Continuing our series of interviews previewing this year’s ACPGBI annual meeting (9-11 July), we talked to Professor Dion Morton (Barling Professor of Surgery, University of Birmingham) about some of the topics he will raise when he presents this year’s BDRF Lecture: ‘What have randomised trials done for patients with colorectal disease?’.
“The opportunity that randomised controlled trials provide is high quality evidence that will impact directly on clinical practice. Surgery as a whole has been slow to develop evidenced-based medicine and has perhaps disadvantaged patients as a consequence, because the evidence has been insufficient to institute advances in a timely fashion,” Professor Dion Morton explained. “Historically, surgeons have relied on individual cohort data and if a procedure or device has worked then they perform more of them, but that is an inefficient and ineffective way of demonstrating benefit. In recent years delivery of surgery – and science as a whole – has increasingly required large, multidisciplinary teams and that increasing complexity requires more rigorous evaluation. There is still a large variation in practice and high-quality, evidence-based data derived from randomised controlled trials can encourage and persuade surgeons to change their practice. We need more rigorous evidenced-based surgery to ensure we are performing the right procedures, on the right patients.”
Despite the apparent advantages of randomised controlled trials, Professor Morton explained that there are difficulties in delivering such trials, namely: the time, rigour, effort and cost. In order to change practice, a randomised controlled trial should be comprehensive and should evaluate and demonstrate practice across the world, not just in one or two centres in the UK. For example, he highlighted three randomised trials of laparoscopic surgery (CLASSICC, COLOR and COST) that included thousands of patients at hundreds of hospitals in the UK, US and Europe. These trials changed surgical practice across the world by providing high-quality evidence to policy makers and stake-holders to encourage them to invest in the development of, and training for, laparoscopic surgery.
“The NHS is a uniform system for delivering healthcare and lends itself well for delivering randomised controlled trials,” he added. “We also have the a highly-developed infrastructure for establishing trials and colorectal surgery in the UK has developed more high-quality practice changing trials in the last five years than anywhere else in the world. Colorectal surgery, like other surgical specialties, is about to undergo transformational change in how it is delivered and performed. To safely implement and execute these changes, randomised controlled trials will be an essential component.”
He cited several examples of UK-based, colorectal surgery randomised controlled trials that have changed practice, not only in the UK, but across the world. Firstly, the CREST trial (led by Professor James Hill, Manchester) was the first trial in the world to show the benefits of endoluminal stenting devices for large bowel obstruction for colorectal cancer. Secondly, the first ever trial of robotic surgery, ROLARR (led by Professor David Jayne, Leeds), demonstrated the safety and potential benefits of this new technology particularly for difficult rectal resections (eg. patients with obesity, male patients with a narrow pelvis, low cancers). Interestingly, similar benefits have also been shown in gynaecology. In terms of peri-operative care, the DREAM trial showed dexamethasone reduces postoperative vomiting in patients undergoing elective gastrointestinal surgery. Conversely, the orthopaedic CSAW surgical trial revealed that some procedures are not benefitting patients and therefore should not be performed. The outcomes from these studies, Professor Morton added, would be impossible to show without a randomised controlled trial.
“During my presentation in Birmingham I will emphasise that the benefits of rigorous, high-quality, randomised controlled trials exceed the primary endpoint as they bring together large groups including hundreds of surgeons thinking about, learning and evolving their practice from their experience participating in the trial. This is something that drug trials do not achieve, surgical trials have the capacity to evolve and improves practice through the process of carrying out the study,” Professor Morton explained. “In addition, introducing complexity to clinical trials can allow us to measure multiple interventions at the same time. Historically, clinical trials have measured one intervention against another, but we are now designing trials where there maybe eight or more different interventions all been randomised between the patients. This means we do not concentrate on the benefits of one intervention but the potential benefits of a combination of interventions.”
An example of such a trial is ROSSINI II (led by Mr Thomas Pinkney, Birmingham), which will simultaneously investigate three in-theatre interventions and patients will be randomly allocated to receive all, none or some of these in any combination. In total, the trial is looking at eight different ways of preventing wound infection at the same time.
“These are the types of questions we can ask within complex trials,” Professor Morton added.
In his lecture, he will also highlight that surgical trials need to move away from being carried out in a very selective group of patients, so trials become embedded in day-to-day practice. This means that all patient data should be made available so that the trial data can be compared to the background data.
“This so-called ‘big data’ is about taking advantage of the huge amount of patient data and using it within trials. This will blur the boundaries between routine clinical practice and research-based clinical activity. This integration will benefit patients more quickly,” he explained. “Randomised controlled trials will continue to play an essential role in the development of evidenced-based surgery, but they will become embedded into routine clinical practice and be comparable to the registry data from those procedures.”