A 64 year old woman with knee pain: case presentation
http://www.100md.com
《英国医生杂志》
1 Department of Medicine, University of Ottawa, 202-1 Stewart Street, Ottawa K1N 6N5, Canada, 2 Ottawa Health Research Institute, Ottawa Hospital, 1053 Carling Avenue, Ottawa, K1Y 4E9, Canada, 3 Cochrane Musculoskeletal Review Group, University of Ottawa
Correspondence to: ptugwell@uottawa.ca
Mrs Patell is a 64 year old woman with osteoarthritis of the right knee. A year ago, her general practitioner had recommended up to 4000 mg/day of paracetamol (acetominophen) for pain localised at the medial compartment of the knee. Mrs Patell had been taking the paracetamol but had forgotten that her general practitioner had mentioned that it should not be taken with alcohol. She has recently seen a newspaper article warning people about the risks of chronic use of analgesics. The article warns against liver damage when taking high doses of paracetamol in association with alcohol.
After reading the newspaper article she became worried, because she enjoys a cocktail before dinner and shares a bottle of wine while eating with her husband. She knew she needed to take something; her pain was intolerable with lower doses and the pain also limits her abilities to help her husband with routine tasks around the house.
This case is fictional but was developed from several real cases. It was commissioned to contribute to the special issue on harms that will be published on 3 July 2004.
Her friend recommended trying non-steroidal anti-inflammatory drugs for pain relief because these drugs have no alcohol restrictions. The non-steroidal anti-inflammatory provided better relief than the paracetamol, but her pharmacist warned her that it could cause serious bleeding problems. Confused, Mrs Patell seeks advice from her general practitioner.
Her general practitioner agrees that she needs a chronic analgesic but points out that there are benefits and risks from high doses of both paracetamol and non-steroidal anti-inflammatories. He explains the concept of the "number needed to treat" and tells her that, compared with placebo, the number needed to treat for improvement in pain at rest is 2 with up to 4000 mg of paracetamol and 1 with non-steroidal anti-inflammatories.1 He also tells her that there are case reports of serious liver toxicity in people taking therapeutic doses of paracetamol; some authorities have associated the toxicity with alcohol,2 3 but the association is controversial (Nick Bateman, personal communication, 2004). In addition, a recent review of the literature about non-steroidal anti-inflammatories cites the chance of hospital admission or death due to a serious gastrointestinal event at up to 16 per thousand a year.4
To further explain the evidence about the benefits of the treatment options to Mrs Patell, her general practitioner uses a diagram5and grades the strength of the scientific evidence using platinum (strongest), gold, silver, or bronze (weakest) ribbons (figure).6 The diagram shows the effect on 1000 people with no treatment, 4000 mg/day of paracetamol, and non-steroidal anti-inflammatories.
Diagram using a block of 1000 figures to describe benefits and harms of treatment for osteoarthritis of knee. Each dot stands for one person. NSAID=non-steroidal anti-inflammatory drug
If 1000 people did not have any treatment over the next year, then about 90 would improve on their own (silver) and no one would have serious side effects due to treatment. If 1000 people like her took up to 4000 mg/day of paracetamol for one year, then about 730 people would improve (silver), as was the case for Mrs Patell. However, about 1 person, who also consumed the same amount of alcohol as Mrs Patell, might develop liver damage, which could lead to death in a fifth of cases (bronze).2
The evidence suggests non-steroidal anti-inflammatories may work better for her than paracetamol and she would be able to drink alcohol, but they also have a higher chance of serious side effects than paracetamol. Her general practitioner concludes that the choice depends on the importance Mrs Patell places on the improved pain relief with non-steroidal anti-inflammatories compared with their harms. Mrs Patell says she definitely needs treatment, but she is not sure what to choose. She wants more time to think it over and asks for written information. In the meantime, she will go back to using paracetamol and try to reduce her alcohol intake.
Questions
In Mrs Patell's case, which option has the most favourable benefit:harm ratio?
Does it make a difference to you whether you think of the benefit:harm ratio as number needed to treat versus number needed to harm or number/1000 benefiting versus number/1000 harmed?
How would you satisfy Mrs Patell's desire for more information?
Please respond through bmj.com
We thank George Wells and Joan Peterson for help with the calculations.
Competing interests: PT received travel and research support from pharmaceutical companies. AOC receives an unrestricted research grant from the Foundation for Informed Medical Decision Making, which has a licensing agreement with Health Dialog, a company that markets decision aids.
This is the first of a three part case report where we invite readers to take part in considering the diagnosis and management of a case using the rapid response feature on bmj.com. Next week we will report the case progression and in four weeks' time we will report the outcome and summarise the responses.
References
Towheed TE, Judd MJ, Hochberg MC, Wells G. Acetaminophen for osteoarthritis. Cochrane Database Syst Rev 2003;(2): CD004257.
Zimmerman HJ, Maddrey WC. Acetaminophen (paracetamol) hepatotoxicity with regular intake of alcohol: analysis of instances of therapeutic misadventure . Hepatology 1995;22: 767-73.
Holtzman JL. The effect of alcohol on acetaminophen hepatotoxicity. Arch Intern Med 2002;162: 1193.
Rostom A, Dube C, Jolicoeur E, Boucher M, Tugwell P, Wells G, et al. Non-steroidal anti-inflammatory drug-induced gastroduodenal toxicity. In: Evidence Based Gastroenterology and Hepatology. London: BMJ Books (in press).
Man-Son-Hing M, O'Connor AM, Drake E, Biggs J, Hum V, Laupacis A. The effect of qualitative v quantitative presentation of probability estimates on patient decision-making: a randomized trial. Health Expect 2002;5: 246-55.
Tugwell P, Shea B, Boers M, Brooks P, Simon LS, Strand V, Wells G, eds. Evidence-based rheumatology. London: BMJ Books, 2004: xiii-xxix(Peter Tugwell, professor )
Correspondence to: ptugwell@uottawa.ca
Mrs Patell is a 64 year old woman with osteoarthritis of the right knee. A year ago, her general practitioner had recommended up to 4000 mg/day of paracetamol (acetominophen) for pain localised at the medial compartment of the knee. Mrs Patell had been taking the paracetamol but had forgotten that her general practitioner had mentioned that it should not be taken with alcohol. She has recently seen a newspaper article warning people about the risks of chronic use of analgesics. The article warns against liver damage when taking high doses of paracetamol in association with alcohol.
After reading the newspaper article she became worried, because she enjoys a cocktail before dinner and shares a bottle of wine while eating with her husband. She knew she needed to take something; her pain was intolerable with lower doses and the pain also limits her abilities to help her husband with routine tasks around the house.
This case is fictional but was developed from several real cases. It was commissioned to contribute to the special issue on harms that will be published on 3 July 2004.
Her friend recommended trying non-steroidal anti-inflammatory drugs for pain relief because these drugs have no alcohol restrictions. The non-steroidal anti-inflammatory provided better relief than the paracetamol, but her pharmacist warned her that it could cause serious bleeding problems. Confused, Mrs Patell seeks advice from her general practitioner.
Her general practitioner agrees that she needs a chronic analgesic but points out that there are benefits and risks from high doses of both paracetamol and non-steroidal anti-inflammatories. He explains the concept of the "number needed to treat" and tells her that, compared with placebo, the number needed to treat for improvement in pain at rest is 2 with up to 4000 mg of paracetamol and 1 with non-steroidal anti-inflammatories.1 He also tells her that there are case reports of serious liver toxicity in people taking therapeutic doses of paracetamol; some authorities have associated the toxicity with alcohol,2 3 but the association is controversial (Nick Bateman, personal communication, 2004). In addition, a recent review of the literature about non-steroidal anti-inflammatories cites the chance of hospital admission or death due to a serious gastrointestinal event at up to 16 per thousand a year.4
To further explain the evidence about the benefits of the treatment options to Mrs Patell, her general practitioner uses a diagram5and grades the strength of the scientific evidence using platinum (strongest), gold, silver, or bronze (weakest) ribbons (figure).6 The diagram shows the effect on 1000 people with no treatment, 4000 mg/day of paracetamol, and non-steroidal anti-inflammatories.
Diagram using a block of 1000 figures to describe benefits and harms of treatment for osteoarthritis of knee. Each dot stands for one person. NSAID=non-steroidal anti-inflammatory drug
If 1000 people did not have any treatment over the next year, then about 90 would improve on their own (silver) and no one would have serious side effects due to treatment. If 1000 people like her took up to 4000 mg/day of paracetamol for one year, then about 730 people would improve (silver), as was the case for Mrs Patell. However, about 1 person, who also consumed the same amount of alcohol as Mrs Patell, might develop liver damage, which could lead to death in a fifth of cases (bronze).2
The evidence suggests non-steroidal anti-inflammatories may work better for her than paracetamol and she would be able to drink alcohol, but they also have a higher chance of serious side effects than paracetamol. Her general practitioner concludes that the choice depends on the importance Mrs Patell places on the improved pain relief with non-steroidal anti-inflammatories compared with their harms. Mrs Patell says she definitely needs treatment, but she is not sure what to choose. She wants more time to think it over and asks for written information. In the meantime, she will go back to using paracetamol and try to reduce her alcohol intake.
Questions
In Mrs Patell's case, which option has the most favourable benefit:harm ratio?
Does it make a difference to you whether you think of the benefit:harm ratio as number needed to treat versus number needed to harm or number/1000 benefiting versus number/1000 harmed?
How would you satisfy Mrs Patell's desire for more information?
Please respond through bmj.com
We thank George Wells and Joan Peterson for help with the calculations.
Competing interests: PT received travel and research support from pharmaceutical companies. AOC receives an unrestricted research grant from the Foundation for Informed Medical Decision Making, which has a licensing agreement with Health Dialog, a company that markets decision aids.
This is the first of a three part case report where we invite readers to take part in considering the diagnosis and management of a case using the rapid response feature on bmj.com. Next week we will report the case progression and in four weeks' time we will report the outcome and summarise the responses.
References
Towheed TE, Judd MJ, Hochberg MC, Wells G. Acetaminophen for osteoarthritis. Cochrane Database Syst Rev 2003;(2): CD004257.
Zimmerman HJ, Maddrey WC. Acetaminophen (paracetamol) hepatotoxicity with regular intake of alcohol: analysis of instances of therapeutic misadventure . Hepatology 1995;22: 767-73.
Holtzman JL. The effect of alcohol on acetaminophen hepatotoxicity. Arch Intern Med 2002;162: 1193.
Rostom A, Dube C, Jolicoeur E, Boucher M, Tugwell P, Wells G, et al. Non-steroidal anti-inflammatory drug-induced gastroduodenal toxicity. In: Evidence Based Gastroenterology and Hepatology. London: BMJ Books (in press).
Man-Son-Hing M, O'Connor AM, Drake E, Biggs J, Hum V, Laupacis A. The effect of qualitative v quantitative presentation of probability estimates on patient decision-making: a randomized trial. Health Expect 2002;5: 246-55.
Tugwell P, Shea B, Boers M, Brooks P, Simon LS, Strand V, Wells G, eds. Evidence-based rheumatology. London: BMJ Books, 2004: xiii-xxix(Peter Tugwell, professor )