 |
|
||||
| Home | Login | |||||
| About | Editorial board | Articles
|
NSI Publications
|
Search | Instructions | Online Submission | Subscribe | Videos | Etcetera | Contact |
|
||||||||||||||||||||
|
|
|
Neuropathological complications of infective endocarditis : study of autopsy material.
Correspondence Address:
78 autopsy proven cases of infective endocarditis (IE) seen during 1983 to 1995 were retrospectively reviewed. The brain was available for examination in 44 cases. In the remaining cases, brain was not examined because examination of it was not requested due to lack of neurological findings. Brain lesions were observed in 35 out of 44 cases of IE. Assuming remaining 34 cases to be without brain lesions, the brain involvement in IE would be 44.87% (35 out of 78 cases). Mean age of all cases of IE and those with brain lesions was similar i.e. 26.5+/-16.6 years and 26.6+/-13.06 years respectively. Largest number of cases with neuropathological lesions were associated with normal valve IE (48.57%). Mitral valve was most commonly involved in cases with CNS complications (57.14%) (p<0.05). The various types of brain lesions were infarction (68.57%), haemorrhage (57.14%), cerebral micro-abscess (31.42%) and focal meningitis (14.28%). More than one type of lesion was observed in 19 cases, indicating complicated nature of brain lesions in fatal cases of IE. Left sided middle cerebral artery (MCA) territory was the commonest site of infarction and haemorrhage. Staphylococcus aureus appeared to be the most common organism in fatal cases of IE. Normal valve IE with or without CNS complications constitutes a significant group in India and is different from the west as far as the predisposing conditions are concerned.
Despite the availability of a wide range of antibiotics, infective endocarditis (IE) continues to be a major problem all over the world.[1] The main reasons include emergence of newer infective organisms and extracardiac complications. Among the extracardiac complications, involvement of the central nervous system (CNS) is possibly the most serious. The association of IE and neurological complications has been recognised for more than a century. Sir William Osler2 in his series of Gulstonian lectures, underscored the clinical triad of fever, heart murmur and hemiplegia. He was first to suggest that neurological symptoms can be the initial manifestation of IE. The clinical profile and expression of IE has undergone profound changes in the past half century due to natural evolution of the disease, although some changes relate to newer therapies and newer population at risk. Recent studies of IE3-7 have emphasized the changing pattern of neurological complications due to advancing age of the patients, increase of nosocomial infections, increasing incidence of non-rheumatic heart disease, as rheumatic heart disease has virtually disappeared outside the developing countries, intravenous drug abuse and decreasing incidence of infection due to streptococcus viridans with the concomitant increase in infection due to staphylococcus aureus, Group D streptococci and other organisms. It has also been observed that IE is an important cause of stroke in the young.[8] The present study was undetaken to analyse the spectrum and relative frequency of brain lesions in IE in autopsy material.
The patients who succumbed to IE at Nehru Hospital, Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh and in whom consent was available, were subjected to partial or complete autopsy. Brain was examined in the cases in which specific consent was available. Complete autopsy records and the tissues of all cases of IE seen from 1983 to 1995 were re-examined, both grossly and microscopically. Microscopic examination was performed on paraffin embedded section with routine haematoxylin and eosin stain and special stains including those for organisms as necessary. The brain of each case was examined in the standard manner.[9] The blood vessels were examined in detail particularly in cases with haemorrhage. The heart was explored by inflow-outflow tract method in each case and examined grossly and microscopically for the presence or absence of congenital or acquired disease, the size of vegetations and their location. The cases having vegetations composed of fibrin and inflammatory cells, with or without demonstrable bacterial colonies, were considered as infective endocarditis. The data was expressed as mean percentage and standard deviation. The categorical variables were analysed by Chi-square test.
There were 78 autopsy proven cases of IE during the period 1983-1995. The brain was available for examination in 44 cases. In the remaining 34 cases, the examination of the brain was not requested, as none of them had neurological symptoms clinically. Out of 44 cases in whom brain was examined, the lesions were observed in 35 cases. The results are described under group A i.e. all cases of IE irrespective of brain lesion (78 cases) and Group B i.e. cases of IE showing brain lesions (35 cases). A comparative analysis [Table I] showed similar age distribution of cases in both the groups and relative female preponderance in group B. While pre-existing heart disease was observed in both groups, normal valve endocarditis was more frequent in group B cases. Patients of group B exhibited more left sided value (80%) and isolated valve (85.71%) involvement. Isolated mitral value (MV) involvement was most common (60%) followed by involvement of aortic valve (AV) (17.14%) and others. Vegetations : Vegetations measuring 0.5 cm or more were considered as large and less than 0.5 cm as small. Brain lesions were more commonly associated with small vegetations than IE with large vegetation (25 versus 10 cases). Causative Agents : Gram's stain and PAS stain were performed on paraffin embedded sections of vegetation in all 44 cases. Gram positive cocci were found in 26 cases and demonstrable bacterial colonies without further categorisation in another 5 cases. Two cases of aspergillous endocarditis were also detected. However, special stains failed to demonstrate any organisms in 11 cases. The blood culture reports were available in 14 cases. Staphylococcus aureus was positive in 6 and Klebsiella pneumoniae in 2 cases. Culture was sterile in 6 cases. Neuropathological complications : Spectrum of neurological lesions in IE were embolism, infarction, haemorrhage and infective complications like meningitis and cerebral abscess. Single as well as multiple lesions were observed. Multiple lesions were more common than single type (19 versus 16 cases) [Table II]. Infarction : Infarction [Figure - 1] was the most frequently observed brain lesion in 24 cases, of which 15 were associated with other lesions. The middle cerebral artery (MCA) territory was involved in 19 cases with left MCA in 8, right MCA in 5 and both in 6 cases. The posterior cerebral artery (PCA) territory was involved in 2 cases. Infarction in the water shed area of MCA and anterior cerebral artery (ACA) was found in 3 cases. Out of 24 cases, multiple small infarcts were seen in 9, multiple large infarcts in 7, single large infarcts in 7 and single small infarct in one case. On microscopic examination, 15 were recent infarcts (less than 1 week) as characterised by inflammatory cell infiltration and other features. In 4 cases, the infarct was older (over one week), as characterised by collection of macrophages and cystic changes. Five cases showed old as well as recent infarcts. Isolated MV endocarditis was more commonly associated with infarction in 15 cases followed by isolated AV (4 cases) and multiple valve endocarditis (5 cases). Haemorrhage (Mycotic aneurysm/arteritis) : Twenty cases of haemorrhage [Figure - 2] with or without arteritis and 3 cases of arteritis without haemorrhage were noted. The relative proportion of anatomical distribution of haemorrhage in 20 cases was intracerebral haemorrhage (ICH) (9 cases), sub-arachnoid haemorrhage (SAH) + intra-ventricular haemorrhage (IVH) + ICH (7 cases), SAH+ICH (2 cases), SAH (one case) and SAH+IVH (one case). The intracerebral haemorrhage (ICH) of more than 1.5 cm diameter in brain stem and more than 3 cm diameter in rest of the brain was classified as large haemorrhage.10 Accordingly, 7 cases showed small ICH and the remaining 2 cases showed large haemorrhage. Another 7 cases of ICH along with SAH and IVH were invariably of large haemorrhage type. Two cases of ICH associated with SAH were of small haemorrhage type. Out of these 20 cases of haemorrhage, only seven showed mycotic aneurysm [Figure - 3], involving both major arteries (4 cases) and minor branches (3 cases) of the middle cerebral arteries. Majority of these cases of haemorrhage were associated with mitral valve endocarditis (11 out of 20 cases). Nine cases of haemorrhage were noted in endocarditis of normal valve and 7 in endocarditis in RHD. Infection : Twelve cases showed infective complication including 4 cases of focal meningitis with microabscesses [Figure - 4], 7 cases of microabscess and a case of solitary cerebellar abscess with diffuse meningitis. These microabscesses were multiple, small and non-encapsulated. Confluent lesions of this kind are sometimes designated as suppurative encephalitis and were seen typically in association with septicaemia. These lesions were seen in 8 cases of MV endocarditis, 2 cases of AV endocarditis. In 2 cases, other valves were involved. Cerebral microabscesses were most commonly observed in normal valve endocarditis (7 cases) which appears to be a component of septicaemia. The remaining 5 cases had RHD (3 cases), CHD (1 case) and prosthetic valve endocarditis (1 case). A single case of solitary cerebellar abscess with diffuse meningitis was found in association with tricuspid valve (TV) endocarditis. The endocarditis in this case appeared to be secondary to cerebellar abscess leading to thrombophlebitis of lateral sinus and internal jugular vein and direct extension to the right atrium and TV.
Despite advancement in chemotherapeutic measures, IE continues to be as enigmatic as ever with a world wide distribution. The CNS complications today remain remarkably similar to the picture described in the pre-antibiotic era.[1],[6],[17] Most of the recent reports1,[7],[11],[12],[13],[14],[15] indicate the incidence of CNS complications, in the antibiotic era, between 20-40% with an average of 30% as compared to 12-31%1,16,17 in the pre-antibiotic era. The pattern and background of IE in the western countries differs in many respects from those in the developing or underdeveloped countries. In a developing country, like India, with higher incidence of infections as well as RHD, combined with inadequate facilities of detection and poor availability of newer antibiotics, the pattern of IE may not be the same as in the developed countries. The present study comprised of fatal cases of IE with high incidence of brain lesion (35 out of 44 cases in which brain was examined). If we assume that the 34 cases in which the brain was not examined, were free of any brain lesion, even then the percentage of neuropathological complication in this series would be at least 44.87%. However, the figure could have been higher if we had examined those cases as well. The figure seems to be higher than any other series, because these were all fatal cases of IE. The mean age of the patients in this series was 26.[5] years and the majority were under 40 years of age, similar to those reported by Santoshkumar et al7 and Choudhury et al.[18] In contrast, a higher age incidence was reported by Bayliss et al[19] and Salgado et al.[20] This difference in the age distribution of patients was due to decreasing incidence of RHD and greater longevity of the population in latter series from the developed countries. Male preponderance among cases of IE was similar to that observed by Choudhury et al[18] and Salgado et al,[20] but Von Reyn et al[21] showed equal sex distribution. The reason for male preponderance is not clear. In contrast, brain lesions in this study were more commonly found in the females [Table I]. Pre-existing heart disease : Normal valve IE was more common than diseased valve IE in this study in contrast to most Indian studies where the diseased valve IE was more common.[7],[18],[22],[23] Among the diseased valve IE, CHD predominated in this study in contrast to RHD seen in other Indian clinical studies.[7],[18],[22],[23] The difference may be due to smaller sample size in our study but it should also be noted that normal valve IE may remain clinically undetected because of the predominant manifestations of septicaemia, neurological illness and short survival. CNS complications were more commonly found in normal valve IE in this study. 48.57% of cases with brain lesions had normal valve endocarditis. The reasons are unclear but iatrogenic causes like use of infected needles for therapeutic purposes is a strong possibility. Currently cardiac prosthesis and intravenous drug abuse accounted for a significant identifiable source of infection in the western literature,[24] but these were uncommon in our setting and none of the patients in the present series was IV drug abuser. The left sided valve involvement was more commonly associated with brain lesions than right, because the septic emboli from the left sided valve can directly go to the brain. CNS lesions were noted more frequently in mitral valve endocarditis (21 cases) than aortic valve (6 cases) in this study, which is similar to the findings of Pruitt et al15 and LeCam et al,[25] however, but different observation was made by Garvey and Neu.[26] Since this institute is a referral centre of north-west India, the majority of patients had probably received antibiotics prior to hospitalisation, which could be the reason for negative blood culture. Staphylococcus aureus, however, was the most common causative agent. Two cases of aspergillous endocarditis detected by histopathological examination of vegetations appeared to be a manifestation of acute disseminated aspergillosis, a common finding in our autopsy material. Neuropathological complications : Cerebral infarction was the most common neuropathological complication found in this study, similar to the findings of Pruitt et al.[15] The middle cerebral artery territory was most commonly involved. The cerebral infarction was generally due to embolic occlusion of the arteries. In majority of cases (15 out of 24 cases), emboli could not be demonstrated grossly as well as microscopically. The likely explanation is disintegration of the emboli after initial lodgment[27] and involvement of smaller branches. Normal valve IE was more commonly associated with embolic complication in comparison to diseased valve IE. In normal valve IE, the process is rapidly destructive without evidence of healing and results in large friable vegetations. Intracranial haemorrhage remained the second common neuropathological complication of IE and was usually caused by rupture of an aneurysm, presumably secondary to localised arteritis subsequent to a septic embolus. Occasionally, arterial wall destruction secondary to septic embolus may occur before formation of either true or false aneurysm. The middle cerebral arterial system has been reported to be the predominant site of involvement. Mycotic aneurysm was found in nearly equal number of cases in normal valve and diseased valve IE. A similar observation was made by Pankey[28],[29] and Kernohan et al (Cited by Lerner[1]). In contrast, other studies[30],[31],[32],[33],[34] showed higher incidence of mycotic aneurysm in diseased valve IE. Low virulent organism like streptococcus viridans was found in the diseased valve.[1],[15] Large number of mycotic aneurysm in diseased valve IE reported in western literature[30],[31],[32],[33],[34] may be related to subacute course. In the present study, small sized vegetations were more commonly associated with aneurysm and arteritis, in comparison to large ones. Infective complications were more common in normal valve than diseased valve IE, similar to the findings of Pankey et al.[28],[29] Multiple microabscesses were commonest infective complication. A similar observation was made by Jones and Siekert35 Other studies however,[3],[14],[15] showed meningitis to be more common than cerebral microabscesses. Small sized vegetations and mitral valve involvement were commonly associated with infective complications. These were generally a reflection of acute course and septicaemic nature of these cases. High proportion of cases had multiple type of lesions. This is relevant, because such combinations are difficult to document on clinical studies. A significant number of cases of endocarditis, particularly on a normal valve, may be clinically silent. A confirmation of the diagnosis of IE, as well as the various CNS lesions, is more accurate in autopsy series. It is also worth noting that normal valve IE with or without CNS complications constitutes a significant group in India and is different from the west as far as the predisposing conditions are concerned.
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||
