Discussion
In this study, we introduced our singled-center experience of isolated tricuspid valve surgery in the past 20 years. We focused only on patients undergoing RITS after previous left-sided valve surgery, because we believe that this is high risk surgical population. Outcomes of this population might differ from those of patients undergoing first-time isolated tricuspid valve surgery and other cardiac procedures concomitant with redo tricuspid valve surgery. We reported early and long-term outcomes and evolving strategies applied in this population. The evolving strategies referred to innovated surgery techniques: early surgery if patients were considered candidates of surgery, totally endoscopic approach, and preferred TVr with leaflets augmentation when leaflet tethering was severe. The outcomes were encouraging, and both mortality and morbidity decreased since 2015.
The volume of isolated tricuspid valve surgery has significant increased while still remaining relatively rare in comparison to 2.4 million residents with moderate to severe TR [1, 14]. During the study period, more than 15 000 tricuspid valve surgeries were performed at our institute while RITS after LSVS accounted for only 1.1%, and the procedures were rarely performed due to the relatively high in-hospital mortality ranging from 6% to 21% reported in previous reports [15-17]. However, we adopted a more active attitude towards surgery in the recent 5 years for the following reasons: 1) based on our clinical experiences, we found most patients sought surgery due to recurrent symptoms of right heart dysfunction with long-term medical treatment history and diuretics possibly being ineffective; 2) the totally endoscopic access combined with leaflet augmentation, simplified unicaval drainage without snare and beating-heart surgery techniques reduced ICU stay time and reoperation rate; and 3) we found TVr was effective and safe in patients with severe leaflet tethering [11].
Early studies introduced minimally invasive tricuspid valve surgery and reported lower mortality and morbidity; our evolving surgical techniques were different from those studies [10, 18]. More studies reported RITS after LSVS with a higher proportion of TVR due to leaflet structural alterations or excessive leaflet tethering in the context of a long-lasting disease when repair was difficult [10, 16, 18]. We prefer TVr, because TVR carries a higher risk of mortality and morbidity. Zack et al. reported their investigation of the nationwide trend of isolated tricuspid valve surgery and found higher in-hospital mortality for valve replacement than for repair, with a risk-adjusted odd ratio of 2.2 [14]. The present study also indicated that TVR was an independent risk factor with odds ratio (OR) of 6.778 (95% CI: 1.370–33.549,p = 0.019). The common issue [19] that tricuspid valve without coaptation and extensive annular dilation was no amenable to repair may persuade a surgeon to replace, while we believe that the augmentation of both anterior and posterior leaflet might address this issue by minimizing the leaflet tethering and increasing the coaptation. Therefore, we introduced the updated technique of leaflet patch augmentation of both anterior and posterior leaflets by sewing the patch to the leaflets and the annulus to convert the native leaflets into the coaptation zone and (partially) the chordae [13]. Other repair techniques such as ring implantation, artificial chordae, commissurotomy and papillary muscle incision were also performed as needed, and therefore the overall CPB time was long because of these combined repair techniques. Some surgeons might also be concerned about the increased rate of residual or recurrent TR and preferred TVR, our data showed acceptable results with 5.5% residual severe TR. These five patients with TVr had large preoperative jet area of TR with a median of 26.0 (22.0–31.5) cm2 that decreased to 13 (11.5–14.0) cm2 after surgery (p = 0.008).
The current controversies surrounding redo isolated tricuspid valve surgery are concentrated in the following aspects: 1) is surgery necessary and does surgery improve long-term outcomes of this population? 2) which surgical correction is better, repair or replacement? 3) when is the best time to consider surgery? Andrea et al. reported on 3276 patients from 2001 to 2016 and this large data set indicated that surgery did not improve survival of this population using a propensity-matched method [20]. Although patients with medical treatment were not included in the present series, our data indicated that surgical correction had excellent outcomes with 10-year survival rate of 59.2% (95% CI: 43.5–75.5%). Patients with repair had better early outcomes while there was no significant difference in the long-term survival. Another key finding was the increased mortality with higher NYHA function class. Patients with NYHA function class IV had relatively higher risk of in-hospital death with an OR of 8.525 (95% CI: 2.153–33.760,p = 0.002), and NYHA function class IV as an independent risk factor was frequently confirmed by previous studies [17, 18, 21, 22]. Considering the high risk of higher NYHA function class, we adopted early surgery strategies and preoperative optimized medical therapy are recommended to improve heart function.