Sampling considerations

Once a sample is obtained, the forceps remain closed until removed from the body, then are opened over a sterile field and the sample carefully removed by an assistant. A small needle can be used as care must be taken not to crush the sample with forceps nor cut it as these processes can affect pathological analysis. An ideal sample is 3–4 mm² volume, usually pink in colour and sinks in formalin. A grey or white sample that floats might indicate scar tissue or tricuspid valve apparatus.

The samples should be transferred to room temperature 10% neutral-buffered formalin. The exact number of samples obtained will depend on what tests are to be requested. Four to five samples are sufficient for light microscopy but tests such as transmission electron microscopy (TEM) or viral genomic analysis will require more samples.21 Samples for TEM require fixing in 4% glutaraldehyde. This technique may be of use when testing for anthracycline toxicity, infiltrative diseases such as amyloidosis, viral myocarditis and glycogen or lysosomal storage disorders.1 Viral PCR tests require a sample to be flash-frozen with liquid nitrogen and stored appropriately.22

False-negative results are possible, particularly with multifocial or microfocal localised diseases.23 Therefore, the imaging modality chosen should guide sampling of different areas of myocardium as repeated sampling of the same region of tissue increases the likelihood of a false negative and of a complication.24 Information about distribution of abnormal areas of myocardium obtained from CMR may be used to guide biopsy.20 Directly targeted EMB is uncommon, but the addition of intracardiac electrogram (EGM) mapping to guide biopsy was shown to be of benefit in a small group of patients.25 Targeting endomyocardium that exhibits a low amplitude or abnormal EGM trace resulted in significant improvements in diagnostic yield but the technique is more time-consuming and requires a separate electrophysiology catheter.26

In selected cases, biopsy forceps may be used therapeutically, such as the excision of small benign cardiac tumours.27 In patients who are receiving anticoagulation, no guidance exists on a safe international normalised ratio (INR) to undertake EMB, but many centres advise an INR of below 1.5–2.

Complications

Complication rates quoted in published case series vary from 0% to 3.3%,5 8 28–30 but considering many studies feature figures obtained over a very long timeframe and a clear improvement in safety has been demonstrated over the years, the overall complication rate at present is around 1% or lower.8 Rates for specific complications are not known as most published reports come from individual cases or series, however a survey of 6739 patients returned an overall complication rate of 1.17% and a perforation rate of 0.42%.29

Complications include vasovagal syncope, ventricular perforation, pericardial tamponade, heart block, supraventricular and ventricular arrhythmias, accidental arterial puncture, pneumothorax, vascular damage, nerve damage, pulmonary embolism, coronary-cameral fistula formation, bleeding complications and damage to the tricuspid valve, especially in patients undergoing repeated biopsies (figure 3, online supplementary video 1).

Following RV EMB, ongoing monitoring is not normally necessary, but if a complication is anticipated, the patient should remain on a cardiac monitor postprocedure and serial echocardiography should be considered. Following an internal jugular vein (IJV) approach, patients can mobilise immediately following 5 min of pressure on the jugular access puncture, but following femoral vein puncture they should take 1 hour bedrest.

After routine LV EMB with arterial puncture, the patient should remain monitored for at least 30 min. Standard protocols for postprocedure management of radial or femoral artery puncture should be followed. Four weeks’ aspirin after LV EMB has been advocated in patients not formally anticoagulated.17 An example consent form is included in online supplementary material 1.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

A haemo-pneumopericardium caused as a complication of endomyocardial biopsy. White arrows indicate the thin pericardium and black arrows a fluid level produced by blood within the pericardial space.

Perforations are rare and tend to occur less frequently in transplanted patients due to the (usually) minimal pericardial space and the myocardium normally being healthy, as opposed to other indications for EMB, where the heart is affected by a disease process. For this reason, some recommend echocardiography routinely be employed in conjunction with fluoroscopy outside the transplanted population, especially if sampling from the RV free wall (which would normally be avoided in a cardiac allograft).31 The 2007 Joint Scientific Statement stipulates that if perforation is suspected, echocardiography should be used to assess for a pericardial effusion prior to removing the venous sheath and facilities should be available to perform rapid pericardiocentesis or surgical drainage, which may limit EMB to centres with cardiothoracic surgical support.1 Epicardial fat is noted in biopsy samples occasionally, with no suggestion radiologically or symptomatically that a significant perforation occurred during biopsy, so it is possible perforation rates are higher than suspected, but many are not clinically relevant.

Indications

Although EMB has been shown to provide additional information in multiple diagnoses such as dilated cardiomyopathy32 or Fabry’s disease,33 this may not translate to altering the clinical course or management of the patient. Therefore, the 2007 Joint Scientific Statement has been drawn up with an emphasis on whether EMB will provide clinically useful information.

Cardiac transplant rejection

EMB is most commonly performed as routine surveillance for rejection of a transplanted heart, which is not addressed in the more general 2007 statement and worth considering separately. Both acute cellular (ACR) and antibody-mediated rejection (AMR) can be identified using histopathological analysis, although clinical AMR requires additional information including donor-specific anti-HLA antibodies and functional information (echocardiography or CMR). Transplanted patients will have regular surveillance biopsies, according to local guidelines but normally weekly to begin with and becoming less frequent as the months pass. A typical schedule would advise EMB at weeks 1, 2, 3, 4, 6, 8, 12, 16, 22, 28, 36, 44 and 52. As bioptomes are preshaped in the same way, patients risk having the same area of myocardium sampled across multiple biopsies so operators may wish to alter the curvature of the bioptome manually to decrease the chance of sampling previous biopsy sites.

The two principal markers of cellular rejection are lymphocytic infiltrate and myocyte injury. The International Society for Heart and Lung Transplantation biopsy grading system is used to characterise acute cellular rejection (table 1).34

AlloMap molecular expression testing has a very high negative predictive value for ACR, with a negative predictive value range at 6 months found to be 98.1%–100%.35 EMB remains the gold standard investigation to test for presence of rejection but CMR has been shown to have a high negative predictive value and may have a role in non-transplant centres as a screening tool prior to biopsy. A recent analysis of 73 cases found that CMR achieved a 52% positive predictive value and 98% negative predictive value for a positive EMB. Sensitivity was 93% and specificity 78%.36

Use outside the transplant population

In practice, the diagnostic yield of EMB is only one consideration in developing recommendations for specific scenarios. Another is the availability of an effective therapy for the condition being tested for. As such, the AHA/ACC/ESC Scientific Statement devised its advice based on clinical situations, where the diagnostic and prognostic information that could potentially be obtained from EMB outweighs the risks of the procedure.10

Leading on from the 2007 statement, in 2013 an expert panel suggested broadening the indications for EMB in the diagnosis of myocarditis, beyond the specific clinical scenarios detailed in the AHA/ACC/ESC Scientific Statement.2 The ESC Working Group on Myocardial and Pericardial Disease highlighted the fact that the 2007 Scientific Statement did not use diagnostic and prognostic information from modern immunohistochemical and viral genomic testing2; only the Dallas histopathological criteria were used.37 While these newer techniques improve diagnostic sensitivity, there has not been widespread uptake due to a combination of cost and a paucity of effective treatments for myocarditis.38 39 The AHA/ACC/ESC Writing Group acknowledge very few centres are capable of handling and interpreting the samples adequately. However, by using immunohistochemical and viral genomic testing to identify virus-negative inflammatory myocarditis, immunosuppression is being increasingly used with encouraging preliminary results.38

The Dallas criteria themselves have been criticised for reasons including a lack of quantification of the inflammatory infiltrate, failing to quantify fibrosis or myocyte damage and a lack of an aetiological characterisation.40

The AHA/ACC/ESC Scientific Statement’s recommendations are summarised below, using the same clinical scenarios figure 4.1

• Download figure
• Open in new tab
• Download powerpoint

Figure 4

Flow diagram illustrating clinical scenarios described in AHA/ACC/ESC Scientific Statement(1). Class of recommendation I: conditions for which there is evidence or there is general agreement that a given procedure is beneficial, useful, and effective; class II describes conditions where there conflicting evidence; class IIA: conditions for which the weight of evidence/opinion is in favour of usefulness/efficacy, class IIB: conditions for which usefulness/efficacy is less well established by evidence/opinion. LV,left ventricle; AV; atrioventricular; HCM, hypertrophic cardiomyopathy,; ARVC, arrhythmogenic right ventricular cardiomyopathy.

New-onset fulminant heart failure

Patients who present with unexplained heart failure of <2 weeks’ duration, causing haemodynamic compromise, should be investigated with EMB (class of recommendation 1, level of evidence B).41 This can offer useful diagnostic and prognostic information. Giant cell myocarditis (GCM), necrotising eosinophilic myocarditis and lymphocytic myocarditis can all present in this manner, but whereas the first two carry a poor prognosis, lymphocytic myocarditis normally has a far more favourable outcome 10

Establishing a diagnosis of GCM would allow administration of immunosuppressive therapy, which has demonstrated an improvement in clinical outcome,42 although it remains a frequently fatal condition. If a patient requires mechanical circulatory support, patients with GCM should receive biventricular support as the right ventricle is likely to be involved and cardiac transplantation should be considered early in the disease course.

Recent onset ventricular arrhythmias, atrioventricular block or refractory heart failure

Patients presenting with new, unexplained heart failure of 2 weeks to 3 months’ duration, associated with a dilated LV, ventricular arrhythmias, early  (AV) block (Mobitz II or complete heart block) or simply failure to respond to conventional heart failure therapy should undergo EMB (class of recommendation 1, level of evidence B).1 Again, these features are suggestive of GCM and can be useful in discrimination from other causes of dilated cardiomyopathy. Fifteen per cent of cases of GCM exhibit ventricular tachycardia at presentation, and AV block is seen in 5%. GCM is often associated with autoimmune disorders or thymoma and may not respond to conventional heart failure treatment.43

Heart failure associated with late development of ventricular arrhythmias, AV block or unresponsive to conventional therapy

Patients presenting with unexplained heart failure of >3 months’ duration associated with a dilated LV and new ventricular arrhythmias, Mobitz II or complete heart block or failure to respond to conventional heart failure therapy within 1–2 weeks should undergo EMB (class of recommendation 2a, level of evidence C). This clinical picture is consistent with a more indolent GCM, granulomatous myocarditis or sarcoidosis. As sarcoidosis can frequently be diagnosed without invasive techniques, EMB would be reserved for occasions when the diagnosis is unclear. Sarcoidosis tends to have a better prognosis than GCM and may respond to steroid therapy, in contrast to combination immunosuppression for GCM.44 Both have similar rates of ventricular arrhythmias and AV block, but sarcoidosis normally has a more chronic course.

Eosinophilia

Despite the improvements in non-invasive imaging techniques in the diagnosis of hypereosinophilic syndrome, EMB remains the gold standard and has utility for diagnosis and serial EMBs can be used to assess response to therapy.45 Patients presenting with a dilated cardiomyopathy in association with suspected allergic reaction and significant eosinophilia should undergo EMB, irrespective of duration of illness (class of recommendation 2a, level of evidence C). Hypersensitivity myocarditis is caused by an allergic reaction and can present in a variety of ways. EMB is indicated if drug hypersensitivity is suspected, to differentiate from granulomatous myocarditis, necrotising eosinophilic myocarditis or GCM, as management might be affected by withdrawing the offending drug and initiating glucocorticoid therapy.45

Restrictive cardiomyopathy

Some causes of restrictive cardiomyopathy are potentially reversible, including haemochromatosis or amyloidosis. Therefore, EMB may be indicated in diagnosis (class of recommendation 2a, level of evidence C), however these conditions are often diagnosed non-invasively with characteristic changes seen on echocardiography or CMR.46

Selected cardiac tumours (except myxomas)

In keeping with the aforementioned indications for EMB, advances in cardiac imaging mean that characterisation of many intracardiac masses can be accomplished successfully with non-invasive techniques. In many circumstances, cardiac surgery is the treatment of choice and of course surgical resection of a mass allows detailed analysis. However, in selected cases, diagnosis might be achieved via EMB and render surgery unnecessary (eg, lymphoma or infiltrative masses where surgical resection may not be feasible). If echo features are suggestive of myxoma, EMB is not recommended as myxomata are associated with a high risk of embolisation.1

In the few cases where a diagnosis cannot be reached non-invasively, and in those where the indication for surgery is unclear, EMB might be considered (class of recommendation 2a, level of evidence C). Transthoracic, transoesophageal or intracardiac echocardiography should be employed in addition to fluoroscopy.47

Specific conditions

Anthracycline toxicity, hypertrophic cardiomyopathy and arrhythmogenic right ventricular cardiomyopathy (ARVC) are all normally diagnosed with non-invasive imaging. EMB may be indicated on rare occasions where diagnosis has proved elusive. EMB in cases of suspected ARVC is uncommon and should be guided by imaging such as CMR to identify regions to biopsy. The right ventricular outflow tract (RVOT) has been found to be the most useful area, although the finding of myocardial fat is often non-specific.48

A recent hypothesis-generating study analysed histopathological data obtained from EMB in patients with systemic sclerosis and cardiac involvement. The degree of fibrosis seen on EMB appeared to correlate with a poorer prognosis in terms of major adverse cardiovascular events (defined as cardiovascular death, an arrhythmic event and heart failure-related hospital admission).49

The emerging recognition of patients presenting with myocardial infarction with non-obstructive coronary arteries (MINOCA) has led to previously missed diagnoses being discovered, most commonly using CMR, which provides a diagnosis in 74% of cases.50 The most common differential within this group is myocarditis presenting with features of an acute coronary syndrome. In the 26% of cases with no abnormality detected, EMB may have a role but it does not appear in any guidelines as of yet and the largest meta-analysis of MINOCA trials does discuss this indication.50

Efficacy

A recent single-centre, retrospective analysis of 851 patients sought to build on the advice given by the 2007 Scientific Statement and specifically attempted to ascertain information relevant to the clinical scenarios outlined above.51 Overall, 25.5% of EMBs provided a diagnosis and 22.6% of EMBs changed clinical course. However, the most common indication (33.6% of cases) was unexplained DCM with evidence of restrictive myocarditis; 12.8% of cases were patients presenting with acute onset fulminant heart failure and when analysing this cohort, 39% of EMBs were diagnostic and 27.5% led to a change in management, confirming the utility of biopsy in this critically unwell subset.51 Viral genomic analysis was not performed in most cases, although when this has been examined in other studies, approximately half of biopsies demonstrated an absence of any viral genome.32

Future uses

The majority of literature pertaining to EMB is based on light microscopy but with increasing use of genomic or proteomic testing the role of EMB may yet expand, especially in the diagnosis of dilated cardiomyopathy. The improved targeting of immunosuppressive therapies and using more advanced immunohistochemical analysis of specimens, may lead to a more prominent role for EMB in the management of acute and chronic myocarditis. Diagnostic sensitivity may improve, alongside new understanding of disease pathophysiology.31 It remains a key research tool as the only way to directly obtain cardiac tissue in vivo apart from surgery and in particular is used to validate non-invasive imaging techniques.