Direct intrawound administration of dimethylsulphoxide relieves acute pain in rats

Abstract

Wounds associated with injuries such as burns can produce moderate to severe pain. Besides causing distress to the patient, unrelieved pain could delay healing owing to stress‐related problems. Thus, pain needs to be treated as early as possible after injury. It was hypothesised that local treatment of wounds with appropriate analgesic drugs could attenuate pain. HOE 140, a bradykinin receptor antagonist, reduced acute inflammatory pain in rats after intrawound administration. In this study, the analgesic effect of dimethylsulphoxide (DMSO) was investigated in a similar hind‐paw incision model in rats. An extremely small quantity (10 µl) of 100% DMSO was administered into the incision site just before closure of the wound. It persistently attenuated guarding behaviour in rats over a period of 3 days without affecting thermal hyperalgesia or allodynia. Accumulated evidence indicates that guarding is equivalent to pain at rest in humans. The possible mechanisms of the analgesic effect could be inhibition of C group of peripheral nerve fibres or even free radical scavenging. Healing of the wound was found to be normal at the end of the study period. In conclusion, DMSO could be useful in the treatment of acute pain resulting from tissue injuries such as burns.

Introduction

Injuries have become a part of our fast‐paced life. Pain associated with tissue damage resulting from burn‐related injuries, accidents or violence leads to considerable anxiety and discomfort 1. The initial sharp pain due to activation of Aδ fibres is soon replaced by a dull aching pain mediated by C fibres 2. Frequently, medical care is directed to wound care and restoration with little attention being paid to actual pain originating from the wound site. Even if attended to, treatment is limited to oral or intramuscular non‐steroidal anti‐inflammatory drugs (NSAIDs) and rarely with weak opioids such as tramadol 3. Untreated pain has been reported to delay wound healing possibly because of stress‐related mechanisms 4.

Following tissue damage, chemical mediators such as bradykinin, cytokines and purines are released from the surrounding tissue, which sensitise the thinly myelinated (Aδ) and unmyelinated (C) peripheral nerve fibres 5, 6. Passage of repeated afferent nerve impulses to the spinal cord produces a form of synaptic plasticity in the dorsal horn called central sensitisation 7. A vicious cycle of pain is thus initiated characterised by decreased threshold and spontaneously occurring pain even in the absence of fresh peripheral stimuli. Countering specific chemical mediators can relieve acute inflammatory pain in rats as has been reported recently using the bradykinin type 2 receptor antagonist HOE 140 8. This peripheral analgesic action can also occur following infiltration/infusion of NSAIDs and/or local anaesthetic agents into surgical wounds 9, 10. In a recent study, patients receiving combined local infiltration analgesia plus spinal anaesthesia during herniorrhaphy had better pain control and were discharged earlier than those administered spinal anaesthesia alone 11. Again, hip arthroplasty performed under local infiltration analgesia provided better pain relief with reference to analgesic consumption, pain intensity on mobilisation and lesser side effects than intrathecal morphine at specific time points after surgery 12.

The role of dimethylsulphoxide (DMSO) in relieving pain following direct intrawound administration has never been investigated. Antinociceptive effect of DMSO following intra‐cerebroventricular and oral administration has been shown recently in thermal escape pain models and during formalin test in mice 13. In patients with complex regional pain syndrome (CRPS) type 1, topical application of 50% DMSO over the affected area had shown favourable results, particularly when associated with inflammatory symptoms 14, 15. DMSO is derived from processing of paper pulp and is used as a preservative during stem cell or organ transplantation. Its direct use for medicinal purposes has only been allowed for the treatment of pain in interstitial cystitis 16. In this study, DMSO was administered directly into the wound following hind‐paw incision in rats. The pain behaviour was evaluated by the guarding behaviour, thermal hyperalgesia and mechanical allodynia. Guarding behaviour reflects continuously occurring background pain in animals, which prevents them from fully supporting their body weight on the affected paw. Thermal hyperalgesia and mechanical allodynia represent hypersensitivity to a high‐intensity thermal stimulus and to innocuous mechanical stimulation, respectively. Guarding would tend to reflect pain at rest (non‐evoked pain) in humans, whereas thermal hyperalgesia and mechanical allodynia are observable in patients after tissue injury as, for example, during the postoperative period 17.

Materials and methods

Adult male Sprague‐Dawley rats weighing 200–225 g were sourced from the Experimental Animal Facility of AIIMS (control, n = 6; DMSO treated, n = 6). All experimental procedures were approved by the Institutional Animal Ethics Committee. The procedure for hind‐paw incision has been reported before 18. The paw was swabbed with povidone‐iodine followed by isopropyl alcohol. Briefly, under inhalation of isoflurane in air and oxygen mixture (1:1 ratio), plantar surface of the right hind paw was incised aseptically by a no. 11 scalpel blade. A 1‐cm long incision was made through the skin and underlying fascia. The flexor digitorum brevis muscle was elevated by a forceps and a 0·5‐cm longitudinal incision was made along the long axis without disturbing the attachments. The tips of a forceps were introduced through the cut and the muscle fibres were separated slightly by releasing the blades. The muscle was replaced back. Finally, the edges of the wound were lifted up to create a pocket and 10 µl of 100% DMSO was administered into the wound by a sterile micropipette. Pure DMSO was previously procured in the form of a sterile solution (Santa Cruz Biotechnology, Santa Cruz, CA). Following 30‐second waiting period, the wound was closed by two mattress sutures using 4‐0 silk (Ethicon®, Johnson & Johnson, Himachal Pradesh, India) and an antiseptic ointment (Neosporin®, Glaxo SmithKline Pharmaceuticals Ltd, Mumbai, India) was applied. The animals were transferred to a warm recovery chamber. In the control group, similar procedure was followed except that no drug was administered into the wound. Following recovery, animals were housed individually in cages with soft clean bedding made of cellulose (ALPHA‐dri, Shepherd Speciality Papers, Milford, NJ).

The pain resulting from this artificial but controlled tissue damage was determined by the guarding behaviour (2 hours, 8 hours and days 1–4), thermal hyperalgesia (2 hours, 8 hours and days 1–7) and mechanical allodynia (2 hours, 8 hours and days 1–7) after surgery in a blinded manner by the observer. The procedure for evaluating guarding has been described before and is expressed by the cumulative pain score 19. The rats were kept over a metal wire mesh (8 × 8 mm²) within Perspex enclosures. After an acclimatisation time of 15 minutes, the position of the plantar surface of both hind paws was observed from below the mesh for a 1‐minute period. If the incised area of the affected paw was completely off the mesh for the maximum time period, a score of 2 was awarded. Similarly, if the concerned area was lightly touching the mesh, 1 was given and 0 if there was full weight bearing on the incision site. Full weight bearing was indicated by blanching of the affected area. The other paw was also observed and marked according to its position. This was repeated once every 5 minutes for a 1‐hour period. The total of 12 scores for each paw was calculated. The difference between the two scores was the final cumulative score, which could range from − 24 to + 24. In naive rats, basal values of guarding lie between 0 and 2. For mechanical allodynia, the ‘up‐down method’ described by Chaplan et al. was used 18, 20. Mechanical allodynia was assessed immediately after studying the guarding behaviour. Rats were allowed to remain over the elevated metal mesh platform covered with plexiglass boxes (15 × 15 × 15 cm³), which permitted testing of the paws by filaments extended through the wire mesh from below. Calibrated von Frey filaments (North Coast Medical Inc., San Jose, CA) of sizes 3·61 to 5·18 were used. These were pressed to the skin on the medial side of the incision in the proximal part of the paw until it buckled 18. The filament was allowed to exert pressure for 7–8 seconds. The first filament was of size 4·31. If withdrawal occurs, the next lighter filament was used to exert pressure and in case, there was no withdrawal, the next heavier filament was used. Testing was continued until four filaments had been applied after the first one that produced a withdrawal response. This was done to eliminate variations in the data with respect to the 50% g withdrawal threshold. If no withdrawal was noted with the heaviest filament, 15 g was recorded. On the contrary, withdrawal on pressure with the lowest filament was recorded as 0·4 g. The procedure of determining thermal hyperalgesia has been reported earlier 21. The rats were placed on a glass platform within Perspex enclosures, which allowed unrestrained movements (Plantar Test Apparatus, UGO Basile, Italy). After acclimatisation for 15 minutes, the thermal stimulus was directed at the plantar surface of the hind paw, at about the middle of the incision site. The rat withdrew it paw on feeling pain and the withdrawal latency to 0·1 second was automatically detected. A cut‐off time of 20 seconds was maintained to prevent damage to the paw. The percent maximum possible effect was derived by the following formula:

$$\begin{array}{cc}& \left[\mathrm{Drug}-\mathrm{induced}\mathrm{latency}-\mathrm{Baseline}\mathrm{latency}/\right.\\ & \left.\mathrm{Cut}-\mathrm{off}\mathrm{latency}-\mathrm{Baseline}\mathrm{latency}\right]\times 100\end{array}$$

In the same group of animals (control, n = 6; DMSO treated, n = 6), the tissue around the incision site was dissected out on day 8 and processed for Cresyl violet (0·5% solution) staining. This is a basic stain for observation of the nuclei; 7‐µm‐thick paraffin sections were cut using a rotary microtome. The sections were taken on glass slides coated with gelatin and mounted using DPX. Images were acquired using E‐600 Nikon microscope connected to a ProgRes CT5 CCD camera (Jenoptik, Jena, Germany).

Data were analysed by the statistical software GraphPad Prism (version 5) using two‐way analysis of variance (ANOVA) followed by post hoc Bonferroni test for comparison between groups using the factors ‘drug’ and ‘time’. P value <0·05 was considered statistically significant. Values are expressed as mean ± standard error of the mean.

Results

The rats recovered consciousness rapidly after anaesthesia inhalation was stopped. Guarding was evaluated initially followed by allodynia and then thermal hyperalgesia (Figure 1). In the control group, guarding was maximum immediately after surgery, similar to clinical situations (Figure 1A). It decreased progressively over the next 4 days and attained values close to the basal level by the end of this time period. Treatment with DMSO not only reduced guarding behaviour over the entire time period (2 hours to day 4) but also brought it back to the basal level by day 4. Significant relief was noted between 2 hours and day 3. This antinociceptive effect could be considered to be long‐lasting in comparison to the control group. General observation also showed that the DMSO‐treated rats were moving around more than those in the control group. However, allodynia or thermal hyperalgesia did not differ between control and DMSO‐treated animals (Figure 1B,C).

Figure 1.

Post‐incisional pain was evaluated in the control and dimethylsulphoxide (DMSO)‐treated groups. (A) Guarding behaviour was determined as the cumulative pain score over a period of 1 hour at every time point starting from 2 hours to day 4 after incision. Higher scores indicate more pain. Guarding was maximum after the injury and then gradually decreased over the next 4 days. It was close to the basal level at day 4. DMSO treatment reduced the score over the entire observation period. Significant decrease was evident at 2 hours, 8 hours and days 1–3. (B) Allodynia or pain sensitivity to innocuous mechanical stimuli was highest immediately after injury in the peri‐incisional area. It is expressed as 50% withdrawal threshold. Higher values indicate less pain. It lessened somewhat in the control and DMSO‐treated groups over the 7‐day observation period. Significant difference between the two groups was not observed. (C) Thermal hyperalgesia is determined by the time required to sense the thermal stimulus and the withdrawal of the affected paw (withdrawal latency). It is represented as the maximum possible effect (%). Higher values indicate more pain. No difference was noted between the control and the DMSO‐treated groups. Values are shown as mean ± SEM. P value < 0·05 is considered statistically significant. *P < 0·05; **P < 0·01; ***P < 0·001.

Histopathological examination of the incision site showed fibroblastic proliferation and resolution of the inflammatory process, which were well advanced in both the control and the DMSO‐treated groups (Figure 2). Healing was equivalent in both the groups at the microscopic and macroscopic levels. No obvious side effects were noted in both the groups.

Figure 2.

Incision site and the surrounding peri‐incisional tissue were removed from the paw and processed for paraffin staining using Cresyl violet stain on day 8. The incision site is filled with healthy granulation tissue (star marked). This is covered by proliferating epidermal tissue in which many keratinocytes are also visible (block arrow). No difference was noted between the control (A) and the DMSO (B)‐treated groups. Scale bar represents 200 µm.

Discussion

Pain has been labelled as a protective reflex. Despite this, continuous pain can adversely affect the quality of life. Relief from pain by systemically administered drugs such as NSAIDs is a common therapeutic procedure but local administration of suitable analgesic drugs for treating pain is rare except for the relief of postoperative pain. Washing wounds with normal saline or water has been recommended 22. Distinct classes of nociceptors such as heat‐sensitive (H), mechano‐heat (MH)‐sensitive and mechanically insensitive (MIS) nociceptors have been identified in the periphery by single‐unit recording from C‐nerve fibres 23. Some of these nerve fibres also respond to chemical mediators of inflammation 24. They do so by expressing specific receptors such as prostaglandin E2 and bradykinin B2 receptors for these chemicals. Thus, the nociceptors are both directly responsive to tissue damage as well as indirectly to chemical mediators released following the damage. It is possible that the subacute phase of pain is caused mostly by chemical mediators.

In this work, administration of an extremely small amount of DMSO (10 µl) immediately after tissue damage produced long‐lasting relief of guarding behaviour over the next 3 days. This was considered to be long‐lasting because in an earlier study in our laboratory under similar conditions, HOE 140, a bradykinin type 2 receptor antagonist, could relieve guarding up to day 2 8. Guarding has been correlated to spontaneous activity of nociceptors in the affected area 25. An earlier study has also observed increased guarding with more severe tissue damage. Thus, it has been hypothesised that guarding could mimic pain at rest in humans following surgical interventions 26, and DMSO could significantly relieve guarding. The mechanism of action could be direct blockade of neural transmission in the C‐nerve fibres 27, and it is also a free radical scavenger. Several oxygen‐derived free radicals such as superoxide (O₂ ⁻) anion and hydroxyl (OH⁻) ion are generated during inflammation, and neutralisation of these relieves pain 28, 29, 30. The transient receptor potential channels mediate the algesic effect of free radicals 31. Thus, it is possible that the significant relief in the guarding behaviour could be due to the antioxidant effect of DMSO. Again, painful peripheral neuropathy resulting from oxaliplatin, a chemotherapeutic agent in rats, was markedly relieved by antioxidants like α‐lipoic acid 32. Similar mechanism of action is thought to account for the beneficial effect of DMSO in CRPS type 1 33.

DMSO could affect guarding behaviour without affecting thermal hyperalgesia or mechanical allodynia. As explained in the methods section, following tissue injury on the plantar surface of the paw, the rat protects the paw from further injury by lifting it and thus avoiding weight transmission. If at all the rat uses the paw, weight transmission occurs through the toes or through the non‐affected parts of the plantar aspect of the paw. Estimation of the guarding behaviour is a measure of weight transmission through the injured area of the right paw in comparison to the same area on the opposite normal paw. Allodynia is the occurrence of mechanical and tactile hypersensitivity in and around the injured area. Experimentally, it is evaluated by pressing the tips of nylon monofilaments to a specified part of the peri‐incisional area and looking for the spontaneous withdrawal of the paw, somewhat similar to the flexor withdrawal reflex in humans. Finally, escape from nociceptive thermal stimuli is a well‐standardised model of estimation of nociception (thermal hyperalgesia). For example, tolerance to experimental heat pain is increased during pregnancy 34. It has been suggested that the underlying mechanisms of guarding (non‐evoked) and both thermal hyperalgesia and mechanical allodynia (evoked varieties) could be different 17. The time course of the occurrence of these pain modalities is also different – while guarding resolves by 4 days after incision, the others run a more protracted course for a week or more. Differential analgesic effect on guarding predominantly and to some extent thermal hyperalgesia but not on allodynia has been reported after systemic administration of antibody to the nerve growth factor 35.

This is the first study on the antinociceptive effect of DMSO after local administration in an animal model of tissue injury; 100% DMSO was used because of the limited scope of application within the small wound size in these rats and to prevent possible run‐off from the wound. The standardisation of the concentration (100%) and dose (10 µl) of DMSO was done in preliminary experiments. Also, DMSO in similar dose and concentration has been previously used for investigating the analgesic effect of candidate drug molecules after intrathecal administration 36, 37. In clinical studies, 50% concentration could be used in therapeutic amounts. It may be speculated that DMSO could lessen pain following its administration in burn injuries, which are associated with one of the most severe types of pain. Its use in postoperative pain relief following surgery could also be investigated.

This study reports on the novel antinociceptive effect of DMSO after intrawound administration. Combined with other analgesic drugs like NSAIDs, this could prevent unnecessary pain after tissue injury. Preliminary studies in our laboratory have shown that intrawound administration of flurbiprofen, a non‐selective cyclooxygenase inhibitor, dissolved in DMSO could produce a short but significantly higher analgesic effect than DMSO alone in rats (unpublished observations). Definitive analgesic effect following topical administration of diclofenac (dissolved in vehicle containing 45·5% DMSO), an NSAID, in osteoarthritis patients has been reported 38. No adverse effect was noted apart from dryness of skin. However, to the best of our knowledge, analgesic effect resulting from intrawound administration has never been investigated in the past. In conclusion, additional studies are required in humans to further investigate this novel analgesic effect of DMSO.