Paravalvular leak

Article written and reviewed by Dr Luca Moretti

A paravalvular leak is regurgitation that passes through a space between the prosthesis and the anchoring tissue rather than through the normal valvular orifice. It may occur after surgical valve replacement or transcatheter implantation and has a very broad clinical spectrum, ranging from a minimal echocardiographic finding to heart failure, transfusion-dependent hemolysis or hemodynamic instability. Its nature is therefore defined mainly by the location of the jet, not simply by the presence of regurgitation.

Distinguishing it from intraprosthetic regurgitation is essential because mechanical valves have physiological washing jets and bioprostheses may show a minimal central jet. When flow originates outside the ring, by contrast, the finding indicates a sealing defect or dehiscence. Artifacts and shadowing can make this distinction particularly difficult in the mitral position, where three-dimensional transesophageal echocardiography is often required.

The clinical significance of a leak depends on regurgitant volume, defect geometry and hemodynamic consequences. A large orifice mainly causes volume overload, whereas a small irregular channel may generate high shear and severe hemolysis despite quantitatively modest regurgitation. Assessment therefore cannot rely on color Doppler jet area alone but must integrate hemodynamics, chamber remodeling, laboratory findings and symptoms.

Treatment is selected according to mechanism and anatomy. A mild stable defect may simply be monitored, whereas endocarditis, annular instability or extensive dehiscence point toward surgery; in selected patients, percutaneous closure may instead avoid reoperation. True clinical success does not consist merely in placing a device, but in abolishing the responsible jet without compromising prosthetic valve function.

Etiology, anatomy and pathophysiology

After surgery, an early leak results from incomplete apposition between the sewing ring and tissue, uneven suture distribution, calcification, friability or technical difficulty. Mitral annular calcification, previous endocarditis and reoperation increase risk. Small defects may be recognized intraoperatively; whether they are acceptable depends on severity, stability and the feasibility of safe correction.

A late leak suggests loosening or rupture of sutures, progression of calcification, tissue fragility or endocarditis with dehiscence. Newly developed paravalvular regurgitation after a period of stability should be considered infectious until adequately evaluated, even in the absence of fever. Abscesses and pseudoaneurysms may extend beyond the annulus and make simple closure insufficient.

After TAVI, the predominant mechanism is malapposition of the frame to the calcified native valve: calcium nodules, an elliptical annulus, sizing, implantation depth and underexpansion can create channels between the skirt and tissue. Newer generations, owing to more effective sealing systems and better CT planning, have reduced moderate or severe regurgitation without eliminating it entirely; a minimal leak may also change over time with frame expansion and remodeling.

The defect location should be described relative to shared anatomical landmarks. In the mitral position, for example, the three-dimensional surgical view places the aorta superiorly and the left atrial appendage laterally, whereas in the aortic position the cusps and coronary arteries are used; because “clock-face” positions change with perspective, the report should always state the orientation adopted. The anatomical map is used to plan access and trajectory, not merely to name the site of the leak.

The channel may be circular, crescent-shaped, serpiginous, calcified or multiple, and its atrial and ventricular orifices may have different dimensions. Device selection therefore cannot be based on maximum width alone because tunnel length, available rim and distance from the occluders must also be considered; with mechanical prostheses, a few millimeters may be enough to cause disc impingement.

Hemolysis results from flow acceleration, turbulence and erythrocyte collision within the jet. Cellular fragments are progressively removed and the body increases reticulocyte production as long as iron stores, folate and bone marrow capacity can compensate, while renal impairment and hemoglobin loss worsen the picture. The degree of mechanical hemolysis therefore depends more on residual shear than on the total regurgitant volume.

Hemodynamic consequences depend on the site of the defect. A mitral leak returns volume to the left atrium, increases pulmonary pressure and promotes ventricular dilatation, whereas an aortic leak causes diastolic regurgitation into the left ventricle; after TAVI, even moderate regurgitation may weigh more heavily on a ventricle that is already hypertrophied and poorly compliant. In right-sided prostheses, systemic congestion and reduced cardiac output predominate.

Clinical presentation, hemolysis and suspicion of infection

The most common manifestation is exertional dyspnea, which may be accompanied by edema, orthopnea and progressive reduction in functional capacity. Because patients often spontaneously limit their activity, severity may increase slowly without being immediately recognized; when dehiscence causes acute regurgitation, pulmonary edema, hypotension or shock may develop. The murmur may also be subtle, especially with eccentric jets or rapid pressure equalization.

Clinical hemolysis presents with fatigue, pallor, jaundice, dark urine and dyspnea worsened by anemia. Laboratory findings typically show reduced hemoglobin, elevated lactate dehydrogenase, low haptoglobin, indirect hyperbilirubinemia and reticulocytosis, with schistocytes supporting a mechanical mechanism; iron or folate deficiency may nevertheless blunt the reticulocyte response and mask the increase in erythrocyte production.

Subclinical compensated hemolysis is not by itself an indication for a procedure. Correction becomes relevant when hemolysis causes symptoms, renal deterioration, transfusion requirement or persists despite supportive care; similarly, an echocardiographically small defect should not be dismissed if it continues to sustain anemia. The true therapeutic target is the high-shear jet, to the point that even a minimal residual jet can prevent normalization.

Fever, chills, weight loss, emboli or a new conduction block increase suspicion of endocarditis and, when clinical stability allows, require multiple blood cultures before antibiotics are started. A new or progressive late leak is itself a sign of possible infectious dehiscence; therefore, the absence of vegetations on the first echocardiogram is not sufficient to exclude prosthetic valve endocarditis.

Assessment should include heart failure, rhythm, ventricular function, pulmonary pressure and the status of other valves, without assuming that the leak explains the entire clinical picture. Anemia, for example, may have gastrointestinal, renal or bone marrow causes, and possible autoimmune hemolysis should be distinguished with appropriate testing. Correction of the defect is therefore justified when the pathophysiological link among leak, symptoms and consequences is convincing.

After TAVI, at least moderate paravalvular regurgitation is associated with a worse prognosis, whereas the impact of mild forms varies with definition, device and population. Imaging should therefore use updated and sufficiently granular grading classes without automatically transferring data from first-generation technologies to current devices, and temporal trajectory should be considered together with the initial grade.

Hemodynamic instability, refractory pulmonary edema, rapidly progressive hemolysis and dehiscence with prosthetic motion are emergencies and require transfer to a center with cardiac surgery and structural intervention capability. Diuretics, vasodilators and transfusion support may be necessary to stabilize the patient but should not delay definitive control of the defect.

Echocardiography and multimodality quantification

Transthoracic echocardiography is the first examination used to identify the jet, assess its severity and measure its consequences, but artifacts can substantially limit accuracy. In the mitral position the jet may be obscured by the prosthesis, whereas in the aortic position multiple windows must be interrogated; the presence of flow outside the ring, together with convergence and vena contracta, helps define its origin. Quantification nevertheless remains multiparametric.

In the aortic position, the percentage of circumference occupied by the jet can contribute to grading, but it assumes relatively regular geometry and loses reliability with multiple or eccentric leaks. Color jet area also depends on Nyquist limit, gain and pressure, so diastolic flow reversal in the aorta, Doppler density, deceleration time and ventricular remodeling must be integrated without assigning decisive weight to a single parameter.

In the mitral position, vena contracta, flow convergence, continuous-wave signal, pulmonary venous flow and ventricular volume are integrated, recognizing that PISA is unreliable for crescentic orifices and non-planar surfaces. The same prosthesis may also have multiple leaks and concomitant intraprosthetic regurgitation; when echocardiography and the clinical picture are discordant, volumetric quantification with CMR may provide complementary information.

Three-dimensional transesophageal echocardiography provides an en face view that defines the number, shape and course of defects, while 3D color facilitates mapping and biplane imaging guides crossing during closure. Because temporal resolution and blooming may alter apparent dimensions, measurements should be confirmed in multiple planes. The procedure therefore requires a truly shared spatial language between the echocardiographer and the interventionalist.

Cardiac CT defines calcification, dehiscence, channel morphology, relationships with coronary arteries and occluders and may help identify abscesses, with multiplanar reconstruction being particularly useful in complex tunnels and access planning. Metal artifacts and rapid heart rate can reduce image quality, and contrast use must be balanced against renal function, which may already be impaired by hemolysis.

When an infectious origin is suspected, PET/CT and CT add information on periprosthetic activity and anatomical extent beyond transesophageal echocardiography alone. Postoperative uptake may, however, reflect sterile inflammation and must be interpreted with expert protocols; the diagnosis of endocarditis remains the result of integrating microbiology, imaging and clinical criteria. Closing an infected defect with a device without controlling the source risks trapping infection and causing procedural failure.

Before treatment, the team must define precisely which leak is the target and what result can be considered clinically effective. Reducing regurgitation from severe to moderate may improve heart failure without necessarily resolving hemolysis, just as closure of the largest defect may leave a small high-velocity jet capable of maintaining transfusion requirements. Intraprocedural imaging therefore verifies not only the residual leak but also leaflet motion, device stability and any pericardial effusion.

Medical therapy and percutaneous closure

Medical therapy has mainly a supportive role: diuretics and heart-failure treatment reduce congestion, while iron, folate and transfusions correct hematologic consequences and renal function should be protected. Beta-blockade may reduce shear in selected cases, but evidence is limited; no drug, however, closes a dehiscence, and support should not become an indefinite substitute for correction in a patient who is a procedural candidate.

Percutaneous closure is considered for clinically significant regurgitation or hemolysis in patients with high surgical risk or a favorable procedural profile. Not all devices are specifically designed for leaks, and use depends on anatomy and availability. Small round channels, crescent-shaped defects and multiple leaks require different configurations, sometimes with more than one device.

Access to a mitral leak may be transseptal, retrograde or, in selected circumstances, transapical, whereas access to an aortic leak is more often retrograde. The trajectory must allow the defect to be crossed without injuring tissue or entrapping material, which is why guidewires and catheters are controlled simultaneously with echocardiography and fluoroscopy. In the presence of a mechanical prosthesis, disc freedom is checked repeatedly before final release.

Device selection takes into account minimum and maximum diameter, tunnel length, shape and supporting rim. Excessive oversizing may deform the annulus, interfere with the prosthesis or promote embolization, whereas an undersized device leaves a residual shunt. Percutaneous success is therefore defined by reduction of the leak to mild or absent without death, embolization, valvular interference or urgent surgery, but the true benefit must also be confirmed during clinical follow-up.

Complications include bleeding, tamponade, stroke, embolization, prosthetic interference, new or persistent hemolysis and residual leak. Residual flow through the device mesh may initially produce shear and later decrease with thrombosis, but worsening anemia requires prompt reassessment; for this reason, the ability to retrieve the device and obtain surgical backup should be anticipated from the planning stage.

Observational series and registries document high feasibility in experienced centers and improvement in symptoms or transfusion requirements when closure is effective. Comparison with surgery remains affected by patient selection and does not amount to a generalizable randomized trial; in this field, operator experience and imaging quality may influence outcome more than the specific device model used.

After closure, echocardiography, complete blood count and hemolysis markers are monitored, while the antithrombotic regimen is defined according to the pre-existing prosthesis, rhythm and device, and a mechanical valve continues to require a VKA. Follow-up should look for migration, residual leak and ventricular recovery, remembering that restoration of iron stores may take time even after the jet has been abolished.

Surgery, choice of intervention and prognosis

Surgery is preferred in active endocarditis, abscess, extensive dehiscence, prosthetic instability, anatomy unsuitable for closure or the need for other corrections. It allows device removal, debridement and annular reconstruction. It may be technically demanding because of calcification and previous operations, but it addresses the mechanism comprehensively when focal closure would be insufficient.

The choice between surgical redo and percutaneous treatment should be multidisciplinary and integrate operative risk, frailty, life expectancy, location, size and number of leaks, presence of infection and the need for bypass surgery or intervention on other valves. High surgical risk favors a transcatheter approach only when anatomy offers a reasonable likelihood of abolishing the jet, because a less invasive procedure destined to leave hemolysis is not necessarily less burdensome in the overall outcome.

A mild asymptomatic leak can be followed with a baseline examination and serial surveillance, but increasing regurgitation, new chamber dilatation, anemia or symptoms reopens the treatment decision. After TAVI, the grade may vary with blood pressure and remodeling and should therefore be compared under similar hemodynamic conditions; after surgery, by contrast, a growing late leak again requires exclusion of endocarditis.

Prognosis depends on regurgitation severity, degree of heart failure, hemolysis, presence of infection and success of correction. Significant paravalvular regurgitation after TAVI and symptomatic surgical dehiscence are associated with adverse outcomes, whereas effective closure may reduce hospitalizations and transfusions; if an important residual leak persists, risk remains. It should nevertheless be remembered that mortality does not depend on the leak alone because these patients often have substantial comorbidities.

Prevention includes dental health and endocarditis prophylaxis when indicated, meticulous technique, preimplant imaging and intraoperative assessment. Annular calcification should be recognized before the procedure to choose the strategy and discuss risk, while after implantation fever and new dyspnea require rapid evaluation because the time between dehiscence and diagnosis may affect the possibility of repair.

Paravalvular leak ultimately requires a three-dimensional interpretation of the problem: anatomy determines the jet, the jet explains hemolysis or volume overload, and clinical consequences define the indication for treatment. Following this chain prevents both intervention on an innocuous image and underestimation of a small channel responsible for systemic disease. Optimal management therefore arises from integration of imaging, clinical hematology, interventional cardiology and surgery.

Grading after TAVI uses more detailed classes than the traditional mild, moderate and severe categories because boundaries between small defects are frequent and geometries are multiple. Assessment considers the number of jets, circumferential extent, vena contracta width, aortic diastolic flow and volume. An intermediate class should not, however, create false precision: image quality and parameter concordance are stated, especially when the decision concerns another procedure.

During mitral leak closure, defect location directly influences selection of the transseptal puncture. A medial and a lateral leak require different height and orientation, and an unfavorable trajectory can make both tunnel crossing and device stabilization difficult; in some cases arteriovenous loops provide greater support at the cost of greater complexity and additional risk. Three-dimensional planning reduces manipulation near discs, the aorta and the atrial wall.

In surgical aortic leaks, proximity to the coronary arteries and conduction system affects device size and deployment technique. After TAVI, by contrast, post-dilation may improve sealing when the frame is underexpanded, whereas a plug is better suited to persistent accessible focal channels; in selected cases a second valve can correct malposition but adds another frame and may compromise the coronary arteries. The solution should therefore derive from the mechanism, not from the grade of regurgitation alone.

Transfusion management should avoid both undertreatment and chronic dependence. Intravenous iron may be necessary when functional loss exceeds absorption; folate supports erythropoiesis; transfusion is tailored to symptoms, ischemia and instability rather than to a universal threshold. Ongoing depletion of stores and renal injury are clinical consequences even before extreme anemia develops and may strengthen the indication for correction.

If a leak persists after the first attempt, the first step is to determine whether the residual flow passes alongside the device, through its mesh or through a second untreated channel. The timing of a possible repeat procedure depends on potential thrombosis of the device, the course of hemolysis and clinical stability; adding devices without reconstructing the new anatomy may increase interference without closing the true target. A new assessment with three-dimensional TEE and CT should therefore precede the decision.

Mechanical prostheses add specific antithrombotic complexity because the VKA cannot be stopped without a strategy, while vascular access, transseptal puncture and devices increase bleeding risk. The therapeutic target, any bridging and resumption of anticoagulation are agreed according to valve position and risk; during deployment, fluoroscopy and echocardiography must also confirm that no disc contacts the plug. Hemostatic safety and unrestricted occluder motion are therefore simultaneous goals, and any change in click or gradient after the procedure requires immediate verification.

Quality of life and dependence on care complement traditional endpoints because reducing transfusions, restoring exercise capacity and decreasing hospitalizations may represent a substantial benefit even when mild residual regurgitation remains. Conversely, an apparently excellent echocardiographic result without functional improvement requires investigation for persistent anemia, ventricular dysfunction or pulmonary disease. Late assessment therefore distinguishes anatomical success from benefit actually perceived by the patient.

A scheduled clinical follow-up also verifies adherence, antithrombotic tolerance and the need for rehabilitation, completing instrumental measurement of the result.