ABC | Volume 112, Nº3, March 2019

Original Article Comparison of Biological and Mechanical Prostheses for Heart Valve Surgery: A Systematic Review of Randomized Controlled Trials Alberto Takeshi Kiyose, 1, 2 Erica Aranha Suzumura, 2 Lígia Laranjeira, 2 Anna Maria Buehler, 2 José Amalth Espírito Santo, 2 Otavio Berwanger, 2 Antonio Carlos de Camargo Carvalho, 1 Angelo Amato de Paola, 1 Valdir Ambrósio Moises, 1 Alexandre Biasi Cavalcanti 2 Universidade Federal de São Paulo (UNIFESP), 1 São Paulo, SP – Brazil Hospital do Coração (HCOR), 2 São Paulo, SP – Brazil Mailing Address: Alberto Takeshi Kiyose • Rua Alabarda, 240 casa 08. Postal Code 04641-020, Jardim dos Estados, São Paulo, SP – Brazil E-mail: atk@uol.com.br, clinicatakeshi@gmail.com Manuscript received August 08, 2017, revised manuscript October 04, 2017, accepted September 05, 2018 DOI: 10.5935/abc.20180272 Abstract Background: The choice of a mechanical (MP) or biological prosthesis (BP) for patients with valvular heart disease undergoing replacement is still not a consensus. Objective: We aimed to determine the clinical outcomes of MP or BP placement in those patients. Methods: We conducted a systematic review and meta-analysis of randomized controlled trials (RCTs) that compared biological prostheses and mechanical prostheses in patients with valvular heart diseases and assessed the outcomes. RCTs were searched in the MEDLINE, EMBASE, LILACS, CENTRAL, SCOPUS and Web of Science (from inception to November 2014) databases. Meta-analyses were performed using inverse variance with random effects models. The GRADE system was used to rate the quality of the evidence. A P-value lower than 0.05 was considered significant. Results: A total of four RCTs were included in the meta-analyses (1,528 patients) with follow up ranging from 2 to 20 years. Three used old generation mechanical and biological prostheses, and one used contemporary prostheses. No significant difference in mortality was found between BP and MP patients (risk ratio (RR = 1.07; 95% CI 0.99-1.15). The risk of bleeding was significantly lower in BP patients thanMP patients (RR=0.64; 95%CI 0.52-0.78); however, reoperations were significantly more frequent in BP patients (RR = 3.60; 95% CI 2.44-5.32). There were no statistically significant differences between BP and MP patients with respect to systemic arterial embolisms and infective endocarditis (RR = 0.93; 95% CI 0.66-1.31, RR = 1.21; CI95% 0.78-1.88, respectively). Results in the trials with modern and old prostheses were similar. Conclusions: The mortality rate and the risk of thromboembolic events and endocarditis were similar between BP and MP patients. The risk of bleeding was approximately one third lower for BP patients than for MP patients, while the risk of reoperations was more than three times higher for BP patients. (Arq Bras Cardiol. 2019; 112(3):292-301) Keywords: Heart Valve Prosthesis; Bioprosthesis; Metal-on-Metal Joint Prosthesis; Heart Valve Prosthesis Implantation/ trends; Review. Introduction In the early 1960's, valve replacement surgery using prostheses completely changed the natural history of patients with valvular heart disease. Approximately 90,000 valve prostheses are implanted in the United States, and 280,000 are implanted worldwide each year. 1 Currently, the total number of biological valve prosthesis implants surpasses that of mechanical prosthesis implants. 2-4 The factors that seem to affect the increased use of biological prostheses include advances in their construction, leading to increased durability, and the fact that they do not require permanent use of oral anticoagulants. 5 However, biological prostheses still present an increased risk of structural deterioration and the need for reoperation, although the surgical risk involved in reoperation has decreased substantially in recent years. 6 Furthermore, in the event of a stenosis disorder, patients with aortic bioprosthesis impairment can be treated with a catheter-implanted prosthesis. 7 A systematic review of randomized trials published in 2000 comparing mechanical and biological valve prostheses suggested that no difference in mortality existed between the two implant types. 8 There was, however, less risk of reoperation with mechanical prostheses but increased risk of bleeding compared to biological prostheses. There are no recent systematic reviews comparing the performance of biological valve prostheses with that of mechanical prostheses. Since the publication of the last review, further randomized studies may have been published that better reflect progress in prosthesis development, surgical techniques and clinical treatments during that time period. The objective of the present systematic review of randomized studies was to compare the effect of biological valve prosthesis use with 292