Objective

Pulmonary thromboembolism (PTE) is a common disease with significant morbidity and mortality. As previously reported, acute PTE has been associated with elevated serum levels of cardiac biomarkers such as troponins I and T — studies have reported an incidence in the range of 16%–70%.1 2 Such rises have been shown to be associated with right ventricular dysfunction3 and increased mortality.1 The majority of previous investigators have attributed such observations primarily to acute right ventricular infarction4–9 or endomyocarditis,10 though myocyte damage due directly to increased myocardial strain has also been proffered as an alternative explanation by some.11

We present a series of autopsy cases of PTE, without other significant cardiorespiratory disease, in which isolated right ventricular disease was identified. We characterize the nature of the pathological changes by way of histological assessment with the use of immunohistochemical stains. The findings in the study group are compared to age- and sex-matched autopsy control cases in which death was attributed to other causes and in whom there was no evidence of significant cardiorespiratory disease.

Design/setting

A retrospective study examining coronial autopsy cases performed at the Department of Forensic Medicine, Sydney, during the period 1995–2009, in which significant PTE had been identified and isolated right ventricular pathology encountered upon routine histological assessment. Cases were identified by the investigating pathologists or alternatively using the departmental case database (using ISYS Search data search software, North Sydney, NSW, Australia). Cases with significant coexisting cardiorespiratory disease were excluded from the study (for the purposes of this study, ‘significant’ cardiorespiratory disease was defined so as to include individuals with a documented antemortem diagnosis, a history suggestive of symptomatic disease or the finding of ‘significant’ disease at autopsy, including valvular heart disease and coronary artery atherosclerosis—with a cutoff of ≥75% cross-sectional area stenosis in any one coronary artery; cases of cardiac hypertrophy were, however, included in the study).

Control cases, without autopsy evidence of either PTE or other significant cardiorespiratory diseases, were selected blindly for each case from the departmental database, having regard to matching for age and sex.

The following gross postmortem data were obtained from the autopsy report in each case: age, sex, body length, weight, heart weight, the thickness of the free wall of the right ventricle (where documented) and the macroscopic extent and distribution of thromboemboli within the pulmonary arterial vasculature in PTE cases. Histological sections of the heart and lungs were also reviewed.

The majority of right ventricular histological sections had been sampled from the anterior right ventricular outflow tract, with the remainder sampled from the anterior and lateral free wall. The tissues had been fixed in 10% buffered formalin and embedded in paraffin, prior to 4 μm sections being cut and stained with H&E in a standard fashion. Immunohistochemical markers were applied following antigen retrieval (pressure cooker; pH 9.0); antibodies used were as follows: anti-human neutrophil elastase, clone NP57 (DakoCytomation, Glostrup, Denmark); anti-human CD20cy, clone L26 (Dako); anti-human CD3, clone SP7 (Thermo Scientific, Fremont, USA) and anti-human CD68, clone KP1 (DakoCytomation). Immunohistochemical signals were visualized using the Dako Envision + Dual Link System-HRP and Dako Liquid DAB + Substrate Chromogen System.

The following histological parameters were evaluated within each histological section of the right ventricular myocardium:

• Extent of inflammation within that section in PTE cases (focal: infrequent, small foci; patchy: several areas of inflammation of variable intensity; widespread: multiple inflammatory foci, becoming more diffuse and diffuse)

• Nature and density of the inflammatory cell infiltrate within the inflammatory foci in PTE cases (qualitative and semiquantitative assessment, scored 0–3, utilizing the system set out in the ‘Updated Sydney System’12—a robust scoring schema with proven reproducibility used to grade inflammation in gastric biopsies)

• Myocytolysis, in association with the inflammatory foci in PTE cases (yes/no)

• Background macrophage recruitment, taking care to exclude intravascular cells (scored semiquantitatively from 0 to 3)13

• Fatty replacement (scored semiquantitatively from 0 to 3)

• Fibrosis (scored semiquantitatively from 0 to 3)

Lung sections were assessed for evidence of organization of PTE, features of recent pulmonary infarction and pulmonary hypertensive vascular changes.

Ethical clearance for the study was granted by the Sydney South West Area Health Service Ethics Review Committee (reference no. X06-0233), and permission to undertake this project was also sought and obtained from the New South Wales State Coroner. Each study and control case was indexed with reference only to the departmental autopsy case number to maintain anonymity.

Statistical analysis

All data are presented as mean±SD or median and range. Statistical comparison of parametric data was performed using a two-tailed unpaired Student t test. The relationship between dichotomous variables was tested using the Fisher's exact test. A p value of <0.05 was considered statistically significant. Statistical testing and data analysis were performed using the GraphPad Prism V. 4.0 (GraphPad Software Inc, La Jolla, CA, USA).

Discussion and conclusions

The observation of mixed inflammatory infiltrates in sections of the right ventricular myocardium from cases of fatal PTE provides a potential pathological explanation for the clinical manifestations of right ventricular dysfunction and raised levels of serum cardiac biomarkers in some instances of PTE.

There has been much debate in earlier studies regarding the likely pathophysiological sequence of events leading to troponin leak and clinically detectable right ventricular dysfunction in association with PTE, and such changes have hitherto been attributed primarily to ischaemic damage.4–8 However, the absence of significant epicardial coronary artery disease in many such instances has given rise to the hypothesis of ‘demand’ or ‘functional’ ischaemia—in essence, relative cardiac ischaemia brought about by way of a mismatch in which right ventricular myocardial work exceeds the capacity for increased coronary flow.11 14 15

The nature of the inflammatory changes observed in this study is fairly consistent and differs from that seen in pure acute ischaemic damage to the myocardium. This observation, taken in conjunction with the observation that the necroinflammatory changes are multifocal rather than zonal as one might expect in ischaemic damage consequent upon compromised small vessel blood flow, suggests that direct cellular injury as a result of right ventricular stretch and strain might be operative at least initially—though quite possibly exacerbated by impaired diastolic coronary blood flow to the right ventricular myocardium consequent upon increased right ventricular pressure and, perhaps, also an element of systemic hypotension and hypoxia with resultant heightened myocardial susceptibility to an increased afterload burden.8 This theory is supported by findings in other studies in which a close correlation was demonstrated between levels of troponin and B-type natriuretic peptide—a marker of myocardial wall strain16—and further support can be found in animal studies in which chemokine-mediated inflammatory myocyte damage has been shown to occur following an experimental pulmonary embolism, the extent of which was limited by an experimental agranulocytosis.17 That troponin levels are also recognized, on occasion, to be elevated in other conditions such as chronic obstructive pulmonary disease18–20—in which increased right ventricular afterload would also be expected—lends further weight to the hypothesis that right ventricular strain is a significant factor in the development of these right ventricular myocardial changes. Interestingly, the authors are also aware (unpublished data) of several instances of identical-appearing isolated right ventricular changes in the absence of PTE, in which there were other potential causes of increased right ventricular strain—including abundant injected foreign matter within the pulmonary blood vessels in intravenous drug users, disseminated malignancy exhibiting a prominent intravascular (‘lymphangitic’) pulmonary growth pattern, acute lung injury and end-stage left ventricular failure.

The incidence of necroinflammatory changes in the right ventricle in cases of PTE is, of course, not apparent from this study. However, previous clinical studies demonstrating a troponin leak in a high proportion of cases of PTE1 2 suggest that such myocardial damage is, in fact, likely to be relatively common. It is considered probable that the cases we have identified represent ‘the tip of the iceberg’, with low sensitivity, no doubt, due primarily to insufficient sampling of the right ventricular myocardium in cases of PTE. However, it remains to be determined whether the right ventricular characteristics we describe represent a small subtype of PTE with an ‘inflammatory phenotype’ or whether such changes are, in fact, physiologically much more common, with manifestations only detectable at autopsy in those with an element of chronicity to the thromboembolic process, perhaps with minor PTE prior to the fatal episode, giving rise to a preterminal increased right ventricular afterload and, thereby, permitting time for the morphological features of associated right ventricular pathology to develop. The peripheral extension of PTE and histological evidence of the organization of thromboemboli observed in the majority of cases identified in this study do point to an element of chronicity of disease, supporting this hypothesis.

The findings in this study will have implications for the clinical management of PTE, underlining the need for clinicians to be mindful of the possibility of myocardial pathology in such instances. Importantly, right ventricular dysfunction in instances of PTE has been shown to have an association with significantly increased morbidity and mortality,18 21–23 and it may be that necroinflammatory changes such as those described in this study are at the root of such disease.

In addition to having relevance to the clinical management of PTE and other causes of right ventricular strain in the acute setting, the degree of right ventricular damage is also likely to have a bearing on the long-term prognosis with regard to persistent right ventricular dysfunction, which is known to affect a significant number of PTE surviviors.24 Previous studies25 have documented that right ventricular myocardial velocities often take up to 6 months to normalise after non-fatal PTE, long after short-term improvements in right ventricular filling have occurred. Our findings suggest that PTE is often associated with an inflammatory infiltrate and milieu that may lend itself to ongoing right ventricular dysfunction. The finding of significantly increased right ventricular fibrosis in study cases compared to controls may possibly be a reflection of earlier episodes of non-fatal PTE with myocyte necrosis and subsequent fibrous replacement.

In the postmortem setting, the findings in this study would suggest that optimal autopsy practice in cases of PTE and other potential causes of increased right ventricular strain should include systematic histological sampling of the right ventricular wall to assess for myocardial pathology.

A potential future direction of research would be to investigate further PTE cases such as these that exhibit an inflammatory myocardial response, with greater sampling of the right ventricular myocardium, placing particular emphasis on the molecular characteristics of the necroinflammatory pathological changes and with careful correlation to the histological age of the thromboemboli.