ABC | Volume 111, Nº6, December 2018

Original Article Andrade et al Prognostic Value of Myocardial Scintigraphy Arq Bras Cardiol. 2018; 111(6):784-793 The prevalence of 30% of ischemia among patients was higher than that found in previous studies. Zellweger et al. 14 detected ischemia in 19% of patients after 60 months of PCI, and Rajagopal et al. 11 in 23% of those evaluated after 3.9 months. The exception was the study by Galassi et al., 12 which included only patients known to undergo incomplete revascularization and, as expected, detected more abnormal perfusions. Similar to previous studies, 9,11 incomplete revascularization as an indication of MPS and the presence of previous AMI were considered independent predictors of ischemia. In contrast, the presence of DM was not independently associated with ischemia, as described by other authors. One possible explanation, given that all patients are asymptomatic, is the valorization of the presence of comorbidity leading to a higher indication of exams. Seventy percent of the diabetics in this study had indication of control MPS. Previous studies that analyzed the role of MPS in the follow-up after percutaneous revascularization used the composite endpoint model, which impaired the comparison of the results. It should be noted that the evaluation of events separately, as performed in this study, is important because the endpoints analyzed (death, cardiovascular death, non-fatal AMI and revascularization) have different clinical relevance and occurred at different frequencies in all the studies described. 9-14 The mortality rate observed was 2% per year, comparable to the rate described by Leon et al., 19 in the 5-year follow-up of patients treated with conventional and drug-eluting stents. However, comparing the different perfusion groups, patients with abnormal MPS without ischemia had a mortality rate of 3.3% per year, higher than that found in patients with abnormal perfusion with ischemia and normal perfusion, respectively, 2% and 1.2%. In addition, the extent of the total perfusion defect was independently associated with death when greater than 6%. In the evaluation of other variables, age greater than 70 years was considered an independent predictor of mortality, which is expected in the natural evolution of coronary disease. Likewise, the presence of DMwas associated with a higher risk of death, similar to data in the literature that showed a more diffuse atherosclerotic involvement among diabetics and a higher propensity to develop restenosis after percutaneous intervention, thus leading to greater mortality in the long term. 20 Acampa et al. 21 had emphasized that patients undergoing pharmacological stress had a higher age group and a higher prevalence of clinical predictors of ischemia compared to those who underwent physical stress, and, therefore, had a poorer prognosis. Similarly, in the current study, the pharmacological stress protocol was used in 70% of the patients who died and was significantly associated with the endpoint risk. Aspects related to MPS indications also directly influenced the results, with a preoperative examination being associated with a greater chance of death. One possible justification for such finding is the risk inherent to the surgical procedure itself, and the potential severity of the underlying pathology. This variable was not addressed by the other studies already cited. 9-14 Although they were not included in the multivariate analysis because of a strong correlation with perfusion scores, the presence of prior AMI and lower EF values were more frequently found among those who died, respectively, 69% × 51%, p = 0.009 and 47 ± 16 × 54 ± 12, p = 0.001. Other studies had already demonstrated the impact of ventricular function on survival of patients with coronary artery disease, among which the Coronary Artery Surgery Study (CASS) is highlighted, which observed an inverse relationship between EF and mortality. In this register, survival rates after 12 years of follow-up of coronary arteries disease with EF ≥ 50%, between 35 and 49% and < 35% were, respectively, 73%, 54% and 21% (p = 0,001). 22 Similar to what was found in the mortality analysis, the outcomes of cardiovascular mortality and non-fatal AMI had a higher incidence in the group with abnormal perfusion without ischemia compared to the others. The absence of statistical significance may be justified by the small number of events, but certainly does not compromise the importance of the findings, especially cardiovascular mortality with p= 0.064, close to what is considered relevant. The only factor independently associated with cardiovascular mortality was the total perfusion defect greater than 6%, and to non-fatal AMI was the presence of DM. Georgoulias et al., 10 after an 8-year follow-up of 246 asymptomatic patients undergoing CPM after PCI, also observed that the occurrence of the composite endpoint, cardiovascular death, and non-fatal AMI was greater the greater the extent of the total perfusion defect. The annual rate of endpoint revascularization was 4.6%, more significant during the 1 st year of follow-up compared to that found in subsequent years, 11.9% × 3.4%, respectively. Leon et al. 19 observed similar results, 20.4% of patients treated with conventional stents, and 11.2% of those treated with drug-eluting stents underwent a new approach in the 1 st year of follow-up; then, the annual rate of revascularization was a constant of 3.5% between the 2 nd and 5 th years. In view of these findings, it should be pointed out that, as suggested by Leon et al., 19 the events taking place in the first year seem to be related to the initial procedure, with markedly reduced rates of conventional therapy to pharmacological therapy, whereas later revascularizations reflect the progression of disease, with constant rate, regardless of the type of stent used. Zellweger et al., 9 in the follow-up of patients undergoing percutaneous intervention, demonstrated that the cumulative rate of composite outcome was statistically higher among patients with ischemia than those without ischemia at MPS, and revascularization corresponded to 65% of these events. Similarly, Galassi et al., 12 in a cohort consisting of asymptomatic patients submitted to incomplete percutaneous revascularization, reported that 42% of the participants performed a new approach at themean follow‑up of 33months, and that the extent of ischemia in the MPS performed 4 to 6 months after the procedure was a predictor of this outcome. In the current study, in addition to the presence and extent of ischemia, incomplete revascularization as an indication of MPS and the interval between percutaneous intervention and MPS before 2 years were also significantly associated with revascularization. These results suggest that 790