This article has Open Peer Review reports available.
Single-dose rofecoxib for acute postoperative pain in adults: a quantitative systematic review
© Barden et al; licensee BioMed Central Ltd. 2002
Received: 5 April 2002
Accepted: 9 June 2002
Published: 9 June 2002
Rofecoxib is a cyclo-oxygenase 2 selective inhibitor. This systematic review of rofecoxib in acute pain examined studies in adults of analgesic efficacy over six hours, the amount and quality of the evidence on extended duration of analgesia, and the quality and quantity of evidence on adverse events.
Cochrane Library (issue 4, 2001), Biological Abstracts (March 2002), MEDLINE (March 2002) and PubMed (March 2002) were searched using rofecoxib as a free text term. The area under the pain relief versus time curve was dichotomized using validated equations to derive the proportion of patients on rofecoxib 50 mg or placebo with at least 50% pain relief over six hours. This was used to calculate the number needed to treat for at least 50% pain relief over six hours for rofecoxib compared with placebo. Information on duration of analgesia and adverse events was also collected.
Five included trials investigated 1,118 patients, of whom 211 received placebo and 464 received rofecoxib 50 mg. The NNT for rofecoxib 50 mg was 2.3 (95% confidence interval 2.0 to 2.6). The weighted mean remedication time was 1.9 hours for placebo (126 patients), 7.4 hours for ibuprofen 400 mg (97 patients) and 13.6 hours for rofecoxib 50 mg (322 patients).
Rofecoxib at 2–4 times the standard daily dose for chronic pain is an effective single dose oral analgesic in acute pain. Limitations in trial reporting constrain conclusions about longer duration of analgesia and adverse event profile.
Acute pain has been studied in single dose designs first proposed by Beecher and colleagues [1, 2] and formalized by Houde and Wallenstein . The problem with single trials is that while they can demonstrate statistical superiority of analgesic over placebo, variation because of random chance means that, if small, they provide a poor estimate of the size of the analgesic effect . Combining results from clinically homogeneous trials in a meta-analysis gives an accurate estimate of the extent of the analgesic effect when sufficient numbers of patients have been randomized [4, 5].
Clinical trials in acute pain normally last four to six hours, because that is the duration of effect for most analgesics, whether injected or as tablets, and for simple analgesics, NSAIDs or opioids. Meta-analysis in acute pain has concentrated on the use of the area under the total pain relief versus time curve (TOTPAR), dichotomized into at least 50% pain relief or not . It is not necessarily the only measure available or the most relevant. The precedent exists at a primary research level for adopting remedication as an outcome. Bullingham and colleagues  and Gibb and colleagues  explored the measure of re-medication in post-surgical patients subject to multiple dosing. A 'rescue factor' design was refined to demonstrate both analgesic efficacy and assay sensitivity. . Farrar and colleagues  extrapolated from re-medication data in an effort to both define and demonstrate analgesic efficacy.
Despite these efforts, time to remedication has never become a standard outcome in analgesic trials. One reason for this might simply be the similarity in duration between commonly used analgesics. For rofecoxib, this might be different. In arthritis 12.5 mg or 25 mg of rofecoxib is given as a single dose once a day to control pain. The acute pain dose of 50 mg is claimed to have prolonged analgesic activity .
The purpose of this systematic review and meta-analysis of rofecoxib in acute pain was threefold. First to examine all acute pain studies of analgesic efficacy in adults over six hours for comparison with other analgesics in acute pain. Second to examine the amount and quality of the evidence presented on extended duration of analgesia. Third to examine the quality and quantity of evidence on adverse events.
QUORUM guidelines were followed . Possible studies for inclusion were sought through searching the Cochrane Library (issue 4, 2001), Biological Abstracts (March 2002), MEDLINE (March 2002) and PubMed (March 2002) using rofecoxib as a free text term. The search strategy used is detailed in Additional File 1. Abstracts were examined for possible inclusion if they were randomized trials conducted in an acute pain setting, used rofecoxib and a placebo or active comparator. Reference lists and review articles were examined for possible additional references, and in-house databases were also searched for papers.
Criteria for inclusion for postoperative pain were: full journal publication, randomized controlled trials which included single dose treatment groups of oral rofecoxib and placebo, double blind design, baseline postoperative pain of moderate to severe intensity, patients over 15 years of age, at least 10 patients per group, and the pain outcome measures of total pain relief (TOTPAR) or summed pain intensity difference (SPID) over 4–6 hours or sufficient data to allow their calculation. Pain measures allowed for the calculation of TOTPAR or SPID were a standard five point pain relief scale (none, slight, moderate, good, complete), a standard four point pain intensity scale (none, mild, moderate, severe) or a standard visual analogue scale (VAS) for pain relief or pain intensity. Also of interest was information on the time to remedication. For adverse events, the primary outcome sought was the proportion of patients experiencing any adverse event, with secondary outcomes of patients experiencing particular adverse events. Although adverse events are often reported inconsistently in acute pain trials , the outcome of any patient experiencing any adverse event was the least inconsistently reported.
Each report which could possibly be described as a randomized controlled trial was read independently by at least three authors and scored using a commonly-used three item, 1–5 score, quality scale . Consensus was then achieved. The maximum score of an included study was 5 and the minimum score was 2. Authors were not blinded because they already knew the literature.
For each trial, mean TOTPAR, SPID, VASTOTPAR or VASSPID values for each treatment group were converted to %maxTOTPAR by division into the calculated maximum value . The proportion of patients in each treatment group who achieved at least 50%maxTOTPAR was calculated using valid equations [16–18]. The number of patients randomized was taken as the basis for calculations, to produce an intention to treat analysis. The number of patients with at least 50%maxTOTPAR was then used to calculate relative benefit and NNT for rofecoxib versus placebo. The same methods were used for adverse events.
Relative benefit and relative risk estimates were calculated with 95% confidence intervals using a fixed effects model . Heterogeneity tests were not used as they have previously been shown to be unhelpful [20, 21], though homogeneity was examined visually . Publication bias was not assessed using funnel plots as these tests have been shown to be unhelpful [23, 24]. The number needed to treat or harm (NNT and NNH) with confidence intervals was calculated by the method of Cook and Sackett  from the sum of all events and patients for treatment and placebo. Cumulative calculation of NNT  was performed by adding studies by year of publication, and alphabetically within a year.
Relative benefit or risk was considered to be statistically significant when the 95% confidence interval did not include 1. NNT or NNH values were only calculated when the relative risk or benefit was statistically significant, and are reported with the 95% confidence interval. Calculations were performed using Microsoft Excel 2001 on a Power Macintosh G4.
Searches identified twelve papers that were excluded. The references and reasons for exclusion are given in Additional File 2. Five studies were included [11, 26–29]. Details of the design, numbers of patients, outcomes, analgesic results, adverse events and quality scores are given in Additional File 3. Four of the five included trials were in a dental pain setting after third molar extraction, and one  was after orthopaedic surgery. The five trials studied 1,118 patients, of whom 211 received placebo and 464 received rofecoxib 50 mg. Comparator analgesics included ibuprofen 400 mg (117 patients in three trials), celecoxib 200 mg (91 patients in one trial), paracetamol 600 mg plus codeine 60 mg (180 patients in one trial) and naproxen sodium 550 mg (55 patients in one trial). Quality scores were 4 for four trials and 5 for one. All the included studies were funded by Merck, the manufacturer of rofecoxib.
At least 50% pain relief
Number, number-needed-to-treat/harm and relative benefit/risk estimate for placebo controlled clinical trials of rofecoxib 50 mg
Number/total (%) with the outcome with
Number of trials
Rofecoxib 50 mg
Relative benefit/risk (95% CI)
NNT/H (95% CI)
At least 50% pain relief
4.8 (3.3 to 7.2)
2.3 (2.0 to 2.6)
Any adverse event
0.69 (0.44 to 1.08)
0.48 (0.26 to 0.88)
-11 (-6 to -40)
0.36 (0.16 to 0.79)
-12 (-7 to -46)
Onset of analgesia with rofecoxib 50 mg appeared to be generally the same as with comparator analgesics. Trials tried to discourage remedication within 90 minutes, and two patients in one trial were excluded because they remedicated within 90 minutes, one on placebo and one on rofecoxib 50 mg . This was not reported consistently (Additional File 3). The percentage of patients remedicating within two hours was given in one study , and within 24 hours in two [11, 27]. Only one study (.( gave cumulative remedication rates over the duration of the study. Median time to remedication was given in three studies [11, 27, 28], and this allowed the calculation of a mean of the three trials weighted by the number of patients. For placebo the weighted mean time to remedication from 126 patients in three trials was 1.9 hours. For rofecoxib 50 mg the weighted mean time to remedication from 322 patients in three trials was 13.6 hours. For ibuprofen 400 mg the weighted mean time to remedication from 97 patients in two trials was 7.4 hours.
Adverse events were not reported consistently (Additional File 3). Two trials reported the proportion of patients having any adverse event. An adverse event was reported by 33% of patients on placebo and 28% of those on rofecoxib 50 mg (not significantly different, Table 1). Nausea and vomiting were reported separately in three dental trials, and both occurred significantly less frequently with rofecoxib 50 mg than with placebo (Table 1). For both there were negative numbers needed to harm (-11 and -12 respectively), showing that for every 11 or 12 patients given rofecoxib, one fewer was nauseous or vomited than with placebo. There were few events, though, with only 44 nauseated patients and only 27 who vomited in the three trials of rofecoxib 50 mg comparable with 34 nauseated patients and only 22 who vomited in the two trials of ibuprofen 400 mg.
The five included studies were of generally high reporting quality (quality scores of 4 or 5), known to be associated with minimal reporting bias [30, 31]. They all had a placebo control, used standardized pain intensity and relief scores in adult patients with established pain of moderate or severe intensity, and measured outcomes over at least six hours. The pain models, third molar dental extraction and orthopaedic surgery, are commonly used in research and choice of model does not appear to influence results in single-dose analgesic studies when patients are able to take oral medication . Comparison of rofecoxib 50 mg with other analgesics over six hours is legitimate, because the same outcome was measured in the same way in the same patients over the same period of time, and with same comparator.
Rofecoxib 50 mg numbers-needed-to-treat compared with common analgesics
Drug and dose
Number of patients in the comparison
Rofecoxib 50 mg
2.3 (2.0 to 2.6)
Diclofenac 50 mg
2.3 (2.0 to 2.7)
Ibuprofen 400 mg
2.4 (2.3 to 2.6)
Morphine 10 mg (IM)
2.9 (2.6 to 3.6)
Paracetamol 1000 mg
3.8 (3.4 to 4.4)
Paracetamol 600/650 mg + codeine 60 mg
4.2 (3.4 to 5.3)
Aspirin 600/650 mg
4.4 (4.0 to 4.9)
For rofecoxib 50 mg this may not fully reflect its efficacy compared with these other analgesics, because rofecoxib 50 mg may have a longer duration of action. In the five included trials duration was inconsistently reported, however, with median remedication times in three trials (but without dispersion) and percentage remedicated at two hours in one trial and 24 hours in two trials. This disparity in reporting allowed only the crudest analysis of weighted mean remedication time. This was 1.9 hours for placebo, 7.4 hours for ibuprofen 400 mg and 13.6 hours for rofecoxib 50 mg. The figures of 7.4 hours for ibuprofen 400 mg and 1.9 hours for placebo is in agreement with a larger unpublished analysis (Barden et al, unpublished observations). The extended time to remedication for rofecoxib 50 mg may represent a real clinical advantage where the problem is not a lack of effective analgesics, but rather effective analgesics ineffectively delivered , and given that using a relatively large dose has no disadvantages.
The efficacy of rofecoxib in these five classical clinical analgesic assays is in contrast to two randomized comparisons of rofecoxib with placebo when given before the operation in adults  and children . The methodological difficulties and lack of effect of preemptive analgesia have been pointed out previously [38, 39]. In these two trials additional difficulties were the large between-individual variation in morphine consumption in adults , and the use of a novel (unlicensed) syrup formulation of rofecoxib in children . Lack of a demonstrable analgesic effect in preemptive studies does not negate a substantial postoperative effect when patients both have pain and can swallow.
There was no consistent pattern of adverse event reporting in the five included trials. Two trials had information on any patient with any adverse event (Table 1), with no difference between rofecoxib 50 mg and placebo. In three trials information on patients suffering nausea and vomiting was available, showing that these adverse events occurred less frequently with rofecoxib 50 mg than placebo. The amount of information (number of patients randomized) was insufficient to make any authoritative statement about adverse events, or compare the adverse event profile of rofecoxib 50 mg in single doses with other single dose analgesics. Adverse events in single-dose studies are often poorly reported, and of limited value .
In clinical practice, NSAIDs are associated with acute renal failure in patients with impaired renal function [41, 42] and with congestive heart failure in older people with a prior history of heart disease . Acute renal failure after surgery occurs in 1 patient in 1000 after major surgery , though it is not clear how much of this is specifically due to the NSAID. For coxibs, especially at high doses, similar caution should be used in patients with pre-existing renal or cardiac conditions as apply to the use of NSAIDs. The slope of the adverse event dose-response may be steeper than for analgesia.
These published reports were insufficiently detailed for calculation of numbers needed to treat for durations longer than six hours. Some information was available for TOTPAR calculated to eight hours, but not in a dichotomous format. TOTPAR is not normally distributed, and mean values are not meaningful . For times longer than six hours, and for duration of analgesia and adverse event profiles, more detailed information from individual patients would be required.
Rofecoxib at a dose 2–4 times the standard daily dose for chronic pain is an effective single dose oral analgesic in acute pain. Limitations in trial reporting constrain conclusions about longer duration of analgesia and adverse event profile.
The work was supported by Pain Research Funds, supported by the Oxford Pain Relief Trust.
- Beecher HK, Keats AS, Mosteller F, et al: The effectiveness of oral analgesics (morphine, codeine, acetylsalicylic acid) and the problem of placebo "reactors" and "nonreactors". J Pharmacol. 1953, 109: 393-400.Google Scholar
- Beecher HK: The measurement of pain. Pharmacological Reviews. 1957, 9: 59-210.PubMedGoogle Scholar
- Houde RW, Wallenstein SL, Beaver WT: Clinical measurement of pain. In: Analgetics New York and London Academic Press. Edited by: G De Stevens. 1965, 75-122.Google Scholar
- Moore RA, Gavaghan D, Tramer MR, et al: Size is everything–large amounts of information are needed to overcome random effects in estimating direction and magnitude of treatment effects. Pain. 1998, 78: 209-216. 10.1016/S0304-3959(98)00140-7.View ArticlePubMedGoogle Scholar
- Ioannidis JPA, Lau J: Evolution of treatment effects over time: Empirical insight from recursive cumulative metaanalyses. PNAS. 2001, 98: 831-836. 10.1073/pnas.021529998.View ArticlePubMedPubMed CentralGoogle Scholar
- McQuay HJ, Moore RA: An evidence-based resource for pain relief. Oxford Oxford University Press. 1998Google Scholar
- Bullingham RE, McQuay HJ, Dwyer D, et al: Sublingual buprenorphine used postoperatively: clinical observations and preliminary pharmacokinetic analysis. Br J Clin Pharmacol. 1981, 12: 117-122.View ArticlePubMedPubMed CentralGoogle Scholar
- Gibbs JM, Johnson HD, Davis FM: Patient administration of iv buprenorphine for postoperative pain relief using "Cardiff" demand analgesic apparatus. Br J Anaesth. 1982, 54: 279-284.View ArticlePubMedGoogle Scholar
- Savarese JJ, Thomas GB, Homesley H, et al: Rescue factor: A design for evaluating long-acting analgesics. Clin Pharmacol Ther. 1988, 43: 376-380.View ArticlePubMedGoogle Scholar
- Farrar JT, Portenoy RK, Berlin JA, et al: Defining the clinically important difference in pain outcome measures. Pain. 2000, 88: 287-294. 10.1016/S0304-3959(00)00339-0.View ArticlePubMedGoogle Scholar
- Malmstrom K, Daniels S, Kotey P, et al: Comparison of rofecoxib and celecoxib, two cyclooxygenase-2 inhibitors, in postoperative dental pain: a randomised placebo- and active comparator-controlled clinical trial. Clin Therap. 1999, 21: 1653-1663. 10.1016/S0149-2918(99)80045-9.View ArticleGoogle Scholar
- Moher D, Cook DJ, Eastwood S, et al: Improving the quality of reports of meta-analyses of randomised controlled: the QUOROM statement. The Lancet. 1999, 354: 1896-1900. 10.1016/S0140-6736(99)04149-5.View ArticleGoogle Scholar
- Edwards JE, Oldman A, Smith L, et al: Oral aspirin in postoperative pain: a quantitative systematic review. Pain. 1999, 81: 289-297. 10.1016/S0304-3959(99)00022-6.View ArticlePubMedGoogle Scholar
- Jadad AR, Moore RA, Carroll D, et al: Assessing the quality of reports of randomized clinical trials: is blinding necessary?. Control Clin Trials. 1996, 17: 1-12. 10.1016/0197-2456(95)00134-4.View ArticlePubMedGoogle Scholar
- Cooper SA: Single-dose analgesic studies: the upside and downside of assay sensitivity. In: The design of analgesic clinical trials (Advances in Pain Research and Therapy Vol. 18). Edited by: Max MB, Portenoy RK, and Laska EM. 1991, New York Raven Press, 117-124.Google Scholar
- Moore A, McQuay H, Gavaghan D: Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesics. Pain. 1996, 66: 229-237. 10.1016/0304-3959(96)03032-1.View ArticlePubMedGoogle Scholar
- Moore A, McQuay H, Gavaghan D: Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesics: Verification from independent data. Pain. 1997, 69: 127-130. 10.1016/S0304-3959(96)03251-4.View ArticlePubMedGoogle Scholar
- Moore A, Moore O, McQuay H, et al: Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesics: Use of pain intensity and visual analogue scales. Pain. 1997, 69: 311-315. 10.1016/S0304-3959(96)03306-4.View ArticlePubMedGoogle Scholar
- Morris JA, Gardner MJ: Calculating confidence intervals for relative risk, odds ratios and standardised ratios and rates. In: Statistics with confidence – confidence intervals and statistical guidelines London British Medical Journal. Edited by: Gardner MJ and Altman DG. 1995, 50-63.Google Scholar
- Gavaghan DJ, Moore RA, McQuay HJ: An evaluation of homogeneity tests in meta-analyses in pain using simulations of individual patient data. Pain. 2000, 85: 415-424. 10.1016/S0304-3959(99)00302-4.View ArticlePubMedGoogle Scholar
- Higgins J, Thompson S, Deeks J, et al: Statistical heterogeneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practice. J Health Surv Res Policy. 2002, 7: 51-61. 10.1258/1355819021927674.View ArticleGoogle Scholar
- L'Abbé KA, Detsky AS, O'Rourke K: Meta-analysis in clinical research. Ann Intern Med. 1987, 107: 224-233.View ArticlePubMedGoogle Scholar
- Tang J-L, Liu JLY: Misleading funnel plot for detection of bias in meta-analysis. J Clin Epidemiol. 2000, 53: 477-484. 10.1016/S0895-4356(99)00204-8.View ArticlePubMedGoogle Scholar
- Sterne JA, Gavaghan D, Egger M: Publication and related bias in meta-analysis: power of statistical tests and prevalence in the literature. J Clin Epidemiol. 2000, 53: 1119-1129. 10.1016/S0895-4356(00)00242-0.View ArticlePubMedGoogle Scholar
- Cook D, Sackett DL: On the clinically important difference. ACP J Club (Ann Intern Med). 1992, 117: A16-17.Google Scholar
- Ehrich EW, Dallob A, De Lepeleire I, et al: Characterization of rofecoxib as a cyclooxygenase-2 isoform inhibitor and demonstration of analgesia in the dental pain model. Clin Pharmacol Ther. 1999, 65: 336-347.View ArticlePubMedGoogle Scholar
- Morrison BW, Christensen S, Yuan W, et al: Analgesic efficacy of the cyclooxygenase-2-specific inhibitor rofecoxib in post-dental surgery pain: a randomized, controlled trial. Clin Therap. 1999, 21: 943-953. 10.1016/S0149-2918(99)80016-2.View ArticleGoogle Scholar
- Chang DJ, Fricke JR, Rea Bird S: Rofecoxib versus codeine/acetaminophen in postoperative dental pain: a double-blind, randomised, placebo- and active comparator-controlled clinical trial. Clin Therap. 2001, 23: 1446-1455. 10.1016/S0149-2918(01)80119-3.View ArticleGoogle Scholar
- Reicin A, Brown J, Mea Jove: Efficacy of single-dose and multidose rofecoxib in the treatment of post-orthopedic surgery pain. Am J Orthop. 2001, 40-49.Google Scholar
- Khan KS, Daya S, Jadad AR: The importance of quality of primary studies in producing unbiased systematic reviews. Arch Intern Med. 1996, 156: 661-666. 10.1001/archinte.156.6.661.View ArticlePubMedGoogle Scholar
- Moher D, Pham B, Jones A, et al: Does quality of reports of randomised trials affect estimates of intervention efficacy reported in meta-analyses?. Lancet. 1998, 352: 609-613. 10.1016/S0140-6736(98)01085-X.View ArticlePubMedGoogle Scholar
- McQuay HJ, Carroll D, Moore RA: Injected morphine in postoperative pain: a quantitative systematic review. J Pain Sympt Manag. 1999, 17: 164-174. 10.1016/S0885-3924(98)00126-2.View ArticleGoogle Scholar
- Collins SL, Moore RA, McQuay HJ, et al: Single dose oral ibuprofen and diclofenac for postoperative pain. Cochrane Database Syst Rev. 2000, CD001548:Google Scholar
- Moore A, Collins S, Carroll D, et al: Single dose paracetamol (acetaminophen), with and without codeine, for postoperative pain. Cochrane Database Syst Rev. 2000, CD001547:Google Scholar
- Bruster S, Jarman B, Bosanquet N, et al: National survey of hospital patients. Br Med J. 1994, 309: 1542-1546.View ArticleGoogle Scholar
- Huang JJ, Taguchi A, Hsu H, et al: Preoperative oral rofecoxib does not decrease postoperative pain or morphine consumption in patients after radical prostatectomy: a prospective, randomized, double-blinded, placebo-controlled trial. J Clin Anesth. 2001, 13: 94-97. 10.1016/S0952-8180(01)00219-7.View ArticlePubMedGoogle Scholar
- Pickering AE, Bridge HS, Nolan J, et al: Double-blind, placebo-controlled analgesic study of ibuprofen or rofecoxib in combination with paracetamol for tonsillectomy in children. Br J Anaesth. 2002, 88: 72-77. 10.1093/bja/88.1.72.View ArticlePubMedGoogle Scholar
- McQuay HJ: Pre-emptive analgesia: a systematic review of clinical studies. Ann Med. 1995, 27: 249-256.View ArticlePubMedGoogle Scholar
- Moiniche S, Kehlet H, Dahl JB: A qualitative and quantitative systematic review of preemptive analgesia for postoperative pain relief: the role of timing of analgesia. Anesthesiol. 2002, 96: 725-741.View ArticleGoogle Scholar
- Edwards JE, McQuay HJ, Moore RA, et al: Reporting of adverse effects in clinical trials should be improved. Lessons from acute postoperative pain. J Pain Symptom Manag. 1999, 18: 427-437. 10.1016/S0885-3924(99)00093-7.View ArticleGoogle Scholar
- Griffin MR, Yared A, Ray WA: Nonsteroidal antiinflammatory drugs and acute renal failure in elderly persons. Am J Epidemiol. 2000, 151: 488-496.View ArticlePubMedGoogle Scholar
- Henry D, Page J, Whyte I, et al: Consumption of non-steroidal anti-inflammatory drugs and the development of functional renal impairment in elderly subjects. Results of a case-control study. Br J Clin Pharmacol. 1997, 44: 85-90. 10.1046/j.1365-2125.1997.00631.x.View ArticlePubMedPubMed CentralGoogle Scholar
- Page J, Henry D: Consumption of NSAIDs and the development of congestive heart failure in elderly patients. Arch Intern Med. 2000, 160: 777-784. 10.1001/archinte.160.6.777.View ArticlePubMedGoogle Scholar
- Forrest JB, Camu F, Greer IA, et al: Ketorolac, diclofenac, and ketoprofen are equally safe for pain relief after major surgery. Br J Anaesth. 2002, 88: 227-233. 10.1093/bja/88.2.227.View ArticlePubMedGoogle Scholar
- McQuay H, Carroll D, Moore A: Variation in the placebo effect in randomised controlled trials of analgesics: All is as blind as it seems. Pain. 1996, 64: 331-335. 10.1016/0304-3959(95)00116-6.View ArticlePubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2253/2/4/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article: verbatim copying and redistribution of this article are permitted in all media for any purpose, provided this notice is preserved along with the article's original URL.