ABC | Volume 112, Nº3, March 2019

Original Article Kiyose et al A systematic review for valvular prostheses Arq Bras Cardiol. 2019; 112(3):292-301 Figure 4 – Forest plots showing the effects of biological versus mechanical prostheses on the risk of systemic arterial embolism (A) and the risk of endocarditis (B). 16.7% 6.4% 18.6% 41.7% 1.63 [0.84, 3.19] 0.44 [0.13, 1.54] 1.00 [0.54, 1.85] 1.05 [0.58, 1.91] 26.3% 11.3% 15.7% 53.2% 0.60 [0.38, 0.94] 0.87 [0.36, 2.12] 1.22 [0.60, 2.47] 0.80 [0.51, 1.25] 5.1% 5.1% 1.41 [0.34, 5.80] 1.41 [0.34, 5.80] 100.0% 0.93 [0.66, 1.31] 6.2% 32.1% 24.3% 62.6% 1.05 [0.18, 6.06] 0.63 [0.30, 1.32] 2.14 [0.91, 5.07] 1.11 [0.46, 2.67] 8.6% 2.2% 21.6% 32.4% 1.64 [0.37, 7.23] 0.32 [0.02, 6.14] 1.55 [0.62, 3.87] 1.42 [0.67, 3.01] 5.0% 5.0% 2.12 [0.30, 14.94] 2.12 [0.30, 14.94] 100.0% 1.21 [0.78, 1.88] 1.3.1 Aortic valve Edinburgh 2003 Stassano 2009 Veterans Affairs 2000 Subtotal (95% CI) Test for overall effect: Z = 0.16 (p = 0.87) Heterogeneity: Tau 2 = 0.12; Chi 2 = 3.46, df = 2 (p = 0.18); I 2 = 42% 1.3.2 Mitral valve Edinburgh 2003 Vallejo 1981 Veterans Affairs 2000 Subtotal (95% CI) Test for overall effect: Z = 0.99 (p = 0.32) Heterogeneity: Tau 2 = 0.05; Chi 2 = 2.94, df = 2 (p = 0.23); I 2 = 32% 1.3.3 Mitral + aortic valve Edinburgh 2003 Subtotal (95% CI) Test for overall effect: Z = 0.47 (p = 0.64) Heterogeneity: Not applicable Total (95 CI) Test for overall effect: Z = 0.41 (p = 0.68) Test for subgroup differences: Chi 2 = 0.92, df = 2 (p = 0.63); I 2 = 0% Heterogeneity: Tau 2 = 0.06; Chi 2 = 8.69, df = 6 (p = 0.19); I 2 = 31% 1.4.1 Aortic valve Edinburgh 2003 Stassano 2009 Veterans Affairs 2000 Subtotal (95% CI) Test for overall effect: Z = 0.24 (p = 0.81) Heterogeneity: Tau 2 = 0.32; Chi 2 = 4.45, df = 2 (p = 0.11); I 2 = 55% 1.4.2 Mitral valve Edinburgh 2003 Vallejo 1981 Veterans Affairs 2000 Subtotal (95% CI) Test for overall effect: Z = 0.91 (p = 0.36) Heterogeneity: Tau 2 = 0.00; Chi 2 = 1.06, df = 2 (p = 0.59); I 2 = 0% 1.4.3 Mitral + aortic valve Edinburgh 2003 Subtotal (95% CI) Test for overall effect: Z = 0.75 (p = 0.45) Heterogeneity: Not applicable Total (95 CI) Test for overall effect: Z = 0.85 (p = 0.40) Test for subgroup differences: Chi 2 = 0.41, df = 2 (p = 0.82); I 2 = 0% Heterogeneity: Tau 2 = 0.01; Chi 2 = 6.23, df = 6 (p = 0.40); I 2 = 4% 0.1 0.2 0.5 1 2 5 10 Favours biological Favours mechanical 0.1 0.2 0.5 1 2 5 10 Favours biological Favours mechanical A Systemic arterial embolism Study or Subgroup Weight Risk Ratio Risk Ratio IV, Random, 95% CI IV, Random, 95% CI B Endocarditis Study or Subgroup Weight Risk Ratio Risk Ratio IV, Random, 95% CI IV, Random, 95% CI monitoring processes currently available, the difference in the risk of bleeding for patients treated with mechanical versus biological prostheses may be lower than that found in this review. In our systematic review, three of the four trials considered used the first generation biological prostheses and single disc mechanical prostheses. 13,17,20 Although uncontrolled studies suggest that second and third generation biological prostheses have greater durability, 24 in the study by Stassano et al., 21 which included both modern biological and modern mechanical prostheses, the increased risk of reoperation was similar to that observed in the other trials. Concordance and discordance in relation to other studies We found a single meta-analysis that included three trials comparing old generation biological and mechanical prostheses, which was published by Kassai et al. 13 15 years ago. In the current review, we identified an additional study 21 that compared modern prostheses. In addition, the randomized studies of the Veterans Affairs group 14 and the Edinburgh group 18 presented new publications with extended follow-up periods of 15 and 20 years, respectively. 17,20 Our results, as well as adding an additional study, reflect long-term follow‑up, which is fundamental for better characterising the clinical progress of patients undergoing prosthetic valve implantation. 298