Isolated tricuspid valve surgery: first outcomes report according to a novel clinical and functional staging of tricuspid regurgitation

Abstract

 

Open in new tabDownload slide

OBJECTIVES

The goal of this study was to assess the applicability of a novel classification of patients with tricuspid regurgitation based on 5 stages and to evaluate outcomes following isolated surgical treatment.

METHODS

All patients treated with isolated tricuspid valve repair or tricuspid valve replacement (TVR) from March 1997 to January 2020 at a single institution were retrospectively reviewed. Patients were divided according to a novel clinical-functional classification, based on the degree of regurgitation together with symptoms, right ventricular size and function and medical therapy. A total of 195 patients were treated; however, 23/195 were excluded due to lack of sufficient preoperative data.

RESULTS

A total of 172 patients were considered; of these, 129 (75%) underwent TVR and 43 (25%) had tricuspid valve repair. The distribution of patients showed that 46.5% of patients who underwent tricuspid valve repair were in stage 2, whereas 51.9% who underwent TVR were in stage 3. TVR patients were in more advanced stages of the disease, with dilated right ventricles, more pronounced symptoms and development of organ damage. Hospital mortality was 5.8%, in particular 0% in stages 2 and 3 and 15.3% in stages 4 and 5 (P < 0.001). Both intensive care unit and hospital stays were significantly longer in more advanced stages (P < 0.001). Patients in stages 4 and 5 developed more postoperative complications, such as acute kidney injury (3.7–10% in stages 2 and 3 vs 44–100% in stages 4 and 5; P < 0.001) and low cardiac output syndrome (15–50% in stages 2 and 3 vs 71–100% in stages 4 and 5; P < 0.001).

CONCLUSIONS

Patients in more advanced stages had higher hospital mortality and longer hospitalizations. Timely referral is associated with lower mortality, short postoperative course and mostly valve repair.

INTRODUCTION

Tricuspid valve (TV) pathology has been disregarded for a long time due to the firm belief that treatment of left-sided heart disease would lead to resolution of tricuspid insufficiency [1]. Therefore, conservative treatment has always been the gold standard. However, with the advent of new transcatheter interventions, more attention has been paid to the TV [2]. In fact, due to increasing interest in TV disease, surgical and percutaneous treatments have recently been in the spotlight.

A growing number of studies have been published in the literature reporting high morbidity and mortality for the surgical treatment of isolated tricuspid regurgitation (TR) [3–5]. Poor outcome of tricuspid valve repair (TVr) or tricuspid valve replacement (TVR) is mainly related to a delayed surgical approach usually associated with dilation and/or dysfunction of the right ventricle [6]. Patients are often managed pharmacologically, mainly with diuretics, for a long period and referred for surgery only after development of hepatorenal dysfunction and refractoriness to medical therapy. At this point, patients have prominent right ventricular (RV) dilation and dysfunction, resulting in complicated and poor postoperative response, further strengthening the belief of a high-risk procedure, with high early and late mortality [4, 7, 8]. This vicious circle should be interrupted early on in the progression of TV disease [9].

A novel clinical and functional staging of TV regurgitation has recently been proposed [10, 11]. The progression of morphological changes to the TV, annulus and right ventricle has been described and associated with the onset of symptoms. The classification proposed by Latib et al. [10] and mentioned by Taramasso et al. [11] identifies a number of parameters and factors that may be useful in better stratifying surgical risk, such as TR grade, symptoms, RV remodelling and function and medical therapy.

The goal of our study was to understand the distribution of patients who had isolated TV surgery among the 5 stages of this classification and to assess their in-hospital outcomes accordingly.

METHODS

We conducted a retrospective, single-centre study that included 195 patients affected by TV regurgitation and treated with isolated TV surgery from March 1997 to January 2020 at San Raffaele University Hospital, Milan, Italy. All consecutive patients were individually reviewed. The ethical committee of our institution approved the study and waived individual informed consent for this retrospective analysis. The most relevant baseline factors for classification were analysed specifically. All patients had undergone both preoperative transthoracic echocardiography and transoesophageal echocardiography (TOE) upon hospitalization. TOE was routinely used to better define the mechanism of TR. Within the study period, and as a common practice within our institution, patients were admitted to the hospital 48–72 h before surgery and were adequately studied within this time frame. Echocardiographic parameters together with all preoperative characteristics and data were entered into a dedicated database. Symptoms were graded, ranging from none to overt RV failure and organ damage, on the basis of the New York Heart Association functional class, together with the presence at presentation of peripheral oedema, ascites, repeated previous episodes of hospitalization for RV failure and altered laboratory values specific for hepatorenal function. TR was graded as 1+ (mild), 2+ (moderate), 3+ (moderate-to-severe) and 4+ (severe). In the more recent period of the study, this grading involved a multiparametric approach based on the current European Association of Echocardiography recommendations [12–14]. Overall, 55/195 (28.2%) patients were evaluated with newer and more extensive echocardiographic approaches. RV remodelling and function were defined based on preoperative transthoracic echocardiography and TOE values, such as RV end-diastolic diameter, tricuspid annular plane systolic excursion, tissue Doppler imaging and systolic pulmonary artery pressure. Medical therapy was assessed based on daily diuretic use, dosage and combination therapies and was retrospectively judged as low-dose, moderate-to-high dose and the frequent need for intravenous diuretic use [15, 16]. Based on a combination of all factors, patients were divided according to the 5 stages of the classification (Fig. 1).

Figure 1:

Classification and determinants/characteristics for distribution of patients into the 5 stages. IV: intravenous; RHF: right heart failure; RV: right ventricle; TR: tricuspid regurgitation.

Open in new tabDownload slide

Surgery was performed using standard techniques including bicaval cannulation or peripheral venous cannulation based on the surgeon’s preference. TV surgery was performed either on beating heart (BH) or arrested heart, using a standard median sternotomy approach or a right anterior thoracotomy [17, 18].

All patients underwent TOE in the operating room after they were weaned from cardiopulmonary bypass and transthoracic echocardiography before discharge in order to assess biventricular function and the outcomes of either TVr or TVR. The primary end points of the study were in-hospital mortality, intensive care unit (ICU) postoperative length-of-stay (LOS) and hospital LOS. The secondary end points of the study were development of postoperative in-hospital complications and general assessment of the distribution of our patient population according to the 5 stages.

Statistical analyses

Statistical analyses were conducted using SPSS (IBM, Armonk, NY, USA) and Stata Software version 15 (Statacorp, LLC, College Station, TX, USA). Categorical variables were expressed as absolute number and percentages. Normal distribution of continuous variables was assessed with the Shapiro–Wilk test. Continuous variables normally distributed were expressed as mean ± standard deviation. Continuous variables not normally distributed were reported as median and interquartile range. Differences between stages for categorical variables were tested with the χ² test or the Fisher’s exact test, as appropriate. Differences between stages for continuous variables were tested with the analysis of variance or Kruskal–Wallis test, as appropriate.

A logistic regression model was employed to assess differences between stages for in-hospital mortality. A Poisson regression model or negative binomial regression model, whenever there was overdispersion, was used to assess differences between stages in terms of ICU LOS and hospital LOS.

The Dunnett test with the Bonferroni correction was run to perform multiple comparisons between stages. Univariate and multivariate analyses were performed to assess predictive factors for hospital mortality. A P-value of <0.05 was used to define statistical significance.

RESULTS

A total of 195 patients who underwent isolated TV surgery at our institution were reviewed and, based on preoperative characteristics, divided into the 5 stages. A total of 23/195 (12%) patients were excluded from the study due to lack of sufficient preoperative data, which prevented them from being adequately assigned to the correct stage (Fig. 2). None of our patients met the criteria for stage 1, because by definition the TR grade is to be considered less than moderate and therefore patients undergoing isolated surgical treatment had at least moderate-to-severe TR. Furthermore, an extremely limited number of patients treated were part of the end-stage disease group (stage 5). The distribution of patients in stages is shown in Table 1. A total of 172 patients were considered for inclusion in this study, among which 129 (75%) underwent isolated TVR and 43 (25%) underwent TVr. The main aetiologies were functional TR in 75 patients (43.6%), primary TR in 40 patients (23.3%, among whom 2 had carcinoid disease, 4 had congenital anomalies, 14 had rheumatic disease and 20 had degenerative involvement), endocarditis in 15 patients (8.7%), failure of previous TVrs in 14 cases (8.1%) and bioprosthesis degeneration in 4 patients (2.3%), a mix of aetiologies was seen in 11 patients (6.4%) and traumatic TR occurred in 10 patients (5.8%).

Figure 2:

Flow chart of the study population. preop: preoperative; TV: tricuspid valve; TVR: tricuspid valve replacement.

Open in new tabDownload slide

A BH operation was carried out in 110 patients (64%). A BH rather than arrested heart approach was chosen based on the surgeon’s preference. In general, a BH operation was preferred in patients with more advanced RV dysfunction to avoid any further myocardial damage potentially due to cardioplegic arrest and in patients undergoing TVR. On the other hand, patients with patent foramen ovale or interatrial septal defects were preferably treated with an arrested heart approach. No difference between surgical technique and in-hospital outcomes was noted in our patient population.

In case of TVR, a bioprosthesis was used in 123 patients (95.4%) and a mechanical prosthesis was preferred in the remaining 6 (4.6%). Among these, 4 patients occurred early within our experience. The remaining 2 patients were both young women with a history of extremely rapid bioprosthesis degeneration. Among the 43 patients submitted to repair for the different aetiologies reported, a ring annuloplasty was performed in 37 of them (86%), with or without concomitant leaflet repair (including the edge-to-edge or clover techniques), whereas in the remaining 6 no ring was used and the repair comprised suture annuloplasty and/or leaflet repair. Considering the overall population, the majority of patients were classified either in stage 3 (46.5%) or stage 4 (36%). However, when analysing separately patients who underwent either repair or replacement, 46.5% of patients who underwent TVr were in stage 2, whereas more than half of patients with TVR were in stage 3 (51.9%).

When considering the overall distribution, patients in stages 2 and 3 were younger and in less advanced stages of the disease (Table 1). In fact, patients presented with fewer signs and symptoms of RV failure, with ascites being almost absent and atrial fibrillation not being the main presenting rhythm; lesser organ involvement and development of hepatorenal syndrome were seen, as underlined by higher albumin levels and lower frequency of chronic kidney disease. Furthermore, patients in stages 4 and 5 were more frequently patients who had previously undergone cardiac operations (redo operations). Regarding echocardiographic parameters, patients in stages 2 and 3 had less dilated and less dysfunctional left ventricles, better RV function and lower systolic pulmonary artery pressure.

Moreover, TVr could be performed in most patients in stage 2.

Postoperative outcomes were significantly different between stages. More specifically, advancement from one stage to the next was correlated with higher mortality and development of postoperative complications. Indeed, mortality in the overall population was 5.8% (10/172 patients), but it was 0% in stages 2 and 3 and 15.3% in stages 4 and 5 (P < 0.001), and it was uniquely seen in patients who underwent TVR and who were in stage 4. At univariate analysis, chronic kidney disease and left ventricular ejection fraction emerged as predictors of hospital death, but at multivariate analysis, only left ventricular ejection fraction was identified as an independent predictor of this event (odds ratio 0.92, P = 0.015, confidence interval 0.85–0.98) (Table 2).

Patients in more advanced stages experienced more postoperative complications, such as acute kidney injury (3.7–10% in stages 2 and 3 vs 44–100% in stages 4 and 5; P < 0.001) and low cardiac output syndrome with the need of high doses of inotropic support (15–50% in stages 2 and 3 vs 71–100% in stages 4 and 5; P < 0.001). The incidence of a new postoperative permanent pacemaker overall was 10.4% (18/172), and all cases occurred after TVR (18/129, 13.9%) (Table 3). All implanted prostheses functioned adequately postoperatively (mean gradient among stages 3.6 ± 1.3 mmHg) and all TVrs had good outcomes, with the exception of 2 patients who were discharged with residual moderate (2+/4+) TR. No difference among stages was reported.

As shown in the Central Image, both ICU and hospital LOSs were significantly longer in patients in more advanced stages.

DISCUSSION

This retrospective, single-centre study evaluated the applicability of a novel clinical and functional classification of patients with TR proposed by Latib et al. [10, 11] to our isolated TV surgery series. To the best of our knowledge, this is the first study assessing the outcomes of isolated TVr or TVR according to this 5-stage preoperative classification. Overall, this classification reflects the population of patients affected by isolated TR that we have been treating surgically. In fact, in our series, around 40% of patients were in advanced stages 4 and 5: they showed more pronounced RV dilation and dysfunction and presented with important symptoms and organ involvement/damage. These characteristics resulted in more in-hospital deaths, a higher rate of postoperative complications and longer stays in the ICU and hospital. Furthermore, these patients were more frequently redo cases and were more likely to undergo valve replacement because of the important geometric deformation of the TV with severe tethering, in addition to annular dilation secondary to advanced RV remodelling. Our findings are in line with those from a great number of retrospective studies that have emphasized poor clinical outcomes in patients treated surgically for severe TR late after left-sided heart surgery [19, 20]. High in-hospital mortality and poor follow-up results seem to be related to the high-risk profile of the treated patients, who are frequently older, with advanced functional class and concomitant morbidities rather than to TV surgery per se [21].

In the current era, a great number of technologies and new devices have been investigated and proposed to treat patients who are considered too high risk for surgery [22]. However, few data are available to date, and none of the current technologies effectively solve this growing problem [23]. On the other hand, a timely surgical referral could dramatically change the surgical risk as well as the late outcome of this challenging population. The lack of data reporting positive outcomes following early TV surgery for isolated TR has contributed to the uncertainty as to when and how to treat this condition. Timing remains controversial. Our data further support current European Society of Cardiology/European Association of Cardio-thoracic Surgery guidelines recommending early referral for surgery in patients with isolated TV regurgitation, even asymptomatic, in case of progressive RV dilation or initial dysfunction, well before advanced RV remodelling/failure and organ damage [24, 25]. In our opinion, the 2014 American Heart Association/American College of Cardiology guidelines [23] are not able to change the poor outcomes associated with stages 4 and 5 surgical referrals. Indeed, they suggest, for instance, that reoperation for isolated TV disease may be considered (IIb) for persistent symptoms due to severe TR in patients who had previously undergone left-sided heart surgery. Such symptoms typically occur in the very advanced stages of the disease. Our data support the concept that the timing of surgical referral should be chosen by looking not mainly at the symptoms but essentially at the stage, which includes the degree of TR, RV size and function, geometry of the TV and the dose of diuretics. This observation is confirmed by the fact that stages 2 and 3 were associated with better outcomes compared to those of stages 4 and 5, but basically no individual predictors of mortality were identified at multivariate analysis, with the exception of left ventricular ejection fraction. We speculate that the stage is an expression of a number of parameters that are more important for risk stratification than the individual clinical or echocardiographic variables.

In our experience, particularly patients belonging to stage 2 had no hospital deaths and had a smooth, short postoperative course following TVr, which could be performed in the majority of cases. These findings are in line with the results of studies that have identified as determinants of good outcomes both left and RV function and pulmonary pressures, emphasizing the importance of timely referral before the development of end-stage cardiac impairment [26]. Therefore, early referral and intervention in patients with isolated TR are of utmost importance, as demonstrated by our surgical experience. Both TVr and TVR, when performed in the early stages of the disease, are feasible and safe, with a low rate of postoperative complications and few reported deaths. Repair should be pursued whenever possible, being associated with fewer deaths, a smoother postoperative course and no/low need of a pacemaker and, according to our series, its likelihood is higher in the early stages of the disease.

Whenever early interventions are not performed and patients in advanced stages are treated, due to the high risk of further exacerbating RV dysfunction and low cardiac output syndrome, a more protective perioperative management should be considered. Preoperative and/or postoperative support, with either inotropes or mechanical support devices, has been shown to rapidly result in haemodynamic improvement and contribute to a favourable survival [27, 28].

Limitations

Our study has limitations, mainly related to the fact that it is a retrospective single-centre study. Thorough assessment of preoperative baseline characteristics and echocardiographic parameters was completed to try and perform the most accurate division possible into stages. However, retrospective adjudication has its limitations, which need to be kept in mind. The sample size is relatively small due to the few isolated TV surgical procedures performed, and partly due to the lack of sufficiently complete data, especially systematic echocardiographic parameters. Furthermore, the fact that most patients were evaluated using older echocardiographic techniques to grade TR may have affected the division into stages. Another limitation is that we focused on the hospital outcomes of the patients. Further studies should address the late results at follow-up.

Presented at the 34th Annual Meeting of the European Association for Cardio-Thoracic Surgery, Barcelona, Spain, 8–10 October 2020.

ACKNOWLEDGEMENTS

This article was written with the support of the Alfieri Heart Foundation (data collection and management).

Conflict of interest: none declared.

Author contribution

Alessandra Sala: Writing—original draft. Roberto Lorusso: Supervision. Marta Bargagna: Data curation. Guido Ascione: Methodology. Stefania Ruggeri: Formal analysis. Roberta Meneghin: Data curation. Davide Schiavi: Data curation. Nicola Buzzatti: Data curation. Cinzia Trumello: Data curation. Fabrizio Monaco: Supervision. Eustachio Agricola: Supervision. Ottavio Alfieri: Supervision. Alessandro Castiglioni: Supervision. Michele De Bonis: Writing—review & editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Emilio Monguió, George M. Palatianos and the other, anonymous reviewer(s) for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

 
• BH

  Beating heart

 
• ICU

  Intensive care unit

 
• LOS

  Length-of-stay

 
• RV

  Right ventricle

 
• TOE

  Transoesophageal echocardiography

 
• TR

  Tricuspid regurgitation

 
• TV

  Tricuspid valve

 
• TVr

  Tricuspid valve repair

 
• TVR

  Tricuspid valve replacement