ABC | Volume 114, Nº3, March 2020

Original Article Campos et al. Takotsubo syndrome recurrence Arq Bras Cardiol. 2020; 114(3):477-483 Figure 1 – Prisma Flowchart of the studies selection for the composition of the systematic review. 164 identified articles (PubMed: 154; LILACS: 6; SciELO: 4). 4 articles selected from the references of other articles. 160 articles after deleting duplicates. 88 articles after analysis of the titles. 13 articles after analysis of the abstracts. 6 selected studies after full reading. 72 removed – 1 because it was written in German; – 29 because they are identified as non-observational study types; – 42 because they did not meet the objective proposed. 75 removed – 9 because they were not written in English; – 48 because they were identified as non-observational study types; – 4 because they did not present recurrence rate of at least 3% and/or five or more patients with recurrence; – 14 for failing to meet the objective proposed. 7 eliminated for not meeting at least 80% of the STROBE criteria checklist. Table 1 – Score and percentage of articles quality based on STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) criteria Reference Study design Publishing journal Points on STROBE % Looi et al., 9 Prospective cohort. Journal of Heart, Lung and Circulation. 18 81.8 Templin et al., 3 Case-control. The New England Journal of Medicine. 19 86.3 Patel et al., 10 Retrospective cohort. Journal of Cardiac Failure. 19 86.3 Elesber et al., 11 Retrospective cohort. Journal of the American College of Cardiology. 19 86.3 Vriz et al., 12 Prospective cohort. Journal of Cardiovascular Medicine. 18 81.8 Nishida et al., 13 Case-control. Heart and Vessels. 20 90.9 Nishida et al. 13 presented a case-control study. Data from 251 patients who composed the BOREAS Registry (with 15 participating countries) from June 1999 to March 2012, were analyzed. Patients were divided into two groups, those with apical ballooning (type A), classic TTS presentation, and those with non-apical ballooning (non-A type), which included all other presentation forms of the syndrome. During a follow‑up of 2.6 ± 2.8 years, the recurrence rate was 2.8% (7/251), with no significant difference between A and non-A groups (2.8 and 2.9 %, respectively). In the univariate analysis, low Body Mass Index (BMI) (p = 0.048), midventricular (p = 0.01), and concomitant right ventricular involvement (p = 0.06) were associated with TTS recurrence. Only BMI (hazard ratio [HR] 0.75; 95% confidence interval [CI] 0.54–0.99; p = 0.048) and midventricular obstruction (HR 14.71; 95% CI 1.87–304.66; p = 0.01) remained significantly associated with TTS recurrence. Discussion TTS recurrence rate is variable in the literature and the factors associated with it were not clearly defined as well. This statement was clear when, recently, that 62-year-old female patient who had a history of being resuscitated for cardiac arrest a year before during an elective surgery and had a diagnosis of TTS, and who recovered to normal left ventricular (LV) function, now 1 year after the index episode, came to the office of one of the authors asking for a cardiovascular risk evaluation for another elective plastic surgery. After careful selection, data from the six studies were analyzed, in which factors with possible association with TTS recurrence were considered. The female gender was more prone to recurrence. Proximity to the first episode of the syndrome was a factor described by some authors as predisposing to a greater chance of recurrence. Low BMI and 479