Article Text

Download PDFPDF

254 Construction of pre-operative prediction model and its use in gynaecological oncology using cardiopulmonary exercise testing and routine health data
  1. NL Whitham1;2,
  2. J Knight1;2 and
  3. N Wood2
  1. 1Lancaster University, Bailrigg, UK
  2. 2Royal Preston Hospital, Fulwood, UK


Introduction/Background*Our ever-growing population is putting greater strain upon the NHS with more complex medical problems, and strained resources. Comorbidities predispose patients to postoperative complications, impacting recovery and survival, length of stay, and mortality rates. The one size fits all attitude to treatment is no longer the best approach to tackle illness, as technology develops patient care will transform.

CPET assesses the bodies neurohumoral stress response to surgery, however not all patients complete CPET, with mobility and contraindications an issue. The ability to quantify morbidity and mortality risks enables discussions regarding appropriateness of surgical interventions, discuss likely scenarios and quality of life (QOL).

Methodology The aim is to create a pre-operative prediction model using cardiopulmonary exercise testing (CPET) and routine health data (RHD). The model can be utilised in conjunction with CPET, identifying patients in greater need of high dependency care (HDU), and at greater risk of complications.

All gynaecological oncology patients undergoing CPET from 2011 onwards are included in the retrospective analysis in one centre, which includes those over 65 years and those with multiple comorbidities. RHD, and CPET data will be collated, assessing links between the data with known clinical outcomes, producing a risk prediction tool that will then be used on a prospective cohort of patients.

Result(s)*Risk stratification tools allow shared decision making with personalised perioperative risks giving better patient experience and post-operative QOL.

RHD and CPET is currently being collated and analysed using R Studio.

Conclusion*The hope is to create a prediction model to use in conjunction with CPET to better guide care and improve patient outcome. If shown to better predict high risk patients it may be possible to improve care by prediction model alone, meaning all patients can be assessed, be cost effective, and be a more personalised approach to patient care.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.