Methodological quality
The average overall quality rating was 0.81 ± 0.53, with ratings ranging from 0 to 1.81. Appendix A illustrates the average scores on the elements of the checklist. The analysis revealed lower scores related to the internal validity for both bias and selection bias, and for power analysis, which is related to the quality of reporting. The low values shown are due to the study type being a retrospective with no randomized samples validity studies. Acceptable interrater agreement was found (κ = 0.89; %‐agree = 94.9).

Follow up

The mean follow up period, calculated in 9 papers, was 35.66 [27.50- 43.81] months 14-18,20,21,25,26. The longest follow up period was 55.2 months 17. Follow up was 100% complete in 9 studies 14-17,19,20,23-25.

Main endpoints

Figure 2A shows that the log IRR of survival between AVR group and noAVR group was 0.58 [0.28, 0.87] (p-value = 0.0001; I-square = 24.16%, p-value = 0.25; Egger’s test: 0.18 [-0.18, 0.55], p-value = 0.002). This suggests that the overall survival is significantly better in the AVR group compared to the noAVR group. The funnel plot is shown in Figure 2B (funnel plot asymmetry test: p-value = 0.10). Moreover, the meta regression revealed that low LVEF is related to higher survival rates in the AVR group (p-value = 0.04) when compared to preserved LVEF (Figure 3A ). Conversely, LVEF has no impact on survival in the noAVR group (p-value = 0.18), as shown in Figure 3B .