*Case Report* **Sudden Death from Primary Cerebral Melanoma: Clinical Signs and Pathological Observations**

**Alfonso Maiellaro <sup>1</sup> , Antonio Perna <sup>2</sup> , Pasquale Giugliano <sup>3</sup> , Massimiliano Esposito 4,\* and Giuseppe Vacchiano 5,\***


**Abstract:** Primary cerebral tumors rarely provoke sudden death. The incidence is often underestimated with reported frequencies in the range of 0.02 to 2.1% in medicolegal autopsy series. Furthermore, primary cerebral melanoma is an uncommon neoplasm. It represents approximately 1% of all melanoma cases and 0.07% of all brain tumors. This neoplasm is very aggressive, and its annual incidence is about 1 in 10 million people. In the present study, a 20-year-old male was admitted to hospital with vomiting, headache, paresthesia and aggressive behavior. A computed tomography (CT) scan of the head was performed showing a hyperdense nodule in the right parietal lobe with inflammation of the Silvian fissure. A complete autopsy was performed 48 h after death. A blackish material was displayed on the skull base, and posterior fossa. Microscopic examination diagnosed primary brain melanoma. A systematic review of the literature was also performed where no previous analogous cases were found. The forensic pathologist rarely encounters primary cerebral melanoma, and for these reasons, it seemed appropriate to describe this case as presenting aspecific clinical symptoms and leading to sudden unexpected death. Histopathological observations are reported and discussed to explain this surprising sudden death caused by a primary cerebral melanoma.

**Keywords:** primary cerebral melanoma; sudden death; clinical signs; pathological observations

#### **1. Introduction**

Sudden unexpected death due to a primary nervous neoplasm is not frequent in medico-legal activity [1] because the incidence of undiagnosed primary central nervous system tumors has decreased with the emergence of advanced neuroimaging and other improved diagnostic techniques. Occasionally, the forensic pathologist can encounter a cerebral neoplasm that had been undiagnosed or not suspected prior to death. Eberhart et al. [2] observed 11 cases of primary central nervous system tumors resulting in sudden death over a period of 20 years (1980 to 1999). Glioblastomas [3–6], oligodendragliomas [7], astrocitomas [8], ependymomas [9], extramedullary plasmacytomas [10], colloid cysts of third ventricle [11,12] and meningiomas [13], are reported in the literature to cause sudden death with vague or short-term symptoms and limited healthcare access. Current studies of sudden death from brain tumors concern glioblastoma multiforme or astrocytomas. In the present study, a systematic review of sudden death and brain tumors from 1980 to 2021 was conducted, and it was concluded that although articles had been analyzed over a period of 40 years, there are still few articles today that report this type of death. We found only one case of sudden death associated with melanoma brain metastases [14] and

**Citation:** Maiellaro, A.; Perna, A.; Giugliano, P.; Esposito, M.; Vacchiano, G. Sudden Death from Primary Cerebral Melanoma: Clinical Signs and Pathological Observations. *Healthcare* **2021**, *9*, 341. https:// doi.org/10.3390/healthcare9030341

Academic Editor: Pedram Sendi

Received: 24 February 2021 Accepted: 15 March 2021 Published: 17 March 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

it seemed appropriate to describe the histopathological manifestations of this surprising case of sudden death due to primary cerebral melanoma. Glitza [15], in an autopsy series, identified up 80% of patients with melanoma with metastatic involvement of the central nervous system (CNS); however, while the cerebral metastases of melanoma are often observed, primary cerebral melanoma is very rare. Several histogenetic theories on the origin of meningeal melanocytes have been put forward [16]. It has been postulated that during embryogenesis, leptomeningeal melanocytes coming from the multipotential cells of the neural crest can develop into mesodermal and neural elements. Primary cerebral melanomas develop once melanocytes become neoplastic. metastases [14] and it seemed appropriate to describe the histopathological manifestations of this surprising case of sudden death due to primary cerebral melanoma. Glitza [15], in an autopsy series, identified up 80% of patients with melanoma with metastatic involvement of the central nervous system (CNS); however, while the cerebral metastases of melanoma are often observed, primary cerebral melanoma is very rare. Several histogenetic theories on the origin of meningeal melanocytes have been put forward [16]. It has been postulated that during embryogenesis, leptomeningeal melanocytes coming from the multipotential cells of the neural crest can develop into mesodermal and neural elements. Primary cerebral melanomas develop once melanocytes become neoplastic.

The aim of this study was to present a case report and a systematic review of this rare pathology, in order to improve the overall knowledge of forensic pathologists on sudden death from brain tumors, especially in the case of cerebral melanoma. A complete autopsy, comparing radiological findings and histological presentation, could provide a comprehensive overview of such sudden deaths. The aim of this study was to present a case report and a systematic review of this rare pathology, in order to improve the overall knowledge of forensic pathologists on sudden death from brain tumors, especially in the case of cerebral melanoma. A complete autopsy, comparing radiological findings and histological presentation, could provide a comprehensive overview of such sudden deaths.

#### **2. Materials and Methods 2. Materials and Methods**

#### *2.1. Case Description 2.1. Case Description*

A 20-year-old male was admitted to hospital with vomiting and headache; he was a cannabis user and cigarette smoker (40 cigarettes a day). He referred paresthesia in the upper limbs, psychomotor agitation and aggressive behavior. The neurological examination pointed out a palpebral ptosis and a right eye ophthalmoplegia including paralysis of the left facial nerve. A CT scan of the brain was performed and showed a hyperdense nodule in the right parietal lobe and along the Sylvian fissure, the leptomeningeal spaces were described as affected by an inflammatory process (Figure 1). Fifteen minutes after the CT scan, he died. The study of the medical record and the sudden onset of symptoms defined death as "sudden". The accuracy of the autopsy examination, the histological investigation (H&E and immunohistochemistry), attributed his death to a rare form of primary cerebral melanoma. A 20-year-old male was admitted to hospital with vomiting and headache; he was a cannabis user and cigarette smoker (40 cigarettes a day). He referred paresthesia in the upper limbs, psychomotor agitation and aggressive behavior. The neurological examination pointed out a palpebral ptosis and a right eye ophthalmoplegia including paralysis of the left facial nerve. A CT scan of the brain was performed and showed a hyperdense nodule in the right parietal lobe and along the Sylvian fissure, the leptomeningeal spaces were described as affected by an inflammatory process (Figure 1). Fifteen minutes after the CT scan, he died. The study of the medical record and the sudden onset of symptoms defined death as "sudden". The accuracy of the autopsy examination, the histological investigation (H&E and immunohistochemistry), attributed his death to a rare form of primary cerebral melanoma. No ethical committee was required. Written informed consent was obtained from his relatives.

**Figure 1.** Pre-mortem CT scan of the brain. The right parietal lobe showed inflammatory processes with a hyperdense nodule. **Figure 1.** Pre-mortem CT scan of the brain. The right parietal lobe showed inflammatory processes with a hyperdense nodule.

No ethical committee was required. Written informed consent was obtained from his relatives.

#### *2.2. Autopsy Findings*

A complete autopsy was performed 48 h after death. External examination revealed abrasions on the hands and the upper limbs. The autopsy revealed the dura was full of clots. The brain weighed 1550 g and was diffusely swollen. In the subarachnoid space of the fronto-parietal lobes and along the interhemispheric cleavage of the parietal left lobe, some nodular blackish rounded soft formations with blood were observed. Moreover, in the temporal and parietal lobes, similar formations were present (Figure 2). A soft and blackish mass (4.5 × 7 cm) enveloped the brain stem and was spreading to the inferior temporal poles and to the cerebellar lobes. There was no herniation of the temporal lobe, unci or cerebellar tonsils. The section of the brain revealed on edematous soft parenchyma. Blackish soft material present in subarachnoid space of the fronto-parietal right lobe invaded the cerebral cortex. The brain ventricles were filled with blood. From the dorsal face of the brain stem, striae of blackish material invaded the nervous tissue (Figure 3). Blackish material was also present along the venous sinus of the skull base and in the posterior fossa (Figure 4). The internal examination of the other organs was unmarkable. *Healthcare* **2021**, *9*, 341 3 of 14 *2.2. Autopsy Findings*  A complete autopsy was performed 48 h after death. External examination revealed abrasions on the hands and the upper limbs. The autopsy revealed the dura was full of clots. The brain weighed 1550 g and was diffusely swollen. In the subarachnoid space of the fronto-parietal lobes and along the interhemispheric cleavage of the parietal left lobe, some nodular blackish rounded soft formations with blood were observed. Moreover, in the temporal and parietal lobes, similar formations were present (Figure 2). A soft and blackish mass (4.5 × 7 cm) enveloped the brain stem and was spreading to the inferior temporal poles and to the cerebellar lobes. There was no herniation of the temporal lobe, unci or cerebellar tonsils. The section of the brain revealed on edematous soft parenchyma. Blackish soft material present in subarachnoid space of the fronto-parietal right lobe invaded the cerebral cortex. The brain ventricles were filled with blood. From the dorsal face of the brain stem, striae of blackish material invaded the nervous tissue (Figure 3). Blackish material was also present along the venous sinus of the skull base and in the posterior fossa (Figure 4). The internal examination of the other organs was unmarkable.

**Figure 2.** Brain examination. In the temporal and parietal lobes, some nodular blackish rounded **Figure 2.** soft formations with Brain examination. In the temporal and parietal lobes, some nodular blackish rounded soft blood were observed. formations with blood were observed. *Healthcare* **2021**, *9*, 341 4 of 14

*2.3. Histological Analysis* 

**Figure 3.** Posterior face of the brain revealed stem striae of blackish material that invaded the nervous tissue. **Figure 3.** Posterior face of the brain revealed stem striae of blackish material that invaded the nervous tissue.

**Figure 4.** Skull base-posterior fossa and venous sinus examination showed a blackish material.

During the autopsy, brain samples were collected that were fixed in 10% buffered formalin. After washing, the water was removed, and they were embedded in paraffin. The obtained blocks were cut (4 µm thickness), using a microtome (Dako, Glostrup, Denmark), sectioned, and stored at room temperature. Sections were stained with Hematoxylin and Eosin (H&E) and Von Gieson's method. The immunohistochemical analysis of the samples was performed through antimelanoma-antibody (HMB45+), Melan A-antibody, CD3-antibody, CD20+-antibody, PenCK-antibody, Ki67+, as suggested by Tosaka

**Figure 4.** Skull base-posterior fossa and venous sinus examination showed a blackish material. **Figure 4.** Skull base-posterior fossa and venous sinus examination showed a blackish material.

**Figure 3.** Posterior face of the brain revealed stem striae of blackish material that invaded the

#### *2.3. Histological Analysis 2.3. Histological Analysis*

nervous tissue.

During the autopsy, brain samples were collected that were fixed in 10% buffered formalin. After washing, the water was removed, and they were embedded in paraffin. The obtained blocks were cut (4 µm thickness), using a microtome (Dako, Glostrup, Denmark), sectioned, and stored at room temperature. Sections were stained with Hematoxylin and Eosin (H&E) and Von Gieson's method. The immunohistochemical analysis of the samples was performed through antimelanoma-antibody (HMB45+), Melan A-antibody, CD3-antibody, CD20+-antibody, PenCK-antibody, Ki67+, as suggested by Tosaka During the autopsy, brain samples were collected that were fixed in 10% buffered formalin. After washing, the water was removed, and they were embedded in paraffin. The obtained blocks were cut (4 µm thickness), using a microtome (Dako, Glostrup, Denmark), sectioned, and stored at room temperature. Sections were stained with Hematoxylin and Eosin (H&E) and Von Gieson's method. The immunohistochemical analysis of the samples was performed through antimelanoma-antibody (HMB45+), Melan A-antibody, CD3-antibody, CD20+-antibody, PenCK-antibody, Ki67+, as suggested by Tosaka et al. [17] for the diagnosis of primary leptomeningeal melanoma. Sections were observed using a Zeiss Axioplan light microscope (Carl Zeiss, Oberkochen, Germany). Subsequently, images were obtained using a Zeiss AxioCam MRc5 digital camera (Carl Zeiss, Oberkochen, Germany). In the present study, the microscopic examination of the brain showed, on the right parietal lobe and in brain-stem tissues, many atypical and pleomorphic cells with voluminous discolored nuclei and melanin-pigment in the cytoplasm. These cells were arranged like a "cordon" or irregularly like a "vortex". Moreover, a vascular proliferation and spotty hemorrhages were also observed. In the cerebral cortex tissues, edema and ischemic dark neurons, and near them, a lymphocytic and macrophage reaction were seen. The leptomeningeal arteriolar vessels showed clots, partially reorganized. Around them, we observed many pleomorphic and atypical cells with melanin pigment in the cytoplasm (Figure 5).

et al. [17] for the diagnosis of primary leptomeningeal melanoma. Sections were observed using a Zeiss Axioplan light microscope (Carl Zeiss, Oberkochen, Germany). Subsequently, images were obtained using a Zeiss AxioCam MRc5 digital camera (Carl Zeiss, Oberkochen, Germany). In the present study, the microscopic examination of the brain showed, on the right parietal lobe and in brain-stem tissues, many atypical and pleomorphic cells with voluminous discolored nuclei and melanin-pigment in the cytoplasm. These cells were arranged like a "cordon" or irregularly like a "vortex". Moreover, a vascular proliferation and spotty hemorrhages were also observed. In the cerebral cortex tissues, edema and ischemic dark neurons, and near them, a lymphocytic and macrophage reaction were seen. The leptomeningeal arteriolar vessels showed clots, partially reorganized. Around them, we observed many pleomorphic and atypical cells with melanin

**Figure 5.** Histological findings. (**a**) neoplastic voluminous cells with melanin in the cerebral tissue (E.E. 60×); (**b**) neoplastic cell population: numerous, voluminous melanoblasts with marked nuclear pleiomorphism; (**c**) neoplastic cells with melanic pigment antimelanoma-antibody (HMB45 + 200×); (**d**) neoplastic cell population near a small cerebral artery (E.E. 200×). **Figure 5.** Histological findings. (**a**) neoplastic voluminous cells with melanin in the cerebral tissue (E.E. 60×); (**b**) neoplastic cell population: numerous, voluminous melanoblasts with marked nuclear pleiomorphism; (**c**) neoplastic cells with melanic pigment antimelanoma-antibody (HMB45 + 200×); (**d**) neoplastic cell population near a small cerebral artery (E.E. 200×).

#### **3. Systematic Review 3. Systematic Review**

A systematic review was conducted according to the PRISMA guidelines [18]. A systematic review was conducted according to the PRISMA guidelines [18].

Pubmed and Google Scholar were used as search engines from 1 January 1980 to 1 February 2021 to evaluate the association between sudden death and brain tumor. meSH was used for the following words: (sudden death) AND (primary cerebral tumor). Pubmed and Google Scholar were used as search engines from 1 January 1980 to 1 February 2021 to evaluate the association between sudden death and brain tumor. meSH was used for the following words: (sudden death) AND (primary cerebral tumor).

#### *3.1. Inclusion and Exclusion Criteria 3.1. Inclusion and Exclusion Criteria*

pigment in the cytoplasm (Figure 5).

The following exclusion criteria were used: (1) review, (2) articles not in English, (3) animal studies, (4) abstract, (5) editorial, (6) poster, and (7) communications at conferences. The inclusion criteria were as follows: (1) Original Article, (2) Case Report, and (3) Articles in English. The following exclusion criteria were used: (1) review, (2) articles not in English, (3) animal studies, (4) abstract, (5) editorial, (6) poster, and (7) communications at conferences. The inclusion criteria were as follows: (1) Original Article, (2) Case Report, and (3) Articles in English.

#### *3.2. Quality Assessment and Data Extraction*

A.M. and A.P. initially evaluated all the articles, evaluating the title, the abstract, and the whole text. Once, P.G. and M.E. reanalyzed the articles chosen independently. In cases of conflicting opinions between the articles, they were submitted to G.V.

#### *3.3. Characteristics of Eligible Studies*

A total of 388 articles were collected (66 from PubMed and 322 from Google Scholar). Of them, 27 duplicates were removed. A total of 339 articles did not meet the inclusion criteria. In conclusion, 22 articles were included in the present systematic review (Figure 6).

A.M. and A.P. initially evaluated all the articles, evaluating the title, the abstract, and the whole text. Once, P.G. and M.E. reanalyzed the articles chosen independently. In cases

A total of 388 articles were collected (66 from PubMed and 322 from Google Scholar). Of them, 27 duplicates were removed. A total of 339 articles did not meet the inclusion criteria. In conclusion, 22 articles were included in the present systematic review (Figure

of conflicting opinions between the articles, they were submitted to G.V.

*3.2. Quality Assessment and Data Extraction* 

*3.3. Characteristics of Eligible Studies* 

6).

**Figure 6.** Flow diagram illustrating included and excluded studies in this systematic review.

#### **Figure 6.** Flow diagram illustrating included and excluded studies in this systematic review. **4. Results**

**4. Results**  Most of the studies were case reports (n. 16), some cases series (n. 4), and a few original articles (n. 2). Most of the tumors related to sudden death were matched with glioblastoma multiform or astrocytomas in the different world health organization (WHO) grades. Other tumors related to sudden death were oligodendrogliomas, gliomas, adenomas, and colloid cysts. The most frequent autopsy findings were cerebral edema, solid tumor mass, and hemorrhagic infarction of surrounding tissue. The most frequent cause of death was due to intracranial hypertension from the tumor; sometimes it was due to hemorrhage caused by the tumor (especially in cases of glioblastoma multiform); only one case was due to respiratory arrest due to the location of the tumor near the center of breathing control (Table 1). Most of the studies were case reports (n. 16), some cases series (n. 4), and a few original articles (n. 2). Most of the tumors related to sudden death were matched with glioblastoma multiform or astrocytomas in the different world health organization (WHO) grades. Other tumors related to sudden death were oligodendrogliomas, gliomas, adenomas, and colloid cysts. The most frequent autopsy findings were cerebral edema, solid tumor mass, and hemorrhagic infarction of surrounding tissue. The most frequent cause of death was due to intracranial hypertension from the tumor; sometimes it was due to hemorrhage caused by the tumor (especially in cases of glioblastoma multiform); only one case was due to respiratory arrest due to the location of the tumor near the center of breathing control (Table 1).





**Table 1.** *Cont.*

*Healthcare* **2021**, *9*, 341


#### **5. Discussion**

Sudden death due to brain tumors has a low incidence and is often underestimated [18]. This systematic review highlights that currently there is no study with a large number of cases of sudden death from brain tumors. Furthermore, most of the case reports report glioblastoma multiforme. This case report is the first to focus on sudden death from cerebral melanoma and it is important that forensic pathologists recognize this cause of sudden death.

Primary cerebral melanoma is a very uncommon neoplasm derived from melanocytes present in the leptomeninges. This pathology, first described by Virchow [32], is very rare, with an annual incidence of approximately 1 in 10 million people [33] and represents about 1% of all melanoma cases and 0.07% of all brain tumors [34]. Indeed, many studies have demonstrated that in central nervous systems (CNS) metastases of melanoma are also frequently present because the melanoma has a high risk of spreading to CNSs. The melanoma cells, in fact, share with vascular cells numerous cell surface molecules; they are highly angiogenic and possess a higher degree of "stemness" than other solid tumors [15]. According to Hayword [35], we can identify a primary cerebral melanoma when the following occur: (a) no malignant melanoma outside the CNS; (b) non-malignant neoplasm in other parts of the CNS; (c) histological validation of melanoma.

Age is also an additional factor to distinguish a primary cerebral melanoma from metastatic melanoma. Primary cerebral melanoma develops primarily in patients under 50 years of age and rarely metastasize to other organs. Metastatic melanomas develop mainly in the elderly, have a rapid clinical course and multiple intracerebral diffusion [36]. In the presented case, no other melanomas were detected at the clinical and forensic examinations of the body; a solitary blackish mass (4 × 7 cm) was observed around the brain stem. Leptomeningeal blackish soft rounded formations in the fronto-parietal lobes were observed and the patient was under 50 years old. The histological observations confirmed, with the immunohistochemical analysis, the diagnosis of melanoma. In addition to the blackish mass observed around the brain stem, we also observed in the subarachnoid space in the right parietal lobe some rounded blackish nodules invading the nervous tissue. We interpreted these modifications as the primitive cerebral melanoma located in the brain stem subarachnoid space, where we found a blackish mass at autopsy. The meningeal vessels of the left parietal lobe were thrombosed and around these vessels we observed many atypical pleomorphic cells with cytoplasmatic melanin pigment; the neoplasm had probably first developed from the leptomeningeal space around the brain stem, and after, spread to the subarachnoid space of the left parietal lobe. In fact, it should be borne in mind also that melanocytes far outnumber leptomeninges of the skull base and cervical cord [37,38] and the most common locations of primary cerebral melanoma are the anterolateral face of the spinal cord and the postero-lateral face of the brain stem [39]. In agreement with this author, primary cerebral melanoma was observed also in the posterior fossa in the cerebellopontine angle [40,41] and Arantes [42] reported 13 cases of primary malignant melanoma derived from the pineal body. Greco Castro [43] observed this neoplasm in the temporal lobe near the Sylvian fissure and Quillo-Olvera [44] reported that the primary cerebral melanoma can be found frequently in the cerebral lobe (53.1%) in the posterior fossa (17.3%) and pineal region (13.6%). We believe that primary cerebral melanoma may occur in any leptomeningeal location, although our reported case is in agreement with the observations of Troya-Castilla. The clinical manifestations are generally non-specific, related to an increased cerebral pressure, or cerebral hemorrhages, or a neurological dysfunction and the early diagnosis of this neoplasm is a very difficult challenge for clinicians [45–48]. The aggressive behavior, the vague symptoms, and the unspecific neuroimaging make the prognosis of this tumor very difficult [49]. Several studies [50,51] reported a survival of ~4 months from diagnosis of cerebral nervous system melanoma, and once the cancer spreads to the leptomeninges, the overall median survival is generally only 10 weeks [52,53]. In our case, two months after the first symptoms, the patient died. The patient had a history of drug addiction, presenting psychomotor

agitation and aggressive behavior, as can be observed in many cerebral neoplasms [54–57]. Morais described a case of delirium in a patient with brain melanoma metastases [58] and other authors reported headache and diplopia in a 27-year-old man affected by primary meningeal melanocytoma in the anterior cranial fossa [59]. However, we can ascribe the first presented symptoms (headache and vomit) to increased intracranial pressure, probably due to the increased neoplastic mass in the posterior fossa and a subarachnoid hemorrhage. However, the symptoms observed before death (paralysis of the III and VIII cranial nerve) clearly showed a suffering brain stem. The involvement of this important brain structure also explains the sudden death due to an alteration of the hypothalamic cardiovascular regulatory centers. It is well-known that stimulation of the hypothalamus can lead to anatomic cardiovascular disturbances and cardiac arrest [60,61].

#### **6. Conclusions**

The surprising case of sudden death stresses the need for a careful analysis in all patients with psychiatric disorders or no-specific neurological signs, especially when these manifestations show-up unexpectedly in a healthy person. Despite the fact that modern genetic [62–64], radiological [65–67], and therapeutic treatments [68–73] have revolutionized the approach to primary cerebral melanoma, autopsy can still provide a useful support in determining the exact location, diffusion, and histological pattern of this neoplasm, and the cause of death. In human beings, melanocyte exists in the uvea, cerebral parenchyma, leptomeninges, and mucous membranes. Primary cerebral melanoma is rare and has an estimated incidence of 0.005 cases per 100,000 and develops in around 1% of all melanoma cases. Brain melanoma metastasis is much more common than primary brain melanoma. Primary brain melanomas originate from melanocytes of the leptomeninges and histopathologically show strongly pigmented cells with prominent nucleoli [74]. Through this systematic review, the present case is the first described in literature in which a primary brain melanoma caused a sudden death. It is, therefore, of crucial importance to report this case to the scientific community.

**Author Contributions:** Conceptualization, A.M. and A.P.; methodology, P.G.; software, M.E.; validation, G.V., A.M. and A.P.; formal analysis, P.G.; investigation, M.E.; resources, G.V.; data curation, A.M.; writing—original draft preparation, A.P.; writing—review and editing, M.E.; visualization, G.V.; supervision, P.G.; project administration, A.M.; funding acquisition, G.V. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** All procedures performed in the study were in accordance with the ethical standards of the institution and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

**Informed Consent Statement:** Informed consent was obtained from the relatives.

**Data Availability Statement:** All data are included in the main text.

**Acknowledgments:** The authors thank the Scientific Bureau of the University of Catania for language support.

**Conflicts of Interest:** The authors declare no conflict of interest.

**Ethical Approval and Consent to Participate:** All procedures performed in the study were in accordance with the ethical standards of the institution and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from relatives.

#### **References**


### *Article* **Anaphylactic Death: A New Forensic Workflow for Diagnosis**

**Massimiliano Esposito 1,† , Angelo Montana 1,† , Aldo Liberto <sup>1</sup> , Veronica Filetti <sup>2</sup> , Nunzio Di Nunno <sup>3</sup> , Francesco Amico <sup>1</sup> , Monica Salerno 1,\*, Carla Loreto <sup>2</sup> and Francesco Sessa <sup>4</sup>**


**Abstract:** Anaphylaxis is a life-threatening or fatal clinical emergency characterized by rapid onset, and death may be sudden. The margin of certainty about the diagnosis of anaphylactic death is not well established. The application of immunohistochemical techniques combined with the evaluation of blood tryptase concentrations opened up a new field of investigation into anaphylactic death. The present study investigated eleven autopsy cases of anaphylactic death, carried out between 2005 and 2017, by the Departments of Forensic Pathology of the Universities of Foggia and Catania (Italy). An analysis of the medical records was carried out in all autopsies. Seven autopsies were carried out on males and four on females. Of the eleven cases, one showed a history of asthma, one of food ingestion, two of oral administration of medications, six did not refer any allergy history, and one subject was unknown. All cases (100%) showed pulmonary congestion and edema; 7/11 (64%) of the cases had pharyngeal/laryngeal edema and mucus plugging in the airway; only one case (9%) had a skin reaction that was found during external examination. Serum tryptase concentration was measured in ten cases, and the mean value was 133.5 µg/L ± 177.9. The immunohistochemical examination using an anti-tryptase antibody on samples from the lungs, pharynx/larynx, and skin site of medication injection showed that all cases (100%) were strongly immunopositive for antitryptase antibody staining on lung samples; three cases (30%) were strongly immunopositive for anti-tryptase antibody staining on pharyngeal/laryngeal samples; and eight cases (80%) were strongly immunopositive for anti-tryptase antibody staining on skin samples. We conclude that a typical clinical history, blood tryptase level >40 µg/L, and strongly positive anti-tryptase antibody staining in the immunohistochemical investigation may represent reliable parameters in the determination of anaphylactic death with the accuracy needed for forensic purposes.

**Keywords:** anaphylactic death; diagnostic workflow; immunohistochemical investigation; blood tryptase level

#### **1. Introduction**

The term anaphylaxis was introduced in 1902 by Portier and Richet [1], and it refers to a serious, generalized or systemic, allergic or hypersensitivity reaction [2]. It can be a life-threatening or fatal clinical emergency with airway and circulatory impairments [3–6]. It is usually associated with skin and mucosal alterations (widespread hives, pruritus, and swollen lips/tongue/uvula) and gastrointestinal disorders (vomiting, diarrhea, and abdominal cramps) [7,8]. In particular, anaphylaxis is due to a systemic reaction mediated

**Citation:** Esposito, M.; Montana, A.; Liberto, A.; Filetti, V.; Nunno, N.D.; Amico, F.; Salerno, M.; Loreto, C.; Sessa, F. Anaphylactic Death: A New Forensic Workflow for Diagnosis. *Healthcare* **2021**, *9*, 117. https:// doi.org/10.3390/healthcare9020117

Academic Editor: Mariano Cingolani Received: 18 December 2020 Accepted: 20 January 2021 Published: 22 January 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

by vasoactive amines, released from mast cells, and basophils sensitized by immunoglobulin E (IgE) [7,9–12]. Conversely, anaphylactic shock (AS) is an anaphylactic reaction characterized by critical organ hypoperfusion after exposure to a previously encountered antigen [11,12]. The incidence and prevalence of anaphylaxis are difficult to establish. However, the incidence ranges from 1.5 to 7.9 per 100,000 person-years, but there has been an increase in admissions with anaphylaxis over the last two decades [3]. Moreover, the prevalence is 0.3% in the European population [3]. According to Chaudhuri et al. [13], the incidence of anaphylaxis in the United States ranged from 1.21% to 15.04% in the population. Furthermore, the risk of severe anaphylaxis has been estimated to be 1–3 per 10,000 person-years, while the risk of death due to anaphylactic shock is about 1–3 per million per year. Food and medications are responsible for most anaphylaxis reactions. However, virtually any agent capable of directly or indirectly activating mast cells or basophils can cause this syndrome [9–18]. Food is the cause of anaphylaxis in children most of the time, and drugs are major causes in adults, and are also the most frequent cause of anaphylaxis in hospitalized patients [12,19–22]. A higher frequency of anaphylaxis has been shown in adult females to food and non-steroidal anti-inflammatory drugs (NSAIDs) [8,12]. The common drugs responsible for anaphylaxis reactions are antibiotics, muscle relaxants, non-steroidal anti-inflammatory drugs, and radioactive contrast media [8,13]. Risk factors for severe anaphylaxis with hospitalization are old age combined with comorbidities such as cardiovascular disease (CVD) and chronic obstructive pulmonary disease [14,15,22–26].

Tryptase is an abundant secretory granule-derived serine proteinase contained in mast cells. The tryptase enzyme is the only protein that is specific for human mast cells, and tryptase plasma levels reflect the clinical severity of anaphylaxis. Elevated levels of serum tryptase occur in both anaphylaxis and anaphylactic reactions, but a negative test does not exclude anaphylaxis [27].

In this study, an investigation of the serum tryptase levels combined with the immunohistochemical expression of tryptase in specimens from the lungs, glottis, and skin (site of medication injection) in eleven autopsy cases was performed to clarify and discuss their significance in anaphylactic death.

#### **2. Materials and Methods**

#### *2.1. Sample Collection*

A retrospective analysis of the autopsy records of the Departments of Forensic Pathology of the University of Foggia and Catania (Italy), was carried out between 2005 and 2017. From the analysis of death scene investigations and autopsy reports, together with the information gathered from the police, eleven cases of anaphylactic death origin were selected. Cases with weak or missing information about the manner of death were excluded. Decomposed bodies were also excluded from the study. All procedures performed in this study were in accordance with the ethical standards of the institution and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all relatives.

The deceased were four men and seven women, ranging in age from 16 to 69 years (average: 47 years ± 17.69). In all cases of anaphylactic death, sections of lungs, glottis and the skin site of medication injection were collected. Eleven cases were selected as controls. These included the following cases: seven cases of sudden cardiac death and four cases of fatal motor vehicle crashes. All control cases were selected for their negative clinical histories for manifestations of asthma or allergies. All autopsies were performed within four days after death was determined, and all cadavers were stored at −4 ◦C.

#### *2.2. Histological Analysis*

A routine microscopic histopathological study was performed using hematoxylin– eosin (H&E) staining. Specimens from the lungs, glottis and skin were fixed in 10% buffered formalin, as previously described [28]. After an overnight wash, specimens were dehydrated in graded ethanol, cleared in xylene and paraffin-embedded. Tissue paraffin

blocks were then cut (4 µm thickness) using a microtome and sections were mounted on silane-coated slides (Dako, Glostrup, Denmark) and stored at room temperature. Sections then were stained with H&E and observed using a Zeiss Axioplan light microscope (Carl Zeiss, Oberkochen, Germany) for morphological examination. Finally, representative micrographs were captured using a Zeiss AxioCam MRc5 digital camera (Carl Zeiss, Oberkochen, Germany).

#### *2.3. Immunohistochemical Staining*

Immunohistochemical investigation of samples from the lungs, glottis and skin was performed using anti-tryptase antibodies. For the immunohistochemical analysis, specimens were processed as previously described [28,29]. In particular, sections were dewaxed in xylene, rehydrated with graded ethanol and then incubated for 20 min in 0.3% H2O2/methanol solution to block endogenous peroxidase activity. After rinsing for 20 min with phosphate buffered saline (PBS), slides were pre-treated to facilitate antigen retrieval and to increase membrane permeability to antibodies using a microwave oven (750 W) (5 min × 3) in capped polypropylene slide-holders with citrate buffer (10 mM citric acid, 0.05% Tween 20, pH 6.0; Bio-Optica, Milan, Italy) and then incubated overnight at 4 ◦C with anti-tryptase monoclonal antibodies (Agilent Dako, Copenhagen, Denmark) diluted 1:100 in PBS. The detection system used was the LSAB+ kit (Dako, Copenhagen, Denmark) incubated for 10 min at room temperature, a refined avidin–biotin technique in which a biotinylated secondary antibody reacts with several peroxidase-conjugated streptavidin molecules. The positive reaction was visualized by 3,3-diaminobenzidine (DAB) peroxidation (DAB substrate Chromogen System; Dako) according to standard methods [30]. The sections were counterstained with Mayer's hematoxylin (Histolab Products AB, Göteborg, Sweden) mounted in GVA (Zymed Laboratories, San Francisco, CA, USA). The sections were observed and photographed as described above.

#### *2.4. Evaluation of Immunohistochemistry (IHC)*

The anti-tryptase immunoreaction was identified as either negative or positive. Immunohistochemical positive staining was defined by the presence of brown chromogen detected on the edge of the hematoxylin-stained cell nucleus, distributed within the cytoplasm or in the membrane via evaluation by light microscopy. Positive controls consisted of tissue specimens with known antigenic positivity. Sections treated with PBS without any primary antibody served as negative controls. Seven fields of about 600,000 µm<sup>2</sup> , randomly selected from each section, were considered for morphometric and densitometric analysis. The percentage of the areas (morphometric analysis) stained with anti-tryptase antibody was expressed as % positive dark brown pixels of the analyzed fields. Moreover, the levels (high/low) of staining intensity of positive areas (densitometric analysis) were expressed as densitometric count (pixel<sup>2</sup> ) of positive dark brown pixels in the analyzed fields. These parameters were calculated using software for image acquisition (AxioVision Release 4.8.2-SP2 Software, Carl Zeiss Microscopy GmbH, Jena, Germany). Data are expressed as mean ± standard deviation (SD). Digital micrographs were taken and fitted as previously described. The samples were also examined with a confocal microscope, and a three-dimensional reconstruction was performed (True Confocal Scanner, Leica TCS SPE).

#### *2.5. Statistical Analysis*

Statistical analysis was performed using GraphPad Prism 7.0 (GraphPad Software, Inc., La Jolla, CA, USA). The Shapiro–Wilk normality test was used for the calculation of the distribution of the samples. Unpaired t-tests were used for the comparison between the levels of tryptase staining intensity of positive areas (pixel<sup>2</sup> ) of cases of anaphylactic death and controls. *p*-values of less than 0.05 (*p* < 0.05) were considered significantly different.

#### *2.6. Serum Tryptase Assay*

Samples of femoral blood were obtained via a transcutaneous femoral approach (from the femoral artery) in eleven post-mortem examinations, of which all subsequently underwent full autopsy. Serum was derived from whole blood by centrifugation, decanted into plastic test-tubes and stored at −80 ◦C. Samples were shipped on ice to the Industrial Bio-Test (IBT) Reference Laboratory (Florence, Italy) for analysis. Information regarding the cause of death was hidden from the reference laboratory performing the assays. Serum tryptase levels were determined using a competitive immunofluorescent enzyme assay with monoclonal anti-human tryptase antibodies against both the A and B structural types of tryptase. These antibodies were incubated with a serum aliquot; the sample was washed, and enzyme-labelled anti-tryptase was added, followed by incubation. The sample was washed a second time, the developer was added, and the fluorescence in the aliquot was measured. The amount of fluorescence given off by the sample was directly proportional to the concentration of tryptase in the sample. Through a radioimmunoassay method, which only detected the β form of the tryptase molecule. Eleven cases were selected as controls. These included the following cases: seven cases of sudden cardiac death and four cases of death after motor vehicle crashes.

#### **3. Results**

Table 1 shows the clinical history of all selected cases, the cause of the anaphylactic reaction, and the interval between the onset of symptoms and death. According to the results shown in Table 1, 72.7% of our cases did not have a history of allergy; only 1/11 had a history of asthma and celiac disease; 5/11 (45.4%) died within 1 h, and 6/11 (54.6%) within 1 min. The causes of anaphylaxis were medications (6/11), injected contrast medium (3/11), food (1/11), and latex (1/11).


**Table 1.** The circumstantial data of the selected cases.

The post-mortem diagnosis was based on: (1) the circumstantial evidence, including the history of exposure to a likely allergen prior to death and clinical presentation; (2) postmortem findings suggesting an anaphylactic reaction, such as laryngeal edema, mucous plugging in the airways, erythematous skin rash and edema, eosinophilia in the mucosa and submucosa of the respiratory and the gastrointestinal tracts, and marked pulmonary congestion and edema; (3) toxicology test of serum concentrations of tryptase in femoral blood samples (from the femoral artery); (4) histological examination of all the organs using H&E; and (5) immunohistochemical examination for anti-tryptase antibody staining. The standard upper limit of total serum tryptase level has been established and was set at 40 µg/L in the Office of the Departments of Forensic Pathology.

The cases were analyzed regarding: (1) circumstantial evidence, including history of exposure; (2) post-mortem examination findings, including histological study and toxicology testing; and (3) cause of death. The data that were analyzed were extracted from the police investigation report, medical records, interviews of the victim's family members and reports by forensic pathologists or investigators, and forensic autopsy protocols.

On the basis of these results, a new workflow as a useful tool in anaphylaxis deaths was elaborated.

#### *3.1. Autopsy Findings*

All cases showed pulmonary swelling and edema during autopsy. Macroscopic examination during the autopsies revealed that 64% of the cases had pharyngeal/laryngeal edema and mucus plugging in the airways. Only one case (9%) had a skin reaction that was found during the external examination. The results of toxicological analyses performed on ante- and post-mortem samples (blood and urine) were negative for alcohol, drugs and medications.

#### *3.2. Histological and Immunohistochemical Analysis*

All cases displayed pulmonary congestion and edema during the histological examination. The glottis was sampled in six cases, and the skin was sampled in eight (only in cases of transdermal administration of medication). An immunohistochemical examination of anti-tryptase antibody staining on lung samples was performed in ten autopsies, glottis in six, and skin site of injected medications in eight. All cases showed strong immunopositivity for anti-tryptase antibody staining on lung samples (10/10), on pharyngeal/laryngeal samples (7/7), and on skin samples (8/8) (Table 2). Samples from the lung, skin site of injected medications and glottis showed a strong and diffuse anti-tryptase immunolabeling. In particular, in lung specimens, anti-tryptase was found in mast cells of the connective interstitium and bronchiolar structure (Figure 1a). The skin site of medication administration also showed strong mast cell antibody immunolabelling in the connective derma (Figure 1b). Moreover, the glottis of these cadavers exhibited an overexpression of anti-tryptase antibody staining scattered in the laminar connective tissue at the vocal fold level (Figure 1c). The Shapiro–Wilk normality test showed that the level of tryptase staining intensity of positive areas (pixel<sup>2</sup> ) in cases of anaphylactic death and controls differs significantly from a normal distribution. As shown in Figure 2, the level of tryptase staining intensity revealed that in the tissue of cadavers who had died from anaphylactic shock the tryptase immunostaining level was much higher compared to controls (*p* < 0.05). In particular, these results of tryptase immunostaining were confirmed in lung tissue (Figure 2a), skin tissue (Figure 2b), and glottis tissue (Figure 2c). Figure 3a summarizes the histological results (H&E), while in the other quadrants, the anti-tryptase immunohistochemical staining results are shown (Figure 3b–d). Figure 4 shows the anti-tryptase immunohistochemical results in lung samples by confocal laser scanning microscopy or with a light microscope (Figure 4a–d).


**Table 2.** Concentration of total serum tryptase, autopsy findings and histological examination. H&E: hematoxylin–eosin; IHC: immunohistochemistry.

**Figure 1.** (**a**) Lung specimens from a cadaver who had died of anaphylactic shock; anti-tryptase antibody staining is strongly expressed in mast cells (black arrows) in the peribronchial interstitium. The insert shows the immunostaining software image analysis of Figure 1a, in which a highly immunostained area (red color) was detected (magnification: 20×; scale bar: 5 µm). (**b**) Skin specimens of the gluteus where medication administration occurred from a cadaver who had died of anaphylactic shock; anti-tryptase immunolocalization (black arrows) was demonstrated in the derma of the medication injection site. The insert shows the immunostaining software image analysis of Figure 1b, in which a highly immunostained area (red color) was detected (magnification: 20×; scale bar: 5 µm). (**c**) Glottis specimens of a cadaver who had died of anaphylactic shock showed strong anti-tryptase immunoexpression in mast cells (black arrows). The insert shows the immunostaining software image analysis of Figure 1c, in which a highly immunostained area (red color) was detected (magnification: 20×; scale bar: 5 µm).

**Figure 2.** Comparison of the densitometric analysis (pixel<sup>2</sup> ) of the tryptase immunostained area expressed by positive, dark brown pixels in the analyzed fields for: (**a**) lung tissues (*n* = 10) of cadavers who had died of anaphylactic shock vs. controls (ctrl); (**b**) skin tissue (*n* = 8) of cadavers who had died of anaphylactic shock vs. ctrl; (**c**) glottis tissue (*n* = 7) of cadavers who had died of anaphylactic shock vs. ctrl. Data are presented as mean ± standard deviation (SD) (*p* < 0.05).

**Figure 3.** (**a**) H&E examination of lung samples shows capillary congestion and severe alveolar edema (amplification: 20×; scale bar: 5 µm). (**b**) IHC examination of lung samples with strong anti-tryptase immunopositivity (magnification: 40×; scale bar: 5 µm). (**c**) IHC examination of pharyngeal samples with strong anti-tryptase immunopositivity (magnification: 40×; scale bar: 5 µm). (**d**) IHC examination of skin samples with strong anti-tryptase immunopositivity (magnification: 20×; scale bar: 5 µm.

**Figure 4.** Comparison of anti-tryptase antibody reaction between confocal laser scanning microscopy (**a**,**c**) and light microscopy (**b**,**d**) shows strong anti-tryptase immunopositivity in: (**a**,**b**) lung tissue; (**c**,**d**) pharyngeal tissue. Magnification: 40×; scale bar: 5 µm.

#### *3.3. Serum Tryptase Analysis*

Routine toxicology tests were performed in all eleven autopsies. Serum tryptase analysis ranged from 40.5 µg/L to 640 µg/L, and the mean value was 133.5 µg/L ± 177.9.

Table 2 shows a summary of total serum tryptase, autopsy findings, and histological examinations. All cases showed pulmonary congestion and edema during autopsy and the histological examination.

#### **4. Discussion**

Today, there is no specific forensic workflow in cases of death from anaphylactic shock. A systematic approach would allow forensic pathologists to arrive at a confident diagnosis of death from anaphylactic shock. Through a retrospective analysis of eleven deaths from anaphylactic shock, the aim of this study was to propose a new forensic workflow, leading to a more accurate diagnosis.

The present study shows that a typical clinical history, high levels of serum tryptase (>40 µg/L), and strong positivity for anti-tryptase antibody staining are highly suggestive for establishing the diagnosis of anaphylactic death.

Anaphylaxis is a life-threatening syndrome [3–6]. The anaphylactic reaction is mostly triggered by food and drugs, but it may be provoked by any agent capable of activating mast cells or basophils [9,24]. In the UK, about half of the 20 fatal reactions recorded each year due to anaphylactic shock are iatrogenic; the rest are caused by food ingestion or insect venom. Respiratory or cardiac arrest occurs within 30 min for food, 15 min for venom, and 5 min for iatrogenic reactions [30]. A history of exposure to anaphylactic stimuli and

clinical features such as hypotension are important to identify death from anaphylactic shock [31].

Several epidemiologic and experimental studies [9,28,32–46] have underlined the importance of immunohistochemical analyses and the concentrations of serum tryptase; however, based on the literature, this article is the first study that combines the two parameters for a specific diagnosis of anaphylactic death.

This study showed that the symptoms of the anaphylactic reaction occurred within one hour (Table 1): one minute in the case of injected contrast medium reaction, in cases of anaphylaxis during anesthesia, the shock occurred within one minute; in cases of medication, anaphylaxis shock occurred both within one minute (60%) and within one hour (40%). These results are in agreement with previous studies [3–6].

The autopsy procedure has to be careful, with an accurate external examination, searching for injection sites of stinging or biting invertebrates, as well as blood, vitreous, and urine collection. It is essential to examine the stomach contents, above all in suspected cases of anaphylactic shock from food [47]. Autopsy findings, such as the formation of mucus plugs, congestion and intra alveolar hemorrhages, and congestion and edema of major organs, are not exhaustive or specific for the diagnosis of fatal anaphylaxis [37]. Immunohistochemical analysis using anti-tryptase antibodies is also not exhaustive for the diagnosis of death from anaphylactic shock [48–50].

According to Turillazzi et al. [39], in all cases, the larynx and pharynx were opened with forceps following the posterior median line and the glottis was observed; the sides were stretched outward to study the mucosa. Then, following the Ghon technique, abdominal organs were removed using the bloc method, taking care to preserve the integrity of vascular structures. All autopsies showed pulmonary congestion and edema of the lungs. When squeezing the lungs, in all cases, an abundant, reddish-colored liquid was observed. Macroscopic examination during the autopsies showed glottis edema and mucus plugging in the airways in 64% of cases. Only one case had a skin reaction that was found during the external examination.

In our study, the results of the histological and immunohistochemical analyses showed generalized stasis with areas of acute pulmonary emphysema in all autopsies, and in deaths of subjects over 40 years old, eosinophilic cross-bands ranging from segments of hypercontracted to coagulated sarcomeres in heart samples. The immunohistochemical examination of anti-tryptase antibody staining on samples from the lungs, glottis, and skin of medication injection sites revealed strong positivity for anti-tryptase antibody staining for all sampling sites in all cases. In particular, in lung specimens, anti-tryptase was found in mast cells of the connective interstitium and bronchiolar structure. Skin sites of medication administration also showed strong mast cell antibody immunolabelling in the connective derma. Moreover, the glottis of these cadavers exhibited a high level of anti-tryptase antibody staining scattered in the laminar connective tissue at the vocal fold level.

According to the guidelines on autopsy practice for suspected acute anaphylaxis of the Royal College of Pathologists [49], serum tryptase samples should always be collected, even if an autopsy is performed days or even weeks after death. Despite an average serum tryptase concentration, anaphylactic death cannot be completely excluded. Different sampling techniques can impact post-mortem tryptase levels [49,51]. A recent study demonstrated that the level of tryptase is significantly lower in samples collected via transcutaneous aspiration compared with femoral/external iliac vein samples [38]. In fact, for post-mortem tryptase analysis, a sample from a clamped femoral/external iliac vein should be defined as the gold standard [38,52,53]. There are doubts about the variability of serum tryptase by post-mortem interval (PMI). Mast cells present in the respiratory tract and heart and post-mortem cell lysis might influence the release of tryptase; for this reason, peripheral blood (i.e., femoral blood) is preferable to central blood [49,53]. After death, mast cell tryptase is very stable with a long half-life, and it could be measured up to four days after death [49].

There are four different tryptases (α, β, γ, and δ), but only the α and β form are medically necessary; during an anaphylaxis reaction, they are released by mast cells [10]. Tryptase has proinflammatory effects such as the promotion of tissue edema and remodeling, chemokine secretion, and neutrophil recruitment [10]. The tryptase level can be increased by cell autolysis or liquefaction [54]. Higher values of tryptase serum have also been found in other types of death, such as sudden infant death syndrome (SIDS), amniotic fluid embolism, and heroin-related deaths [53–59]. β-tryptase is a more reliable indicator of acute mast cell activation. It is emitted at the same time as histamine, but the release is slower, making this marker more suitable for post-mortem investigation [60]. In 1998, Edison et al. [51] proposed that the cut-off level of tryptase should be 10 µg/L. Subsequently, in 2007, Edston et al. [45] modified the value to over 20 (44.5 µg/L) in femoral samples. In 2011, Mayer et al. [44] recommended a cut-off level over 45 µg/L. In 2014, McLean-Tooke et al. [30] modified the cut-off level to 110 µg/L in aortic samples. Finally, in 2017, Xiao et al. [48] established a cut-off level of 43 µg/L using femoral samples. There are few studies on the change in the cut-off of tryptase levels in cases of cardiac blood samples. However, it is easy to consider that the cut-off is the same as for aortic sampling (110 µg/L) with the same reliable margin (sensitivity of 80% and specificity of 92.1%). According to Tse et al. [42], sensitivity reaches 100% when the cut-off of tryptase is between 11.4 and 30 (µg/L), but specificity is low; specificity reaches 100% when the cut-off is above 70 (µg/L). In the case of aortic blood samples, the cut-off is 110 µg/L (sensitivity 80% and specificity 92.1%) [30,37].

In this retrospective analysis, tryptase serum determination was performed as part of all autopsies (11/11). The concentration ranged from 40.5 µg/L to 640 µg/L with a median of 133.5 µg/L ± 177.9. All autopsies were performed within four days after death and all cadavers were stored at −4 ◦C; this did not change the validity of the test. In fact, storing a corpse at −4 ◦C after death does not affect tryptase levels, as has been shown by previous studies. Sravan et al. [50] performed autopsies three days after death with storage at 4 ◦C. Edston et al. [45] published their study in which the mean time between death and autopsy was 3.861days. Tse [42] reported two cases in which there was an analysis of tryptase levels at three days and six days after death.

The combination of anamnestic information, autopsy findings, tryptase serum determination, and immunohistochemical testing can help to make a diagnosis of anaphylactic reaction as the cause of death in patients who died suddenly with unspecific symptoms.

The post-mortem diagnosis of anaphylactic shock is a challenge, and it is often achieved by exclusion. A limitation of this study is the small sample size of the analysis. In this regard, we suggest future studies to confirm our observations.

A sampling of serum tryptase is mandatory [49]. However, a literature review revealed that there are many doubts about its cut-off, sampling site (central or peripheral), and changes during the post-mortem interval (PMI). Histological and immunohistochemical investigation, through the use of the confocal microscope, help in the diagnosis. The results of the present study suggest that through the use of the blood tryptase concentration, together with the immunohistochemical investigation for anti-tryptase antibody staining in samples from the lung, glottis, and skin (at the site of administration of medications and contrast medium), it is possible to realize a very reliable diagnostic workflow of anaphylactic death (Figure 5). In fact, previous studies reported in the literature [61–68] have not clearly expressed how to establish a specific diagnosis of anaphylactic death. This diagnostic workflow should be used to establish an anaphylactic reaction as the cause of death with a large margin of certainty.

**Figure 5.** Proposed workflow to detect anaphylactic shock in fatal cases. In red text the goal of the flow chart.

**Author Contributions:** Conceptualization, M.E. and A.M.; methodology, V.F. and C.L.; software, C.L.; validation, N.D.N., A.L. and M.S.; formal analysis, V.F.; investigation, C.L.; resources, M.E.; data curation, F.A.; writing—original draft preparation, M.E.; writing—review and editing, M.E.; visualization, M.S.; supervision, C.L. and F.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** All procedures performed in the study were approved by the Scientific Committee of the University of Foggia, (code: 09/2018) and were performed in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** All data are included in the main text.

**Acknowledgments:** We wish to thank the Scientific Bureau of the University of Catania for language support.

**Ethical Approval and Consent to Participate:** All procedures performed in the study were in accordance with the ethical standards of the institution and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from the relatives.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

