53 Cystic echinococcosis

Cystic echinococcosis (CE)\cystic hydatid disease is one of the most widespread and important global helminth zoonoses. The parasite Echinococcus granulosus is maintained in a wide spectrum of intermediate hosts, including sheep, goats, camels, cattle, pigs and equines. A number of wild intermediate hosts occur, including cervids in the northern part of the North American continent and Eurasia, marsupials in Australia and wild herbivores in East and southern Africa. The application of a range of molecular techniques to the characterization of the parasite has confirmed the existence of mostly host-adapted strains and genotypes of the parasite and several new species have been proposed. The ubiquitous domestic dog serves as the most important definitive host for the transmission of the parasite throughout its wide geographical range.

A wide range of diagnostic techniques, including necropsy, arecoline purgation, coproantigen ELISA and DNA based tests are available for detecting E. granulosus infection in the definitive host. In intermediate animal hosts, diagnosis at post mortem still remains the most reliable option. In humans, imaging techniques including ultrasound, nuclear magnetic resonance (NMR) or computer aided tomography (CAT-scan) provide not only a method of diagnosis but also reveal important clinical information on the location, condition, number and size of the hydatid cysts in man. Of these ultrasound is the most widely used diagnostic technique and is the only imaging technique for screening of populations in rural areas, where the disease is most common. A classification system has been developed which can be used to assess the likely development of a cyst and hence guide the clinician in treatment options for the patient. Treatment relies on surgery and/or percutaneous interventions, especially ‘Puncture, Aspiration, Injection, Re-aspiration’ (PAIR) and/or antiparasitic treatment with albendazole (and alternatively mebendazole).

CE is largely a preventable disease. Successful elimination programmes have focused on frequent periodic treatments of dogs with anthelmintics and the control of slaughter of domestic livestock. In many regions elimination or even control remains a problem as the parasite is endemic over vast areas of low income countries where there may be limited resources for control. In some areas, such as former communist administered countries, the parasite is resurgent. New tools are becoming available to control the parasite, including a highly effective vaccine in sheep which prevents the infection in sheep and breaks the transmission cycle. In addition cost effective methods are being developed which may be appropriate in low income countries where financial resources are not available for intensive control programmes that have been successful in high income countries.

Perhaps one of the first references to hydatid disease may be that of Hippocrates (379 bc) with other early references including the works of Galen (ad 139–200), Aretaeus (ad 7–79), and Rhazes (ad 860–932). The cyst was often regarded as enlarged degenerating glands, pus, or end blood vessels (reviewed in Dew (1928); Schwabe (1986)).

Redi (1624–94) first recognized that the cyst was of animal origin followed by Pallas who considered them to be a bladder forming parasite. Goeze divided this genus into visceral and intestinal taeniasis. The name Echinococcus was introduced into zoology by Rudolphi in 1801. In 1808, Rudolphi described three species within the genus Echinococcus: E. hominis, E. simiae, and E. veterinorum. In 1855, Kuchenmeister described two forms of Echinococcus: E. scolicipariens and E. altricipariens in which protoscolices and daughter cysts were formed, respectively. Von Siebold first experimentally infected dogs with Echinococcus veterinorum = Taenia echinococcus (Von Siebold 1853). This was followed by Naunyn in 1863 in Berlin and Krabbé in Iceland who independently infected dogs with adult worms derived from human protoscolex material. Haubner in 1855 first infected sheep. At that time, Virchow recognized that the condition known as colloid carcinoma was caused by a larval cestode. Subsequently, Leuckart described a form nominated as Taenia multilocularis, but did not regard it as a distinct species. Although Leuckart provided the first clear account of the life cycle before the end of the nineteenth century, speciation within the genus Echinococcus (Rudolphi 1801) was not solved until the mid-twentieth century.

In Iceland in 1864 Krabbé wrote a 16-page pamphlet which explained the life cycle. Control was initiated in 1869. Dog numbers were limited by taxation and some were treated with areca nut, but the two most important factors in reducing prevalence were a change from wool to lamb production and the reading of Krabbé’s pamphlet. Of 15,888 autopsies carried out between 1932 and 1982, 214 had CE. All but eight were born before 1900. The last non-latent case was found in 1960 in a 23 year old woman, but the last two latent cases were found in 1984 and 1988: born respectively in 1905 and 1920. The last two cases of ovine echinococcosis were observed in 1979 (Dungal 1957; Schwabe 1969; Beard 1973a, b).

The introduction of E. granulosus into Australia preceded knowledge of the life cycle, and it was considered prior to 1867 that ‘hydatids’ in humans was caused by eating undercooked sheep meats. From that time, information from Europe spread rapidly and it was recommended in 1898 in the sixth Annual Report of the New Zealand Department of Agriculture that all dogs should be treated with areca nut. Several attempts were made to introduce control there and elsewhere, but, except for Iceland, they were without success until the second half of the twentieth century (Gemmell 1990).

For the first half of the twentieth century controversy existed if the alveolar form of Echinococcus was a separate species. It was not until the 1950s that the cestode observed in a rodent on St Lawrence Island by Rausch and Schiller (Rausch and Schiller 1951) was considered to be conspecific with the parasite causing alveolar hydatid disease in Eurasia (Rausch 1953). Initially nominated as E. sibericensis this was re-nominated as E. multilocularis Leuckart, 1863 (Vogel 1955). Two further species have been regarded as valid. Taenia oligartha Diesing, 1863 was transferred to the genus Echinococcus in 1910 by Lühe and the life cycle of E. oligarthrus was determined experimentally (Sousa and Thatcher 1969). A fourth species, E. vogeli, was subsequently described by Rausch and Bernstein (1972) in Ecuador.

Several species which were nominated during the twentieth century are now considered synonymous with E. granulosus (Schantz et al. 1975, 1976). In addition a further species has been described from the Tibetan plateau, E. shiquicus (Xiao et al. 2005).

There is also strain variation in the species now recognized as E. granulosus and there are proposals for taxonomic revision. This chapter will be primarily concerned with CE, caused by the larval stage of E. granulosus, and possibly some of the other newly elevated species of Echinococcus which were previously considered part of the taxon E. granulosus.

The suborder, Taeniata Skriabin (Schults 1937) and (subclass Eucestoda, order Cyclophyllidae) consists of a single family, Taeniidae (Ludwig 1886), to which belong the eucestodes of the greatest medical significance (Rausch 1993). Two monotypic subfamilies, Taeninae (Stiles 1896) and Echinococcinae (Abuladze 1960) are recognized. The latter subfamily contains a single genus Echinococcus (Rudolphi 1801).

Recently using molecular techniques, the phylogeny of E. granulosus has advanced considerably. Findings may vary depending on which gene is sequenced and if mitochondrial DNA or nuclear DNA is used for strain typing. The various strains/subspecies defined by molecular biology are often named after their main intermediate host (e.g. sheep strain, cattle strain etc.). However except for the horse strain, host specificity of each strain/subspecies or newly identified species does not occur. Until recently 10 strains of E. granulosus were proposed: G1–G10 (Thompson and McManus 2002; Thompson 2008) (Table 53.1). A taxonomic revision has now been recommended with some of these strains being elevated to species. E. granulosus strains G1 (the common sheep strain), G2 (Tasmanian sheep strain) and G3 (the buffalo strain) are very close genetically, with no more than 6 base pair variation in the cox1 gene and one variant between G2 and G3. This compares to at least 30 base pair differences between these and other genotypes. Likewise 12s Ribosomal DNA does not discriminate between G2 and G3 strains (Vural et al. 2008; Rinaldi et al. 2008). Consequently these may turn out to be minor variations of E. granulosus sensu stricto. Indeed from an epidemiogical and biological point of view each of these strains readily infects sheep, buffalo and man and a number of other species. In cattle they often produce non fertile cysts. Globally the G1 strain is the most widespread and is responsible for the vast majority of cases of human CE.

Molecular evidence now places the G4 strain in a separate species: E. equinus. The molecular variation between E. equinus and E. granulosus sensu stricto appears to be at least as great as that between E. multilocularis and E. granulosus. In addition E. equinus appears to infect only one group of intermediate hosts—equines and there are no reported cases of human infection with this parasite (Thompson 2008).

E. granulosus G5 strain (the cattle strain) has major genetic differences from E. granulosus sensu stricto and elevation to species status—E. ortleppi has been proposed. E. ortleppi produces highly fertile hydatid cysts in cattle and infects humans. The proposal to elevate this parasite to the status of E. ortleppi confirms the original taxonomic proposal in 1943 (Lopez-Neyra and Soler Planas 1943).

E. granulosus strains G6 (the camel strain), G7 (the pig strain), G8 (North American cervid strain) and G10 (the Scandanavian cervid strain) appear to be a genetic clade. E. granulosus G9 was reported from a series of patients in Poland (Scott et al. 1997). Further studies suggested no significant differences between the mitochondrial sequence data between this strain and the pig strain (G7) (Kedra et al. 1999). Presently these strains are still considered strains of E. granulosus although there may be sufficient genetic to justify elevating this group to species status. E. canadensis has been proposed for G6, G7 and G8 strains (Nakao et al. 2007). Alternatively the cervid strains G8 and G10 may represent one species (E. canadensis) with G6 and G7 a second species (E. intermedius) (Thompson 2008). Recent studies of the nuclear genome tend to confirm the findings of the mitochondrial genome analysis and support the proposal that the strains G6-G10 should be two species with G6, G7 and G9 being E. intermedius with G8 and G10 being E. canadensis (Saarma et al. 2009). The taxonomic status, however, is not settled and other proposals may be made with new molecular tools and a greater understanding of the parasites’ biology. All these strains appear to be infective to man. Human CE caused by the pig strain has been reported from Europe and Turkey (Turcekova et al. 2003; Pawlowski and Stefaniak 2003; Schneider et al. 2007; Snabel et al. 2009,). The camel strain G6 has been isolated from an increasing number of human CE cases including from Africa (Dinkel et al. 2004; Casulli et al. 2009), the Middle East (Sadijadi 2006) western China (Bart et al. 2006) and South America (Guarnera et al. 2004) The cervid strains, whilst infective to humans appear to be relatively benign (Castrodale et al. 2002).

A further development has confirmed that lions are definitive hosts for a genotype of Echinococcus, provisionally designated E. felidis. This parasite appears to be identical to the E. felidis isolated from lions previously by Ortlepp (1937). To date, hydatid cysts of the E. felidis genotype have only been reported in warthogs (Phaeoceros spp) (Huttner et al. 2009). It is unknown if E. felidis is infective to man.

These Echinococcus genotypes have a global distribution (Fig. 53.1)

Echinococcus spp. are small tapeworms 2–7 mm in length rarely possessing more than five proglottids. They require a definitive (final) host in which the adult develops in the small intestine, and an intermediate host in which the metacestode usually develops in the visceral organs. The principal definitive host of E. granulosus is the domestic dog with domestic animals, such as sheep, goats, cattle, camels, and pigs, as intermediate hosts. Several wild life cycles exist including for example the dingo (Canis familiaris dingo) and macropodid marsupials, particularly the swamp wallaby (Wallabia bicolour) in Australia (Jenkins 2006), the wolf and wild cervidae in the northern hemisphere (Thompson 2008) and the lion and warthog in Africa (Hüttner and Romig 2009) (Fig. 53.2). Humans are aberrant intermediate hosts and become infected by ingestion of Echinococcus spp. eggs.

Ingested protoscolices develop as tapeworms in the small intestine. Reproduction is sexual. Development involves proglottidization and formation of new reproductive units or proglottids and their maturation. Onset of egg production is between 34 to 58 days. It has been estimated that gravid proglottids are shed every 7–14 days. The number of eggs per proglottid is variable ranging between 100 and 1,500 but possibly averages about 600 eggs. Limited studies on the longevity of the worms in dogs suggest it to be approximately 8 months to 1 year (Aminzhanov 1975). Indirectly the longevity has also been estimated through transmission models which also suggests approximately 1 year (Torgerson et al. 2003b).

Taeniid eggs are similar in size (30–40 μm), shape, and structure, and cannot be differentiated morphologically. Taeniid eggs are immediately infective after being passed by a definitive host and the larval stage or oncosphere of Echinococcus hatch in the small intestine and rapidly and actively penetrate it, aided by hook movement and possibly secretions. It has been suggested that transfer may occur via lymphatic or venous migration (Heath 1971). Once established, the oncosphere rapidly undergoes a series of changes to develop into a metacestode (Thompson 1995).

The metacestode of E. granulosus is typically unilocular, subspherical and fluid-filled. It consists of a thin germinal layer supported by a strong acellular laminated layer of variable thickness. Asexual proliferation of the germinal layer and brood capsule formation takes place endogenously. Pouching of the wall may give rise to secondary chambers and a multilocular appearance. In some cases daughter cysts may develop within the primary cyst (Rogan et al. 2006). Reproduction is asexual and thus has an almost unlimited generative capacity (Whitfield and Evans 1983). Brood capsules develop in the perinuclear layer and within them protoscolices are generated. Fertile cysts may occur in about 195 days in mice, 10–12 months in pigs, or 1 year in sheep.

There are no clinical signs of E. granulosus infection in dogs. In intermediate hosts there are often no obvious clinical signs, but evidence suggests that E. granulosus may cause significant reductions in productivity such as reduce fertility and wool, meat and milk production (Torgerson 2003a). Although many studies that suggest such losses were not well designed, such potential losses should be considered when examining the societal burden of echinococcosis (Carabin et al. 2005).

The sedimentation and counting technique (SCT) is considered the gold standard (Eckert et al. 2001a). The carnivore intestine is opened, incubated in physiological saline and the intestinal mucosa is scraped with a spatula. For mass screening, the intestinal scraping technique (IST), is a less laborious method with a sensitivity of 78% compared to the SCT (Eckert 2003). Deep mucosal scrapings (total 15 per intestine) are squashed into a thin layer and examined microscopically: semi quantitative estimation of the worm burden is possible. Safety precautions must be strictly followed using both these diagnostic strategies (Eckert et al. 2001c). An obvious major disadvantage of necropsy is it requires the killing of the dog and is therefore only useful for surveillance in strays and unwanted dogs.

Oral administration of arecoline hydrobromide results in the purgation of intestinal contents after 30–60 minutes. This material can then be examined for the presence of Echinococcus. This technique was used during the eradication campaign in New Zealand for mass surveillance (Gemmell 1990). The technique has nearly 100% specificity and has been used for quantitative studies. Safety precautions during field work and parasite identification in the laboratory are essential and time consuming (Eckert et al. 2001c). Arecoline can also cause serious, adverse reactions in dogs requiring strict veterinary supervision. The sensitivity is poor with reports of 38% and 65% (Schantz 1997; Ziadinov et al. 2008). Nevertheless, it is a potentially useful technique, particularly when evaluating other diagnostic procedures as it can prove the presence of infection.

The development of ELISA for the detection of antigens in the faeces of dogs was initially developed by several groups (Allan et al. 1992; Deplazes et al. 1992). Sensitivity of coproantigen assays is generally good with moderate to high worm burdens (>100 worms), but less in animals with low burdens. Coproantigens can also detect pre-patent infections (Jenkins et al. 2000; Allan and Craig 2006). Although sensitivity and specificity have often been defined in select groups of animals (e.g. experimentally infected animals), the actual field performance is less certain because of cross reactivity with antigens from other helminths. Test parameters will also vary with the population on which the test is used. Today coproantigen tests remain a useful procedure for population studies providing the potential pitfalls are fully understood when utilizing such tests.

Highly specific polymerase chain reactions (PCR) have been developed for use on faeces or isolated eggs to confirm the presence of Echinococcus spp. (Stefanic et al. 2004; Mathis and Deplazes 2006; Trachsel et al. 2007). As PCR is costly and time-consuming, it cannot presently be considered suitable for routine diagnostic or large-scale purposes. Copro-PCR is best used for confirmatory purposes of coproantigen-positive samples. Alternatively, it can be used as the method of choice for identification of the morphologically indistinguishable taeniid eggs recovered from faecal or environmental samples.

A significant problem is that large volumes of faeces are required to obtain a high sensitivity. PCR procedures may also suffer from significant inhibition (Mathis and Deplazes 2006). One approach, to overcome these limitations is to undertake the PCR on isolated taeniid eggs (Mathis et al. 1996). A study in Kyrgyzstan suggests the sensitivity of egg isolation followed by PCR is 78% for E. granulosus infections. Attempts to improve the sensitivity beyond this could include repeated faecal sampling and/or using PCR techniques that do not rely on egg isolation thus detecting pre-patent infections (Al-Sabi et al. 2007). The dynamics of copro-DNA excretion is dependent on loss of parasite stages (protoscoleces during the first days after inoculation and immature stages at the end of pre-patency) whereas coproantigen concentrations are related in this infection period to parasite metabolic activity (Al-Sabi et al. 2007). Direct detection of DNA in faeces of E. granulosus infected dogs revealed 15 dogs positive by copro-PCR during the prepatent period of 58 dogs proven to be infected by E. granulosus indicating a sensitivity of 26% during this phase of the infection (Lahmar et al. 2007b). A further important limitation of copro-PCR is that formalin-fixed faecal material is not suitable due to DNA degradation. However, faecal material stored in 70% ethanol or at −20 °C or −80 ^o can be used (Al-Sabi et al. 2007).

Serology in dogs for intestinal E. granulosus infections (Gasser et al. 1990) has been investigated. Problems reported with sensitivity, specificity and previous infections have prevented serology being employed in surveillance studies (Gasser et al. 1993; Craig et al. 2003; Eckert et al. 2001b).

Diagnosis is rarely indicated in the individual intermediate hosts but population surveys are important to obtain baseline epidemiological information, particularly in respect of control (Torgerson and Heath 2003). Generally, it is undertaken at post mortem (Torgerson et al. 1998). Sheep slaughtered for meat are often young which inevitably have a lower prevalence than older animals (Torgerson and Heath 2003). Furthermore, the better conditioned animals are sent to the abattoir with a smaller likelihood of infection. Thus, abattoir studies, especially if not age-stratified can produce substantial underestimates of the true prevalence. Ultrasound (US) examination can be used in sheep to detect hepatic cysts but the sensitivity is poor (Sage et al. 1998; Lahmar et al. 2007a).

Immunodiagnosis in sheep presents problems of sensitivity and specificity, limiting its applicability. Indirect haemaglutination, double diffusion and ELISA have all been attempted (Gatti et al. 2007). Poor specificity can often be attributed to cross reactivity with other taeniids (such as Taenia ovis and T. hydatigena). Recent studies using hydatid fluid antigen have reported a sensitivity and specificity of 89 and 90%, respectively. The specificity could be improved by using purified fractions of hydatid cyst fluid (S2B) but the sensitivity was decreased (Gatti et al. 2007). Previous studies using either hydatid cyst fluid, antigen B (AgB) purified from hydatid cyst fluid or protoscolex antigens have reported sensitivities ranging from 36% to 90% and specificities ranging from 65% to 96% (Yong et al. 1984; Ibrahem et al. 1996; Kittelberger et al. 2002).

At the end of the nineteenth century, the only drug available to treat dogs for tapeworms was the areca nut, Areca catechu. In 1924, a synthetic salt, arecoline hydrobromide, was developed and used for treating dogs. Up to 11 treatments may be needed to free some dogs from worms. Praziquantel was evaluated in the 1970s and found to be highly effective (Gemmell and Johnstone 1981). The drug was equally effective when given with or without food and can be successfully incorporated into a medicated dog food of high acceptability (Pu-Sheng 1993). It can also be given by injection. The drug does not have a practical ovacidal effect. Praziquantel has now been used in millions of doses without toxicity or drug resistance (Gemmell and Johnstone 1981). Praziquantel remains the drug of choice and cheap supplies of the product are now available as the patent has expired. Although other effective products, such as epsiprantal (Arru et al. 1990) have become available, there appears to be no reason not to continue recommending praziquantel as the drug of choice to treat Echinococcus infections in dogs.

Benzimadazoles such as mebendazole have been shown to be effective in sheep and pigs if treated daily 50 mg/kg for 3 months (Gemmell et al. 1981). It has been observed that in some areas CE in sheep has declined following the widespread introduction of albendazole for the control of nematodes (Santos et al. 2008) resulting in revived interest in using such products for the control of CE in livestock. There have been promising results with albendazole (Santos et al. 2008) and with oxfendazole combined with nitazoxanide (Gavidia et al. 2009).

Population studies suggest that if dogs are exposed to E. granulosus they will develop a protective immunity (Torgerson 2006a) and exploiting this with a vaccine could be a valuable control tool. Early research with crude antigens to vaccinate dogs resulted in little demonstrable success (Gemmell et al. 1986b). More recently claims were made for significant protection using experimental vaccines (Zhang and McManus 2008). Detailed analysis of these claims (Torgerson 2008) demonstrated that to date there is little progress in developing a vaccine in dogs.

There has been substantial progress in the development of an effective vaccine against E. granulosus in the intermediate host, particularly in sheep using the EG95 oncosphere antigen (Lightowlers et al. 2003). Presently the vaccine comprises a single recombinant oncosphere antigen that is expressed in E. coli and the adjuvant Quil A. It induces complement-fixing antibodies that kill the invading oncosphere early in an infection. In the majority of vaccinated animals, no hydatid cysts occur following a challenge infection but a small number of viable cysts may occur in some vaccinated animals. Animals are vaccinated subcutaneously on two occasions 1 month or more apart, and induce protection against infection which lasts for at least 12 months. A third injection given 6–12 months later induces a high and long-lasting protection against challenge infections. The vaccine has proved effective in vaccine trials carried out in ruminants in New Zealand, Australia, Argentina, Chile and China. GMP production scale-up of the vaccine has been undertaken in New Zealand and China and it is expected that the vaccine will be become available through these sources for implementation as part of hydatid control programs worldwide (Lightowlers and Heath 2004).

Human infection has been recorded in one year old children and in adults of more than 80 years of age (Drolshammer et al. 1973; Chaouachi et al. 1989; Utrilla et al. 1991). Abdominal CE infection is positively correlated with age, whilst surgical pulmonary cases are not (Macpherson et al. 2004). The peak age at the time of diagnosis is 30–50 years with females often over represented (Romig 1990; Schantz et al. 2003; Yang et al. 2006; Yolasigmaz et al. 2006; Yu et al. 2006). The incubation period and clinical picture is dependent on the organ involved (Grossi et al. 1991). The sites of predilection of 1,802 cysts recorded in the Australian Hydatid Register were liver (63%), lung (25%), muscles, (5%), bones (3%), kidney (2%); spleen, brain (1%), and heart, breast, prostate, parotid and pancreas (< 1%) (Grove et al. 1976). Similar figures are given from a series of 3,736 cases hospitalized in Western China: liver (65%), lung (27%), abdominal cavity (2%), brain (3%) and other locations (3%) (Wen and Yang 1997). Most patients have single cysts and the right lobe is more commonly affected than the left lobe of the liver (Romig 1990; Ammann and Eckert 1995). ‘Collossal’ cysts have been recorded particularly in patients with limited access to health care but the average size may be 1–10 cm (Todorov et al. 1992a).

Cysts have been reported to grow 1–50 mm per year (30%), 6–15 mm per year (43%)or > 30mm per year (11%) or to persist without change for years, or even collapse (4/75 cysts) (Romig et al. 1986; Mufit et al. 1998). Ultrasound (US) follow-up after mass surveys showed that half of the cysts failed to change in size despite having been diagnosed during the so-called evolutionary growth phase; in the other 50% increases were variable, ranging from a few millimetres to some centimetres (Wang et al. 2006). Cysts can spontaneously rupture collapse or disappear. A number of studies indicate that liver cysts grow at a slower rate than lung cysts (Larrieu and Frider 2001). The exact sequence of qualitative cyst changes during its natural history is still unclear (Rogan et al. 2006). Protoscolices may be found in daughter cysts of diameter 0.5 cm (Alvarez et al. 1991).

Symptomatology depends on:

Leakage following diagnostic puncture or spontaneous or post-traumatic rupture or fissure may induce mild or serious complications, such as urticaria, asthma, or anaphylaxis shock, due to allergic reaction to Echinococcus antigens (Vuitton 2004). Membrane nephropathy has also been described (Ammann and Eckert 1995). Secondary peritoneal echinococcosis may result from spontaneous, traumatic rupture, or rupture during surgery of a hepatic cyst, but is rare from a pulmonary cyst (Ammann and Eckert 1995). The consequences of secondary CE are recognized several years later. Cysts that become encapsulated and calcified, although potentially viable, may remain asymptomatic indefinitely accounting for a relatively high proportion of cases being asymptomatic during ultrasound surveys (Ammann and Eckert 1995). Spontaneous cure of CE does occur due to collapse and resolution, calcification, or cyst rupture into the bronchial tree or bile ducts (Ammann and Eckert 1995).

Ultrasound examination (US), a widely used technique, can confirm the diagnosis of abdominal echinococcosis and indicate if lesions are active. Further information can be found in many sources (WHO-IGWE 2003; Macpherson et al. 2003; Rogan et al. 2006; Kern et al. 2006; Bresson-Hadni et al. 2006). Pulmonary echinococcosis cannot normally be detected by US. US for mass screening is generally considered to have a sensitivity of 88–98% and a specificity of 93–100% for abdominal lesions (Macpherson and Milner 2003; Macpherson et al. 2003b). However, the sensitivity of this technique for detecting all forms of echinococcosis is somewhat lower. Using technical modifications, a proportion of lung cysts might be detected (El Fortia et al. 2006).

Nuclear magnetic resonance (NMR) or computer aided tomography (CAT-scan) can be used to confirm the diagnosis, especially for lung and brain cysts, and to perform a pre-therapeutic assessment of the lesions, for all locations (Turgut et al. 2007). Unfortunately, in many remote endemic areas such facilities are not available. In this case, serological back-up tests may be required to give additional information regarding the nature of the lesion detected by US.

US has been used to study the evolution of cysts over time, the structural integrity of cysts following chemotherapy (WHO-IWGE 2003) and for use in screening and surveillance programmes (Macpherson et al. 2003a). A classification system has been developed which can be used to assess the likely development of a cyst and hence guide the clinician in treatment options for the patient (Fig. 53.3) (WHO-IWGE 2003).

The widespread use of the WHO classification provides a better understanding of the epidemiology, biology, public health importance, control and treatment of CE globally, through:

The WHO classification differs from Gharbi’s original classification (Gharbi et al.1981) by introducing a cystic lesion (CL) stage and by reversing the order of CE Types 2 and 3 (Table 53.2; WHO-IWGE 2003; Brunetti et al. 2009). The number of cyst types remains unchanged from Gharbi’s classification and the types are readily categorized into active, transitional and inactive stages. The inclusion of the CL stage reflects the detection of early, undifferentiated cysts found during mass screening. CL cysts are not included as a Type of CE as they have to be further evaluated before being classified as CE. The first CE Types are CE1 and 2 and such cysts are active, usually fertile cysts containing viable protoscoleces. CE3 are cysts entering a transitional stage where the integrity of the cyst has been compromised either by the host or by chemotherapy. CE4 and 5 are inactive cysts that have normally lost their fertility and are degenerative. Data on long-term follow-up of cysts treated with albendazole and percutaneous puncture (PAIR) provide ground for a further sub-classification of CE3 transitional cysts into CE3a (detached endocyst) and CE3b (predominantly solid with daughter cysts) (Junghanss et al. 2008). This has important implications for clinical decision making and prognosis.

A large number of serological assays have been developed for the diagnosis of human echinococcosis (reviewed in Carmena et al. 2006). Native crude antigens such as hydatid cyst fluid have generally proved to be sensitive. In a variable proportion of CE patients, no specific antibodies could be demonstrated with a variety of tests, especially those with lung cysts. The highest sensitivity, observed for liver cysts, may be due to complex parasite-host interactions modulating the immune system (Riganò et al. 2001) Specificity is a problem with cross reactions with other helminths (Eckert and Deplazes 2004). Specificity was considerably increased on genus or even species level by purification of components such as AgB in hydatid cyst fluid of E. granulosus metacestode material. Various groups have developed diagnostic tests based on recombinant AgB with better test performance than native AgB (reviewed by Mamuti et al. 2005; Carmena et al. 2006), and synthetic peptides mimicking defined epitopes have also been investigated. Sensitivity and specificity of tests based on AgB have largely been evaluated on known hydatid disease patients or known healthy individuals where sensitivity has varied between 0.45 and 0.92. Specificity is reported as between 0.71 and 1 with cross reactivity against AE, cysticercosis, Schistosoma and Toxocara (Carmena et al. 2006).

Serology also detects exposure to the parasite before disease development and exposure without parasite establishment. This, with the less than ideal level of test specificity explains why sometimes positive serological results are found during mass screening in endemic areas in patients without any cyst disclosed by US. The problems associated with detecting false positives prevent serology alone from being used for screening for CE.

There are numerous important reviews on the therapeutic management of CE. However, it still is a controversial subject (Dziri et al. 2004). Few controlled studies on the treatment of CE with a proper follow-up of patients are available (Vuitton and Wen 2007). Consensus opinion was reached within the WHO-IWGE and the main conclusions published in 1996 (WHO-IWGE 1996) with updated recommendations published in 2009 (Vuitton 2009; Brunetti et al. 2009). Surgery was the only option until the late 1970s. Complementary or alternative options are now available, which include non-surgical interventions and chemotherapy with anti-parasitic drugs. Treatment indication should be based on a multidisciplinary discussion and depends on cyst type, number and location, and presence or absence of cyst complication. A proper and long-term follow-up of the patients should assess the efficacy of the treatment, detect treatment complications, and timely disclose recurrences.

To avoid protoscolex spillage during operation, viable components of the parasitic cyst must be inactivated with hypersaline solution or hydrogen peroxide prior to any cyst opening. Use of formalin must be totally forbidden; but none of the protoscolicides are totally safe (Ammann and Eckert 1995).

The main complication is sclerosing cholangitis, which is possible when there is communication between the cyst and the biliary tree (Ammann and Eckert 1995). There is still debate on the necessity, modalities and time schedule of anti-parasitic chemotherapy before and/or after interventional treatments, including surgery (Vuitton and Wen 2007).

The principal methods include hepatic resection; pericystectomy and cystectomy for hepatic cysts; and opening, sterilization and partial removal of the cysts, also called partial cystectomy; pericystectomy and lobectomy for pulmonary cysts (Ming-Quian 1993). Laparoscopic surgery is a technical option in selected cases but possibly increases protoscolex spillage. Cysts should be removed as much as possible; however, the more radical the intervention the higher the operative risk, but with the likelihood of fewer relapses, and vice versa (WHO-IGWE 1996). Following the cleavage plane between the inner layer of host’s reaction facing towards the parasite and the outer layer, or ‘adventitia’, limits the damage to liver parenchyma when dissecting around the cyst and allows a safer removal of the cyst by total cystectomy (Wu et al. 2004). Postoperative complications may occur in between 10 and 25% of cases (Kammerer and Schantz 1993). Some of the untoward sequelae that may be expected are summarized in Table 53.2. Recurrence rates vary from 2 to 25% (Romig 1990; Todorov et al. 1992b) and are more frequent after partial operations. Absence of long-term follow-up and/or studies restricted to small series, in selected patients, and/or in highly specialized settings of developed countries, usually explains the most optimistic results. In an extensive review in China, 92, 7, 0.8 and 0.2% had 1, 2, 3, and 4 or more operations (Menghebat et al. 1993). In a study in Kenya, among 663 patients, there were 2 deaths, one intraoperative and one postoperative, after surgery, and 47 patients had repeated operations because of postoperative complications and/or recurrences (Cooney et al. 2004).

Besides recurrence, biliary leakage is the most frequent complication and the most difficult to manage. Lethality associated with the first operation is about 2% (Kammerer and Schantz 1993), but it increases substantially after subsequent surgery (Amir-Jahed et al. 1975).

Percutaneous puncture (PAIR) for inoperable cases is currently accepted as an alternative to surgery in selected cases. This includes puncture of the cyst, aspiration of the fluid content of the cyst, introduction of a protoscolicide, such as hypertonic saline or, preferably, alcohol and re-aspiration. It is carried out under ultrasonic guidance (Kammerer and Schantz 1993; Ammann and Eckert 1995; Filice et al. 2000). Detailed practical guidelines have been published by the WHO-Informal Working Group (WHO-IWGE 2001). In experienced hands, PAIR represents the first choice of treatment in middle-sized non-complicated cysts (Filice et al. 1990). Long term follow-up of patients is now available (Akhan et al. 1996). A meta-analysis has supported the efficiency, safety and usefulness of the procedure, in selected indications (Smego et al. 2003). A very limited number of anaphylactic shocks, usually reversible with appropriate resuscitation, have been reported. In most series, recurrence of the cysts is lower than after surgery. PAIR can be proposed for type CE1, and selected cases among CE2, and CE3 cysts. It is contraindicated if there is communication of the cyst with the biliary tree. Drainage may be associated with PAIR for large cysts (Filice et al. 1990). To treat cysts with numerous daughter cysts, modified techniques using larger tubes and vacuum aspiration through a small surgical incision have been described, but the rate of long-term recurrences seems to be high (Brunetti et al. 2009). Prior to, and for a period following percutaneous interventions, benzimidazoles should be administered. In the most accepted administration schedule, albendazole (ABZ) is given the day before and during 1 month after (Brunetti et al. 2009). ABZ should not be given when PAIR is performed during pregnancy.

Benzimidazole compounds are potential parasitocidal agents against E. granulosus. ABZ is currently the drug of choice to treat CE, either alone or as an adjunctive/prophylactic therapy together with surgery or percutaneous interventions. The drug is given orally at a dosage of 10–15 mg/kg/day (normally rounded to 400mg twice daily), accompanied by a fat-rich meal to increase bioavailability. It should be given continuously (Brunetti et al. 2009). Alternatively, mebendazole (MBZ) is another benzimidazole compound which may be used if ABZ is not available or not tolerated by the patient. MBZ daily dosage is of 50 mg/kg body weight, in three divided doses during fat-rich meals. Both drugs should be avoided during pregnancy, especially for the first trimester (WHO-IWGE 1996). Main side-effects are haematological and hepatic toxicity, and alopecia. Blood count and serum aminotransferases should be monitored every 2 weeks for the first 3 months, then monthly. Optimal dosage and optimal durations have never been formally assessed for either of the drugs (WHO-IWGE 1996; Ammann and Eckert 1996). Serum blood levels of ABZ sulfoxide or of MBZ should be monitored 1 month after commencement of treatment and then every 3 months (WHO-IWGE 1996).

Most studies have used echographic methods to assess benzimidazole efficiency (el-Mufti et al. 1993; Wen et al. 1993). Difficulties have been found in ascribing death of cysts to the treatment. Efficacy with standard 800 mg ABZ daily for 3–6 months produces an average of 30% cure and 45% response. It is more effective in younger patients and for small CE1 and CE3a cysts with a thin wall without infection or communication. It may be less effective for CE2 (Horton 1997). Three available randomized controlled trials showed that ABZ alone had a better effect on hydatid cysts than placebo or MBZ (Gil-Grande et al. 1993; Franchi et al. 1999; Keshmiri et al. 2001). Complete disappearance of all cysts was never reached and therefore chemotherapy is not the ideal treatment when used alone. In addition, its efficacy is lower for lung cysts or bone cysts than for liver cysts. This therapeutic option is indicated for inoperable patients, for patients with multiple cysts in liver or in two or more organs, and for patients with peritoneal cysts (Brunetti et al. 2009).

Praziquantel has been tested for protoscolicidal efficacy in vitro. In one in vivo study involving 101 patients, there was very little difference in the viability of protoscolices between the treated and untreated patients as determined at surgery (Jia-Zhong et al. 1993). One prospective controlled trial compared ABZ and praziquantel (25mg/kg/d) versus ABZ alone (Mohamed et al. 1998) and concluded that the combined treatment was more effective than ABZ alone. Combined treatment is usually administered in severe cases with multiple cysts and/or several locations, and in bone CE.

E. granulosus has both sylvatic and domestic cycles. It is the latter transmission cycle that is the most common and poses the greatest threat to human health. The highest incidences in man are seen where there is a close association with man and domestic livestock, often using working dogs. A common source of infection for dogs is offal from infected sheep. The resultant high infection levels in these dogs then pose a risk to humans. The cohabitation with dogs and feeding of uncooked viscera is a known risk factor for human CE (Campos-Bueno et al. 2000).

The potential for domestic transmission of E. granulosus is highest in countries where the level of education may be poor, veterinary and medical services inadequate and where home slaughter is commonly practiced. In such circumstances, the rates of infection in dogs can reach between 20 and 50% with perhaps an excess of 50% of the sheep population being infected. Dog contact is a major risk factor as they are the definitive host (Tiaoying et al. 2005; Yu et al. 2006; Moro et al. 2008). Dogs themselves are more likely to become infected if they are young, allowed to roam, fed on raw offal, offal in the community is not disposed of properly, the dogs not receiving anthelmintic treatment or the dogs’ owners being ignorant of the disease (Macpherson 2005; Budke et al. 2005a, b; Buishi et al. 2005, 2008; Ziadinov et al. 2008).

As a generalization, human CE is linked to the prevalence in domestic livestock, particularly sheep and livestock husbandry practices (Bai et al. 2002; Torgerson et al. 2002; Schantz et al. 2003; Moro et al. 2008; Ahmadi and Hamidi 2008). Occasionally other species, such as camels and pigs, may also be intermediate important hosts (Kedra et al. 1999). In Europe, autothochonous CE is generally rare in central and northern Europe, although there are some foci in eastern Europe which are believed to be mainly through transmission of the pig strain (Romig et al. 2006). The most intensely infected areas are Spain, where in some districts human incidence rates are 1.1–3.4/100,000/year (Carmena et al. 2008) and Italy, particularly Sardinia where annual human incidence rates are 3.5/100,000 (Castiglia et al. 2004). CE is also an emerging problem in Greece, Bulgaria and Romania where incidences of 3.3/100,000/year have been recorded (Todorov and Boeva 1999). There is also a small focus in Wales in the UK.

CE is a significant problem across much of the Middle East and North Africa. In Jordan the annual incidence is 2.9/100,000 (Kamhawi 1995), in Tunisia 15/100,000 (Majorowski et al. 2005), 10–30/100,000 in rural parts of and in Turkey 0.67–6.6/100,000 (Altintas 2003). In Central Asia surgical incidence rates are commonly between 10 and 20/100,000 (Torgerson et al. 2006).

Certain communities in Tibet have some of the highest incidences of CE. In some villages US prevalences range between 5–10% (Budke et al. 2004; Yu et al. 2008). Similar disease burdens have also been recorded in transhumant pastoral communities in East Africa, such as the Turkana and Masai in Kenya and Tanzania, Toposa in Sudan and the Dassanetch and Nyangatom in southern Ethiopia (Macpherson et al. 1989  ;  Magambo et al. 2006).

In Latin America there are large endemic areas throughout the Andean regions of Peru, Argentina, Chile, Uruguay southern Brazil and sporadic cases being reported elsewhere such as Mexico (Moro and Schantz 2006).

In the US and Canada, CE tends to be sporadic and rare with cases occasionally being reported in certain groups of native Americans or particular ethnic groups (Moro and Schantz 2006).

In Australasia CE was introduced with European colonization (Jenkins 2005) and became a problem in large sheep rearing areas. Successful control programmes have resulted in the elimination of the parasite from New Zealand and Tasmania (Craig and Larrieu 2006). In continental Australia the parasite has established a wild life cycle between macropod marsupials and dingoes. Therefore prospects for elimination are now considered bleak. There are frequent descriptions of transmission within the domestic cycle and human CE is recorded not uncommonly (Jenkins 2005, 2006).

The risks associated with infection are illustrated by the deteriorating situation in Central Asia. Prior to the breakup of the Soviet Union, CE in man was at relatively low levels. Following independence of the Central Asian republics there was widespread structural and economic reform. This resulted in privatization of farms, abandonment of centralized meat processing facilities and a return to small subsistence-type agricultural practices. Veterinary services collapsed due to a lack of government funding. This resulted in an epidemic of human CE, with the annual incidence of surgical cases reported by hospitals in excess of 4–5 times the number reported prior to 1991 (Torgerson et al. 2006). A similar pattern is also emerging in other former communist countries like Bulgaria (Todorov and Boeva 1999). Unregulated slaughtering of domestic animals results in greater possibilities for dogs to become infected. Dogs are more numerous than previously as livestock units are smaller and greater numbers of dogs are required for shepherding. Small farms are also generally located closer to human population centers compared to the large state run farms previously. Thus there are more dogs with a higher prevalence of infection in greater contact with humans (Torgerson et al. 2006; Shaikenov et al. 2003). Another noticeable feature of these emerging epidemics in former socialist countries is the high numbers of children with the disease, with perhaps up to one third of surgeries in children under 14, indicating recent transmission (Torgerson et al. 2003a; 2006).

There is a growing body of evidence that CE may be a cause and consequence of poverty within endemic zones (Torgerson et al. 2001, 2003a, 2009b; Budke et al. 2004, 2005c). It is uncertain if this is cause and effect. Greater poverty may render individuals more susceptible to infection or as CE is a devastating disease, infection may lead to poverty such as through loss of employment.

Surprisingly dog contact is not always a risk factor for transmission of CE, especially in climates where eggs may survive in the environment for long periods of time (Carmona et al. 1998; Bai et al. 2002; Torgerson et al. 2003a). Lack of association with contact may be a consequence of infection occurring many years earlier and recall is difficult for subjects of epidemiological studies. Also of importance is the fact that in many highly endemic regions dog ownership and dog contact is almost universal, so this factor cannot discriminate between infected and non infected individuals (Torgerson et al. 2009) and hence it is the epidemiology of infection of the dog (see above) that becomes important. In some societies, such as in North Africa, the Middle East and Turkey dogs are considered unclean and hence there is a reluctance to have close contact with them (Dowling et al. 2000). Despite this CE infection amongst many of these people is evidence for indirect transmission of echinococcosis to man through contaminated food or water supplies (Carmona et al. 1998; Dowling et al. 2000; Larrieu et al. 2002; Torgerson et al. 2003a; Tiaoying et al. 2005). Epidemiological and experimental evidence has demonstrated that taeniid eggs can be transmitted considerable distances by mechanical carriers such as insects (Gemmell 1990). Thus, in highly endemic areas it is quite possible for individuals to contract CE even in the absence of dog contact.

Women are often reported to have a higher incidence of infection than men (Romig 1990; Schantz et al. 2003; Yang et al. 2006; Yolasigmaz et al. 2006; Yu et al. 2006) and this may be because they are more likely to tend to the household dogs and be involved with food preparation. In some populations men are found to have a higher incidence (Torgerson et al. 2003a) and this may reflect that in some societies men are more likely to be treated because they are more economically active, rather than an actual increased risk for men. Increasing age is also often reported as a risk factor (Romig 1990; Yu et al. 2006; Ahmadi and Hamidi 2008) with the peak age being 30–50 years. This is most likely due to continuous infection opportunities over time and the chronic and asymptomatic nature of abdominal CE (Macpherson et al. 2004).

Absence of CE in children may indicate that transmission has ceased following successful implementation of control measures (Zanini et al. 2009). Changes over time to more advanced cyst types seen in all age groups, following the introduction of control measures may also suggest cessation of transmission. Increasing numbers of paediatric cases and or small early cyst types in various age groups may indicate a breakdown of control measures (Torgerson et al. 2002, 2003a).

CE presents a considerable societal and economic burden to a number of societies where E. granulosus is endemic. In terms of animal health the disease can lead to significant losses of production due to liver condemnations, lowered milk, meat and wool production or decreased fertility in animals (Torgerson 2003a). This particularly impacts on societies with low socio economic development which have the highest prevalences in livestock.

The societal burden in terms of human health is also of major consideration. In financial terms the costs of the disease include costs of surgical and medical treatment, the costs of convalescence such as loss of income and the costs of providing supportive care for individuals with chronic ill health. The direct treatment costs can vary depending on the country in which treatment is received. This can be as low as US$524 in a country such as Jordan (Torgerson et al. 2001) (2001 figures), to >$13,000 in the UK (Torgerson and Dowling 2001). This high variation reflects the different costs of medical care in different countries. In high income countries the costs of treatment are greater. However, the impact to individuals of such costs is likely to be much greater in countries where incomes are low as even the relatively low costs (by international comparison), may still be beyond the resources of patients from these areas. One of the consequences is that the incidence of CE in low income countries is much more likely to be underreported.

A number of studies have estimated the economic costs of CE in various countries and it is often considerable (Torgerson et al. 2000, 2001; Torgerson and Dowling 2001; Majorowski et al. 2005). There is also a financial estimate of the global costs of CE which suggests it could be as much as $3 billion annually (Budke et al. 2006) if estimates for underreporting of the disease are accounted for.

Although financial estimates of societal burden are important for zoonoses that have major impacts on both livestock productivity and animal health, other measures have also been calculated. The WHO preferred measure is the Disability Adjusted Life Year (DALY) (see Chapter 4, this volume). Estimates for DALYs have been made in a highly endemic region of the Chinese province of Sichuan and this indicates a loss of 0.81 DALYs per person in the population due to echinococcosis (Budke et al. 2004). Thus the health impact is considerable as the average numbers of DALYs lost to Chinese residents due to all causes is 0.18 DALYs. A preliminary estimate of the global burden on echinococcosis has also been made which indicates a loss of approximately 1 million DALYs (Budke et al. 2006). This is of a similar magnitude to the global burden of Trypanosomosis and considerably more than diseases such as Dengue or Leprosy.

Burden estimates present important methods in developing cost effective means of controlling this often neglected zoonoses.

Some experience has now been gained on applying control based on studies of field trials, control programmes and, more recently, mathematical modelling (Craig and Larrieu 2006).

Priority status for control include:

It should always be remembered that these two goals are not the same.

There are two models. The first creates, through specific legislation, a national or regional executive authority with responsibility for the control programme. The second utilizes an existing government organization (e.g. Ministry of Health or Agriculture). To an extent, the former is likely to be funded through a dog tax and latter through the legislature.

Depending on the programme to be adopted, areas in which legislation may be needed include:

Control of CE has always involved a combination of routine anthelmintic treatment of dogs, control and reduction of stray dog populations, supervision of the slaughter of livestock and subsequent disposal of offal, and education of the public. Gemmell and Roberts (1996) have previously described 5 options. The option of do nothing may be considered if there are more pressing public health issues that demand the use of resources. A slow track approach may focus on education and upgrading of facilities and development and possibly supplying drugs to treat dogs and such an approach may take many decades to complete. A third approach using arecoline testing as an educational approach is no longer

appropriate due to the advent of better diagnostic technologies and the hazards and welfare issues surrounding the use of arecoline. The fourth option was combining this with dog euthanasia, which may also have substantial resistance in many societies. The fifth option was based on the intensive treatment of dogs with praziquantel. The pre-patent period of E. granulosus is approximately 6 weeks and hence this has usually been the recommended treatment interval.

With new technology (e.g. the sheep vaccine) additional options are becoming available. Mathematical models have been developed to simulate control options (Torgerson and Heath 2003). Six weekly anthelmintic treatment is highly effective but is expensive and therefore less suitable for use in poor countries. Simulation models suggest it may be possible to lengthen the interval between anthelmintic treatments to at least 3 months and still reduce prevalence rates in dogs and livestock to less than 1% within 10–15 years (Torgerson 2003b). This does depend on treating perhaps 75% or more of the dog population. If only 60% or less is treated then failure to eliminate the parasite is possible. This idea has been supported by field studies in Uruguay (Cabrera et al. 2002) and New Zealand (Gemmell 1990). The lengthening of the treatment intervals to beyond the pre-patent period can work because the mean time to reinfection is often considerably longer than six weeks. Six monthly anthelmintic treatment only reduces the levels of echinococcosis substantially if the treatment rate is well in excess of 90%, which is unlikely to occur in practice.

Providing at least 75% of sheep are vaccinated, echinococcosis will be reduced considerably, but not for several years after implementation and intensive vaccination of sheep may become an option when the EG95 vaccine is commercially available. Alternatively an option of twice yearly anthelmintic treatment of dogs and vaccination of sheep should be considered. Providing 60% of sheep are vaccinated and 60% of dogs treated there is high probability of success within a time frame of 15–20 years (Torgerson and Heath 2003).

Recent work has demonstrated that old sheep contain most of the infective larval biomass (Torgerson et al. 2009c). This presents opportunities for a faster track approach to control by culling old sheep and hence instantly reduce the infection pressure to dogs by 80–90% resulting in a rapid cessation of transmission (Fig. 53.4). Alternatively suitable chemotherapy in sheep (Gavidia et al. 2009) might achieve a similar effect. Economic analysis should be an important priority before control is initiated (Torgerson 2003a) to develop the most cost-effective means of control.

Four phases of control can be recognized: a ‘preparatory’ and/or ‘planning’ phase; ‘attack’; ‘consolidation’ and, where appropriate, ‘maintenance of elimination’ phase (Gemmell 1987). During the costly ‘attack’ phase, the control measures are applied non-discriminately to the entire host population at risk. As soon as the parasite reaches certain low level it becomes more cost-effective to cease the overall attack and to target control in the ‘consolidation’ phase. If appropriate, once the parasite has been almost eliminated, the ‘maintenance of elimination’ phase may be entered. Here all specific activities are suspended or disbanded and ‘vigilance’ is permanently maintained through the normal meat inspection services.

The control programmes selected for review in this section, differ from one another in administration, resources used, methods applied, and rate of decline in transmission. The changes that occurred in the prevalence of E. granulosus in adult sheep are illustrated in Fig. 53.5.

The successful programme initiated in the nineteenth century in Iceland has been described previously.

In New Zealand in 1908, it was made compulsory for owners to register their dogs. In 1935, a strong educational campaign was introduced. In 1937 by an amendment to the Dog Act (1908), all owners who registered their dogs received sufficient arecoline hydrobromide to treat them four times a year. In 1940, it was made illegal to feed raw offal to dogs. No change was detected in the prevalence of echinococcosis in animals or humans 20 years later (Gemmell 1990).

Control in Uruguay was initiated in 1965 with the creation of a National Commission. Funding was initially provided by a dog tax and owners were also expected to purchase tablets and treat their dogs. No evidence could be found that any of the activities attempted between 1965 and 1991 modified the level of transmission of echinococcosis between dogs and sheep (Cabrera et al. 1995; Parada et al. 1995). Recent evidence from serological and ultrasound surveys also suggests that in some departments, up to 2% of the rural population may have asymptomatic CE (Bonifacino et al. 1991; Paolillo et al. 1991; Cohen et al. 1991). Since 1990, the programme has been transformed to Option 5).

Two programmes using a ‘slow-track’ approach have been completed in New Zealand and Tasmania (Craig and Larrieu 2006).

In both programmes, transmission of CE (Tables 53.3 and 53.4) ceased to children within about 7–12 years of their introduction, although the ‘maintenance of eradication’ phase was not reached until about 35 years of their introduction. Both New Zealand and Tasmania are now believed to be free of Echinococcus.

Cyprus initiated stray dog euthanasia in 1970 and a dog-testing programme with arecoline surveillance in 1972. In 1974, partition occurred, but control continued in the government-controlled area. There, transmission between animals rapidly ceased and in 1985 eradication was assumed (Economides et al. 1998). More recently, small foci have reappeared confirming the need to maintain ‘vigilance’ even when the parasite becomes difficult to find with current diagnostic tools (Christofi et al. 2002).

The Falkland Islands introduced a 6 weekly dog-dosing programme in 1970 with praziquantel. Here the slope of the decline in the prevalence of E. granulosus in sheep and duration of the ‘attack’ phase was similar to that in Cyprus without a dog euthanasia programme. In this programme, farm owners assumed responsibility for regular dog treatments. While transmission of CE to humans has ceased, it is not yet known whether or not elimination has been achieved (Craig and Larrieu 2006).

In Chile, programmes with Option 5 were introduced in Region 12 in 1979 and Region 11 in 1982. The results obtained are illustrated in Fig. 53.5 and show clearly that this option is being successfully pursued (Anon. 1994) and should lead to the transformation from the costly ‘attack’ programme to a permanent ‘consolidation’ phase within 20 years.

In Argentina, control programmes using Option 5 have been funded by provincial ministries of health (Gemmell and Varela Diaz 1973). Few have been sustained, but three have demonstrated that control is feasible. For example, in a pilot trial in the Province of Neuquén in 1970, 28% of the dogs, 71.4% of sheep, and 90% of cattle harboured E. granulosus, but 17 years later when the trial was terminated, these prevalences were 2.1%, 6.9%, and 13.8% respectively, and transmission of CE to children had almost ceased (Kaczorkiewicz 1988). Similar programmes have been undertaken in the Provinces of Rio Negro (Craig and Larrieu 2006) with evidence of a reduction in prevalence in humans or animals or both.

Comparing the results obtained from programmes using Options 2 with 3, it seems that little progress is likely to be made in echinococcosis control simply by introducing legislation directing owners to treat their dogs and preventing them from gaining access to raw offal, even if a substantial educational programme is involved. From the results of the island programmes using Option 3, it is clear that control and even elimination is not only feasible, but also rapidly benefits the whole community. The important contributing factors include:

Comparing results obtained from programmes selecting Options 3, 4, and 5, it is clear that the costly ‘attack’ phase may be shorter with options 4 and 5 than with 3, provided that this phase can be transformed to the ‘Consolidation’ phase. It seems that this can only readily be achieved if effective survey and surveillance policies and animal health movement legislation can be applied. In their absence, there may be a trend to relax the dosing programme with unpredictable results. It seems that the most rapid progress can best be achieved by using animal health rather than human health administrations. Advantages also include:

In conclusion, control is expensive and long-term and the ‘planning’ phase is now regarded as essential if the legislature is to be expected to support a programme financially. Modelling provides a criterion for deciding if a control programme can succeed in eliminating the parasite and underpins a benefit–cost analysis of control options (Lawson 1994; Torgerson 2003a, b; Budke et al. 2005c). Several indicators of economic performance are available, including the net present value, the internal rate of return, but the most usual method is the benefit–cost ratio including the cost per DALY averted. This is calculated as the total discounted benefit over a fixed time period. If the benefit–cost ratio is greater than unity and/or the cost per DALY averted is low, then the benefits of a control action outweigh its costs and the programme can be recommended to the legislature as efficient from an economic point of view, with a relatively high priority as control has now been shown to be feasible.