Article Text

Original research
Determinants of clinical outcomes of surgery for isolated severe tricuspid regurgitation
  1. Sung Jun Park1,
  2. Jin Kyung Oh2,
  3. Seon-Ok Kim3,
  4. Seung-Ah Lee4,
  5. Ho Jin Kim1,
  6. Sahmin Lee4,
  7. Sung Ho Jung1,
  8. Jong-Min Song4,
  9. Suk Jung Choo1,
  10. Duk-Hyun Kang4,
  11. Cheol Hyun Chung1,
  12. Jae-Kwan Song4,
  13. Jae Won Lee1,
  14. Dae-Hee Kim4,
  15. Joon Bum Kim1
  1. 1 Department of Thoracic and Cardiovascular Surgery, Asan Medical Center, Seoul, The Republic of Korea
  2. 2 Cardiology, Chungnam National University Sejong Hospital, Sejong, The Republic of Korea
  3. 3 Department of Clinical Epidemiology and Biostatistics, Asan Medical Center, Songpa-gu, The Republic of Korea
  4. 4 Division of Cardiology, Asan Medical Center, Seoul, The Republic of Korea
  1. Correspondence to Dr Joon Bum Kim, Department of Thoracic and Cardiovascular Surgery, Asan Medical Center, Seoul 05505, Korea (the Republic of); jbkim1975{at}amc.seoul.kr; Dr Dae-Hee Kim, Division of Cardiology, Asan Medical Center, Seoul, The Republic of Korea; daehee74{at}amc.seoul.kr

Abstract

Objectives Although the incidence of patients with isolated tricuspid regurgitation (TR) is increasing, data regarding the clinical outcomes of isolated TR surgery are limited. This study sought to investigate the prognostic implications according to procedural types, and to identify preoperative predictors of clinical outcomes after isolated TR surgery.

Methods Among consecutive 2610 patients receiving tricuspid valve (TV) procedure, we analysed 238 patients (age, 59.6 years; 143 females) who underwent stand-alone TV surgery (repair, 132; replacement, 106) for severe TR. Primary outcome was the composite of all-cause mortality and heart transplantation. Clinical outcomes between the repair and the replacement groups were compared after adjusting with the inverse probability of treatment weighting (IPTW) method.

Results During follow-up (median, 4.1 years), 53 patients died and 4 received heart transplantation. Multivariable analysis revealed that age (p=0.001), haemoglobin level (p=0.003), total bilirubin (p=0.040), TR jet area (p=0.005) and right atrial (RA) pressure (p=0.022) were independent predictors of the primary outcome. After IPTW adjustment, there were no significant intergroup differences in the risk of primary outcome (HR 1.01; 95% CI 0.55 to 1.87). In the subgroup analysis, tricuspid annular diameter was identified as a significant effect modifier (p=0.012) in the comparison between repair versus replacement, showing a trend favouring replacement in patients with annular diameter >44 mm.

Conclusions The outcomes of stand-alone severe TR surgery were independently associated with the severity of TR and RA pressure. In selected patients with severe annular dilation >44 mm, replacement may become a feasible option.

  • tricuspid valve disease
  • valve disease surgery

Statistics from Altmetric.com

Introduction

Despite renowned high operative risks as high as 10% in early mortality,1–3 large-scale registry data have shown steady increments in the volume of isolated tricuspid valve (TV) surgery, but without improvements in early outcomes.4 Isolated TR has only recently begun to be recognised for its clinical significance,5 and there are limited data regarding the long-term clinical outcomes following isolated TR surgery. The lack of sufficient evidence to base the optimal timing of surgery likely leads to delayed referral to surgery until the developments of overt right-heart failure or end-organ insults, which might have adversely contributed to the discouraging early postoperative outcomes.6 Unlike left-sided valve diseases, isolated functional tricuspid regurgitation (TR) are often the result, not the cause of lung or right heart disease, and the prognosis is related to the underlying lung or right heart disease. Previous studies have suggested several laboratory markers as prognostic factors following stand-alone TV surgery.2 7 However, comprehensive analysis incorporating quantitative echocardiographic measures has been challenged by limitations in its reproducibility because of the complex geometry of the right ventricle (RV).

In TV surgery, valve repair is generally preferred over valve replacement; however, TV replacement may be a more favourable option over repair in the presence of challenging TV pathologies such as severe annular dilation or significant leaflet defects.6 Nonetheless, published data are limited to refer in the selection of procedural type (repair vs replacement) conditional to underlying anatomy. Therefore, clinical studies involving a reasonably sized cohort with comprehensive quantification of the right cardiac chambers may offer valuable information to guide the decision-making process in isolated TR surgery. In the present study, we sought to identify the determinants for adverse outcomes following isolated severe TR surgery and to determine the impact of procedural type on clinical outcomes. As part of study aims, we explored parameters which can guide a selection of the type of surgery.

Methods

Study population and endpoint

We reviewed the medical records of adult (aged ≥18 years) patients who received TV repair or replacement for isolated severe TR between January 1996 and June 2018 at Asan Medical Center (Seoul, Korea). Diagnosis of severe TR required all of the following criteria: (1) distal jet area >10 cm2; (2) vena contracta width >0.7 cm and (3) proximal isovelocity surface area (PISA) radius >0.9 mm. The exclusion criteria were as follows: (1) underwent concomitant procedures for the aorta, coronary arterial bypass grafting or heart valves other than the tricuspid valve; (2) constrictive pericarditis or structural heart defects; (3) primary intention of surgery unrelated to TV procedures (eg, myxoma, pulmonary thromboembolism) or (4) underwent TV surgery for infective endocarditis or primary TV anomalies.

The primary outcome was the composite of all-cause mortality and heart transplantation. The secondary outcome was the composite of all-cause mortality, heart transplantation, readmission due to congestive heart failure (CHF) or tricuspid reoperation. Early death was defined as in-hospital or 30 days mortality. The follow-up data were obtained through 31 October 2018. Vital status was based on the medical records and confirmed by accessing the Korean National Registry of Vital Statistics.8

Echocardiographic assessment

Baseline echocardiography was performed in all patients within 2 months prior to the index surgery. All echocardiographic images were retrospectively reappraised by two experienced imaging cardiologists. Semi-quantitative TR assessment was done using jet area, vena contracta width and proximal convergence method,9 10 and integrative qualitative grading was determined by echocardiography experts according to the American Society of Echocardiography guidelines criteria.11 Right atrial (RA) pressure was estimated by the inferior vena cava diameter and its response to inspiration,12 13 and subdivided into three categories. The tricuspid annular diameter was measured on the RV-focused view as the maximal distance between the insertion hinge points of the septal and anterior tricuspid leaflets14 (online supplemental figure 1). RV dimensions were estimated at end-diastole, and the RV cavity areas were calculated by tracing endocardial border from the lateral tricuspid annulus to medial tricuspid annulus.13 RV systolic function was assessed by the RV fractional area change (FAC).

Supplemental material

Statistical analysis

Categorical variables were compared using the χ2 test or Fisher’s exact test and are presented as frequencies and percentages. Continuous variables, expressed as mean±SD, were compared using the Student’s t-test. Kaplan-Meier analyses were used to assess the conditional probability of the primary and secondary outcomes, and log-rank tests were used to assess the intergroup differences. The Cox proportional hazard model was used to identify the prognostic factors following surgery, and the proportional hazards assumption was assessed using the Schoenfeld residual, which yielded no evidence to suggest rejecting the assumption about the long-term outcomes. To reduce the potential treatment-selection bias, inverse probability of treatment weighting (IPTW) method was used. The adjustment with IPTW was performed based on the trimmed stabilised weight with robust SEs, the baseline profiles were well-balanced with most of the covariates having standardised mean differences of <10%. To examine the effect of repair versus replacement in different subsets of patients according to echocardiographic findings, subgroup analysis was conducted incorporating multiple variables (age, haemoglobin, total bilirubin, TR jet area and RA pressure), which were significant in the multivariable model. Adjusted hazards were calculated using Cox proportional hazard models. In the IPTW-adjusted cohort, subgroup analysis was also conducted. To assess the annular diameter-dependent effects of surgical mode on the primary outcome, a Cox proportional hazard model was fit with the use of an interaction term for the annular diameter and procedural types. Linear, quadratic, natural spline and restricted cubic spline models with the number of knots (3, 4 or 5 knots) function were considered and compared based on the Akaike information criterion. The linear model was chosen as the most suitable. Missing baseline values were treated through multiple imputation using the Markov Chain Monte Carlo method. Parameter estimates (HR) were computed by pooled estimate from 10 multiple imputed datasets in the multivariable analysis, whereas the propensity score was estimated as the mean of the predicted probabilities of 10 multiple logistic regressions. All reported p values are two-sided, and those <0.05 were considered to indicate statistical significance. R software V.3.4.0 (R Foundation, Vienna, Austria) was used for statistical analysis.

Patient and public involvement

The current study was written without patient or public involvement. Patients were not involved in the study design, data analysis or writing the manuscript of the current study. Patients were not invited to contribute to the writing or editing of this document for readability or accuracy.

Results

Patients

A total of 2610 adult patients that underwent TV procedures between January 1996 and June 2018 were identified from the cardiac surgical database of Asan Medical Center. After applying the exclusion criteria, 238 patients (age, 59.6±12.4 years; 143 females) who underwent TV surgery by TV repair (n=132) or TV replacement (n=106) were included (online supplemental figure 2).

The baseline demographic, clinical and operative profiles of the patients are shown in table 1. A total of 147 patients (61.8%) had atrial fibrillation and 94 patients (39.5%) had prior cardiac surgery. Of cases with severe TR, 78.2% were caused by functional component and 17.6% were associated with prolapse or flail motion of leaflets. Annuloplasty was performed in 85.6% of patients in the repair group, while mechanical prosthesis was used in 78.3% in the replacement group.

Table 1

Baseline demographic, echocardiographic and laboratory profiles

Clinical outcomes

Postoperative outcomes are summarised in table 2. Early death occurred in 11 patients (4.6%) and there was no significant difference between the repair group (5.3%) and the replacement group (3.8%; p=0.78). During the median follow-up duration of 49 months, there were 42 (17.6%) additional cases of mortality and 4 patients (1.7%) underwent heart transplantation (figure 1A, online supplemental table 1). A total of 31 patients required rehospitalisation for CHF, and 8 underwent tricuspid reoperation (figure 1B, online supplemental table 1).

Table 2

Early outcomes

Figure 1

Clinical outcomes. (A) The primary outcome and (B) the secondary outcome during the overall study period. CHF, congestive heart failure.

Risk factors for adverse outcomes following surgery

The final multivariable model identified that older age (HR, 1.05; 95% CI, 1.02 to 1.08; p=0.001), lower blood haemoglobin level (HR, 0.83, 95% CI, 0.73 to 0.94; p=0.003), higher total bilirubin level (HR, 1.50; 95% CI, 1.02 to 2.20; p=0.040), larger TR jet area (HR, 1.03; 95% CI, 1.01 to 1.06; p=0.005) and elevated RA pressure (HR, 1.14; 95% CI, 1.02 to 1.25; p=0.022) were significant predictors for the primary composite outcome (table 3). Figure 2 shows the primary composite outcomes stratified by the TR jet area and the RA pressure. Among all the study subjects with a colour Doppler jet area of >10 cm2, the patients with a jet area of >30 cm2 showed worst long-term outcomes, while the those with a jet area of <20 cm2 showed most favourable outcomes. Long-term outcomes also differed when the RA pressure was stratified by the pressure of 15 mm Hg.

Figure 2

The primary outcomes stratified by (A) tricuspid regurgitant (TR) jet area and (B) right atrial (RA) pressure.

Table 3

Univariable and multivariable analyses for the primary outcome

The detailed results of the univariable and multivariable analyses for the secondary outcome are provided in online supplemental table 2. Similar variables were identified as prognostic factors for the secondary outcome.

Repair versus replacement

After adjustment with IPTW, the baseline profiles were well-balanced between the repair and replacement groups (online supplemental table 3). There were no significant differences in the risk of primary composite outcome between the two groups in both the original cohort (5-year heart transplantation-free survival rate, 79.9% vs 78.4%; p=0.75) and the IPTW-adjusted cohort (HR, 1.01; 95% CI, 0.55 to 1.87; p=0.97) (figure 3A,B). The risk of secondary outcome was also similar in the original (p=0.5) and the IPTW-adjusted cohort (HR, 1.23; 95% CI, 0.72 to 2.10; p=0.45) (figure 3C,D).

Figure 3

Repair versus replacement. (A) The primary outcome in the original cohort and (B) the IPTW-adjusted cohort. (C) The secondary outcome in the original cohort and (D) the IPTW-adjusted cohort. IPTW, inverse probability of treatment weighting.

Subgroup analysis

In the subgroup analyses adjusted by multiple variables, which were identified as significant determinants for the primary outcome in the final multivariable model, the tricuspid annular diameter was identified as a significant effect modifier (p=0.008) in the comparison between repair and replacement (figure 4A). In the subgroup of patients with tricuspid annular diameters >44 mm, replacement was superior to repair in terms of the risk for primary composite outcome (HR, 7.52; 95% CI, 1.96 to 28.78; p=0.003). When the tricuspid annular-diameter was examined as a continuous variable, there was a trend favouring TV replacement as compared with repair, as the tricuspid annular diameter increases (figure 5A). Similar trend was also exhibited among the IPTW-adjusted cohort, and the cut-off point was obtained between 43 and 44 mm of tricuspid annular diameter (figure 5B).

Figure 4

Adjusted hazard of primary outcome with repair compared with replacement according to echocardiographic findings. (A) Adjusted by variable which were significant in the multivariable analysis and (B) IPTW-adjusted. TVP, tricuspid valvuloplasty; TVR, tricuspid valve replacement; TR, tricuspid regurgitation; RA, right atrial; ITPW, inverse probability of treatment weighting.

Figure 5

Adjusted tricuspid annular diameter-dependent hazard of primary outcome with repair compared with replacement.

The tricuspid annular diameter was also identified as a significant effect modifier (p=0.012), in the subgroup analyses of IPTW-adjusted cohort (figure 4B). In the subgroup of patients with tricuspid annular diameters >44 mm, the risk for primary outcome tended to favour replacement over repair (HR, 2.51; 95% CI, 0.91 to 6.91; p=0.075); for those with diameters of 44 mm or less, there was a trend of favouring repair over replacement (HR, 0.44; 95% CI, 0.16 to 1.20; p=0.11).

Discussion

The present study is one of the largest observational studies to date regarding the treatment of isolated severe TR. The present study has two main findings. First, the long-term outcomes following surgery for isolated severe TR were significantly affected by the preoperative severity of TR as measured by jet area and the RA chamber pressure. Second, although there was no significant difference in the overall survival between TV repair and replacement, TV replacement showed more favourable clinical outcomes compared with TV repair when the TV annular diameter was >44 mm.

The present study intended to exclusively enrol patients who received TV surgery primarily for the treatment of isolated severe TR. We excluded the patients who underwent concomitant aortic or left-sided cardiac surgery, and those who had congenital heart defects in whom the primary goal of the surgery was the repair of the aforementioned defects. Furthermore, patients with infective endocarditis or congenital TV anomalies were excluded because surgery may have been regarded as the only therapeutic option for them. In the management of patients with isolated severe TR, the benefits of corrective surgery remain uncertain; also, if surgery is unavoidable, decisions regarding the timing of surgery or procedural type (repair vs replacement) are made based on the individual physician’s experience rather than empirical evidence. Such issues are more apparent in functional TR without marked structural lesions in the TV. By employing such rigorous patient selection criteria, we believe that we have retrieved a homogenous study cohort that replicates specific clinical situations in which indefinite and difficult decisions need to be made.

In the present study, the preoperative TR jet area was identified as a significant prognostic factor following isolated TV surgery. Of note, even when all study patients had severe TR, long-term survival showed significant differences according to the TR jet area. The TR jet area alone has only limited diagnostic value in evaluating the TR severity due to the inaccuracy in eccentric or wall impinging regurgitant jets and significant overlaps between mild and moderate TR.11 Nonetheless, considering that a colour Doppler jet area of >10 cm2 is generally consistent with severe TR and the majority of the study subject were associated with functional TR without eccentric jet flow,11 these findings may give strength to the need for the extended severity scale of TR beyond severe (ie, very severe), which had been proposed by Hahn et al previously.15 16 However, our results should be interpreted with caution and appraised by further studies involving other quantitative measures such as effective regurgitant orifice area because haemodynamic factors such as preload can affect the appearance of the central jet of the TR.17 18

Initial experiences with isolated TV replacement reported poor operative mortality rates ranging from 15% to 50%19–21; however, those results may have been confounded by the greater number of comorbidities, and recent comparative studies showed that the operative mortality rates were similar between repair and replacement when the imbalances in comorbidities were adjusted.3 6 Very recently, Axtell et al 22 reported that TV surgery does not improve long-term survival compared with conservative treatment, because referral to isolated TR surgery was delayed until the developments of overt right-heart failure or end-organ damage, which might have adversely contributed to the discouraging postoperative outcomes. Likewise, the present study showed that early mortality rates were not significantly different between the repair group and the replacement group (5.3% vs 3.8%; p=0.78). Beyond the early operative results, these two groups also showed comparable long-term results in terms of the composite outcome of all-cause death and heart transplantation as well as the composite of death, readmission due to CHF and tricuspid reoperation. The cross-over of the Kaplan-Meier estimates of these two groups at 5 years may be partly explained by the prosthetic valve-related complications.

A recent meta-analysis by Wang et al 23 showed that repair surgery was significantly reduced in-hospital mortality compared with replacement. Although, in our study, there were no significant differences in the risk of primary composite outcome between the two groups. This may come from the underpowered of analysis for less frequent events, subgroup analysis showed that the maximal annular diameter of TV was as a significant effect modifier in the comparison between TV repair and replacement. The adjusted outcomes were superior in the replacement group when the preoperative tricuspid annular diameter was >44 mm in the multivariable analysis, although without statistical significance in the IPTW-adjusted cohort. Given that similar results consistently maintained irrespective of the statistical methods, it suggests that replacement might be more beneficial than repair in patients with severe annular dilatation who are deemed to be unamenable to repair. Functional TR is regarded as a progressive disease with ongoing dilatation of the annulus, as an expression of congestive heart failure.24 25 With an absence of anatomic fibrous annulus, the TV is highly susceptible to annular dilatation in a continuum of right ventricular dilatation.25 Remodelling annuloplasty has been the dominant treatment for functional TR, which is associated with left-sided valvular lesions. Concomitant tricuspid annuloplasty can be performed without additional risks during left-sided surgery to prevent further dilatation of tricuspid annulus.6 However, patients undergoing stand-alone tricuspid surgery are distinct from those receiving concomitant tricuspid procedures during left-sided surgery. Many of those who are referred to stand-alone TV surgery may have more severe TR and severely dilated tricuspid annulus, and a large proportion of this population has a history of left-sided valve surgery. In patients undergoing stand-alone TV surgery, regardless of repairability, surgery is expected to resolve the right heart failure and the associated symptoms through an effective and durable correction of severe TR. In advanced stages of TR associated with severe annular dilatation, TR can increase over time postoperatively even after initially successful repair.6 Incomplete and unsustainable results of repair may partly explain the modifier effect of the maximal annular diameter in the comparison between TV repair and replacement.

Limitations

This study has several limitations. First, this study is a single-centre, retrospective observational study so that the lack of randomisation may lead to selection bias. The IPTW analysis to overcome the potential selection bias and baseline differences between the groups provided a well-balanced study population. Although it is generally superior to the covariate adjustment method, the inherent problems originated form the retrospective analysis cannot be eliminated. Second, our study had an extremely long treatment period (22 years), there have been changes in TV surgical techniques as well as concomitant treatment over the past years, which may affect postoperative clinical outcomes. Third, quantification of regurgitant flow, degree of TV tethering and tricuspid annular diameter are important to evaluate the severity of valve regurgitation, but evaluation of these parameters had the inherent limitations of technical or measurement errors. Among the echocardiographic parameters, there were missing values in the RV FAC and tricuspid annular diameter. The missing data were handled with multiple imputations using the Markov Chain Monte Carlo method, but the results after multiple imputations may still be biased. Furthermore, postoperative echocardiography parameters could not be assessed. Therefore, important additional factors such as significant postoperative TR or early postoperative RV dysfunction, were not comprehensively integrated into this risk model.

Conclusions

The outcomes of isolated severe TR surgery were significantly associated with the severity of TR as measured by jet area and the RA chamber pressure. These parameters may be useful in predicting the prognosis of isolated severe TR and guiding the decision-making process in its surgical management. In selected patients with enlarged annulus who are considered difficult to repair, valve replacement may become a feasible option. Although our findings may provide additional clinical evidence into the patents undergoing TV surgery, overall findings should be confirmed or refuted through further clinical investigations.

Key questions

What is already known on this subject?

  • Isolated tricuspid regurgitation (TR) recently begun to be recognised for its clinical significance, and there are limited data regarding the long-term clinical outcomes following surgery.

What might this study add?

  • The present study is one of the largest observational studies to date regarding the treatment of isolated severe TR.

  • The long-term outcomes of stand-alone severe TR surgery were independently associated with the preoperative severity of TR as measured by jet area and the right atrial chamber pressure.

  • TV replacement showed more favourable clinical outcomes compared with TV repair when the TV annular diameter was >44 mm.

How might this impact on clinical practice?

  • Even when all study patients had severe TR, long-term clinical outcomes were differed significantly according to the severity of TR as stratified by jet area.

  • In stand-alone tricuspid surgery, repair versus replacement should be selected in thorough consideration of degree of tricuspid annular dilation.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • SJP and JKO contributed equally.

  • D-HK and JBK contributed equally.

  • Correction notice This article has been corrected since it was published Online First. Dae-Hee Kim has been added as a corresponding author and the equally contribution statement has been linked to authors Dae-Hee Kim and Joon Bum Kim.

  • Contributors Study concept and design: SJP, JKO, D-HK, JBK. Data acquisition and outcome measure: S-AL, HJK, SL, S-HJ, J-MS, SJC, D-HK, CHC. Data analysis and interpretation: SOK. Manuscript drafting: JKO. Critical review and revision: SJP, JKO, D-HK, JBK, J-KS, JWL. Approval of final version: D-HK, JBK.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval The study protocol was approved by the Asan Medical Center Institutional Review Board (2020–0718) and all patients provided written informed consent. Research was performed in accordance with the Declaration of Helsinki.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on reasonable request.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles