Background: Despite existing clinical, laboratory and electrocardiographic characteristics suggestive of acute pericarditis, there is no multivariate model as a diagnostic tool. Objective: To develop a clinical score for diagnosis of pericarditis as the cause of acute chest pain, using admission data. Methods: In a diagnostic case control study, we compared 45 patients of the Chest Pain Registry diagnosed of pericarditis (confirmed by magnetic resonance imaging or the presence of pleural effusion in echocardiography) versus 90 patients with an alternative confirmed diagnosis, randomly selected from our registry. Six clinical characteristics, 16 chest pain characteristics and 4 complementary exams were tested. Logistic regression was used to derivate a probabilistic model composed by independent predictors of pericarditis. Results: Among 17 variables associated with pericarditis, 5 remained independent predictors: age, pain aggravation with thorax movement; positive troponin; diffuse ST segment elevation and C reactive protein. Each independent predictor was attributed a score proportional to its regression coefficient. The final score presented discriminatory capacity represented by Cstatistic of 0,97 (95% CI = 0,93 to 1,0). The best cutoff point was defined as > 6 points, with sensitivity of 96% (95% CI = 85 to 100), specificity of 87% (95% CI = 78 to 93), positive likelihood ratio of 7,2 (95% CI = 4,2 to 12) and negative likelihood ratio of 0,05 (95% CI = 0,01 to 0,2). Conclusion: The proposed multivariate score is accurate for diagnosis of pericarditis. It needs to be further validated in an independent sample.
IntroductionCoronary anatomy is one of the strongest risk predictors in Acute Coronary Syndromes (ACS), which justifies early coronary angiography. Diagnostic scores for predicting outcomes are usually superior to clinical judgment. Despite being validated for prognosis, the GRACE score has been used to discriminate patients with high or low probability of anatomical severity.ObjectiveTo test the hypothesis that the GRACE score actually predicts anatomical severity.MethodsThe study was carried out by assessing consecutive patients with ACS who underwent invasive angiography. Severe anatomical disease was defined as obstructive involvement (≥ 70% in diameter) in (1) left main coronary artery or (2) double or triple vessel disease involving proximal left anterior descending artery or (3) subocclusion. The GRACE score was evaluated under numerical and dichotomous tests.ResultsA total of 733 patients were evaluated, aged 63 ± 14 years, 61% male and GRACE score of 119 ± 37. Obstructive coronary disease was observed in 81% of the patients, classified as one, two or three vessel disease, or left main coronary artery involvement in 28%, 23%, 26% and 4%, respectively. The area under the ROC curve of the GRACE score was 0.65 (95% CI = 0.61 - 0.69) for predicting severe disease. The cutoff point below which the first GRACE tertile is defined (109) was used to dichotomize low-risk (N = 318) and medium-high-risk (N = 415) samples. This standard definition of intermediate-high risk by the GRACE score (> 109) revealed sensitivity of 67% in detecting severe anatomy (95% CI = 61% - 72%) and specificity of 50% (95% CI = 46% - 55%), resulting in positive likelihood ratio of 1.3 (95% CI = 1.2 - 1.5) and negative likelihood ratio of 0.66 (95% CI = 0.55 - 0.80). There was a weak correlation between GRACE and anatomical scores such as SYNTAX (r = 0.36, P < 0.001) and Gensini (r = 0.36, P < 0.001).ConclusionDespite statistical association with extent of anatomical coronary disease, the GRACE Score is not accurate to predict severity of disease before coronary angiography.
Background Bleeding and hospital death have an independent association in observational records of acute coronary syndromes (ACS), leading to interpretation of this relationship as causal. However, association does not guarantee causality, and better exploitation of this phenomenon is necessary. Purpose To describe the association between bleeding and death of patients with ACS, exploring causality through the cascade of events that separate these two phenomena. Methods Patients consecutively admitted to Coronary Unit by objective criteria of ACS were included prospectively. Major bleeding during hospitalization was defined according to types 3 and 5 of the Universal Bleeding Classification. Logistic regression and sequence analysis of events were used to evaluate the association between bleeding and death. Results A total of 1104 patients were studied, age 65±14 years, 58% male, 23% with ST elevation. The incidence of major bleeding was 4.7% (52 cases). Bleeding patients presented mortality of 31% (16 deaths), compared to 4.7% death in the non-bleeding group (RR=6.6, 95% CI: 4.0–11). Those who bled had a GRACE score significantly higher than those free of bleeding (157±39 versus 121±38, P<0.001). After adjustment for this score, bleeding remained strongly associated with death (OR=4.5, 95% CI: 2.1–9.7, P<0.001). However, among the 16 deaths that occurred after bleeding, in 56% of the patients the death was due to hemorrhage, while the remainder was a consequence of the myocardial injury of the infarct or natural evolution of noncardiac morbidities. Conclusion The independent association between major bleeding and death in acute coronary syndromes is only partially mediated by causality. In the same proportion, deaths coexist with bleeding without a causal relationship. Funding Acknowledgement Type of funding source: None
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.