Key messages

  • Assess and document characteristics of pain in order to individualise and monitor effectiveness of treatment
  • Use paracetamol first: it is effective when taken regularly in appropriate doses and has a good safety profile
  • Consider the range of adverse effects and serious drug interactions with tramadol when selecting therapy where pain requires an opioid or opioid-like analgesic
  • Before prescribing COX-2 selective or conventional NSAIDs, review risk of peptic ulcer, cardiac disease or renal impairment
    • COX-2 selective NSAIDs are not more effective than conventional NSAIDs and have a similar range of adverse effects

Assess the pain to guide treatment choice and review its effectiveness

Obtain a ‘pain history’ in addition to the physical examination1,2

A pain history attempts to determine the mechanisms producing pain and factors influencing the painful experience. The history should consider:

  • location—is there more than one site affected?
  • what makes the pain better or worse?
  • character of the pain (e.g. throbbing, aching, sharp, burning)
  • intensity/severity of pain
  • does the pain radiate anywhere?
  • timing—is the pain continuous or does it fluctuate or relate
  • to specific events (e.g. sleep, movement)?
Proper assessment and control of pain requires patient involvement1

Pain is always subjective so measuring pain must rely on recording the patient’s report. A common method uses visual analogue scales which can be a sensitive and consistent means of assessing the effectiveness of analgesia on an individual basis.3

The pain scale

Figure 1: Pain relief scale

Pain is best treated early because established or severe pain is more difficult to treat1

Pain relief scales (as above) may be more convenient than pain intensity scales because patients have the same baseline relief (zero) whereas they could start with different baseline intensities (e.g. moderate or severe).3

The effectiveness of analgesia should be reviewed regularly. Inadequate pain relief can result in the patient progressing from acute pain to chronic pain.1,2 Although beyond the scope of this discussion, non-drug therapies should also be employed to modify pain.

Paracetamol is the drug of first choice in acute pain

Paracetamol is an effective analgesic when used regularly and in appropriate doses

Good efficacy and safety, together with its low cost relative to other analgesics, makes paracetamol the drug of first choice for managing acute pain conditions.

Systematic reviews of randomised controlled trials have confirmed that paracetamol is an effective analgesic for a wide range of painful conditions.4,5 Paracetamol continues to be recommended as a first choice analgesic for osteoarthritis.2,6

The dosage of paracetamol is often insufficient, leading to perceptions that it is ineffective. Regular use of therapeutic doses provides relief from mild to moderate pain.

Paracetamol has an excellent safety record at therapeutic doses

Paracetamol is well tolerated and toxicity at therapeutic doses is extremely rare. Most cases of paracetamol-induced hepatotoxicity have resulted from either deliberate self-poisoning or accidental overdoses with therapeutic intent.7

The risk of paracetamol toxicity increases in those who are fasting or dehydrated (poor fluid intake > 24 hours), with concurrent illness (fever, vomiting, or diarrhoea) causing dehydration, or with underlying hepatic disease.

Achieving pain relief when paracetamol is inadequate

Paracetamol should be used as an adjunct when other forms of analgesia are employed

The WHO ‘analgesic ladder’ is a stepwise approach to managing pain, starting with a non-opioid analgesic (with or without non-drug therapies) and moving up to potent opioid analgesics. Throughout any progression, the non-opioid analgesic should be continued.

NSAIDs can be added to, or substituted for, paracetamol when pain is of an inflammatory nature

NSAIDs are particularly useful in pain resulting from inflammation. Combining paracetamol with an NSAID may enable lower doses of the NSAID to be used,8 thus decreasing the risk of gastrointestinal adverse effects.

 Add a weak opioid or opioid-like analgesic when paracetamol alone is inadequate…

An alternative to using an NSAID (particularly in pain without inflammatory origins) is to add a weak opioid (such as codeine) to paracetamol or an opioid-like analgesic such as tramadol.

…but avoid combinations of paracetamol and dextropropoxyphene

Avoid paracetamol/dextropropoxyphene combinations (Capadex, Di-Gesic, Paradex)—they are no more effective than paracetamol alone.9,10 Furthermore, these fixed-dose combinations are generally taken at a frequency that increases the risk of adverse effects from accumulation of dextropropoxyphene and its cardiotoxic metabolite. Dextropropoxyphene can also cause dependency.

Tramadol has potentially serious adverse effects and drug interactions which make it less preferable

Consider adverse effects and drug interactions when choosing between add-on analgesics

The efficacy of paracetamol 1000 mg/codeine 60 mg and tramadol 100 mg is comparable.3,11 Thus the prescriber needs to consider safety issues when choosing between these alternatives.

Typical opioid adverse effects associated with codeine (particularly constipation but also nausea, drowsiness, dizziness and dry mouth) are well-known to prescribers and greatly influence how well the patient tolerates the paracetamol/codeine combination at higher codeine doses.

Tramadol is associated with serious adverse effects…

Similarly, tolerability can be an issue with tramadol. Almost one-third of patients stopped taking tramadol in clinical trials, predominantly due to adverse effects.12

Nausea occurs in at least 1 in 10 patients,13 often limiting therapy.14 Serious adverse effects include hallucinations, raised blood pressure and hypersensitivity reactions.15 Tramadol may induce seizures, particularly in patients with epilepsy or a recognised risk of seizure.2,8

The Australian Adverse Drug Reactions Advisory Committee (ADRAC) has received 20 reports of serotonin syndrome with tramadol15; this syndrome is more likely to occur with high doses of tramadol.16

 …abuse and withdrawal reactions have been reported…

Case reports have revealed dependence, abuse and withdrawal symptoms with tramadol, mainly after long-term treatment14 of several months or longer. ADRAC has received 11 reports of withdrawal symptoms with tramadol.15

…and clinically important drug interactions

Concomitant use of tramadol with drugs which lower the seizure threshold, such as tricyclic antidepressants, selective serotonin re-uptake inhibitors (SSRIs), bupropion, or opioids, may precipitate convulsions.14,15

Additionally, combining tramadol with drugs which increase serotonin activity in the brain increases the risk of serotonin syndrome; such drugs include tricyclic antidepressants, SSRIs, venlafaxine, monoamine oxidase inhibitors (including moclobemide), pethidine, St. John’s wort and sibutramine.2,8,14

As tramadol may increase the effect of warfarin,15 INR should be monitored in patients taking warfarin who either commence or cease tramadol therapy.

Review analgesic effectiveness and patient tolerability of the drug therapy regularly

For patients in whom tramadol is used, it would seem prudent to initiate therapy with the immediate-release dosage form (to assess a patient’s response) and only switch to the modified-release formulation if tolerability is established.

Little difference between COX-2 selective NSAIDs and other NSAIDs for analgesia and adverse effects

Consider the overall benefit/harm of both gastrointestinal and nongastrointestinal adverse effects when prescribing all NSAIDs

COX-2 selective NSAIDs are not more effective than other NSAIDS. Therefore, the overall benefit/harm profile should be considered before prescribing.

The safety profile for COX-2 selective NSAIDs depends on the cumulative effects of both gastrointestinal and non-gastrointestinal adverse effects; these data remain controversial and incomplete at present.17

It remains unresolved whether COX-2 selective NSAIDs are prothrombotic compared to other NSAIDs. On current evidence, prescribing COX-2 selective NSAIDs preferentially over conventional NSAIDs is not justified in patients requiring prophylactic low-dose aspirin.18

Review the patient’s risk of peptic ulcer, cardiac disease or renal impairment before prescribing NSAIDs,  including COX-2 selective NSAIDs

There is no evidence that COX-2 selective NSAIDs can be used in patients with active peptic ulcer disease. Gastrointestinal complications (perforation, obstruction, or bleeding) do occur with celecoxib and rofecoxib. Improved gastrointestinal safety has been questioned, particularly with celecoxib use exceeding 6 months.19 Meloxicam has less COX-2 specificity at higher doses8 and its gastrointestinal safety has not been assessed adequately in clinical outcome trials.

Caution is advised if NSAIDs or COX-2 selective NSAIDs are used in patients at risk of heart failure or acute renal failure, particularly the elderly or in those taking ACE inhibitors and thiazide diuretics concurrently. Assess renal function and blood pressure prior to prescribing and during therapy in those considered at risk.


Associate Professor Milton Cohen, Rheumatologist and Pain Physician, St. Vincent’s Hospital Campus, Sydney

  1. NHMRC. Acute pain management: information for general practitioners. Canberra: Commonwealth of Australia; 1999.
  2. Writing Group for Therapeutic Guidelines: Analgesic. Therapeutic Guidelines: Analgesic, Version 4, 2002. North Melbourne: Therapeutic Guidelines Ltd; 2002.
  3. Bandolier Extra. Acute pain. Feb 2003. (
  4. Moore A, et al. Pain 1997;70:193–201.
  5. de Craen AJM, et al. BMJ 1996;313:321–5.
  6. Abramson SB. Arthritis Rheum 2002;46:2831–5.
  7. Prescott LF. Am J Ther 2000;7:99–114.
  8. Australian Medicines Handbook 2003.
  9. Collins SL, Edwards JE, Moore RA, McQuay HJ. Single dose dextropropoxyphene, alone and with paracetamol (acetaminophen), for postoperative pain (Cochrane Review). In: The Cochrane Library, Issue 2, 2003. Oxford: Update Software.
  10. Po ALW, Zhang WY. BMJ 1997;315:1565–71.
  11. NSW Therapeutic Assessment Group. Low back pain 2002. (Available at
  12. Schnitzer TJ, et al. J Rheumatol 2000;27:772–8.
  13. Tramal Product Information. MIMS online. ( accessed May 2003)
  14. Kaye K, Theaker N. Tramadol–a position statement of the NSW Therapeutic Assessment Group. September 2001. (Available at
  15. ADRAC. Aust Adv Drug React Bull 2003;22:2.
  16. ADRAC. Aust Adv Drug React Bull 2001;20:14.
  17. Jones R. BMJ 2002;325:607–8.
  18. National Institute for Clinical Excellence. Guidance on the use of cyclo-oxygenase (Cox) II selective inhibitors. London: NHS Executive, July 2001.
  19. Juni P, et al. BMJ 2002;324:1287–8.