Study population

From the available records, we retrospectively identified consecutive patients admitted to the Department of Cardiology between January 2016 and September 2021 in whom the presence of severe TR was confirmed during the index hospitalization. The patients whose baseline clinical and echocardiographic data were not available were excluded.

Data collection

The data were collected from the electronic database and included: patient age, sex, comorbidities and previously performed cardiac procedures, the NYHA HF class, laboratory parameters, pharmacologic treatment, and procedures performed during the index hospitalization.

Routine transthoracic echocardiography (TTE) was performed and analyzed by a certified cardiologist during the index hospitalization, and the echocardiographic data were collected from the electronic database. The date of death was obtained from the patients’ family members and from the Polish national public registries. The primary end point was all‑cause mortality, and the period analyzed was from the day when the echocardiographic evidence of severe TR was obtained until death or the last follow‑up in May 2022.

Echocardiography

To obtain the analyzed parameters, 2‑dimensional TTE enriched with M‑mode and color Doppler was performed according to the guidelines^4,5. Tricuspid annular plane systolic excursion (TAPSE) was measured in M‑mode at the junction of the TV and RV free wall. The severity of TR and the function of other valves were assessed using qualitative, semiquantitative, and quantitative approaches provided in the guidelines.^5,9 Severe TR was defined based on echocardiographic parameters of TR as effective regurgitant orifice area (EROA) of 40–59 mm², TR regurgitant volume (TRvol) 45–59 ml, or vena contracta width (VCW) 7–13.9 mm. Massive TR was defined as EROA 60–79 mm², or TRvol 60–74 ml, or VCW 14–20.9 mm. Torrential TR was defined as EROA equal to or above 80 mm², TRvol equal to or above 75 ml, or VCW equal to or above 21 mm.¹⁰ Quantification of the chambers was performed in apical 2- and 4‑chamber views or RV‑focused apical 4‑chamber view for the left and right chambers, respectively. The left ventricular ejection fraction was assessed with the Simpson method.¹¹ Wall motion was measured using the 16‑segment model.¹¹ The probability of pulmonary hypertension (PH) was evaluated using systolic pulmonary artery pressure (SPAP), tricuspid regurgitation pressure gradient (TRPG), and echocardiographic signs suggesting PH regarding the ventricles, pulmonary artery, inferior vena cava (IVC), and right atrium (RA).^12-14 Additionally, the TAPSE/SPAP ratio was calculated for all patients.

We analyzed the prognostic significance of echocardiographic parameters associated with the RV function and remodeling,^15,16 that is, RV dysfunction defined as TAPSE below 17 mm, diameter and collapse of the IVC, where a value above 21 mm was considered IVC dilatation, flattening of the interventricular septum, presence of the echocardiographic features of PH and the TA dilatation, defined as TA diameter above 40 mm.

Statistical analysis

Statistical analysis was carried out with IBM SPSS Statistics package (version 28.0; IBM, New York, New York, United States). The Shapiro–Wilk test was performed to assess distribution of the continuous variables. The normally distributed variables are presented as mean (SD) and compared with the t test. The non‑normally distributed variables are presented as median with interquartile range (IQR) and compared with the Mann–Whitney test. Categorical variables are presented as a number and percentage and compared using the χ² test or the Fisher exact test. The survival analysis was performed using the Kaplan–Meier survival analysis, receiver operating characteristic (ROC) curve analysis, and nonadjusted and adjusted Cox regressions for evaluation of the factors associated with mortality. Adjusted Cox regression models included age, sex, diabetes mellitus (DM), chronic kidney disease (CKD), anemia, and echocardiographic parameters which were significant at univariable Cox regression. Statistical significance was established at 2‑sided P value below 0.05.

Patient characteristics

Overall, we analyzed 172 patients (93 women [54.1%]) with severe or greater TR. Of those, 126 individuals (73.2%) had severe TR, 33 (19.2%) had massive TR, and 13 (7.6%) had torrential TR. Median (IQR) EROA was 58 mm² (49–74 mm²), median TRvol was 57 ml (48–70.3 ml), and median VCW was 10 mm (8–12 mm). Baseline echocardiographic characteristics of TR are provided in Table 1. The mean (SD) age of the patients was 76.4 (10.5) years. The most common comorbidity was atrial fibrillation (n = 146; 84.9%). The remaining comorbidities included hypertension (n = 117; 68%), CKD (n = 93; 54.1%), coronary artery disease (CAD; n = 78; 45.3%), DM (n = 53; 30.9%), and respiratory diseases such as chronic obstructive pulmonary disease (COPD; n = 23; 13.4%) or asthma (n = 7; 4.1%). Some patients had a history of myocardial infarction (n = 42; 24.4%) and prior stroke or transient ischemic attack (n = 24; 14%). Peripheral edema or ascites were found in 79 (45.9%) and 33 (19.2%) patients, respectively. The median (IQR) EuroSCORE II and TRI‑SCORE values were 4.68% (2.88%–8.05%) and 14% (5%–34%), respectively.

The past procedures included 23 systemic valve interventions (13.4%), of which transcatheter aortic valve implantation was performed in 3 patients (1.7%) and mitral transcatheter edge‑to‑edge repair (TEER) in 2 patients (1.2%). Nine patients (5.2%) underwent coronary artery bypass grafting, 32 (18.6%) had percutaneous coronary intervention (PCI), and 4 (2.3%) tricuspid valvuloplasty (isolated). Almost a half (n = 81; 47.1%) of the patients were provided with an implantable device, of which 78 (45.3%) had a ventricular lead placed.

Interventions

During the index hospitalization, the following procedures were performed: cardiac surgery (aortic valve replacement, mitral valve replacement, or mitral valve repair) in 3 patients, percutaneous treatment in 7 patients, including PCI in 6 patients (3.5%) and 1 mitral TEER, implantation of a CIED in 26 patients (15.1%), battery replacement in 11 patients (6.4%), the CIED lead removal in 5 patients (2.9%), and catheter ablation in 2 patients (1.2%). Coronary angiography was performed in 43 patients (25%), documenting CAD in 21 individuals (12.2%). That included 11 patients (6.4%) with multivessel disease. The data concerning pharmacologic treatment, laboratory results, and echocardiographic parameters are presented in Table 2.

Follow‑up

Median (IQR) follow‑up was 24 (8.5–41) months. Clinical status at 1 year was available for 90.1% of patients, and at 2 years for 86% of patients. The overall mortality was 29.7% at 1 year and 47.3% at 2 years. Higher grade of TR, TAPSE below 17 mm, TAPSE/SPAP ratio below 0.26, and right atrial area (RAA) equal to or above 35 cm² were associated with worse survival in the Kaplan–Meier survival analysis with log‑rank test values of P = 0.04, P = 0.002, P = 0.02, and P = 0.02, respectively (Figures 1, 2, 3, 4). The characteristics of the patients according to the TAPSE value are presented in Table 2. The ROC curve analysis regarding 2‑year mortality showed the area under the curve (AUC) values for the cutoff of the RAA at 35.95 cm² (AUC = 0.612; P = 0.03); SPAP of 52.5 mm Hg (AUC = 0.627; P = 0.046), and IVC diameter of 2.45 cm (AUC = 0.621; P = 0.03).

The nonadjusted Cox regression results are presented in Table 3. Meaningful results were found for TAPSE below 17 mm (hazard ratio [HR], 1.905; 95% CI, 1.25–2.902; P = 0.003), median TAPSE/SPAP equal to or above 0.26 (HR, 0.502; 95% CI, 0.277–0.908; P = 0.02), TRPG above 36 mm Hg (HR, 1.52; 95% CI, 1.006–2.297; P = 0.047), anemia (HR, 1.869; 95% CI, 1.214–2.877; P = 0.005), and DM (HR, 1.915; 95% CI, 1.274–2.88; P = 0.002). The results for CAD and COPD were insignificant.

The adjusted Cox regression models were developed for age, sex, diabetes, CKD, anemia, and all echocardiographic parameters (Models 1–5), and Model 6 included age, sex, diabetes, CKD, anemia, and all significant echocardiographic parameters from models 1–5. In model 5, statistically signficant results were obtained for DM (HR, 2.048; 95% CI, 1.261–3.324; P = 0.004), anemia (HR, 2.161; 95% CI, 1.22–3.827; P = 0.008), RAA equal to or above 35 cm² (HR, 1.774; 95% CI, 1.101–2.856; P = 0.02), and TAPSE below 17 mm (HR, 2.054; 95% CI, 1.273–3.317; P = 0.003). All significant results are shown in Table 4.

The main echocardiographic findings included enlargement of the left (94.2%) and right (92.4%) atria and RV (73.3%). Dilatation of the pulmonary trunk was observed in nearly half of the patients, and more than a half had segmental contractility dysfunction. The remaining echocardiographic parameters are presented in Table 2.

Limitations

A major limitation of our study is a relatively small number of included patients and its retrospective nature. Another limitation is the fact that RH catheterization was not routinely performed in the studied patients and full characteristics of PH could not be performed. Although consecutive patients were enrolled, some selection bias cannot be excluded. Furthermore, this is a single‑center experience.

Conclusions

Patients presenting with severe TR are characterized by large comorbidity burden and high mortality despite intensive HF management. The choice of treatment method should be made based on disease‑specific risk‑stratification tools, such as TRI‑SCORE scale, to avoid the risk of underestimation. RV function is one of the most important predictive factors in TR, and echocardiographic parameters of RV function (TAPSE <⁠17 mm and TAPSE/SPAP ratio <⁠0.26) are associated with mortality.