased anxiety form increased risk-taking behaviors would further strengthen the characterization of this model of TBI and polytrauma.

Men sustain TBI at a rate far higher than women do.³ However, the consequences of those injuries may be different for the sexes. For example, a recent survey of military personnel that had suffered mTBI suggested that women experienced higher rates of some sensory changes, including sensitivity to light and changes in taste and smell.² Other analyses have found greater mortality after TBI among women, although these differences may be less when multiple injury severity is taken into account.¹ Likewise, functional recovery after TBI does not appear to be strongly linked to sex.⁴ These data are in contrast to sex-related differences in the prevalence of various pain syndromes that often show a predilection for women over men,⁴² including pain after surgical trauma.³⁶ Our data failed to identify sex-linked differences in nociceptive sensitization after mTBI, and no clear difference in the additive effect of TBI to the sensitization observed after tibial fracture. On the other hand, there was a prominent effect of sex on sensitization after tibial fracture alone as reported previously.⁴⁶ Similar to our previous study, male fracture mice had deficits in nonspatial working memory. We went on to evaluate memory and other pain- or trauma-related behavioral comorbidities in both sexes for interaction of limb trauma and TBI.⁴⁶ Sex impacted anxiety, risk-taking behavior, and memory to differing extents in our studies. We did observe both sexes to have deficits in nonspatial working memory after limb trauma, TBI, or both. These results in combination with the clinical observations suggest that the effects of sex on TBI outcomes might be dependent on the specific outcomes being examined, as they may involve different underlying mechanisms.

Conclusions

Nociceptive sensitization in regions distant from the brain can be modeled after closed-head mTBI, including sensitization resulting from the combination of mTBI with peripheral injury. This type of model provides a tool critical to studying the very common occurrence of chronic pain after TBI, including multi- or polytrauma. Our animal data are consistent with clinical reports in suggesting that brain centers involved in descending modulation are involved in enhanced pain sensitization after mTBI, while we do not at this point understand the details of these changes. A better understanding of the specific involved brain centers, the modulation of their activities and a more detailed understanding of the mechanisms of injury to the brain after mTBI might help prevent or reduce chronic pain in this population.