Heart valve disease presenting in adults: investigation and management

Murmur alone

Limited evidence showed that murmur is an indicator of valve disease. But the evidence also showed that a substantial proportion of people with a murmur do not have valve disease confirmed by a reference test. The committee agreed that 'innocent' murmurs can occur, particularly during the teenage and young adult years and in pregnancy. These are difficult to differentiate from pathological murmurs by clinical examination alone. The evidence was not strong enough to recommend that everyone with a murmur should be referred for echocardiography. The committee agreed that this would be a change in practice, would increase pressure on echocardiography services and would offer uncertain benefit. However, when the nature of the murmur, family history, age or medical history suggest possible valve disease, echocardiography should be considered to establish a diagnosis.

Systolic murmur with a reduced second heart sound

Evidence suggested that the presence of a systolic heart murmur plus a reduced second heart sound had good specificity for aortic stenosis confirmed by echocardiography. The recommendation specifies ejection systolic murmur because this, combined with a reduced second heart sound, is a classic indicator of aortic stenosis and is most often present in severe aortic stenosis. Although this was based on only a few studies, the committee agreed that people with these features should be referred for echocardiography. Because of the limited evidence identified, this recommendation was limited to those in whom heart valve disease was considered a possible explanation of these signs based on the nature of the murmur, family history, age or medical history.

Murmur with other symptoms or signs

Studies showed that echocardiography detected valve disease in a higher proportion of people with murmur plus other signs and symptoms (abnormal ECG, angina, breathlessness, peripheral oedema) than in people with murmur alone. That is, murmur plus other signs or symptoms had a higher specificity for echocardiography confirmed valve disease. Again, this was based on a few studies only, so the committee agreed that the nature of the murmur, family history, age or medical history should also suggest valve disease as a possibility.

Echocardiography

A peak aortic jet velocity more than 5 m/s was a risk factor for increased mortality (all‑cause and cardiac or cardiovascular) and sudden death in people with asymptomatic severe aortic stenosis who had not had a valve intervention. An aortic valve area less than 0.6 cm² was also associated with increased all‑cause mortality, both before and after valve intervention in adults with asymptomatic severe aortic stenosis.

A left ventricular ejection fraction (LVEF) less than 55% was the best marker of early myocardial decompensation, being linked to increased mortality in adults with asymptomatic severe aortic stenosis.

Raised B-type natriuretic peptide (BNP), particularly when 2 to 3 times the normal level, was a risk factor for all‑cause mortality, before and after valve intervention, for people with asymptomatic severe aortic stenosis and a preserved ejection fraction. The committee agreed that this would also apply to N-terminal proBNP (NT-proBNP), which is more widely used currently in the UK than BNP.

Some of these indicators were broadly in line with current practice and the experience of the committee. In addition, the evidence for increased mortality was strong, including for BNP. Therefore, the committee agreed that these indicators of poorer prognosis should prompt a discussion about the possible need for referral for intervention in people with asymptomatic severe aortic stenosis. Recommendations were limited to considering referral because the evidence was low to very low quality.

There was some evidence of increased mortality in people with asymptomatic severe aortic stenosis and a global longitudinal strain less than 14.7% or 15%, even when ejection fraction was preserved. However, there is some concern about reproducibility of measurements. The committee agreed that further research in this area would help to inform future guidance, so they made a recommendation for research.

Stress testing and stress echocardiography

Despite limitations in the quality of the evidence, the committee agreed that there was enough to show that symptoms revealed during exercise testing predict a poor outcome in people with asymptomatic severe aortic stenosis. They noted that some people may not report symptoms because they have adapted, for example, by reducing their activity. Exercise testing may therefore reveal these symptoms, which is an indication for intervention.

There was evidence from 2 studies, but with limitations, that no increase in valve area on dobutamine stress testing was associated with worse outcome in symptomatic low-flow low-gradient aortic stenosis. Point estimates and confidence intervals from both studies were consistent with this being a risk factor for poor outcome. Severe aortic stenosis is suggested if a person with low-gradient aortic stenosis has an LVEF less than 50% and a valve area less than 1 cm² at rest. Based on the evidence and the committee's experience, this can be confirmed on dobutamine stress testing if their valve area stays below 1 cm² and their mean gradient rises above 40 mmHg. This would therefore be an indication for intervention.

Cardiac MRI and cardiac CT

The evidence showed that a higher aortic valve calcium score measured by cardiac CT indicates a worse prognosis for people with aortic stenosis. This could be because it is an index of the severity of aortic stenosis or because it is a marker of more widespread vascular disease. This was supported by the knowledge and experience of the committee, who noted that a more calcified aortic valve is associated with more severe aortic stenosis. However, the mechanism of aortic stenosis in bicuspid aortic valves or in rheumatic disease is different, and cardiac CT would not be as relevant for monitoring valve calcium.

Most of the evidence suggested that myocardial fibrosis was associated with increased risk of a poor outcome in severe aortic stenosis. This was in line with the committee's experience that myocardial fibrosis in general, not only in aortic stenosis, is associated with a worse prognosis. Furthermore, myocardial fibrosis in people with severe aortic stenosis indicates early decompensation and the possible need for early intervention to stop progression, because mid‑wall fibrosis cannot be reversed or improved by intervention. The committee agreed that follow up should be enhanced and further assessment should be offered in those with mid‑wall fibrosis to check for symptoms and enable earlier aortic valve intervention to improve prognosis.

Echocardiography

The committee agreed that it is established practice to consider intervention for people with severe aortic regurgitation and reduced cardiac function. Severity is defined in line with British Society of Echocardiography guidelines. People with aortic regurgitation are often younger than people with other types of valve disease and benefit from timely intervention.

Evidence showed that when LVEF was less than 55%, the risk of cardiovascular mortality or heart failure after intervention was higher. End systolic diameter index (ESDI) is also a measure of systolic dysfunction. Evidence showed an increased risk of left ventricular systolic dysfunction or death when ESDI was more than 24 mm/m². The committee agreed that either of these 2 indicators of early myocardial decompensation should prompt discussion of possible intervention for asymptomatic severe aortic regurgitation. Recommendations were limited because of the evidence included being low to very low quality.

There was not enough evidence to include BNP level as an indicator for referral for intervention for people with asymptomatic severe aortic regurgitation. The committee agreed to make a recommendation for research to inform future practice.

Stress testing and stress echocardiography

No evidence was identified for stress testing and stress echocardiography in adults with asymptomatic severe aortic regurgitation. The committee agreed that further research could answer questions about when to intervene in this population. Therefore, they made a recommendation for research to identify prognostic factors in this population on stress testing.

Echocardiography

Evidence showed that an LVEF less than 60% was a risk factor for increased cardiac mortality after intervention for asymptomatic severe mitral regurgitation. An ESDI greater than 22 mm/m² was associated with onset of symptoms, left ventricular dysfunction, or death without intervention. This is broadly equivalent to the non‑indexed ESD threshold of 45 mm used in current practice. The committee agreed that either of these indicators of early myocardial decompensation should prompt consideration of an intervention for people with asymptomatic severe mitral regurgitation. Recommendations were limited to considering an intervention because the evidence was low to very low quality. The evidence on valve morphology, atrial fibrillation and pulmonary hypertension was not robust enough to include these as independent indicators for referral for intervention. However, the evidence suggested that these were associated with increased mortality, so the committee agreed their presence should be considered when discussing the possibility of intervention.

There was not enough evidence to include BNP level as an indicator for referral for intervention for people with asymptomatic severe mitral regurgitation. The committee agreed to make a recommendation for research to inform future practice.

Stress testing and stress echocardiography

Evidence from 2 studies showed that an increase of systolic pulmonary artery pressure (SPAP) to more than 60 mmHg on exercise was associated with worse outcomes in people with mitral regurgitation (asymptomatic or asymptomatic/mildly symptomatic, moderate or severe). This agreed with the committee's experience. Although there is limited evidence that in severe mitral regurgitation, intervening before symptoms develop results in better outcomes, the committee agreed that this may be better. Evidence from 1 study showed that SPAP above 60 mmHg on exercise was associated with symptoms developing during follow up.

There was not enough evidence for the committee to make a recommendation about symptomatic non‑severe mitral regurgitation. The single small study identified suggested that an increase in effective regurgitant orifice area of 13 mm² or more on exercise may indicate a worse outcome for this group. But the committee were not confident in this result and so made a recommendation for research to inform future practice.

Aortic stenosis when surgery is suitable

Evidence from 8 randomised controlled trials (RCTs) showed no large or clear differences for most outcomes between transcatheter aortic valve implantation (TAVI) and surgery for adults with non‑bicuspid aortic stenosis, including mortality outcomes and quality of life. However, a benefit of TAVI was identified for major bleeding and atrial fibrillation at 30 days, and length of hospital stay after the intervention. Absolute effects for other outcomes also suggested a benefit, but there was more uncertainty based on the confidence intervals. A harm of TAVI was identified for pacemaker implantation at 30 days. Although absolute effects also suggested a possible harm of TAVI in terms of mortality and rehospitalisation, the direction and size of the effect was much more uncertain for these outcomes and no clear difference between the 2 groups could be identified.

Only 1 study reported data beyond 5 years, but only for all‑cause mortality. The health economic model developed as part of the guideline looked for cost effectiveness over a lifetime. Therefore, it included evidence about the impact of complications in the long term, beyond 5 years, given the longer life expectancy for younger people with lower surgical risk. The results of the health economic model showed that TAVI at current prices was cost effective for people at high risk of surgery but not for people at low or intermediate risk. The committee agreed that if surgery is an option, it should be offered to those with severe aortic stenosis who are low or intermediate risk. Although all of the evidence identified was for non‑bicuspid aortic stenosis, it was agreed that the recommendation should also apply to bicuspid aortic stenosis, because suitability of surgery does not depend on the type of aortic stenosis. TAVI is also considered to be more difficult in bicuspid aortic stenosis.

Aortic stenosis when surgery is unsuitable

Evidence showed benefits for TAVI for people with inoperable non‑bicuspid severe aortic stenosis compared with pharmacological management at 1 to 5 years. These included benefits in all‑cause mortality, cardiac mortality, need for another intervention during follow up, and hospital admission. However, at 30 days TAVI was associated with increased mortality, stroke or transient ischaemic attack, major bleeding, and major vascular complications. The committee noted that TAVI is the only intervention available for some people with symptomatic severe aortic stenosis. They agreed that pharmacological management is not sufficient to help symptoms in severe aortic stenosis and for some aortic stenosis can be fatal without an intervention. TAVI can improve outcomes in many cases. Two UK‑based studies indicated that TAVI offers a good balance of benefits and costs in adults who cannot have surgery. The committee agreed to recommend TAVI, if suitable, for those with non‑bicuspid severe aortic stenosis if surgery is unsuitable. TAVI is the only option for this group and was deemed cost effective in this population.

All of the evidence identified was for non‑bicuspid aortic stenosis. TAVI is considered to be more difficult for bicuspid aortic stenosis and the committee could not extrapolate the evidence to cover this population.

Invasiveness of surgery

Evidence was identified from 14 RCTs comparing minimally invasive surgery for aortic valve replacement with standard surgery by median sternotomy across different aortic valve disease populations. Some harms of minimally invasive surgery were observed, and 1 health economic study suggested that minimally invasive surgery was less cost effective than median sternotomy. However, the RCTs were small and a small number of events were observed for many outcomes. The health economic study was limited for the same reasons because it was based on 1 of the RCTs and was limited to a 12‑month time-horizon. Although the committee agreed it is likely there would not be a large difference in outcomes after 12 months, this may be too short to draw conclusions about cost effectiveness over a lifetime. The committee highlighted that in their experience there was no difference between minimally invasive surgery and median sternotomy when done by those with expertise. The committee were also aware of certain advantages of minimally invasive surgery, for example, smaller incisions. The committee agreed not to limit the use of minimally invasive surgery and to recommend a choice with the decision based on patient characteristics and preferences. A lack of expertise in minimally invasive surgery locally should not be used as a reason for not performing a minimally invasive procedure and adults should be referred to a centre where there is expertise if this procedure is agreed as most suitable.

Despite no direct evidence for bicuspid aortic stenosis, aortic regurgitation (bicuspid or non‑bicuspid) and mixed aortic valve disease (aortic stenosis and regurgitation in the same person), the committee agreed that the type of aortic valve disease would not affect decisions about the invasiveness of surgery and the evidence could be extrapolated to any aortic valve disease.

TAVI for non-bicuspid aortic stenosis when surgery is unsuitable

The committee agreed that the use of TAVI is increasing, particularly when surgery is unsuitable and there are no other options for interventional procedures. It would be rare not to perform TAVI in these circumstances, but palliative care with pharmacological management is sometimes agreed. Therefore, the committee considered that the recommendation would represent a minimal change in practice and would not increase the number of TAVI procedures.

Surgery for aortic stenosis when this is suitable

The committee agreed that TAVI is usually reserved for when surgery is unsuitable or carries high risks of mortality. But data from the UK TAVI registry suggests that in recent years the procedure has been expanded to groups of people with lower surgical risk. The recommendation to offer TAVI to those with high surgical risk should have a moderate impact, as only 1.9% of surgeries are currently done in this group.

The recommendation to offer surgery instead of TAVI to those with intermediate and low surgical risk should increase the number of surgeries and reduce TAVIs in this group. This will ultimately improve NHS efficiency.

Minimally invasive surgery or median sternotomy for aortic valve disease

Data suggests that between 5% and 10% of surgical isolated aortic valve replacements are done by minimally invasive surgery. If the recommendation leads to an increase in the number of aortic valve replacements being done by minimally invasive surgery, this could represent an important change in practice. There may be no increase in the short term, as more training in these procedures will be needed, but only in the long term when more centres will have the expertise and capacity of offering minimally invasive surgery.

Return to recommendations

Repair or replacement when surgery is suitable

Evidence from 3 RCTs demonstrated few differences between surgical repair and surgical replacement in those with severe mitral regurgitation. (One study included both primary and secondary mitral regurgitation; the other 2 studies covered secondary mitral regurgitation only). The largest effect was for the need for reintervention for secondary mitral regurgitation, with fewer repeat interventions needed in the replacement group. Overall, the included evidence was limited; all studies were very small, with very few events reported for most outcomes and substantial uncertainty in the effects reported. Most outcomes were graded as very low quality. The lack of stronger evidence is likely to be because surgical repair has been preferred to replacement in mitral valve surgery for the past few decades. This was based on observational evidence and because randomising to repair or replacement in people for whom repair is suitable was thought to be unethical. Based on these limitations, the committee made recommendations reflecting current practice for those with severe mitral regurgitation requiring an intervention, with surgical repair recommended in those for whom it is suitable and replacement when repair is not suitable.

The committee noted that there are differences in the aetiology and treatment of primary and secondary mitral regurgitation. Although valve intervention is the next step for primary mitral regurgitation and an indication for intervention, for secondary mitral regurgitation the underlying heart failure is usually treated first. Therefore, the committee recommended that an intervention should be offered for severe primary mitral regurgitation and considered for secondary mitral regurgitation after optimisation of medical management.

Invasiveness of surgery

Evidence from 5 RCTs comparing minimally invasive surgery with median sternotomy for mitral regurgitation or mixed/unclear mitral valve disease demonstrated few differences. The studies were limited by small participant numbers and a small number of events for many reported outcomes. There was substantial uncertainty for most reported outcomes, a lack of long-term data for many outcomes, and most outcomes were graded as low or very low quality. Overall, when any larger differences were observed (for example, length of stay), these were for a benefit of minimally invasive procedures. A single health economic study suggested the cost of minimally invasive surgery was less per person than median sternotomy. However, the committee did not consider the included evidence to be strong enough to support recommending 1 type of surgery over the other. They agreed that median sternotomy and minimally invasive surgery should be options for those with mitral regurgitation requiring mitral valve surgery, with the decision being based on patient characteristics and patient preferences. A lack of expertise in minimally invasive surgery locally should not be used as a reason for not performing a minimally invasive procedure and patients should be referred to a centre where there is expertise if this procedure is agreed as most suitable.

Transcatheter mitral valve repair in primary mitral regurgitation when surgery is unsuitable

No clinical evidence was identified comparing transcatheter mitral valve repair with medical management for primary mitral regurgitation when surgery is not suitable. The committee noted that the lack of evidence may be because it is well established that medical management does not improve outcomes and transcatheter mitral valve repair is useful when surgery cannot be performed. One health economic study, based on a non‑randomised registry, reported that transcatheter repair was cost effective compared with medical management in those with severe mitral regurgitation when surgery was not suitable. This study had limitations because it included people with secondary mitral regurgitation and used data from a prospective, single‑arm registry with a control group obtained retrospectively. A second Japanese study on a mixed population with secondary and primary mitral regurgitation found transcatheter repair with the MitraClip device to be cost effective. This study had some limitations too as the relative treatment effects were informed from a propensity score matching study rather than an RCT.

A health economic model developed as part of this guideline did not find MitraClip to be cost effective for adults with secondary mitral regurgitation. However, the committee agreed that it was plausible that MitraClip would offer more benefits for people with primary mitral regurgitation because they are likely to have less residual disease affecting quality of life after the intervention. The committee agreed to recommend that transcatheter mitral valve repair should be considered for primary severe mitral regurgitation with symptoms when surgery is unsuitable.

Transcatheter mitral valve repair in secondary mitral regurgitation when surgery is unsuitable

Evidence was included from 3 RCTs comparing transcatheter mitral valve repair with medical management for secondary mitral regurgitation. Two of these were clearly in a population in which surgery was not suitable and covered the use of the MitraClip device; the third study covered a Carillon device rather than MitraClip and the population was unclear. Outcomes from all 3 studies were pooled if possible, in the clinical review, but the health economic modelling was limited to the population in which surgery was not suitable.

The clinical review highlighted uncertainty in the results for 3 outcomes (all‑cause mortality, cardiac mortality and onset/exacerbation of heart failure at 1 to 3 years or 2 to 3 years). Some studies demonstrated a benefit of transcatheter repair, some a harm (lack of benefit) and some no difference. One UK health economic study based on the results of the COAPT trial, which enrolled people with very severe secondary mitral regurgitation deemed inoperable, found that transcatheter edge-to-edge repair with MitraClip device had an incremental cost per quality-adjusted life year (QALY) of about £30,000.

A health economic model was developed as part of the guideline to investigate the cost effectiveness of using the MitraClip device when surgery is not suitable. The model demonstrated that transcatheter mitral valve repair had a low chance of being cost effective at £20,000 per QALY gained, with an incremental cost‑effectiveness ratio of £30,000 per QALY gained. These results are in line with the UK study identified in the literature review. The health economic model was largely based on results from the COAPT trial, which covered transcatheter mitral valve repair in severe secondary mitral regurgitation. This trial demonstrated substantial benefits over medical management alone when surgery was unsuitable. However, it was not considered to be cost effective at the current list price. For this reason, edge-to-edge mitral valve repair was not recommended over medical management.

Repair or replacement when surgery is suitable

Edge-to-edge repair is not widely available in the NHS. Therefore, this recommendation may lead to a change in practice and increase the amount of percutaneous mitral intervention in those for whom it is suitable.

Invasiveness of surgery

The recommendations are in line with current practice.

Transcatheter mitral valve repair in primary mitral regurgitation when surgery is unsuitable

Transcatheter mitral valve repair is rarely done for primary mitral regurgitation when an intervention is needed and surgery is unsuitable, so the recommendation may lead to a change in practice. This procedure has only recently been commissioned by the NHS and its use is likely to increase now based on this commissioning. The recommendation is unlikely to increase use much beyond this.

Transcatheter mitral valve repair in secondary mitral regurgitation when surgery is unsuitable

Transcatheter mitral valve repair is not currently used for secondary mitral regurgitation because it has not been commissioned by the NHS for this. The recommendation is unlikely to lead to a change in practice.

Return to recommendations

Anticoagulant and antiplatelet treatment after surgical biological valve replacement

Evidence from a population without atrial fibrillation demonstrated an increased risk of major bleeding with vitamin K antagonist compared with single antiplatelet therapy (aspirin). No clear reduction in mortality or thromboembolic events was observed with vitamin K antagonist. Therefore, the committee agreed that anticoagulation should not be offered after surgical biological valve replacement unless there are other indications for anticoagulation. This covers both vitamin K antagonists and direct‑acting oral anticoagulants (DOACs) because there was no evidence to show that DOACs are safe. One small study in people with atrial fibrillation suggested there may be no clear differences in outcomes between DOACs and vitamin K antagonists, and it is not common practice to use DOACs for this group. The committee agreed that if there is already an indication for anticoagulation or antiplatelet therapy, for example, because of atrial fibrillation, the existing NICE guidelines for these indications should be followed.

Despite 1 study demonstrating a potential reduction in arterial thromboembolic events and vascular mortality with combined anticoagulant and antiplatelet therapy compared with anticoagulant therapy alone after surgical biological valve replacement, there was uncertainty around this result. This uncertainty, combined with further study limitations, including issues with the target international normalised ratio used and the selective population, meant that the study could not be used to inform general recommendations for surgical biological valve replacement.

There was a lack of evidence comparing anticoagulant or antiplatelet therapy with no treatment after surgical biological valve replacement, so the committee made a recommendation for research. They made another recommendation for research to investigate the long-term effect of anticoagulant or antithrombotic therapy on valve function and outcomes after biological valve replacement because no long-term data was available.

Single antiplatelet therapy after TAVI

Evidence from 4 studies demonstrated a clinically important benefit of single antiplatelet therapy (aspirin) compared with dual antiplatelet therapy in reducing major and minor bleeding in the short-to-medium term. Based on this, the committee agreed that single rather than dual antiplatelet therapy should be considered after TAVI. As aspirin is used in practice, and this was used in all of the studies, aspirin was recommended, with clopidogrel specified as the alternative if aspirin was not tolerated.

The committee were also aware of observational evidence that antiplatelets reduced the risk of valve thrombosis and improved valve durability over the long term. There was also evidence from 1 study demonstrating harms of DOACs compared with single antiplatelet therapy for most reported outcomes, including mortality, bleeding and withdrawal because of adverse events. This further supported the recommendation for single antiplatelet therapy. Because of the lack of evidence comparing anticoagulant and antiplatelet therapy with no treatment after TAVI, the committee made a recommendation for research.

Valve repair

No evidence was identified comparing different anticoagulant and antiplatelet treatments in adults who have had valve repair. The committee made a recommendation for research comparing anticoagulant and antiplatelet treatments with placebo after valve repair.

Anticoagulant and antiplatelet treatment after surgical biological valve replacement

Practice is currently variable, with some centres offering vitamin K antagonists after surgical biological valve replacement. Therefore, the recommendation will lead to a change in practice in some centres.

Single antiplatelet therapy after TAVI

It is unusual for people not to receive at least single antiplatelet therapy after TAVI and many people receive dual antiplatelet therapy. The recommendation was not thought to represent a change in practice in terms of the number of people who receive some form of antiplatelet therapy after a transcatheter procedure.

Return to recommendations