One size does not fit all: performance of the 0/1-h and 0/2-h cardiac troponin t algorithms in a contemporary cohort of chest pain patients. (3rd May 2023)

• Record Type: Journal Article
• Authors: Obling, L; Alquezar, A; Garcia-Osuna, A; Vila-Perales, M; Rodriguez-Sotelo, L; Bragagnini, W; Pena-Ortega, P; Romero, C J; Padrosa, M; Alvarez-Nuno, S; Moreno, C; Ibanez, Y; Sionis, A; Ordonez-Llanos, J; Duran-Cambra, A
• Is Part Of: European heart journal. Volume 12(2023)Supplement 1
• Publication Date: 2023-05-03
• Subjects: 616.1205
• DOI: 10.1093/ehjacc/zuad036.094
• Languages: English
• ISSNs: 2048-8726

Abstract

Abstract: Funding Acknowledgements: Type of funding sources: None. Background: A rapid and efficient disposition of patients with chest pain and a suspected acute coronary syndrome (ACS) diagnosis represents a clinical challenge in the emergency department (ED). The latest European clinical guidelines recommend the use of diagnostic algorithms at 0/1 and 0/2 hours, but their performance in different patient populations needs to be further investigated. Purpose: To evaluate the performance of the 0/1-h and 0/2-h high-sensitivity cardiac troponin T (hs-cTnT) algorithms in a cohort of patients admitted to an ED in an academic center in Spain. Methods: We performed a single-center, prospective study in patients admitted with suspected non-ST elevation ACS (NSTEACS). Hs-cTnT (Roche Diagnostics) was measured at admission (t0) and one (t1) and two hours (t2) later; hs-cTnT differences between times were named Delta 0/1h (D-0/1) or Delta 0/2h (D-0/2). Patients were sub-grouped according to the following: "Rule out" when the time of pain-debut was >3 hours and t0 hs-cTnT <5 ng/L; for remaining patients, t0 hs-cTnT <12 ng/L and D-0/1 <3 ng/L or t0 <14 ng/L and D-0/2 <4 ng/L; "Rule-in" when hs-cTnT at t0 ≥52 ng/L or in remaining patients when D-0/1 ≥5 ng/L or D-0/2 ≥10 ng/L); "Observe" were those not fulfilling described rules, in whom further diagnostic testing was needed. Final diagnosis was adjudicated by two independent cardiologists according to the Fourth Universal Definition of Myocardial Infarction. Performance of both algorithms was made by quantification of the negative predictive value (NPV) and the positive predictive value (PPV) in the "Rule-out" and "Rule-in" group, respectively. Results: 286 patients were enrolled in the study. Median age was 70 (higher than in previous validation studies of rapid rule-in and rule-out algorithms; see Table 1 for baseline characteristics). The overall prevalence of NSTEACS was 17% (n = 49/286). A hs-cTnT<5 ng/L at t0 correctly ruled out NSTEACS in all patients (n = 48) fulfilling the rule; by contrast, a t0 value >52 ng/L existed in 39 patients, but only 69% were NSTEACS. The 0/1-h algorithm demonstrated an NPV of 97.3% and a PPV of 72% while the 0/2-h algorithm achieved an NPV of 100% and a PPV of 72.9% for the "Rule-out" and "Rule-in" group, respectively (Figure 1). Both algorithms allocated same percentage of patients (~30%) in the "Observe" group. Overall, the algorithms performed well, but in the 0/1-h algorithm, four patients with final diagnosis of NSTEACS were wrongly allocated to the "Rule-out" group. This did not occur in the 0/2-h algorithm where the same patients were allocated to the "Observe" group. Conclusion: In our prospective cohort, the performance of the two recommended rapid rule-in and rule-out algorithms were good, but the 0/2-h algorithm achieved a better NPV and allowed for a more precise disposition of patients with respect to the 0/1-h algorithm. Table 1: Baseline characteristics Figure 1: ESC hs-cTnT algorithms