Abstract 12496: Prognostic Value of Glasgow Prognostic Score and Its Modified Scores on 5-Year Outcome in Patients With Percutaneous Coronary Intervention

Background: Glasgow Prognostic Score (GPS) is an inflammatory score commonly used in clinical practice. In recent years, its modified scores have emerged, including modified GPS (mGPS) and hs-CRP-modified GPS (hs-mGPS). However, the long-term predictive values of these three scores (GPS, mGPS, and hs-mGPS) in coronary heart disease (CHD) remains unclear. Aim: To assess the predictive values of GPS, mGPS, and hs-mGPS on long-term survival in patients following percutaneous coronary intervention (PCI). Methods: Consecutive 10,724 patients with PCI were enrolled in the year of 2013. The primary endpoint was 5-year all-cause death. Results: Finally, 8909 patients with complete clinical data and no loss of follow-up were included. According to the original scores, the risk of all-cause death in patients with a high score of GPS, mGPS, and hs-mGPS was significantly higher than that in patients with a low score. All GPS, mGPS, and hs-mGPS had predictive values for all-cause death with relatively lower AUC of 0.534, 0.522, and 0.545, respectively. In further analysis, we improved hs-mGPS by using hs-CRP 2 mg/L and albumin 40 g/L, and established the CHD-hs-mGPS. The AUC of CHD-hs-mGPS was 0.554 which was significantly higher than mGPS (P = 0.017) and was numerically higher than GPS and hs-mGPS in the total population. Of these four scores (GPS, mGPS, hs-mGPS, and CHD-hs-mGPS), only the CHD-hs-mGPS had a predictive value for both ACS and non-ACS subgroups. Conclusion: For CHD patients, GPS, mGPS, and hs-mGPS had significant long-term predictive value for all-cause death. The score we modified, CHD-hs-mGPS, simultaneously had a predictive value in the total population, ACS and non-ACS subgroups. These inflammatory scores can help early identification of high-risk patients in the patients with established CHD following PCI.