In My Signals, health and social care staff and service users tell us what research is important to them and why they feel others need to know about it. Join the conversation on Twitter and tell us which Signals have interested, excited or surprised you, using #MySignals. You can find the latest NIHR Signals here.
In this collection, Jane Blazeby and Natalie Blencowe, academic surgeons at the University of Bristol, share their selection of Signals which can help to better inform surgical decisions.
Evidence use in surgery
Clinical decisions about whether to undergo surgery and the types of operations to perform may be based on high quality evidence, but when it’s not available, surgeons have to rely on other information, such as that from case series (which are usually very biased), or even their own personal experience.
This means decisions for surgery vary between surgeons and centres, operations are frequently conducted in different ways (and it is unknown which is best) and there is widespread variation in management around the time of operation. To address these issues surgical teams are working with clinical trials units, methodologists and patient partners to design and conduct large scale surgical trials. If designed and conducted appropriately, they provide reliable information on short and long-term risks and benefits and NHS costs. Information is critical to patient choice and individual decision-making, as well as policy decisions in the NHS.
Another issue is “technology creep”, where new practices may be introduced without evaluation, with popular opinion assuming that new must be better. This can lead to widespread uptake of a new technology based on unreliable evidence (such as case series or short term outcomes). An important part of surgical research is the comparison of new technologies, such as minimal access techniques, with standard treatments.
These signals have been selected to illustrate how rigorous evaluation of a new technology within a randomised controlled trial (RCT) provides short and long term clinical and patient centred outcome data. In the examples, different types of outcomes (often adverse events and quality of life data) may be contradictory. Although this can make it difficult to reach a clear cut conclusion about an intervention, the trials provide useful comparative treatment profiles over time to inform decision-making.
For example, some patients may choose to risk short term adverse events for the increased chance of a long term benefit, while others may chose the opposite. Provision of robust comparative information helps surgeons and patients to weigh up the different profiles of risks and benefits of the treatments available.
It is important to note, however, that the trial data can only really allow patients to make an informed choice if they are of sufficient quality (i.e. they are not biased), and if surgeons provide them with the information in an understandable way.
Here are some example of trials that can be used in practice, and of reviews that are relevant to surgical care.
In this NIHR signal, four well designed multi-centre Randomised Controlled Trials (RCTs) (all at low risk of bias) were successfully completed. Patients with severe aortic stenosis were randomised to either open aortic valve replacement or to minimally invasive insertion of an aortic valve (TAVI) device. This was guided into position in the heart through a small incision in the groin or chest.
Two years after the procedure, more people who had the TAVI device were still alive with fewer initial complications. However there were higher risks of vascular complications, permanent pacemaker implantation and leakage around the valve. These different treatment profiles now provide more information to inform patients and health providers. Forthcoming data on costs and outcomes from an on-going UK trial will further inform this debate.
Piles are common and cause unpleasant problems with bleeding and itching around the anus. Historically, first line treatment involved the placement of rubber bands around the piles (either in the outpatient clinic or during sigmoidoscopy).
A new operation, called haemorrhoidal artery ligation (which is undertaken in the operating theatre and requires a general anaesthetic) was compared with simple banding in an RCT performed in 12 hospitals (372 patients). It found that although the new technique was associated with lower recurrence rates of piles at 12 months than standard bands, it had some serious complications and, because of the need for anaesthesia (and time in the operating theatre), was much more expensive.
This NIHR trial provides accurate data to allow patients, surgeons and healthcare providers to make an informed choice. It highlights serious issues with the new procedure that need careful consideration before widespread uptake occurs.
Where part of the gut has been operated upon, surgeons are often concerned about leakage from the site of the surgery. Traditionally patients are kept from eating and drinking for several days following surgery. Historically, this was thought to protect the repaired wall of the gut and reduce the chances of leakage. However, it means that patients must endure a prolonged period of fasting.
This systematic review and meta-analysis combined data from 15 studies, eight of which were RCTs, to challenge the dogma. The RCTs were assessed to be at low or unclear risk of bias. It found no significant difference in the risk of anastomotic leak, re-operation, mortality, or readmission amongst patients exposed to early or late feeding. Length of stay and post-operative pneumonia were both significantly lower amongst patients undergoing early feeding.
Although this review was published in 2016, many surgeons still routinely prohibit oral intake for several days after stomach surgery. Indeed neither of us had previously seen this review, despite being academics and upper GI surgeons. There is a need to raise surgeons’ awareness of research findings and their relevance to clinical practice. There is also a need to train them to use the data with confidence or to participate in new research to further address the questions.
Abdominal wall hernias are common and, if symptomatic, patients are usually offered surgical repair. The best method for performing the operation is uncertain; options include using stitches or mesh to repair the tissue. This systematic review summarised results from 10 RCTs (and 1215 patients) comparing these two techniques. It found that hernia recurrence within six months to five years after surgery was 64% less likely after mesh repair than after stitching.
However, the quality of the RCTs was variable, with all of them at high or unclear risk of bias. Only a few of the studies measured post-operative pain, and those that did found pain levels were higher amongst patients receiving mesh. Data regarding cost-effectiveness were not collected, which may be important given that mesh is more costly than stitching.
Although we can be reasonably confident that recurrence rates can be reduced by using mesh, the lack of data about post-operative pain and other patient-centred outcomes is concerning -especially given that recent research about surgical repair of inguinal hernias found the rate of chronic pain was much higher than previously thought.
Further high quality RCTs are therefore still required. Despite the poor quality of available evidence highlighted by this review, many surgeons routinely use mesh to repair abdominal wall hernias. This may be because surgeons are unaware of this research, are unable to interpret its findings, or have disregarded them.
About the authors: Natalie Blencowe and Jane Blazeby
Natalie Blencowe (@NatalieBlencowe) and Jane Blazeby (@BlazebyJ) are academic surgeons at the University of Bristol. Jane is an honorary consultant general surgeon with an interest in upper gastro intestinal surgery and a Professor of Surgery. Natalie is an NIHR clinical lecturer and an honorary specialty registrar in general surgery.
Jane directs the MRC ConDuCT-II (Collaboration and innovation in difficult RCTs in invasive procedures) Hub for Trials Methodology Research, co-directs the Royal College of Surgeons Trials Centre in Bristol and is lead of the Surgical Innovation theme of the Bristol Biomedical Research Centre. She is a chief investigator of a large trial in bariatric surgery (NIHR HTA By-Band-Sleeve). She supports surgeons to design, conduct and lead clinical trials.
Jane and Natalie are very involved in collaborations to design and deliver trials, and work closely with trials methodologists to develop new and better ways to conduct RCTs. They believe that whilst trials ultimately create evidence to influence decisions, the process of trial participation itself is important for surgeons to understand, and, subsequently implement the findings of that trial (so NHS care becomes evidence based).
You can get NIHR Signals delivered to you by email by joining our mailing list.