|Home | About | Journals | Submit | Contact Us | Français|
Earlier this year the National Institute for Clinical Excellence (NICE) issued guidelines for use of diathermy in tonsillectomy, based on a nationwide audit that is still continuing. A letter issued by Mr Andrew Dillon, Chief Executive, and Professor Bruce Campbell, C Chair of the Interventional Procedures Committee, advised that 'all surgeons should consider how best to minimize their use of diathermy during the tonsillectomy' [www.nice.org.uk/page.aspx?o=203699]. As the surgeon who does most of the adult tonsillectomies in my hospital, I have found this advice both puzzling and disruptive. An immediate consequence was that the 'risk management' section of our trust expressed grave concern that diathermy (bipolar in our case) was being used in our department. Within a fortnight a departmental policy was created stating that diathermy may be made available only if all attempts at controlling bleeding with ligatures have failed. When does that point arrive? If one tries hard enough and keeps on ligating every red object in the tonsillar bed, all bleeds eventually stop. The surgeon now feels guilty even about mentioning diathermy.
Having looked at the 'evidence', I wonder why NICE was in such a hurry to release this interim guideline before reaching definitive conclusions. Take the chart providing figures for 'cold steel dissection with diathermy haemostasis' (the method I use). No distinction is made between monopolar and bipolar types even though the audit form differentiates between the two. We are told that patients operated in this way are 0.7% more likely to return to theatre with postoperative bleeding than those in whom ligatures only are used. Since the document later states that the monopolar type is twice as harmful as the bipolar type, we might reasonably assume that in patients receiving bipolar diathermy for haemostasis the excess risk of returning to theatre will be even less than 0.7%. Should we really change our practice on the basis of this very small difference and lose the advantages that bipolar haemostasis has to offer—notably, a much quicker operation and less preoperative blood loss (both matters on which the 'interim guidance' is silent)? The practical advantages of bipolar dissection have been well discussed by Silveira et al.1 In presenting its haemorrhage figures, NICE states that patients operated upon by trainees are roughly twice as likely to bleed as those operated on by non-training grades and consultants. Unfortunately, this difference is not allowed for in discussion of individual methods. If, for example, cold steel tonsillectomy with ligature haemostasis is done mainly by experienced surgeons while the younger generation favour diathermy, that might explain the apparent advantages of the former. Many aspects of the interim guidance are in conflict with recent studies. Both Belloso et al.2 and Timms3 have reported lower rates of postoperative bleeding with coblation than with cold-steel dissection and bipolar haemostasis. Moreover, in a review of the published work, Leinbach et al.4 showed no significant difference in postoperative bleeding between monopolar dissection and cold steel dissection.
By presenting its confusing interim guidance NICE breached a basic principle of audit that changes are effected only after the data collection phase of the audit is over; otherwise the final assessment is falsified. For example, whereas before the guidelines nearly all the tonsillectomies at our hospital involved diathermy, the procedure is now almost extinct - yet the data collection phase is yet not over. As it happens, the number of post-tonsillectomy bleeds has slightly increased, though (as with some of the differences in the NICE document) this may be due to chance.
The interim guidance [www.nice.org.uk/interimtonsillectomyguidance] was issued in response to a request from the Chief Medical Officers of England, Scotland and Northern Ireland to review urgently the use of diathermy tonsillectomy, and followed an interim analysis of the results of the National Prospective Tonsillectomy Audit which was carried out in England and Wales. The results1 suggested that there was a higher risk of secondary haemorrhage requiring readmission to hospital and return to theatre after tonsillectomy using diathermy techniques or coblation compared with techniques which use no diathermy either for dissection or haemostasis.
This advice was issued in conjunction with an accompanying letter from Professor Richard Ramsden, Chairman of the Audit Steering Group, in which he summarized the results [www.nice.org.uk/pdf/diathermytonsillectomyletterrichardramsden.pdf]. The interim guidance advised that all surgeons should consider how best to minimize their use of diathermy during tonsillectomy, particularly when diathermy is being used for both dissection and haemostasis. It highlighted that the risk may be particularly high for monopolar diathermy and surgeons should consider discontinuing this method. The risk may also be higher with currently available disposable diathermy equipment for tonsillectomy and again surgeons should consider discontinuing use of such equipment. In addition the Institute advised that the National Prospective Tonsillectomy Audit should continue and that all patients having tonsillectomy should be included.
In advising the Institute on this matter, the Interventional Procedures Advisory Committee were aware of the methodological challenges of assessing interim data, some of which have been raised by Mr Shahzad. The Institute will issue full guidance when a systematic review of the literature together with a more detailed analysis of the audit data is available.
In view of the concerns raised by the British Association of Otorhinolaryngologists, Head and Neck Surgeons' audit we consider that the Institute has taken a measured and appropriate approach to ensuring patient safety.