Discussion
In recent years, significant improvements have been made in minimally invasive aortic valve surgery, including direct right mini-incision[4] and thoracoscopic-assisted right mini-incision[5], and robotic-assisted aortic valve surgery[6], all of which are characterized by less trauma, less bleeding, and better recovery compared with conventional aortic valve surgery[7]. With similar repair results, total thoracoscopic AVR can also be used for complex valvular lesions, with the advantage of minimally invasive valve surgery. The upper mini-sternotomy group had significantly reduced 24h chest drainage, reduced pain, decreased ICU stay and hospital stay when compared to conventional AVR, which was consistent with the previous findings of Totaro P et al.[8] and Paparella D et al[4]. Furthermore, total thoracoscopic AVR had more favorable perioperative assessment data compared to the upper mini-sternotomy group, in except for a prolonged CPB as well as AC time. However, a study by Tokoro M et al. indicated that all-thoracoscopic AVR had similar CPB and AC times compared to conventional AVR[9]. Cresce GD et al. showed that the type of valve may have a significant effect on CPB and AC time, and we were failed to perform a subgroup analysis because of the insufficient number of procedures[10]. In addition, since Vola et al. first reported total thoracoscopic AVR in 2014, only a few centers have performed further studies on this procedure. We think that factors including the anatomical characteristics of the patient’s aorta, the insufficient number of aortic valve procedures performed by the surgeon, and the insufficient exposure of the surgical field by the assistant during the procedure limit the development of total thoracoscopic AVR procedures.
The first is patient selection based on the anatomical characteristics of the aortic valve structure, in our experience, a wider sinotubular junction region may be more favorable for thoracoscopic exploration, and a narrower junction region may result in poorer surgical field of view exposure. The severity of calcification of the diseased aortic valve may also render total thoracoscopic AVR more difficult, leading to slightly unfavorable surgical outcomes. In addition, the height of the aortic valve position may be another factor affecting the exposure of the surgical field. In contrast, the conventional AVR, upper mini-sternotomy AVR, is less demanding on the anatomical characteristics of the aortic valve and is generally applicable. Technical improvements in thoracoscopic devices may be a viable solution in the future.
The second is the limitation on the number of heart valve procedures the surgeon has completed. Based on the valve surgery learning curve assessment, after completing a sufficient number of right heart valve procedures, one can attempt mitral valve surgery and subsequently aortic valve surgery[11], which is also a process of re-familiarization with the cardiac anatomy. Our previous study identified the safety as well as reliability of total thoracoscopic mitral valve repair with superior early postoperative outcomes, with the exception of prolonged AC and CPB times[12]. Similarly, the increased AC and CPB times are reasonable due to the limitations of the aortic valve anatomic region as well as the sinotubular junction region. Our study identified that total thoracoscopic MVP provided equally satisfactory surgical results compared to conventional MVP and that stabilization could be achieved gradually after completion of the 75th procedure, which was similar to the results of previous learning curve studies of minimally invasive valve surgery. Although there were significant differences between different studies, the overall tendency of change in CUSUM was similar, with the number of total thoracoscopic MVPs required to overcome the learning curve ranging from 64 to 116[13-15]. However, we failed to perform an assessment of the learning curve of CUSUM (Cumulative Sum) based on total thoracoscopic AVR due to the insufficient number.
The third is the insufficient cooperation between the surgeon and the assistant, decreased cooperation will directly lead to inferior surgical field exposure, which in turn will affect suturing and knotting results.
It maybe noteworthy that one case of perivalvular leak was observed in the total thoracoscopic group at the 3-month follow-up, as the incidence of perivalvular leakage after AVR was less than 0.01 in our center, this may be the result of inferior suture angles and weak knotting.