Acute anaphylaxis is an uncommon but well-recognized cause of sudden death. The mechanism of death is usually attributed to either profound hypotensive shock resulting from peripheral vasodilatation or asphyxia secondary to upper airway edema and/or bronchospasm. Antemortem clinical manifestations vary and usually include combinations of symptoms such as generalized flushing, urticaria, angioedema, vomiting and diarrhea, cough, shortness of breath, and loss of consciousness.
In severe reactions, cardiorespiratory arrest can occur within minutes of symptom onset. Asphyxia can follow upper airway obstruction as a result of pharyngeal or upper airway edema or lower airway obstruction as a result of bronchospasm, in some cases with mucous plugging. Cardiac arrest can follow respiratory arrest or can occur without respiratory difficulty as a result of either direct effects of mediators of anaphylaxis on the heart or profound shock resulting from peripheral vasodilatation often combined with angioedema causing loss of intravascular fluid.
Postmortem diagnosis of fatal anaphylaxis is frequently hampered by relative lack of specific pathologic findings in these deaths. In some cases, upper airway edema, petechial hemorrhages, and hyperinflated lungs have been reported, along with nonspecific visceral congestion and pulmonary edema.1
In the present study, we investigated postmortem findings after anaphylactic reactions to drugs at autopsies performed in Turkey between January 2001 and June 2006. The autopsy findings from 36 cases of fatal anaphylaxis were described in detail.
MATERIALS AND METHODS
The study included all fatal anaphylactic reactions sent to the Council of Forensic Medicine between January 2001 and June 2006. Inclusion criteria were (1) death following documented anaphylactic reaction to a known allergic drug, death known to be associated with anaphylaxis; (2) no other cause of death determined at a thorough autopsy examination. Cases that did not fulfill the above mentioned criteria were excluded from the study.
The Council of Forensic Medicine is an official organ of the Ministry of Justice and the only official expert in forensic medicine in Turkey. The duties and responsibilities of the Council of Forensic Medicine are determined by the law. Each year approximately 85,000 reports about scientific and technical subjects related to forensic sciences requested by the courts and the district attorneys are written by the council. The Council of Forensic Medicine includes specialized departments such as morgue and toxicology departments and branches such as the first branch, which deals with autopsy, toxicological analyses, and medical and legal records.
Details of autopsies performed in the cities of Turkey since 2001 have been recorded in a database in the first branch, which deals with the cause of death. The first branch includes a general surgeon, a cardiovascular surgeon, a neurosurgeon, a gynecologist, an internist, a cardiologist, a hematologist, an immunologist, a pathologist, and a forensic specialist. The study included the cases in which the cause of death was documented as anaphylaxis or anaphylactic reactions to a single dose of certain drugs in the reports from the first branch of the Council of Forensic Medicine.
There were 36 cases of fatal anaphylaxis due to drugs or intravenous contrast medium identified by the first branch of the Council of Forensic Medicine in Turkey between January 2001 and June 2006.
All cases in which the source of allergy was identified had a history of collapse shortly after injection of the suspected allergen. Death occurred within 1 hour of the onset of anaphylaxis in 27 cases. The mechanisms of death were attributed to sudden circulatory collapse in 27 cases in which profound hypotension was clearly documented at the time of death. Five cases survived the initial attack, but died subsequently of hypoxic brain injury. The mechanisms of death were difficult to ascertain in the remaining 4 cases as they presented with “sudden collapse” and died before medical assessment was performed.
About 30 deaths occurred within a health unit, such as an emergency department, a clinic, or a pharmacy. Out of these 30 deaths, 19 were due to treatment of an infection; 6 due to treatment of pain such as headache and muscle pain and 5 were due to prophylaxis of acute rheumatic fever. Two postoperative deaths and 1 intraoperative death occurred within the hospital setting. Two deaths occurred at home and 1 in a bawdyhouse.
We found that the majority of cases of fatal anaphylaxis (72.2%) were caused by antibiotics. The cause of anaphylaxis was penicillin in 12 deaths, cephalosporin in 10 deaths, aminoglycosides in 2 deaths, amphenicol in 1 death, and lincosamide in 1 death. A total of 7 cases of fatal anaphylaxis were caused by the analgesics methimazole and diclofenac. Anaphylaxis was due to meglumine amidotrizoate in 1 case, H2 receptor antagonist in 1 case, and propofol in 1 case. Drugs that caused anaphylaxis are summarized in Table 1.
A drug called Ceftriaxone was prescribed to a woman (born in 1981), who was working in a bawdyhouse. Five minutes after the intramuscular injection of the second dose, nausea and vomiting started; she was short of breath, lost her consciousness, and died.
A drug called Gentamycin was administered to a man (born in 1977); 5 to 10 minutes later, his face swelled and he experienced shortness of breath, rapidly lost his consciousness, and died.
None of the cases were known to have previous allergic reactions to the agent responsible for the fatal event. Death was due to antibiotics in 26 cases, analgesics in 7 cases, H2 receptor antagonist in 1 case, an intraoperative drug in 1 case, and intravenous contrast medium in 1 case. Clinical findings are summarized in Table 2.
The principal autopsy findings are summarized in Table 2. The most common autopsy findings were nonspecific pulmonary congestion and edema, which were reported in 29 cases (80.5%). Upper airway edema was present in 11 cases (28.2%), while cutaneous edema was seen in 2 cases (5.5%). Hyperinflation of the lungs and/or mucous plugging of airways, suggesting an asthmatic component, were noted in 5 cases, 4 of which were immediate deaths. Petechial hemorrhages, also suggesting an asphyxial component of death, was found in 3 cases, all of which were immediate deaths. Of 36 cases, 5 (13.8%) showed signs of global cerebral ischemia at autopsy. In each of these 5 cases, time elapsing from the onset of anaphylaxis to death ranged from several hours to 3 days. Atherosclerotic coronary vascular disease was reported in 15 cases (41.6%), including 3 cases with severe disease, 3 with moderate involvement, and 9 with mild involvement. Another important pathology identified at autopsy was carcinoma of the liver. In 4 cases, no positive findings relevant to the cause of death were identified at autopsy and the diagnosis of fatal anaphylaxis was made on the basis of clinical history and eye witness records.
Histology was performed in all autopsies. One case had upper airway edema. Histology of the bronchi demonstrated mucosal edema, inflammation, and epithelial sloughing in 3 of 5 cases with mucous plugging identified macroscopically.
This is the first study to report postmortem findings, clinical reports, and eye witness accounts in Turkey in an unselected series of fatal anaphylactic reactions to drugs. This study did not show the frequency of anaphylactic deaths from drugs at autopsies in Turkey. Because the courts and the district attorneys ask the Council of Forensic Medicine how death happens only when there is a suspicious situation, all anaphylactic deaths are not sent to the Council of Forensic Medicine. The reported 36 cases were representative of deaths thought to be caused by anaphylaxis or anaphylactic reactions.
Anaphylaxis is an acute, systemic type 1 hypersensitivity reaction that occurs after reexposure to an antigen. Anaphylactic reaction is caused by cross-linkage of antigen-specific IgE molecules bound to the surfaces of tissue mast cells and peripheral blood basophiles, which then undergo degranulation and release a variety of potent mediators including histamine and tryptase.2 Other mediators, such as certain prostaglandins and leukotrienes, are also generated via the arachidonic acid metabolism cascade. These mediators act on receptors that induce mucus production, increase vascular permeability, and stimulate smooth muscle constriction in various organs. Clinical manifestations of anaphylactic reactions can vary from organ-specific symptoms such as urticaria, acute respiratory distress, abdominal pain, diarrhea, and vomiting to upper airway edema and often marked facial angioedema. Fatal cases occur either due to severe bronchospasm with respiratory arrest or hypertensive shock with rapid circulatory collapse, in which case death may occur prior to the onset of any other symptoms.
Epidemiological studies on life-threatening and lethal anaphylactic reactions in the United States, the United Kingdom, and the other European countries have revealed an incidence of 1 to 3 per 10,000 people for severe anaphylaxis and 1 to 3 per million people for fatal cases.3 In Turkey, only when how death has occurred is not clear, the courts and the district attorneys send witness accounts, medical records, and autopsy reports to the first branch of the Council of Forensic Medicine and ask them to determine the cause of death. Otherwise, they do not request reports about deaths, the cause, and the manner of which are clear. This study, although limited in scope, did not demonstrate a high incidence of anaphylactic deaths in Turkey.
Most of the cases in this study were known to have a history of drug allergies and an asthmatic component was documented in 7 cases. In fact, the majority of the cases of fatal anaphylaxis (61.1%) were caused by penicillin and cephalosporin. The chemical structures of these drugs are very similar. For this reason, both drugs cause an allergic reaction and cross reactivity.4
A study by Pumphrey and Roberts1 demonstrated that 21 of 56 deaths were due to drug reactions. Likewise, Low and Stables5 demonstrated that 10 of 18 deaths were due to drug reactions. The median age was 63 years in the former study and 53.5 years in the latter study. In this study, the median age was 26.7 years. Turkey is a developing country and has a high rate of younger population than developed countries. In Turkey, using antibiotics without physicians' prescriptions is a significant problem. In fact, in the present study, we found that antibiotics and analgesics recommended by neighbors or friends were used in some cases.
The relationship between asthma and severe anaphylactic reactions, particularly to ingested food such as nuts, cows milk, and eggs, are well established. In a study by Sampson et al,6 all 13 cases of fatal and near-fatal food anaphylaxis occurred in children and adolescents who were known to be asthmatics. Another study by Bock et al7 revealed that 30 of 32 cases of fatal anaphylactic reactions were due to nuts. Most of the victims were adolescents or young adults, and all but 1 case had asthma with known food allergies. The mechanisms of death were most commonly attributed to severe bronchospasm with respiratory arrest. Given the similarities in clinical presentation, it is not uncommon for anaphylactic reactions in asthmatic patients to be mistaken for severe asthma attacks. In this study, 7 cases that had asthma accompanied by infection of the respiratory tract used antibiotics and died instantly. Because there is a relationship between asthma and anaphylactic reactions, care should be taken with the choice of drugs for the treatment of infections in asthmatic patients.
At autopsy a significant number of cases showed nonspecific pulmonary congestion and edema (80.5%), which is consistent with the literature.1,5 Nonspecific pulmonary congestion and edema are the important nonspecific findings at autopsy.
We found that 19 of the 36 anaphylactic deaths had no macroscopic postmortem findings suggestive of anaphylaxis. If the probability of identifying deaths as anaphylactic is greater, then recognizable features can be established at autopsy, but the true fraction of deaths with negative findings is likely to be higher. Other than single case reports,8,9 there has been no previous systematic study of postmortem findings of anaphylactic deaths from all causes.
We found that upper airway edema was present in 11 cases (28.2%). Similarly, in a study of 43 fatal drug-related reactions, Delage and Irey10 reported upper airway edema and bronchial obstruction in only 22% and 25% of cases, respectively. In another study, Pumphrey and Roberts1 showed that upper airway edema was more common in deaths related to food allergens (77% of immediate deaths) than in those after reactions to venom or drugs (40% and 30% of immediate deaths, respectively).
In view of the frequent absence of any specific findings at autopsies, other ancillary tests have been proposed to increase the diagnostic specificity of anaphylactic deaths. Of these tests, serum tryptase measurements have been most widely used. The peak level of serum tryptase occurs within 2 hours of the onset of anaphylaxis and then declines following first-order kinetics with a half-life of around 2 hours.11 Therefore, blood samples should ideally be collected within hours of symptom onset. Cases in which death is delayed, the assay should be performed on samples taken during the acute phase of the illness. Once the sample has been collected, the tryptase levels remain relatively stable and do not appear to be significantly affected by the temperature at which samples are stored.12
Elevated levels of serum tryptase have consistently been demonstrated in postmortem samples from cases of suspected anaphylactic death.13 A study by Randall et al,12 however, reported tryptase levels to be above the normal threshold of 1 mg/L in 31 of 49 autopsy cases in which there was no evidence to suggest anaphylactic causes for the deaths. In Turkey, serum tryptase level measurement is not performed routinely at autopsy. It is a significant deficiency likely to affect autopsy results in anaphylactic reactions.
A study by Edston and van Hage-Hamsten14 investigated the optimal cut-off value for tryptase levels. They found that a level of 10 mg/L was ideal in terms of high sensitivity (86%) and specificity (88%). In a study by Greenberge et al,15 serum tryptase levels were measured in 7 elderly cases of fatal anaphylaxis and 4 cases were found to be high tryptase levels. Therefore, serum tryptase measurements are useful in the evaluation of sudden unexplained deaths particularly when the clinical history is suggestive of an anaphylactic reaction. A raised serum tryptase level on its own, however, is not diagnostic of fatal anaphylaxis.12
Russell and Smith reported a case of hypotension and cardiac arrest from spinal anesthesia and general anesthesia. Having suspected of anaphylaxis, they performed mast cell tryptase test 90 minutes after deep hypotension and skin test 12 weeks later. High tryptase levels on repeated measurements confirmed mastocytosis and explained that hypotension and cardiac arrest were the result of spinal anesthesia. They suggested that measurements of serum tryptase levels made weeks after a fatal event can be used to distinguish mastocytosis from anaphylaxis when serum tryptase levels were persistently high throughout the event although no allergic agents were detected.16
However, another study showed that tryptase levels were not an effective marker in diagnosis of IgE-dependent food allergy in children with allergic reaction to one or more food allergens. In fact, patients with IgE-dependent food allergy may have elevated levels of tryptase with no clinical manifestation. Clinicians should use caution when using serum tryptase to refute or support a diagnosis of anaphylaxis. Serial tryptase measurement increases sensitivity and specificity.17,18
In conclusion, this is the first study on anaphylactic deaths from drugs based on autopsies performed in Turkey. The results of this study are largely consistent with those reported from other studies in the world literature and confirm that anaphylaxis is an uncommon cause of sudden death. Unavailability of postmortem measurement of serum tryptase levels routinely in Turkey is an important deficiency. However, nonspecific autopsy findings, typical clinical history, eye witness records can be diagnostic aids.