The comprehensive search yielded 2520 articles, of which 19 articles were potentially eligible after excluding duplicates and irrelevant articles by reading titles and abstracts. After full-text review, four articles were excluded because they did not provide data on Cabrol-related morbidity, mortality or graft complications. In addition, a study by Coselli et al. [32] was excluded from the quantitative analysis due to the lack of a description of the duration of follow-up. Therefore, 14 studies with a total of 833 patients were included in this meta-analysis. (Fig. 1) illustrates the selection process of the 14 articles. The pooled preoperative and perioperative characteristics are shown in (Table 1). The mean follow-up time after Cabrol surgery was 4.4 years (range 1.7–8.6 years) for a total of 3727.3 patient-years. The characteristics of the included studies are summarized in (Supplementary Appendix 3) [7,8,9,10, 12, 20,21,22,23,24,25, 33,34,35]. Consistently, classical Cabrol technique is still used in the majority of studies, with 80.7% (672/833) of the total patients in 11 studies using it [7,8,9,10, 12, 20, 21, 24, 25, 33, 34]. The proportion of reoperation was 34.5% (126/365) in the 8 studies [9, 12, 20,21,22,23, 33, 35]. 27.7% (96/347) of connective tissue disease in 7studies [9, 12, 20,21,22,23, 33]. Aortic dissection in 6 studies was 42.2% (136/322) [9, 12, 21,22,23, 33].

Flow chart of the selection process for studies included in the systematic review and meta-analysis

10 studies were included and the combined all-cause early mortality was 9.0% [7, 9, 12, 20,21,22,23,24,25, 33].

7 studies were included and the combined bleeding reoperation rate was 4.9% (17 patients) [12, 20,21,22,23,24, 33].

10 studies were included, with a combined late mortality of 3.63% (per patient-year) using the random effects model, with a 95% confidence interval of (2.79–4.73) and a heterogeneity I2 of 37.1% [7, 9, 12, 20,21,22,23,24,25, 33] ( Fig. 2).

Forest plots and 95% confidence intervals for combined late mortality

The definition of aortic root reoperation followed the description of reinterventions in the guidelines for reporting mortality and morbidity after cardiac valve interventions [27]. 11 studies were included, with a random effects model combined root reoperation of 0.64% (per patient-year), a 95% confidence interval of (0.35–1.16) and a heterogeneity I2 of 7.6% [7,8,9, 12, 20,21,22,23,24,25, 33] (Fig. 3).

Forest plots and 95% confidence intervals for combined aortic root reoperation

8 studies were included, with a random effects model combined hemorrhage of 0.57% (per patient-year), 95% confidence interval of (0.25–1.31) and heterogeneity I2 of 0% [8, 9, 12, 20,21,22, 24, 33] (Fig. 4).

Forest plots and 95% confidence intervals for Hemorrhage

8 studies were included, with a random effects model combined thromboembolism of 0.66% (per patient-year), a 95% confidence interval of (0.16–2.74) and a heterogeneity I2 of 76.2% [8, 10, 12, 20,21,22,23, 33] (Fig. 5).