Pulmonary Siderophages and Unexpected Infant Death

Abstract

It has been proposed that the presence of siderophages in the lungs of infants who die unexpectedly should be considered a marker of a previous hypoxic event, which may preclude a diagnosis of sudden infant death syndrome. The authors retrospectively reviewed all infant deaths (<1 year old) going to autopsy at the Denver Office of the Medical Examiner from January 1999 to January 2001. Lung sections were stained with Prussian blue, and siderophages were counted in 20 high-power fields per lobe sampled. Cell counts were performed by two independent pathologists who were blinded to history and cause of death, with good reproducibility. Iron stain results were then categorized by average number of siderophages per 20 high-power field (category 1 = <5, category 2 = 5-100, category 3A = 100-500, category 3B = >100 in a single lobe, category 4 = >500). The results were subsequently correlated to case history, autopsy findings, and cause/manner of death. Forty-three cases were reviewed. The causes of death included sudden infant death syndrome (16), asphyxia (5), undetermined (6), and other (16). Those deaths were categorized by the above criteria as follows: category 1. (32), category 2. (6), category 3. (4), and category 4. (1). All sudden infant death syndrome deaths were in category 1. Categories 1 and 2 also included deaths in which hypoxia might have been present before death because of such factors as pneumonia and congenital heart disease. Categories 3 and 4 included a known homicidal asphyxia in which repeated episodes of intentional smothering were documented, 2 probable asphyxias, 1 nonaccidental trauma, and 1 undetermined. All 5 cases had questionable circumstances surrounding the death of the infant. Pulmonary siderophages were described in only 1 of the 43 autopsy reports. It was concluded that pulmonary siderophages can be markedly increased in cases of repeated asphyxia. Siderophages may also be increased in cases where hypoxia may have been present for another reason, but not to the same degree. Siderophages are not increased in sudden infant death syndrome. Because iron-laden macrophages often are not recognized on routine examination with hematoxylin and eosin staining, iron stains may be helpful in the evaluation of infant deaths. If siderophages are present in increased amounts without an obvious explanation, further investigation is warranted.

The frequency and significance of alveolar haemosiderin-laden macrophages in sudden infant death

“…Whatever the mechanism of previous pulmonary haemorrhage in the NAI-related deaths in our series, we found that alveolar HLMs were significantly more common in NAI than in unexplained SUDI with no features of NAI, which is consistent with findings of previous studies [8,10,16]. The advantages of the methodology of our study are first, the large size of the study group, this being the largest study of HLMs in SUDI, all of which were examined using a common autopsy protocol at a single centre, and secondly, that no attempt was made to quantify alveolar HLM counts, such scoring systems often being subject to considerable inter-and intraobserver variability; instead, we merely reported the presence of alveolar HLMs in at least one Perls stained lung section in an otherwise unselected population of SUDI referred to our centre for autopsy, thus removing potential subjective bias in the interpretation of the extent and/or severity of HLMs/haemosiderin content on histological examination of the lungs.…”

“…Furthermore, consistent evidence is provided by a mouse model, based on intranasal instillation of blood, in which haemosiderin staining within alveolar macrophages was first detected at day 3, peaked at day 7, and persisted through to the end of the study period after 2 months [12]. Previous studies have attempted to examine the significance of HLMs in the setting of sudden unexpected death in infancy (SUDI) and sudden infant death syndrome (SIDS), and published series in the English-language literature of pulmonary HLMs in sudden infant death are summarised in Table 2 [8,10,[14][15][16][17][18][19][20][21]. However, data comparison between different studies is hampered by inconsistent use of terminology and lack of standardised classification and scoring systems for haemosiderin content or HLMs.…”

“…However, data comparison between different studies is hampered by inconsistent use of terminology and lack of standardised classification and scoring systems for haemosiderin content or HLMs. For example, whilst some papers deal exclusively with alveolar HLMs [8,10,14,17], others comment only on interstitial HLMs [15,20,21] -and even then definitions seem to vary among different studies; moreover, some studies make no attempt to indicate specific anatomical distributions of HLMs [16,18,19]. Furthermore, where attempts have been made to quantify the numbers of HLMs present, different scoring systems have been developed, therewith precluding direct comparisons between studies [16,18,19].…”

“…Moreover, in the latter study the NAI cases comprised traumatic homicides in which the mechanism of pulmonary haemorrhage was likely to be different to that observed in the asphyxial deaths included in the aforementioned papers [1]. Several studies have attempted to quantify pulmonary HLMs in post-mortem lung sections, although a precise anatomical distribution of the HLMs is not specifically mentioned in some of these [16,18,19]. The latter two studies both used different semiquantitative scoring systems for haemosiderin content (''iron scores'') and pulmonary ''siderophage [HLM] counts'' respectively, both demonstrating higher haemosiderin scores/HLM counts in deaths not caused by SIDS.…”

“…The latter two studies both used different semiquantitative scoring systems for haemosiderin content (''iron scores'') and pulmonary ''siderophage [HLM] counts'' respectively, both demonstrating higher haemosiderin scores/HLM counts in deaths not caused by SIDS. In one study [18] these comprised explained natural and non-natural but not NAI-related deaths (none of the three NAI cases included in that paper showed particularly high iron scores), whereas in the other study [16] high HLM counts were more common in deaths suspicious for NAI, whether due to trauma or repeated episodes of inflicted asphyxia.…”