Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
"Auditing" in routine surgical practice
- Richard G Fiddian-Green (15 June 2002)
-
Ethical Problems of Surgical RCTs
- Daniel Polowetzky, RN, ACRN, 1250 Broadway, New York, New York 10009, USA (17 June 2002)
-
At last, sense and diamonds have been recovered from an ocean of nonsense
- Ahmed N. Ghanem, MD< FRCS Ed (20 June 2002)
-
Randomized trials in surgery : problems and solutions
- abd hamid mat sain (24 June 2002)
-
Randomised trials in surgery
- Richard G Faber (25 June 2002)
-
Trials must be done, we must make them happen
- Martin Elliott (26 June 2002)
-
Surgical trials: further thoughts
- Stephen A Bridgman, Nicola Maffulli (5 July 2002)
-
Need for an integrated approach
- Rajan Madhok, Helen H. G. Handoll (8 July 2002)
-
Insufficient evidence to warrant increased use of surgery?
- Robert W Leckridge (24 July 2002)
|
|
|||
|
Richard G Fiddian-Green, None None
Send response to journal:
|
Different factors influence outcome in different institutions and even in the same institution. Training and experience are not the only important variables. A consultant going on holiday, a poor locum, an incompetent registrar, a shortage of ICU nurses, winter illnesses, volumes of fluid and blood transfused during anaesthesia, the anaesthetist and the type of anaesthetic given, to name just a few variables, may also influence outcome. One or more of these variables is responsible for the very wide range in outcomes reported especially for complex operation such as a pancreatico-duodenectomy. Without comprehensive data collection and approariate statistical analysis it is impossible to establish the cause or causes of poor outcomes with any confidence. That applies to the Bristol and St George's deaths after cardiac surgery as well. Can there be any justification for not including continuous "auditing" in every surgeon's and every institution's practices? |
|||
|
|
|||
|
Daniel Polowetzky, RN, ACRN, Clinical Instructor Visiting Nurse Service of New York, 1250 Broadway, New York, New York 10009, USA
Send response to journal:
|
A discussion of the difficulties associated with surgical randomized controlled trials should mention inherent ethical problems. These cannot obviously be separated from issues of scientific rigour or methodology. For example, one of the key methods of avoiding bias in nonsurgical randomized controlled trials is the use of placebo with accompanying "double blinding". It would seem that such methodology when used in Type 3 trials (those comparing surgical with nonsurgical interventions)is unavoidably unethical, e.g "sham surgery", where there is absolutely nothing to be gained for the patient randomized to the "placebo" arm of the study. Further discussion of this ethical dimension would be welcomed. |
|||
|
|
|||
|
Ahmed N. Ghanem, MD< FRCS Ed, Consultant Urological Surgeon King Khalid Hospital, P O Box 1120
Send response to journal:
|
Sir, I read with interest this excellent article [1] on which I congratulate and commend the authors for diving deep and wide in order to recover sense and diamonds from an ocean of irrelevant nonsense. I fully support the reported views and recommendations. It is a shame that such landmark articles are not immediately implemented by concerned bodies to become standard law of medical research and advancement, in particular surgical research. The authors correctly concentrated on problems facing therapeutic surgical studies, and the need to differentiate from drug studies. However, it has perhaps already been realized that prospective observational diagnostic studies do significantly contribute to evidence based medicine (EBM), with or without randomization or statistics. Such studies also start from the day of encountering the first case, does not require the approval of anybody or committee and only patients’ consent is required when and as interference with approved diagnostic or therapeutic procedures is indicated. Contributions of such diagnostic studies to EBM include, identifying the problems of currently received protocols, proposing new ones with possible novel diagnostic signs or tests and suggestions for resolving problems as demonstrated in the authors’ article [1]. Observing new links between known diseases of idiopathic aetiology may reveal new patho- aetiology of importance in advancing new surgical therapy or modifying old one that had previously proved a failure because of overlooked such patho- aetiology or incorrect timing of the procedure. As many have been, and now more than ever might be, fishing for a conflict of interest in such research, please allow me give them one: it the “type lead poisoning” that inflicts an author to the degree of addiction, causing strong desire to contribute to science advancement and betterment of Global society as charity work. This is not just gas. Some do put their money where their mouth is, costing much of own money, time, efforts and health with no gain whatsoever in terms of career advancement or money making because it is simply of little concern or past it- as it never happened in the past when it was most needed. If, however, such unexpected gain or offer might occur in future it might be welcomed as unconditional grant, mostly spent on advancing current and further future research as judged ethical, appropriate and fit. Yours sincerely Ahmed N. Ghanem, MD (Uro), FRCS Ed. References 1. McCulloch P, Taylor I, Sasako M, Lovett B, Griffin D. Education and debate. Randomised trials in surgery: problems and possible solutions. BMJ 2002; 324: 1448-1451 (15 June) |
|||
|
|
|||
|
abd hamid mat sain, Associate Professor dept of surgery, sch of med sciences, universiti sains malaysia, kubang kerian, kelantan, malaysia 1
Send response to journal:
|
Dear Sir, The article on randomized trials in surgery by McCulloch et al is very pertinent indeed during this illuminated medical era of demand for scientific evidence to justify therapy. Peculiarities and difficulties in the practice of surgery in relation to clinical experimentation is well described. However, the authors have asserted that randomized controlled trials(RCT) are appropriate especially where a clear and clinically important choice exists between contrasting alternatives.On the contrary, the very place of RCT is mainly in the evaluation of potential marginal benefits of one therapy over the other based on theoretical assumption(1). The RCT is essentially conducted to prove or otherwise this assumption. Furthermore, the availability of adequate numbers of those cases is crucial for the design of RCT. There are many clinical situations in surgery where the theoretical advantages seem so overwhelming in favour of one therapeutic approach as compared to the other. Acute appendicitis as a main differential diagnosis in an obese female is an apt example. Conventionally, acute appendicitis is always thought as a condition which is essentially diagnosable by clinical examination alone despite so many attempt at introducing various levels of investigations. It was always stated that 10-20% of "white appendix" in conventional appendicectomy is acceptable. With the advent of laparoscopy equipped with modern optical technology, it is quite easy to see the enormous advantage of applying diagnostic laparoscopy in the clinical situation described above(2). The magnitude of theoretical advantage is so overwhelming to the extent that any attempt at performing an RCT in this situation may seem unethical. The example of hand-washing by Semmelweis is perhaps an excellent parallel in history. It is to be highlighted that many technical improvement in surgical therapy is merely a change in techniques and tools of doing the same thing. Quite often these techniques and tools should be added to the existing armamentaria to be used sequentially rather than competitively as needs dictate. In conclusion, as emphasized also by the authors, despite the apparent theoretical strength of RCT, which is perhaps applicable only in voluminous surgical problems, prospective cohort remains to be a more ethically appropriate design for a meaningful scientific appraisal of many surgical situations. References 1. Pollock AV. Historical Evaluation : Methods, Attitudes, Goals. In Principles and Practice of Research. 2nd Ed., Springer-Verlag, 6 2. Memon MA. Laparoscopic appendicectomy: current status. Ann R Coll Surg Engl 1997 Nov 79:393-402 |
|||
|
|
|||
|
Richard G Faber, Consultant surgeon Royal Berkshire Hospital RG1 5AN
Send response to journal:
|
Editor Congratulations on publishing this superb but politically incorrect paper. I fully support the whole thesis of the authors. I wish to add one point in support. Randomised trials and informed consent are mutually incompatable in operative surgery. By the time the surgeon has explained the options to the patient the vast majority decide to opt for one or the other arm and only a tiny minority are prepared to be randomised. |
|||
|
|
|||
|
Martin Elliott, Cardiac Surgeon & Director of Transplantation The Great Ormond Street Hospital for Children, London WC1N 3JH
Send response to journal:
|
McCulloch et al make a valuable point. Surgical research has been beset by poor design and the literature abounds with retrospective reports, and single centre experience. In paediatric cardiac surgery, dealing as it does with a very heterogenous group of patients, the problems are even more evident. Given the public expectations of us in particular, it really behoves us to try and get our data right. Whilst I find much with which to agree with in McCullochs article, I find myself more moved by the arguments of Lilford et al (BMJ 2000; 320:43)in favour of tracker studies. There are similarities between these two papers, each emphasising the need for flexibility and the use of observational data. Both require the use of randomisation. Tracker studies, however, seem to me better to reflect the pressures of real life. Other correspondents have argued that the consent for surgery process obstructs or confounds the request for randomisation. Surgeons should not object but rather find solutions. We must compete less and cooperate more. Compared with the progress in cooperation made by oncologists over the last 25 years, surgeons are still in the dark ages. |
|||
|
|
|||
|
Stephen A Bridgman, Senior Lecturer School of Medicine, University of keele, Medical Research Unit, Thornburrow Drive, Hartshill, Stoke-, Nicola Maffulli
Send response to journal:
|
Dear Sir, We found McCulloch et al’s article on randomised controlled trials (RCTs) in surgery useful and of great interest.1 We support many of their arguments. We add to their analysis some additional observations from our experience. A major issue is funding.2,3 Surgeons are partly to blame for this, and is a reflection of their general lack of epidemiological expertise. For instance, while the epidemiological and trial design reviews of research proposals may be excellent, surgical reviewers might comment, inappropriately, that they do not see the need for a particular trial as the options trialled may not be part of their normal practice. Such views do not help the surgical community. RCTs often need large numbers of patients to provide a scientifically satisfactory answer. By nature, surgeons tend to be competitive individualists. We therefore strongly support McCulloch et al’s view about the need to facilitate collaboration to answer questions in surgery that may have a major impact on public health and public health services. The current competitive UK University Research Assessment Exercise does not provide the environment which facilitates inter-unit collaboration, and does not provide adequate recognition for collaborating units in RCTs. The lack of commercial pressure to undertake research is another factor. Companies introducing new devices do not generally need to evaluate their interventions by RCTs. In our field, Trauma and Orthopaedics, there are hundreds of different manufacturers of essentially the same device(s). The high costs of RCTs disadvantage manufacturers who would like to better evaluate them: they would not be able to compete if they undertook randomised evaluations and their competitors did not. McCulloch et al were brave to raise the almost unspeakable: some surgeons may be strongly influenced by financial gain. In this respect, surgeons are no different than commercial companies: profit is king, not science and health. A challenge for the Surgical Colleges and their membership is to move forward from their traditional “gentleman’s club” culture to a truly scientific and health focussed one. They need to collaborate on priorities for research, lobby for the funds to sponsor these trials, include participation in trials as part of their accreditation of units and individuals, and ensure that academic surgeons are adequately trained in epidemiological methods. A barrier for academic surgeons is the inequality in their financial compensation compared to non -academic surgeons. These surgeons put themselves out on a limb, earn much less than jobbing surgeons, and meet barriers in the NHS to pursue excellence in surgery through trials and adequate research. Surgical trials involve an element of “skill” that pharmaceutical trials do not, and probably require more funding than pharmaceutical trials. In particular, McCulloch et al mention the need to video operations. We strongly support this - such videos should be considered part of the archive of evaluations of surgical interventions so that later scientists can satisfy themselves first hand. Finally, the NHS still has a strong element of commissioning by “bean” counting, irrespective of the scientific evidence justifying policies or interventions in individual cases. We spend millions on waiting-list initiatives. Perhaps, the NHS is missing a trick by not employing academic and evidence-based trained surgeons and epidemiologists to work on evidence-based demand management, with the reward of research funds for trials for those identifying key questions that will impact on demand. Dr Stephen Bridgman, Head of Surgical Trials, Epidemilogy and Public Health Reseach Unit, University of Keele, and Director of Public Health Newcastle-under-Lyme Primary Care Trust and Professor Nicola Maffulli, Professor of Orthopaedic Surgery University of Keele, and Consultant Orthopaedic Surgeon North Staffordshire Hospital Trust References Johnson AG, Dixon JM. Removing bias in surgical trials. BMJ, 1997, 314, 916-7. McCulloch P, Taylor I, Sasako M, Lovett B, Griffin D. Randomised trials in surgery: problems and possible solutions. BMJ, 2002, 324, 1448- 1451. Bridgman SA, Elder J, Gray R, Lilford R. Funding is important for randomised trials of surgery. BMJ, 315, 310, 1997. |
|||
|
|
|||
|
Rajan Madhok, Medical Director and Director of Public Health North and East Yorkshire and Northern Lincolnshire Health Authority, YO10 5DG, Helen H. G. Handoll
Send response to journal:
|
Dear Editor, As Cochrane reviewers appraising randomised controlled trials (RCTs) on surgical interventions for orthopaedic trauma and through our involvement with the promotion of evidence based orthopaedic surgery in Teeside, we would like to endorse and extend McCullouch et al.’s observations on RCTs in surgery [1]. Though issues specific to surgical trials, mentioned by McCullouch et al, also apply, most of the trials we have reviewed have methodological defects that could have been avoided. For example, concealment of study allocation is always possible, yet this was confirmed in just 2 of the 44 RCTs included in a review of surgical treatment of wrist fractures in adults [2]. Tackling any "lack of education in clinical epidemiology", and various other measures proposed by McCullouch et al. will go some way towards addressing the current state of affairs in research in surgery but more is needed. Surgeons should realise that using the right tools for clinical research is comparable to selecting and using the right instruments for an operation. Proper attention to study design, conduct, analysis and reporting is equally crucial. And, overall we need an integrated programme incorporating Research, Audit and Training - the RAT model [3]. This proposes that, given the aim of medical practice is to improve patient care by ‘Doing things that matter’, we have the responsibility to do three things: a) find out what matters - through primary or secondary research, b) apply the findings - and audit practice, and c) train clinicians if the right things are not being done the right way, every time. Keeping research, audit and training separate does little to improve patient care ultimately; it is better to develop integrated programmes starting with the most common conditions in the specialty. References 1. McCulloch P, Taylor I, Sasko M, Lovett B, Griffin. Randomised trials in surgery: problems and possible solutions. BMJ 2002; 324: 1448- 51. 2. Handoll HHG, Madhok R. Surgical interventions for treating distal radial fractures in adults (Cochrane Review). In: The Cochrane Library, Issue 2, 2002. Oxford: Update Software. 3. Madhok R, Stothard J. Promoting evidence based orthopaedic surgery: an English experience. Acta Orthop Scand Suppl (in press). |
|||
|
|
|||
|
Robert W Leckridge, Associate Specialist Glasgow Homeopathic Hospital G12 0XQ
Send response to journal:
|
This paper points out that less than 10% of published research in surgery is of the RCT methodology. The York Centre for Reviews and Dissemination studied the evidence for homeopathy in its recent Effective Health Care Bulletin. The Centre only studied "systematic reviews" of RCTs and considered other "evidence" to be unworthy of note. The BMJ reported this study with the headline "Use of homeopathy in the NHS not justified" based on the conclusion that there was insufficient RCT evidence of effectiveness of homeopathy "to warrant significant changes in the current provision of homeopathy." If surgery cannot be justified on the same basis ie insufficient RCT evidence, then should we now conclude there is not enough evidence to "warrant significant changes in the current provision of surgery"? and where does this leave the arguments about reducing waiting lists for surgical treatments and arguing for the development of "operation factories" and sending patients abroad for surgical treatments? |
|||