Mushroom Poisoning—A 17 Year Retrospective Study at a Level I University Emergency Department in Switzerland

The consequences of mushroom poisoning range from mild, mostly gastrointestinal, disturbances to organ failure or even death. This retrospective study describes presentations related to mushroom poisoning at an emergency department in Bern (Switzerland) from January 2001 to October 2017. Gastrointestinal disturbances were reported in 86% of the 51 cases. The National Poisons Information Centre and mycologists were involved in 69% and 61% of the cases, respectively. Identification of the mushroom type/family was possible in 43% of the cases. The most common mushroom family was Boletaceae (n = 21) and the most common mushrooms Xerocomus chrysenteron (n = 7; four being part of a cluster), Clitocybe nebularis, Lepista nuda and Lactarius semisanguifluus (n = 5 each, four being part of a cluster). Poisonous mushrooms included Amanita phalloides (n = 3, all analytically confirmed), Boletus satanas (n = 3), Amanita muscaria (n = 2) and Amanita pantherina (n = 2). There were no fatalities and 80% of the patients were discharged within 24 h. Mushroom poisoning does not appear to be a common reason for emergency consultation and most presentations were of minor severity and related to edible species (e.g., due to incorrect processing). Nevertheless, poisonous mushrooms and severe complications were also recorded. Collaboration with a poison centre and/or mycologists is of great importance, especially in high risk cases.


Introduction
Of the estimated 5000 existing mushroom species, only 200-300 have been established to be safely edible, while 50-100 are known to be poisonous to humans and the toxicity profile of most other species has not been investigated [1,2]. Depending on the species, consumption of poisonous mushrooms can cause various clinical signs and symptoms, ranging from mild, mostly gastrointestinal, disturbances to organ failure and death [2][3][4][5]. Poisonous mushrooms contain a variety of different toxins whose potency is influenced by many extrinsic and intrinsic factors present around the mushroom [6]. Main categories of mushroom toxins include protoplasmic poisons (result in generalized destruction of cells, followed by organ failure), neurotoxins (which cause neurological symptoms such as coma, convulsions and hallucinations), and gastrointestinal irritants (produce nausea, vomiting, abdominal cramping and diarrhoea) [6][7][8]. Patients with early symptoms (typically between thirty minutes and six hours) often have a favorable outcome, while delayed symptoms (i.e., after six hours) are associated with a higher risk of severe complications [3,8,9]. Although mushroom poisoning make up only a small fraction of the total number of emergency consultations for poisoning [10], this poses an important seasonal and regional problem for public health, with reported mortality rates of 8-12% [11,12] and higher rates (up to 15-30%) with mushrooms containing amatoxin [1,2].
In Switzerland, the National Poisons Information Centre Tox Info Suisse provides 24-h medical advice nationwide in cases of (suspected) poisoning. In recent decades, there has been a steady growth in mushroom-related calls, with an autumnal peak seen each year [13]. In 2016, 446 (1.4%) of the 32063 documented consultations for poisoning were related to mushroom poisoning (56% in adults, 43% in children and 1% of unknown age) [10]. The most commonly reported family was Boletaceae (6.3%), while 34.9% of the mushrooms were self-harvested and not further identified [14]. In addition to the medical advice offered by Tox Info Suisse, the Swiss Mushroom Control Association (Schweizerische Vereinigung Amtlicher Pilzkontrollorgane, VAPKO) offers checkpoints for a voluntary assessment of self-harvested mushrooms, in order to check edibility. In 2014, 4504 harvests were checked by the French-speaking part of VAPKO; 37.9% of those contained inedible, 12.5% poisonous and 1.7% lethal mushroom species [15]. Despite possible regional differences, only a study from the French speaking part of Switzerland has investigated the epidemiology and clinical characteristics of patients presenting with mushroom poisoning at an emergency department (ED) in Switzerland [3], and there are no data available from the German speaking part of the country. The aim of this study was to describe the characteristics and management of presentations related to mushroom poisoning at an emergency department in Bern, Switzerland, during a 17 year period.

Materials and Methods
This was a retrospective single centre study of patients presenting at the ED of Bern University Hospital (the Inselspital) with signs/symptoms related to mushroom exposure or intoxication during a 17 year period (from 1 January 2001 to 31 October 2017). The ED of the Inselspital is both a primary care facility (walk-in patients) and a tertiary referral centre for other hospitals in the region and provides a catchment area for approximately two million people-with about 46,000 (2017) emergencies admissions a year (≥16 years of age).
Patients admitted to the ED are registered in a specific clinical program: The program Qualicare (Qualidoc, Trimbach, Switzerland) was used for the period from 1 January 2001 to 31 May 2012 and a second program-E.care (E.care BVBA, Turnhout, Belgium)-from 1 June 2012 to 31 October 2017. These databases were used to identify all patients presenting with (suspected) mushroom poisoning. Cases were retrieved using a comprehensive full-text search algorithm with mushroom poisoning, related terms and a large number of specific mushroom names, including their German translation, as search terms. Each identified case was reviewed by one of the authors of the study. We included each patient requiring medical evaluation at the ED for mushroom poisoning, regardless of the circumstances of exposure (i.e., accidental or intentional) and the severity of the complaints. We excluded patients with no history of mushroom consumption and cases where the complaints could be clearly attributed to a cause other than mushroom poisoning based on the available final diagnosis in the medical report and the documented evaluation of the physician assessing the patient.
The following parameters were extracted (if available) from the charts of included patients: age, sex, type of transport to the ED (e.g., by ambulance), date of admission and discharge, mushroom species and how they were obtained (e.g., self-harvest, purchased from commercial sources), circumstances of exposure (e.g., accidental, recreational, suicide), symptoms, interval from ingestion to onset of symptoms, laboratory findings (including detection of α-amanitin in urine), consultation of a mycologist or Tox Info Suisse, treatment provided, type of discharge from the ED (e.g., inpatient admission, outpatient therapy), length of hospital stay, and outcome (e.g., complications, death). As in previous studies [3,9], we differentiated between early onset of symptoms (i.e., within six hours of ingestion) and delayed onset (i.e., >six hours). Furthermore, we distinguished between number of "clusters" and number of "patients/cases" [16], i.e., patients admitted together at the ED because they consumed the mushrooms jointly were regarded as one cluster, but each patient individually as one patient/case. Data were analyzed descriptively; numerical data are presented as arithmetic mean and standard deviation (SD) and nominal data as proportion (%). Detection of α-amanitin in urine was performed at Zurich University Hospital using enzyme-linked immunoassay (ELISA) until August 2016 and afterwards liquid chromatography-tandem mass spectrometry (LC-MS/MS) [17]. Data were analysed descriptively using Microsoft Excel (Microsoft, Redmond, WA, USA). The study was approved by the local ethics committee (No. 2017-02243).

Results
During the study period (1 January 2001 to 31 October 2017), 62 patients were admitted to the ED of Inselspital due to (suspected) mushroom poisoning. Eleven patients were excluded because mushroom consumption could not be confirmed, or the complaints could clearly be attributed to a cause other than mushroom poisoning. Finally, 51 patients were included in our study. The mean (±SD) age of the study population was 48.4 ± 16.9 years (median 48 years), with a range from 15-79 years. Twenty-nine patients (56.9%) were female and 22 (43.1%) male. The annual number of presentations over the study period ranged from zero to ten patients, with a peak in 2008 ( Figure 1). The seasonal variation over the period of 17 years showed a peak in autumn (September, October; 27 patients, 52.9%) and in summer (July, August; 13 patients, 25.5%) ( Figure 2 ingestion) and delayed onset (i.e., >six hours). Furthermore, we distinguished between number of "clusters" and number of "patients/cases" [16], i.e., patients admitted together at the ED because they consumed the mushrooms jointly were regarded as one cluster, but each patient individually as one patient/case. Data were analyzed descriptively; numerical data are presented as arithmetic mean and standard deviation (SD) and nominal data as proportion (%). Detection of α-amanitin in urine was performed at Zurich University Hospital using enzyme-linked immunoassay (ELISA) until August 2016 and afterwards liquid chromatography-tandem mass spectrometry (LC-MS/MS) [17]. Data were analysed descriptively using Microsoft Excel (Microsoft, Redmond, WA, USA). The study was approved by the local ethics committee (No. 2017-02243).

Results
During the study period (1 January 2001 to 31 October 2017), 62 patients were admitted to the ED of Inselspital due to (suspected) mushroom poisoning. Eleven patients were excluded because mushroom consumption could not be confirmed, or the complaints could clearly be attributed to a cause other than mushroom poisoning. Finally, 51 patients were included in our study. The mean (±SD) age of the study population was 48.4 ± 16.9 years (median 48 years), with a range from 15-79 years. Twenty-nine patients (56.9%) were female and 22 (43.1%) male. The annual number of presentations over the study period ranged from zero to ten patients, with a peak in 2008 ( Figure 1). The seasonal variation over the period of 17 years showed a peak in autumn (September, October; 27 patients, 52.9%) and in summer (July, August; 13 patients, 25.5%) ( Figure 2).    ingestion) and delayed onset (i.e., >six hours). Furthermore, we distinguished between number of "clusters" and number of "patients/cases" [16], i.e., patients admitted together at the ED because they consumed the mushrooms jointly were regarded as one cluster, but each patient individually as one patient/case. Data were analyzed descriptively; numerical data are presented as arithmetic mean and standard deviation (SD) and nominal data as proportion (%). Detection of α-amanitin in urine was performed at Zurich University Hospital using enzyme-linked immunoassay (ELISA) until August 2016 and afterwards liquid chromatography-tandem mass spectrometry (LC-MS/MS) [17]. Data were analysed descriptively using Microsoft Excel (Microsoft, Redmond, WA, USA). The study was approved by the local ethics committee (No. 2017-02243).

Cluster Size (Number of Persons
Involved)

h
Only one patient of this cluster consulted our ED

4-12 h
Only one patient of this cluster consulted our ED Although the total number of cluster patients was 37, data were not available for ten of those patients who did not consult our ED. Therefore, 27 of the cluster patients are described in our study (Table 1; ID 2 -7,12-14,16,21-28,35,37,40,41,45-48,50).

Discussion
During the study period of 17 years, 51 cases related to mushroom poisoning presented in the ED, with a seasonal peak in autumn and 27 cases as part of a cluster (14 clusters, with only one or some of the persons involved consulting our ED in some cases). In three quarters of the cases, the mushrooms were self-harvested. A mycologist and/or Tox Info Suisse were involved in the majority of the cases, and the exact mushroom type or family could be identified in almost half of the cases. In the majority of the cases, the symptoms were gastrointestinal disturbances of minor severity, and the patients were discharged within 24 h after symptomatic therapy was provided. Although there were no cases of death or liver failure, there were three cases of analytically confirmed Amanita phalloides poisoning, and, in one case, acute renal failure, thrombotic thrombocytopenic purpura with retinal haemorrhages and left side hemiplegia as complications. Other severe complications included a case of DIC after ingestion of Boletus satanas.
Boletaceae was the most commonly identified mushroom family in our study, as in the report from Tox Info Suisse [10] and the study from the French part of Switzerland [3], which might indicate that this mushroom is commonly consumed in Switzerland. As in previous studies [2,29], most presentations were related to edible species, which is probably associated with incorrect handling during collection, transport and storage of these mushrooms (e.g., collection in plastic bags, long period of transport at high temperature, long-term storage of mushroom dishes) [29]. Cases with poisonous and potentially life-threatening species were less common and included Amanita phalloides, Boletus satanas, Amanita muscaria and Amanita pantherina. Among the main mushroom toxins categories, gastrointestinal irritants were the most common in our study, followed by neurotoxins and protoplasmic poisons.
A check list of some important points regarding patient history and management in cases of (suspected) mushroom poisoning is contained in the supplement (Table A1, data from [30][31][32][33][34]).
Although not routinely available, analytical methods such as urinary ELISA can be used to detect amatoxin in cases of suspected amatoxin poisoning. This method has good sensitivity (up to 100%), especially if performed within 36 h after ingestion [35,36], although the amatoxin-level does not correlate with the disease severity [37]. In our study, urine was tested for α-amanitin in more than one third of the cases. Of these, the great majority were negative, which might be because of the ingestion of species lacking amatoxin or because the test was performed in the first six hours after ingestion, since detection in urine is only reliable after six hours post-ingestion [36,37], or too late after the meal. The characteristic sequence of the amatoxin-toxidrome with hepatic and multisystem organ failure typically appearing two to six days after mushroom consumption [1,8] was seen in one of the patients in our study with analytically confirmed Amanita phalloides poisoning (Table 1, ID 14), who developed nausea and vomiting after a latency of 12 h and liver and kidney function deterioration on the fourth day and after apparent clinical improvement. Coagulopathy, as also seen in this case in our study, is also reported as possible complication in this stage, which spans four to seven days [1]. The other two patients in our study with analytically confirmed Amanita phalloides poisoning developed none (ID 35) or only moderate (ID 19) transaminase elevation. Despite this, those patients were hospitalised for several days, most probably to assure regular follow-ups in order to exclude a delay onset hepatic and/or multisystem organ failure.
Treatment of Amanita phalloides poisoning includes administration of activated charcoal for gastrointestinal decontamination and silibinin [37], a milk thistle extract that has been shown to inhibit amatoxin uptake by the OATP1B3 transporters located in hepatocyte membranes [1,38,39] and to lower mortality in cases of amatoxin poisoning [40]. The combination of penicillin G and silibinin is not recommended, since it seems to worsen outcome compared to silibinin alone [41]. In case of hepatotoxicity, recommendations also include administration of N-acetylcysteine, similarly to paracetamol intoxication [37,40]. In our study, all three patients with Amanita phalloides ingestion received activated charcoal, silibinin and N-acetylcysteine. Despite the current recommendation, penicillin G was administered in combination with silibinin in four cases in our study, which might be because older guidelines were applied (all four cases with penicillin G administration were from 2001). Furthermore, since some patients with suspected poisoning received silibinin as precautionary measure while awaiting the analytical results, an in-house test could contribute to avoiding unnecessary administration of drugs and reducing length of hospital stay.
Similarly to other studies [3,4,29], in the majority of our cases the symptoms began with a latency of less than six hours, which is usually associated with a favourable outcome, since mushrooms that only cause acute toxicity are rarely life-threatening [3,8,9]. However, early symptoms do not exclude ingestion of potentially dangerous mushrooms, since patients often ingest more than one type of mushroom and even amatoxin-containing mushrooms can (rarely) cause symptoms before six hours [31,34]. This is also represented in our findings, as one of three patients with analytically confirmed amatoxin poisoning presented with a latency of less than 6 h ( Table 1, ID 19). In case of amatoxin-containing mushroom poisoning, earlier onset of gastroenteritis after consumption may correlate with more severe hepatotoxicity [42].
Identification of the specific mushroom(s) involved can contribute to appropriate diagnostic and treatment. In cases of mushroom poisoning, the exact mushroom species is successfully identified in the range of 5-27% of patients [3,31], with only one study [16] reporting higher rates of almost 90%. In our study, we were able to identify the exact mushroom type or family in almost half of the cases, a rate higher than most of the reported rates in the literature, most probably due to the close contact between the Inselspital and Tox Info Suisse/mycologists. The authors of the study in Italy [16] also report that the high identification rate was due to the available mycologist service in cases of suspected mushroom poisoning, while, in the study from the French speaking part of Switzerland [3], a mycologist was contacted in only one third of the cases. These findings highlight the importance of collaboration between EDs and the local poison control centres and/or mycologists, in order to enable mushroom identification and professional consultation, especially in cases of high risk.
Similarly to our findings, analysis of mushroom poisoning data from Tox Info Suisse also shows marked seasonal fluctuation [2] and most published cases of mushroom poisoning are recorded in the autumn (September and October) and late summer (July and August) [2,3,11,13,16,29]. This autumnal peak is most probably associated with the levels of rainfall, temperature and humidity during this period in Switzerland, which facilitates the growth of mushrooms and is also part of the wild mushroom season (May-October) [2].
The limitations of our study include the retrospective design. Although we aimed to identify all mushroom poisoning cases during the study period using the extensive search of our emergency unit specific software, it is possible that some cases were missed, e.g., if a patient was falsely diagnosed or did not mention the mushroom consumption, or that in some cases the symptoms were attributable to another cause that remained unrecognized despite thorough diagnostic at the ED. Moreover, although it is to assume that the patients remained free of symptoms if there was no re-presentation following their discharge, no information was available regarding their further outcome after leaving the ED. It is also possible that in some cases the treatment did not contribute to the clinical improvement (e.g., when administered as precautionary measure while awaiting the analytical results). Thus, it is not possible to provide an accurate evaluation of the undertaken therapeutic measures, Furthermore, it is possible that the diagnosis of mushroom poisoning was made after discharge from the ED or that patients from peripheral hospitals were directly admitted to a ward, thus resulting in cases that could not be found in our ED database, and it is also possible that some patients developed symptoms or complications after being discharged. Moreover, we cannot report in how many cases identification or exclusion of amatoxin-containing mushrooms was provided by the Tox Info Suisse/mycologist before the analytical results were available, since the exact time a mushroom specialist was contacted is not known. As a result, it is also not possible to describe how the patient management was adjusted (e.g., stopped or intensified) based on the mushroom specialist consultation in each case. Finally, our data may not be generalised to other regions in Switzerland, inter alia due to differences in local weather. The main strengths of our study include its long duration of 17 years. In addition, we frequently consulted a mycologist and Tox Info Suisse and this contributed to our high rate of identification of the mushrooms.

Conclusions
Based on our findings, ED consultations due to mushroom poisoning do not seem to be very common and mostly include mild symptoms associated with gastrointestinal irritants, usually with favorable outcome. However, although less common, cases of poisonings with potential lethal mushrooms and severe complications were also recorded. Collaboration with the local poison centre and a mycologist is of great importance for the identification of the mushroom(s) involved and decisions on patient management, especially in cases of high risk. The data of our study can contribute to the development of preventive strategies, such as medical personnel training and information material for the public (e.g., using websites or other media) on the correct handling of and potentially life-threatening mushrooms, especially during the high risk season for mushroom poisoning.

Physical examination
Classification of poisoning based upon clinical presentation (e.g., gastroenteritis, liver failure, seizures, cholinergic poisoning, hallucinations, renal failure) helpful in guiding further treatment

Asservation of mushrooms for analysis
Whenever possible, samples of all ingested mushrooms should be obtained for potential identification by a trained mycologist. Asservation of remainders of dishes, fresh mushrooms, peeling and waste.
Whole mushrooms are preferred, but identification can be made on parts of the mushroom, especially the cap. Further storage is facilitated by wrapping the mushrooms in wax paper, placing it in a paper bag, and refrigerating the sample. Storage in plastic bags should be avoided.

Contact with local poison control centre/mycologist
Consultation with a medical toxicologist/ mycologist is advised to determine likely species ingested based on clinical findings, identification of mushrooms, and specific treatment.