Abstract Number: Esoc2026a2327 the Crescendo Paradox: High-Risk Phenotype With Low-Risk Scores — A Retrospective Study of Strictly Defined Crescendo Tia

Abstract Background Crescendo Transient Ischemic Attack (TIA) is a high-risk clinical pattern. Although traditionally associated with high stroke risk, it remains unclear whether Crescendo TIA (cTIA) represents a distinct vascular risk profile compared with Single TIA (sTIA), or whether standard risk stratification tools such as the ABCD2 score accurately reflect its clinical urgency. Methods This retrospective single-center study included all patients diagnosed with TIA at Shafa Hospital (2018–2023). Strictly defined cTIA (≥2/24h, ≥3/72h, or 4/2 weeks; n=100) was compared with sTIA (n=479). Recurrent non-crescendo cases (n=29) were excluded due to higher chronic stroke burden (44.8%, P=0.004), indicating a separate phenotype. Demographics, risk factors, symptom duration, and ABCD2 scores were analyzed. Results Risk profiles were similar. No significant differences were found in age 60 (45% vs 44%, P=0.841), male sex (50% vs 51%, P=0.901), hypertension (51% vs 51%, P=0.991), diabetes (30% vs 27%, P=0.597), hyperlipidemia (26% vs 19%, P=0.158), ischemic heart disease (24% vs 24%, P=0.999), smoking (12% vs 12%, P=0.899), prior stroke (22% vs 16%, P=0.200), or opium use (24% vs 18%, P=0.388). However, cTIA presented with shorter symptoms: 46% lasted 10 minutes vs. 15% in sTIA (P0.001). Consequently, fewer cTIA patients were classified as "High Risk" by ABCD2 ≥4 (49.0% vs 63.5%, P=0.006). Conclusion Crescendo TIA is a distinct phenotype of brief, repetitive ischemia, not explained by traditional risk factors. The ABCD2 score paradoxically underestimates its risk due to short symptom duration. Recognizing the crescendo pattern itself should drive urgent triage over reliance on risk scores. Conflict of interest Ali Mahyaei: nothing to disclose, Fatemeh Keshavarz: nothing to disclose, Aliakbar Zamanigerashi: nothing to disclose, Prof.Farhad Iranmanesh: nothing to disclose.