Statistics from Altmetric.com
We were interested to read the paper by Watts and Armstrong1 that looked at the effectiveness of double Tubigrip in grade 1 and 2 ankle sprains. The authors' aim was to carry out a well designed randomised controlled trial that would determine whether functional recovery was achieved more rapidly with double Tubigrip or with no treatment. We have some reservations about their study design, the data collected, and the conclusions. Of the 485 patients who were eligible to enter only 197 completed the trial. This is equal to a 59% drop out rate. Basic validity of a controlled trial requires that at least 70% of those eligible complete the trial.2
The main outcome measures were crude by any standards. The only significant difference shown by their study was in the need for analgesia. But the question “ did you take pain killers?” can hardly be regarded as a sensitive tool for determining the amount of pain suffered by the patients. Had both groups of patients been issued with blister packs of a standardised analgesic and the number of tablets taken observed then a more accurate estimation of pain suffered could have been made. No attempt was made to re-evaluate the patients for ankle instability, which is the major complication of lateral ligament injuries. If we were to accept these findings and stop using double Tubigrip in ankle sprains, such information would be vital.
We were also concerned at the one week follow up. The authors themselves state that grade 1 and 2 ankle sprains can take 10 days to recover to a level where the patient can return to work and yet the mean number of days off work was less than four in both groups. This would suggest that these were extremely mild ankle sprains and not reflective of the whole spectrum of this injury. Therefore the accuracy of grading the injuries acutely rather than at review several days later must also be questioned.
We feel that overall this study adds little to the literature on ankle sprain management, and certainly does not provide any evidence for a change of current accepted practice.
We welcome the opportunity to respond to the comments made by Drs Lewis and Atkinson on our recently published paper.1
Firstly, as our Consort diagram clearly shows, 85 patients declined randomisation, were not entered and therefore did not form part of the study population. We included this information to illustrate the clarity of our study and because we felt it was interesting to note the number of patients who did not want to take part in a study aiming to recruit 400 patients. Our “drop out” rate was 203 of 400 patients—that is, 50.75%. This weakness is covered fully in our discussion, where we pointed out that our study might be best considered as a pilot study. Even so our study, with 197 patients, is still the largest to tackle this research question. Previous studies have also suffered from disappointing follow up rates2 and while using a funded research assistant may help, the fact remains that ankle sprains have a good prognosis3,4 with many patients not even seeking medical attention.5 Getting patients to consider their follow up telephone questionnaire or appointment as anything other than an inconvenience when they themselves feel significantly better is a difficult problem to solve.
We feel that Drs Lewis and Atkinson have missed the point when they criticise one of our chosen outcome measures. The fact that there was no difference between the groups is surely the point? We did find that significantly more analgesia was required in the double Tubigrip group and therefore obviously needed to discuss that finding. As regards the “crudeness” of that question, we return to the point made above on clarity. Our study was carried out without funding and with the invaluable help of non-medical A&E receptionists. For the reasons above we chose not to bring patients back for follow up appointments but to telephone them with a standardised questionnaire instead. As anyone who has designed a questionnaire will know, the wording and reading age is vital and we had ours checked by an educational psychologist, hence the choice of such a basic question as “Did you take painkillers for your ankle injury?”. Blister packs can get lost, patients may use them for other things or give them away and, of course, the patient needs to attend for follow up.
Do Drs Lewis and Atkinson routinely review all grade 1 and 2 ankle sprains for instability? In a study in which 150 patients with ankle sprains (not graded) were given no support but treated with elevation for 24 hours followed by motion exercise and weight bearing according to ability, 8% of patients had residual symptoms that they considered “inconvenient” at one year.6 The key to preventing chronic problems after ankle sprains is appropriate rehabilitation7 and the advice given to these patients is vital.
Drs Lewis and Atkinson also criticise our choice of follow up at one week and the fact that our patients seemed to return to work rather earlier than expected, suggesting that the accuracy of grading acute injuries rather than review several days later is questionable. However, the work used to quote anticipated recovery times for the different grades of injury was carried out almost a quarter of a century ago.8 As we said in our paper, the factors affecting patients' desire for an early return to work/activities is complex but there have clearly been massive social changes over the past 25 years that are highly likely to influence these variables. We believe our patients were graded accurately and we continued reinforcing the grading system throughout the trial period. Also, deferring assessment and grading of sprains for several days is not an option as there is agreement that the excellent prognosis of these injuries means patients presenting more than 24 hours after injury should be excluded from trial entry.
We feel that Drs Lewis and Atkinson's final comments are unjustified. Research does not have to be (and probably never can be) perfect to be published, but what is vital is that the authors describe and explain their study's shortcomings. Our study was transparent and honest and remains the largest one in this area to date. We feel that by stimulating discussion we have certainly added something to this debate.
Finally, we find Drs Lewis and Atkinson's last line ironic. Where is the evidence to make double Tubigrip “accepted practice” in the first place?
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.