Tratamento de trombose de prótese valvar: revisão sistemática e metanálise

Abstract

Introduction: Prosthetic valve thrombosis (TPV) is a serious complication that threatens the lives of patients with prosthetic heart valves. Thrombolysis (TH) or surgery (SU) can be used to treat obstructive TPV, best initial therapeutic option is unknown. Methods: We conducted a systematic review in PubMed, Web of Knowledge, HINARI, LILACS and EMBASE, including papers indexed until August 22th, with at least 10 patients. The methodological quality of studies was assessed using the Newcastle-Ottawa Scale. All analysis were performed using Comprehensive Meta Analysis Software. The pooled results were compared by the random effects analysis for the outcomes: death, stroke, bleeding, embolic events and success rate. We evaluated publication bias and performed sensitivity analysis, heterogeneity (I2), meta-regression and subgroup analysis. Results: Forty seven studies were included (2055 patients). Mortality for SU was 18.1% (14.7-22.0; I2 62%), while mortality for TH was 6.9% (6.0-9.6; I2 0%) (I2 0%) (p <0.001).We find the following results for other outcomes, for SU and TH respectively: embolic events 4.1% (2.7-6.4) and 13.9% (11.6-16.7) (p<0.001); stroke 4.3% (2.7-6.6) and 6.3% (4.5-8.2) (p=0.18); success rate 81.9% (77.1-85.9) and 80.9% (75.6-85.3) (p=0.76); bleeding 4.6% (2.9-7.1) e 7.3% (5.5-9.6) (p=0.08); death or stroke 18% (14.1-21.2) e 13.4% (11.0-16.1) (p=0.02). Subgroup analysis evaluating exclusively mitral valve and another subgroup analysis for left side valve thrombosis find similar results, with lower mortality for TH than CI: 6.3% vs 20.5% for mitral valve and 6.3% vs 17.8% for left side valve thrombosis. Conservative subgroup analysis comparing TH studies with more than 40 patients with SU studies without those four studies with mortality higher than 30% found mortality rate of 13.8% (11.8-16.1) for SU group and 8.5% (6.5-11.0) in TH group (p<0.001). Publication bias evaluation for SU studies was negative (p=0.84). It was positive for TH studies (p=0.003). Using Trim and Fill to fix publication bias, the death rate imputed for TH was 8.3% (6.7%,10.4%), a value that is still lower than surgical results. Conclusion: Mortality rates for surgery in patients with TPV is very high, counterbalancing stroke rates of thrombolysis treatment. Our subgroups analyses makes difference in mortality found in this meta-analysis robust