The Emergence, Epidemiology, and Etiology of Haff Disease

Haff disease is a type of human rhabdomyolysis characterized by the sudden onset of unexplained muscular rigidity and an elevated serum creatine kinase (CK) level within 24 h after consuming cooked freshwater or seawater products^[1]. Haff disease was first identified in the summer and autumn of 1924 near the Königsberger Haff shores along the Baltic coast^[2-4]. The clinical manifestations of this disease mainly involve the sudden onset of rhabdomyolysis with severe muscle pain. In most patients, this is accompanied with myoglobinuria, significant increases in the serum levels of CK and myoglobin (Mb), and potential increases in the levels of other muscle enzymes [e.g., lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase (ALT)]. Epidemiological studies revealed that patients with Haff disease had consumed cooked fish, including burbot (Lota lota) and pike (Esox spp.), or other aquatic products within 24 h prior to the onset of illness.

Similar outbreaks of Haff disease have occurred in many countries since its initial identification. The identified patients had consumed various types of cooked fish or other aquatic products, including burbot (L. lota), buffalo fish (Ictiobus cyprinellus), crayfish (Procambarus clarkii), freshwater pomfret (Colossoma brachypomus), Cambaroides spp., Atlantic salmon (Salmo salar), and boxfish (Ostraciontidae spp.). Increases in human standards of living have led to the increasing availability of fish for consumption and, consequently, increasing reports of Haff disease. However, the specific cause of Haff disease remains unclear. Chemical analyses of suspicious fish samples, blood samples, and urine samples from cases for possible toxins, drugs, and hazardous elements produced negative results or toxin levels below the toxicity thresholds. Currently, the following criteria are used to define cases of Haff disease: (1) a history of consuming cooked aquatic products within 24 h prior to the onset of illness, (2) a markedly elevated (i.e., fivefold or greater increase over the normal levels) serum creatine phosphokinase level, and (3) a CK muscle/brain (CK-MB) fraction of < 5%^[5].

Several studies regarding the clinical manifestations, epidemiological characteristics, and pathogenic factors of Haff disease have been conducted since it was first reported in 1924. These studies have laid a solid foundation to confirm the pathogenic and influencing factors of Haff disease, to determine the means of preventing and controlling the occurrence of this disease, and to support further studies. This paper reviews the previous studies of Haff disease that have been conducted worldwide while suggesting future research directions according to trends in the development of related disciplines.

The Epidemiology of Haff Disease　　In the 9-year period after the first report of Haff disease in 1924, more than 1,000 cases with similar symptoms were reported along the Haff shores during the summer and autumn seasons^[1]. In the United States, most patients with Haff disease had consumed buffalo fish (I. cyprinellus) before the onset of illness^[6-8]. In 2008, an outbreak of 27 cases of Haff disease was reported in the Amazon region of Brazil, where the affected patients had consumed silver carp (Mylossoma spp.), black-finned colossoma (Colossoma macropomum), or freshwater pompano (Piaractus brachypomus)^[9]. From June 2016 to April 2017, 67 cases of Haff disease associated with the consumption of Olho-de-boi (Seriola spp.) or badejo (Mycteroperca spp.) were reported in Salvador, Brazil^[10]. In Japan, 13 sporadic cases of Haff disease were reported between 1990 and 2008, and all patients had a history of consuming boxfish (Ostraciontidae)^[11-12]. These data demonstrate the particular epidemiological characteristics of Haff disease, including a dietary history of cooked fish or other aquatic products within 24 h of the onset of illness and a location near a coastline, lake, or river.

In China, the number of cases of Haff disease has increased gradually since the initial report of 6 confirmed cases associated with the consumption of crayfish in Beijing during August 2000^[13]. In 2009, a reported outbreak of 54 cases in Guangdong was attributed to the consumption of freshwater pomfret (C. brachypomus)^[14]. However, most Chinese cases involved a dietary history of crayfish. Regarding the timing and location of Haff disease onset in China, cases have been reported during July and August of each year, which correlates strongly with the crayfish consumption season^[15-18]. This pattern is also consistent with the observed concentration of cases in areas with crayfish production and high-volume consumption, such as Jiangsu, Anhui, and Hubei provinces^[15]. Ma Shaolei et al. performed a spatial autocorrelation analysis to determine the incidence and spatial structure of crayfish-related rhabdomyolysis syndrome in Nanjing during 2016. Notably, a global spatial autocorrelation of the high prevalent regions revealed that cases were concentrated mainly in urban areas adjacent to the Yangtze River^[17]. The literature reviewed for this report included 1,118 patients with age ranging from 4 to 97 years, of which 484 (43.29%) were male. This finding was consistent with another study on the communal consumption of crayfish, which found no statistical difference in the incidence of Haff disease between female and male subjects^[19].

Clinical Features　　As previously noted, Haff disease is characterized mainly by the sudden idiopathic onset of rhabdomyolysis, which refers to a variety of pathological changes caused by hereditary or acquired factors such as crush wounds, drug toxicants, striated muscle cell damage, cell membrane channels, and/or an abnormal energy supply. These conditions damage the cell membrane integrity and expose the contents of cells, which include enzymes such as Mb and CK, and toxic ionic and small molecule substances. Accordingly, rhabdomyolysis is often accompanied with life-threatening metabolic disorders and acute renal failure (ARF)^[20-23]. The observed association of Haff disease with a history of consuming freshwater or seawater foods within the previous 24 h excludes other potential etiologies of rhabdomyolysis syndrome such as excessive exercise, trauma, alcoholism, hereditary disease, drug use, infection, or metabolic abnormalities^[24,25].

Tables 1 and 2 summarize the existing reports of Haff disease in the literature, including the serum levels of CK and creatine kinase isoenzyme, which are important for the diagnosis of Haff disease. Although most reported patients were discharged after hospitalization for symptomatic treatment, the tables describe the incidence of complications or other symptoms and related deaths. Additional information about the affected subjects is also presented in both tables.

Haff disease is associated with varying degrees of muscle pain, which is concentrated mainly in the neck and shoulders or lower back^[7,40], and this pain is usually the initial presenting sign^[57]. Severe cases also experience muscle tingling that is exacerbated by activity^[19,28]. In some cases, reddish-brown- or brown-colored urine is observed, and symptoms may be combined with limb or general weakness, chest discomfort, headache and dizziness, and/or gastrointestinal symptoms (e.g., nausea, vomiting, abdominal pain, and diarrhea)^{[16,26,28,30-32,34,36,39-42,44,46,48,54,58]}. Table 3 presents the distribution of clinical manifestations in 641 cases reported in China that met the above-stated criteria for Haff disease.

Regarding laboratory tests, elevated serum levels of various kinases, including CK, CK isoenzyme, lactate dehydrogenase, and aspartate aminotransferase are observed in Haff disease^{[7,15,16,19,26,28,30,31,37-41,44-46,48,58-60]}. Of these, the increase in CK is the most evident, with peak values as high as 39,164 U/L^[47]. Additionally, a routine urinalysis may reveal blood or protein in the urine^{[7,19,30,32,34,38-40,44]}, with ARF in some cases^[19]. In some cases, routine blood testing revealed an increased white blood cell count, elevated neutrophil percentage, and hemoglobin and platelet levels within the normal ranges^{[7,39, 40]}. Moreover, slightly elevated liver function markers (ALT) and normal renal function were observed in some cases^[39,40]. Cardiologic evaluations may reveal a cardiac troponin T level of < 0.1 ng/mL and normal electrocardiography findings^{[34, 44]}. Regarding muscle function, electromyography may reveal myogenic damage^[31], while a pathological analysis of the painful muscle may reveal rhabdomyolysis^[33].

Despite the above findings, the prognosis of Haff disease is good. With supportive treatment, including rest, intravenous fluid hydration, and bicarbonate infusion, the symptoms resolve, and the levels of muscle enzymes return to normal. However, a few serious cases may develop renal and other organ failures. In 2015, one case of rhabdomyolysis syndrome in Yancheng, Jiangsu, was attributed to the consumption of crayfish. The patient subsequently developed septic shock, multiple organ dysfunction syndrome, and acute renal failure. Although he became critically ill, he was discharged after treatment in the intensive care unit (ICU)^[38]. Furthermore, a 31-year-old man in Nanjing was diagnosed with Haff disease after consuming cooked crayfish. After a 5-hour period of hospitalization, the patient developed an acute lung injury. He was treated with intravenous fluid hydration and bicarbonate infusion and was discharged after 9 days. It remains unknown whether the lung was the direct target of the toxin in that case^[56].

Haff disease is rarely associated with mortality. In 2013, one male patient in Shanghai developed muscle pain, fatigue, muscle rigidity, oliguria, brown-colored urine, and shortness of breath 12 h after consuming crayfish. A laboratory examination revealed a CK level of 358.6 U/L. Following a misdiagnosis of lumbar disc herniation, he was admitted to the hospital with multiple organ failure but died after receiving treatment in the ICU^[35]. Generally, an early and accurate diagnosis and provision of active therapy to prevent complications are key factors required to improve patient outcomes.

Causes of Haff Disease　　Three categories of factors are known to cause rhabdomyolysis^{[20, 61, 62]}: physical factors, nonphysical factors, and genetic factors. Physical factors include squeezing and trauma, exercise and excess muscle activity, status epilepticus, electric shock, high fever, and heatstroke. Nonphysical factors include drugs, such as statins, poisons, bacterial (e.g., Legionella) and viral infections (influenza A, B), and electrolyte imbalances (e.g., hypokalemia, hypophosphatemia, hyponatremia, and hypocalcemia). Genetic factors include autoimmune diseases (e.g., polymyositis and dermatomyositis), endocrinological and genetic metabolic diseases (e.g., diabetes), and factors that can cause rhabdomyolysis.

Studies of the risk factors of Haff disease have been conducted consistently since the discovery of the disease. Descriptive epidemiological studies have suggested an etiology associated closely with the consumption of cooked seawater or freshwater foods. Furthermore, analytical epidemiological studies (e.g., case-control and cohort studies) and chemical analyses of collected biological samples have been conducted to determine the risk factors for Haff disease in cooked aquatic products, and a toxicity test has been developed to establish an animal model of Haff disease. One unmatched case-control study conducted in Nanjing in August 2010 involved the administration of a telephone/face-to-face questionnaire survey on food consumption in 20 patients with rhabdomyolysis syndrome and in 25 patients who had shared food with the patients. Subsequently, the subjects’ dietary histories, disease statuses, background diseases, alcohol consumption during the suspected crayfish meals, physical training habits, and allergic histories were investigated, and an association was found between the consumption of > 10 pieces of crayfish and an increased risk of disease^[63]. In 2009, Huang Qiong et al. conducted a retrospective cohort study on an outbreak of Haff disease caused by freshwater pomfret in Guangdong Province. Of the 3,857 Jiubei residents, 159 people had consumed pomfrets, and 50 members of this subgroup became ill. A dose-response relationship between fish consumption and elevated serum levels of muscle enzymes (CK, CK-MB, aspartate aminotransferase) were observed. Residents who did not eat fish were unaffected^[14]. Although these studies demonstrate a close association between the consumption of cooked aquatic products and the occurrence of Haff disease, the specific etiological factors require further investigation.

Previous studies have screened food samples and biological samples from patients for drugs and toxicants known to cause rhabdomyolysis syndrome. In one study by Buchholz et al., recovered leftovers and uncooked buffalo fish from the same lot were tested for active sodium channel biotoxins [e.g., ciguatoxin (the toxin of ciguatera) or saxitoxin (the toxin associated with paralytic shellfish poisoning)], cyanobacterial toxins (e.g., the blue-green algal toxins microcystin or nodularin), organophosphates, and arsenic. However, the tests produced negative results or values below the toxicity thresholds^[1]. Huang Qiong et al. tested fish, water, and soil samples for microcystins, toxins (including scorpion toxin, chelating base, histamine, ractopamine, F^-, CL^-, NO^3-, NO^2-, and SO₄^2-), and 26 heavy metals (Be, Na, Mg, Al, K, Ca, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Mo, Ag, Cd, Sn, Sb, Ba, Tl, Pb, Th, and U) using chemical analysis techniques and methods such as liquid chromatography-mass spectrometry (MS), ion chromatography, gas chromatography-MS, inductively coupled plasma-MS, and atomic force spectroscopy. Similarly, the results were either negative or below the toxicity thresholds^[14]. In a cytotoxicity and hemolysis analysis of palytoxins, Chen Xiaofeng et al. revealed that methanol/water gills, intestines, and glands extracts from crayfish were not inhibited effectively by ouabain, indicating that the rhabdomyolysis-inducing toxin was not an allergen or palytoxin^[64]. According to Chen Yan et al., a chemical analysis of crayfish samples purchased from a market and of blood and urine samples from 2 cases did not reveal any possible toxins, drugs, and hazardous elements among at least 200 screened compounds with known relevance to rhabdomyolysis (e.g., macrolides, sulfonamides, polyether antibiotics, β-agonists, organophosphorus pesticides, aniline, phenylamine, m-nitroaniline and p-nitroaniline, arsenic, cadmium, and other heavy metals)^[63]. He Ying^[65] found that although cyanobacteria, equisetum, and heavy metals induced increases in the serum CK level in the Institute of Cancer Research (ICR) mice, the related changes in enzyme activities were insufficient to induce Haff disease and did not cause symptoms of rhabdomyolysis in the ICR mice. No substances currently known to cause rhabdomyolysis were detected in samples of aquatic products consumed by patients with Haff disease or in the urine and blood samples of these patients.

Accordingly, researchers conducted animal experiments and established a mouse model of rhabdomyolysis to determine the substances in cooked aquatic products that cause Haff disease. Buchholz et al. separated the extracts of buffalo fish samples into water-soluble, nonpolar lipid and polar lipid fractions, which were administered intraperitoneally and orally to laboratory mice. Mice-fed hexane-soluble products exhibited behavioral changes consistent with muscle impairment and red-brown-colored urine in the bladders^[1]. Huang Qiong et al. found that samples of freshwater pomfret from the same reservoir could induce muscle damage in mice^[14]. Moreover, some batches of crayfish induced rhabdomyolysis syndrome in some mice in a dose-dependent manner^[41].

Some Chinese scholars have speculated that rhabdomyolysis syndrome may be related to an allergic reaction. Accordingly, LIN Jiang Wei conducted a series of studies of crayfish poisoning but ultimately did not detect an association between sea anemone toxin and allergic reactions. Moreover, Huang Qiong et al.^[41] observed that crayfish liver and pancreas could induce a decrease in the serum total immunoglobulin G antibody level, suggesting a weak or nonexistent association between rhabdomyolysis and allergy.

Professor Adgren suggested that the manifestations of Haff disease were consistent with a disease known as Chastek paralysis, which was first observed on silver fox farms in the United States. Subsequently, Ender and Helgebostad demonstrated that Chastek paralysis could be induced by a diet deficient in vitamin B1. Interestingly, Green and Evans reported that certain types of fish, particularly carp, contain a substance that inactivates vitamin B1. Moreover, foxes with Chastek disease recovered rapidly when treated with high doses of vitamin B1^[2]. However, the potential association between Haff disease and vitamin B1 deficiency has not been studied further.

In summary, the etiology of Haff disease remains unknown. Further study is required to determine whether cooked aquatic products contain an unknown heat-stable, tissue-specific, and non-neurotoxic pathological factors or whether the consumption of cooked aquatic foods activates particular signaling pathways that induce rhabdomyolysis.

The literature demonstrates an evident global increase in the number of reports of Haff disease since its first report in 1924. A review of this literature demonstrated that the existing research was based mostly on case reports from hospitals. Accordingly, the current body of knowledge faces several shortcomings. First, since the main clinical manifestation of Haff disease is muscle pain, many potential cases with mild symptoms may not have been detected. Second, researchers have used different approaches to verify their hypotheses regarding the etiology of Haff disease, and accordingly, the study was not conducted systematically. Different researchers may not have considered the relevant case conditions and contributing factors because of limitations specific to their scenarios. Researchers may also have insufficient data on the health risks associated with disease-causing food cultures and ecological environment factors. For example, in a study of Haff disease in Beijing in 2000, the researcher lacked trace data about the Cambaroides spp. consumed by the subjects and did not collect food or biological samples for further studies of the causative factors. Third, few studies have explored rhabdomyolysis in experimental animals, such as mice, and such research remains in the initial exploration stage. To date, all experimental materials have been sampled from patients’ leftovers and from reservoirs where the aquatic foods were caught. Any previously drawn conclusions will require further validation because the epidemiological hypothesis and directions of animal experiments remain unclear. Therefore, future studies of the health hazards, epidemiological characteristics, and influencing factors related to Haff disease and new animal models are required. This research can be approached through the following aspects:

(1) The epidemiological characteristics and influencing factors, including the collection and classification of previous cases of Haff disease. Researchers could construct a Haff disease case monitoring system and perform retrospective investigations in high-incidence provinces. Epidemiological investigations should use standard questionnaires to collect case information, and biological, food, and environmental samples should be collected. Descriptive epidemiological studies to clarify the health hazards and epidemiological characteristics of the disease should be developed systematically, and case-control and cohort studies should be conducted to determine the influencing factors.

(2) Research on the health risk factors associated with the culture, transportation, sales, and ecological environments of disease-causing aquatic foods (e.g., crayfish, burbot, buffalo fish). In China, Haff disease is mainly caused by the consumption of crayfish, and ecological studies should investigate the reservoirs and wild fishing areas of these organisms. Risk factors associated with breeding, transportation, and sales practices should be explored for possible health risk factors.

(3) Establishment of an animal model of rhabdomyolysis syndrome and animal test study. An animal model of rhabdomyolysis syndrome should be established using experimental animals of different species and germ lines. Pilot experiments based on epidemiological studies should be conducted in these animal models to verify the hypotheses of epidemiological investigations and to screen positive samples.

In conclusion, Haff disease, an idiopathic form of rhabdomyolysis syndrome associated with the consumption of freshwater marine products, has important impacts on human health, socioeconomic development, and stability. Haff disease can easily induce a social panic, given the acute incidence and potential severity and its association with the consumption of cooked aquatic foods. An understanding of the influencing and pathological factors associated with Haff disease would contribute significantly to preventing the occurrence of disease and would positively influence healthy dietary choices and the formulation of relevant industrial standards.

Author Contributions　　Conceived and designed the manuscript, LU Xuan Cheng, WANG Rui, and PEI Pei; searched and read the papers, PEI Pei, LU Shuang Shuang, LI Xiao Yan, and LIU Zhe; wrote the manuscript, PEI Pei; critically revised the manuscript, LU Xuan Cheng and WANG Rui; and provided the final approval of the manuscript to be published, LU Kai, LU Xuan Cheng, and WANG Rui.

Competing Interests　　None of the authors have competing interests to declare.