Introduction
There is still no perfect treatment for paediatric aortic valve disease. Aortic valve replacement (mechanical valve and biological valve) has not become the first choice for paediatric aortic valve disease due to the risk of patient-prosthesis mismatch, anticoagulant complications of mechanical valves, and premature biological valve failure.Reference Sharabiani, Dorobantu and Mahani1 Currently, the main treatment options for paediatric aortic valve disease are aortic valve repair and Ross procedure, with Ross procedure being the gold standard for complex paediatric aortic valve disease. After extensive clinical research and practical experience, the efficacy of Ross procedure has been confirmed.Reference Sharabiani, Dorobantu and Mahani1,Reference Mazine and El-Hamamsy2 However, Ross procedure is a complex procedure with significant complications, which poses risks of degeneration and reoperation for both the aortic and pulmonary valves.Reference Charitos, Takkenberg and Hanke3 Patients with pulmonary valve disease, Marfan syndrome, and other conditions are unable to undergo Ross procedure.Reference Morita and Kurosawa4 In recent years, Ozaki procedure has gradually become a new option for treating adult aortic valve disease due to its excellent haemodynamic performance, low reoperation rate, and no need for anticoagulation.5–Reference Krane, Boehm and Prinzing7 Wiggins et al. also confirmed the safety and efficacy of Ozaki surgery in treating paediatric aortic valve disease in their early research.Reference Wiggins, Mimic and Issitt8 To evaluate the efficacy and prognosis of Ozaki procedure for paediatric aortic valve disease and compare it with Ross procedure, we conducted a systematic review and meta-analysis of all studies comparing Ozaki procedure and Ross procedure in treating paediatric aortic valve disease.
Methods
Literature search
This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses.Reference Page, McKenzie and Bossuyt9 The study was registered in PROSPERO (CRD420251132818). Two authors searched for potentially relevant literature in PubMed, EMBASE, and Web of Science databases, with a cut-off date of 20 August 2025. Keywords: Aortic valve neocuspidisation, Ozaki, AVNeo, Ross. There was no language restriction. The detailed search strategy is provided in Supplementary Table 1.
Study selection and data collection
Inclusion criteria: 1. Clinical trials or observational studies; 2. Studies comparing the outcomes of Ozaki procedure and Ross procedure for paediatric aortic valve disease; 3. Studies including at least 10 paediatric patients with aortic valve disease who underwent Ozaki procedure and Ross procedure. Exclusion criteria: 1. Systematic reviews, meta-analyses, conference abstracts, case reports, animal studies; 2. Duplicate data or duplicate publications; and 3. Studies lacking primary indicators of postoperative outcomes or insufficient data.
Data collection: 1. Basic information of the literature: first author, publication year, research country, research design; 2. Baseline data: number of patients, gender of patients, age of patients, previous surgical history, type of aortic valve disease, valve leaflet anatomy, aortic valve ultrasound data, and preoperative comorbidities; 3. Primary outcome indicators: all-cause mortality (in-hospital, longest follow-up time), reoperation, moderate or severe aortic valve stenosis, or regurgitation; and 4. Secondary outcome indicators: cardiopulmonary bypass time, aortic cross-clamping time, ICU time, ventilator assistance time, and postoperative complications. When the included continuous variables are the median between quartiles, we use specific calculations to convert the data into mean (m) ± standard deviation (s).Reference Wan, Wang and Liu10,Reference Luo, Wan and Liu11
Quality assessment
We used the Cochrane tool for assessing bias risks in the randomised controlled trials included in the literature. For cohort studies, the Newcastle-Ottawa Scale was used to detect the quality of the included literature. The two authors independently assessed the quality of eligible studies, and disagreements were resolved through consensus.
Statistical analysis
Kaplan–Meier curves were plotted using StataMP-64 software. Statistical analysis was conducted using Review Manager 5.3 (Cochrane Collaboration, Oxford, UK). The I2 statistic was used to test for heterogeneity among the outcomes. I2 ≤ 50% indicates no significant heterogeneity among the studies, and a fixed-effects model was used to evaluate the overall effect size and 95% confidence interval for each outcome event. If I2 > 50% indicates significant heterogeneity among the studies, a random-effects model was used. P < 0.05 was considered statistically significant.
Results
literature
Through systematic literature search, we identified a total of 232 articles (PubMed (n = 59), EMBASE (n = 91), and Web of Science (n = 82)). After excluding duplicate articles (n = 95), 137 articles remained. After screening the titles and abstracts of these articles, 7 articles remained. The remaining 7 articles were subjected to full-text search, and the articles were evaluated according to predetermined inclusion and exclusion criteria. Ultimately, 4 articles were included in the meta-analysis. The process of including studies is shown in Figure 1.
Flow diagram of included studies.
All four included studies were retrospective cohort studies. Detailed information on the included studies is provided in Table 1. The quality assessment of the four included studies is provided in Supplementary Table 2. The four included articles involved a total of 243 patients (117 Ozaki operations and 126 Ross operations). The baseline characteristics of the patients in the included studies are presented in Supplementary Table 3.
Characteristics of included literatures
RCS = retrospective cohort study.
Kaplan–Meier curve of freedom from reoperation
The studies by Komarov, Wang, and Zhang all used the Kaplan–Meier method to construct the curves of patients free from reoperation in two groups. We reconstructed individual patient data from the published Kaplan–Meier curves,Reference Liu, Zhou and Lee16 and after combining the three sets of data, we used the Kaplan–Meier method to construct new curves of freedom from reoperation and cumulative incidence of reoperation (Figure 2). The new curves suggest that the rate of freedom from reoperation in the Ozaki procedure is lower than that in the Ross procedure, but there is no statistically significant difference between the two (p = 0.075).
Kaplan−Meier curve (a) freedom from reoperation. (b) cumulative incidence of reoperation.
Primary outcome indicators
In-hospital all-cause mortality
The in-hospital all-cause mortality of the Ozaki procedure ([odds ratio = 1.38; 95% confidence interval: 0.38, 5.07, p = 0.63], Figure 3(a)) was similar to that of the Ross procedure, and the difference between the two was not statistically significant.
Forest plot of odds ratios for primary outcome indicators.
All-cause mortality during the follow-up period
The all-cause mortality during the follow-up period of the Ozaki procedure ([odds ratio = 1.85; 95% confidence ratio: 0.54, 6.32, p = 0.32], Figure 3(b)) was similar to that of the Ross procedure, and there was no statistically significant difference between the two.
Reoperation on the aortic valve
The reoperation on the aortic valve in Ozaki procedure ([odds ratio = 10.48; 95% confidence interval: 2.22, 49.40, p = 0.003], Figure 3(c)) was higher than that in Ross procedure, and the difference between the two was statistically significant.
Reoperation on the pulmonary valve
Patients without reoperation on the pulmonary valve after Ozaki procedure ([odds ratio = 0.21; 95% confidence interval: 0.03, 1.23, p = 0.08], Figure 3(d)) and the rate of reoperation on the pulmonary valve after the Ross procedure was 6.3%. The difference between the two was not statistically significant.
Cumulative reoperation
The cumulative reoperation rate of the Ozaki procedure ([odds ratio = 2.29; 95% confidence interval: 0.93 – 5.66, p = 0.07], Figure 3(e)) was higher than that of the Ross procedure, but the difference between the two was not statistically significant.
Moderate or severe aortic valve stenosis or regurgitation
The incidence of moderate or severe aortic valve stenosis or regurgitation after Ozaki procedure ([odds ratio = 15.58; 95% confidence interval: 5.64, 43.06, p < 0.00001], Figure 3(f)) was higher than that of Ross procedure, and the difference was statistically significant.
Secondary outcome indicators
Postoperative complications
The incidence of postoperative complications after Ozaki procedure ([odds ratio = 0.24; 95% confidence interval: 0.04, 1.62, p = 0.14], Figure 4(a)) was not statistically significant compared with that of Ross procedure.
Forest plot of odds ratios for secondary outcome indicators.
Cardiopulmonary bypass time
The cardiopulmonary bypass time of Ozaki procedure ([odds ratio = −32.09; 95% confidence interval: −45.05, −19.14, p < 0.00001], Figure 4(b)) was shorter than that of Ross procedure, and the difference was statistically significant.
Aortic cross-clamping time
The aortic cross-clamping time in the Ozaki procedure ([odds ratio = −20.39; 95% confidence interval: −43.68, 2.90, p = 0.09], Figure 4(c)) showed no statistical significance compared with the Ross procedure.
Ventilator assistance time
The ventilator assistance time after Ozaki procedure ([odds ratio = 1.71; 95% confidence interval: −42.70, 46.13, p = 0.94], Figure 4(d)) showed no statistical significance compared with that of Ross procedure.
Intensive care unit time
The ICU time of Ozaki procedure ([odds ratio = −0.38; 95% confidence interval: −0.93, 0.16, p = 0.17], Figure 4(e)) was not statistically significant compared with that of Ross procedure.
Discussion
The Ozaki procedure, first reported in 2011, is a surgical method that uses autologous pericardium treated with glutaraldehyde to reconstruct aortic valve leaflets.Reference Ozaki, Kawase and Yamashita5 It can be widely applied to adult patients with aortic valve disease, including stenosis, regurgitation, stenosis with regurgitation, and infective endocarditis. Studies have shown that the Ozaki procedure retains the natural expansion characteristics of the aortic root, increases the effective orifice area, and has haemodynamic properties similar to native valves.Reference Krane, Boehm and Prinzing7,Reference Koechlin, Schurr and Miazza17,Reference Yamamoto, Iino and Shintani18 It has excellent biocompatibility and valve function and does not require anticoagulation after surgery, with a low reoperation rate.Reference Ozaki, Kawase and Yamashita6,Reference Iida, Fujii and Akiyama19 Therefore, the Ozaki procedure is a promising surgical method. In recent years, some centres have begun to use the Ozaki procedure to treat paediatric patients with aortic valve disease, and some have achieved good results.Reference Wiggins, Mimic and Issitt8 The Ozaki procedure provides a new treatment option for paediatric patients with aortic valve disease and may become an alternative to the Ross procedure. Therefore, we conducted a systematic review and meta-analysis of all studies comparing the Ozaki procedure with the Ross procedure for the treatment of paediatric aortic valve disease.
This meta-analysis included four studies. The baseline characteristics of patients in the included studies were different. The choice of surgical procedures in different research centreas might be influenced by specific patient pathologies (such as stenosis-dominant and regurgitation-dominant) or aortic root anatomical features, leading to baseline mismatch. Such differences may cause the prognostic indicators (such as reoperation rate and haemodynamic outcomes) to be affected by confounding factors. Due to the small number of included studies, we were unable to conduct sufficient subgroup analyses to correct this bias, which suggests that future studies should focus on the standardisation of surgical indications and baseline matching.
Our meta-analysis shows that both the Ozaki procedure and the Ross procedure have relatively low in-hospital all-cause mortality and all-cause mortality during the follow-up period, with no statistically significant difference between the two. Our meta-analysis indicates that the incidence of moderate or severe aortic valve stenosis or regurgitation after the Ozaki procedure is significantly higher than that of the Ross procedure (42% vs 6.4%), with a statistically significant difference. The reoperation on the aortic valve after the Ozaki procedure is also higher than that of the Ross procedure (12% vs 0.8%), with a statistically significant difference. It has only been 14 years since the Ozaki procedure was first reported, and there are currently no reports on the long-term effects and reoperation rates of the Ozaki procedure. However, Ozaki, Pirola, and others have reported the mid-term results of the Ozaki procedure. Ozaki evaluated the mid-term results with a maximum follow-up time of 118 months, and the freedom from reoperation rate was 95.8%.Reference Ozaki, Kawase and Yamashita6 Pirola et al. reported the mid-term results with a follow-up of 5 years, and no patients required reoperation intervention.Reference Pirola, Mastroiacovo and Arlati20 Both studies demonstrated excellent mid-term efficacy of the Ozaki procedure, but both were conducted on adult patients with aortic valve disease. For paediatric patients with aortic valve disease, the aorta changes with the patient’s growth, while the pericardial patch has no growth potential. After the Ross procedure, the autologous pulmonary valve still has growth potential and grows with the patient’s growth, which may be one of the main reasons for the higher reoperation rate of the aortic valve after the Ozaki procedure compared to the Ross procedure. Patients after the Ross procedure are at risk of requiring a second pulmonary valve surgery due to pulmonary valve disease. Our meta-analysis shows that the incidence of second pulmonary valve surgery after the Ross procedure is 6.3%, while there are no patients requiring reoperation due to pulmonary valve disease after the Ozaki procedure. For the overall reoperation rate, there is no statistical difference between the Ozaki procedure and the Ross procedure (11.9% vs 7.1%), and both have high freedom from reoperation rates (88.1% vs 92.9%). Recent evidence suggests that in paediatric patients, autologous pericardium may have better anti-calcification and growth adaptation potential.Reference Jiang, Cui and Li21 The materials used in the Ozaki procedure (autologous pericardium vs xenogeneic pericardium) may affect valve durability. However, in the included literature of this study, the reports on the materials used were incomplete and inconsistent, and a quantitative analysis of the impact of material type on the comparison results could not be conducted. Future studies should report in detail and distinguish the types of materials used and conduct long-term follow-up to compare the durability differences between autologous pericardium and xenogeneic pericardium in paediatric Ozaki procedures, in order to optimise surgical decision-making.
For secondary outcome measures, our meta-analysis showed that the cardiopulmonary bypass time of the Ozaki procedure was shorter than that of the Ross procedure, with a statistically significant difference. This is related to the complexity of Ross procedure. However, Ozaki procedure uses standardised measuring instruments and standardised valve leaflet moulds, making the surgical process simple and repeatable, so it does not increase the cardiopulmonary bypass time of patients. For aortic cross-clamping time, ventilator assistance time, ICU time, and postoperative complication rate, there was no statistical difference between Ozaki procedure and Ross procedure.
Study limitations
Our meta-analysis has certain limitations. This meta-analysis is based on aggregated data from different studies and cannot obtain patient-level data. Additionally, among the four studies we included, the sample sizes were all relatively small, and the aggregated sample size was also small. It is necessary to conduct multi-centre studies to expand the sample size and use propensity score matching research methods or randomised controlled trials to reduce bias. Moreover, it is necessary to further investigate the long-term outcomes of the surgery.
Conclusions
In conclusion, our meta-analysis showed that there was no statistically significant difference in the rates of freedom from reoperation and all-cause mortality between the Ozaki procedure and the Ross procedure in the treatment of paediatric aortic valve disease. However, the Ozaki procedure was associated with a higher incidence of moderate or severe aortic valve stenosis or regurgitation and a higher rate of reoperation on the aortic valve, while the Ross procedure was associated with a higher rate of reoperation on the pulmonary valve. This meta-analysis could only incorporate four direct comparison studies, which objectively reflects the current scarcity of high-quality comparative evidence in this field. Although we attempted to quantitatively synthesise the available data, the generalisability of the results is still to be interpreted with caution due to the limited number of included studies and sample sizes. The findings of this study should be regarded as preliminary and exploratory evidence rather than definitive conclusions guiding clinical practice. However, this meta-analysis still provides a trend reference under the best available evidence at present, and more prospective, multicentre studies are needed in the future for further validation.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1047951126111299.
