|Home | About | Journals | Submit | Contact Us | Français|
Summary: Tapeworms (Cestoda) continue to be an important cause of morbidity in humans worldwide. Diphyllobothriosis, a human disease caused by tapeworms of the genus Diphyllobothrium, is the most important fish-borne zoonosis caused by a cestode parasite. Up to 20 million humans are estimated to be infected worldwide. Besides humans, definitive hosts of Diphyllobothrium include piscivorous birds and mammals, which represent a significant zoonotic reservoir. The second intermediate hosts include both freshwater and marine fish, especially anadromous species such as salmonids. The zoonosis occurs most commonly in countries where the consumption of raw or marinated fish is a frequent practice. Due to the increasing popularity of dishes utilizing uncooked fish, numerous cases of human infections have appeared recently, even in the most developed countries. As many as 14 valid species of Diphyllobothrium can cause human diphyllobothriosis, with D. latum and D. nihonkaiense being the most important pathogens. In this paper, all taxa from humans reported are reviewed, with brief information on their life history and their current distribution. Data on diagnostics, epidemiology, clinical relevance, and control of the disease are also summarized. The importance of reliable identification of human-infecting species with molecular tools (sequences of mitochondrial genes) as well as the necessity of epidemiological studies aimed at determining the sources of infections are pointed out.
Tapeworms of the genus Diphyllobothrium Cobbold, 1858 (Cestoda: Diphyllobothriidea), commonly called “broad tapeworms” or “fish tapeworms,” have been known as intestinal parasites of humans for a long time. Some of the milestones in the history of human diphyllobothriosis are summarized in Table Table11.
In the early 1970s, diphyllobothriosis was estimated to affect 9 million humans globally, with 5 million in Europe, 4 million in Asia, and the rest in America (164). More recent data indicate that 20 million people are infected worldwide (27, 105), but no recent estimation concerning the global prevalence of this parasitosis has been done.
Nevertheless, in the last years, some studies showed a decline of human diphyllobothriosis in several countries, particularly in North America (39, 40, 79), Asia (87, 174), and most of Europe (45; B. Wicht, R. Peduzzi, and J. Dupouy-Camet, unpublished data). The number of human cases in areas where the prevalence of diphyllobothriosis was highest, such as Finland and Alaska, has decreased considerably during the last decades (39, 40).
In contrast, diphyllobothriosis has shown a reemergence in some countries such as Russia (136), South Korea (87), Japan (66, 174), and South America (Brazil) (46, 47, 153). Several cases have also been recently reported from the regions where a disappearance of the disease had been expected, such as Alpine lakes in Switzerland, northern Italy, and eastern France (Haute-Savoie) (see “Europe” below).
Although the broad fish tapeworm has been recognized as a human parasite for a long time, many aspects of its biology and epidemiology, including the spectrum of hosts causing human infection, clinical relevance, and present distribution, still remain poorly known. The importance of the disease is emphasized by its current recrudescence in some regions of the most developed countries throughout the world. Therefore, an updated overview of the current state of knowledge is presented, with focus on epidemiology (infective sources) of the disease and its reliable diagnosis based on molecular methods. Another aim of the review is to map existing gaps in our understanding of different aspects of diphyllobothriosis to promote future research and attract the attention of public health authorities.
Eggs released into the stool are unembryonated and possess an operculum on the narrower end (Fig. (Fig.1).1). The first-stage larva (oncosphere) is covered with a ciliated outer envelope, thus forming a coracidium, which hatches in the water. The motile coracidium swims and attracts potential first intermediate hosts (158).
Approximately 40 species of the genera Acanthodiaptomus, Arctodiaptomus, Diaptomus, Eudiaptomus, Eurytemora, and Boeckella (Copepoda: Diaptomidae), Cyclops, and probably Mesocyclops (Copepoda: Cyclopidae) serve as the first intermediate hosts (48, 97, 157). The coracidium penetrates the intestinal wall of the copepod and develops into the procercoid, which lacks a differentiated anterior end with attachment organs (scolex) but possesses a posterior appendage (cercomer) that contains six embryonic hooks (28).
Second intermediate hosts include freshwater, anadromous, or marine fish. Through the ingestion of infected copepods, the procercoid enters their tissues and develops into the plerocercoid stage (28). The sites of development may differ according to the fish species, with the larvae being localized in almost any organ and frequently even free in the abdominal cavity. Plerocercoids usually lie unencapsulated in the host tissue (Fig. (Fig.2A),2A), but they may be enclosed in connective tissue cysts (40). From the epidemiological point of view, the presence of larvae in muscles, liver, and gonads is of particular importance, but plerocercoids from viscera may migrate to the muscles after the death of the host. In addition, larvae of Diphyllobothrium dendriticum, normally encapsulated in the viscera, were also found unencapsulated in the musculature (29, 61).
Major groups of fish that may serve as source of human infection are briefly listed below.
Most common intermediate hosts, especially of D. latum, are predatory fish such as perch (Perca fluviatilis), pike (Esox lucius) (Fig. 2A to C), and burbot (Lota lota) in Europe and pikeperch or walleye (Sander canadensis and S. vitreus) in North America (7). The recrudescence of human diphyllobothriosis in the Alpine region corresponds to heavy infection of perch with D. latum plerocercoids. For example, in Lake Geneva, 4 to 10% of perch fillets examined between 2003 and 2005 contained D. latum plerocercoids (116). The infection rate of perch in Lake Maggiore reached up to 14% in 2005 and 2006 (167).
Plerocercoids of Diphyllobothrium cestodes from salmonids have often been identified as being D. latum. However, this identification is questionable, especially in the case of whitefish (Coregonus spp.) (171). It is possible that many, if not most, records from salmoniform fish (salmon, trout, and whitefish, etc.) actually belonged to other species. Evidence inferred from molecular data is necessary to confirm previous identifications of plerocercoids from salmoniform fish as being D. latum. Pacific salmons such as cherry, pink, chum, and sockeye salmon (Onchorhychus masou, O. gorbuscha, O. keta, and O. nerka, respectively) harbor Diphyllobothrium nihonkaiense in the northern Pacific Ocean (16, 39, 40, 48, 56, 107, 172, 181).
Whitefish (Coregonidae) do not harbor plerocercoids of D. latum, but they are frequently infected with larvae of other Diphyllobothrium species, especially D. dendriticum (Fig. (Fig.2D)2D) and D. ditremum (Creplin, 1825) (7, 171).
There are very few reliable data on the occurrence of plerocercoids of Diphyllobothrium in brackish-water and marine fish (5, 152). In South Korea, five human cases of “D. latum” infection were attributed to the consumption of raw flesh of redlip mullet (Liza haematocheila) (33). Common snook (Centropomus undecimalis), a marine fish that can enter fresh waters (55), was eaten raw, together with Atlantic salmon (Salmo salar) in sushi and sashimi, by patients infected with human broad tapeworms during an outbreak in São Paulo, Brazil (146). Plerocercoids of Diphyllobothrium pacificum were found in a number of predatory fish (152), which are supposed to be paratenic rather than second intermediate hosts (M. Tantaleán, personal communication). Japanese anchovy or “shirasu” (Engraulis japonica) and Japanese sardine (Sardinops melanostictus) have been suspected to be the source of human infection with Diplogonoporus tapeworms (11, 80).
Plerocercoids develop rapidly into adults in the definitive hosts’ intestine, yielding their first eggs 2 to 6 weeks later (48, 158). Most Diphyllobothrium species are characterized by a relatively low specificity at the adult stage, which implies that humans may become infected with parasites normally maturing in carnivore mammals or even in fish-eating birds (25).
Diphyllobothrium tapeworms are among the largest parasites of humans and may grow up to 2 to 15 m in length as adults in the intestine; the maximum length (up to 25 m) was reported for tapeworms with as many as 4,000 segments (158). The growth rate may be as high as 22 cm/day, or almost 1 cm/h (83). These parasites may live up to 20 years or longer; a patient with an infection more than 25 years old was reported by Dogiel (41).
Species of Diphyllobothrium are characterized by a scolex with a paired slit-like attachment groove (bothrium) on the dorsal and ventral surfaces, dividing it into two lips or leaves (6) (Fig. 3A and B). A proliferative zone (neck) is usually present posterior to the scolex. The remaining body (strobila) is composed of a high number of segments (proglottids [singular, proglottid]), each containing one set (or, rarely, two sets) of genital organs of both sexes (36) (Fig. (Fig.3C3C and and4A4A).
Testes are numerous and oval to spherical. The bilobed ovary (germarium) lies in the posterior one-third of each segment. The vitellarium is formed by numerous follicles distributed throughout the segments (Fig. (Fig.3C).3C). The vagina and the cirrus sac containing the male copulatory organ, a muscular cirrus, open medially into a common genital atrium on the ventral surface, anterior and median to the uterine pore (82) (Fig. (Fig.4B).4B). The uterus is tubular and extends far anterior to the ovary; its outer coils form a rosette that leads to the uterine pore (36). Eggs are operculate (Fig. (Fig.1),1), are unembryonated when laid, and measure 35 to 80 μm in length and 25 to 65 μm in width, depending on the species (36). In fact, there is a large overlap in size among many taxa; host species and intensity of infection may also influence egg size (8).
Many species in the genus Diphyllobothrium have been described since Linnaeus proposed Taenia lata as the first representative of this group. The taxonomic composition of the genus has changed many times, as have opinions on the validities of individual species (36, 72, 149). Notwithstanding uncertain systematics of many taxa, it is unquestionable that besides D. latum, several other species are implicated in human infections in circumpolar regions and in the Pacific area. A total of 14 out of more than 50 Diphyllobothrium species, currently considered to be valid, have been reported from humans (14, 36, 72).
Phylogenetic relationships within the genus are not well known, because DNA sequences of only a few taxa, especially those infecting humans, are available (11, 150, 169). In most analyses, D. pacificum and/or D. stemmacephalum (type species of the genus) represent the most basal taxa of the genus (12, 22, 81, 150). Diphyllobothrium nihonkaiense is the basal taxon to the clade formed by D. latum, D. dendriticum, and D. ditremum, which also contains members of the genera Ligula and Digramma, the adults of which are parasites of fish-eating birds (44). Such a topology of the cladogram implies that the genus Diphyllobothrium represents a para- or polyphyletic assemblage of taxa that do not form a monophyletic group (12, 22, 169).
Molecular data also indicate that Diplogonoporus balanopterae (Lönnberg, 1891) [synonym, Diplogonoporus grandis (Blanchard, 1894)] is a species of Diphyllobothrium with doubled genitalia in a segment (11, 12). This human-infecting parasite is therefore included in our review.
Basic information on the species of Diphyllobothrium and Diplogonoporus reported for humans is presented in Tables Tables2,2, ,3,3, and and4,4, including the most frequent synonyms to prevent confusions in reporting individual taxa (for an extensive list of synonyms, see references 36 and 72). The species are divided on the basis of their life cycles, i.e., freshwater (the life cycle is completely realized in fresh water), anadromous (salmonids entering from the sea to freshwater), and marine (all the cycle takes place in the sea) (65). For each one, most pertinent papers are mentioned to facilitate the search for more data.
Diphyllobothriosis is known to occur widely in the world. Since this disease is generally considered to be a mild illness and is not systematically reportable, our understanding of its global distribution is somewhat fragmentary, based on a restricted number of surveys or clinical case reports. Human infections with Diphyllobothrium tapeworms are generally associated with cold waters, because most cases were reported from the Palaearctic region and some parts of North America. However, clinical cases from South America, especially from its Pacific coast, are also known.
Human diphyllobothriosis is still present in western Europe, but compared with data from previous studies (69, 85, 158), it has shown a marked decrease in the historical Baltic areas of endemicity (Estonia, Latvia, and Lithuania) as well as in Poland, Romania, Sweden, and Norway (Wicht et al., unpublished). In Finland, where the number of human cases was very high (159), human infections with D. latum showed a decrease until the 1980s, and the rate is currently about 20 cases/year (128; Wicht et al., unpublished).
Reports of diphyllobothriosis have increased in sub-Alpine areas around the great Swiss, Italian, and French lakes, where raw or undercooked perch (Perca fluviatilis) is consumed. More than 200 cases were documented in a survey from 1987 to 2002 (45), and 330 cases were documented from a survey conducted between 2002 and 2007 (Wicht et al., unpublished). An outbreak of diphyllobothriosis in Geneva has recently been reported by Jackson et al. (67). In some countries previously considered to be disease free (Austria, Czech Republic, Belgium, The Netherlands, and Spain), sporadic cases have been reported over the last 6 years, which were presumably linked to the consumption of raw imported fish (Wicht et al., unpublished).
Diphyllobothrium latum has been considered to be the principal species infecting humans in Europe, whereas D. dendriticum is present in northern Europe (45). Molecular analyses of samples isolated from clinical cases in Finland, France, and Switzerland indicate the presence of D. nihonkaiense locally acquired from imported salmons (124, 169, 179; Wicht et al., unpublished).
The first case of D. latum in humans in North America was reported by Nickerson (115), in a child born in Minnesota. Subsequent investigations led to the conclusion that this species was introduced by immigrants from Scandinavian regions of endemicity (96, 163), but there is convincing evidence that infection with D. latum is primarily a worldwide zoonosis and that the tapeworm occurred in indigenous people and dogs in North America prior to its colonization (27, 39, 40). Until 1982, diphyllobothriosis was a reportable disease in the United States. The Centers for Disease Control and Prevention (CDC) estimated that about 125 to 200 cases occurred during the period of 1977 to 1981. Most cases occurred in the Great Lakes region, central Canada (Manitoba), and Alaska, although human infections elsewhere have been reported. However, there has been a drastic decline in reports of D. latum over the last 100 years (39, 40).
The following species of Diphyllobothrium were documented as adults from humans in North America: D. latum, D. dendriticum, D. dalliae, D. lanceolatum, D. ursi, D. alascense (133, 134), and, just recently, D. nihonkaiense (169).
The first case of human diphyllobothriosis was documented in Argentina in 1911, in a young Russian immigrant who had just arrived in the country (34). The second case was reported as late as in 1950 from Chile (114) and was attributed to the infection of introduced rainbow trouts (Oncorhynchus mykiss) due to water contamination with tapeworm eggs shed by tourists and immigrants from North America.
Diphyllobothrium latum and D. dendriticum were supposed to have been brought to South America via European immigrants (148). South American diphyllobothriosis caused by D. pacificum is an ancient disease, because eggs of this species (morphologically diagnosed) were found in coprolites from 2000 to 3000 BC (138). Diphyllobothrium pacificum was first reported from humans in Peru (17), and multiple human cases were then documented in Peru and Chile.
Most human infections caused by D. pacificum and, allegedly, D. latum have been reported from Chile and Peru, with other cases in Argentina, Brazil, and Ecuador (17, 18, 39, 40, 54, 145-147), including recent outbreaks of diphyllobothriosis in Rio de Janeiro and São Paulo (46, 47, 153).
Diphyllobothriosis is frequently reported in Japan, especially along the coast of the Sea of Japan, averaging about 100 cases per year since the 1970s (120). In this country, as many as 11 species of human diphyllobothriid cestodes, including Diplogonoporus grandis, have been reported (39, 40, 72, 80). Recent taxonomic studies strongly suggest that the majority of human infections are due to D. nihonkaiense (39, 40, 172, 174). About 45 human cases of diphyllobothriosis and one case of diplogonoporiasis have also been reported from South Korea (32, 87, 88).
Infections by Diphyllobothrium are common in Russia, especially in the Far East, including the Amur River region, where D. nihonkaiense (reported as D. klebanovskii) is regarded as an important zoonotic species (76, 91, 106-108, 136). The parasite is widespread in all major river basins east of the Urals, including those of Lena, Kolyma, and Indigirka (151). The coastal areas of the Okhotsk Sea, where human prevalence ranges from 1.0 to 3.3% (91), are also endemic foci (12).
Rare clinical cases attributed to different species of Diphyllobothrium have been reported from the Middle East, Saudi Arabia, India, Indonesia, Malaysia, Mongolia, and Taiwan (1, 3, 31, 38, 51, 93, 99, 109, 121, 141, 142). However, some cases were probably imported from areas of endemicity such as Japan (3). In addition, species of Spirometra that cause sparganosis (105) may have been confused with Diphyllobothrium because the genus was synonymized by some authors (149).
There are no reliable recent reports on the occurrence of broad fish tapeworms in humans from Africa and Australia.
Diphyllobothriosis is associated with the ancestral habit of eating raw or poorly cooked fish. This includes the consumption of raw salted or marinated fillets in Baltic or Scandinavian countries, “carpaccio” (very thin slices of raw fish, also called “carpaccio di persico” in northern Italy and “carpaccio d'omble chevalier”); “tartare maison,” made with raw salmon; “poisson du lac façon nordique” in French-speaking areas of Europe; and “gefilte fish” (balls of finely chopped fish mixed with crumbs, eggs, and seasonings, cooked in a broth) in Jewish populations. In Latin America, lightly marinated fish known as “ceviche” (or “cebiche”) (pieces of raw fish freshly marinated in lemon and salt) is a common dish (17, 27). In Japan, “sushi” (bite-size pieces of cold cooked rice topped with fish, egg, or vegetables and wrapped in seaweed) and “sashimi” (slices of fresh raw fish) are traditional dishes served with soya sauce for dipping. Among the three major types of sushi, “nigiri,” a piece of sliced raw fish fillet on a vinegared rice ball, is the commonest dish (112). Specialties such as sweetish “ayu” and cherry trout, “sakura mad,” are still popular in Japan even though they can serve as sources of human infection (70). In developed countries, the popularization of restaurants serving uncooked fish is thought to be responsible for the increase of the number of human cases of fish-borne parasites.
Occupational risks are exemplified by the often high infection rates in fishermen who have the habit of eating roe and liver of their fresh catches and in women who taste foods under preparation that contain raw fish (39, 40). Salmons are probably the most common hosts that transmit diphyllobothriosis, although broad fish tapeworms may also be transmitted by whitefish, trout, pike, and other species (19, 39, 40) (Fig. (Fig.2A2A).
The recent increase of human cases in some regions accentuates the necessity to carry out epidemiological surveys in order to provide reliable data on the actual distribution of diphyllobothriosis within the population. As an example of such a campaign, epidemiological monitoring realized by one of the authors in Switzerland, a country where D. latum is considered to be the principal causative agent of diphyllobothriosis, is briefly described.
In the first part of the study, physicians, veterinarians, and medical laboratories were invited to collect samples of Diphyllobothrium diagnosed in their routine practice. After standard morphological identification, they were asked to preserve a part of fresh stool samples containing eggs, as well as tapeworm segments, in 96% ethanol for further investigations. When possible, a questionnaire reporting clinical data, type of fish eaten, and cooking habits was completed for each patient. In the successive phase, parasites were analyzed using molecular techniques. The partial nucleotide sequence of the cytochrome c oxidase subunit I (cox1 or COI) gene was obtained and compared with reference sequences available in GenBank. The survey allowed us to identify three cases caused by exotic species of Diphyllobothrium locally acquired from imported fish (166, 167). According to anamnestic data, the other patients were infected with D. latum acquired mainly from perch fished in local lakes.
At the same time, parasitological inspections of fish sold in restaurants, fisheries, and supermarkets in Ticino and Geneva put in evidence the presence of D. latum plerocercoids (identified with molecular methods) in 9.5% and 5% of fillets of perch originating from Switzerland, Poland, Estonia, and Russia (170).
The fecundity (reproductive potential) of Diphyllobothrium parasites is extremely high: one worm is estimated to produce up to 1 million eggs per day (158). This implies that the environment can be easily contaminated if basic hygienic or sanitary rules, such as using toilets and effective treatment of sewage waters, are not realized.
The problem of water contamination with tapeworms’ eggs is improved by the ability of most Diphyllobothrium species to mature in nonhuman hosts. Because of their generally broad host specificity, their life cycles are maintained in nature independently from humans (39, 40). Therefore, dehelminthization of the human population does not necessarily eliminate the parasite from concerned areas. Sylvatic cycles involving bears, foxes, seals, gulls, and other fish-eating birds and mammals probably play a crucial role in water contamination (2, 39, 139, 154). The close contact between dogs, cats, and humans may represent a risk of transmitting zoonotic agents, but some surveys revealed a low infection rate of these hosts (48). For example, coprological examinations of 505 and 296 dogs from Switzerland and Finland, respectively, revealed the presence of D. latum in only 0.4% of dogs examined (125, 143).
The high vagility of animals serving as a reservoir of adult tapeworms may result in the dissemination of parasites to new geographical areas similarly to the import of fish intermediate hosts such as Pacific salmons, rainbow trout, or whitefish (167). Sporadic human carriers may also cause a heavy infection load of plerocercoids in fish populations, even in lakes of considerable size (39). Fish are fundamental reservoirs of Diphyllobothrium because plerocercoids may survive in their body from several months up to a few years (39, 40).
Well-established, new endemic foci of the disease often originate from population transfers from infected areas due to emigrations and war, etc. In such circumstances, people often retain their dietary habits, which is indispensable for the epidemiological cycle of the disease (39, 158).
Since human diphyllobothriosis is in most cases mild or asymptomatic, and because patients can be easily treated with praziquantel, the general tendency of physicians and medical laboratories is to identify the parasite only to the genus level. From the clinical point of view, this attitude is, in principle, correct. Species identification in routine laboratories may actually be difficult because procedures involving molecular techniques require trained personnel. Nevertheless, some aspects concerning the importance of species identification deserve to be discussed.
The diagnosis of Diphyllobothrium species as well as the detection of their sources for human infection (through anamnestic data) are of great importance with respect to epidemiology. The identification of hosts and parasites at the species level would contribute to a better understanding of the present distribution of different taxa. In light of recent case reports, human infections with exotic (imported) species of Diphyllobothrium might be present in a number of countries, being actually underestimated. Imported parasites may also enhance the probability of the appearance of outbreaks even in countries with a high level of medical care (39, 67).
Diagnosis of human broad tapeworms is based largely on finding eggs of the typical ovoid shape with an operculum on a narrowed pole and a size of 35 to 80 by 25 to 65 μm or segments with medially situated genital pores (Fig. (Fig.1,1, ,3C,3C, and and4B).4B). Morphology-based diagnoses are cheap and relatively easy but in most cases do not enable identification at the species level. Some taxa can be differentiated from one another only on the basis of the shape and size of the scolex, which is usually absent in clinical samples. In addition, after treatment, most samples are not suitable for morphological evaluation and identification because of the damage of tapeworm tissues.
Most cases of diphyllobothriosis are correctly diagnosed at least at the genus level, but there have been misidentifications with flukes (or trematodes [Digenea]), which may also possess operculate eggs of a similar size, or with segments of taenias (Taenia saginata or T. solium) that may have a similar shape (13, 105). Many samples are identified automatically as being D. latum, but D. nihonkaiense and other taxa may be misidentified (39, 87, 169).
Some species can be differentiated on the basis of the shape and size of the embryonic hooks of the oncosphere (95, 175). However, embryonated eggs are not present in fresh stool of definitive hosts, and embryonation requires several days in the water (158).
Morphological identification of plerocercoids in fish is often difficult (98), but a simple key is available for the three main species that are parasitic in the Holarctic fish, namely, D. latum, D. dendriticum, and D. ditremum (7). Plerocercoids differ from each other in the body surface (wrinkled or smooth), the length of microtriches, the retraction of the scolex, and the number of subtegumental longitudinal muscles (7, 9) (Fig. 2B to D). Identification of Diphyllobothrium plerocercoids in marine fish is more problematic (5, 156, 178), and molecular-based diagnostics will be necessary to confirm previous records, including those of D. nihonkaiense from marine fish in Peru (152).
Procercoids in copepods cannot be identified at the species level because of their morphological similarity.
Molecular methods have been widely used in diagnoses because of their specificity and the possibility of evaluating a large number of samples in a short time. At present, they represent the most reliable tool to identify clinical samples of Diphyllobothrium at the species level. They can also be applied to determine different ontogenetic stages of parasites such as segments of adult tapeworms and their eggs as well as plerocercoids in fish to trace the origins of sources of human infections.
Restriction fragment length polymorphism with the endonucleases SmaI, HinfI, and HhaI used as species-specific markers enabled D. nihonkaiense and D. latum to be distinguished (102). The sequencing of nuclear and mitochondrial DNAs provided essential data for the identification and the phylogeny of Diphyllobothrium tapeworms. Phylogenetic trees based on sequences of ribosomal genes (18S rRNA and 28S rRNA) and ITS1-5.8S-ITS2 regions have been used to elucidate the relationships among some taxa (22, 92). However, these data are not useful for the routine discrimination of all Diphyllobothrium species (180). The characterization of the complete mitochondrial genomes of D. latum and D. nihonkaiense provided essential information as to the utility of coding and noncoding regions for the parasite's identification (78, 111, 122). In particular, the cox1 gene sequence appeared to be an appropriate target for species identification of human broad tapeworms mainly because of its higher mutation rate than that of nuclear genes (167, 169, 180).
Samples to be identified with molecular techniques (eggs, larval stages, and adult parasites) should be preserved in pure ethanol, whereas DNA extraction from native fecal samples should be performed immediately. Fixatives containing formalin, widely used for the storage of clinical samples, as well as denatured alcohol should be avoided because they damage DNA. It is possible, in some cases, to amplify short DNA regions from parasites already fixed in such solutions (20, 89).
When molecular analysis (PCR) is made directly from fecal samples, a preliminary concentration of eggs (formol-ether concentration), without the addition of formalin, is suggested to obtain sufficient amounts of DNA (140). Sonication (two to three times for 10 s each at medium intensity) allows the disruption of egg shells and release of their content. Larvae and adult specimens conserved in ethanol should be washed carefully with phosphate-buffered saline before genetic analysis.
DNA extraction can be made either with a commercial kit or with the classic phenol-chloroform method, which seems to yield larger amounts of DNA, especially from small-sized procercoid larvae (165). Eluting DNA in water, instead of in buffers, can be useful in cases of small amounts of DNA because it can be concentrated using a vacuum centrifuge (i.e., SpeedVac; Savant Instruments, Inc.).
The most appropriate targets for the identification of Diphyllobothrium at the species level are the cox1 (21, 166, 180), tRNAPro, tRNAIle, tRNALys, NADH dehydrogenase subunit 3 (NADH3) (179), and cytochrome b (cob) (Wicht et al., unpublished) genes.
Adult tapeworms lie folded in loops of the small intestine of their vertebrate host. Attachment to the intestinal wall usually takes place at the level of the ileum and less commonly in the jejunum or other levels. Rarely, the worms attach in a bile duct (101). Despite the large size of most Diphyllobothrium species and, thus, their mechanical effect on the host, many infections with this parasite are reported to be asymptomatic (70, 101). In about one out of five infections, diarrhea, abdominal pain, or discomfort occurs; other symptoms of diphyllobothriosis may include fatigue, constipation, or pernicious anemia (60, 160) and, sometimes, headache and allergic reactions. Less commonly, massive infections may result in intestinal obstruction, and migrating segments can cause cholecystitis or cholangitis. Other symptoms including pain in the tongue at eating have been associated to diphyllobothriasis.
Prolonged or heavy D. latum infection may cause megaloblastic anemia due to a parasite-mediated dissociation of the vitamin B12-intrinsic factor complex within the gut lumen, making B12 unavailable to the host (162). Approximately 80% of the B12 intake is absorbed by the worm, with a differential absorption rate of 100:1 in relation to the absorption by the host. About 40% of D. latum-infected individuals may show low B12 levels, but only 2% or less develop clinical anemia, which is hyperchromic and macrocytic and may be associated with low platelets or low white blood cell counts. Severity of the disease is known to be directly associated with worm burden and by-products produced by tapeworms (60). This deficiency may produce damage to the nervous system, including peripheral neuropathy or central nervous system degenerative lesions. Diphyllobothrium-associated pernicious anemia is rarely reported nowadays (43), and anemia is also rare or nonexistent in the small D. pacificum tapeworm. After successful treatment, B12 levels come back to normal ranges in a period of several months.
Although the symptoms are also generally mild, infection by D. nihonkaiense can have a substantial emotional impact on patients and their families, because segments are evacuated over a long period of time. An in-depth study of eight clinical cases has shown that severe clinical symptoms can lead to specialized consultations and expensive complementary analyses, resulting in an average cost of €400 for the management of a single diphyllobothriosis case (37).
Human Diphyllobothrium infection becomes patent (begins to pass eggs in stools) after approximately 15 to 45 days after ingestion of plerocercoid larvae. Typically, the presence of the tapeworm is noted because of the expulsion of chains of segments with the stools (Fig. (Fig.4A).4A). The total length of D. latum usually ranges between 3 and 12 m (3,000 to 4,000 segments), and adult parasites may survive for decades in the human host.
The aim of preventive and control measures must be to break the life cycle of the parasite. Theoretically, any point of the life cycle can be attacked. In practice, measures should be focused on the following three principal points: (i) prevention of water contamination (see above), (ii) treatment of people harboring the parasite, and (iii) prevention of transmission of infective larvae from fish to humans.
Sewage treatment plants and the use of sanitary facilities represent the most effective sanitary measures to avoid water contamination (161). Treatment of infected patients and prevention of food risk are discussed in detail in the next section.
Adult tapeworms are easily treated with praziquantel (Table (Table5).5). A single oral dose of 25 mg/kg is highly effective against human D. latum infections (26, 63). A lower dose of 10 mg/kg was effective against human infections with D. pacificum (63, 94) but showed a poor effect against D. latum in experimentally infected golden hamsters (26). Diphyllobothrium nihonkaiense seems to be more sensitive to praziquantel than D. latum and equally or more sensitive to praziquantel than D. pacificum.
Oral administration of a single dose of praziquantel at 5 to 10 mg/kg was reported to be effective and safe for D. nihonkaiense infections, but a single administration of a 25- to 50-mg/kg dose is usually applied (117-119). However, some studies revealed lower effectiveness of this medicament and a low proportion of tapeworms with a detached scolex (80).
Side effects of praziquantel are usually mild and do not require treatment, although they may be more frequent and serious in patients with a heavy worm burden. The following side effects, in order of severity, have been observed: malaise, headache, dizziness, abdominal discomfort with or without nausea, rise in temperature, and, rarely, urticaria (63). Such symptoms can, however, also occur with the infection itself.
Niclosamide (a single dose of 2 g in adults and 1 g in children older than 6 years) is an alternative antihelminthic drug effective for diphyllobothriosis. Side effects are infrequent because it is not absorbed from the gastrointestinal tract. The availability of niclosamide is, however, limited in many countries.
Intraduodenal gastrographin (used for contrast-enhanced intestinal radiographs) has been reported to be efficacious in the treatment of large cestodes including Taenia saginata, T. solium, D. nihonkaiense, D. yonagoense (D. stemmacephalum), and Diplogonoporus grandis (110). The first case of treatment by oral administration of gastrografin (181) resulted in finding the expelled strobila more than 6 m long with the scolex, which was morphologically identical to D. nihonkaiense. However, this method is not considered to be the best choice of therapy because the insertion of the duodenal tube is painful, the therapy is expensive, and fluoroscopic images are needed. On the other hand, the advantage of this method is the discharge of a complete, living worm with the scolex and thus one suitable for species identification (80).
The best prophylaxis is to avoid the consumption of raw, smoked, or pickled fish. Fish should be well cooked; alternatively, freezing for 24 to 48 h at −18°C also prevents the infection. The Fish and Fishery Products Hazards and Controls Guide recommends an internal temperature below −20°C for 7 days or −35°C for 15 h to kill the parasites (52). Based on currently available data, these recommendations may appear stringent because they were developed for parasites that are considered to be the most resistant to freezing (52).
On an individual basis, infections with Diphyllobothrium tapeworms can easily be prevented by eating well-cooked fish or deep-frozen fish (at least −10°C for 24 h) or by placing the fish in a concentration of brine (12% NaCl) (126). Cooking fish at a temperature of 55°C kills plerocercoid larvae in 5 min (40), whereas freezing at −10°C kills them within 8 to 72 h, depending on the thickness of the fish flesh (53). Eguchi (48) observed that plerocercoids survived for 4 h but died after 12 h at −8°C; all plerocercoids died after 6 h at −10°C. Smoking of fish does not kill the parasite (19).
The U.S. Food and Drug Administration (FDA) suggested that fish intended for raw or semiraw (such as marinated dishes) consumption should be blast frozen to −35°C or below for 15 h or be regularly frozen to −20°C or below for 7 days. Similarly, according to the European Union Hazard Analysis and Critical Control Points, marine fish for raw consumption should be frozen at −20°C for more than 24 h. Salting of fish also results in reduced infectivity, but it may take several days or weeks depending on the size of the fish and the volumes of salt used.
To avoid new outbreaks of human diphyllobothriosis, salmon and other fish should not be eaten raw, at least not until it has been frozen under the conditions discussed above. Salmons are now transported worldwide only on ice, and this is the way fish helminths are usually introduced to new areas and may infect humans anywhere (167, 171, 180). It is therefore necessary to inform consumers about the risks linked to some culinary habits. Inspection of fish sold to the public should also help detect infected species and orientate public health measures, particularly by identifying infected species and areas of endemicity.
Human diphyllobothriosis seems to be generally declining in many areas where it represented an actual medical problem several decades ago. However, this parasitic disease should not be regarded as neglected, because new foci and human infections with exotic species have recently appeared, even in countries with a high standard of medical care. The increasing popularity of eating uncooked or raw fish, uninspected import of Pacific salmon and other fish, as well as global climate changes represent factors that might lead to a rapid and massive recrudescence of diphyllobothriosis in the near future.
Certain projections indicate that the future worldwide demand for fish will increase substantially (50), which may result in more intensive exploitation of the marine environment for food (35). A higher demand for fish may also increase the risk of diphyllobothriosis by increasing the harvesting and the export of products from areas where this zoonotic disease is endemic. Higher risks for urban populations may also arise because of the incentive for exporters to ship fresh (nonfrozen) fish by air to gain a competitive edge in the market (35, 104, 113). Europe and Canada, both still areas where diphyllobothriosis is endemic, supply about one-third of the U.S. demand for seafood (35).
Seemingly unrelated environmental changes may also have unexpected effects on the epidemiology of this zoonosis. It has been suggested that the increase in human infections with D. pacificum in northern Chile during the period of 1975 to 2000 was related to the cyclic appearance of El Niño phenomena in the Eastern Pacific, which affected not only fish populations but also the primary definitive host of this tapeworm, the sea lion (145).
The species-specific identification of clinical samples is not essential for the effective treatment of most human infections by Diphyllobothrium. However, it is important from an epidemiological perspective, because there is an urgent need to detect the most important sources of plerocercoids, in particular those of D. pacificum and other marine taxa transmitted by yet-unknown sea fish. Molecular methods have been proven to be a powerful tool for the specific identification of causative agents of human disease, but a fast, cheap, and simple molecular-based diagnostic method for the routine laboratory evaluation of clinical samples is still unavailable.
Treatment of human cases does not seem to represent a serious problem at present, unlike a generally low awareness of the infection risk when eating raw or undercooked fish. Better education of all population segments, such as consumers, health professionals, fishermen, and sellers, particularly in the regions under potential risk of infection, is necessary (153). A more rigorous sanitary inspection of fish products before they are transported throughout the globe should also be applied.
Many aspects of the biology, epidemiology, and control of the broad fish tapeworm are still poorly known and require intensive research. However, it is hoped that warning of the potential risk of reemergence of diphyllobothriosis due to changing eating habits, globalization of the food market, and climatic change will help in a more effective control of this parasitic disease on the global scale.
We are indebted to numerous persons who kindly provided valuable information as well as those who sent clinical samples of tapeworms for molecular evaluation. Special thanks are due to Manuel Tantaleán, Universidad Peruana de Cayetano Heredia, Lima, Peru, for providing valuable information on D. pacificum and Andrea Gustinelli, University of Bologna, Bologna, Italy, for material. Helpful suggestions and critical remarks of anonymous reviewers are also greatly appreciated.
This study was partly supported by the Grant Agency of the Czech Republic (project no. 524/04/0342 and 524/08/0885) and the Institute of Parasitology (project no. Z60220518 and LC 522).