Introduction
Tricuspid regurgitation is most often secondary and develops as a result of severe left-sided heart valve defects [1]. Functional regurgitation is characterized by no primary structural lesions of the valve on the contrary to, for example, infective endocarditis [2], rheumatic disease or congenital tricuspid valve diseases [3]. However, even secondary changes are structural in nature. In particular, secondary functional tricuspid regurgitation is a consequence of tricuspid annulus enlargement, right ventricle remodeling and increase in the distance between the papillary muscles and tricuspid annulus [4]. Tricuspid leaflets and chords themselves are indeed anatomically intact. Most of techniques for correction of functional tricuspid regurgitation are aimed at one of pathophysiological aspects (tricuspid annulus enlargement). Multiple annuloplasty techniques have been proposed which can be conceptually divided into 2 groups (implantation techniques and suture annuloplasty). Despite numerous data on advantages of implantation techniques, especially in long-term period [5], suture techniques (DeVega and its modifications) are still widely used [6]. At the same time, there is heterogeneity in scientific data, formation of study groups and surgical methods (rigid or soft ring [7], choice of implant size, technical modifications of suture techniques, etc.). There are no detailed surgical recommendations for choosing correction technique depending on the type of insufficiency and anatomy of tricuspid valve and right ventricle [8]. Discussions about indications for correction depending on severity of regurgitation and tricuspid annulus dimension continue [9].
The purpose of the study was to compare hemodynamic stability of suture and ring tricuspid valve annuloplasty for functional insufficiency in patients with left heart valve defects.
Material and methods
There were 746 tricuspid valve repair procedures for the period from January 2009 to December 2016. We excluded 70 patients with infective endocarditis, rheumatic tricuspid valve disease, congenital tricuspid valve disease and degenerative valve disease. Therefore, 676 patients with moderate-to-severe tricuspid insufficiency were recruited. A rigid ring was implanted in 294 patients, DeVega procedure was performed in 382 patients.
We used propensity score matching (PSM) to reduce the risk of bias associated with selection of patients for a particular type of correction, as well as to create homogeneous study groups. PSM included such parameters as age, gender, diabetes mellitus, chronic lung diseases, functional class of heart failure, pulmonary artery pressure, right ventricular size, tricuspid annulus diameter, and atrial fibrillation. Thus, 2 groups by 159 patients were distinguished. Flowchart diagram of the study is shown in Fig. 1.
Fig. 1. Flow chart of the study.
TR — Tricuspid Regurgitation
Surgeries were carried out through a median sternotomy. Standard establishing of cardiopulmonary bypass was applied (ascending aorta, superior and inferior vena cava). In all cases, Custodiol was used for myocardial protection. Support ring size was determined using an appropriate sizer. Anterior tricuspid leaflet area and intercommissural distance were considered. DeVega annuloplasty was performed using multifilament suture 2/0 with synthetic pads within the first and the last stitches. Degree of tricuspid annulus narrowing was determined using a hydraulic test.
Statistical analysis was performed using SPSS software (version 26.0; IBM SPSS Inc., Armoronk, NY). Continuous quantitative variables are presented as arithmetic mean and standard deviation. Categorical and nominal variables are presented as percentages (%). The Mann-Whitney test was applied to compare two independent samples of continuous variables, χ2 test or Fisher’s exact test — for categorical variables. The Wilcoxon test for paired measurements was used to compare two dependent samples. PSM was used to ensure homogeneous groups of patients by baseline clinical and demographic parameters. Kaplan-Meier freedom from recurrent moderate-to-severe tricuspid regurgitation was evaluated. Differences were assessed using the Log-rank test and considered significant at p-value <0.05.
Results
Preoperatively, age, left ventricular contractility and severity of concomitant diseases (diabetes mellitus, chronic obstructive pulmonary disease) were similar in both groups. The groups differed in other indicators characterizing right heart damage and surgical risk factors. There were more patients with previous surgery in the rigid ring group. Preoperative characteristics of patients are summarized in Table 1.
Table 1. Baseline clinical, demographic and echocardiographic characteristics of both groups
Variable |
Characteristics of overall groups |
Characteristics of patients after PSM |
||||
Rigid ring (n=294) |
Suture annuloplasty (n=382) |
p-value |
Rigid ring (n=159) |
Suture annuloplasty (n=159) |
p-value |
|
Age, years |
55.9 (19—75) |
57.2 (19—78) |
0.2638 |
55.4 (22—73) |
57.5 (19—76) |
0.7821 |
Male, n (%) |
95 (32.3) |
146 (38.2) |
0.0089* |
47 (29.5) |
30 (18.8) |
0.4043 |
BMI, kg/m2 |
24.9±2.1 |
22.4±3.1 |
0.0210* |
23.1±3.2 |
23.3±2.8 |
0.3891 |
BSA, m2 |
1.61±0.2 |
1.58±0.18 |
0.6120 |
1.57±0.17 |
1.57±0.17 |
0.7822 |
Risk factors: |
||||||
Insulin-dependent diabetes mellitus, n (%) |
11 (3.7) |
16 (4.2) |
0.0064* |
8 (5.03) |
7 (4.4) |
0.8231 |
COPD, n (%) |
34 (11.5) |
15 (3.9) |
0.3941 |
23 (14.4) |
20 (9) |
0.4932 |
Stroke, n (%) |
2 (0.7) |
6 (1.57) |
0.0038* |
0 |
0 |
— |
CKD, n (%) |
16 (5.4) |
12 (3.1) |
0.0310* |
11 (6.9) |
12 (12.5) |
0.9712 |
AF, n (%) |
186 (63.2) |
204 (53.4) |
0.0241* |
107 (67.3) |
88 (55.3) |
0.3172 |
Redo surgery, n (%) |
3 (1) |
4 (1.04) |
0.4121* |
0 |
0 |
— |
NYHA class ≥3, n (%) |
228 (77.5) |
282 (73.8) |
0.0311* |
101 (63.5) |
112 (70.4) |
0.2793 |
Primary heart valve disease: |
||||||
Mitral valve, n (%) |
201 (68.3) |
256 (67) |
0.2830 |
100 (62.8) |
98 (61.6) |
0.6541 |
Aortic valve, n (%) |
93 (31.7) |
126 (33) |
0.1621 |
59 (37.1) |
61 (38.3) |
0.4511 |
Echocardiography: |
||||||
LV EDD, cm |
5.1±2.3 |
5.3±1.9 |
0.4220 |
5.0±1.9 |
4.8±1.1 |
0.9622 |
LV EF, % |
56.8±9.6 |
54.1±7.2 |
0.1220 |
54.3±7.6 |
52.9±8.1 |
0.6015 |
Left atrium, cm |
5.1±1.8 |
5±2.3 |
0.0001* |
4.9±1.7 |
4.7±2.4 |
0.4216 |
PA systolic pressure, mm Hg |
55.4±12.2 |
57.3±8.2 |
0.0450* |
54.8±11.3 |
57.3±8.2 |
0.5688 |
Tricuspid regurgitation grade |
2.74±0.3 |
2.57±0.2 |
0.0001* |
2.8±0.4 |
2.57±0.2 |
0.7867 |
Moderate tricuspid regurgitation (grade 0-2), n (%) |
102 (34.7) |
148 (38.7) |
0.0001* |
54 (33.9) |
66 (41.5) |
0.2318 |
Severe tricuspid regurgitation (grade 3-4), n (%) |
192 (65.3) |
234 (61.3) |
0.0001* |
105 (66) |
93 (58.4) |
0.5628 |
Tricuspid annulus, cm |
3.76±1.3 |
3.74±1.1 |
0.4014 |
3.7±1.3 |
3.69±1.1 |
0.5609 |
Note. BMI — body mass index, BSA — body surface area, COPD — chronic obstructive pulmonary disease, CKD — chronic kidney disease, AF — atrial fibrillation, LV EDD — left ventricular end-diastolic dimension, LV EF — left ventricular ejection fraction, PA — pulmonary artery.
Perioperative and immediate postoperative results are presented in Table 2. Sizes of implanted rings are summarized in Table 3. In-hospital mortality and incidence of early recurrent tricuspid insufficiency were similar in both groups.
Table 2. Perioperative data
Variable |
Characteristics of overall groups |
Characteristics of patients after PSM |
||||
Rigid ring (n=294) |
Suture annuloplasty (n=382) |
p-value |
Rigid ring (n=159) |
Suture annuloplasty (n=159) |
p-value |
|
Aortic valve replacement, n (%) |
5 (1.7) |
11 (2.9) |
0.0260* |
2 (1.3) |
1 (0.6) |
0.6321 |
Mitral valve replacement/repair, n (%) |
201 (68.3) |
256 (67.1) |
0.7812 |
104 (65.4) |
110 (69.2) |
0.2412 |
Aortic valve replacement + mitral valve replacement/repair, n (%) |
88 (30) |
115 (30) |
0.1254 |
53 (33.3) |
48 (30.2) |
0.2354 |
CPB time, min |
127±20.7 |
113±26.4 |
0.2368 |
112±18.7 |
100±19.8 |
0.8412 |
Aortic cross-clamping time, min |
90±9.7 |
93±7.8 |
0.0001* |
83±11.2 |
84±13 |
0.7564 |
ICU-stay, days |
2.1±0.6 |
2.3±0.8 |
0.7831 |
2±0.5 |
2.2±0.3 |
0.6556 |
In-hospital mortality, n (%) |
8 (2.72) |
9 (2.35) |
0.2181 |
4 (2.5) |
3 (1.9) |
0.5462 |
Respiratory failure followed by prolonged ventilation, n (%) |
15 (5.1) |
10 (2.6) |
0.0254* |
3 (1.8) |
4 (2.5) |
0.3467 |
Stroke, n (%) |
3 (1) |
4 (1) |
0.3465 |
2 (1.2) |
1 (0.6) |
0.5234 |
Pacemaker implantation, n (%) |
13 (4.4) |
13 (3.4) |
0.8312 |
10 (6.3) |
4 (2.5) |
0.6587 |
LV EF at discharge, % |
54.9±11.6 |
51.8±9.3 |
0.2345 |
50±10.3 |
54±7.8 |
0.2451 |
Tricuspid regurgitation grade ≥3 at discharge, n (%) |
7 (2.4) |
10 (2.6) |
0.6821 |
2 (1.2) |
2 (1.2) |
0.9999 |
Tricuspid regurgitation grade at discharge |
0.9±1.1 |
1.0±0.9 |
0.6821 |
0.8±0.7 |
1.0±0.5 |
0.6581 |
Note. CPB — cardiopulmonary bypass, ICU — intensive care unit.
Table 3. Ring sizes
Ring number |
Before PSM (n=294) |
After PSM (n=159) |
No. 30. n (%) |
42 (14.3) |
18 (11.6) |
No. 32. n (%) |
157 (53.4) |
91 (57.3) |
No. 34. n (%) |
93 (11.6) |
48 (30.1) |
No. 36. n (%) |
2 (0.7) |
2 (1) |
Right heart changes after PSM are described in Table 4. Ring annuloplasty and suture technique resulted satisfactory outcomes in early and mid-term postoperative period (within 1 year) regarding recurrent tricuspid insufficiency. Moderate-to-severe regurgitation at discharge was diagnosed in 2.4% (n = 7) and 2.6% (n=10) of patients in both groups, respectively.
Table 4. Echocardiography parameters of the right heart and pulmonary circulation
Variable |
Rigid ring group after PSM |
Suture annuloplasty group after PSM |
||||||
Prior to surgery |
30 days after surgery |
Long-term outcomes (up to 5 years) |
p-value |
Prior to surgery |
30 days after surgery |
Long-term outcomes (up to 5 years) |
p-value |
|
RV, cm |
2.8±1.4 |
2.9±1.8 |
2.5±1.9 |
0.0835 |
2.6±1.2 |
2.9±1.7 |
2.5±2.1 |
0.0897 |
PA systolic pressure, mm Hg |
54.8±11.1 |
37.2±116.5 |
41±112.2 |
0.0100* |
55.6±9.1 |
36.7±8.9 |
40.9±11.2 |
0.0093* |
Tricuspid annulus, cm |
3±0.9 |
2.8±1.1 |
3.1±10.7 |
0.3511 |
3.1±0.6 |
3±0.9 |
2.7±1.2 |
0.2415 |
Tricuspid regurgitation grade |
2.7±0.6 |
— |
— |
— |
2.6±0.9 |
— |
— |
— |
Note. RV — right ventricle.
Both techniques did not increase the risk of pacemaker implantation. Aortic cross-clamping time was significantly higher in the ring annuloplasty group (p = 0.0001). This difference disappeared after PSM.
Mean long-term postoperative follow-up period was 72.4±16 months (range 8 — 120). Long-term results were assessed in 299 (94%) patients of both groups after PSM. There were no redo cardiac surgeries. Echocardiography data on the right heart and pulmonary circulation are summarized in Table 4. Significant decrease in pulmonary artery pressure was accompanied by stable postoperative dimensions of the right ventricle and tricuspid annulus in both groups.
Actuarial freedom from recurrent tricuspid insufficiency is shown in Fig. 2. Ten-year freedom from recurrent severe tricuspid insufficiency was 53.7% and 32.4% after ring and suture annuloplasty, respectively. This difference was significant in favor of rigid ring implantation (plog-rank = 0.01).
Fig. 2. Freedom from severe (grade 3—4) tricuspid regurgitation.
Discussion
Multiple studies were devoted to correction of secondary tricuspid regurgitation. There are data on the advantages of implantation techniques and more stable results for rigid and semi-rigid rings [10, 11]. Nevertheless, Veen K. et al. [12] performed a systematic review of 14 studies devoted to ring and suture tricuspid annuloplasty in 2020. The authors found no differences in the incidence of recurrent tricuspid regurgitation in long-term postoperative period. Earlier, Parolari A. et al. [13] reported a significant advantage of rigid ring implantation. Overall 15-year freedom from recurrent tricuspid regurgitation was 57.3% that is similar to our 10-year outcomes after ring annuloplasty (53.7%).
There are 2 main conclusions from this study. First, ring implantation has been shown to be beneficial within 10 years. Secondly, despite the advantages over suture annuloplasty, even ring implantation did not provide a stable hemodynamic result in almost half of patients 10 years after surgery. These results can be explained by right ventricular remodeling (Table 4). There was no significant reverse remodeling of the right ventricle despite left-sided heart valve correction followed by gradual decrease in pulmonary artery pressure. Similarly, isolated annuloplasty for secondary mitral regurgitation is not followed by stable results due to left ventricular remodeling and dysfunction [14]. Surgical correction of only one anatomical aspect involved in valve failure (fibrous annulus enlargement) does not provide a favorable hemodynamic result. Thus, additional subvalvular procedures may be essential for stable hemodynamic result rather annuloplasty technique per se. Some subvalvular corrections were described in experimental models [15]. Development and clinical evaluation of such procedures should be the subject of further research.
Conclusion
Both techniques can effectively and safely eliminate secondary functional tricuspid regurgitation during left-sided heart valve surgery and ensure good immediate results.
Within 10 years after surgery, ring annuloplasty ensures significantly higher freedom from recurrent tricuspid insufficiency.
At the same time, even a rigid ring did not provide an excellent freedom from regurgitation in long-term postoperative period. Therefore, development of reconstructive surgical techniques in addition to annuloplasty is required to improve long-term outcomes in patients with functional tricuspid insufficiency.
The authors declare no conflicts of interest.