Patients
A retrospective analysis was performed on 137 consecutive patients who underwent isolated and initial AVR at Yokosuka General Hospital Uwamachi from January 2009 to December 2020. Full sternotomy (CAVR), partial sternotomy (PAVR), and trans-right axillary AVR with Stonehenge technique (SHAVR) were performed in 67 (49%), 31 (23%), and 39 patients (28%), respectively. The choice of the approach was left to the surgeon’s preference. The following conditions were excluded from SHAVR: previous pneumectomy, ascending aorta enlargement (diameter > 50, shaggy aorta, difficulty of differential lung ventilation, deformity of the thoracic cage, and significant anatomical shift of aortic root to the left side. Patients who needed an emergency AVR because of endocarditis were excluded if they had a periannular abscess requiring annulus implantation and reconstruction with pericardium. With propensity score matching, the patients were adjusted with age, gender, comorbidity, New York Heart Association (NYHA) class, preoperative hemoglobin, and preoperative EF. Well-matched 22 (PAVR vs. CAVR) and 28 (SHAVR vs. CAVR) pairs were evaluated for surgical outcomes. Our selection process and the number of selected patients are shown in Figure 2. The Institutional Review Board approved this retrospective observational study. The approval included a waiver of informed consent.
Data collection
Medical records and preoperative examinations were used to collect the variables. Preoperative characteristics, such as sex, body mass index (BMI), NYHA classification, European System for Cardiac Operative Risk Evaluation (EuroSCORE II) scores, and comorbidities, including hyperlipidemia, hypertension, diabetes mellitus, chronic kidney disease (CKD), peripheral arterial disease, ischemic heart disease, old stroke, chronic obstructive pulmonary disease, and atrial fibrillation, were compared between the groups [21-23]. Anemia was defined as preoperative hemoglobin (Hb) <13 g/dL and <12 g/dL in males and females. CKD was defined as estimated glomerular filtration rate (eGFR) < 60 ml/min/1.73m2 in preoperative blood test. Surgical variables, including operation time, CPB time, surgical procedure, intensive care unit (ICU) and hospital stay, and hospital death, and postoperative complications, such as stroke, pneumonia, mediastinitis, bleeding, postoperative aortic regurgitation (≥moderate), tracheotomy, and postoperative hemodialysis, were also compared between the groups.
Surgical procedure
Conventional surgical aortic valve replacement was performed in all procedures without using a suture less aortic valve prosthesis.
Stonehenge technique
SHAVR was performed as described in previous reports [18, 20]. An 8 cm skin incision was made in the right anterior axillary line along the edge of the pectoralis major muscle, followed by the fourth or third intercostal thoracotomy. CPB was established via the subclavian artery or femoral artery for arterial cannulation and the femoral vein for venous cannulation. A longitudinal incision was made on the pericardium 2–3 cm away from the phrenic nerve. The pericardium was pulled up using several braided polyester sutures to put the ascending aorta and aortic root close to the chest wall according to the references [20], which improved the surgical view shortening the distance from the aortic root to the chest wall. Aortic cross-clamping was performed using a Cygnet Flexible Clamp (Vitalitec Inc., Plymouth, MA, USA) from within the wound. To protect myocardium, cold blood cardioplegia was administered in an antegrade fashion into the aortic root and the coronary ostia approximately every 20 min (Figure 3).
Partial sternotomy
Partial upper or lower sternotomy was performed standardized [4, 24, 25]. The 8 cm midline skin incision was made. The upper or lower sternotomy was chosen based on each patient’s anatomical position of the aortic valve and surgeons’ preference (upper sternotomy; n =24 [77.4%], lower sternotomy; n = 7 [22.6%]). The sternum was incised from the sternal notch down to the left or right fourth intercostal space during an upper sternotomy. However, in lower sternotomy, the sternum was incised from the left or right second intercostal space to the xiphoid. For CPB, the ascending aorta, subclavian artery, or femoral artery was used for arterial cannulation. The right atria or femoral vein was used for venous cannulation. The left ventricle was vented through a small cannula placed in the right upper pulmonary vein. Through the incision, the aortic cross-clamp was applied. Cold blood cardioplegia was administered for myocardial protection in an antegrade fashion into the aortic root and the coronary ostia approximately every 20 min.
Conventional approach
The conventional full sternotomy was performed. CPB was established with the ascending aorta, subclavian artery, or femoral artery for arterial cannulation and the right atria or femoral vein for venous cannulation. For myocardial protection, cold blood cardioplegia was administered into the aortic root in an antegrade fashion and the coronary ostia approximately every 20 min.
Statistical Analysis
Continuous data were expressed as the median (25–75 interquartile range). The Mann–Whitney U test was used to compare continuous data between groups in this study. Categorical data are expressed as frequencies (%) and evaluated by using the Chi-square test or Fisher’s exact test.
Propensity score methodology was adopted to reduce the confounding in the statistical comparison of surgical outcomes in two groups by accounting for differences in baseline patient characteristics, using a 1:1 nearest-neighbor-matching algorithm with a ±0.2 caliper and no replacement. It produced 22 (PAVR vs. CAVR) and 28 (STAVR vs. CAVR) pairs of propensity score-matched observations. The propensity score was calculated using a multivariate logistic regression model with an indication for the selection of patients for the two groups, using the preoperative variables. The preoperative patient variables considered clinically relevant were included as explanatory covariates, namely age, comorbidities (hypertension, hyperlipidemia, diabetes mellitus, CKD, peripheral arterial disease, ischemic heart disease, atrial fibrillation, chronic obstructive pulmonary disease, and old stroke), preoperative hemoglobin, preoperative LVEF, and NYHA class. The risk factors for prolonged CPB time > 2 h and extended hospital stay of more than 2 weeks in MIAVR (SHAVR and PAVR) groups were included in the univariate analysis, and any variable with a P-value ≤ 0.1 was incorporated in the multivariate logistic regression model. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan) or Prism version 8.0 (GraphPad Software, San Diego, CA, USA). A P-value < 0.05 was considered statistically significant.