Article Text
Abstract
The endomyocardial biopsy(EMB) is a valuable tool that is inadequately utilised, except in monitoring orthotopic, homograft, heart grafts. Performed on a regular, programmed schedule, or on an emergent basis when needed, the EMB is the gold standard for monitoring graft function ( with regard to cellular rejection), often before clinical symptoms develop. This paper addresses these points and discusses handling of tissues and some studies for possible antibody mediated rejection when the lack of morphologic features of cellular rejection does not fit with the patients clinical presentation, days or months after the surgical procedure. In the hands of a skilled operator the EMB procedure is relatively painless and free of signicicant complications.
Statistics from Altmetric.com
Endomyocardial biopsy (EMB) is the gold standard for monitoring the graft heart following orthotopic heart transplantation. An established modality for the diagnosis of cardiac disease, the EMB has gained increasing “popularity” as the standard for monitoring patients post-cardiac transplant.1 2 At most centres, patients undergo regularly scheduled biopsies, performed every week initially and then every three months for the rest of the patient’s life. The analysis and diagnosis of EMB based graft rejection is critical for the appropriate and timely modulation of immunosuppression protocols.
Appropriate interpretation of an EMB therefore needs a good understanding and awareness of biopsy findings, grading systems, artefacts and pitfalls so that a complete and accurate report may be formulated. The biopsy report, critical for the management of these patients, needs an organised approach. This paper will discuss the approach to the post-transplant EMB, its artefacts and the diagnosis of graft rejection. A review of the English language literature was performed using PubMed (US National Library of Medicine) to identify previously published papers on endomyocardial biopsies. The following search terms were used: heart transplantation, endomyocardial biopsy, heart biopsy.
History
The early EMBs used the open approach with biopsy of the epicardial surface of the heart. These were followed by the use of the Vim–Silverman needle through a limited thoracotomy or transthoracic approach.1 2 They were associated with complications such as cardiac tamponade, and their use was somewhat restricted. In 1962, Sakakibara and Konno introduced a biopsy catheter or “bioptome” for the sampling of the endomyocardium through an endovascular approach rather than the previous transpericardial and transmural approach1 3 4 The benefit of this bioptome was the ability to obtain adequate EMB samples with fewer complications. The new transcatheter approach was followed by the introduction of the Stanford bioptome, both of which utilised the transvascular approach, under fluoroscopic control. While access initially was through arteries with biopsy of the left side of the interventricular septum, the technique soon evolved to a transjugular approach through the neck veins, excluding the necessity for a “cutdown”. Today, virtually all biopsies are performed using this transvenous approach, via the neck; access through the jugular (or subclavian) vein gives good tissue samples. Today’s catheters have been modified further so that the catheter lies inside a sheath, thereby allowing greater flexibility and manoeuvrability in addition to the ability to perform repeated biopsies at the same sitting using the same venous puncture.5 6
The EMB today has widespread diagnostic use; however its regular use is most frequently for the monitoring of orthotopic heart transplantation. EMBs are performed sequentially to monitor orthotopic heart transplants and have played a major role in bringing cardiovascular pathology into the realm of contemporary surgical pathology.
Indications for endomyocardial biopsy
There are many clinical indications for endomyocardial biopsies. These can be divided broadly into diagnostic and allograft heart monitoring biopsies (table 1). Diagnostic biopsies are performed primarily for the confirmation of clinically suspected entities such as myocarditis, dilated cardiomyopathy and rare or unusual forms of cardiomyopathies, such as the storage and genetically determined cardiomyopathies or for the determination of the aetiology of clinically diagnosed heart failure. The endomyocardial biopsy also helps to differentiate between restrictive and infiltrative myocardial diseases.7 8 It is critical to differentiate hypertropic cardiomyopathy from a myocarditis or dilated cardiomyopathy since the management is quite different. Endomyocardial biopsies are helpful in the evaluation of other inflammatory cardiac conditions, such as drug toxicities and hypersensitivity/inflammatory reactions.7 Of the many new drugs brought into use each year, few have had cardiac studies associated with them. Today, some of the new anticancer agents such as Herceptin (trastuzumab; Genentech, San Francisco, California, USA) have been reported to show cardiotoxic effects. Doxorubicin, a classic chemotherapeutic agent used for treating many tumours, has known cardiotoxicity. The morphological basis of the toxicity of doxorubicin was established using EMB.
The most significant indication for the EMB today is for the monitoring of cardiac allograft rejection; this is the focus of this paper. EMB remains the gold standard for monitoring undiagnosed heart failure, intractable tachyarrhythmias and allograft rejection. It is also used for monitoring of side effects of new/experimental drugs.
Biopsy protocols: The EMB protocols for monitoring graft hearts vary from institution to institution. Most institutions have in-house post-transplant immunosuppression and monitoring protocols. At our institution this is comprised of weekly biopsies in the first month, biweekly biopsies for two months after transplant and quarterly biopsies subsequently, for the lifetime of the individual. Additional biopsies are undertaken when indicated by the sudden onset of heart failure or worsening of symptoms, or when a patient has significant rejection which is followed up by a repeat biopsy one or two weeks later. Depending on the results of this repeat or follow-up biopsy, subsequent biopsies may have to be undertaken before the scheduled intervals. In some instances, such as non-compliance with immune suppressive medications, significant rejection may develop quite rapidly and necessitate an unscheduled EMB.
Handling of endomyocardial biopsies
The tissue for good interpretation of an endomyocardial biopsy should be comprised of at least four, preferably more, “good” pieces of tissue (table 2). “Good” refers to the fact that the tissue should be comprised entirely of endomyocardium and not of too much scar tissue (from previous biopsies) or of thrombus (fresh, organising or organised), from previous biopsy sites (fig 1). The entire sample is submitted for histopathology and the processed tissues are nearly entirely sectioned. Multiple sections, going virtually through the entire block, are cut and mounted on glass slides (15 slides with ribbons of two or three sections on each).1 8 9 10 At our institution, alternate slides are stained with H&E. Slides 6 and 15 may be stained with connective tissue stains. The unstained slides are all saved. These slides are saved so that if there is difficulty in interpreting the biopsy, the unstained sections may be stained and further information obtained from appropriate adjacent sections. This may be essential in the cases of suspected vasculitis, Quilty lesions and occasionally in grading rejection. The connective tissue stains may at times be very helpful in deciding between previous endomyocardial biopsy sites related inflammatory infiltrate and the presence of rejection. One of these saved sections is now used for the C4d paraffin stain on all cases.
Technical considerations
The easy access to venous return to the heart from the head and neck makes access to the right ventricle an attractive approach and is the favoured approach today, in virtually all centres. A catheter is introduced through the jugular vein into the superior vena cava, the right atrium, and tricuspid valve into the right ventricle.4 5 The interventricular septum is the preferred biopsy site, since this tissue is thicker than the adjacent ventricular free wall and because of its continuity with the left ventricle. Its thickness and location make it ideal for biopsies and minimise the chances of perforation and catastrophic events. A significant problem associated with this develops over a period of time and is related to the fact that the bioptome floats into the right ventricle and towards the septum; it often approaches more or less the same area every time. The biopsy site therefore can and does end up showing progressively greater scarring. This makes interpretation difficult, and makes diagnosis difficult because the traumatised myocardium has an inflammatory infiltrate; it has extensive scar tissue, and at times the bioptome brings up only scar tissue. In addition, this tougher tissue means that the patient has a greater degree of discomfort, if not pain, associated with each biopsy. Today the EMB is treated as a routine procedure; it must however be borne in mind that it is an invasive procedure and can be associated with complications. Fortunately significant complications are rare in the hands of most experienced operators.1 4 6
Artefacts
Before beginning to interpret the biopsy, one must be aware of artefacts that can be seen and that can affect interpretation. These artefacts may be classified as follows:
(1) Those related to the procedure, that is biopsy artefact.
(2) Those related to the pre-transplant period.
(3) Those related to the biopsies themselves.
The procedure related artefacts are also relatable to the technique of handling the tissue once it is removed (fig 2). The biopsy itself, that is the procedure, results in the tissue, especially the muscle fibres at the edges or the torn edges, hypercontracting and leading to contraction band like artefact (fig 2E,G). If the tissue is left sitting on a wet sponge, for an excessive period of time, it leads to misleading findings: “separation of fibres”, interstitial oedema and other changes, which are difficult to interpret. The tissues must at all times, especially before paraffin embedding, be handled gently to avoid mechanical injury such as crush artefact and “waist” artefact due to compression of the tissue at its centre.1 7
The pre-biopsy artefacts are related to the ischaemic interval between explanting the heart from the donor and implanting the graft into the recipient. This change is generally not identifiable at the time of transplant but can be seen in the first few biopsies. This depends on the length of the graft anoxia period, and the longer the period, the greater the chances of this artefact. This is comprised of small subendocardial foci of muscle fibre loss with no good evidence of inflammation. This is akin to subendocardial ischaemic damage and is related to subendocardial ischaemia.
Biopsy artefact: The third category of artefact is related to previous biopsy damage (fig 2A–G). These changes may be early or late. Early changes are comprised of the presence of variable amounts of thrombus on the biopsy site and are followed by evidence of organisation of this thrombus (fig 2C–D). The thrombus therefore shows an ingrowth of fibroblasts and endothelial cells and slowly gets organised with deposition of collagen and later significant fibrosis. The endocardium shows progressive fibrosis and thickening while the subjacent myocardium shows increasing interstitial fibrosis and changes in muscle fibre size, with the fibres becoming smaller rather than larger and an appearance of small areas of muscle fibre disarray (fig 2D–F). The disarray is seen only at the immediate biopsy site and is related to the increased fibrosis. This myocardium also shows evidence of mononuclear inflammatory cell infiltrates, comprised of lymphocytes and macrophages (fig 2F). These mononuclear cells can persist for weeks and months (fig 2F).1 7 Occasionally, it may be difficult to differentiate rejection with muscle fibre damage from previous biopsy sites. One must keep in mind that the previous biopsy site is associated with increased fibrosis. The previous biopsy site related damage is generally close to the endocardial surface, in contiguity with the endocardium at some point; a review of adjoining sections may be helpful as may the use of connective tissue stains to demonstrate this.
Endocardial inflammatory cell infiltrates or Quilty lesions: The biopsy often shows variable infiltrates of endocardial mononuclear cells (in 10–20% of cases). These are usually flat lesions but may grow “out” into the endocardium and appear as nodular lesions confined to the endocardium (Quilty A) or into (deep) the myocardium (Quilty B).11 The former largely grow into the lumen of the chamber (fig 3A). A more worrisome lesion is the similar lesion which grows into the subjacent myocardium—the Quilty B lesion (fig 3B). While the Quilty A need not be associated with a subendocardial component, the subendocardial form (Quilty B) is invariably associated with an endocardial involvement (fig 4). The concern about this Quilty B stems from the difficulty in differentiating this lesion form a significant grade of rejection (fig 4). These infiltrates are comprised mainly of T lymphocytes and a few B cells and have a significant degree of vascularity. Of unknown aetiology, they seem to be not significant and not related to acute rejection. In differentiating a nodular aggregate of mononuclear cells in the biopsy, it is often helpful to look at adjacent sections and look for a connection with the overlying endocardium. Quilty B lesions and Grade 2 lesions have in the past been misdiagnosed, and likely still are! The concern that these lesions may represent a post-transplant lymphoproliferative disorder (PTLD) is wrong since the PTLD are predominantly B-cell lesions while the Quilty are T-cell lesions.12 13
“Occluded arterioles”: Some arterioles may look as though they are occluded and possibly thrombosed, suggesting a graft vasculopathy, with a cellular mass in the lumen (fig 5). In fact, an elastic trichrome stain would show that the lumen is filled with prolapsed intima from adjacent sections. If necessary, an immunohistochemical stain for endothelial cells may be undertaken (such as CD31). This lesion must not be construed as an occluded vessel or graft vasculopathy.
The approach to post-transplant, endomyocardial biopsies is similar to that of endomyocardial biopsy interpretation in general. One should therefore start in a similar systematic manner, though the emphasis here is twofold:
To find interstitial mononuclear cells, primarily lymphocytes.
To find evidence of muscle fibre damage.
Once these are noted, the next attempt is to try and grade the infiltrate and damage. This is done by noting the number of parts of the biopsy tissue that show similar damage and the extent of this damage. For this, there are now grading systems that use this approach and will help grade the rejection (tables 3 and 4).14 The old system of grading had four grades: Grade 1A and 1B; Grade 2; Grade 3A and Grade 3B; and Grade 4 (table 3). The new system is primarily a grading of rejection with emphasis on whether or not it is significant and needs to be treated or not. Grades 1 and 2 do not need treatment, while Grade 3 does need treatment (table 4).
Histochemistry stains performed
The routine stains are H&E and elastic trichrome (table 6). Occasionally, one may need immunohistochemistry for lymphocytes and macrophages. The use of the C4d stains for possible identification of antibody mediated rejection (AMR) or non-cellular rejection is still somewhat debatable. Some cases may need additional special stains based on the preoperative diagnosis, that is, the reason for the transplant. Recurrence of lesions has been noted in amyloid cardiomyopathy, in sarcoid cardiomyopathy, and in Chagas disease related cardiomyopathy.1 7
Stains for identification of microorganisms may be needed and these pertain especially to the identification of cytomegalovirus (CMV) and perhaps other viruses.
In a very occasional biopsy, it may be necessary to differentiate myocarditis from rejection. The myocarditis is usually associated with a significant infiltrate of plasma cells, or if associated with a hypersensitivity reaction, with significant numbers of eosinophils in addition to other cells.
Identification of rejection
The immune response: Implantation of a graft heart with its vasculature connected starts a complex process involving several mechanisms, which would culminate in graft rejection. The initiation of immune suppression allows the graft to survive and makes medication compliance critical for the continued survival of the new heart. Immunological barriers remain a critical concern in transplantation. For a detailed review of alloimmune reactions and responses, there are many excellent reviews.15 16 17 Briefly though, T lymphocyte activation plays a pivotal role in the hosts response to the allograft heart. The response starts with the activation of the antigen presenting cells of the host and the donor and can then follow several signalling pathways. B cells get activated when the antigens react with the B cell receptors in lymphoid tissues, mediated by B and T cells. Inflammatory mediators and the complement pathway are activated and are essential to the immune response. Cell mediated rejection is characterised by the presence of mononuclear (inflammatory) cells in the myocardial interstitium. These cells usually are the T lymphocytes, plasma cells and activated macrophages.
There is now increasing awareness that graft rejection may be cell and antibody mediated.16 17 While there is concern about the occurrence and significance of antibody mediated rejection (AMR), there is no consensus around its recognition or diagnosis. For demonstration of AMR, it is necessary to show capillary injury, endothelial cell swelling and aggregation of intravascular macrophages. For this demonstration of capillary endothelial damage, positive staining for C4d or C3d fragments of complement by endothelial cells is critical. The stains can now be performed on paraffin embedded tissues.
Immune suppression has advanced significantly in the last 20–30 years. This has been made possible by the effective blocking of lymphocyte activation pathways.14 15 16 17 Cyclosporin has been the centrepiece of all immune suppression protocols, and has been associated with significant improvement in graft survival. There are many newer agents and several immune suppression protocols. The use of tacrolimus has helped make more improvements in survival.
Interpretation of the biopsy
The aim of the EMB, post-heart transplantation is to: identify rejection; determine the presence of infection, if any; and determine the development of any post-transplant neoplasia or post-transplant lymphoproliferative disease. Graft vasculopathy can develop at any time, however in most patients, it is seen several months post-heart transplantation. The identification of rejection is based on the presence of mononuclear inflammatory cells in the interstitium, away from previous biopsy sites and not associated with Quilty lesions11 12 (table 5).
As noted above, interpretation of the biopsy must start at the endocardium. All sections must be examined since an inflammatory infiltrate may be present in the first section or the last (15th) section and in between. It is therefore essential to review every stained section, in the same complete manner:
Examine the endocardium.
Examine the myocardial fibres
Examine the interstitium, including the vessels in it.
The endocardium usually shows fibrosis which may be minimal, mild or more pronounced and is most commonly related to healed biopsy sites. The endocardium may show a scattered infiltrate of mononuclear cells or may show aggregates of mononuclear cells; the latter likely represent a Quilty A lesion (fig 3A,B).
The interstitium: Acute rejection is characterised by an inflammatory infiltrate and associated with variable myocyte damage. The inflammatory infiltrate may also be related to previous biopsy sites or may be related to cellular rejection. The infiltrate is comprised predominantly of lymphocytes, and may have macrophages mixed with it. In the most severe grades of rejection the infiltrate is polymorphic and is associated with neutrophils and some eosinophils. Severe rejection may also be associated with interstitial oedema and with red blood cells in the interstitium (not related to mechanical vascular trauma). The T cells seen in this rejection reaction are a mix of CD4 and CD8 cells. Counting these cells is of no help in determining the significance or grade of the rejection.
Myocyte damage seen with this infiltrate is seen as scalloping of the sarcolemma of myofibres or the apparent, progressive loss of muscle fibre “content” and diminution in size, so that the damaged myofibres look smaller than their neighbours. Note must be made of the types of cells comprising the infiltrate. Most frequently the cells are mononuclear cells.
Cellular rejection is graded, as noted above. The most widely used system was that designed by Dr Margaret Billingham (Stanford University). In 1990 the ISHLT grading system (table 3) was introduced, followed in 2006 by the newer, grading system also sponsored by the ISHLT (tables 3 and 4).14 In looking at the interstitial inflammatory infiltrate, it is therefore essential to note whether the infiltrate is comprised of single cells or large clusters of cells, whether these cells are present in the interstitium, in a perivascular manner, or are present around myocytes. The significant infiltrate is the one which surrounds myocytes and is associated with myocyte damage. The greater the infiltrate, the greater the damage to the myocytes, the more significant is the rejection18 19 (table 4). In reviewing all of these features, one must, as in any other histological tissue, be on the lookout for other inflammatory cells, such as polymorphonuclear cells or plasma cells, and for the presence of viral inclusions in endothelial cells (eg, in CMV infection).
The interstitium may show large, bulky looking clusters of mononuclear cells which start at the endocardium and extend into the myocardium, with apparent loss of muscle fibres. Small vessels may be present in these clusters of lymphocytes. However, these are most likely cells of a Quilty B lesion (fig 3B, fig 4A–F).
Grade 0: No evidence of rejection (mononuclear cells seen among fat cells or in granulation tissue do not represent rejection).
Grade 1R: Mild rejection with perivascular or interstitial mononuclear cell infiltrates. Muscle fibre damage is either not seen or only one focus is noted (fig 6A,B, fig 7A,B).
Grade 2R: Two or more foci of mononuclear cells in the interstitium, with associated myocyte damage. These two foci may be seen in one biopsy or in any of the biopsies (from the same sitting) (fig 8).
Grade 3R: A diffuse infiltrate of mononuclear cells, usually in most (if not all) of the pieces of tissue, with multifocal muscle fibre damage; with or without interstitial oedema, red blood cells in the interstitium, vasculitis and the presence of neutrophils and/or eosinophils. The endocardium may show an infiltrate of similar cells (fig 9A,B)
Heart–lung transplant: Heart–lung transplant is an effective form of treatment for patients with complex congenital heart disease and those with end-stage heart–lung disease. These patients are maintained on the same immune suppression protocols. Acute lung rejection is far more common in them than is cardiac rejection, and hence it is usual for them to have cardiac biopsies only when indicated, rather than on an established protocol. There is no good explanation for the relative lack of cardiac rejection. The protective role of the bronchial lymphoid tissues may be a factor.1 7 20
Other occasionally seen artefacts
Foreign body granulomas
The biopsy sites as they heal may occasionally show multinucleate giant cells. These should be examined in detail, and the presence of foreign bodies or synthetic fibres, or cotton fibres noted. These are often related to the bioptome being placed on paper or on fabric, and then handled slightly carelessly, so that these materials are transported back into the endocardium, when the bioptome is reinserted for a second biopsy.6 7
The occasional biopsy shows the presence of scattered calcification in the fibrotic endomyocardium. The cause of this is not known but the calcification appears to be associated with previous muscle fibre damage and later calcification.
Other lesions
Occasional cases are reported to show evidence of infection after heart transplantation. This is associated with the immune suppression therapy. Reported organisms include CMV, Toxoplasma gondii and fungi. The organisms may be present within the myocytes and have a similar appearance to the lesions elsewhere. No inflammatory reaction is reported. Today with CMV serological typing of patients and donors, CMV infection is much less likely. Graft recipients on long term immunosuppression, are a hundred times more likely to develop malignant neoplasms of other tissues as compared to the population at large.
Post-transplant lymphoproliferative disorders
These are uncommon and aggressive lesions with an incidence of about 2%. Their incidence has likely diminished significantly, since we have not seen one in many years. They appear to have been associated with some immunosuppression protocols, such as those using cyclosporine or OKT3.21 When present they are seen as localised masses or as disseminated involvement of the heart, lymph nodes and extra nodal areas.
AMR and C4d staining
In biopsies where there is no evidence of cellular rejection, but the treating team is significantly concerned about the patient’s condition and graft rejection, one of the possibilities is antibody mediated rejection. To detect antibody mediated rejection, the C4d for paraffin embedded tissues (C4dpar; Ventana, Tucson, Arizona, USA) is available. Interpretation is, however, not easy and there are no standardised approaches to interpretation available at this time. We have developed a grading system for C4dpar stained biopsies, which we are comfortable with (table 7). However, we are not yet certain as to the clinical significance of the three grades. The C4dpar stains the capillary endothelial cell membrane in paraffin embedded tissues. Staining of endothelial cells in larger vessels is of no significance with regard to AMR/graft rejection (fig 10). At this time the C4d results must be used with caution and with a good understanding between the pathologist and the treating team, with regard to the role and significance of the C4d stain.12
Conclusion
EMBs are critical for the appropriate monitoring of the graft heart. Interpretation of the endomyocardial biopsy follows the same process that one uses for any other tissues. It is essential to initially survey the entire slide and then survey the individual biopsies in further detail at a higher magnification. The primary features to look for, are the presence of mononuclear cells in the interstitium and evidence of myocyte damage in the surrounding myocytes. Today, the endomyocardial biopsy is the gold standard for the determination of graft rejection. It is important for the pathologist to have a good working relationship with the treating team, so that there is clarity regarding the interpretation of the rejection grading system. No other rejection monitoring tests/techniques offer significant reliability at this time. The role of the pathologist is therefore critical in the continuum of care of the post-transplant patient.
Take-home messages
The endomyocardial biopsy is a reliable tool for diagnosis of graft rejection in cardiac transplants.
The biopsy procedure is associated with artefacts and a reasonable awareness of these artefacts is essential in order to not overcall cellular rejection.
Healing and healed biopsy sites may show significant mononuclear cell infiltrates. These must not be misinterpreted as representing active rejection.
The “Quilty” lesion may be present with significant numbers of lymphocytes in the subendocardium with what looks like damage to and loss of muscle fibres. This must not be misinterpreted as active rejection.
REFERENCES
Footnotes
Competing interests None.
Provenance and Peer review Not commissioned; externally peer reviewed.