What Do We (Not) Know About How Paracetamol (Acetaminophen) Works?

Journal of Clinical Pharmacy and Therapeutics (2010) 35, 617–638 doi:10.1111/j.1365-2710.2009.01143.x

REVIEW ARTICLE What do we (not) know about how paracetamol (acetaminophen) works?

K. Toussaint* PharmD,X.C.Yang*PharmD, M. A. Zielinski* PharmD,K.L.Reigle* PharmD,S.D.Sacavage*PharmD, S. Nagar PhD and R. B. Raffa PhD Temple University School of Pharmacy, Philadelphia, PA, USA

SUMMARY BACKGROUND What is known and background: Although paracetamol (acetaminophen), N-(4-Hydroxyphe- The recent (June, 2009) convening of an FDA joint nyl)acetamide, is one of the world’s most widely meeting of the drug safety and risk management used analgesics, the mechanism by which it pro- advisory committee with the anaesthetic and life duces its analgesic effect is largely unknown. This support drugs advisory committee and the non- lack is relevant because: (i) optimal pain treatment prescription drugs advisory committee, which had matches the analgesic mechanism to the as its primary topic area for discussion the safe use (patho)physiology of the pain and (ii) modern drug of acetaminophen (acetaminophen: para-acetylam- discovery relies on an appropriate screening assay. inophenol; paracetamol: para-acetylaminophenol; Objective: To review the clinical profile and pre- Tylenol: para-acetylaminophenol; APAP: N-acetyl- clinical studies of paracetamol as means of gain- para-aminophenol) and a recent article (1) and ing insight into its mechanism of analgesic action. accompanying commentary (2) in this journal Methods: A literature search was conducted of about the relative risk of paracetamol vs. aspirin, clinical and preclinical literature and the infor- resurrect the following interesting fact: the mech- mation obtained was organized and reviewed anism of action of paracetamol, one of the world’s from the perspective of its contribution to an most widely used analgesics, is not fully known. understanding of the mechanism of analgesic action of paracetamol. Modern pain therapy Results: Paracetamol’s broad spectrum of analgesic and other pharmacological actions is pre- In contrast to the older view in which pain was cat- sented, along with its multiple postulated egorized according to its subjective ‘degree’ (using mechanism(s) of action. No one mechanism has terms such as mild, moderate, severe, etc.), the more been definitively shown to account for its anal- modern view categorizes pain according to its gesic activity. mechanistic ‘type’, i.e. according to the underlying What is new and conclusion: Further research is (patho) physiology (e.g. nociceptive, inflammatory, needed to uncover the mechanism of analgesic neuropathic, etc.) and the biochemical mediators action of paracetamol. The lack of this knowledge (e.g. prostaglandins, substance P, glutamate, etc.) (3). affects optimal clinical use and impedes drug In many painful conditions, the underlying injury is discovery efforts. actually multi-faceted and the pain is transmitted by multiple primary and secondary afferent pathways Keywords: acetaminophen, mechanism of action, and by a variety of neurotransmitters and modula- paracetamol tors (‘mixed’ pains). The pain can result from Received 11 September 2009, Accepted 12 October 2009 increased activity in excitatory pathways involving, Correspondence: R. B. Raffa, Temple University School of for example substance P, glutamate, etc. decreased Pharmacy, 3307 N. Broad Street, Philadelphia, PA 19140, USA. activity in inhibitory pathways involving, for Tel.: +1 215 707 4976; fax: +1 215 707 5228; example noradrenaline or serotonin (5-HT) or both e-mail: [email protected] mechanisms (4). In addition, the underlying pain *These authors contributed equally to this work.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd 617 618 K. Toussaint et al.

(patho)physiology may be time-variant, i.e. the type and toxicity of paracetamol and preclinical inves- can change from one to another because of the tigations of its mechanism of analgesic action. development of central or peripheral sensitization Several comprehensive reviews of the early litera- (or both) or other such phenomena. The modern ture were obtained, as well as an extensive collec- view of pain is better able to explain the otherwise tion of primary literature. In addition, valuable seemingly contradictory clinical observation that recent reviews of mechanism of action were avail- addition of a so-called ‘weak’ analgesic [such as an able, as well as data from one of the author’s (RBR) non-steroidal anti-inflammatory drug (NSAID) or laboratory. paracetamol] to a so-called ‘strong’ analgesic (such as an opioid) can sometimes achieve superior pain Analysis relief. According to the new view of pain, optimal pain treatment results from matching the type of Each of the sources was reviewed for its relevance pain with the drug that has the appropriate mecha- to the mechanism of analgesic action of paraceta- nism of analgesic action. Recent advances in the mol. In many cases, the information was obtained understanding of pain transmission pathways, from studies that did not have investigation of the genetic polymorphisms (5) and analgesic pathways mechanism as the primary outcome, but the results suggest that future pharmacotherapy might be able were determined to either reflect on the mechanism to target a patient’s unique pain with the fewest or to suggest further avenues of investigation. adverse effects. Until that time, clinicians are faced with the ‘analgesic challenge’ – namely, trying to Assessment treat pain with the currently available drugs. As novel, more targeted analgesics are awaited, the cli- Each item was evaluated for its relevance and nician can optimize treatment of pain by using strength of the evidence. In at least one case, an individual drugs, or combinations, which incorpo- author (authority) was contacted with a series of rate the appropriate mechanism(s) of action. questions that shed additional light on a particular mechanistic hypothesis that originally seemed to be only weakly supported. Implications of not knowing drug mechanism The search for a new drug that represents an RESULTS improvement of an existing drug such as paracetamol requires a mechanistic assay for screening com- Paracetamol is an aniline (a.k.a. phenylamine, pounds (derived from combinatorial chemistry, aminobenzene, C6H7N aromatic amine) derivative. molecular modelling or other source) or for testing It is an active metabolite of two other anilines the activity of such compounds. Without knowledge [Greek for ‘black’], acetanilide and phenacetin, and of the molecular target of the drug, it is impossible to this played a role in the history of its use. Acetan- setupsuchanassay.Likewise,withoutknowledgeof ilide was first synthesized in 1852 and serendipi- thedrug’smechanism, it isimpossibletodetermine if tously found to have antipyretic effects by Cahn the new drug works the same way as the existing one. and Hepp (6). While studying the effect of naph- Therefore, until the mechanism of action of paracet- thalene on intestinal parasites, Cahn and Hepp amolisknown,drug-discoveryeffortsarestalledand requested naphthalene from the local pharmacy, a great deal of effort and money will be expended but were inadvertently sent the incorrect material. studying and trying to address paracetamol’s They noted that the delivered substance (acetani- adverse effects rather than finding a replacement. lide) behaved differently than expected and had antipyretic properties. The mistake was soon real- ized and taken advantage of. Acetanilide was METHODS marketed as an antipyretic under the clever trade name ANTIFEBRIN. Unfortunately, acetanilide Identification of studies was soon discovered to be quite toxic (causing Computerized literature searches were conducted cyanosis due to methemglobinenia), prompting a using keywords related to clinical use, attributes search for a safer substitute.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 619

Phenacetin and paracetamol, both derivatives of and it was introduced to the UK in 1956 as pre- acetanilide, were studied during the 1880s and scription only PANADOL (7). Found to be ade- 1890s (7, 8). During this period, Hinsberg and quately safe at therapeutic doses and free of the Treupel (9) showed that paracetamol was as effec- gastrointestinal bleeding side effects associated tive as phenacetin as an antipyretic, but von Mer- with acetylsalicylic acid, paracetamol gained over- ing (7, 9) concluded that paracetamol was more the-counter status in the US in 1960 (12) and was toxic, so phenacetin began to be used. By the early added to the British Pharmacopoeia in 1963 (7). 1900s, phenacetin’s analgesic effects were recog- By the mid-1960s, paracetamol use increased nized and it began to be used as an analgesic for dramatically overtaking the use of acetylsalicylic mild to moderate pain in addition to its use as an acid (7). Combination products containing parac- antipyretic (7). etamol were also introduced at this time. In 1978, In the 1940s, Brodie and Axelrod at the National paracetamol sales surpassed those of acetylsalicylic Institutes of Health (NIH) (10, 11) and Smith and acid in the UK (8). Paracetamol gained further Williams (9) at St Mary’s Hospital in London, popularity in the 1980s when acetylsalicylic acid studied the metabolism of phenacitin and acetani- was associated with Reye’s syndrome in children lide. Both groups found that acetanilide and with viral illness, making it the antipyretic and phenacetin are metabolized to paracetamol (see analgesic of choice for children (9). Fig. 1). Paracetamol turned out to be mainly Currently, paracetamol is one of the most widely responsible for the antipyretic and analgesic effects used analgesics in both children and adults. It is of phenacitin and acetanilide, whereas another included in clinical guidelines for multiple pain metabolite, p-phenetidine, turned out to be conditions [e.g. (13, 14)]. responsible for much of the toxiciy. Paracetamol was ﬁrst marketed in the US in 1950 as a combi- APAP Pharmacokinetics nation product as TRIOGESIC, which also con- tained aspirin (acetylsalicylic acid) and caffeine (9). The pharmacokinetics of paracetamol have been TRIOGESIC was removed from the market 1 year well described in humans (15). Paracetamol later when it was erroneously associated with exhibits an oral bioavailability of 88%, a total body agranulocytosis. clearance of 5 mL ⁄ min ⁄ kg and a volume of distri- Paracetamol was reintroduced into the US mar- bution 0Æ8L⁄ kg. About 3% of the drug is excreted ket in 1955 as prescription only TYLENOL (9, 12) unchanged in the urine. Paracetamol is not highly bound to plasma proteins. The mean peak plasma concentration upon a 20 mg ⁄ kg oral dose is O O 20 lg ⁄ mL, and time to achieve this peak is on HN C CH HN C CH 3 3 average 0Æ33 h. Paracetamol crosses the blood– brain barrier. A 0Æ6 mg i.v. dose in rats achieves Acetanilide Phenacetin quantiﬁable brain concentrations that are 10–20% of blood levels (16). Highest brain concentrations

OC2H5 are observed in the frontal cortex at 15 min post-

12 dose, and in the cerebellum at 120 min post-dose. 34 A steady-state brain concentration of 5Æ8 lg ⁄ mL was reported in mice upon an i.v. infusion of O 12Æ1mg⁄ kg (with a loading dose of 46Æ3mg⁄ kg) NH HN C CH NH 2 3 2 (17). Human studies have evaluated paracetamol concentrations in the cerebrospinal ﬂuid (CSF), and Aniline p-Phenetidine successful PK-PD models have been generated with the assumption that CSF concentrations are OH OC H 2 5 closest obtainable data to the effect-site concentra- Paracetamol tions. Thus, a study with 40 mg ⁄ kg paracetamol Fig. 1. Chemical structures and pathways involved in nasogastric administration in paediatric patients the discovery of paracetamol. reported a Cmax in the CSF of 15 lg ⁄ mL; an EC50

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 620 K. Toussaint et al. for antipyretic effect was estimated with PK-PD was only effective in preventing hyperalgesia, not (pharmacokinetic–pharmacodynamic) modelling at reversing it. Direct injection into the brain (intra- 9Æ6 lg ⁄ mL (18). Similar CSF concentrations have cerebroventricular administration) suppresses been reported in independent studies (19). A meta- thermal and mechanical hyperalgesia (25). Direct analysis of human studies reported an APAP EC50 injection into the spinal cord (intrathecal adminis- of 4Æ63 mg ⁄ L for antipyresis and 9Æ98 mg ⁄ L for tration failed to reduce streptozotocin-induced analgesia (20). From these and numerous other hyperalgesia when given 1-month post-induction studies on paracetamol PK-PD, one can conclude (38). Paracetamol (1000 mg) given i.v. significantly that adequate paracetamol levels are achieved to reduced electricity-induced hyperalgesia in healthy support a central site of action. In addition, pos- young adults (32). tulated mechanisms of action must be consistent Several studies have also reported paracetamol with the levels of paracetamol actually achieved in to be effective in a variety of models of allodynia the central nervous system. [painful sensation from normally non-painful stimuli (24)] when given systemically or intrathecally to mice or rats (39–42) (Table 2). ANALGESIC SPECTRUM Paracetamol has been tested for the treatment According to a meta-analysis from the American of migraine headaches [e.g. (43, 44)], and it is Academy of Family Physicians, <1000 mg of oral widely used, but its effectiveness is not so notable paracetamol can reduce over 50% of mild to as to give particular insight into its mechanism of moderate acute non-specific type pain (21). This action. includes orthopaedic pain and tension headaches Likewise, there has not been substantial evi- (21), ankle sprain pain (22), acute low back pain dence showing paracetamol to be particularly (13) and multiple others. advantageous in treating menstrual pain. In a Paracetamol can be effective even against severe randomized, active-controlled, single-blind, paral- pain if it is administered intravenously. For exam- lel-designed study, subjects treated with a heat ple, in a repeated-dose, randomized, double-blind, wrap experienced a greater decrease in pain, placebo-controlled, three-parallel group study of cramping, fatigue and mood swings than did the orthopaedic surgery patients that compared 1000 mg paracetamol group (45). 1000 mg i.v. acetaminophen (at 6-h intervals over 24 h) to placebo (23), pain intensity (measured CLINICAL SPECTRUM using a 100-mm visual analogue scale) and pain relief (measured using a five-point verbal scale), As possible insight into paracetamol’s analgesic and rescue i.v. patient-controlled morphine use mechanism of action, it is instructive to examine its (time to use and quantity) were significantly spectrum of other pharmacological activities. Spe- decreased in the paracetamol group compared cifically, with regard to fever (and body tempera- with the placebo group. The median time to first ture), inflammation and platelets. rescue medication (patient-controlled i.v. morphine) was 3 h for paracetamol compared with Fever and body temperature <1 h for placebo and the amount of rescue morphine was significantly lower in the paracetamol It is well known that paracetamol is antipyretic. It group than in the placebo group. reduces fever in multiple species (46–51). A central Several studies have reported paracetamol to be site of antipyretic action against induced fever was effective in a variety of models of hyperalgesia demonstrated in rabbits by direct injection into the (enhanced sensitivity to pain) (24–36) (Table 1). By organum-vasculosum-lamina-terminalis (OVLT) systemic or local administration, paracetamol has located in the anterior wall of the third intracere- been reported to be antinociceptive against chem- bral ventricle (49). ical-induced hyperalgesia in mice and rats (26, 27, It is less well known that paracetamol can 29, 30, 34–37), formalin-induced pain-related also lower afebrile body temperature (47, 52–56) behaviour (30) and thermal hyperalgesia (33), (Table 3). For example, several studies, employ- although Bianchi (27) reported that paracetamol ing different methods and routes of administra-

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 621

Table 1. Test for analgesic activity of paracetamol against hyperalgesia

Subjecta Routeb Dose Study design Results Reference

Hyperalgesia mR (1) Noxious thermal stimulus. Tail In the absence of ischaemia, APAP had 25 i.c.v. hyperalgesia induced by 49 C water. no effect on latencies of ischaemia ⁄ tail 0Æ05–0Æ5mg⁄ kg Tail flick latency measured flick. In the presence of ischaemia, (2) Mechanical noxious stimulus. Tail APAP at minimum 0Æ2mg⁄ kg abol ischaemia induced by inflatable cuff ished reperfusion hyperalgesia. i.c.v. at the base of tail. Tail flick latency dosing 2–3 times less than systemic measured pretourniquet and post- dosing tourniquet release mR Chemical-induced hyperalgesia. Brewer’s Central hyperalgesia reduction at 26 oral yeast intraplantar injection. APAP 25 mg ⁄ kg. Peripheral hyperalgesia 25–100 mg ⁄ kg given 1 h prior yeast. Inflammatory reduction at 50–100 mg ⁄ kg and in hyperalgesia, edema and nociceptive creased in threshold of non- threshold measured inflammed paws. Paw edema and tail threshold not affected mR Chemical-induced hyperalgesia. Formalin APAP prevented hyperalgesia, but not 27 oral intradermal injection to tail. Effect on effective when hyperalgesia already 25 mg ⁄ kg hindpaw nociception threshold by established thermal stimulation measured mR Chemical ⁄ heat-induced hyperalgesia. Dynorphin level in frontal cortex 28 i.p. Opioid antagonists pretreatment. significantly decreased with APAP. 400 mg ⁄ kg APAP injected 15-min Naloxone reversed APAP effect on post-antagonists. Hyperalgesia hotplate, but not in formalin test triggered with formalin or hotplate. Dynorphin-A level measured by brain autopsy radio immunoassay mM Chemical-induced hyperalgesia. APAP showed dose-dependent 29 oral Intraplantar formalin, intrathecal antinociception in both phases (0–5; 10–300 mg ⁄ kg substance P and glutamate. Measured 20–40 min) in formalin only. duration (s) of nociceptive response Glutamate and substance P nociception reversed in phase 2 mR Chemical-induced hyperalgesia. APAP oral (300–400 mg ⁄ kg) reduced 30 oral Intraplantar formalin to rats. Oral nociceptive behaviour (biting ⁄ licking) 100–400 mg ⁄ kg APAP 40 min prior formalin, i.v. in both phases (0–5, 20–40 min) more i.v. 5 min prior, Intraplantar 30 min in phase 2, post-formalin. Same with 1mL⁄ kg prior, s.c. in the back 30 min prior and i.v. Intraplantar required very high intraplantar–20 mg ⁄ kg IT 10 min prior dose to inhibit behaviour i.t. 200 lg mR Chemical-induced hyperalgesia. Induce APAP at 300 mg ⁄ kg had very weak 31 oral arthritis and neuropathic pain in rats activity in reduction of thermal 100–600 mg ⁄ kg with Freund’s adjuvant intraplantar hyperalgesia and mechanical in hindpaw Mechanical allodynia, allodynia thermal and joint hyperalgesia mea sured. APAP 30 min prior measure using von Frey filaments and radiant heat source

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 622 K. Toussaint et al.

Table 1. (Continued)

Subjecta Routeb Dose Study design Results Reference mR Chemical-induced hyperalgesia. APAP given 4 weeks after 38 i.t. 75 mg ⁄ kg streptozotocin intraperito streptozotocin. APAP did not fl 1–7 mg neal to mimic diabetic neuropathy. decrease hyperalgesia Measured hind-paw-withdrawal threshold from anesthesimeter tip (hyperalgesia developed in 3–6 weeks) fR Thermal-induced hyperalgesia. Hindpaws APAP inhibited the drop of heat 33 i.p. of rats immersed in water bath tem threshold dose-dependently 30 mg ⁄ kg perature 51 C for 20 s. Hyperalgesia detected after 60 min. APAP 20-min post-injury. Noxious heat threshold measured mR Freund’s adjuvant-induced inflammatory APAP increased nociceptive threshold. 34 oral hyperalgesia. Adjuvant intraplantar to APAP was weakest 65 mg ⁄ kg 2·d left hindpaw to induce PGE2 and among other investigative drugs TNF alpha release in spinal cord. including tramadol. APAP decreased Randall-Sellito-paw-withdrawal test PGE2 release compared with measured withdrawal latency. adjuvant alone. APAP alone did not Measured PGE2 and TNF alpha in decrease TNF alpha level compared spinal fluid with placebo (P =0Æ007) mR Chemical-induced hyperalgesia. i.t. Sub Oral APAP decreased magnitude of 35 oral stance P-induced hyperalgesia. APAP hyperalgesia, thermal response 300 mg ⁄ kg 60 min prior substance P latency, and spinal PGE2 level. APAP i.t. i.t. suppressed hyperalgesia 10–200 lg dose-dependently. mR Chemical-induced hyperalgesia. Carra APAP reversed hyperalgesia and 36 i.p. geenan (CG) intraplantar to hind increased threshold above basal level. 500 mg ⁄ kg paw. APAP intraplantar to hind paw APAP also raised threshold in non- 30 min prior or 2 h post-CG inflammed paw. Effect reversed by naloxone H Electric noxious stimulation pain in APAP significantly reduced 32 i.v. duced electrically and secondary hyperalgesia 1000 mg hyperalgesia induced by von Frey filaments aM, mouse; R, rat; Rb, rabbit; H, human; f, female; m, male. bi.p., intraperitoneal; s.c., subcutaneous; i.v., intravenous; i.c.v., intraverebroventricular; i.t., intrathecal; supp, suppository; APAP = APAP; ASA, aspirin; OVLT, organum vasculosum of the lamina terminalis. tion, have shown that paracetamol produces reported with a single 1000 mg dose of paraceta- hypothermia in mice when the drug is adminis- mol in patients with subarachnoid haemorrhage or tered intravenously (160 mg ⁄ kg, 2Æ5 C decrease) head trauma (57) and oral or suppository paracet- (47), intraplantarily (100–300 mg ⁄ kg with 0Æ4–2 C amol given 6 g daily to stroke patients lowered decrease respectively) (53) or intracerebrovascu- afebrile body temperature by 0Æ3 C (56), a decrease larly (dose, 0Æ25 C decrease) (52). attributed with reducing relative risk by 10–20%. The data in humans are mixed. Effective and However, oral paracetamol (650–1300 mg) was rapid reduction in brain temperature (2 C) was reported to not lower core body temperature in

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 623

Table 2. Test for analgesic activity of paracetamol against allodynia

Subject Route Dose Study design Results Reference

Allodynia mM Chemical-induced allodynia. Formalin Rubbing behaviour fl in 1st phase 37 s.c. injected to upper lip. APAP injected (0–3 min) and 2nd phase 25–400 mg ⁄ kg in the neck 20 min prior formalin. (15–39 min) after formalin. Rubbing frequency measured Dose-dependent only during 2nd phase mR Chemical-induced allodynia. APAP ED50 = 1100 lM ⁄ kg in 42 i.v. Vincristine via osmotic pump mechanical allodynia. Analgesic 1–5 mL ⁄ kg 30 lg ⁄ kg ⁄ day for 14 days. efficacy 72%. Therapeutic index Measured thermal, mechanical 12, better performance than nociception, cold and mechanical ibuprofen, aspirin and celecoxib allodynia mM Cancer-induced allodynia. Osteolytic Oral APAP decreased minimum 41 oral murine sarcoma cells injected into number of pain related responses 0Æ3, 3, 30, 300, 3000 mg ⁄ kg distal femur. APAP 2 weeks post- in dose-dependent manner from tumour-implant. Level of cancer 0Æ3 to 3000 mg ⁄ kg. Plateau effect at pain measured 3000 mg ⁄ kg mR Post-surgical (left sciatic nerve APAP suppressed tactile allodynia 39 i.t. exposition and gut ligatures). APAP and thermal hyperalgesia 20 mM IT to hindpaw. Von Frey filaments or hot plates (28 C) measured withdrawal threshold mR Post-surgical (partial sciatic nerve APAP produced dose-dependent 40 s.q. ligation). Induced neuropathy. anti-allodynic effect and 100 mg APAP SQ to hindpaw 15 min prior antihyperalgesic effect von Frey test of mechanical allodynia

See Table 1 for abbreviations. normothermic cardiac (54) or stroke (3900 mg intrathecally at 200 lg ⁄ kg (63) reduced inflam- daily) patients (55). Thus, paracetamol-induced matory pain, but had no effect on edema and in hypothermia appears to be clinically insignificant a randomized, double-blind, placebo-controlled when given at therapeutic daily dose <4 g. trial no significant improvement was seen in the paracetamol (1000 mg four times daily) group when assessed 2 and 12 weeks into treatment Inflammation (65). The relatively poor anti-inflammatory effect Paracetamol has been reported to suppress var- of paracetamol is a characteristic distinction from ious inflammation-related substances in animals the NSAIDs and might be a reflection of differ- [e.g. (58–61)] and in inflamed dental tissue ent mechanism of action. (1000 mg pretreatment and 4000 mg post-surgery in patients with two-third molar extractions) Platelet aggregation (61), but paracetamol is generally not considered to display very effective anti-inflammatory Because of the common impression that paraceta- action in the clinical setting (58–64) (Table 4). mol lacks clinically relevant antiplatelet action, it is For example, paracetamol given i.p. or orally often used to avoid the bleeding risk associated at 100 mg ⁄ kg (62), i.v. at 100–300 mg ⁄ kg or with aspirin and other NSAIDs.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 624 K. Toussaint et al.

Table 3. Test for analgesic activity of paracetamol against hypothermia

Subject Route Dose Study design Results Reference

Hypothermia mM Mice were pretreated with APAP Significant temperature decrease APAP 52 i.c.v. hepatotoxic inducers. Crossover was 20 min post-ICV and lasted for 10 min. An 500 mg ⁄ kg done. Brain, liver, blood samples approximate drop of )0Æ25 C. Correlation collected. Rectal temp recorded. Brain between APAP level and tissue was most and liver APAP levels compared with significant in plasma and brain, not in liver. to degree of hypothermia Hypothermia is induced by APAP parent drug in brain and not metabolites in liver mM Basal body temperature and brain One hour post-APAP: 100, 200, 300 mg ⁄ kg 53 intraplantar PGE2 levels were measured before lowered basal body temperature by 0Æ4, 0Æ8 100–300 mg ⁄ kg and after administration of APAP in and 2 C respectively. No hypothermic control mice and COX-1 ⁄ COX-2 KO action is seen in the KO mice mice PGHS-1 weight, Fever induced by LPS (from Escherichia APAP pre-LPS: <160 mg ⁄ kg – no effect, 47 PGHS-1 KO coli). APAP injected into wild-type >160 mg ⁄ kg –2Æ5 C drop in 60 min. LPS mice and knock-out mice preplacebo and induced 1 C increase similar to APAP- i.v. post-placebo ⁄ LPS administration. untreated 80, 120, 140, Core temp recorded before and after APAP post-LPS: (80 mg ⁄ kg) temperature 160 mg ⁄ kg LPS. Blood and brain samples col return to baseline after 30 min then re lected and monitored hourly bounded to febrile state after 30 min, (160 mg ⁄ kg) temperature dropped 4 C below baseline and lasted until end of experiment. Brain and plasma PGE2 level in placebo and LPS mice unchanged by APAP 160 mg ⁄ kg H Study patients underwent APAP did not affect core temperature when 54 oral hypothermic cardiopulmonary by patients are normothermic. APAP did not 650, 1300 mg pass APAP given to all study patients affect onset of hyperthermia experience hypothermia, normothermia and hyperthermia during post-operative period H Stroke-induced hyperthermia. Patients For ischaemic stroke patients, hypothermia is 55 oral were randomized to receive APAP. more prevalent in APAP patients and the 3900 mg CBT measured every 30 min for 24 h amt of time of hyperthermia is reduced. However, compared with placebo, difference is statistically insignificant ()0Æ16 C) H Acute ischaemic stroke hyperthermia. (two High dose APAP decreased baseline 56 oral, supp randomized, double-blinded, phase II temperature by 0Æ27 Cin24h.A0Æ3 C drop 3000–6000 g trials). High dose APAP with low correlates to 10–20% risk reduction. Fewer dose APAP vs. placebo in patients patients in high-dose APAP suffered from with acute ischaemic stroke. High severe stroke than in high-dose ibuprofen dose APAP with ibuprofen (2400 mg ⁄ day) with placebo in patients with acute ischaemic stroke. Majority of patients received txs within 12 h onset. NIHSS measured severity of stroke

See Table 1 for abbreviations.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 625

Table 4. Test for antiinﬂammatory activity of paracetamol

Subject Route Dose Study design Results Reference

Inflammation mR Chemical-induced polyarthralgia. i.v.: No change in number of Fos-LI 59 i.v. Intradermal Freund’s adjuvant to tail. neuron with APAP 150 mg ⁄ kg. Oral 150 mg ⁄ kg Pressure applied to ankle joint after (10 days): decrease in number, no oral 3 weeks. Fos-LI neurons number in symptoms improvement. Oral (14 days): 150, 500 mg ⁄ kg lumbar spinal cord measured. no change in number Number correlates to signs ⁄ After 3 weeks Oral (14 days) 41% symptoms of polyarthritic pain decrease in number in gray matter, 38% dorsal horn, 46% in ventral horn mR Chemical-induced inflammation. APAP reduced more number of c-Fos 58 i.v. Carrageenixn intraplantar to paws. neurons at higher dose than at lower 75–150 mg ⁄ kg APAP 15 min prior carrageenin. dose. More marked effect in deeper Number of c-Fos-LI neurons laminae than superficial at the dorsal measured horn mR Chemical-induced inflammation carra Non-nitrated APAP – no edema 62 i.p. geenin i.p. (edema), 150 lL reduction, minimum of 100 mg ⁄ kg for 2Æ5–100 mg ⁄ kg (hyperalgesia) into hindpaw of rats. antinociception. ED50 = 62 Nitrated oral Edema size ⁄ volume and nociception APAP– reduced edema, minimum of 25–100 mg ⁄ kg threshold measured 2Æ5mg⁄ kg for antinociception. ED50 = 44 mR Chemical-induced hyperalgesia. Give i.t.: APAP increased vocal threshold dose- 63 i.v. APAP 2 h post-carrageenin. dependently, highest at 200 lg ⁄ kg. 100–300 mg ⁄ kg Nociception assessed with i.v.: APAP increased vocal threshold simi i.t. mechanical noxious stimulation. larly across the three doses. APAP failed 50, 100, 200 lg Vocalization threshold (squeak) and to reduce edema volume if given >2 h edema volume measured before and post-injury after carrageenin and treatment H Osteoarthritis Pilot Study. APAP APAP withdrawal resulted in pain score of 64 oral compared with NSAIDs in reduction 18 and after resuming APAP it dropped to 4000 mg of anti-inflammation in OA patients. 9 (50% decrease). Both APAP and WOMAC osteoarthritis index pain NSAIDs have similar outcomes score used (5–25). Total effusion volume measured during knee pain with each APAP withdrawal ⁄ resumption H Surgical removal of 2 impacted mandibular APAP suppressed PGE2 significantly after 61 oral third molar. Pretreat with APAP 1 hour 80–180 min. APAP had nearly no effect on 1000 mg prior surgery. Continue to treat post- TXB2 suppression surgery same dose q6h. Microdialysis biopsy done to measure PGE2 (COX-1, 2) and TXB2 (COX-1) fR Chemical-induced inflammation. Zymosan APAP significantly increased exudates TNF 60 tube feed induced. Air pouch created on back alpha level and not exudates IL-1 beta 100 mg ⁄ kg APAP given 1 h later. Exudates (IL-1 beta, level TNF alpha, PGE2) in pouch collected and measured. APAP plasma conc. measured

See Table 1 for abbreviations.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 626 K. Toussaint et al.

There is some evidence of antiplatelet activity of separation (larger Therapeutic Index) might be paracetamol in human blood samples using in-vitro possible. and ex-vivo assays [e.g. (66–71)], but other studies More speculatively, paracetamol has been asso- suggest a lack of antiplatelet action (66–74) ciated with asthma. Because the prevalence of (Table 5). Such an action, when present, is believed asthma in developing countries (76) increased in to be reversible (shorter acting), in contrast to the parallel with increased paracetamol use, a causal irreversible action of aspirin and NSAIDs (66, 70). relationship has been postulated. The Peer Educa- At least two recent clinical trials report that par- tion in Pregnancy Study (76), which examined acetamol did not interrupt platelet aggregation women at high risk of having a child with asthma, when given at 1000 mg (73) or 3000 mg i.v. (74). reported that use of paracetamol by these women Paracetamol might (68) or might not (72) interact during mid to late pregnancy was significantly with NSAIDs on this endpoint. related (after control for potential confounders) to The relatively poor inhibition of platelet aggre- wheezing in the offspring during the first year of gation by paracetamol is another characteristic life. The Danish National Birth Cohort, a popula- distinction from the NSAIDs that might be a tion-based study of 100 000 newborns recruited reflection of a different mechanism of action. between 1996 and 2003, reported that prenatal exposure to paracetamol was associated with wheezing and asthma in the offspring at 18 months TARGETS FOR IMPROVEMENT and 7 years of age. The use of paracetamol during pregnancy was associated with an increase in rate Efficacy and potency of hospitalizations for asthma up to 18 months of Despite the demonstration of some degree of age and the use of paracetamol during the first effectiveness against a very wide variety of pains, trimester was associated with increased severity of paracetamol’s common clinical application is pri- asthma attacks at 7 years of age (77). marily limited to use against types of pain that are The Nurses Health Study reported that women generally described as mild to moderate. It is not who used paracetamol more than 1 day ⁄ month had clear whether this limitation on clinical analgesic a significantly higher risk of developing hyperten- efficacy is imposed by the drug’s level of intrinsic sion (78) (it is not clear why the women were taking analgesic activity ⁄ action or whether the limitation paracetamol every day, but it suggests that there was is imposed by not being able to administer higher already anunderlying health condition). Asubgroup doses (because of the fear of inducing serious of the Physicians Health Study (79) examined the risk adverse effects). A non-opioid, non-NSAID anal- of hypertension in 8229 male physicians (ages 53–97) gesic with greater clinical efficacy than paracetamol without prior hypertension who reported analgesic would be highly desirable. use. After adjusting for potential confounding vari- ables, there was no significant increase in hypertension (defined as BP 140 ⁄ 90 mmHg) associated with Adverse effects use of acetaminophen. Paracetamol displays an excellent safety profile Whether or not asthma or hypertension is rela- within its usual therapeutic range. However, the ted to paracetamol use, high-dose hepatotoxicity is major negative aspect of paracetamol is its ability sufficient reason to desire an alternative. Unfortu- to induce serious, even fatal, hepatotoxicity above nately, without knowledge of the mechanism of the usual therapeutic range. The mechanism of paracetamol’s analgesic action, it cannot be known this toxicity is well known and results from the if the maximum possible separation between ther- depletion of endogenous glutathione and sub- apeutic and toxic doses has been achieved already sequent shunting of paracetamol metabolism from in paracetamol. benign to toxic pathways (75). The risk is greater when the liver is compromised by disease or POSTULATED MECHANISMS OF ACTION excessive alcohol use. Importantly, the mechanism of paracetamol’s toxicity appears to be distinct The major mechanisms that have been proposed to from its mechanism of analgesia, so that greater account for the analgesic action of paracetamol are

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 627

Table 5. Test for antiiplatelet activity of paracetamol

Subject Route Dose Study design Results Reference

Antiplatelet Human Aspirin inhibition of COX and platelet High dose APAP did not show interference 72 blood function. Measured COX-inhibition with ASA-induced inhibition of COX or In-vitro interference by APAP platelet function 1, 3, 6 mM Human In-vitro: collagen, epinephrine, arachidonate, Arachidonate-induced aggregation inhibited 66 blood ionophore A23187 were added to induce by low dose APAP reﬂected by decreased In-vitro platelet aggregation TXB2 production. APAP did not inhibit 0–1Æ3mM Ex-vitro: blood collected prior and post- norepinephrin-induced aggregation Ex-vivo APAP administration to measure changes Ex-vivo platelet aggregation and 14 C-5HT 650, 1000 mg in platelet aggregation and secretion of secretion reduced only when APAP reached 14 C-5HT high plasma concentration APAP may be a reversible platelet aggregation inhibitor H Randomized, double-blinded and cross-over. Adrenaline-induced platelet aggregation 70 i.v. Venous blood collected at 0 h prior APAP, inhibited by APAP for 2 h compared with 30 mg ⁄ kg and at 2, 24 and 48 h post-APAP. PT, aPTT, ketorolac for 24 h. Platelet dysfunction lasted factor V ⁄ VII, Hg, Haematocrit, Platelet, and longer with ketorolac than with APAP APAP bleeding time, TXB2 were measured. APAP caused reversible inhibition of platelet aggregation i.v. compared with ketorolac and decrease in maximal TXB2 level H Arachidonate-induced platelet aggregation. After 90 min, plasma TXB2 level shows 68 i.v. Platelet function measured after 5, 90 min Diclofenac alone vs. APAP + Diclofenac: 15 mg ⁄ kg and 24 h. TXB2 levels measured 44Æ1 vs. 10Æ6(P <0Æ003) Difference abolished after 22 h (P <0Æ90). Similarly in ADP level Human Lipopolysaccharide (LPS)-induced PGE2 and APAP inhibited PGE2 and TXB2 production 71 blood TXB2 production. APAP at different conc. dose-dependently, 44 and 94 respectively. Ex-vivo added to blood samples of each individual At therapeutic plasma conc. 100–300 lM 0Æ05–150 mM inhibited more PGE2 than TXB2 H Randomized, placebo, double-blinded study. 10 min post-i.v. at 15 mg ⁄ kg, arachidonic 67 i.v. Arachidonate-induced platelet aggregation. acid level and TXB2 levels dropped. APAP 15, 22Æ5, TXB2, arachidonate, and APAP plasma dose-dependently inhibited platelet 30 mg ⁄ kg concentration measured aggregation H Platelet function (photometric aggregometry) APAP at high conc. prolongs PFA-100. 69 i.v. and TXB2 level measured APAP demonstrated dose-dependent TXB2 15, 22Æ5, inhibition from conc. 10 lg ⁄ mL and up 30 mg ⁄ kg H Surgically induced pain (& bleeding). APAP, APAP alone did not affect platelet 74 i.v. diclofenac given prior surgery. Photometric aggregation instead increase in LFTs. Both 3000 mg aggregometer measured platelet APAP and diclofenac decreased TXB2 in aggregation, APAP plasma level, LFTs, 1 h but not direct platelet aggregation and TXB2 also measured H Single rescue dose of APAP has no effect on 73 i.v. platelet aggregation indicated by signs of 1000 mg bleeding

LFT, liver function test. See Table 1 for abbreviations.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 628 K. Toussaint et al. the subject of a recent comprehensive review and COX-1 details can be found in it (80). There is evidence for and against each proposed mechanism. Our pur- The antinociceptive (animal equivalent of analge- pose is to present a succinct overview of the sic) activity of paracetamol is diminished in COX-1 available evidence and to identify what questions knockout mice (mice that are genetically modified remain to be addressed. such that they do not produce COX-1), but not in COX-2 knockout mice (83), suggesting that the analgesic activity of paracetamol requires COX-1 Cyclooxygenase (EC 1.14.99.1, COX) inhibition but not COX-2. This is seemingly strong evidence The first well known proposal of a mechanism of in support of inhibition of COX-1 as the mechanism action for paracetamol was made by Sir John Vane, of analgesic action of paracetamol. In the same who discovered that the mechanism of action of study, it was shown that unlike diclofenac, which aspirin involved inhibition of cyclooxygenase (81). reduces prostaglandin synthesis both centrally and Ever since, paracetamol has periodically been peripherally, paracetamol only reduces prosta- proposed to inhibit one or more of the cyclo- glandin synthesis centrally. Even if paracetamol oxygenase (prostaglandin synthase, PGHS) inhibits peripheral COX-1, it is argued that it has enzymes COX-1 (PGHS-1), COX-2 (PGHS-2) and too low a potency to produce pharmacological COX-3 (PGHS-1b). COX enzymes catalyse conver- effects (86). In contrast to the antinociceptive end- sion of arachidonic acid to prostanoids and other point, hypothermic and antipyretic activity of chemical mediators involved in inflammation, paracetamol is not affected in COX-1 knockout fever, pain, platelet aggregation and mucous pro- mice (47), suggesting that these effects of paracet- duction in the gastrointestinal tract. COX-1 is con- amol are not related to COX-1. stitutively active. COX-2 is inducible and quickly However, the argument that paracetamol inhib- upregulated in areas of inflammation and during its COX-1 centrally, but not peripherally, raises fever. Inhibitors of COX-1 and COX-2 (aspirin and some challenging questions about distribution to other NSAIDs) inhibit pain, inflammation and the CNS vs. the periphery (Table 6) and is this fever (82). At some point, each of the three COX different from the NSAIDs; the local conditions enzymes has been proposed to be the target of (e.g. pH) at the sites of distribution within the CNS; paracetamol’s analgesic action. whether there is a difference between ‘central’ and Paracetamol generally lacks the clinically mean- ‘peripheral’ COX-1; and the site of paracetamol’s ingful anti-inflammatory and antiplatelet effects hypothermic and antipyretic effects. These displayed by NSAID COX inhibitors. A common questions and others make it difficult to conclude hypothesis to explain this difference is that parac- that paracetamol’s mechanism of action is direct etamol acts centrally as a COX inhibitor, whereas inhibition of COX-1. the other COX inhibitors act both centrally and peripherally (83). Support for this theory includes COX-2 evidence that paracetamol inhibits the conversion of arachidonic acid to PGE2, PGF2 and thrombox- There have been suggestions that COX-2, but not ane-A2 in microglia exposed to lipopolysaccharide COX-1 or COX-3, is the isozyme involved in the (84) at concentrations 3-fold lower in microglia antipyretic and hypothermic properties of parac- than in peripheral macrophages; evidence that etamol, as there is no loss of antipyretic activity in paracetamol is more potent centrally than periph- COX-1 knockout mice (47). A recent study in erally as a COX inhibitor. However, paracetamol humans reports that paracetamol inhibited COX-2 does not affect 24-h body temperature elevation in ex vivo whole blood samples to a comparable during the luteal phase of the menstrual cycle in extent as NSAIDs and COX-2 specific inhibitors. humans, a process thought to be prostaglandin An affect of COX-2 inhibitors that paracetamol mediated. This suggests that paracetamol’s effect does not display is alteration of fluid balance in the on fever ⁄ temperature does not involve inhibition kidneys, which may point to a mechanism of action of prostaglandins (and so might not its analgesic separate from selective COX-2 inhibitors (82). effect) (85). However, despite the intense interest in COX-2 as

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 629 an analgesic mechanism following the discovery of although it has been shown to be an active COX VIOXX and other ‘selective COX-2 inhibitors’, the protein in canines, the ‘COX-3’ gene sequence of connection to paracetamol’s analgesic action has rats appears to be subject to a frame-shift mutation, not been established. For this postulated mecha- resulting in the transcription of a protein that is nism, a difficult question to answer is: if paraceta- dissimilar to that of COX-1 or COX-2 in both mol shares the same mechanism of analgesic action genetic sequence and function. as the COX-2 inhibitors, why does it not have the However, COX-3 does not mediate antinocicep- same adverse effect profile? tion in rats, COX-3 is not detected in humans, and the low expression level and kinetics make it is unlikely that COX-3 has clinical relevance for par- ‘COX-3’ acetamol’s mechanism of analgesic action (88). A cyclooxygenase enzyme identified in canines was originally thought to be encoded by a separate Peroxidase gene than either COX-1 or COX-2 and thus repre- sented a new COX isozyme, named ‘COX-3’ (87). It has been proposed that rather than inhibiting This new enzyme was shown to be inhibited by COX enzymes directly, as do the NSAIDs, parac- paracetamol, implying that it might be involved in etamol might inhibit the enzymes indirectly. This paracetamol’s mechanism of analgesic action. theory is based on the fact that paracetamol is a Because COX-3 is a variant of COX-1, most evi- phenol, and phenols are powerful reducing agents. dence supporting a central COX-1 mechanism of According to this view, paracetamol acts as a action also supports the COX-3 theory. However, reducing agent to inactivate COX enzymes by

Table 6. Pharmacokinetic data for paracetamol

Reference Study design Results

15 Human Plasma Cmax is 20 lg ⁄ mL after a 20 mg ⁄ kg p.o. dose 18 Human, paediatric; Cerebrospinal ﬂuid (CSF) ⁄ plasma 40 mg ⁄ kg administered nasogastric C ratio = 1Æ18, probably because of

low CSF V; CSF Cmax =0Æ1 mmol ⁄ L=15 lg ⁄ mL; time lag in CSF C; meta-analysis of other data showed plasma C0Æ06–0Æ13 mmol ⁄ L (9–20 lg ⁄ mL) associated with antipyretic effect; PD modelling showed CSF C closer to effect C, but still a lag; EC50 = 0Æ064 mmol ⁄ L= 9Æ6 lg ⁄ mL 20 Meta-analysis, human studies EC50 = 4Æ63 mg ⁄ L for antipyresis EC50 = 9Æ98 mg ⁄ L for analgesia 19 Human, paediatric with Cplasma 0–33 mg ⁄ L CCSF 0–21 mg ⁄ L scaling to a 70 kg body; 30–40 mg rectally 16 Rat 0Æ6mg⁄ rat i.v. APAP crosses the BBB Blood conc. 5- to 11-fold higher than brain; highest brain C seen in cerebellum at 120 min and frontal cortex at 15 min post-dose 17 Mice (ddY male), i.v. infusion Plasma Css = 8Æ7 lg ⁄ mL, brain LD 46Æ3mg⁄ kg and MD 12Æ1mg⁄ kg Css = 5Æ8 lg ⁄ mL for target plasma C 10 lg ⁄ mL

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 630 K. Toussaint et al. converting them to their inactive oxidized forms. cells. Another subtype is inducible (iNOS) and Specifically, paracetamol’s reducing capability plays a role in inflammatory processes (96). G might disrupt the tyrosyl radical step of the COX L -nitro-L-arginine (L-NO-ARG), a non-specific pathway (89). inhibitor of NOS subtypes, 7-Nitroindazole (7-NI), Another mechanism of decreasing peroxide tone a specific inhibitor of nNOS and L-N6 (1-Imino- can be through scavenging and reducing free per- ethyl)lysine (L-NIL), a specific inhibitor of iNOS, all oxides such as peroxynitrate. Paracetamol pos- potentiate paracetamol antinociception (97), sug- sesses similar peroxide scavenging potency as gesting that paracetamol might be an inhibitor of other known reducing agents such as uric acid and NOS. L-NO-ARG and 7-NI also potentiate the antioxidants such as ascorbic acid (vitamin C) (90). effects of paracetamol when administered intra- This mechanism of action might be counteracted thecally, but L-NIL does not, suggesting that iNOS (overwhelmed) in the presence of high concentra- may be involved only in any peripheral actions of tions of oxidative peroxides, often released in areas paracetamol (98). of inflammation (91). This hypothesis is supported Paracetamol has been reported to lack direct by evidence that paracetamol is able to efficiently inhibitory effect on cNOS or iNOS in vitro (99), but eliminate excessive peroxide tone caused by low- perhaps the effect is indirect, as paracetamol dose peroxynitrate, but not peroxide tone caused inhibits expression of the iNOS gene in response to by high-dose peroxynitrate, by H2O2 or by tert- lipopolysaccharide and interferon gamma in RAW butyl-OOH (90). This suggests that paracetamol 264.7 macrophages (96). may inhibit mild to moderate peroxide stimulation of PGHS, but is overwhelmed at higher concen- Cannabinoid trations or by different types of peroxides that are present in inflammatory responses. Another study The discovery that a receptor mediates the effects found that paracetamol inhibits COX-2 in intact of cannabinoids (marihuana-like compounds) cells at concentrations far below the known IC50 in prompted a search for the endogenous ligands humans, but does not affect COX-2 in broken cell (endocannabinoids) (100). This search led to preparations. This effect can be completely blocked anandamide (101), which is the ethanol amide of by an increase in intracellular hydroperoxides, arachidonic acid. Further research led to the rec- further supporting the theory the inactivation of ognition that the endocannabinoid system is evo- paracetamol by overwhelming peroxide tone (92). lutionarily old, occurring in invertebrates as well as The hypothesis that paracetamol produces its in vertebrates (see (102) and references therein). analgesic effect by being a phenol and acts as a Endocannabinoids are involved in, and exogenous reducing agent raises the question: are all, or at least cannabinoid substances can modify, a variety of some other, phenols and reducing agents analgesic? physiological processes, including pain, motor activity, cognitive function, sleep and appetite (102). In addition, the endocannabinoid system is Nitric oxide synthase involved in ‘crosstalk’ with a variety of other Inhibition of the enzyme nitric oxide synthase receptor systems, including 5-HT (serotonin),

(NOS) has been another hypothesized mechanism NMDA and vanilloid (TRPV1) (102). of analgesic action of paracetamol. Nitric oxide The endocannabinoid system has recently been (NO) is produced in response to activation of the proposed to be involved in the mechanism of NMDA (N-methyl-D-aspartate) receptor and it analgesic action of paracetamol. Paracetamol itself ampliﬁes neuronal activity and facilitates nocicep- does not bind to cannabinoid receptors (103, 104), tion (93, 94). Inhibition of NO synthesis can atten- but one of its metabolites displays cannabinoid-like uate nociception, depending on the pain stimulus activity (105). Accordingly, paracetamol could (95). There are several subtypes of NOS. One sub- activate the endocannabinoid system by acting as a type is constitutively active (cNOS) and can be pro-drug. further divided into neuronal NOS (nNOS), located Speciﬁcally, paracetamol is deacetylated to form in the central and peripheral nervous systems, and a primary amine, p-Aminophenol, which subse- endothelial NOS (eNOS), located in endothelial quently undergoes conjugation with arachidonic

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 631 acid in the brain to form N-Arachidonoylphenol- entering the spinal cord. These descending projec- amine (AM404) (105). The conjugation of arachi- tions exert an inhibitory (analgesic) effect on the donic acid with p-Aminophenol is catalysed by the incoming pain signal before it is transmited to enzyme FAAH (fatty acid amide hydrolase). Inhi- higher CNS centres. bition of FAAH has been reported to suppresses Although paracetamol does not have affinity for the antinociceptive effect of paracetamol in mice 5-HT receptors or neuronal 5-HT reuptake sites (104), suggesting that paracetamol’s analgesic (103, 118–120), there is evidence to suggest an action is related to AM404 production. As AM404 indirect mechanism. Paracetamol administration lacks significant affinity for cannabinoid receptors, (100–400 mg ⁄ kg p.o. and 200–400 mg ⁄ kg i.p.) the interaction must be indirect. The nature of such increases 5-HT levels in various regions of (rat) a mechanism was suggested when AM404 was brain (cortical, pontine, hypothalamus, striatum, noticed to have structural similarity to agonists at hippocampus, brain stem) (121–123) and sub- the vanilloid receptor (106) and paracetamol was sequent down regulation of 5HT2A receptors (121, shown to have binding affinity for vanilloid 122, 124, 125). When the spinal 5-HT pathway is receptors in human cells (107). Further research has lesioned in rats, using 5,6-Dihydroxytryptamine demonstrated that AM404 is an activator of the (5,6-DHT) injected intrathecally, paracetamol- vanilloid subtype 1 receptor (TRPV1 or previously induced antinociception in the formalin test is known as VR1; a proven CB1 receptor agonist), and reduced, whereas lesioning the noradrenergic an uptake inhibitor of the endocannbinoid pathway, using 6-Hydroxydopamine (6-OHDA) anandamide (108, 109). TRPV1 itself is involved in has no effect (126) on paracetamol antinociception. pain (110) and perhaps in thermoregulatory (111) Likewise, depletion of 5-HT in cortical and pontine pathways. Acetaminophen’s analgesic properties regions (12% and 19% of baseline) using p-Chlor- are blocked by the administration of CB1 receptor ophenylalanine significantly decreased paraceta- antagonists at doses that inhibit known CB1 mol-induced antinociception in rats in the hot-plate receptor agonists (104, 112). This provides further test, shown by a decrease in pain threshold from evidence for the cannabinoid hypothesis. 32Æ8% to 11% maximum possible effect (121). In addition to their antinociceptive action, Pharmacological approaches have yielded less cannabinoids also reduce body temperature in rats consistent results. Tropisetron, a 5HT3 receptor (113). However, administration of a CB1 receptor antagonist, when administered IT inhibits the an- antagonist does not change paracetamol’s hypo- tinociceptive effects of paracetamol in various pain thermic effect (Scott Rawls, PhD, Temple Univer- models, suggesting 5HT3 receptor involvement (63, sity School of Pharmacy, personal communication). 119, 127, 128). However, granisetron and ondase-

If paracetamol has a signiﬁcant cannabinoid tron, also 5HT3 receptor antagonists, when given IT component, then it is reasonable to expect it to or SQ, failed to block paracetamol antinociception produce effects similar to those of cannabinoid in rats (118, 120, 129); 5-HT3 receptor antisense compounds. In an effort to support a cannabinoid deoxynucleotides also failed to inhibit acetamino- hypothesis, it has been claimed that paracetamol phen antinoception (120). can produce feelings of relaxation, tranquility and Although these results seem counterintuitive, instill a feeling of well-being (114, 115). However, tropisetron, unlike granisetron and ondansetron, to date these have not been supported by objective has afﬁnity for 5HT1B, 5HT2A, 5HT2C receptors in trials and are currently considered anecdotal (116). addition to 5HT3 receptors (129), leading to studies involving these receptor subtypes. One study, using rats and the paw pressure test, concluded 5-HT (5-hydroxytryptamine, serotonin) acetaminophen antinociception is blocked by IT

There is substantial evidence that paracetamol’s penbutolol (5HT1B antagonist), ketanserin (5HT2A mechanism of analgesia in some manner involves antagonist) and mesolergine (5HT2C antagonist), the descending serotonergical pathway. 5-HT but not WAY100635 (5HT1A antagonist) (129). neurons, largely originating in raphe nuclei located There is conﬂicting data about the involvement of in the brain stem (117, 118) send projections down 5HT1A receptor as WAY100635 has also been to the spinal cord that synapse on afferent neurons shown to inhibit acetaminophen antinociception

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 632 K. Toussaint et al.

(30) and to increase acetaminophen antinociception analgesics (NSAIDs) was discovered. It was natural (117). The latter studies utilized different pain and reasonable to postulate that paractamol’s assays (paw pressure and hot plate respectively) mechanism of analgesic action is the same. The first and species (mouse or rat). postulate, inhibition of COX-1, is difficult to rec- Two studies in human volunteers lend support to oncile with paracetamol’s lack of anti-inflamma- a 5-HT-related analgesic mechanism for paraceta- tory action (but reasonable explanations were, and mol. Both studies utilized volunteers that were rapid continue to be, offered), was followed by inhibition metabolizers of tropisetron to ensure homogeneity of COX-2, then inhibition of the putative COX-3 of the study population and to avoid long washout (subsequently shown to be highly unlikely). When periods. The first study (n = 26) found that the it was recognized that ‘weak’ is not a helpful des- analgesic effect of 1000 mg oral paracetamol against ignation for analgesic mechanism, other physio- electrical stimulation of the median nerve was sig- logical pathways have been proposed. The nificantly reduced by the 5-HT receptor antagonists consensus appears to suggest a primarily central tropisetron (5 mg) and granisetron (3 mg) adminis- (CNS) site of action (interestingly, this is similar to tered i.v. (130). A follow-up study (n = 18) by the the earliest proposal, but for a different mecha- same group found that tropisetron completely nism). The persistent difficulty in elucidating par- inhibited the analgesic effect of 1000 mg oral paracetamol’s mechanism of analgesic action might be acetamol in a cold pressor test (131). because of it having more than one mechanism, perhaps acting at multiple sites, and perhaps acting synergistically, neither one of which by itself Self-synergism accounting for the overall effect. From the beginning, the focus of the search for paracetamol’s analgesic mechanism has concentrated CONCLUSION on the central nervous system. When administered intraventricularly (i.c.v.), acetaminophen produces Paracetamol, as all drugs, has desirable and unde- no significant analgesia (115, 132). This finding lead sirable characteristics. It has benefits related to pain to attempts to inject acetaminophen into the spinal relief and it has risks related to adverse effects. cord (i.t.), which produced marked dose-related Hence, there is reason and motivation to under- antinociception (132). It was further demonstrated stand how it produces its analgesic action. This that combined administration of acetaminophen knowledge would also be useful for the design or into the brain and spinal cord produced synergistic discovery of a similar drug with an even larger antinociception (132). This property of acetamino- therapeutic window. phen has been dubbed ‘self-synergy’. Opioid The mechanism of paracetamol’s adverse effects receptor antagonists administered by i.t. injection is fairly well understood. In contrast, none of the attenuate the synergism observed with combined proposals about paracetamol’s mechanism of acetaminophen administration (132, 133). Remem- analgesic action are completely satisfactory. Some bering that acetaminophen itself does not bind to proposals have been discounted, others remain opioid receptors (103), this suggests that endo- promising, but to date lack sufficient evidence to genous opioids contribute to the antinociceptive conclude that they are definitive. So far, no single effects of acetaminophen at the spinal level. A caveat mechanism has been able to describe all of its is that acetaminophen binding has been investigated actions sufficiently. However, it is reasonable to at only 10 lM. As acetaminophen reaches signifi- conclude that paracetamol likely has a pharmaco- cantly higher concentrations in the CNS at analgesic logical mechanism that interacts with a variety of doses (123), opioid binding at higher concentrations physiological pathways, likely within the central should be tested. nervous system. As long as paracetamol’s analgesic mechanism of action remains an enigma, assessment of its SUMMARY benefit ⁄ risk ratio and design or discovery of drugs Paracetamol was already in wide use before with a similar mechanism, but greater benefit ⁄ risk the molecular mechanism of the other ‘weak’ ratio, will be impeded.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 633

16. Lambrecht FY, Durkan K, Yildirim Y, Acar C (2006) REFERENCES Labeling of acetaminophen with I-131 and biodis- 1. Steiner TJ, Voelker M (2009) Gastrointestinal toler- tribution in rats. Chemical and Pharmaceutical Bulletin ability of aspirin and the choice of over-the-counter (Tokyo), 54, 245–247. analgesia for short-lasting acute pain. Journal of 17. Fukuda M, Kitaichi K, Abe F et al. (2005) Altered Clinical Pharmacy and Therapeutics, 34, 177–186. brain penetration of diclofenac and mefenamic acid, 2. Loke YK (2009) Aspirin or paracetamol – what’s but not acetaminophen, in Shiga-like toxin II-treated good for you? Journal of Clinical Pharmacy and mice. Journal of Pharmacological Sciences, 97, 525–532. Therapeutics, 34, 247–248. 18. Anderson BJ, Holford NH, Woollard GA, Chan PL 3. Raffa R (2006) Pharmacological aspects of successful (1998) Paracetamol plasma and cerebrospinal fluid long-term analgesia. Clinical Rheumatology, 25(Sup- pharmacokinetics in children. British Journal of pl. 1), S9–S15. Clinical Pharmacology, 46, 237–243. 4. D’Mello R, Dickenson AH (2008) Spinal cord 19. van der Marel CD, Anderson BJ, Pluim MA, de Jong mechanisms of pain. British Journal of Anaesthesiolo- TH, Gonzalez A, Tibboel D (2003) Acetaminophen gy, 101, 8–16. in cerebrospinal fluid in children. European Journal 5. Kadiev E, Patel V, Rad P et al. (2008) Role of phar- of Clinical Pharmacology, 59, 297–302. macogenetics in variable response to drugs: focus 20. Gibb IA, Anderson BJ (2008) Paracetamol (acet- on opioids. Expert Opinion in Drug Metabolism and aminophen) pharmacodynamics: interpreting the Toxicology, 4, 77–91. plasma concentration. Archives of Disease in Child- 6. Sneader W (2005) Drug discovery a history. West hood, 93, 241–247. Sussex: John Wiley & Sons Ltd. 438–439. 21. Sachs CJ (2005) Oral analgesics for acute nonspecific 7. Spooner JB, Harvey JG (1976) The history and usage pain. American Family Physician, 71, 913–918. of paracetamol. The Journal of International Medical 22. Dalton JD Jr, Schweinle JE (2006) Randomized Research, 4, 1–6. controlled noninferiority trial to compare extended 8. Clissold SP (1986) Paracetamol and phenacetin. release acetaminophen and ibuprofen for the treat- Drugs, 32(Suppl. 4), 46–59. ment of ankle sprains. Annals of Emergency Medicine, 9. Prescott LF (2000) Paracetamol: past, present, and 48, 615–623. future. American Journal of Therapeutics, 7, 143–147. 23. Sinatra RS, Jahr JS, Reynolds LW, Viscusi ER, 10. Brodie BB, Axelrod J (1948) The fate of acetanilide in Groudine SB, Payen-Champenois C (2005) Efficacy man. The Journal of Pharmacology and Experimental and safety of single and repeated administration of Therapeutics, 94, 29–38. 1 gram intravenous acetaminophen injection (par- 11. Brodie BB, Axelrod J (1949) The fate of acetophe- acetamol) for pain management after major ortho- netidin in man and methods for the estimation of pedic surgery. Anesthesiology, 102, 822–831. acetophenetidin and its metabolites in biological 24. Loeser JD, Treede RD (2008) The Kyoto protocol of material. The Journal of Pharmacology and Experi- IASP Basic Pain Terminology. Pain, 137, 473–477. mental Therapeutics, 97, 58–67. 25. Gelgor L, Cartmell S, Mitchell D (1992) Intracere- 12. Black M (1984) Acetaminophen hepatotoxicity. broventricular micro-injections of non-steroidal Annual Review of Medicine, 35, 577–593. anti-inflammatory drugs abolish reperfusion hy- 13. Chou R, Huffman LH, American Pain S, American peralgesia in the rat’s tail. Pain, 50, 323–329. College of P (2007) Medications for acute and 26. Bianchi M, Panerai AE (1996) The dose-related chronic low back pain: a review of the evidence for effects of paracetamol on hyperalgesia and noci- an American Pain Society ⁄ American College of ception in the rat. British Journal of Pharmacology, Physicians clinical practice guideline. Annals of 117, 130–132. Internal Medicine, 147, 505–514. 27. Bianchi M, Panerai AE (1997) Formalin injection in 14. Zhang W, Doherty M, Leeb BF et al. (2007) EULAR the tail facilitates hindpaw withdrawal reflexes evidence based recommendations for the manage- induced by thermal stimulation in the rat: effect of ment of hand osteoarthritis: report of a Task Force paracetamol. Neuroscience Letters, 237, 89–92. of the EULAR Standing Committee for International 28. Sandrini M, Romualdi P, Capobianco A, Vitale G, Clinical Studies Including Therapeutics (ESCISIT). Morelli G, Pini LA, Candeletti S (2001) The effect of Annals of The Rheumatic Diseases, 66, 377–388. paracetamol on nociception and dynorphin A levels 15. Hardman JG, Limbird LE, Gilman AG, eds (2001) in the rat brain. Neuropeptides, 35, 110–116. Goodman and Gilman’s the pharmacological basis of 29. Choi SS, Lee JK, Suh HW (2001) Antinociceptive therapeutics. 10th edn, New York: McGraw-Hill. profiles of aspirin and acetaminophen in formalin,

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 634 K. Toussaint et al.

substance P and glutamate pain models. Brain 40. Dani Ml, Guindon Je, Lambert C, Beaulieu P (2007) Research, 921, 233–239. The local antinociceptive effects of paracetamol in 30. Bonnefont J, Alloui A, Chapuy E, Clottes E, Escha- neuropathic pain are mediated by cannabinoid lier A (2003) Orally administered paracetamol does receptors. European Journal of Pharmacology, 573, not act locally in the rat formalin test: evidence for a 214–215. supraspinal, serotonin-dependent antinociceptive 41. Saito O, Aoe T, Yamamoto T (2005) Analgesic ef- mechanism. Anesthesiology, 99, 976–981. fects of nonsteroidal antiinflammatory drugs, acet- 31. Nagakura Y, Okada M, Kohara A et al. (2003) aminophen, and morphine in a mouse model of Allodynia and hyperalgesia in adjuvant-induced bone cancer pain. Journal of Anesthesia, 19, 218–224. arthritic rats: time course of progression and effi- 42. Lynch JJ, Wade CL, Zhong CM, Mikusa JP, Honore cacy of analgesics. The Journal of Pharmacology and P (2004) Attenuation of mechanical allodynia by Experimental Therapeutics, 306, 490–497. clinically utilized drugs in a rat chemotherapy-in- 32. Koppert W, Wehrfritz A, Korber N, Sittl R, Albrecht duced neuropathic pain model. Pain, 110, 56–63. S, Schuttler J, Schmelz M (2004) The cyclooxygenase 43. Leinisch E, Evers S, Kaempfe N et al. (2005) Evalu- isozyme inhibitors parecoxib and paracetamol ation of the efficacy of intravenous acetaminophen reduce central hyperalgesia in humans. Pain, 108, in the treatment of acute migraine attacks: a double- 148–153. blind, placebo-controlled parallel group multicenter 33. Bolcskei K, Horvath D, Szolcsanyi J, Petho G (2007) study. Pain, 117, 396–400. Heat injury-induced drop of the noxious heat 44. Lipton RB, Baggish JS, Stewart WF, Codispoti JR, Fu threshold measured with an increasing-temperature M (2000) Efficacy and safety of acetaminophen in water bath: a novel rat thermal hyperalgesia model. the treatment of migraine: results of a randomized, European Journal of Pharmacology, 564, 80–87. double-blind, placebo-controlled, population-based 34. Bianchi M, Martucci C, Ferrario P, Franchi S, study. Archives of Internal Medicine, 160, 3486–3492. Sacerdote P (2007) Increased tumor necrosis factor- 45. Akin M, Price W, Rodriguez G Jr, Erasala G, Hurley alpha and prostaglandin E2 concentrations in the G, Smith RP (2004) Continuous, low-level, topical cerebrospinal fluid of rats with inflammatory hy- heat wrap therapy as compared to acetaminophen peralgesia: the effects of analgesic drugs. Anesthesia for primary dysmenorrhea. The Journal of Reproduc- and Analgesia, 104, 949–954. tive Medicine, 49, 739–745. 35. Crawley B, Saito O, Malkmus S, Fitzsimmons B, 46. Clark WG, Cumby HR (1976) Antagonism by anti- Hua XY, Yaksh TL (2008) Acetaminophen prevents pyretics of the hyperthermic effect of a prostaglan- hyperalgesia in central pain cascade. Neuroscience din precursor, sodium arachidonate, in the cat. The Letters, 442, 50–53. Journal of Physiology, 257, 581–595. 36. Rezende RM, Franca DS, Menezes GB, dos Reis 47. Li S, Dou W, Tang Y, Goorha S, Ballou LR, Blatteis WG, Bakhle YS, Francischi JN (2008) Different CM (2008) Acetaminophen: antipyretic or hypo- mechanisms underlie the analgesic actions of par- thermic in mice? In either case, PGHS-1b (COX-3) is acetamol and dipyrone in a rat model of inflam- irrelevant Prostaglandins & Other Lipid Mediators, 85, matory pain. British Journal of Pharmacology, 153, 89–99. 760–768. 48. Crawford IL, Kennedy JI, Lipton JM, Ojeda SR (1979) 37. Luccarini P, Childeric A, Gaydier A-M, Voisin D, Effects of central administation of probenecid on Dallel R (2006) The orofacial formalin test in the fevers produced by leukocytic pyrogen and PGE2 in mouse: a behavioral model for studying physiology the rabbit. The Journal of Physiology, 287, 519–533. and modulation of trigeminal nociception. The 49. Huang W-T, Wang J-J, Lin M-T (2004) Antipyretic Journal of Pain, 7, 908–914. effect of acetaminophen by inhibition of glutamate 38. Matsunaga A, Kawamoto M, Shiraishi S, Yasuda T, release after staphylococcal enterotoxin A fever in Kajiyama S, Kurita S, Yuge O (2007) Intrathecally rabbits. Neuroscience Letters, 355, 33–36. administered COX-2 but not COX-1 or COX-3 50. Devi BP, Boominathan R, Mandal SC (2003) Evalu- inhibitors attenuate streptozotocin-induced ation of antipyretic potential of Cleome viscosa mechanical hyperalgesia in rats. European Journal of Linn(Capparidaceae) extract in rats. Journal of Eth- Pharmacology, 554, 12–17. nopharmacology, 87, 11–13. 39. Cui JG, Zhang X, Zhao YH, Chen C, Bazan N (2006) 51. Tomazetti J, Avila DS, Ferreira AP et al. (2005) Baker Allodynia and hyperalgesia suppression by a novel yeast-induced fever in young rats: characterization analgesic in experimental neuropathic pain. Bio- and validation of an animal model for antipyretics chemical and Biophysical Research Communications, screening. Journal of Neuroscience Methods, 147, 29– 350, 358–363. 35.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 635

52. Massey TE, Walker RM, McElligott TF, Racz WJ 63. Alloui A, Chassaing C, Schmidt J, Ardid D, Dubray (1982) Acetaminophen-induced hypothermia in C, Cloarec A, Eschalier A (2002) Paracetamol exerts mice: evidence for a central action of the parent a spinal, tropisetron-reversible, antinociceptive ef- compound. Toxicology, 25, 187–200. fect in an inflammatory pain model in rats. European 53. Ayoub SS, Botting RM, Goorha S, Colville-Nash PR, Journal of Pharmacology, 443, 71–77. Willoughby DA, Ballou LR (2004) Acetaminophen- 64. Brandt KD, Mazzuca SA, Buckwalter KA (2006) induced hypothermia in mice is mediated by a Acetaminophen, like conventional NSAIDs, may prostaglandin endoperoxide synthase 1 gene- reduce synovitis in osteoarthritic knees. Rheumatol- derived protein. Proceedings of the National Academy ogy (Oxford, England), 45, 1389–1394. of Sciences of the United States of America, 101, 11165– 65. Case JP, Baliunas AJ, Block JA (2003) Lack of effi- 11169. cacy of acetaminophen in treating symptomatic 54. Stevens T, Fitzsimmons L (1995) Effect of a stan- knee osteoarthritis: a randomized, double-blind, dardized rewarming protocol and acetaminophen placebo-controlled comparison trial with diclofenac on core temperature after coronary artery bypass. sodium. Archives of Internal Medicine, 163, 169–178. American Journal of Critical Care: An Official Publica- 66. Lages B, Weiss HJ (1989) Inhibition of human tion, American Association of Critical-Care Nurses, 4, platelet function in vitro and ex vivo by acetami- 189–197. nophen. Thrombosis Research, 53, 603–613. 55. Kasner SE, Wein T, Piriyawat P et al. (2002) Acet- 67. Munsterhjelm E, Munsterhjelm NM, Niemi TT, aminophen for altering body temperature in acute Ylikorkala O, Neuvonen PJ, Rosenberg PH (2005) stroke: a randomized clinical trial Stroke. A Journal Dose-dependent inhibition of platelet function by of Cerebral Circulation, 33, 130–134. acetaminophen in healthy volunteers. Anesthesiol- 56. Dippel DW, van Breda EJ, van der Worp HB, van ogy, 103, 712–717. Gemert HM, Kappelle LJ, Algra A, Koudstaal PJ 68. Munsterhjelm E, Niemi TT, Syrjala MT, Ylikorkala (2003) Timing of the effect of acetaminophen on O, Rosenberg PH (2003) Propacetamol augments body temperature in patients with acute ischemic inhibition of platelet function by diclofenac in vol- stroke. Neurology, 61, 677–679. unteers. British Journal of Anaesthesia, 91, 357–362. 57. Mellergard P (1992). Changes in human intracere- 69. Munsterhjelm E, Niemi TT, Ylikorkala O, Silvanto bral temperature in response to different methods M, Rosenberg PH (2005) Characterization of inhi- of brain cooling. Neurosurgery, 31, 671–677; discus- bition of platelet function by paracetamol and its sion 677. interaction with diclofenac in vitro. Acta Anaesthes- 58. Honore P, Buritova J, Besson JM (1995) Aspirin and iologica Scandinavica, 49, 840–846. acetaminophen reduced both Fos expression in rat 70. Niemi TT, Backman JT, Syrjala MT, Viinikka LU, lumbar spinal cord and inflammatory signs pro- Rosenberg PH (2000) Platelet dysfunction after duced by carrageenin inflammation. Pain, 63, 365– intravenous ketorolac or propacetamol. Acta An- 375. aesthesiologica Scandinavica, 44, 69–74. 59. Abbadie C, Besson JM (1994) Chronic treatments 71. Sciulli MG, Seta F, Tacconelli S, Capone ML, Ric- with aspirin or acetaminophen reduce both the ciotti E, Pistritto G, Patrignani P (2003) Effects of development of polyarthritis and Fos-like immu- acetaminophen on constitutive and inducible pro- noreactivity in rat lumbar spinal cord. Pain, 57, stanoid biosynthesis in human blood cells. British 45–54. Journal of Pharmacology, 138, 634–641. 60. Mamuk S, Melli M (2007) Effect of aspirin, parac- 72. Rao GH, Reddy KR, White JG (1982) Effect of etamol and their nitric oxide donating derivatives acetaminophen and salicylate on aspirin-induced on exudate cytokine and PGE2 production in inhibition of human platelet cyclo-oxygenase. Pro- zymosan-induced air pouch inflammation in rats. staglandins, Leukotrienes, and Medicine, 9, 109–115. European Journal of Pharmacology, 561, 220–225. 73. Scharbert G, Gebhardt K, Sow Z, Duris M, Deusch 61. Lee Y-S, Kim H, Brahim JS, Rowan J, Lee G, Dionne E, Kozek-Langenecker S (2007) Point-of-care platelet RA (2007) Acetaminophen selectively suppresses function tests: detection of platelet inhibition peripheral prostaglandin E2 release and increases induced by nonopioid analgesic drugs. Blood Coag- COX-2 gene expression in a clinical model of acute ulation & Fibrinolysis : An International Journal in inflammation. Pain, 129, 279–286. Haemostasis and Thrombosis, 18, 775–780. 62. al-Swayeh OA, Futter LE, Clifford RH, Moore PK 74. Silvanto M, Munsterhjelm E, Savolainen S et al. (2000) Nitroparacetamol exhibits anti-inflammatory (2007) Effect of 3 g of intravenous paracetamol on and anti-nociceptive activity. British Journal of post-operative analgesia, platelet function and liver Pharmacology, 130, 1453–1456. enzymes in patients undergoing tonsillectomy un-

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 636 K. Toussaint et al.

der local anaesthesia. Acta Anaesthesiologica Scandi- 87. Chandrasekharan NV, Dai H, Roos KLT, Evanson navica, 51, 1147–1154. NK, Tomsik J, Elton TS, Simmons DL (2002) COX-3, 75. Carreia MA (2007) Drug biotransformation. In: Basic a cyclooxygenase-1 variant inhibited by acetami- and Clinical Pharmacology. Katzung BG, ed. nophen and other analgesic ⁄ antipyretic drugs: McGraw-Hill: New York, 50–63. cloning, structure, and expression. Proceedings of the 76. Persky V, Piorkowski J, Hernandez E et al. (2008) National Academy of Sciences of the United States of Prenatal exposure to acetaminophen and respira- America, 99, 13926–13931. tory symptoms in the first year of life. Annals of 88. Snipes JA, Kis B, Shelness GS, Hewett JA, Busija Allergy, Asthma & Immunology : Official Publication of DW (2005) Cloning and characterization of cyclo- the American College of Allergy, Asthma, & Immunol- oxygenase-1b (putative cyclooxygenase-3) in rat. ogy, 101, 271–278. Journal of Pharmacology and Experimetnal Therapeu- 77. Rebordosa C, Kogevinas M, Sorensen HT, Olsen J tics, 313, 668–676. (2008) Pre-natal exposure to paracetamol and risk of 89. Ouellet M, Percival MD (2001) Mechanism of acet- wheezing and asthma in children: a birth cohort aminophen inhibition of cyclooxygenase isoforms. study. International Journal of Epidemiology, 37, 583– Archives of Biochemistry and Biophysics, 387, 273–280. 590. 90. Schildknecht S, Daiber A, Ghisla S, Cohen RA, 78. Forman JP, Stampfer MJ, Curhan GC (2005) Non- Bachschmid MM (2008) Acetaminophen inhibits narcotic analgesic dose and risk of incident hyper- prostanoid synthesis by scavenging the PGHS-acti- tension in US women. Hypertension, 46, 500–507. vator peroxynitrite. The Journal of the Federation of 79. Kurth T, Hennekens CH, Sturmer T, Sesso HD, American Societies for Experimental Biology, 22, 215– Glynn RJ, Buring JE, Gaziano JM (2005) Analgesic 224. use and risk of subsequent hypertension in appar- 91. Boutaud O, Aronoff DM, Richardson JH, Marnett ently healthy men. Archives of Internal Medicine, 165, LJ, Oates JA (2002) Determinants of the cellular 1903–1909. specificity of acetaminophen as an inhibitor of 80. Smith HS (2009) Potential analgesic mechanisms of prostaglandin H2 synthases. Proceedings of the Na- acetaminophen. Pain Physician, 12, 269–280. tional Academy of Sciences of the United States of 81. Vane JR (1971) Inhibition of prostaglandin synthesis America, 99, 7130–7135. as a mechanism of action for aspirin-like drugs. 92. Lucas R, Warner TD, Vojnovic I, Mitchell JA (2005) Nature New Biology, 231, 232–235. Cellular mechanisms of acetaminophen: role of 82. Warner TD, Mitchell JA (2002) Cyclooxygenase-3 cyclo-oxygenase. The Journal of the Federation of (COX-3): filling in the gaps toward a COX contin- American Societies for Experimental Biology, 19, 635– uum? Proceedings of the National Academy of Sciences 637. of the United States of America, 99, 13371–13373. 93. Bujalska M (2004) Effect of nitric oxide synthase 83. Ayoub SS, Colville-Nash PR, Willoughby DA, Bot- inhibition on antinociceptive action of different ting RM (2006) The involvement of a cyclooxygen- doses of acetaminophen. Polish Journal of Pharma- ase 1 gene-derived protein in the antinociceptive cology, 56, 605–610. action of paracetamol in mice. European Journal of 94. Riedel W, Neeck G (2001) Nociception, pain, and Pharmacology, 538, 57–65. antinociception: current concepts. Zeitschrift fur 84. Greco A, Ajmone-Cat MA, Nicolini A, Sciulli MG, Rheumatologie, 60, 404–415. Minghetti L (2003) Paracetamol effectively reduces 95. Godfrey L, Bailey I, Toms NJ, Clarke GD, Kitchen I, prostaglandin E2 synthesis in brain macrophages Hourani SM (2007) Paracetamol inhibits nitric oxide by inhibiting enzymatic activity of cyclooxygenase synthesis in murine spinal cord slices. European but not phospholipase and prostaglandin E syn- Journal of Pharmacology, 562, 68–71. thase. Journal of Neuroscience Research, 71, 844. 96. Ryu YS, Lee JH, Seok JH et al. (2000) Acetamino- 85. Baker FC, Driver HS, Paiker J, Rogers GG, Mitchell phen inhibits iNOS gene expression in RAW 264.7 D (2002) Acetaminophen does not affect 24-h body macrophages: differential regulation of NF-kappaB temperature or sleep in the luteal phase of the by acetaminophen and salicylates. Biochemical and menstrual cycle. Journal of Applied Physiology, 92, Biophysical Research Communications, 272, 758–764. 1684–1691. 97. Bujalska M, Gumulka WS (2001) Effect of cycloox- 86. Hinz B, Cheremina O, Brune K (2008) Acetamino- ygenase and NO synthase inhibitors on antinoci- phen (paracetamol) is a selective cyclooxygenase-2 ceptive action of acetaminophen. Polish Journal of inhibitor in man. The Journal of the Federation of Pharmacology, 53, 341–350. American Societies for Experimental Biology, 22, 383– 98. Bujalska M (2003) Effect of cyclooxygenase and 390. NO synthase inhibitors administered centrally on

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 Paracetamol’s analgesic mechanism? 637

antinociceptive action of acetaminophen (Part II). 112. Ottani A, Leone S, Sandrini M, Ferrari A, Bertolini A Polish Journal of Pharmacology, 55, 1001–1011. (2006) The analgesic activity of paracetamol is pre- 99. Raffa RB (2002) Acetaminophen and metabolite: vented by the blockade of cannabinoid CB1 recep- lack of effect on nitric oxide synthase (constitutive tors. European Journal of Pharmacology, 531, 280–281. or inducible). Headache, 42, 237–238. 113. Ovadia H, Wohlman A, Mechoulam R, Weidenfeld 100. Devane WA, Dysarz FAd, Johnson MR, Melvin LS, J (1995) Characterization of the hypothermic effect Howlett AC (1988) Determination and character- of the synthetic cannabinoid HU-210 in the rat. ization of a cannabinoid receptor in rat brain. Relation to the adrenergic system and endogenous Molecular Pharmacology, 34, 605–613. pyrogens. Neuropharmacology, 34, 175–180. 101. Devane WA, HanuA˚ ¡ L, Breuer A et al. (1992) Iso- 114. Bertolini A, Ferrari A, Ottani A, Guerzoni S, Tacchi lation and structure of a brain constituent that binds R, Leone S (2006) Paracetamol: new vistas of an old to the cannabinoid receptor. Science, 258, 1946–1949. drug. CNS Drug Reviews, 12, 250. 102. Fride E (2002) Endocannabinoids in the central 115. Abbott FV, Hellemans KGC (2000) Phenacetin, nervous system – an overview. Prostaglandins, Leu- acetaminophen and dipyrone: analgesic and kotrienes and Essential Fatty Acids, 66, 221. rewarding effects. Behavioural Brain Research, 112, 103. Raffa RB, Codd EE (1996) Lack of binding of acet- 177–186. aminophen to 5-HT receptor or uptake sites (or 116. Orr KK, Matson KL, Cowles BJ (2006) Nonpre- eleven other binding ⁄ uptake assays). Life Sciences, scription medication use by infants and children: 59, PL37–PL40. product labeling versus evidence-based medicine. 104. Mallet C, Daulhac L, Bonnefont Jrm et al. (2008) Journal of Pharmacy Practice, 19, 286–294. Endocannabinoid and serotonergic systems are 117. Roca-Vinardell A, Ortega-Alvaro A, Gibert-Rahola needed for acetaminophen-induced analgesia. Pain, J, Mico JA (2003) The role of 5-HT1A ⁄ B autore- 139, 190–200. ceptors in the antinociceptive effect of systemic 105. Hogestatt ED, Jonsson BAG, Ermund A et al. (2005) administration of acetaminophen. Anesthesiology, Conversion of acetaminophen to the bioactive 98, 741–747. N-acylphenolamine AM404 via fatty acid amide 118. Sandrini M, Pini LA, Vitale G et al. (2003) Differ- hydrolase-dependent arachidonic acid conjugation ential involvement of central 5-HT1B and 5-HT3 in the nervous system. Journal of Biological Chemistry, receptor subtypes in the antinociceptive effect of 280, 31405–31412. paracetamol. Inflammation Research : Official Journal 106. Zygmunt PM, Chuang H-h, Movahed P, Julius D, of the European Histamine Research Society, 52, 347– Ho¨gesta¨tt ED (2000) The anandamide transport 352. inhibitor AM404 activates vanilloid receptors. 119. Pelissier T, Alloui A, Caussade F, Dubray C, Cloarec European Journal of Pharmacology, 396, 39–42. A, Lavarenne J, Eschalier A (1996) Paracetamol 107. Ralevic V, Kendall DA, Jerman JC, Middlemiss DN, exerts a spinal antinociceptive effect involving an Smart D (2001) Cannabinoid activation of indirect interaction with 5-hydroxytryptamine3 recombinant and endogenous vanilloid receptors. receptors: in vivo and in vitro evidence. The Journal European Journal of Pharmacology, 424, 211–219. of Pharmacology and Experimental Therapeutics, 278, 8– 108. Beltramo M, Stella N, Calignano A, Lin SY, 14. Makriyannis A, Piomelli D (1997) Functional role of 120. Libert F, Bonnefont J, Bourinet E et al. (2004) Acet- high-affinity anandamide transport, as revealed by aminophen: a central analgesic drug that involves a selective inhibition. Science, 277, 1094–1097. spinal tropisetron-sensitive, non-5-HT(3) receptor- 109. De Petrocellis L, Bisogno T, Davis JB, Pertwee RG, mediated effect. Molecular Pharmacology, 66, 728– Di Marzo V (2000) Overlap between the ligand 734. recognition properties of the anandamide trans- 121. Pini LA, Sandrini M, Vitale G (1996) The antinoci- porter and the VR1 vanilloid receptor: inhibitors of ceptive action of paracetamol is associated with anandamide uptake with negligible capsaicin-like changes in the serotonergic system in the rat brain. activity. FEBS Letters, 483, 52–56. European Journal of Pharmacology, 308, 31–40. 110. Karai L, Brown DC, Mannes AJ et al. (2004) Deletion 122. Pini LA, Vitale G, Ottani A, Sandrini M (1997) of vanilloid receptor 1 expressing primary afferent Naloxone-reversible antinociception by paraceta- neurons for pain control. The Journal of Clinical mol in the rat. The Journal of Pharmacology and Investigation, 113, 1344. Experimental Therapeutics, 280, 934–940. 111. Szallasi A, Di Marzo V (2000) New perspectives on 123. Courade JP, Besse D, Delchambre C et al. (2001) enigmatic vanilloid receptors. Trends in Neuro- Acetaminophen distribution in the rat central ner- sciences, 23, 491–497. vous system. Life Sciences, 69, 1455–1464.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638 638 K. Toussaint et al.

124. Srikiatkhachorn A, Tarasub N, Govitrapong P 129. Courade JP, Chassaing C, Bardin L, Alloui A, (1999) Acetaminophen-induced antinociception via Eschalier A (2001) 5-HT receptor subtypes in- central 5-HT(2A) receptors. Neurochemistry Interna- volved in the spinal antinociceptive effect of tional, 34, 491–498. acetaminophen in rats. European Journal of Phar- 125. Sandrini M, Vitale G, Ruggieri V, Pini LA (2007) macology, 432, 1–7. Effect of acute and repeated administration of par- 130. Pickering G, Loriot MA, Libert F, Eschalier A, Bea- acetamol on opioidergic and serotonergic systems une P, Dubray C (2006) Analgesic effect of acet- in rats. Inflammation Research : Official Journal of the aminophen in humans: first evidence of a central European Histamine Research Society, 56, 139–142. serotonergic mechanism. Clinical Pharmacology and 126. Tjolsen A, Lund A, Hole K (1991) Antinociceptive Therapeutics, 79, 371–378. effect of paracetamol in rats is partly dependent on 131. Pickering G, Esteve V, Loriot MA, Eschalier A, spinal serotonergic systems. European Journal of Dubray C (2008) Acetaminophen reinforces Pharmacology, 193, 193–201. descending inhibitory pain pathways. Clinical 127. Pelissier T, Alloui A, Paeile C, Eschalier A (1995) Pharmacology and Therapeutics, 84, 47–51. Evidence of a central antinociceptive effect of par- 132. Raffa RB, Stone DJJR, Tallarida RJ (2000) Discovery acetamol involving spinal 5HT3 receptors. Neuro- of ‘‘self-synergistic’’ spinal ⁄ supraspinal antinoci- report, 6, 1546–1548. ception produced by acetaminophen (paracetamol). 128. Alloui A, Pelissier T, Dubray C, Lavarenne J, The Journal of Pharmacology and Experimental Thera- Eschalier A (1996) Tropisetron inhibits the antino- peutics, 295, 291. ciceptive effect of intrathecally administered par- 133. Raffa RB, Walker EA, Sterious SN (2004) Opioid acetamol and serotonin. Fundamental & Clinical receptors and acetaminophen (paracetamol). Euro- Pharmacology, 10, 406–407. pean Journal of Pharmacology, 503, 209–210.

2010 The Authors. JCPT 2010 Blackwell Publishing Ltd, Journal of Clinical Pharmacy and Therapeutics, 35, 617–638