Propensity score analysis of outcomes following minimal access versus conventional aortic valve replacement

OBJECTIVES: Minimal access aortic valve replacement has become routine in many institutions. Aim of this study was to compare the clinical outcomes between conventional and minimal access aortic valve replacement. METHODS: We retrospectively analysed the data of 2103 patients who underwent primary, isolated aortic valve replacement (AVR) in our institution between January 2001 and May 2012 with a minimal access AVR (MAAVR) via the upper partial ministernotomy approach (n = 936) or conventional AVR (CAVR) via the full sternotomy approach (n = 1167). After propensity score matching considering potential confounders [age, sex (female), weight, height, preoperative serum creatinine level, previous myocardial infarction, LV-EF and aortic valve pathology (isolated AS)], 585 matched patients were included in each group. RESULTS: Mean age (65 ± 10.5 vs 65.7 ± 11.5 years, P = 0.23), gender (females 37.2%, P = 0.9), aortic cross-clamp time (65.6 ± 18.4 vs 64.3 ± 19.8 min, P = 0.25) and postoperative blood loss [median (IQR) 400 (224-683) vs 400 (250-610) ml, P = 0.83) were similar in MAAVR and CAVR group. Thirty-day mortality was also not significantly different (1.5 vs 1.7%, P = 0.74, respectively). In contrast, CPB times were significantly longer in MAAVR (93.5 ± 25 vs 88 ± 28 min, P < 0.001). Intraoperative and postoperative autologous blood transfusions were significantly lower in MAAVR (927.2 ± 425.6 vs 1036.4 ± 599.6 ml, P < 0.001 and 170.2 ± 47.6 vs 243.5 ± 89.3 ml, P < 0.001, respectively). Intubation time was significantly shorter in MAAVR [median (IQR) 7 (5-11) vs 8 (6-14) h, P = 0.01). The incidence of renal insufficiency (creatinine ≥1.5 mg/dl) and respiratory insufficiency (need for non-invasive ventilation, reintubation or tracheotomy) was significantly lower in MAAVR (9 vs 16%, P < 0.001 and 8.5 vs 11.8%, P = 0.03, respectively). CONCLUSIONS: In comparison with CAVR, our study shows that MAAVR is a safe and effective procedure associated with low mortality rate and good long-term survival rates. In addition to that, MAAVR was associated with shorter ventilation times, lower rate of autologous blood transfusion, as well as a lower rate of postoperative respiratory and renal insufficiency. Because of the superior cosmetic results, we therefore advocate MAAVR as the procedure of choice for primary isolated AVR.