| |||||||||||||||||||||||||||||||
The implicated mushrooms were harvested from beneath pine trees on the sandy coast of southwestern France, between late fall and midwinter. T. equestre is widely disseminated throughout the world and is also known as T. flavovirens and colloquially as "bidaou" or "canari" in France, "riddarmusseron" in Sweden, "shimokoshi" in Japan, and "man on horseback" or "yellow-knight fungus" in the United States (Figure 1).3,4 We investigated the rhabdomyolysis apparently induced in 12 humans by several consecutive meals of T. equestre by administering equivalent doses of extracts of this mushroom to mice.
|
Seven women (age range, 22 to 60 years) and five men (age range, 24 to 61 years) were hospitalized between 1992 and 2000 with severe rhabdomyolysis approximately one week after eating wild mushrooms. All 12 patients had eaten at least three consecutive meals that included T. equestre, and none had a history of trauma, other known underlying cause, or medication use that could explain the occurrence of rhabdomyolysis. All patients reported fatigue and muscle weakness accompanied by myalgia, mainly in the upper part of the legs, 24 to 72 hours after their last meal containing mushrooms. The weakness worsened over a period of three to four days, leading to stiffness of the legs and the production of dark urine. These signs were accompanied by facial erythema, mild nausea without vomiting, and profuse sweating in eight of the patients. No fever was noted, and five patients had hyperpnea. The findings on physical examination, which included pulmonary auscultation and a neurologic examination, were unremarkable.
Initial screening tests showed evidence of rhabdomyolysis, with a mean maximal serum creatine kinase activity of 226,067 U per liter in the women and 34,786 U per liter in the men (Figure 2). No hepatic injury was evident. 
-Glutamyltransferase values were normal (range, 5 to 24 U per liter), and maximal average levels of aspartate aminotransferase and alanine aminotransferase were 8104 and 1392 U per liter, respectively, in the women and 1173 and 325 U per liter, respectively, in the men. Despite the intensity of the clinical rhabdomyolysis, electrolyte levels, including potassium values, were normal, and no renal failure occurred. Coagulation tests were normal. Additional studies were negative for parasites or other microorganisms (coxsackievirus, toxoplasma, toxocara, trichinella, hepatitis B and C viruses, and human immunodeficiency virus), as well as for systemic diseases (as assessed by complement-fixation tests and tests for circulating nuclear antibodies).
|
In six patients, samples of quadriceps muscle were obtained for histologic analysis. On light microscopy, the fascicular architecture of the muscles was well preserved, the myofibrils had a "nibbled" appearance, and in some cases the fibers were separated from each other by edema in the absence of vacuoles or the accumulation of glycogen or lipid — signs of a direct muscle injury. In the three patients who died, samples of the psoas and other muscles (from the arms, myocardium, and diaphragm) also showed evidence of acute myopathy.
Over the next 15 days, in all but three of the patients, the serum enzyme values gradually normalized and most symptoms disappeared, although the muscular weakness persisted for several weeks. In the three patients who died, the increasing dyspnea at rest was the first symptom of further deterioration and was followed by increasing rales at both lung bases, leading to admission to the intensive care unit. All three patients had hyperthermia (a temperature of up to 42°C); signs of acute myocarditis, including cardiac arrhythmia, cardiovascular collapse, and widening of the QRS complex without severe acidosis (pH, 7.37, with a serum bicarbonate level of 16 to 20 mmol per liter); and evidence of renal dysfunction, including elevated levels of blood urea nitrogen (30 to 52 mg per deciliter [10.7 to 18.7 mmol per liter]) and serum creatinine (1.4 to 2.5 mg per deciliter [126 to 224 µmol per liter]), with hyperkalemia (potassium, 6.0 to 7.2 mmol per liter) and hypocalcemia (calcium, 5.6 to 8.3 mg per deciliter [1.4 to 2.07 mmol per liter]) and a normal total protein level. The three patients had creatine kinase values of 632,000, 138,900, and 295,700 U per liter, with isoenzyme MB making up 0.5 to 0.7 percent of the total. Despite intensive physiological care, including, in one case, continuous venovenous hemofiltration, all three patients died. Autopsy revealed myocardial lesions identical to the muscular lesions in one patient, renal lesions in one patient, and no hepatic lesions.
Methods
Additional studies were necessary to demonstrate that T. equestre was the cause of the rhabdomyolysis in the 12 patients. Since it would have been unethical to administer T. equestre extracts to human subjects, we chose an established model of myonecrosis in mice.5,6,7 Extracts of T. equestre were prepared and then administered by gavage (gastric intubation) in a dose equivalent to that ingested by the patients. The animals were then assessed for evidence of rhabdomyolysis. Two protocols were employed: a dose-response study with T. equestre alone and one in which mice received extracts of either T. equestre or the nontoxic mushroom Pleurotus ostreatus.
Mushroom Extracts
Specimens of T. equestre collected in southwestern France were identified by qualified mycologists.2,3,4 The taxonomic denomination of T. equestre is synonymous with that of T. flavovirens (Fries) Lundell.3 Specimens of P. ostreatus obtained commercially were confirmed as such by qualified mycologists. Then 500 g of T. equestre was frozen, ground, and mixed with 200 ml of ultrapure water. The resulting mixture was lyophilized to yield 60 g of powder. Extracts were then obtained. A concentrate of 1.35 g of cold aqueous extract was obtained from 5 g of powdered T. equestre, and a concentrate of 1.65 g of boiled aqueous extract was obtained from 5 g of powdered T. equestre. A concentrate of 1.3 g was obtained from 10 g of powdered T. equestre after chloroform–methanol (vol/vol) extraction, and a concentrate of 50 mg of chloroform–methanol lipid-free extract was then obtained from 200 mg of the chloroform–methanol extract. We obtained a cold aqueous extract, a boiled aqueous extract, and a chloroform–methanol (vol/vol) extract of P. ostreatus in a similar manner.
Experimental Intoxication
Adult male Swiss mice (mean [±SE] weight, 30±5 g) (Depre), were randomly divided into several groups. In each group, T. equestre powder, T. equestre or P. ostreatus extract, or solvent alone was administered to each animal in a standardized fashion.
Twenty-four or 48 hours after the final dose, the mice were anesthetized with ether, and blood was collected from the retroorbital sinus and centrifuged at 2000xg for 15 minutes at 5°C to recover serum. The serum was frozen and stored at –20°C until analysis. Creatinine levels were determined by the Jaffé reaction (Merck kit 3385). Aspartate aminotransferase and alanine aminotransferase activities were determined with the use of enzyme kits (Merckotest, Merck).8 Creatine kinase was measured with a commercial kit (Enzyline kit, Biomérieux). The mice were then killed, and samples of striated muscle, liver, and tissues were obtained.
In the first protocol, three groups of three mice each were given 1 ml of powdered T. equestre suspension in water once a day by gavage for three days. The highest total dose was 6 g per kilogram of the body weight of the mouse (0.18 g per mouse), corresponding to a hypothetical toxic dose in a 60-kg person of 3 kg of fresh mushrooms eaten over the course of six meals in a period of three days (72 hours). Creatine kinase activity was determined in serum collected 48 hours after the final dose.
In the second protocol, five groups of five mice each were given 0.3 ml of tricholoma or pleurotus extract dissolved in water or dimethylsulfoxide once a day by gavage for three days. This dose corresponded to a total dose of 0.18 g of powdered T. equestre suspension. The positive control consisted of p-phenylenediamine (dose, 70 mg per kilogram per day for three days), which is a potent myotoxic compound in mice.6 Levels of aspartate aminotransferase, alanine aminotransferase, creatine kinase, and creatinine were determined in serum collected 96 hours after the final dose.
Statistical Analysis
Data are presented as means ±SE. The results were analyzed with use of the Wilcoxon rank-sum test, a nonparametric statistical test chosen because of the small number of animals.
Results
Mice treated with T. equestre powder (Table 1) had a concentration-dependent increase in the serum creatine kinase level (up to a mean of 380±25 U per liter). At a total dose of 4 g and 6 g per kilogram, the increase was significant when compared with the levels in controls (mean, 145±40 U per liter; P=0.01 for both comparisons), although there was variability between mice.
|
|
Two mice — one that received boiled aqueous extract and one that received chloroform–methanol lipid-free extract — died 72 hours after the last dose. These animals were not autopsied because by the time the deaths were discovered autolysis had set in.
Discussion
Rhabdomyolysis is a rare but potentially fatal condition. Muscle compression is the most common cause, but neither muscle ischemia nor unconsciousness was noted before the onset of symptoms in our patients. Deliberate intoxication with substances including cocaine, amphetamines, alcohol, theophylline, phenothiazines, p-phenylenediamine,9,10 antihistamines, and antihyperlipidemic drugs was ruled out.11,12 Medications that could cause dermatomyositis or polymyositis (penicillamine, phenytoin, levodopa, and quinidine) were ruled out by screening tests. Finally, immunologic screening as well as muscle biopsies did not indicate the presence of other systemic disorders or McArdle's disease.13
Toxic rhabdomyolysis has been described after the ingestion of small wild birds that had eaten water hemlock (Conium maculatum), a direct muscle toxin.14,15 Our patients had not eaten such birds. Although mushroom poisoning is not known to produce rhabdomyolysis, this series of cases clearly associates rhabdomyolysis with the ingestion of T. equestre.
Since 75 percent of the patients with large increases in creatine kinase survived, a genetic muscular susceptibility may be unmasked by the direct muscle toxicant contained in T. equestre when the amount of mushrooms ingested exceeds a certain threshold. Therefore, physicians should be aware of the possibility of severe rhabdomyolysis after repeated consumption of T. equestre. At this time treatment is supportive, and hospitalization is recommended for patients with dyspnea, signs of acute myocarditis, or even mild renal failure. Our experiments in animals confirmed the involvement of T. equestre in the etiology of rhabdomyolysis. Mice receiving T. equestre extract had increased creatine kinase levels, whereas those receiving P. ostreatus extract did not.
Several metabolites have been isolated from various tricholoma species — triterpenoids,16 sterols,17 indoles,18 and acetylenic compounds19 — but their muscle toxicity is unknown. The yellow pigment of T. equestre, 7,7' bi-physcion, has been identified.20,21 However, since this pigment is minimally soluble in water, we think it is unlikely to be the toxic compound. Since all extracts of T. equestre were toxic to mice, inducing increases in creatine kinase levels, the toxic compound appears to be extracted equally well by water and chloroform–methanol. It remains to be identified.
Supported by grants from the Conseil Général d'Aquitaine and the University Bordeaux 2 Victor Segalen.
We are indebted to Dr. John Fowler (United Kingdom) for his assistance.
Source Information
From the Poison Center (R.B.) and the Department of Infectious Diseases (J.M.R., M.D., D.N.), University Hospital Pellegrin, Bordeaux; the Departments of Toxicology (I.B., E.E.C.) and Mycology (G.D.), Faculty of Pharmacy, University Bordeaux 2, Bordeaux; Histotox, La Rochelle (J.P.P.); the Departments of Intensive Care (C.G.) and Internal Medicine (S.D.), University Hospital St. André, Bordeaux; the Department of Gastroenterology, Hospital of Montbrisson, Montbrisson (J.C.C.); and the Department of Internal Medicine, University Hospital Pitié–Salpetrière, Paris (P.G.) — all in France.
Jacques Beylot, M.D., Department of Internal Medicine, University Hospital St. André, Bordeaux, France, was also an author.
Address reprint requests to Dr. Creppy at the Department of Toxicology, University Bordeaux 2, 146 rue Leo-Saignat, 33076 Bordeaux, France, or at edmond.creppy{at}tox.u-bordeaux2.fr.
References
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
