Glen Martin, PhD student
Transcatheter aortic valve implantation and hospital re-admissions: An analysis of the British Cardiovascular Interventional Society National Dataset
- Aortic Stenosis (AS) is one of the most commonly acquired diseases of cardiac valves and occurs as a consequence of age related degeneration and calcification of the aortic valve. The mainstay of treatment of symptomatic AS is surgical aortic valve replacement; although multi-morbid elderly patients who were not considered surgical candidates were managed medically, with >90% mortality at 2 years reported.
- The development of Transcatheter Aortic Valve Replacement (TAVI) has provided a treatment option for these multi-morbid patients. Although TAVI has improved clinical outcomes over conservative treatment, long-term morbidity in a non-trial setting is poorly studied. An investigation of such outcomes is required to increase understanding of TAVI and to facilitate in improving the procedure where necessary.
- My PhD has a number of main aims. Firstly, I aim to investigate the causes and risk factors associated with clinical outcomes following TAVI. Secondly, to investigate the potential use of current and/or new risk models that can predict clinical risk for patients undergoing TAVI.
- Finally, I aim to develop statistical methodology to synthesise multiple existing CPMs in order to facilitate the prediction of patient risk at a local level, though utilising scientific information and research across populations/settings.
Start: September 2014
End: September 2017
Funded by: Medical Research Council through the HeRC PhD Studentship
- This project will be undertaken using the British Cardiovascular Interventional Society TAVI dataset (BCIS-TAVI) that records all TAVI procedures undertaken in the United Kingdom. A total of 95 variables are collected to the time of discharge from hospital. Mortality information is available in the dataset, provided by mortality tracking undertaken by the Medical Research Information Service.
- The dataset comprises a range of variables detailing patient demographics, risk factors for intervention, procedural details and adverse outcomes up to the time of hospital discharge. Logistic regression will be used to understand the relationships between pre-procedural characteristics and post-procedural outcomes.
- Predictive modelling techniques will be used to aim to predict the risk/probability of mortality given the pre-procedural patient demographics and risk factors. The performance of existing and newly derived models will be assessed using calibration and discrimination measures.
- Computer simulation methods will be used to investigate new statistical methodology around the aggregation of existing CPMs.
Benefits and Outcomes
Ideally, this work will inform the treatment of patients with AS through a greater understanding of the risks associated with the TAVI procedure. Risk factors for any potential adverse outcomes will be better understood outside the trial setting.
The derivation of a CPM to predict patient risk will help inform clinical decision-making. This has the potential to benefit cardiac teams in choosing treatment allocation and help patients understand the risks and potential complications of the procedure.
The performance of currently available CPMs that estimate mortality risk in patients with aortic stenosis was poor when applied to a cohort of TAVI patient’s independent from those in which they were developed. The development of new TAVI-specific CPMs is recommended, either by adapting existing CPMs or developing new risk scores in populations of interest.
Dr Matthew Sperrin
Dr Mamas Mamas