STUDY OBJECTIVE: Numerous investigators have evaluated the ECG algorithm described by Sgarbossa et al to predict acute myocardial infarction in the presence of left bundle branch block and have arrived at divergent conclusions. To clarify the utility of the Sgarbossa ECG algorithm, we perform a systematic review and meta-analysis of these trials. METHODS: A structured search was applied to MEDLINE and Scopus databases, beginning with the year that the algorithm was derived (1996). Two reviewers independently screened citations, assessed for method quality, and extracted data (individual study characteristics, screening performance, and interobserver agreement) with a standardized extraction tool. We assessed qualifying studies for heterogeneity and generated summary estimates for the sensitivity, specificity, and positive and negative likelihood ratios with fixed-effect models. RESULTS: We identified 11 studies with 2,100 patients that met criteria for at least 1 component of the analysis. Ten studies with 1,614 patients reported a Sgarbossa ECG algorithm score of greater than or equal to 3. These yielded a summary sensitivity of 20% (95% confidence interval [CI] 18% to 23%), specificity of 98% (95% CI 97% to 99%), a positive likelihood ratio of 7.9 (95% CI 4.5 to 13.8), and a negative likelihood ratio of 0.8 (95% CI 0.8 to 0.9). The summary diagnostic odds ratio revealed homogeneity. Seven studies with 1,213 patients reported a Sgarbossa ECG algorithm score of greater than or equal to 2. These yielded sensitivities ranging from 20% to 79% and specificities ranging from 61% to 100%. Positive likelihood ratios ranged from 0.7 to 6.6 and negative likelihood ratios ranged from 0.2 to 1.1. The summary diagnostic odds ratio revealed heterogeneity. Intra- and interobserver agreement was substantial. Sensitivity analysis using the highest-quality studies yielded similar results. CONCLUSION: A Sgarbossa ECG algorithm score of greater than or equal to 3, representing greater than or equal to 1 mm of concordant ST elevation or greater than or equal to 1 mm ST depression in leads V1 to V3, is useful for diagnosing acute myocardial infarction in patients who present with left bundle branch block on ECG. The scoring system demonstrates good to excellent overall interobserver variability. A score of 2, representing 5 mm or more of discordant ST deviation, demonstrated ineffective positive likelihood ratios. A Sgarbossa ECG algorithm score of 0 is not useful in excluding acute myocardial infarction.
STUDY OBJECTIVE: Numerous investigators have evaluated the ECG algorithm described by Sgarbossa et al to predict acute myocardial infarction in the presence of left bundle branch block and have arrived at divergent conclusions. To clarify the utility of the Sgarbossa ECG algorithm, we perform a systematic review and meta-analysis of these trials. METHODS: A structured search was applied to MEDLINE and Scopus databases, beginning with the year that the algorithm was derived (1996). Two reviewers independently screened citations, assessed for method quality, and extracted data (individual study characteristics, screening performance, and interobserver agreement) with a standardized extraction tool. We assessed qualifying studies for heterogeneity and generated summary estimates for the sensitivity, specificity, and positive and negative likelihood ratios with fixed-effect models. RESULTS: We identified 11 studies with 2,100 patients that met criteria for at least 1 component of the analysis. Ten studies with 1,614 patients reported a Sgarbossa ECG algorithm score of greater than or equal to 3. These yielded a summary sensitivity of 20% (95% confidence interval [CI] 18% to 23%), specificity of 98% (95% CI 97% to 99%), a positive likelihood ratio of 7.9 (95% CI 4.5 to 13.8), and a negative likelihood ratio of 0.8 (95% CI 0.8 to 0.9). The summary diagnostic odds ratio revealed homogeneity. Seven studies with 1,213 patients reported a Sgarbossa ECG algorithm score of greater than or equal to 2. These yielded sensitivities ranging from 20% to 79% and specificities ranging from 61% to 100%. Positive likelihood ratios ranged from 0.7 to 6.6 and negative likelihood ratios ranged from 0.2 to 1.1. The summary diagnostic odds ratio revealed heterogeneity. Intra- and interobserver agreement was substantial. Sensitivity analysis using the highest-quality studies yielded similar results. CONCLUSION: A Sgarbossa ECG algorithm score of greater than or equal to 3, representing greater than or equal to 1 mm of concordant ST elevation or greater than or equal to 1 mm ST depression in leads V1 to V3, is useful for diagnosing acute myocardial infarction in patients who present with left bundle branch block on ECG. The scoring system demonstrates good to excellent overall interobserver variability. A score of 2, representing 5 mm or more of discordant ST deviation, demonstrated ineffective positive likelihood ratios. A Sgarbossa ECG algorithm score of 0 is not useful in excluding acute myocardial infarction.
Authors: Santanu Guha; Rishi Sethi; Saumitra Ray; Vinay K Bahl; S Shanmugasundaram; Prafula Kerkar; Sivasubramanian Ramakrishnan; Rakesh Yadav; Gaurav Chaudhary; Aditya Kapoor; Ajay Mahajan; Ajay Kumar Sinha; Ajit Mullasari; Akshyaya Pradhan; Amal Kumar Banerjee; B P Singh; J Balachander; Brian Pinto; C N Manjunath; Chandrashekhar Makhale; Debabrata Roy; Dhiman Kahali; Geevar Zachariah; G S Wander; H C Kalita; H K Chopra; A Jabir; JagMohan Tharakan; Justin Paul; K Venogopal; K B Baksi; Kajal Ganguly; Kewal C Goswami; M Somasundaram; M K Chhetri; M S Hiremath; M S Ravi; Mrinal Kanti Das; N N Khanna; P B Jayagopal; P K Asokan; P K Deb; P P Mohanan; Praveen Chandra; Col R Girish; O Rabindra Nath; Rakesh Gupta; C Raghu; Sameer Dani; Sandeep Bansal; Sanjay Tyagi; Satyanarayan Routray; Satyendra Tewari; Sarat Chandra; Shishu Shankar Mishra; Sibananda Datta; S S Chaterjee; Soumitra Kumar; Soura Mookerjee; Suma M Victor; Sundeep Mishra; Thomas Alexander; Umesh Chandra Samal; Vijay Trehan Journal: Indian Heart J Date: 2017-03-23
Authors: Andrea Di Marco; Marcos Rodriguez; Juan Cinca; Antoni Bayes-Genis; Jose T Ortiz-Perez; Albert Ariza-Solé; Jose Carlos Sanchez-Salado; Alessandro Sionis; Jany Rodriguez; Beatriz Toledano; Pau Codina; Eduard Solé-González; Monica Masotti; Joan Antoni Gómez-Hospital; Ángel Cequier; Ignasi Anguera Journal: J Am Heart Assoc Date: 2020-07-04 Impact factor: 5.501
Authors: Beatrice von Jeinsen; Stergios Tzikas; Gerhard Pioro; Lars Palapies; Tanja Zeller; Christoph Bickel; Karl J Lackner; Stephan Baldus; Stefan Blankenberg; Thomas Muenzel; Andreas M Zeiher; Till Keller Journal: PLoS One Date: 2016-05-05 Impact factor: 3.240