Valve pathology: a dying trade
- Correspondence to Dr Mary N Sheppard, CRY Centre for Cardiac Pathology, Imperial College London, SW3 6NP, UK;
- Accepted 8 July 2011
- Published Online First 6 September 2011
- occupational lung disease
- pulmonary pathology
- lung transplantation
- airways disease
- lung cancer
- sudden cardiac death
- vascular disease
- cardiothoracic pathology
At present, there are only a handful of cardiac pathologists in the UK and we have recently had several retiring. There are no full time academic cardiac pathologists. Most pathologists carrying out autopsies will be familiar with the heart and ischaemic heart disease since this is the main cause of death in the developed world. However, when it comes to non-ischaemic heart disease, the general pathologist is on less firm ground. There has been a marked fall in hospital autopsy rates in both USA and UK1 and general pathologists are doing fewer cases during their training, so their knowledge is by necessity limited. It is useful to remember that pathologists found the study of the whole heart removed at autopsy gave them many of the advances in congenital heart disease, cardiomyopathies and atherosclerosis. Most of us tend to forget that many of the major advances in heart surgery in both children and adults only came about from what was learnt at postmortem. It is essential to continue to nurture expertise in cardiovascular pathology. Without this specialist knowledge, how are we to understand the evolving mechanisms of atherosclerosis, hypertension, cardiomyopathies, sudden cardiac death, heart failure and transplantation. With the decline in the autopsy rate, we no longer have access to material for research and genetic analysis and yet genetic investigations are becoming an essential part of cardiac pathology as we are finding that more and more diseases are inherited.2 There has been rapid advances in cardiac procedures particularly for our ageing population. How are we to understand the consequences of new procedures such as angioplasty and stent insertion, ablation for cardiac arrhythmias, transcatheter replacement of aortic valves (eg, transcatheter aortic valve implantation) or the use of the transcatheter mitral clip to treat mitral regurgitation? Clinical knowledge and expertise are needed to address these evolving clinical scenarios.3
Cardiac pathologists remain popular with clinical colleagues and are seen as a rich source of teaching slides, but they are often left in the shadows by their surgical colleagues who view them as contributing less to the diagnostic workload of a department.4 Are cardiac pathologists to be absorbed into anatomic pathology as a mediocre ‘jack of all trades’ or should they be highly specialised and academic, run research laboratories and get their own research support? I believe that they should be the latter within high volume cardiothoracic centres specialising in inherited cardiac disease, cardiac surgery, transplant and congenital heart disease. General pathologists may consider themselves ‘cardiac’, but when presented with complex postsurgical cases, complex congenital cases, cardiac transplant and sudden cardiac death due to non-ischaemic heart disease, their limitations are obvious. In fairness, most pathologists today recognise this and more and more cases are referred to a specialist.
Valvular heart disease remains common in industrialised countries. The decrease in the prevalence of rheumatic heart diseases has been accompanied by an increase in that of age-related degenerative valve diseases. Aortic stenosis and mitral regurgitation are the two most common types of valvular disease in Europe. The burden of heart valve disease in older people has an important impact on patient management, given the high frequency of comorbidity and the increased risk associated with intervention in this age group. Rheumatic heart disease remains prevalent in developing countries.5 Knowledge of the different types of valves and their replacements/repair is therefore essential to address in those patients who die postoperatively, which enables good detailed clinicopathological correlation.6
The article by Beck et al highlights the problems facing cardiac pathology. When it comes to surgical specimens, the processing of cardiac valves is seen as low priority and this is not in contention. What this article emphasises is that heart valves should not be examined either macroscopically or microscopically unless specifically asked for by a clinician. This I strongly disagree with. First, routine cases such as calcified tricuspid and bicuspid aortic valves and regurgitant floppy mitral valves should be examined macroscopically and a report should be issued without histology. This is to confirm the clinical diagnosis for the surgeon and document the indications for surgery/repair. I believe this is important for imaging and surgical audit. It is also important for the teaching of junior pathologists who will thus become familiar with common valve diseases, which will help them with their autopsy work. More valves are being replaced, so knowledge of the type of valve replacement used and of the advances in technology is required. Replacement valves are now frequently removed for dysfunction that has an anatomical cause in the specimen, such as calcification, cuspal tear, thrombosis, endocarditis or material failure. Clinical information with imaging is essential so that pathologists become familiar with these complications. Replacement valves—both mechanical and bioprosthetic—should also be described macroscopically and a clinical history obtained to provide a report. Obvious cusp tear and calcification can be described and documented without histology. However, thrombosis should be processed for histology since it can also harbour infection. Infective endocarditis (IE) is a difficult and complex disease. It can be lethal if it is not aggressively treated with antibiotics alone or in combination with surgery. The epidemiology of this condition has substantially changed over the past four decades, especially in industrialised countries. Once a disease that affected young adults with previously well-identified valve disease—mostly chronic rheumatic heart disease—IE now tends to affect older patients and new at-risk groups, including intravenous drug users, patients with intracardiac devices and patients exposed to healthcare-associated bacteraemia. As a result, skin organisms (eg, Staphylococcus spp.) are now reported as the pathogen in these populations more often than oral streptococci, which still prevail in the community and in native-valve IE. Moreover, progress in molecular diagnostics has helped to improve the diagnosis of poorly cultivable pathogens, such as Bartonella spp. and Tropheryma whipplei, which are responsible for blood-culture-negative IE more often than expected. Epidemiological data indicate that IE mostly occurs independently of medico-surgical procedures and that circumstantial antibiotic prophylaxis is likely to protect only a minute proportion of individuals at risk.7 Prosthetic IE (prosthetic valve endocarditis) and endocarditis in patients with pacemakers and other devices (cardiac device related IE) are becoming more frequent. The number of Staphylococcus aureus IE is increasing, especially in patients with diabetes or patients on chronic haemodialysis. The change in the underlying population and the increase in the number of cases caused by very virulent organisms explain why the disease still carries a poor prognosis and a high mortality. The variety of clinical manifestations and complications, as well as the serious prognosis, makes it mandatory that IE patients are treated in experienced hospitals with a collaborative approach between different specialists, involving cardiologists, infectious disease specialists, microbiologists, surgeons, neurologists, radiologists and pathologists.8 Beck et al mention that all the valves with histological evidence of IE had positive culture but fail to mention the organism isolated. They also fail to mention the macroscopic appearance of the valves, which is important. Given the histological changes described, it must have been virulent bacterial infection. Patients with IE who require surgery and suffer complications are referred to tertiary hospitals more frequently than patients with an uncomplicated course, so obviously the pathologists in these centres will see the more severe cases. It is well established that culture-negative endocarditis occurs in 10–30% of cases. A focus on PCR and molecular techniques on excised valve tissue especially is needed in culture-negative cases.9 10 Because of the highly variable clinical manifestations of IE, different sets of diagnostic criteria have been used to standardise case definitions of IE. The higher sensitivity of the Duke criteria, as compared with the von Reyn criteria, was demonstrated by valve histological verification.11 Clinical diagnosis remains challenging, but this paper gives the impression that it is easily done prior to the removal of the valve, which oversimplifies the clinical scenario. We are also not told anything about the echocardiography imaging used for diagnosis in this study, which is also important.12 Vegetations may not be so obvious and can be quite subtle. In a recent study of IE, vegetations were found on aortic and mitral valves in less than half the cases while aetiology was not established in 33%.13 Thus, the diagnosis may not be as clear-cut as Beck et al would lead us to believe.
Despite recent advances in diagnostic techniques, many cases of IE remain culture-negative. As a matter of fact, blood culture-negative endocarditis (BCNE) may account for up to 31% of all cases of endocarditis. Among 759 patients with BCNE, a causative microorganism was identified in 62.7% and a non-infective aetiology in 2.5%. Blood was the most useful specimen, providing a diagnosis for 47.7% of patients by serological analysis (mainly Q fever and Bartonella infections). Broad-range PCR of blood and Bartonella-specific western blot methods diagnosed seven additional cases. PCR of valvular biopsies identified 109 more aetiologies, mostly streptococci, T whipplei, Bartonella species and fungi. Primer extension enrichment reaction and autoimmunohistochemistry identified a microorganism in five additional patients. A non-infective cause of endocarditis, particularly neoplastic or autoimmune disease, was determined by histological analysis or by searching for antinuclear antibodies in 19 (2.5%) of the patients. This study highlights that broad-spectrum 16S and 18S ribosomal RNA PCR should be performed on valvular biopsies and should yield dramatic results.10 Thus, I believe that pathological examination of cardiac valves to demonstrate vegetations and valvular inflammation remains the gold standard for the diagnosis of IE and the role of the pathologist is often decisive, especially when bacteriologists fail to isolate a microorganism or when a microorganism that has been isolated may be a contaminant. This study emphasises that the pathologist may play an important role in the identification of previously unknown infectious agents.10 14 I do not believe like Beck et al that cardiac pathologists no longer play a significant role in diagnosis and have no influence on therapy. Obvious cases like those in their study may seem easy, but it is the challenging cases that need our expertise.
While surgically removed valves for IE most likely have large vegetations or perforations, I believe describing these vegetations/perforations in the specimen is important in correlating with imaging modalities and benefits all concerned in addition to verifying the surgical indications. IE still has a significant mortality, and realising the extent of disease within the valve macroscopically and microscopically is important if the patient dies and relatives raise issues about the outcome. The surgeon is on firmer ground when he or she is armed with the pathological evidence in discussions with the family and it also helps to bring closure to the bereavement.
IE is also an evolving disease influenced by the demographics of the population, medical interventions, comorbidities and valve replacements. The paper is also vague concerning ‘failed valve replacement’. Prosthetic valve endocarditis is associated with a high mortality despite diagnostic and therapeutic improvements and its incidence is increasing and can be 20–30% of all IE episodes. Blood cultures remain negative in up to 30% of cases and fungi can be frequent pathogens with molecular diagnosis possible on surgical specimens.15 In addition, what may look like vegetations on a valve may in fact turn out to be a potentially malignant disease such as lymphoma reported by us in a homograft valve.16
I fully understand that in a time of resource limitations, non-malignant diseases such as valve pathology must take a back seat in a general department, but totally excluding examination of these specimens will lead to a suboptimal service to these patients. Personally, if my valve is going to be removed I would want it examined by a specialist pathologist who will liaise closely with the clinicians and give me a report even if the description is only macroscopic, which will be the case for most ‘routine degenerative’ specimens. However, as in all routine pathology, there are always the unexpected challenging cases that will be missed if these valves are cast aside routinely. Also, the loss of these specimens will be detrimental to the training of future pathologists.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.