Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Ketorolac, not yet the gold standard
- Peter Leman (22 November 2000)
-
Ethical considerations
- Eddy Lang (24 November 2000)
-
Author's reply
- Timothy H Rainer (24 November 2000)
-
Author's reply
- Timothy Rainer (29 November 2000)
-
Other NSAIDs are cheaper, safer, and just as effective as ketorolac
- Brendon Smith (30 November 2000)
-
Use a balance approach to get the best possible result
- Nanda Gopal Mandal (1 December 2000)
-
In defence of morphine for severe pain following limb injury
- Stephen Milner (1 December 2000)
-
Why are we trying to replace opioids with ketorolac?
- Kevin McDonald, J R Worthington (12 December 2000)
-
Ketorolac and cancer pain
- Sundar Santhanam (18 December 2000)
-
ketorolac vs morphine in A+E
- Simon Clarke (16 January 2001)
|
|
|||
|
Peter Leman, Consultant in Emergency Medicine St Thomas' Hospital, London
Send response to journal:
|
Editor, The recent paper regarding the effectiveness of ketorolac and morphine for treating pain after limb injury [1], with the accompanying editorial [2] presents some interesting findings. The authors concluded that there was no difference in analgesic efficacy between the two drugs, and that there were fewer side effects with intravenous ketorolac. Unfortunately, the authors do not appear to have performed a power calculation prior to commencement of the study and thus the possibility of a Type II error has not been ruled out. Thus, although the authors showed no difference in analgesic efficacy, such a difference may still exist. Of note is that from Table 2 the hazard ratios favoured morphine for pain relief at rest from 50%, 75% and 100% reduction in pain. Thus, the case for equal analgesic efficacy has not been proven. The validity of the results presented as pain relief with activity ('the research nurse gently moving the injured limb') is of doubtful statistical, clinical or possibly ethical validity. Non-random sampling, daylight oversampling, improper blinding procedures, inadequate descriptions of the 33 non-participants, and an ethnically homogenous sample cast serious doubt on the external validity of the studies findings. The population studied does not reflect the normal emergency department population with a clinical indication for intravenous morphine administration. Over one third of patients had soft tissue injuries only and a further third had minor hand, foot or clavicular fractures. Thus, very few patients with severe injury were actually studied. The generalisability of these results to patients with severe injury would thus be open to question. Though the admission costs of admitted patients were studied, no basis, time-frame, or cost assumptions were explicated. Even the perspective from which this cost analysis was conducted was never stated. Also, no mention has been made of continuing analgesic requirements, rescue medication use or the continuing side effect profile of further NSAID analgesia compared with the decreasing sedation seen with further morphine administration. If these morphine related side effects were severe, then the use of naloxone could have been a useful outcome measure. It is perhaps the lack of antidotes to rare NSAID related adverse effects that have led to ketorolac being banned in France, Germany, Greece, Portugal and the Netherlands. On a final note, the preparation time for each drug when compared had a median difference of 2.0 minutes, the confidence interval (0-5.0) includes zero and thus the possibility of no difference in preparation time between the two drugs exists. However, the p value given of 0.0002 doesn't seem to reflect this degree of uncertainty. It is also of statistical interest why repeated measures ANOVA was not employed as it the standard approach in analysing pain reduction over time. The '50%' or '75%' pain reduction metric employed is both artificial and unvalidated. Overall, this interesting and well-crafted study has explored critically the concept of a gold standard in emergency analgesia. Before we move towards rejection of this gold standard however, more robust studies should be performed, that focus on more severe injury, in less selected patient groups, with evaluations that explicitly and vigorously analyse cost and effectiveness in a statistically responsible manner. Dr Peter Leman MB BS MRCP FRCSEd FFAEM DFPHM MSc
Associate Professor Gregory Luke Larkin MD MSPH FACEP
[1] Rainer TH, Jacobs P, Ng YC, Cheung NK, Tam M, Lam PKW, Wong R, Cocks RA. Cost effectiveness analysis of intravenous ketorolac and morphine for treating pain after limb injury: double blind randomised controlled trail. BMJ 2000;321:1247-51. [2] Jelinek GA. Ketorolac versus morphine for severe pain. BMJ;321:1236-7. |
|||
|
|
|||
|
Eddy Lang, attending physician Sir Mmortimer B. Davis Jewish General Hospital
Send response to journal:
|
Editor, The concern I would like to raise regarding the interesting and well- designed cost-effectiveness trial reported in this issue of BMJ relates to the way in which pain was measured. As noted in the methods section, pain with activity was measured when the nurse caring for the patient performed a series of repeated movements to the injured limb in order to assess this aspect of patient discomfort. The investigators note that this movement was "gentle" but I question whether it was truly necessary to include this type of assessment. Does it really add to our appreciation of analgesic efficacy? This study was endorsed by the hospital's research ethics committee but I wonder if this aspect was overlooked. I know that if I were asked to participate in a trial where my painful and fractured limb would be mobilized a number of times in a 6-hour period, I might choose not to provide consent. The 10% rate on refusal to participate is very interesting in this regard. |
|||
|
|
|||
|
Timothy H Rainer, Associate Professor Chinese University of Hong Kong
Send response to journal:
|
EDITOR Leman and Larkin express reservations about the methodology, validity, generalisability, cost and statistical analysis of our paper. I am not sure whether they had access to the electronic long version of the paper or to the peer review summaries and replies (extra: Prepublication history) which appeared only on the BMJ's website, but important data omitted from the 'short' version is in the public domain and presented at this site. Firstly, this study included cost-effectiveness, analgesic efficacy (at rest and with activity), adverse events and patient satisfaction outcomes. As such we sought to address the problem of treating pain from the perspective of many interested parties, including health administrators, physicians and patients, which must be approaching the ideal for clinical research. Secondly, to publish power calculations for each of these outcomes would have been ideal but placed too many demands on publishing space. A pilot study was performed on 40 subjects prior to initiating the main study in which we sought to identify factors relevant to all the outcomes detailed above but especially with respect to our primary purpose - a cost- effectiveness analysis. As a crude indicator of differences in cost we studied the mean difference in time spent in the department between the two groups, the time taken to prepare or treat related effects, and assessed costs related to these two effects. Data was heavily skewed but after log transformation, the mean difference in time spent in the department between the two groups for time spent in the department was 0.17 (SD 0.48), and for the effect of drugs was 0.62 (SD 0.52). In order to achieve 80% power using a two group t-test, with a 0.05 two sided significance level, we needed 120 patients in each group using the first method, and 13 patients in each group using the second! In this crude pilot study, we identified some additional departmental and pharmacy costs which suggested that cost differences between ketorolac and morphine could be even greater than initially suspected such that even smaller sample sizes would be sufficient. Results from this pilot study were complex but consistently 'favoured' ketorolac. With respect to analgesic efficacy, our pilot study suggested a discrepancy in effect between ketorolac and morphine depending upon whether we assessed pain at rest or with activity. This discrepancy is unexplained but was present again in the main study. The difference between clinical and statistical relevance is discussed in the long version of the paper. Based on our pilot data the difference in mean rate of change in pain score between ketorolac and morphine at rest was 2.5 (SD6.4) suggesting the need for 100 subjects per group to exclude a Type II error. However, Todd showed that, in the opinion of clinicians, anything less than an 18mm (SD 16) difference in pain score was unlikely to be clinically significant [1]. In order to demonstrate such a clinical difference with 80% power using a two group t-test, and with a 0.05 two sided significance level, we would need 13 subjects per group. Clearly in our study we achieved this. We found that 30 minutes after the administration of analgesia, the difference in pain score between ketorolac and morphine with activity was > 20mm and favoured ketorolac (data not shown). At rest the difference between ketorolac and morphine was not > 14 mm at any time point, so even if statistical significance had been achieved with greater numbers, it is doubtful that this would have been clinically relevant. Our conclusions were that no great difference in analgesic effect was evident between the two drugs with the methods and doses used in this study, although with activity ketorolac may be marginally better. Thirdly, with respect to the questionable ethical validity of moving an injured limb and thus causing pain. Clearly limbs need to be moved even if only minimally in order to perform an adequate clinical examination, to apply supporting splints, and to perform appropriate radiological imaging. If it is ethical to allow procedures which are even slightly painful as part of clinical management, then it should be ethical, with the patient's consent, to study pain with movement. Our research nurse was trained to assess pain from small, limited range of movements and over very brief periods of time. Fourthly, consecutive samples were taken and so random sampling was not appropriate. Night-time sampling and higher levels of blinding were not practically nor financially feasible. The 33 non-participants did not wish to participate in the study and so we did not feel it appropriate to collect detailed data. Both soft tissue and bony injuries to the hand, foot and clavicle can be extremely painful and should not be excluded from such studies. The generalisability of our data was commented upon in the long version of our paper. Fifthly, we provided detailed explanations of the cost-effectiveness methods and analyses which are presented in the long version of our paper, but because of restrictions in space, the editorial staff of the BMJ could not include all relevant data in print. If patients have moderate to severe adverse effects which do not affect their conscious level with associated respiratory depression or hypotension then it is not appropriate to use naloxone to reverse the effect. Nausea, vomiting and dizziness should be treated with anti-emetics etcetera. Sixthly, a median (not a mean) difference with confidence intervals including zero may yet be highly significant statistically. The summary measures and statistical methods used in this article were more appropriate than ANOVA although this could have been used for some of the analysis. The methods of describing pain are practical, relevant and have been frequently used by experts in pain research. Finally, Leman and Larkin need to make their minds up. Either this study has been well-crafted or, as they imply, our cost-effectiveness analysis with its statistical analysis is unresponsible! 1. Todd KH, Funk JP. The minimum clinically important difference in physician-assigned visual analog pain score. Acad Emerg Med 1996;3;142-6 |
|||
|
|
|||
|
Timothy Rainer, Associate Professor Chinese University of Hong Kong
Send response to journal:
|
EDITOR Lang's concerns regarding the ethics of assessing pain with activity within an emergency department are considerate and important. Clearly, in clinical practice, limbs need to be moved even if only minimally in order to perform an adequate clinical examination, to apply supporting splints, and to perform appropriate radiological imaging. However is additional pain, over and above basic essential clinical procedures, necessary for research purposes? Firstly, for many patients, including a proportion in our study, pain associated with an injury may be minor to moderate at rest but increase to severe levels with movement. If the assessment of pain had been limited to times at rest many of our subjects would have been excluded from the study and the true evaluation of analgesic effect when patients most needed it would never have been assessed. Their pain would have been undervalued. Secondly, our study demonstrated a small difference between the effectiveness of ketorolac and morphine in relieving pain at rest with pain on movement. Whether this is clinically significant is unclear, but it does suggest that analgesics that are effective at relieving rest pain may not be so effective at relieving pain on movement. If we are to treat our patients effectively, then we need to study them at critical times that are as close to real life as reasonably possible. Therein lies the difficulty because what is reasonable for one person may be unreasonable for another. Thirdly, if it is ethical to allow procedures which are even slightly painful as part of clinical management, then it should be ethical, with the patient's consent, to study pain with movement. Our research nurse was extremely gentle and cautious and my personal opinion having seen her working is that this is ethically reasonable. Fourthly, almost all the patients who declined to be involved in the study also declined immediate strong intravenous analgesia. The reason for exclusion was that they did not want any analgesia for their pain, rather than that they did not want their injury moved more than clinically necessary. All patients were free to leave the study at any time if they wished but none chose to do so. |
|||
|
|
|||
|
Brendon Smith, Staff Specialist, Emergency Department Bankstown Hospital, Australia
Send response to journal:
|
The recent randomised study (1), and accompanying editorial (2), comparing ketorolac and morphine for analgesia in limb injuries after injury provide important data on role of NSAIDs. They conclude that ketorolac has fewer adverse effects compared to morphine and is cheaper overall despite higher drug costs. The choice of NSAID is of interest, as studies have shown that NSAIDs are generally equivalent in analgesic effect, and specifically that ketorolac is no better than ibuprofen in treating acute pain (3,4,5) . Ketorolac is substantially more expensive ($47.59 per 30mg injection v $0.12 per 400mg orally for ibuprofen in Australian prices) (6). Ketorolac has greater toxicity with a relative risk of upper GI bleeding of 24.7 compared to 2.1 for ibuprofen over baseline (7) , and a relative risk of 10.3 for peptic ulcer complications compared to ibuprofen, which is he safest NSAID (8). IV administration is not always indicated and suppository formulations are an alternative option. It may be more prudent to use equipotent agents that are both cheaper and safer than ketorolac to improve on the benefit described. REFERENCES 1.Rainer TH, Jacobs P, Ng YC, Cheung NK, Tam M, Lam PKW, et al. Cost effectiveness analysis of intravenous ketorolac and morphine for treating pain after limb injury: double blind randomised controlled trial. BMJ 2000; 321: 1247-1251 2. Jelinek, GA. Ketorolac versus morphine for severe pain. BMJ 2000 321: 1236-1237. 3. Wright JM, Price SD, Watson WA. NSAID use and efficacy in the emergency department: single doses of oral ibuprofen versus intramuscular ketorolac. Ann Pharmacother. 1994;28(3):309-12. 4. Turturro MA, Paris PM, Seaberg DC. Intramuscular ketorolac versus oral ibuprofen in acute musculoskeletal pain. Ann Emerg Med. 1995;26(2):117-20. 5. Neighbor ML, Puntillo KA. Intramuscular ketorolac vs oral ibuprofen in emergency department patients with acute pain. Acad Emerg Med. 1998;5(2):118-22. 6. Australian Schedule of Pharmaceutical Benefits. 7. Garcia Rodriguez LA, Cattaruzzi C, Troncon MG, Agostinis L. Risk of hospitalization for upper gastrointestinal tract bleeding associated with ketorolac, other nonsteroidal anti-inflammatory drugs, calcium antagonists, and other antihypertensive drugs. Arch Intern Med. 1998;158(1):33-9. 8. Henry D, Lim LL, Garcia Rodriguez LA, Perez Gutthann S, Carson JL, Griffin M, Savage R, et al. Variability in risk of gastrointestinal complications with individual non-steroidal anti-inflammatory drugs: results of a collaborative meta-analysis. BMJ. 1996;312(7046):1563-6. |
|||
|
|
|||
|
Nanda Gopal Mandal, Specialist Registrar Southampton
Send response to journal:
|
Editor Rainer and colleagues 1 found that the use of intravenous morphine was associated with dose related side effects and delayed discharge. They also noted that ketorolac was a better alternative and cost effective. They used upto 30-mg of ketorolac as an initial dose. However, the initial recommended dose of ketorolac is only 10 mg 2. Thus, the 30-mg dose was unnecessary and it could be associated with more dose-related side effects. In their study, O’ Hara and colleagues 3 found that both 10 mg and 30 mg of ketorolac were equal in analgesic effect to 12 mg of morphine. It may be due to the fact that further analgesia could hardly be achieved when the dose of ketorolac is increased beyond 10 mg. An alternative approach to deal with this problem is to introduce the concept of “balanced analgesia,” which is a common practice in the management of acute post-operative pain. This can be achieved by using a combination of drugs, which act at different sites along the pain pathway. The combination of drugs may act synergistically and thus their individual dose can be reduced thereby, decreasing the incidence of side effects. As the ketorolac is associated with minimum side effects it could be used in maximum allowable dose as the first line of analgesic. If the pain is not controlled after administrating the maximum recommended dose of ketorolac then a tritrating dose of morphine could be supplemented. This approach could minimise the dose-related side effects of morphine and avoid unnecessary use ketorolac. This may prove to be cost effective too. References: 1. Rainer TH, Jacobs P, Ng YC, Cheung NK, Tam M, Lam PKW, Wong R, Cocks RA. Cost effectiveness analysis of intravenous ketorolac and morphine for treating pain after limb injury: double blind randomised controlled trial. BMJ 2000; 321: 1247-1251. 2. British Medical Association, Royal Pharmaceutical Society of Great Britain. British National Formulary. London: BMA, RPS, 2000. (No 39.) 3. O’ Hara DA, Fragen RJ, Kinzer M, Perberton D. Ketorolac tromethamine as compared with morphine sulfate for treatment of postoperative pain. Clin Pharmaco Ther 1987; 41: 556-9. Dr Nanda Gopal Mandal, |
|||
|
|
|||
|
Stephen Milner
Send response to journal:
|
Editor - Rainer et al conclude that ketorolac is as effective as morphine for pain due to limb trauma, with a lower incidence of side effects and higher cost effectiveness (1). These findings may be in part related to the design of their study. The subjects studied included some with severe long bone fractures, and others with soft tissue injuries and more minor fractures. The data analysis does not clarify whether ketorolac was as effective as morphine in the subgroup of subjects with long bone fractures who might be expected to suffer the most severe pain. Whilst the dosage regimen for morphine used in this study was in accordance with the advice given in the British National Formulary (2), a maximum titrated intravenous dose of 15mg of morphine sulphate is sometimes insufficient to relieve severe pain after a long bone fracture in young, fit sportsmen with a high lean body mass. With a higher maximum permitted dose of 20mg or 25mg, morphine may have been shown to be a superior analgesic for severe pain. The side effects profile of ketorolac is undoubtedly superior to that of intravenous morphine, and Rainer et al state that this is the main reason that ketorolac was more cost effective. However, concomitant administration of an anti-emetic such as cyclizine effectively reduces opiate-related nausea and vomiting, adds insignificant cost, and is routine in many Accident and Emergency Departments. If all these factors are taken into account, the case against morphine becomes less clear cut. No single analgesic will be suitable for every patient with severe pain following limb trauma, and it may be that a combination of analgesics would prove superior. It is important to remember that the most effective form of analgesia following a long bone fracture, and the one with the most favourable side effect profile, is splintage. However, I believe that morphine still has a place in the management of severe pain associated with long bone fractures. Stephen Milner 1. Rainer TH, et al. Cost effectiveness of intravenous ketorolac and morphine for treating pain after limb injury: double blind randomised controlled trial. BMJ 2000; 321: 1247-51. 2. British Medical Association, Royal Pharmaceutical Society of Great Britain. British National Formulary. BMA, RPS, 2000. (No. 40) |
|||
|
|
|||
|
Kevin McDonald, Pharmacist- critical care; Director- emergency medicine Ottawa Hospital, Ottawa, Canada, J R Worthington
Send response to journal:
|
Editor: We are writing in response to the recently published study on the cost effectiveness of intravenous ketorolac to treat limb injury pain in an emergency department in Hong Kong (1), as well as the accompanying editorial (2). This study is the first we have found that directly compares the rate of onset of pain relief of intravenous ketorolac to that of an intravenous opioid, in this case morphine. A complete (100%) reduction in pain at rest was achieved at rest in 92% of the ketorolac treated patients in a mean time of 20 minutes and in 93% of the morphine treated patients at a mean time of 15 minutes (difference not significant). Similar reductions in pain score with activity were considerably slower. Interestingly, when ketorolac is compared to other NSAIDs in treating pain in the emergency department (primarily musculoskeletal injuries), intramuscular ketorolac has been shown to be no faster and no more effective than oral ibuprofen (3-5). Assuming equivalent efficacy, oral ibuprofen would likely prove more cost effective than ketorolac (drug cost 1/100th, no IV access required, faster to give, less nursing time, no syringe costs). As for rate of onset, the accompanying editorial (2) includes a reference that intravenous NSAIDs relieve the pain of renal colic faster than NSAIDs given by other routes (IV vs IM and IV vs PR); review of the original literature indicates that ketorolac was not one of the NSAIDs tested, and of those tested, the differences were apparent only at 10 and 20 minutes but were not significant at 30 minutes. The same reference reviewed studies of NSAIDs used to treat other pain conditions and no other conclusions about differences between routes of administration could be drawn (6). In the design of the study, ketorolac or morphine doses were given every 5 minutes until adequate pain relief was achieved or an adverse effect was observed. The mean dose of study medications administered is not presented, supplemental doses of analgesics were not reported (the emergency physician was permitted to administer extra doses of medication if pain control was not considered adequate), and no reports of other pain control measures such as splinting or ice was mentioned. Based on the methodology, complaints of nausea or drowsiness (mild but frequently observed adverse effects) would stop further doses of morphine, meaning that optimal pain control with this agent may not have been achieved. Suboptimal pain control may have affected the patients’ perceived satisfaction score of morphine, rather than the adverse effects. It seems that a lot of attention has been paid in research to replace opioid use with injectable ketorolac. Statements in this study such as the risk of dependence to morphine makes opioids “dangerous” (1), which is echoed in the editorial (2) should not be considered in the situation of acute pain syndromes; the DSM-IV criteria for dependence will not be reached for a patient presenting to the ER with a painful injury who is given an opioid. The emergency physician, however, may feel more comfortable administering an injectable NSAID, such as ketorolac, to a patient who they believe is drug seeking, but these patients were excluded from the study. The main danger with opioids are the more severe adverse effects, such as respiratory depression, which can be prevented with care and be well managed in the emergency department. Studies should start examining the role of NSAIDs can play in targeting a different mechanism for pain and how they may save costs by reducing the dose of opioids required, subsequently reducing their adverse effects through this sparing effect, rather than replacing them. Evidence is emerging that the combination may achieve better pain control (7). Given what this study adds to other studies using intravenous or intramuscular ketorolac for pain syndromes in the emergency department setting (all pain syndromes), we feel the following statements can be made at this time: i) Ketorolac by injection (IM or IV) has not been shown to be superior (faster or more effective) to other NSAIDs administered by another route (oral or rectal). ii) Evidence is inconclusive that NSAIDs achieve the same level of pain control as effectively as opioids. iii) Used as single agents, NSAIDs tend to have fewer side effects in this acute setting. iv) Used as single agents, NSAIDs may have a cost advantage over opioids in this acute setting. v) An NSAID given by any route has an opioid sparing effect when used in combination. vi) Optimal pain control may be best achieved with early administration of an NSAID, followed by titration of opioids to the desired level of pain control. (1) Rainer TH, Jacobs P, Ng YC, et al. Cost effectiveness analysis of intravenous ketorolac and morphine for treating pain after limb injury: double blind randomised controlled trial. BMJ 2000;321:1247-1251. (2) Jelinek GA. Ketorolac versus morphine for severe pain: ketorolac is more effective, cheaper, and has fewer side effects. BMJ 2000;321:1236- 1237. (3) Neighbor ML, Puntillo KA. Intramuscular ketorolac vs oral iburofen in emergency department patients with acute pain. Acad Emerg Med 1998;5:118-122. (4) Turturro MA, Paris PM, Seaberg DC. Intramuscular ketorolac versus oral ibuprofen in acute musculoskeletal pain. Ann Emerg Med 1995;26:117- 120. (5) Wright JM, Price SD, Watson WA. NSAID use and efficacy in the emergency department: single doses of oral ibuprofen versus intramuscular ketorolac. Ann Pharmacother 1994;28:309-312. (6) Tramer MR, Williams JE, Carroll D, Wiffen PJ, Moore RA, McQuay HJ. Comparing analgesic efficacy of non-steroidal anti-inflammatory drugs given by different routes in acute and chronic pain: A qualitative systematic review. Acta Anaesth Scand 1998;42:71-79. (7) Cordell WH, Wright SW, Wolfson AB, et al. Comparison of intravenous ketorolac, meperidine, and both (balanced analgesia) for renal colic. Ann Emerg Med 1996;28:151-158. |
|||
|
|
|||
|
Sundar Santhanam, Specialist Registrar in Oncology Dept. of Oncology, Osborne Building, Leicester Royal Infirmary, Leicester LE1 5WW.
Send response to journal:
|
Dear Sir, Many of the issues explored by the Rainer et al, in the acute emergency department setting, are also relevant to other settings such as the palliative care (1). Morphine remains the gold standard in cancer patients with moderate to severe pain. Despite the prophylactic use of laxatives and antiemetics, morphine is often associated with many unpleasant side effects. These side effects such as constipation, nausea, sedation etc., impair the quality of life of the patients but the side effects are often overlooked by physicians (2). Cancer pain, unfortunately, is also opioid resistant in some situations (2). Ketorolac has been shown to be effective in morphine-resistant pain and also has morphine sparing effect (3) (4) (5). It can be given through various routes including continuous subcutaneous infusion (3). In spite of the evidence available, ketorolac and other parentral NSAIDS are under- used in cancer patients. A significant percentage of cancer patients are admitted to the oncology and palliative care wards for pain relief. The regular oral intake of medications by these patients is often poor. This is partly due to the local and systemic effects of advanced cancer itself, and partly due to the side effects of chemotherapy and radiotherapy. An increase in use of ketorolac and other parentral NSAIDS, as a primary as well as opioid adjunctive analgesia, would be potentially cost effective in these settings. This practice could reduce the length of hospital stay because of better pain relief and smaller number of opioid related adverse events resulting from morphine dose reduction. Needless to say, that the quality of life would be better for patients. Ref: (1) Rainer TH, Jacobs P, Ng YC, Cheung NK, Tam M, Lam PKW et al. Cost effectiveness analysis of intravenous ketorolac and morphine for treating pain after limb injury: double blind randomised controlled trial. BMJ 2000; 321: 1247. (2) Ahmedzai S. New approaches to pain control in patients with cancer. Eur J Cancer.1997; 33 Suppl 6:S8-14, 1997. (3) Myers KG, Trotman IF. Use of ketorolac by continuous subcutaneous infusion for the control of cancer-related pain. Postgrad Med J.1994; 70(823): 359-62. (4) Staquet M, De wasch G. A double blind comparison of multiple dose regimen ketorolac and buprenorphine in patients with cancer pain. Pain 1990;suppl 15: s353. (5) Joishy SK. Walsh D. The opioid-sparing effects of intravenous ketorolac as an adjuvant analgesic in cancer pain: application in bone metastases and the opioid bowel syndrome. J Pain Symptom Manage. 16(5): 334-9, 1998 |
|||
|
|
|||
|
Simon Clarke
Send response to journal:
|
EDITOR- Rainer et al [1] produced an interesting paper comparing the cost effectiveness of intravenous morphine with ketorolac in relieving pain after blunt limb injury. However we have a number of concerns regarding the study. Firstly, the morphine doses were given relatively rapidly. it is well recognised that side-effects from intravenous morphine are directly related to speed of administration [2] and in our practice titrating the dose at approximately 1mg per minute results in a much lower incidence of adverse effects than those quoted in the paper. Secondly, the dose of ketorolac given was relatively large. The current recommendations in the British National Formulary for giving parenteral ketorolac are for an initial bolus of 10mg, followed by subsequent doses after a minimum of 2 hours because it is recognised that with larger doses there is a significant risk of serious gastrointestinal effects [3-5]. We believe that with the doses used in the study there is a real risk of such effects. In addition, no mention was made of whether epigastric symptoms were sought either at the time of the study or in follow up. Thirdly, we disagreed with the conclusion regarding the cost analysis: it was stated that the overall mean cost per person including admission was $HK4081.58 more for ketorolac compared to morphine. This is approximately £400 per person, which is a significant difference when the potentially large numbers of such patients passing through an average emergency department are taken into consideration. We think that rather than feeling more confident in using ketorolac as opposed to morphine, doctors should become more confident in using morphine by learning to use it safely. S Clarke, specialist registrar;
Department of Accident and Emergency Medicine, South Manchester University Hospital. Wythenshawe, Manchester. [1] Rainer TH et al; Cost Effectiveness Analysis of Intravenous Ketorolac and Morphine for Treating Pain After Limb Injury: Double Blind Randomised Controlled Trial. BMJ,2000;321;1247-1251 [2] Narcotic Analgesics: the Pure Agonists. In: Pain Management in Emergency Medicine. Appleton-Lange,1987. [3] Strom BL et al; Parenteral Ketorolac and Risk of Gastrointestinal and Operative Site Bleeding. A Postmarketing Surveillance Study. JAMA,1996;275;376-382. [4] McCardle R; Intravenous Analgesia. Critical Care Clinics,1999;15;89-104. [5] Reinhart DI; Minimising the Adverse Effects of Ketorolac. Drug Safety,2000;22; 487-497. |
|||