Risk Calculators Validation for Elective Major General Surgery

Death, Surgery

preoperative parameters, risk prediction, postoperative mortality, elective major general surgery

Nowadays, over 300 million surgical operations take place every year worldwide, which increase at a rate of 33.6% comparing data from 2005 to 2013. According to Surgical Outcomes Monitoring and Improvement Program (SOMIP) reports, which is an Hospital Authority-wide (HA-wide) audit on postoperative outcomes, a growth in major and ultra-major operations performed in our locality is also observed between 2008 and 2016, which leads to an increasing demand of high dependency and intensive care in the postoperative period. With the advancement in surgical technology, increasing surgical complexity and aging population have raised concerns towards perioperative costs and postoperative complications. Therefore, there is a need of an objective tool for risk stratification, which would be useful to guide clinical decision in terms of the magnitude of operation, level of intraoperative monitoring and postoperative placement plan. Various risk scoring systems have been developed nowadays and each has its own limitations. As nowadays, the calculated risk score is commonly used in shared decision making process with patient and among the perioperative team. Risk calculation solely based on preoperative parameters will be more practical for daily clinical use. Therefore, in this study, the investigators would like to validate the postoperative mortality prediction with the risk calculators that are established merely using preoperative variables. Hopefully this would guide the future risk stratification in patients undergoing elective major surgical operation.

Nowadays, over 300 million surgical operations take place every year worldwide, which increase at a rate of 33.6% comparing data from 2005 to 2013. According to Surgical Outcomes Monitoring and Improvement Program (SOMIP) reports, which is an Hospital Authority-wide (HA-wide) audit on postoperative outcomes, a growth in major and ultra-major operations performed in our locality is also observed between 2008 and 2016, which leads to an increasing demand of high dependency and intensive care in the postoperative period. With the advancement in surgical technology, increasing surgical complexity and aging population have raised concerns towards perioperative costs and postoperative complications. An international prospective cohort study revealed that globally 1 in 6 patients experienced a complication before hospital discharge and 1 in 35 patients who experienced a complication subsequently died without leaving the hospital. Therefore, there is a need of an objective tool for risk stratification, which would be useful to guide clinical decision in terms of the magnitude of operation, level of intraoperative monitoring and postoperative placement plan. There are a variety of risk stratification tools available for use in major non-cardiac surgery. Among all, the American Society of Anaesthesiology Physical Status (ASA-PS) evaluation scale is the most commonly used risk evaluation system in the assessment of patients' physical status in the preoperative period. Although ASA-PS is well-validated in previous studies and simple to use, inter-rater reliability and the lack of consideration in the surgical perspective have raised concerns towards the development of risk prediction models to supplement clinical judgements and strengthen operative mortality estimation. In 2013, a qualitative systematic review found that Portsmouth Variation of the Physiological and Operative Score for the enUmeration of Mortality and Morbidity (P-POSSUM) and Surgical Risk Scale (SRS) to be the most reliable multivariate risk scoring systems,, but both were noted to have limitations. P-POSSUM has overcome the issues of risk overestimation and inadequate generalization across various surgical specialties by POSSUM. But the calculation requires 12 physiological and 6 operative variables, some of which requires subjective interpretation e.g. chest X-ray. These makes P-POSSUM labour-intensive for clinical use. Whereas SRS requires fewer data for risk calculation, it has only been validated in a single centre study. In recent years, newer risk prediction models like the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) model and Preoperative Score to Predict Postoperative Mortality (POSPOM) have been developed to provide a more comprehensive perioperative risk prediction for patients undergoing major operation. ACS-NSQIP model is developed based on high-quality clinical data from ACS-NSQIP and is described as a universal risk calculator, which includes a Surgeon Adjustment Score (SAS) that allows further score modification according to surgical performance. However, owing to the high dependence on preoperative laboratory results, ACS-NSQIP often encounters problems where these parameters are not readily available in emergency situations. POSSOM model involves 17 predictor variables. Together with its excellent discrimination and calibration properties demonstrated in its validation cohort and the easily referable rating system, POSSOM is considered a robust tool for 1-year postoperative mortality prediction. However, further reviews on its external validation are yet available. In 2014, a new risk stratification tool, Surgical Outcome Risk Tool (SORT) was developed in the UK to predict 30-day mortality after non-cardiac surgery in adults, based on post hoc analysis of data in the Knowing the Risk study from the observational National Confidential Enquiry into Patient Outcome and Death (NCEOPD). SORT is a multivariate risk scoring system, which includes 6 variables: 1) American Society of Anesthesiologists Physical Status (ASA-PS) grade, 2) urgency of surgery, 3) surgical specialty, 4) surgical magnitude, 5) cancer or non-cancer surgery and 6) age. In 2018, the Combined Assessment of Risk Encountered in Surgery (CARES) surgical risk calculator has been developed based on Singapore local data, which makes use of 9 preoperative parameters namely: 1) age, 2) gender, 3) ASA classification, 4) surgical risk group, 5) emergency surgery, 6) anaemia status, 7) red cell distribution width (RDW), 8) ischaemic heart disease, , 9) congestive heart failure for prediction of postsurgical mortality and need for intensive care unit admission. When the investigators look into each of these existing risk stratification tools, each of the risk calculators possesses its drawbacks when coming into clinical applications. As nowadays, the calculated risk score is commonly used in shared decision making process with patient and among the perioperative team. Risk calculation solely based on preoperative parameters will be more practical for daily clinical use. Therefore, in this study, the investigators would like to validate the postoperative mortality prediction with the risk calculators that are established merely using preoperative variables. Hopefully this would guide the future risk stratification in patients undergoing elective major surgical operation.

No related papers found yet.

Hong Kong