{"id":20747,"date":"2026-07-31T10:33:17","date_gmt":"2026-07-31T10:33:17","guid":{"rendered":"https:\/\/woncaeurope2026.org\/sessions\/ai-enhanced-prediction-and-real-world-risk-assessment-of-hyperglycemia-in-patients-receiving-pcsk9-inhibitors\/"},"modified":"2026-07-31T10:33:17","modified_gmt":"2026-07-31T10:33:17","slug":"ai-enhanced-prediction-and-real-world-risk-assessment-of-hyperglycemia-in-patients-receiving-pcsk9-inhibitors","status":"publish","type":"wsa_session","link":"https:\/\/woncaeurope2026.org\/fr\/sessions\/ai-enhanced-prediction-and-real-world-risk-assessment-of-hyperglycemia-in-patients-receiving-pcsk9-inhibitors\/","title":{"rendered":"AI-enhanced prediction and real-world risk assessment of hyperglycemia in patients receiving PCSK9 inhibitors"},"content":{"rendered":"<p>Proprotein Convertase Subtilisin\/Kexin type 9 inhibitors (PCSK9i) such as alirocumab and evolocumab are increasingly incorporated into primary care lipid-lowering strategies. Recent reports, however, suggest potential short-term disturbances in glucose homeostasis. Evidence characterizing this risk in routine primary care settings is limited.This study assessed the short-term risk of hyperglycemic disorders following PCSK9i exposure and developed a time-series Long Short-Term Memory (LSTM) model to predict hyperglycemia using routinely collected clinical data.A self-controlled case series (SCCS) design was applied to Observational Medical Outcomes Partnership \u2013 Common Data Model (OMOP-CDM) data from XXX Hospital. Incidence rate ratios (IRRs) were estimated for hyperglycemic disorders during PCSK9i exposure, with subgroup analyses for alirocumab and evolocumab and additional adjustment for concomitant hyperglycemia-inducing medications. Predictive performance of single-layer and two-layer LSTM models integrating longitudinal laboratory values and medication histories was compared with Random Forest and Logistic Regression models. SHAP analysis identified key predictors influencing LSTM outputs.PCSK9i exposure was associated with an increased short-term risk of hyperglycemic disorders (IRR 1.69, 95% CI 1.43\u20131.99). Evolocumab showed a significant association (IRR 1.75), while alirocumab showed a nonsignificant trend. The greatest risk occurred 121\u2013150 days after exposure (IRR 2.38). The association remained after adjusting for hyperglycemia-inducing medications. The LSTM models achieved excellent predictive performance (AUROC 0.995 for single-layer; 0.998 for two-layer), outperforming Random Forest (0.969) and Logistic Regression (0.720). SHAP analysis indicated higher BMI, lower lipoprotein(a) and hemoglobin, and elevated creatinine and leukocyte counts as important contributors.Findings suggest that PCSK9i use in primary care may carry a short-term hyperglycemia risk, with agent-specific and time-dependent patterns that warrant clinical attention. The strong discriminatory performance of LSTM models highlights the value of temporal features in real-world risk prediction. These insights may support earlier identification of at-risk patients and guide primary care\u2013oriented monitoring strategies.PCSK9i exposure is associated with a short-term increase in hyperglycemic disorders, and LSTM models provide robust predictive performance with clinically meaningful risk indicators.<\/p>\n","protected":false},"template":"","class_list":["post-20747","wsa_session","type-wsa_session","status-publish","hentry","description-off"],"_links":{"self":[{"href":"https:\/\/woncaeurope2026.org\/fr\/wp-json\/wp\/v2\/wsa_session\/20747","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/woncaeurope2026.org\/fr\/wp-json\/wp\/v2\/wsa_session"}],"about":[{"href":"https:\/\/woncaeurope2026.org\/fr\/wp-json\/wp\/v2\/types\/wsa_session"}],"wp:attachment":[{"href":"https:\/\/woncaeurope2026.org\/fr\/wp-json\/wp\/v2\/media?parent=20747"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}