Predicting risk in emergency surgery
An interview with Mr Malcolm West (MD PhD FRCS) - NIHR Clinical Lecturer in Surgery – University of Southampton and Royal College of Surgeons Senior Clinical Fellow in Advanced & Robotic Colorectal Cancer Surgery – St Mark’s Hospital, London.
Mr West is the senior author of the recently accepted Annals of Surgery publication entitled: Sarcopenia and myosteatosis predict adverse outcomes after emergency laparotomy: a multi-centre observational cohort study.
Emergency surgery is risky. Emergency laparotomy has one of the highest morbidity and mortality rates of all surgical interventions. Nine per cent of patients who have an emergency laparotomy will die within 30 days of their surgery, rising to 19 per cent at the end of one year.
Since 2014, the NELA (National Emergency Laparotomy Audit) score (a perioperative risk predictor) has been used prior to emergency laparotomy surgery to guide surgeons on the risks to individual patients. We know that this and other measures introduced by NELA have helped improve emergency laparotomy outcomes for patients and perioperative decision making, but the mortality rate post-surgery still remains high.
The Wessex Research Collaborative wanted to explore whether there were other predictors that could aid decision making and predict adverse outcomes for emergency laparotomy patients in order to improve perioperative risk stratification, ultimately improving surgical outcomes and survival.
We’ve known about the possible link between two conditions, Sarcopenia (reduced muscle mass) and Myosteatosis (reduced muscle quality), and an increased rate of morbidity post-cancer surgery, so we set out to study the impact that these two conditions had on emergency laparotomy patients.
We setup a surgical trainee led and delivered, observational study across 10 hospitals in Wessex (Hampshire, Dorset and Isle of Wight), and analysed data from 610 emergency laparotomy patients over a 3-month period. We wanted a snapshot of data that we could use to help us determine the outcomes for patients and aid patient, surgical and general perioperative decision making.
The study interrogated pre-operative CT scans of patients admitted to hospital as an emergency, undergoing an emergency laparotomy. Sarcopenia, the decline in muscle mass caused by ageing or a stressor (cancer, for example), and Myosteatosis, the decline in the function of muscle quality, can both be accurately assessed on CT.
The scans were uploaded to a central database at University Hospital Southampton where we were able to analyse them and collate the data.
We setup a surgical trainee led and delivered, observational study across 10 hospitals in Wessex
With expertise from the University of Maastricht, we were able to quantify skeletal muscle index (SMI) and skeletal muscle radiation attenuation (SM-RA) from the scans and used regression modelling to predict the relationship between low SMI (Sarcopenia) and low SM-RA (Myosteatosis) and increased morbidity and mortality rates.
Results showed that there was an increased risk of mortality 30 days after an emergency operation and 1 year after surgery in patients who had either of these conditions. These rates are higher than non-emergency patients, as you might expect, as emergency operations are likely to be performed on people who have been significantly ill for a number of days.
It’s imperative that doctors have accurate objective indicators to enable them to make accurate judgements on patient outcomes and risks. The NELA score has enabled us to improve our decision-making processes and quantify risks before emergency surgery, however this relies on physiological data and some assumptions of what surgeons will find intra-operatively.
It is imperative that doctors caring for emergency patients have objective, individualised data that allows to tailor perioperative management to enable accurate risk prediction and most importantly improve shared decision making.
Our results show that there are ways in which we can refine the NELA score further, allowing us to make improved and more informed decisions on a case-by-case basis.
A key part of our research was deriving a sex-specific threshold value for the data, which hasn’t been done before. We thought that this would be important as we wanted to be able to refine our data as much as possible, so that sex-specific body composition variables further improve outcome predictions.
At the moment this is a research tool. It shows the link between the conditions and the elevated risks, but it can’t yet inform decisions or best-practice. This is the ultimate goal. If we are able to validate this data and develop an automated process for determining sarcopenia and myosteatosis using routine, clinically available CT scans, we might be able to integrate risk prediction models in daily clinical care.
Furthermore, guiding surgical strategy and intra-operative decision making, will further the ethos of individualised care for patients. This might also offer strategies to incorporate perioperative quality improvement care bundles, e.g. targeted nutritional prehabilitation and guide post-operative decision making to improve outcomes in emergency surgical patients. For some, surgery may prove just too risky.
For these patients, this individualised information might guide preoperative conversations with patients and their family instigating more conservative treatment options as opposed to major surgical intervention and also in certain cases palliation.
By conducting this research and creating tools that allow us to paint a more accurate picture of perioperative risk and long-term patient’s outlook, this data, I feel, is crucial to improve overall clinical care and shared decision making for patients undergoing major emergency surgical interventions.
