- Top of page
- Materials and Methods
Abstract: The combination of paracetamol with non-steroidal anti-inflammatory drugs (NSAIDs) is widely used; however, the nature and mechanism of their interaction are still debated. A double-blind, pharmacokinetic/pharmacodynamic, randomized, cross-over, placebo-controlled study was carried out in human healthy volunteers. The aim was to explore the existence of a positive interaction between paracetamol 1 g and ketorolac 20 mg administered intravenously on experimental pain models in human beings. The effects of the paracetamol–ketotolac combination were compared with similar doses of respective single analgesic and to placebo on the sunburn model (UVB-induced inflammation), cold pain tolerance and the nociceptive flexion reflex. The kinetics of the plasma concentrations of paracetamol and ketorolac were measured using 2D-liquid chromatography-mass spectrometry. Thirteen volunteers were screened, and 11 completed the study. Ketorolac significantly decreased primary hyperalgesia to heat stimuli compared with paracetamol (p < 0.014). The combination performed better than paracetamol (p < 0.001) and placebo (p < 0.042), increasing heat pain threshold by 1.5°C. The combination radically reduced primary hyperalgesia to mechanical stimulation (39%) compared with placebo (p < 0.002) and single agents (paracetamol p < 0.024 and ketorolac p < 0.032). The combination also reduced, slightly although significantly, the intensity of pain (10%) for the cold pressor test (versus placebo: p < 0.012, paracetamol: p < 0.019 and ketorolac p < 0.004). None of the treatments significantly affected the central models of pain at this dosage level. No pharmacokinetic interactions were observed. These results suggest a supra-additive pharmacodynamic interaction between paracetamol and ketorolac in an inflammatory pain model. The mechanism of this interaction could mainly rely on a peripheral contribution of paracetamol to the effect of NSAIDs.
Multimodal analgesia involves a combination of different classes of analgesics to improve pain relief and reduce the adverse effects of drugs . The combination of paracetamol with non-steroidal anti-inflammatory drugs (NSAIDs) is widely used in clinical practice although it has been demonstrated to be effective in pain management only in few clinical studies [2–4]. The putative rationale underlying this association is their different site and/or mode of action (central versus peripheral, serotonin and/or cannabinoid systems versus prostaglandin synthesis) . The mechanism of action of paracetamol is yet to be fully elucidated . Much evidence supports the hypothesis of a central analgesic effect , while the other evidence supports a peripheral effect . The prostaglandin H2 synthase (PGHS) , the serotoninergic system [10,11] and/or the cannabinoid system [12,13] have been proposed as potential targets. NSAIDs essentially act by inhibiting prostaglandin biosynthesis at the site of the inflammation . However, some NSAIDs also exhibit a central action . Experimental data suggest that indomethacin , ibuprofen  and ketoprofen  are centrally acting NSAIDs, while ketorolac  or acetylsalicylic acid  are mainly peripherally acting agents. The central effect of the former could be because of pharmacokinetic factors and the ability of one drug to enter the central nervous system (CNS) . Indomethacin, ketoprofen  and ibuprofen  are rapidly detected in the cerebrospinal fluid after a systemic administration, whereas acetylsalicylic acid is not .
There are various methods available in human experimental pain research to quantitatively assess various aspects of pain , and these are useful tools to characterize the analgesic effects of drugs . Models provoking primary and secondary hyperalgesia are relevant and mimic clinical pain, which is often associated with signs of primary and secondary hyperalgesia. Primary and secondary hyperalgesia can be induced by burn injury, capsaicin, freeze injury or ultraviolet irradiation . The sunburn model, in which UVB irradiation induces stable primary and secondary hyperalgesia , was chosen in our study because such stability provides an ideal setting for pharmacokinetic/pharmacodynamic studies. The primary area of hyperalgesia induced by UVB irradiation is explained by the sensitization of peripheral nociceptor terminals and peripheral pain hypersensitivity, while the secondary hyperalgesia involves a central mechanism of sensitization .
This study aimed at experimentally assessing in healthy volunteers: (i) the peripheral and central analgesic efficacy of a combination of paracetamol with ketorolac administered intravenously; (ii) to compare it with single agents and respective placebo. According to the hypothesis, the combination would result in an increased pharmacodynamic effect on an inflammatory model of pain.
- Top of page
- Materials and Methods
The present study aimed to assess the existence of an additive effect of a combination of paracetamol with ketorolac on experimental pain models and to explore the putative site of the interaction. It was observed that the combination significantly decreased UVB-induced heat and mechanical primary hyperalgesia, reflecting a peripheral effect. Additionally, the combination seemed to be superior to ketorolac or paracetamol used separately. Regarding pain tolerance, the combination in comparison with individual analgesic significantly decreased pain intensity during hand immersion in cold water (cold pressor test). The effect was minor although, as the paracetamol–ketorolac combination did not affect the immersion time, in contrast to opioids [36–39]. Nociceptors of cutaneous veins appear to mediate cold pain via the activation of both Aδ- and C-fibres . Ketorolac alone did not have an effect on the cold pressor test, which is in agreement with its mechanism of action on prostaglandin synthesis and with a previous study . Therefore, an enhancement of the effect of paracetamol is postulated, whose mechanism remains to be elucidated.
The results obtained confirm the potential benefit of combining paracetamol with NSAIDs as it is usually carried out clinically. This is in accordance with the evidence regarding a boosted analgesic effect of combining paracetamol and NSAIDs available from animal studies, experimental pain models in human being [28,41] and clinical studies of pain on osteoarthritis , dental surgery  or other selected surgical procedures [2,3]. However, the exact and quantitative nature of the interaction (additive or supra-additive) and its mechanisms are still under debate.
As proposed by a study in a rodent model of visceral pain for assessing the combination of paracetamol with several commonly used NSAIDs, using the isobolographic approach , the results obtained from the present study could suggest that the interaction is of a supra-additive nature. Indeed, neither of the two agents produced a significant effect on primary hyperalgesia to mechanical stimuli nor pain intensity to cold pain, whereas the combined treatments significantly increased these pain thresholds. Ketorolac alone had an effect on thermal hyperalgesia. On the other hand, the combination by itself increased the pain threshold to a greater extent than ketorolac alone. An isobolographic approach would have been necessary to determine the type of interaction (additive, infra-additive, supra-additive and antagonistic) . Pharmacokinetic/pharmacodynamic (PK/PD) modelling combined with an isobolographic technique would have been interesting to provide such information. In our study, the effect of paracetamol alone on hyperalgesia was not significant, making it impossible to derive a dose–response curve. Therefore, the data were insufficient for PK/PD modelling as well as for an isobolographic approach. Although paracetamol has been used for decades, no clear relationship between analgesic effects and plasma concentrations has yet been determined .
Several mechanisms are thought to be involved in the mechanism of paracetamol-induced analgesia . The proposed targets include the central serotoninergic descending pain pathways [10,11] and the cannabinoid system [12,13,46,47]. The action of paracetamol on cyclooxygenase (COX) has been the subject of debate for several years. Furthermore, good evidence suggests that paracetamol has a weak inhibitory influence on peripheral prostaglandin synthesis, which would account for its lack of anti-inflammatory activity. However, it could be a potent inhibitor of prostaglandin synthesis in the CNS. Consequently, it would produce both analgesia and antipyresis . Recent data have demonstrated that the analgesic effects of paracetamol were analogous to those of selective COX-2 inhibitors [9,49]. The enzyme commonly referred to as COX is in fact named PGHS. The PGHS catalyses the two-step oxygenation of arachidonic acid to prostaglandin H2 (PGH2), a prostaglandin precursor . This reaction is catalysed by distinct active sites within the enzyme. The NSAIDs target the COX site, which catalyses the first step formation of prostaglandin G2 (PGG2) from the arachidonic acid. The peroxidase (POX) site catalyses the PGH2 formation from PGG2. It is believed that paracetamol inhibits PGHS by its ability to serve as a reducing cosubstrate for the POX. Elevated peroxide levels evident in inflamed tissues decrease the inhibitory action of paracetamol on PGHS, which explains its lack of anti-inflammatory effect . The inhibitory action of paracetamol on prostaglandin synthesis has also been demonstrated to exist in peripheral sites [8,49]. Thus, paracetamol could have a significant peripheral effect in addition to its main central effect only when it is combined with an NSAID. In the present study, the observed effect on primary hyperalgesia could result from an action of paracetamol on both the central and the peripheral PGHS, and our hypothesis being that ketorolac might enhance the action of paracetamol on PGHS by decreasing the peroxide levels through its anti-inflammatory effect, leading to a supra-additive antinociceptive effect. Therefore, the mechanism responsible for an increased analgesic effect of the combination could well be a result of the distinct mode and site of action.
A pharmacokinetic interaction between paracetamol and ketorolac was excluded by the pharmacokinetic analysis. Indeed, it was found that the pharmacokinetic data could be compared when each analgesic was given alone or in combination, and the obtained pharmacokinetic parameters were consistent with the previously published data [50–52].
The present study confirms the usefulness of the UVB model in investigating the primary antihyperalgesic effect of NSAIDs [30,53]. Additionally, it also confirms the remarkably stable conditions of primary hyperalgesia to thermal and mechanical stimuli lasting for more than 26 hr after UVB exposure, as reported earlier . A different area was irradiated during each study session as tanned skin is less sensitive to UV. No drug effect was observed in the control contralateral skin zone, thereby confirming the prior observations that an inflammatory component is essential to assess NSAID efficacy in experimental pain models .
Two of the volunteers did not develop secondary hyperalgesia. Thus, they did not have allodynia to pinprick outside the irradiated site for one of the four study sessions. The potential reasons for this could be related to the fact that a lower dose of UVB was given, in comparison with Gustorff et al. , and the minimal erythema dose was determined according to the skin type and not individually.
In the present study, no significant effect on secondary hyperalgesia between the treatments and placebo was observed. However, as expected, a trend towards an effect was observed when paracetamol was used. The limitation to the obtained results is probably related to the fact that the study is underpowered for this secondary end-point. Indeed, as our primary end-point was the heat pain threshold in the area of primary hyperalgesia, the present study was not designed to detect a difference in the area of secondary hyperalgesia to pinprick. According to the data obtained from Gustorff et al. , a 30% reduction in secondary hyperalgesia area can be detected with 11 volunteers in the cross-over design. However, as mentioned before, the dosage used for irradiation and the surface exposed was minimal. This explained the lack of power of the present study with the 11 volunteers. The inclusion of more volunteers in our study might probably have given more robust and significant results, on secondary end-points as well, and this could well have provided more information on the mechanisms of the interaction between paracetamol and ketorolac. However, a recent study with 16 healthy volunteers found no significant effect of paracetamol on hyperalgesic areas induced by intradermal electrical stimulation, although a trend was observed, which confirms, in agreement with several previous studies that used the same electrically evoked human pain model, that paracetamol displays a minor analgesic effect in this experimental setting .
In conclusion, the present clinical study in healthy volunteers showed that a combination of paracetamol and ketorolac was superior to either agent alone on an inflammatory pain model. The observed results suggest a supra-additive interaction between paracetamol and ketorolac that should be further investigated. An action on different steps of the prostaglandin synthesis could contribute to the observed increased effect of the combination on primary hyperalgesia. Further studies are warranted to evaluate an action on different steps of the prostaglandin synthesis of the combination.