Abstract
This study aimed to establish the number of deaths in infants under 1 year of age that were being reported for medico-legal examination at a single large academic centre in Hungary, as well as the method of these investigations with special emphasis on histopathology, ancillary techniques and the adherence of our current practice to international recommendations. A single-centre, retrospective audit was conducted on all suspected sudden infant death cases. After the review there were eight infectious background sudden infant death syndrome (SIDS) cases, infectious respiratory tract disease in 14 cases, cardiac septal tumour in one case, and hepatic, possibly metabolic, disorder in one case. Our study has highlighted that even in a single institution there is a huge heterogeneity of approaches which needs standardisation. A reclassification of infant cases according to the San Diego definition resulted in a decreased number of SIDS cases in our material. The San Diego definition and related international recommendations were found to be practical and the classification provides a guide to the standardisation of current practice.
Introduction
Sudden infant death syndrome (SIDS) was defined in 1969 as the sudden death of any infant or young child which is unexpected by history, and in which a thorough post-mortem examination fails to demonstrate an adequate cause of death. 1 Many definitions and criteria have since been proposed, but only a small number have had international acceptance. Autopsy and investigative protocols have also varied,2-4 and in some Western countries death can be attributed to SIDS even though autopsies have not been performed in more than 40% of cases. 5
The recent San Diego definition 6 addressed some of the weaknesses of previous definitions. According to the San Diego definition, SIDS is the sudden and unexpected death of an infant under 1 year of age, with onset of the lethal episode apparently occurring during sleep, that remains unexplained after a thorough investigation. SIDS cases are therefore a subset of sudden unexpected infant death (SUID) but can only be categorised as such after medico-legal investigation. Internationally there are certain standardised autopsy protocols suggested in suspected SIDS that have been validated by different studies. These protocols are similar and accord with European guidelines for medico-legal autopsies, and closely reflect the International Standardised Autopsy Protocol.7,8
This study aimed to establish the number of deaths in infants under 1 year of age that were being reported for medico-legal examination at a single large academic centre in Hungary, as well as the method of these investigations with special emphasis on histopathology, ancillary techniques and the adherence of our current practice to international recommendations. 9
Materials and methods
A single-centre, retrospective audit was conducted on all suspected sudden infant death cases that were admitted to the Department of Forensic and Insurance Medicine, Semmelweis University (Budapest, Hungary) over a 5-year period from January 2010 through December 2015. The department serves approximately 2 million people. All cases had undergone full police investigation, including scene investigation. In total, 24 suspected SIDS cases have been reviewed, which were originally investigated by colleagues with a range of experience in paediatric pathology. Owing to the retrospective nature of the study, the authors were limited by the available data and blocks taken for histology in the case files. The history available on all case files was reviewed and the following details were recorded: demographic details of the infants (age, gender); the circumstances of death; any significant medical history before and after birth; results of scene investigation; and the completeness of post-mortem examination with a special focus on histology and ancillary techniques.
Histopathology review
Available formalin-fixed, paraffin-embedded blocks obtained at autopsy from all cases were collected, and histopathology slides were prepared. Previously stored haematoxylin-eosin-stained slides were also retrieved, special stains (PAS, Masson, Prussian blue) were performed, and the results were reviewed. The histopathology review was performed by a histopathologist who was blinded to the original diagnosis.
DNA isolation and real-time PCR
In the cases where histopathology review confirmed pulmonary infection (e.g. interstitial pneumonitis) DNA was isolated using the Cobas DNA Sample Preparation Kit (Roche Molecular Systems, Mannheim, Germany) following the procedures described in the package insert. For this, 20 ng of isolated DNA and appropriate primers were used in the real-time PCR reaction, which was carried out according to the manufacturer’s instructions (Roche Cobas LightCycler 480 High Resolution Melting Master).
Results
Infant Mortality in Hungary.
Histopathology review
During the review the following criteria were used in the interpretation of pulmonary inflammatory changes: bronchial and peri-bronchial tissue were infiltrated by a higher number of inflammatory cells; purulent exudates-filled larger branches of bronchial tree and consecutive atelectasis occurred distant to obstruction. The interstitial tissue was characterised by widespread lymphoid infiltrates in the walls of alveoli in all sections (interstitial pneumonia). The alveoli showed marked, obvious pulmonary consolidation. Furthermore, specific lung diseases such as established bronchiolitis could be demonstrated. The pre-review diagnosis was: SIDS in 11 cases (without further sub-classification), infectious respiratory tract disease in 10 cases, aspiration in one case, septal cardiac tumour in one case and hepatic disorder in one case. After the review, there were eight SIDS infectious background (IB) cases, infectious respiratory tract disease in 14 cases (including features of interstitial pneumonitis, acute bronchiolitis and bronchopneumonia), cardiac septal tumour in one case, and hepatic, possibly metabolic, disorder in one case.
Ancillary techniques
Microbiological analysis was performed in two cases (8%) with negative results. After histopathology review, suspected cases for mycoplasma or Chlamydia pneumonitis were investigated with real-time PCR; mycoplasma infection could be demonstrated in one case, and chlamydial infection was shown in two cases. Toxicology was performed in four cases (16%) with negative results. Metabolic or radiological screening was not performed in our cases (0%), therefore no data were available for review.
Discussion
It has been reported that infant mortality in Hungary was higher than in many European countries; however, the reported incidence of SIDS has been lower (0.15–0.3/1000 live births) than that of other countries. The historically low incidence of SIDS in Hungary has been supported by evidence obtained from the high rate of death-scene investigation and medico-legal post-mortem examination. 10 Collection of clinical history, examination of the circumstances of death, and a post-mortem examination should be performed in all of our suspected SIDS cases. SIDS is mainly a diagnosis of exclusion, therefore extended histopathological, microbiological and toxicological examination, metabolic and radiological screening cannot be set aside.
After review, six cases originally diagnosed SIDS were reclassified as IB SIDS. In three original SIDS cases sufficient pulmonary infection was found, therefore these cases were removed from among the SIDS cases. In two cases which were originally diagnosed with interstitial pneumonitis, the signs of inflammation could not be demonstrated therefore IB SIDS was considered.
As shown in our results, the interpretation of histopathology findings in SIDS is a highly controversial topic, although a histopathological examination following international recommendations 9 could provide important additional information which can be of help to establish the correct pathological diagnosis in suspected SIDS cases. Thorough sampling is applied in our department, including at least samples of heart, lungs (left two lobes, right three lobes), trachea, larynx, kidney (left and right), pancreas, adrenal, spleen, thymus, cervical and mesenteric lymph nodes, liver, bowels and brain.13-15
There has been discussion on IB in certain cases of SIDS, especially in the heart, submandibular glands, and most importantly in the lungs. Up to 50–80% of SUDI cases that can be explained after autopsy have been attributed to respiratory tract infections.16,17 This is why most interest has focused on lung changes, and especially on evidence of pulmonary inflammation. The classic history of pneumonia with shaking chills followed by high fever, cough and chest pain is rarely observed in infants. Infections of the respiratory tract cause 0.8% of all deaths within the first year of life. Important diseases include various types of pneumonia, mainly bronchopneumonia and interstitial pneumonitis including bronchiolitis. The typical histological features in the lungs which can be seen in cases of suspected SIDS include pulmonary oedema, emphysema, atelectasis, increased number of alveolar macrophages, acute (or sometimes chronic) inflammation of the trachea and bronchi, pneumonia and prominent, hyperplasic bronchus-associated lymphoid tissue. The major problem is that positive findings are dependent upon the criteria threshold of the observer, as there is no a widely accepted general protocol. Peribronchiolar inflammatory infiltration is the main finding, and some histopathologists claim to find an abnormal cellular response in the majority of SIDS material (Figure 1). Others, more used to the less equivocal appearances in adults, fail to attach significance to marginal increases in the cell population of the peribronchial zone. Sometimes the pathologist may have difficulty deciding, in a given instance, whether this slight increase is sufficient or not.18-22 In our review, following thorough histopathology criteria, only pronounced findings in the lung tissue with evidence of pneumonia were accepted as a cause of death (Figure 2).
Discrete interstitial lymphoid infiltrate which is insufficient for a diagnosis of interstitial pneumonitis (H&E, 20×). Interstitial pneumonitis. Diffuse interstitial lymphoid infiltrate with alveolar epithelial damage (H&E, 20×).
Moreover, in a case which was originally diagnosed as aspiration, the signs of aspiration could be demonstrated; however interstitial pneumonitis (possibly chlamydial) was also noted. The presence of gastric contents in the distal lung fields, as seen in two of our cases for which CPR was not attempted, is a well-known phenomenon. Our findings also support the theory that gastric aspiration may be a terminal event only, and it is not directly related to the direct cause of death.27-30
Most of the pneumonitis/bronchiolitis cases have demonstrated the features of viral, chlamydial and mycoplasma infection. In the cases of suspected chlamydial and mycoplasma pneumonia, real-time PCR confirmed the presence of pathogens. Also, it is important to highlight that the usefulness of post-mortem microbiology in the assessment of sudden unexpected deaths in infants and children has been debated by many pathologists. However, some studies support the validity of post-mortem cultures if the specimens are taken during the autopsy following rigorous aseptic conditions. It is not a widely used practice even in our institute, but it is highly recommended to take liquor, blood, nasopharyngeal swabs and faeces for microbiological examination.27-30 Microbiological examination was requested in only two cases of our series, which should be considered insufficient; however, following our histopathology review real-time PCR proved mycoplasma or chlamydia infection in three cases even in DNA which was obtained from paraffin blocks.
Also, there is no doubt that toxicological analysis may reveal a cause of death in a small proportion of cases, ranging from less than 2% to 4% of sudden infant deaths.31,32 There were no significant toxicology findings in our series which could be attributed to the cause of death, but again, it is important to mention that toxicological analysis was performed for only four cases, which is far from optimal.
No category IA SIDS cases were found in our material primarily because of insufficient ancillary testing. In the absence of required facilities, there was no radiological and metabolic screening performed in our cases, though metabolic and radiological screening is now indicated in all suspected SIDS cases. 33 Not only for this reason, material is recommended to be taken during the autopsy for further molecular biology testing. If there is any indication of the presence of a genetically determined disturbance, for example, disorders of cardiac function or metabolism, specific testing must be pursued. 34
Our study has highlighted that even with similar SIDS definitions significant differences are present, which could be possibly due to the differences in the interpretation of findings. 35 It would be appropriate for pathologists to select standard definitions, and to participate in multidisciplinary discussions where significant clinical information is available to enable the most appropriate classifications to be made. It is important to mention that all our cases had been reviewed by the same team, which lowered the number of previously diagnosed SIDS cases using the same criteria. A reclassification of infant cases according to the San Diego definition resulted in a decreased number of SIDS cases in our material. The San Diego definition and related international recommendations6,9 were found to be practical in the standardisation of current practice.
Footnotes
Declaration of conflicting interests
None declared.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.