Pregnancy after Ozaki aortic valve reconstruction in a woman with bicuspid aortic valve: a case report

Learning points.

• The Ozaki procedure maintained stable valve haemodynamics and structure throughout pregnancy and the postpartum period.

• Transient increases in transaortic gradients during pregnancy likely reflect physiological increases in cardiac output rather than valve deterioration.

• Ozaki aortic neocuspidization may be a favourable option for women of childbearing age, avoiding lifelong anticoagulation.

Introduction

Aortic valve stenosis represents the most prevalent valvular heart disease, with symptomatic severe disease carrying mortality rates approaching 50% without intervention.^1–3 While prosthetic valve replacement remains the gold standard, complications including prosthesis-patient mismatch, anticoagulation requirements, and structural valve degeneration pose significant challenges, particularly for young women contemplating pregnancy.^3–5

Summary figure

Clinical course of pregnancy following the Ozaki procedure. Schematic timeline illustrating the clinical course of a woman with congenital aortic valve disease who underwent the Ozaki procedure and subsequently completed a successful pregnancy. The patient underwent balloon aortic valvuloplasty at 5 years of age, developed severe aortic stenosis by 19 years, and underwent Ozaki aortic valve reconstruction at 20 years of age. One-year postoperative follow-up demonstrated normalization of left ventricular dimensions and improvement in the left ventricular outflow tract (LVOT) gradient. Pregnancy at 22 years of age was managed by a multidisciplinary cardio-obstetric team, with stable transvalvular gradients (39/18 mmHg) documented at both 13 and 26 weeks of gestation. The pregnancy progressed uneventfully, culminating in the vaginal delivery of a healthy infant at 39 + 5 weeks. At 18 months postpartum, valve function remained stable without structural deterioration. This case highlights the excellent haemodynamic durability of the Ozaki procedure during the physiological stress of pregnancy and supports its potential advantages in young women of childbearing age, including preservation of native valve function and avoidance of long-term anticoagulation.

The Ozaki procedure, introduced in 2011, employs glutaraldehyde-treated autologous pericardium to create new tri-leaflet aortic valve neocusps.⁶ This technique has demonstrated favourable short- and mid-term outcomes with improved haemodynamics, low complication rates, and minimal anticoagulation requirement.^7–10 However, pregnancy imposes significant haemodynamic, hormonal, and metabolic changes. It is unknown how the pregnancy environment can affect glutaraldehyde-treated autologous pericardium aortic valve. To our knowledge, evidence regarding pregnancy following aortic valve neocuspidization (Ozaki procedure) remains scarce, with the literature currently limited to a single case report.¹⁷

Case presentation

The patient was born with a bicuspid aortic valve and severe stenosis and underwent balloon valvuloplasty at age 5 years (gradient reduced from 85 to 20 mmHg), resulting in moderate residual stenosis and insufficiency managed with angiotensin-converting enzyme (ACE) inhibitors. At 19 years of age, echo showed thickened bicuspid aortic valve, moderate-severe aortic stenosis creating 63/36 mmHg max/mean gradient and moderate-severe aortic regurgitation, with vena contracta 5 mm, aortic regurgitation (AR) slope 3.8 m/s², and flow reversal in the abdominal aorta (Table 1). Cardiac MRI showed severe aortic regurgitation (regurgitant fraction 45%), moderately dilated left ventricle (130 mL/m², z-score +4.8), and low-normal systolic function (left ventricle ejection fraction 55%). The patient underwent the Ozaki procedure at age 20 (Figure 1). Postoperative echocardiography revealed no stenosis and mild regurgitation (Supplementary material).

Table 1.

Echochardiographic data

Timeline	LVEDD	LVESD	FS	AS max gradient	AS mean gradient	AI	Ascending aorta diameter	Aortic arch diameter and z-score
1 year before operation	57	36	37	63	36	Severe	29	29 3.6
4 months post-operation	46	27	41	20	13	Trace	27	29 3.6
Pregnancy 13 weeks	45	25	44	39	18		27	29 3.6
Pregnancy 26 weeks	42	26	38	39	18		27	29 3.6
Delivery	49	33	33	40	23		27	29 3.6
5 months post-delivery	46	25	46	33	15		27	29 3.6
17 months post-delivery	49	30	39	33	18	Trace	27	29 3.6

Abbreviations: LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; FS, fractional shortening; AS max gradient, maximum aortic stenosis transvalvular gradient; AS mean gradient, mean aortic stenosis transvalvular gradient; AI, aortic insufficiency.

Figure 1.

Representative intraoperative photographs demonstrating the key steps of the Ozaki procedure. Representative intraoperative images demonstrating the principal steps of the Ozaki procedure using glutaraldehyde-treated autologous pericardium for aortic valve reconstruction. (A) Preparation of the autologous pericardial patch using the Ozaki template system. The pericardium is marked according to the measured leaflet size to create a patient-specific neocusp. The asterisk indicates the reference point for template alignment, and the arrow denotes the orientation mark. (B) Harvested autologous pericardium after exposure and before excision, with stay sutures providing optimal visualization and handling. The asterisk indicates the underlying cardiac structures. (C) Glutaraldehyde-treated autologous pericardium secured to the Ozaki sizing board, allowing precise leaflet trimming according to the selected template. The arrow indicates the treated pericardial edge. (D) Final intraoperative view of the completed aortic valve neocuspidization, demonstrating three newly constructed autologous pericardial leaflets with excellent symmetry and central coaptation, resulting in a competent tri-leaflet aortic valve.

Cardiac MRI 1 year postoperatively demonstrated a left ventricular ejection fraction of 60%, normal ventricular dimensions (left ventricular end-diastolic volume, 106 mL; z-score, 0), and an aortic regurgitant fraction of 6%. The patient maintained NYHA class I–II functional status on bisoprolol 1.25 mg daily. Prior to surgery, the patient had moderate-to-severe bicuspid aortic stenosis and moderate-to-severe aortic regurgitation with left ventricular dilatation, corresponding to a high-risk pregnancy profile (mWHO class III) associated with a significantly increased risk of maternal cardiovascular morbidity and mortality. Following Ozaki aortic valve reconstruction, serial follow-up demonstrated normalization of left ventricular dimensions and only minimal residual valvular dysfunction, resulting in reclassification to mWHO class I. Consistent with the absence of residual high-risk features and a low CARPREG II score, her predicted risk of maternal adverse cardiovascular events during pregnancy became low.

Preconception genetic screening was normal. A multidisciplinary preconception counselling with a gynaecologist and a cardiologist was done. At age 22, the patient conceived spontaneously while on aspirin 100 mg daily.

First trimester: Echocardiography at 13 weeks showed increased peak gradient (39 mmHg, mean 18 mmHg) with trace regurgitation. The patient was in NYHA II functional state.

Second trimester: Foetal anatomy scan at 21 weeks revealed two small muscular ventricular septal defects and normal growth. Maternal echocardiography at 26 weeks demonstrated stable valve function (39/18 mmHg max/mean gradient, trace regurgitation).

First-trimester combined screening, including nuchal translucency (NT) measurement and maternal serum biochemical markers, demonstrated an NT of 1.3 mm and an estimated risk of 1:2885 for chromosomal abnormalities. Combined second-trimester screening yielded a risk estimate of 1:11,886. Neither non-invasive prenatal testing nor amniocentesis was performed.

Third trimester: Serial echocardiography showed stable gradients. At 33 weeks, the patient reported fatigue without dyspnoea; she was advised to reduce activity. Iron deficiency anaemia (haemoglobin 11.1 g/dL) was treated with intravenous and oral supplements. At 38 weeks, she remained stable with normal cardiac function.

The peripartum strategy was formulated by a multidisciplinary Pregnancy Heart Team consisting of a high-risk obstetrician, a cardiologist, and anaesthesiologists. In accordance with the ESC 2025 cardiac pregnancy guidelines, a term planned vaginal delivery was selected due to the patient’s stable structural and haemodynamic down-stratification to mWHO class I, which minimized the surgical risks and massive fluid shifts associated with a caesarean section. The plan incorporated subacute bacterial endocarditis prophylaxis for her reconstructed valve tissue and an early epidural protocol to safely suppress labour-induced adrenergic surges.

The patient was admitted with a spontaneous onset of labour at 39 + 6 weeks, which was augmented with oxytocin infusion. Anaesthesia was successfully established via an epidural catheter utilizing a step-up titration of lignocaine (Esracain), ropivacaine, and sequential fentanyl, keeping her cardiovascularly stable throughout labour.

During the second stage, the obstetric team identified a prolonged labour pattern alongside acute foetal distress, which manifested as foetal tachycardia. To ensure maternal-foetal safety, quickly shorten the second stage of labour, and eliminate exhausting Valsalva manoeuvres, a low plane vacuum extraction with a medio-lateral episiotomy was successfully performed. The infant was born alive weighing 3088 g with Apgar scores of 6/8 at 1/5 min.

Throughout labour, delivery, and the early postpartum period, the patient remained completely haemodynamically stable. Total estimated blood loss was ∼500 mL, after which she was safely transferred to a regular maternity ward for routine surveillance and intravenous iron therapy to treat her postpartum anaemia, with haemoglobin of 8.2 g/dL. Echocardiogram 2 days post-delivery showed no change from prior exam, with mild aortic stenosis and trace regurgitation. Aspirin was recommended for up to 6 weeks postpartum.

At 18 months postpartum (4 years post-Ozaki), echocardiography showed mild aortic stenosis with minimal regurgitation. Pressure gradient increased slightly postpartum and then decreased. Neonatal echocardiography showed no structural abnormalities; the ventricular septal defects identified prenatally had spontaneously closed. The baby was born healthy and is developing appropriately for age. The patient maintained NYHA class I–II functional status with normal karyotype on genetic microarray testing.

Discussion

This case represents the first documented pregnancy following the Ozaki procedure, demonstrating valve stability throughout gestation and the postpartum period. Pregnancy is associated with elevated oestrogen and progesterone levels that may promote tissue remodelling and alter calcium metabolism. These hormonal shifts are hypothesized to contribute to accelerated bioprosthetic valve calcification and degeneration.¹¹ Pregnancy is also associated with increased calcium demand and altered lipid metabolism, which may further predispose bioprosthetic valves to calcification and valve dysfunction. Pregnancy is also associated with hypercoagulability that may affect bioprosthetic valves.¹²

The preparation of autologous pericardium for the Ozaki procedure differs fundamentally from the techniques used for bioprosthetic heart valves. The Ozaki procedure uses autologous pericardium, harvested directly from the patient during surgery, minimizing immunogenicity and eliminating xenogeneic antigens. In contrast, bioprosthetic valves are typically constructed from xenogeneic tissue—most commonly bovine or porcine pericardium—which requires extensive processing to reduce immune response and risk of calcification. For the Ozaki procedure, the autologous pericardium is treated intraoperatively with 0.6% glutaraldehyde for 10 min, followed by thorough saline rinsing to remove residual glutaraldehyde and reduce cytotoxicity.¹³ In contrast, bioprosthetic valves undergo industrial-scale fixation, often with higher concentrations and longer exposure to glutaraldehyde, and may include additional steps such as decellularization to further reduce immunogenicity and enhance durability.¹⁴ In addition, bioprosthetic valve preparation frequently incorporates advanced modifications to address the limitations of glutaraldehyde fixation, such as neutralization of free aldehyde groups, loading with glycosaminoglycans, polymer coatings (e.g. poly-2-hydroxyethyl methacrylate), and alternative crosslinkers (e.g. genipin, ribose, polyphenols) to improve anticalcification, antithrombogenicity, and endothelialization. The Ozaki procedure, by contrast, relies solely on glutaraldehyde fixation.

There is extensive experience with bioprosthetic valves in pregnancy. There is conflicting evidence regarding the effect of pregnancy on tissue valves, with deterioration during pregnancy reported in several studies but not in others.¹⁵ The differences in preparation of the Ozaki valve may result in a different effect of pregnancy on the valve. It is thus important to investigate the profile of this novel approach during pregnancy and postpartum periods.

In the case presented here, pregnancy seemed to have little or no effect on the valve. The transient increase in aortic gradient during pregnancy (26 to 39 mmHg peak) likely reflects pregnancy-related haemodynamic changes, including a 40%–50% increase in cardiac output and decreased systemic vascular resistance, rather than structural valve deterioration.

In 2025, Ozaki et al. reported mid-term data from the largest cohort of 1196 patients, demonstrating freedom from reoperation of 92% at 10 years, with younger patients showing slightly higher reoperation rates compared to older age groups.¹⁶ Thus, there is little experience with treatment should valve failure occur. The options at this point in time are repeat Ozaki operation (no published data) or surgical aortic valve replacement with either a mechanical valve or biological valve. A mechanical valve has better durability and may be a good option if she is past the fertility age. A bioprosthesis may last for a decade or longer and may be treated in the future with TAVR. Ross operation is another theoretical option.

This case suggests that the Ozaki procedure does not contraindicate pregnancy and may offer advantages over bioprosthetic valves for women of childbearing age by eliminating anticoagulation requirements, potentially reducing pregnancy-associated structural deterioration and maintaining haemodynamic stability during pregnancy’s physiologic stress.

Limitations

As a single case with 4-year post-procedure follow-up, generalizability is limited. Larger series with extended follow-up are needed to establish long-term durability and pregnancy-associated risks. The patient’s young age, preserved ventricular function, and optimal timing of intervention may have contributed to favourable outcomes.

Conclusion

This case demonstrates a favourable outcome of a pregnancy following the Ozaki procedure, with stable valve function and excellent maternal-foetal outcomes. Although based on a single case, these findings suggest that the Ozaki procedure may represent a promising option for selected young women with aortic valve disease contemplating future pregnancy, particularly in view of the avoidance of long-term anticoagulation. Prospective larger registries are warranted to establish comprehensive safety profiles and long-term outcomes.

Lead author biography

I am a senior cardiologist mainly practising in non-invasive cardiology, being the head of two big cardiology ambulatory centres. My research focuses on hyperlipidaemia, metabolic syndrome, and early diagnostics of atherosclerotic diseases. I have authored over 20 journal papers in top life science journals. I have participated in international congresses presenting original studies in different fields of science.

Supplementary material

Supplementary material is available at European Heart Journal – Case Reports online.

Author contributions

Hila Hochler (Conceptualization, Data curation, Methodology, Validation, Visualization, Writing—original draft, Writing—review & editing [lead]), Ilia Davarashvili (Conceptualization, Data curation, Methodology, Resources, Supervision, Validation, Visualization, Writing—review & editing [lead]), Tehila Avitan (Software, Visualization [supporting]), Eldad Erez (Conceptualization, Project administration [lead]), and Amiram Nir (Conceptualization, Investigation [lead])

Consent: Written informed consent for publication of this case report and any accompanying images was obtained from the patient in accordance with COPE guidance.

Funding

No specific funding was received for this work.

Data availability

The data underlying this article are available within the article and its supplementary material. Additional data are not publicly available in order to protect patient confidentiality.