Article Text

Download PDFPDF
Cardiac findings in leptospirosis
  1. Kinjal Shah,
  2. Gayathri P Amonkar,
  3. Rima N Kamat,
  4. Jaya R Deshpande
  1. Department of Pathology, T N Medical College and B Y L Nair Hospital, Mumbai Central, Mumbai, India
  1. Correspondence to Dr Kinjal Shah, Resident Pathologist, T N Medical College and B Y L Nair Hospital, Mumbai Central, 801, Navinasha Co-op soc, Dadar, Mumbai, Maharashtra 400 014 India; drkinjal_shah{at}


Background/aim In leptospirosis, although cardiac involvement in the form of ECG changes and myocarditis is known, it is not considered to be significant. This study analysed cardiac changes in leptospirosis.

Methods Twenty-four hearts from patients who had died from leptospirosis were studied. Detailed gross and light microscopic examination was carried out.

Results Myocarditis was noted in 96% of cases. Endocardial inflammation was seen in 50% of cases. This endocardial inflammation correlates with vasculitis, which is the principal pathogenetic mechanism of the disease.

Conclusions There is definite cardiac involvement in leptospirosis, which even though not symptomatically evident, may add to the morbidity or be contributory to the mortality associated with the disease. In addition, a possibility of dilated cardiomyopathy as a delayed consequence of severe myocarditis remains, and may need evaluation.

  • Autopsy pathology
  • cardiovascular
  • leptospirosis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Leptospirosis is a zoonotic disease caused by pathogenic spirochaetes of the genus Leptospira. The disease is characterised by a spectrum of clinical manifestations ranging from subclinical to fulminant, potentially fatal infection.1 2 Deaths are associated with severe jaundice, oliguric renal failure, haemorrhagic manifestations and cardiac and respiratory failure.2–5 Death from Weil syndrome occurs in more than 10% of cases.6 Cardiac involvement in the form of ECG changes and myocarditis is known.7–11 However, cardiac involvement is generally masked by overt pulmonary haemorrhage or hepatorenal involvement and hence may be underestimated.12

Minor ECG changes have been noted in 48–70% of cases.7–9 13 Conduction system abnormalities, such as first-degree atrioventricular block, represent the most common change.

Histologically, haemorrhages involving the epicardium, endocardium and myocardium and interstitial myocarditis have been demonstrated.10–12 15–18 de Brito et al11 have described inflammatory involvement of the sinoatrial node and atrioventricular node, valvulitis and the presence of acute coronary arteritis and aortitis.

Fatal leptospiral myocarditis was first described in 1935 by Mollaret and Ferroir.13 Deaths have been directly attributed to myocarditis in 4–25% of cases.10 19 Vasculitis is the proposed principal pathogenetic mechanism of the disease.10 11 20

Materials and methods

This is an autopsy study of 24 fatal cases of leptospirosis. The autopsies were performed in a tertiary healthcare centre in Mumbai over a period of 2 years from January 2005 to December 2006. The diagnosis of leptospirosis was suspected from clinical features and postmortem findings and was confirmed by a positive macroscopic latex agglutination test and/or immunoreactivity of postmortem kidney tissues to a mixture of reference rabbit polyclonal antisera reactive with 16 leptospiral strains.

A complete autopsy was performed in each case. Meticulous gross examination of all organs was carried out, with special attention to the cardiac system. After fixation, routine sections were taken from all organs. Sections from the heart included multiple sections from both ventricles, papillary muscles, coronary arteries and aorta. The sections were routinely processed, and 4–5 μm thick tissue sections were stained with H&E and Verhoeff–van Geison stain for light microscopic examination. Particular attention was paid to the severity, type and location of the inflammation.


A total of 24 fatal cases of leptospirosis were analysed. The majority of the cases presented during July, August and September with the highest incidence in August. This coincides with the season of heavy monsoon rains. Age ranged from 2.5 to 70 years (age distribution summarised in table 1). There was a male predominance (83% of the cases), which was attributed to more time spent by males in outdoor activities and hence a higher rate of exposure to flood waters, the latter being the major contributory factor in the upsurge of cases of leptospirosis. There is no known difference in the prognosis for affected male and female patients. Table 2 details the clinical features and investigations. Most patients had febrile illness of 1–10 days' duration and died within 24 h of hospitalisation. Owing to the very short hospital stay, laboratory and radiological investigations were performed in only a few patients.

Table 1

Age distribution of the cases

Table 2

Clinical features and investigations

Table 3 summarises the non-cardiac histological findings at autopsy. Seventy-nine per cent of the patients died from diffuse intrapulmonary haemorrhage. None of the deaths were attributed to myocarditis.

Table 3

Non-cardiac histological findings at autopsy (n=24)

Table 4 outlines the details of gross examination of the heart.

Table 4

Gross examination of the heart (n=24)

The coronary arteries, aorta and the cardiac valves were unremarkable.

Histopathological examination revealed the presence of inflammation in the pericardium, myocardium and endocardium, which was semiquantitatively graded as mild, moderate and severe. Table 5 summarises details of the location, severity and type of inflammation.

Table 5

Cardiac inflammatory findings (n=24)

The pericardium showed haemorrhages and inflammation in 12% and 29% of cases, respectively. In 1/7 (14%) cases, inflammation extended from the myocardium. The predominant feature on histopathological examination was the presence of myocardial inflammation, which was interstitial in all cases. A propensity for perivascular location was noted in 8/23 (35%) cases. Also inflammation that was more dense in the subendocardial and subpericardial myocardium was identified in 4/23 (17%) cases. The inflammatory cells extended into the myofibres in 11/23 (48%) cases. Focal myocyte necrosis was present in 6/24 (25%) cases. Table 6 shows additional features of the myocardium.

Table 6

Other findings in the myocardium (n=24)

Mild to moderate endocardial inflammation was identified in 50% cases. Subendothelial band-like inflammation was present in 58% of these cases. Also fibrous thickening of the endocardium was seen in 6/24 (25%) cases.

Papillary muscles showed mild mixed interstitial, predominantly perivascular inflammation in the myocardium involving 20/22 cases (91%). Endocardium showed predominantly mild mixed inflammation in 9/22 (41%) cases.

The coronary artery and the aorta showed intimitis in 73% and 65% of cases, respectively.


Leptospirosis is an emerging infectious disease with recent large outbreaks in India, South-East Asia, Malaysia, Nicaragua and Brazil.13 Heavy monsoon rains in Mumbai city resulted in an upsurge of cases of leptospirosis. The majority of the fatalities were due to diffuse pulmonary haemorrhage. Cardiovascular involvement was masked by severe respiratory distress and not documented clinically or electrocardiographically in any of the cases. ECG changes and myocarditis are, however, well-documented features of leptospirosis.7–11

On gross examination, the heart was normal in shape. It was enlarged and heavier than usual in 45% of cases. Petechiae or larger foci of pericardial and subendocardial haemorrhage were noted in 33% and 12% of cases, respectively. No haemorrhages were noted in the myocardium on gross examination. Thus, as observed by Arean,10 in our study also haemorrhages were more prominent in the epicardium than in the endocardium and less common in the myocardium (figure 1). As also noted by Arean, the heart was of normal consistency in all our cases, with no gross evidence of endocarditis or pericarditis. Right-sided dilatation of the heart observed in 21% of cases may be due to respiratory insufficiency as a result of the diffuse pulmonary haemorrhage noted in most of our cases. This feature has also been reported by Chakurkar et al18 (table 4).

Figure 1

Gross photograph of the heart showing extensive epicardial haemorrhage.

Microscopy showed pericardial haemorrhages and mild inflammation in 12% and 29% of cases, respectively (table 5). Arean10 has also documented the presence of mild to moderate inflammation in the pericardium (in 38.46% of cases).

Interstitial myocarditis, the predominant histopathological finding, was present in 96% of our cases and in 93.20% cases in the autopsy study of Chakurkar et al18 (figure 2). The incidence of myocardial inflammation was higher in our study and that of Chakurkar et al than in previous studies, where it ranged from 40% to 61.54%.10 15 A noteworthy finding in our study was the propensity for perivascular location of inflammation, which was identified in 35% of cases showing myocarditis (figure 3). Whereas this feature has also been noted by Chakurkar et al, Arean identified perivascular inflammation only in one patient, who also showed fibrinoid degeneration of the arterial intima in some of the myocardial vessels.10 18 Preferential involvement of the myocardium in the subendocardial region was seen in 17% of cases, as also pointed out by Arean10 and de Brito et al.11 Moderate to severe, predominantly mixed inflammation was present in 43% of cases of myocarditis (figure 4). Aschoff-like histiocytes were seen along with the other inflammatory cells in 52% of cases. These Aschoff-like histiocytes, which were also present as collections in a few cases mimicking Aschoff bodies (figure 5). Arean and de Brito et al observed that the inflammatory cells were chiefly lymphocytes and plasma cells and have emphasised the presence of abundant Aschoff-like histiocytes, which were also observed in our study. These histiocytes represent a characteristic response of the mononuclear phagocytic system of the heart to the disease. However, this reactivity is non-specific and can be seen in patients who had died from sepsis.11 In 48% of cases showing myocarditis, splitting and fragmentation of myofibres were noted. This feature has been pointed out by both Arean and de Brito et al in most of their cases. Foci of myocyte necrosis, although reported by Arean in most of his cases, were noted in only 25% of our cases (figure 6). Myocyte splitting, fragmentation and necrosis were seen in a relatively small percentage of our cases, which may be due to a milder degree of myocardial inflammation in many cases. Whereas interstitial oedema was observed only in 12% of cases, it was present in the majority of cases analysed by Arean and de Brito et al. As also identified by Arean, these haemorrhages were focal and seldom infiltrated between muscle fibres (tables 5 and 6). Papillary muscles showed mild mixed interstitial predominantly perivascular inflammation involving 91% of cases. Compared with the right and left ventricular myocardium, the inflammation in the papillary muscle was significantly less severe. Also the papillary muscle did not show any evidence of myocyte degeneration. Our findings in the papillary muscle are in contrast with the observation of Arean that the inflammatory process was much denser in the papillary muscle.

Figure 2

Section from the heart showing interstitial myocarditis (H&E, original magnification 10×).

Figure 3

Section from the heart showing perivascular inflammation (H&E, original magnification 40×).

Figure 4

Section from the heart showing interstitial myocarditis with mixed inflammatory cell infiltrate (H&E, original magnification 40×).

Figure 5

Section from the heart showing interstitial myocarditis with focal collections of Aschoff-like histiocytes (H&E, original magnification 40×).

Figure 6

Section from the heart showing focal myocyte necrosis (H&E, original magnification 40×).

Moderate mixed endocardial inflammation was seen in 50% of our cases. This inflammation was band-like in 58% of cases. Aschoff-like histiocytes formed part of the inflammation in 58% of cases (table 5, figure 7).

Figure 7

Section from the heart showing subendocardial inflammation (H&E, original magnification 40×).

Arean has identified these myocardial changes to be similar to those observed in viral hepatitis and uraemic state. However, in his study, as well as ours, the duration of uraemia was too short to explain the extensive myocardial changes. Also, no gross features suggestive of uraemic pericarditis were observed in our series or by Arean.

Although cardiac involvement occurs in the early leptospiraemic phase, leptospires being present in very small numbers in myocardial interstitium, have not been identified on tissue sections.10 11 Toxin-mediated vascular damage is the principal pathogenetic mechanism.11 16 This may explain the preferential perivascular and subendocardial inflammation as a consequence of ‘intimitis’.

Myocarditis was the cause of death in 4% of the cases in the study by Arean.10 The majority of our patients died from severe pulmonary haemorrhage. The presence of myocarditis in these cases would have worsened the cardiorespiratory problems and contributed to death in at least some of the cases.

The study by de Brito et al11 has emphasised that involvement of the conduction system correlates significantly with myocarditis. ECG evidence of conduction system abnormalities such as first-degree atrioventricular block, sinus block and right bundle branch block have been reported.8 9 21 Rajiv et al9 postulated that these ECG changes indicate that the conduction system may be involved in leptospirosis. Whether the ECG changes identified by various authors are due to interstitial myocarditis or inflammatory involvement of the conduction system is yet to be determined.


Cardiac involvement in the form of myocarditis, although not well documented, is a recognised feature of leptospirosis. However, there was no clinical suspicion of cardiac involvement in most of our cases. The presence of diffuse pulmonary haemorrhage in the majority of cases appears to have masked the cardiac manifestations. Cardiac involvement in leptospiral infections deserves more attention from clinicians. Given the varied manifestations of the disease, there is always the possibility of cases in which cardiac manifestations dominate the clinical picture. Not much has been reported in the literature regarding the possible delayed effects of this leptospiral myocarditis. Artigou et al22 has discussed a case report of acute leptospiral myocarditis with cardiogenic shock, which resolved and presented with dilated cardiomyopathy after 3 years. It would be interesting to carry out long-term follow-up studies to determine the susceptibility of these patients to the eventual development of left ventricular dysfunction and cardiomyopathy. The subendothelial inflammation suggests affinity of endothelial cells for the leptospiral antigen or its toxic products, thereby inciting an inflammatory response related to these areas, with endothelial damage sometimes occurring.


We acknowledge Dr Sheih Wun-Ju, Pathologist/Medical Officer, Infectious Disease Pathology Activity, DVRD/NCID/CDC, Atlanta for performing IHC on kidney sections for leptospira.



  • Competing interests None.

  • Ethics approval The study has been approved by the T N Medical College and B Y L Nair Hospital ethics committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.