11.11.3 Indications and surgical results

 Severe symptomatic organ failure is an indication for surgery. However, in 75% of cases, tricuspid insufficiency is treated primarily for right-sided failure (see Table 12.6), pulmonary hypertension (see Table 12.11) and left-sided disease. Although it was long thought that this treatment was sufficient to reduce regurgitation to levels of no clinical significance, it has been shown that the persistence of significant TI after mitral or aortic surgery and after coronary artery bypass grafting significantly worsens the prognosis, even if the patient is asymptomatic [7,18,22,23]; long-term mortality after left heart surgery is significantly increased in the presence of moderate to severe TI (HR 2.4-3.1) [20]. What's more, TI tends to worsen over the following 3 years, with a concomitant worsening of RV dysfunction [10,11,18]. As correction of moderate to severe TI "en passant" eliminates the leak in 85% of cases and hardly increases the morbidity-mortality of left heart surgery, the surgical attitude has become more interventional in recent years, as right-sided function tends to improve after TI reduction [8,10]. This attitude is reinforced by the high operative mortality (10-25%) of subsequent reintervention to correct persistent TI [26].

 Indications

The currently accepted indications for tricuspid valve surgery are as follows [3,5,11, 19,23,26,28,30].

• Severe symptomatic TI of organic or functional origin that does not respond to medical therapy;
• Severe primary TI, even if asymptomatic, with LV dilatation (basal diameter > 4.0 cm in 4-cavity view) and signs of progressive right-sided dysfunction (TAPSE < 15 mm, S' velocity < 9.5 cm/s, longitudinal systolic strain < -20%, Tei index > 0.45);
• Severe primary or secondary TI in patients requiring left-sided valve surgery;
• Moderate or severe TI with significant annular dilatation (diameter > 4.0 cm in 4 cavities or > 7.0 cm in the surgical field) and/or LV failure in patients requiring left heart surgery;
• Severe traumatic TI, even if asymptomatic (non-emergent).

Plasty is preferable to valve replacement except in cases where the leaflets are too deformed or too restrictive. In the absence of left-sided pathology, severe primary TI in an asymptomatic patient is an indication for surgery in cases of RV dysfunction/dilatation. It is important that anaesthetists are aware of these recommendations, as their echocardiographic knowledge places them at the centre of intra-operative discussions when deciding whether or not to intervene on the tricuspid valve during left heart surgery.

If it is secondary, tricuspid insufficiency develops in parallel with the size of the annulus [15]. Measurement of annulus diameter is preferable to assessment of TI by colour jet because it is not as dependent on haemodynamic conditions as Doppler measurements; the indication for surgery is a diameter > 40 mm or > 21 mm/m² in 4-cavity view (3,5,19). Intraoperative TEE is a necessary guide to the mechanism, indication and outcome of valvuloplasty.

 Surgical procedure

Several correction techniques are available to the surgeon (Figure 11.61 and Figure 11.159) [26]. 

 Figure 11.159: Tricuspid valve surgery. A: Minor residual TI after De Vega plasty for ARF; the mitral valve was replaced by a mechanical prosthesis (yellow arrow). B: Carpentier ring (yellow arrow) and chordal reimplantation (green arrow) for traumatic rupture of the anterior leaflet (same case as Figs 11.117A and 117B). C: Replacement of the tricuspid valve with a bioprosthesis, clearly showing the ring (arrows) and two leaflets.

•  Annuloplasty using narrowing sutures along the insertion of the anterior and posterior leaflets (DeVega plasty).
• Plication of the annulus in its posterior part, completely obliterating the posterior leaflet and converting the tricuspid valve into a bicuspid valve (Kay's bicuspidisation).
• Plasty by insertion of a semi-rigid ring (Carpentier) or a saddle-shaped ring (MC3); this ring is interrupted at the level of the septum to avoid damage to the atrioventricular node and the His bundle.
• Common suture of the tips of the 3 leaflets, transforming the valve into a three-leaf clover shape in diastole (similar to the Alfieri technique for the mitral valve).
• Reimplantation of chords in the event of rupture; resection of prolapse.

Video: Result of tricuspid plasty after traumatic rupture of the anterior leaflet chords; absence of residual leakage and restrictive effect on diastolic flow (in blue).


Video: Result of tricuspid plasty for posterior leaflet tilt in Barlow's disease; minimal residual leakage.

• Replacement of the valve with a prosthesis; a bioprosthesis is implanted rather than a mechanical prosthesis because their survival is identical. In fact, the right heart is less stressed than the left, which is why bioprostheses have a longer lifespan in the tricuspid position than in the mitral or aortic position [9,13]. On the other hand, mechanical prostheses have a much higher rate of thrombosis because the flow is slow due to the very low pressure regime.
• In patients with conduction block, an epicardial lead is implanted at the same time as the tricuspid prosthesis, as it is impossible or inadvisable to pass a pacemaker lead through the epicardium.

 When performed at the same time as left heart surgery, tricuspid repair is performed during rewarming while on ECC after aortic declamping. For isolated tricuspid surgery, access is via a sternotomy or right mini-thoracotomy. The operation is performed on bypass, usually with heart beating (provided there is no PFO) [3].

 Outcomes

After correction, the mean gradient across the tricuspid valve must remain < 3 mmHg [12]. The average failure rate for surgical repair is 14% [22]. Predictors of failure are essentially those that characterise preoperative dilatation of the RV: basal end-diastolic diameter > 4.2 cm, basal end-systolic diameter > 3.7 cm, maximum width > 4.9 cm [17]. Functional results are better (85% of cases without revision at 10 years) and survival at 15 years is doubled with annuloplasty compared to plastic surgery according to De Vega [25]. However, annuloplasty is always preferable to tricuspid valve replacement (TVR), which has inferior outcomes: higher rate of postoperative right-sided failure (28% versus 9%), increased mortality (11% versus 4-7%), anticoagulation, risk of thrombosis (1%/year) [24,25,29]. The prosthesis is a rigid mechanical element that blocks the circular contraction of the base of the RV and limits its longitudinal contraction [22]. As a result, TVR is usually associated with LV failure requiring adequate inotropic support. Right ventricular dysfunction is the most important determinant of postoperative mortality, more important than pulmonary hypertension [23].

Success criteria for repair are more flexible than after mitral valve repair because residual leakage is well tolerated [4,14,16,25]:

• Small residual TI;
• ΔPmax < 4 mmHg;
• Vena contracta < 0.3 cm;
• Absence of PISA and systolic reflux in the IVC and suprahepatic veins.

After correction of a TI, the immediate aftereffects are difficult because the RV encounters PAR as full afterload and no longer benefits from the pressure valve represented by the tricuspid insufficiency. In addition, any mechanical system will block the subtle balance between circular and longitudinal contraction of the RV, the latter being primarily responsible for ejection into the PA. Right-sided failure is common and requires immediate use of beta-catecholamines (dobutamine), inodilators (milrinone, levosimendan) and pulmonary vasodilators, as well as norepinephrine to maintain coronary perfusion and interventricular septal balance (see Tables 12.6 and 12.11).

Surgical treatment of acute bacterial endocarditis in drug users involves removal of the valve. It is sometimes recommended not to immediately implant a prosthetic valve in this septic environment. The haemodynamics are then similar to those of a Fontan: filling of the RV is passive and depends solely on the CVP, which must remain high; spontaneous ventilation is ideal to ensure pulmonary flow [1]. In a second phase (6-9 months), the tricuspid valve can be replaced by a prosthesis.

 Percutaneous procedures

Given the high morbidity and mortality associated with tricuspid  surgery on ECC in high-risk patients, a number of percutaneous transcatheter techniques have been developed in recent years [2,3,21].

• MitraClip™: The device used for the mitral valve is transposable to the tricuspid valve, with a success rate of 80% [6].
• Tric Valve™: implantation of a valve bioprosthesis (TAVI type) at the neck of the vena cava as it enters the RA; this technique prevents vena cava reflux but does not overload and dilate the RV and RA.
• FORMA™: foam-filled (polymer) balloon mounted on a rod anchored to the apex of the RV and placed in the centre of the tricuspid valve; it fills the regurgitant orifice in systole, the leaflets come to rest on it and the valve is sealed.
• Triclip™: The use of a MitraClip allows bicuspidisation of the tricuspid valve at the anteroseptal and/or posteroseptal commissures and reduction of leakage.
• Trialign™: the ring is narrowed by traction between two anchor points (pledgets) at the anteroposterior and septoposterior commissures, achieving plication of the posterior leaflet.
• TriCinch™: traction by a Dacron band between the anteroposterior region of the ring and a stent implanted in the inferior vena cava; the ring is then bent as required.
• Millipede IRIS™: A fully adjustable nitinol ring is placed inside the tricuspid annulus, to which it is attached by micro-screws and then constricted to the desired size.

Two different valve implantation systems have been developed to date.

• TricValve™ (mentioned above): self-expanding nitinol bioprosthesis with pericardial valves placed in the IVC between the RA and the junction of the suprahepatic veins; the aim is to prevent systolic reflux into the IVC and associated congestion in the viscera and lower limbs.
• NaviGate™: orthotopic nitinol bioprosthesis with pericardial valves, equipped with ribs for anchoring to the tricuspid annulus; it is available in several sizes (from 36 to 52 mm).

These techniques are still in their infancy, but the success rate is already >75%; in-hospital mortality is 3.5-5.1% [27,29]. The reduction in TI is less than with conventional surgery, but the improvement in clinical status is certain. Patients are put on lifelong aspirin and anti-vitamin K for 3 months, but many are anticoagulated for atrial fibrillation anyway.

Several of these devices are implanted via the jugular vein, which must remain accessible to the operator. Intra-operative TEE is essential and justifies general anaesthesia. The anaesthetist/echocardiographer has an important role to play in the placement of these devices. Whether the approach is via the IVC or the SVC, access to the tricuspid valve requires a difficult manoeuvre in the RA because the axis of the tricuspid valve is not in line with the vena cava.

 

Operative indications for tricuspid regurgitation

Indications depend on the degree of dilatation of the tricuspid valve, its origin and the surgical context. - Severe symptomatic TI - Severe TI with annular dilatation (D > 4.0 cm in 4 cavities) during surgery for left-sided valve disease - Severe traumatic TI (elective) Repair is straightforward, but replacement is associated with very difficult post-operative sequelae. Operative mortality: 1-2% for concomitant plasty, 4-7% for isolated plasty, 11% for replacement.