Predicting Outcomes After Oesophatectomy
Research type
Research Study
Full title
Outcomes after oesophagectomy for oesophagel cancers: Development and validation of a risk-stratification model
IRAS ID
212275
Contact name
Jayachandran Radhakrishnan
Contact email
Sponsor organisation
Mid Essex Hospitals NHS Trust
Duration of Study in the UK
0 years, 11 months, 28 days
Research summary
Curative surgery for oesophageal cancers is an extremely invasive procedure with high complication rates and prolonged recovery times. 30% of patients who have surgical resection of oesophageal cancers die within 1 year with survivors requiring 18 to 24 months to recover a reasonable quality of life.
The factors associated with poor outcomes in oesophageal resections are not known. A number of studies have attempted to identify these risk factors, but have been limited by small patient numbers, arbitrary selection of risk factors, use of inappropriate outcomes and a suboptimal application of modelling techniques. As a consequence, there is no clinically useful risk stratification model for these patients.
In this study we propose to address these problems, by developing and validating a predictive model using a large patient population, without a priori assumption of risk factors and a robust modelling approach.
We propose to analyse patient and outcome data from about 300 patients with surgical resection for oesophageal cancer. We will collect relevant patient data including demographic data, biochemistry and markers of functional staus, tumour characteristics including histology, spread, type, duration and effectiveness of chemotherapy, and surgical characteristics including type of procedure and other relevant perioperative factors.
Survival at 1 year will be the primary outcome for the predictive modelling. Additional outcomes will include other complication rates including leak rates, duration of ITU and hospital stay, requirement for organ support and time to achieve recovery milestones.
A two pronged analytical approach will be used. Factors associated with the risk of poor outcomes will be identified and will be used to generate hypothesis for future interventional studies. In addition, a predictive model, for clinical use, will be developed and validated to identify patients at risk for poor outcomes. The model will be externally validated in a multi centre study in the future.
REC name
East of England - Essex Research Ethics Committee
REC reference
17/EE/0089
Date of REC Opinion
18 Apr 2017
REC opinion
Further Information Favourable Opinion