Parasitol Res (2004) 94: 275–282 DOI 10.1007/s00436-004-1197-2

O R I GI N A L P A P E R

Hermann Feldmeier Æ Margit Eisele Æ Eric Van Marck Heinz Mehlhorn Æ Ronaldo Ribeiro Æ Jo¨rg Heukelbach

Investigations on the biology, epidemiology, pathology and control of Tunga penetrans in Brazil: IV. Clinical and histopathology

Received: 19 June 2004 / Accepted: 19 July 2004 / Published online: 11 September 2004
Springer-Verlag 2004

Abstract Tungiasis is a parasitic skin disease caused by the sand flea Tunga penetrans. This ectoparasitosis is endemic in South America, the Caribbean and subSaharan Africa, where it is an important but neglected health problem in resource-poor communities. As part of a study on tungiasis-related morbidity in a typical slum in Fortaleza, Brazil, we identified 86 individuals with tungiasis. Lesions were counted, classified according to the stage of development, and clinical pathology was documented. One hundred and nine lesions were biopsied and examined by histological sectioning. The patients had between 1 and 145 lesions (median 14.5), the majority occurring in clusters. In all, 77% of patients reported severe pain at the site of the lesion, and 52% had one or more nails lost or severely deformed. Intense inflammation and/or fissures hindered 45% of the patients from walking normally. Signs of superinfection were observed in 29%, and signs of generalised inflammation in 2% of patients. Clinical pathology was H. Feldmeier (&) Æ M. Eisele Department of Medical Microbiology and Immunology of Infection, Institute for Infection Medicine, Charite´ University Medicine Berlin, Campus Benjamin Franklin, Hindenburgdamm 27, 12203 Berlin, Germany E-mail: [email protected] Fax: +49-30-84453830 E. Van Marck Department of Pathology, University of Antwerp, 2650 Antwerp, Belgium H. Mehlhorn Department of Zoology and Parasitology, Heinrich-Heine-University, 40225 Du¨sseldorf, Germany R. Ribeiro Department of Pharmacology and Physiology, Faculty of Medicine, Federal University of Ceara´, CE-60931-140 Fortaleza, Brazil J. Heukelbach Department of Community Health, Faculty of Medicine, Federal University of Ceara´ and Mandacaru Foundation, CE-60931-140 Fortaleza, Brazil

significantly related to the number of lesions, and the total number of parasites present correlated with the number of fleas occurring in clusters. Clinical pathology was frequently accompanied by a pathological alteration of the epidermis (predominantly hyperplasia, parakeratosis, hyperkeratosis, and spongiosis) and the dermis. Tungiasis causes a broad spectrum of clinical and histopathological alterations, and is a serious health threat in a typical, impoverished community in northeast Brazil. The clinical pathology is closely related to the parasite burden of an individual and the clustering of embedded fleas at certain predilection sites.

Introduction Tungiasis, caused by the sand flea, Tunga penetrans (jigger), is probably one of the oldest plagues of the American continent. It is the only metazoan parasite which emerged in the Americas, and from there spread to other parts of the world (Henning 1904; Heukelbach et al. 2001). Since the 19th century, the sand flea is endemic in many parts of sub-Saharan Africa (Hoeppli 1963; Heukelbach et al. 2001). Records from colonial times indicate that the ectoparasitosis caused considerable harm in native populations and expatriates alike (Gordon 1941; Bruce et al. 1942). Today, tungiasis is a common but neglected health problem in many impoverished communities in Africa, South America and the Caribbean (Arene 1984; Nte and Eke 1995; Heukelbach et al. 2001; Wilcke et al. 2002; Carvalho et al. 2003, Muehlen et al. 2003). The disease has also been reported in some parts of Mexico (Andrade 1952; Ibanez-Bernal and Velasco-Castrejon 1996; Macias and Sashida 2000). It occurred on the Bermuda Islands, but disappeared from the archipelago for reasons unknown (Balfour 1928). It has been estimated that in Brazil alone more than 106 individuals are at risk for severe tungiasis (Heukelbach et al. 2001). In the typical, poor neighbourhoods,

276

the prevalence may be 50% and more in the general population (Muehlen et al. 2003; Carvalho et al. 2003). In view of the wide distribution of tungiasis and its relevance for people living in poverty (Heukelbach et al. 2002a, 2002b), it is surprising that the clinical features and histopathology have never been described systematically. In fact, so far, knowledge of pathology is mainly derived from the examination of travellers returning from endemic areas. However, such patients are atypical as they present only one or two lesions, and show no or only insignificant clinical pathology (Franck et al. 2003). The observations made in travellers are in contrast to case reports from Brazil, in which patients were described who were almost completely covered with embedded fleas, with extremely severe clinical pathology (Valenca et al. 1972; Cardoso 1981; Bezerra 1994; Feldmeier et al. 2003). As part of a larger research project aimed at assessing the public health importance of parasitic skin diseases in impoverished communities in northeastern Brazil, a study was undertaken to determine the pathology related to T. penetrans in patients presenting at a primary health care centre (PHCC). The results indicate that in a typical slum significant clinical pathology can be expected, caused by the combination of a high parasite burden and the occurrence of lesions in clusters at certain predilection sites.

Materials and methods Study area The favela Morro de Sandra’s is a typical shantytown at the outskirts of Fortaleza, the capital of Ceara´ State, northeastern Brazil. During the high transmission season (July to December), roughly one third of the population is affected by tungiasis (Wilcke et al. 2002). The slum is built on a dune close to the beach and has a total population of about 1,500 inhabitants, mostly recent immigrants from the arid, rural hinterland. Sixty percent of the population has a monthly family income of less than two minimum wages (one minimum wage=R$ 240, equivalent to US$ 80). Adult illiteracy is 30%, unemployment rates are high, and violence of all types is common. Ninety-seven percent of the households have electricity, and about 60% have running water (Family Health Program 1999). Many houses are made of improvised construction material and do not have concrete floors. Waste collection and sewage disposal are insufficient, and hygienic conditions are precarious. Most of the streets are unpaved. Thus, in many aspects, Morro de Sandra’s is a typical South American slum. Study population The study was performed at the PHCC serving the population of several shantytowns, including the favela

Morro de Sandra’s. During a period of 8 weeks, all patients from Morro de Sandra’s who went to the health post for any medical reason were eligible for the study. During this period, a total of 86 patients with tungiasis were identified. Clinical examination As tungiasis may occur at any topographic site (Heukelbach et al. 2002b), the whole body surface of the patient was examined for the presence of immature, eggproducing, dead or manipulated fleas. Lesions were classified according to the Fortaleza Classification, a recently elaborated staging system (Eisele et al. 2003). The following findings were considered diagnostic for tungiasis: – flea in ‘‘statu penetrandi’’ (stage I) – a dark and itching spot in the epidermis with a diameter of 1–2 mm, with or without local pain and itching (early lesion, stage II) – lesions presenting as a white halo with a diameter of 3–10 mm, with a central black dot (mature egg-producing flea, stage III) – a brownish-black circular crust with or without surrounding necrosis of the epidermis (dead parasite, stage IV) – circular ‘‘punched-out’’ residue in the keratin layer of the sole of the foot or irregular thickening of the nail rim (stage V). Lesions altered through manipulation by the patient (such as partly or totally eliminated fleas leaving a characteristic, crater-like sore in the skin), and suppurative lesions caused by the use of non-sterile perforating instruments such as needles and thorns were documented as well. The localisation and the number of lesions were noted. Clinical pathology was assessed as follows: – Acute local inflammation: pruritic lesion surrounded by erythema, with or without pain or oedema. – Chronic local inflammation: tenderness, shiny skin with or without desquamation or fissures. – Generalised inflammation: oedematous deformation of at least one toe or finger (defined as the deviation of the normal axis of the toe and finger), or oedema of entire foot or hand. – Superinfection: presence of pustules, suppuration or ulcers. – Physical disability/sequels: difficulty in walking or gripping (if lesions were located on the hands) based on the patient’s statement that pain restricted her/his movements; deformation of the nail; loss of entire nail. As lesions tended to occur in clusters, a cluster was defined arbitrarily as a group of at least five lesions which occurred in close proximity, e.g. at the periungual region of a toe, the heel or the fingertip.

277

In order to assess the degree of clinical pathology, a ‘‘pathology score’’ was defined (max. three points): acute local or chronic local inflammation with no signs of superinfection, disability or sequel=one point; acute local or chronic local inflammation, together with at least one type of disability or sequel=two points; acute local or chronic local inflammation, together with at least one type of disability or sequel and at least one sign of superinfection or presence of general inflammation=three points. Biopsies Local anaesthesia was achieved with lidocaine 2%. Then, lesions were excised under sterile conditions. Biopsies were transferred to 10% formaldehyde. A sterile dressing with antibiotic ointment was applied and changed daily. To prevent secondary infection, all patients received a pair of tennis shoes (many inhabitants of the area do not have shoes and walk barefoot or wear flip-flops). Histopathology Formalin-fixed skin biopsy specimens were routinely embedded in paraffin, sectioned at 5 lm and stained with haematoxylin-eosin (HE) and, if necessary, with Trichrome-Masson’s stain, PAS, Perls iron stain, or Giemsa stain. A total of 109 biopsies were evaluated histopathologically. Statistical analysis Statistical analysis was performed using the StatView software package version 1.5. As the distribution of data was not normal, and because standard deviations varied considerably, nonparametric tests were used. The Wilcoxon signed rank test and the Spearman rank correlation coefficient tests were applied where appropriate. The significant of differences of proportions was determined by Fisher’s exact test. Ethical considerations The study was conducted in accordance with good clinical practice and the Declaration of Helsinki of 1975 (as revised in 1983), and has been approved by the Ethics Committee of the Federal University of Ceara´ State, Fortaleza (Brazil). Prior to the study, meetings with community health workers, community leaders and staff members of the PHCC were held and the objectives explained. Informed written consent was obtained from each patient after explaining the objectives of the study. In the case of minors, the guardians were asked for their consent. After the examination, lesions were treated topically with thiabendazole 5%, the standard treatment for the ectoparasitosis in a primary health care setting.

Table 1 Demographic and parasitological characteristics of the patients Variable

Results

Age in years: median (range) Number of females/males Age females: median (range) Age males: median (range) Number of lesions per patient: median (range) 0–9 years 10–19 years 20–49 years ‡50 years Total number of lesions Total number of viable lesionsa Total number of dead lesions/residuesb Total number of manipulated lesions Number of lesions per cluster: median (range) Total number of lesions occurring in clusters

11 (1–67) 32/54 9.5 (1–48) 13 (3–67) 14.5 (1–145) 15 (1–96) 23 (2–145) 4 (1–128) 27 (6–138) 2,484 912 (36.7%) 1,317 (53.0%) 255 (10.3%) 16 (5–127) 1,380 (55.6%)

a

Stage: I–III (see Materials and methods) Stage: IV–V (see Materials and methods)

b

In the case of superinfection, a dressing with neomycin ointment was applied. Patients and guardians were encouraged not to walk barefoot.

Results Clinical findings The demographic and parasitological characteristics of the patients are shown in Table 1. The parasite burden varied with age, and was highest in the age groups 10–19 years and in the ‡50-years age group. The lowest parasite burden occurred in adults 20–49 years old (median=four lesions; P=0.003 compared to the age group 0–9 years; P=0.004 compared to age group 10–19 years; P=0.02 compared to the age group ‡50 years). There was a difference in the parasite burden between males and females (median 21 versus 11 lesions). However, this was not statistically significant (P=0.4). There was a highly significant correlation between the number of lesions in individual patients and the number of lesions occurring in clusters (rho=0.84; P<0.0001). The topographic distribution of lesions is summarised in Table 2. In all, 2,414 out of 2,484 (97.2%) lesions ocTable 2 Distribution of lesions according to their topographic site (n=2,484) Topographic site

Toe Periungual Other areasa Sole of the foot Heel Dorsum pedis Hand/gluteal region a

Parasite load: median (range)

Total number of lesions at this site

%

6.5 (0–45) 4 (0–74) 1 (0–48) 0 (0–54) 0 (0–4) 0 (0–21)

888 705 417 385 19 70

35.7 28.4 16.8 15.5 0.8 2.8

Such as the tip, the plantar or the dorsal side of the toe

278 Table 3 Clinical findings in tungiasis patients Clinical pathology observed

n(%)

Itching Pain Local inflammation Erythema Edema Generalised inflammation Superinfection Pustules Suppuration Ulcers Fissures Disability/sequels Deformation and/or loss of naila Difficult walking Difficult grippingb

86 (100) 65 (76.6) 61 (70.9) (64.0) (26.7) 2 (2.3) 25 (29.1) 9 (10.5) 19 (22.1) 20 (23.3) 10 (11.6) 65 (75.6) 45 (52.3) 39 (45.3) 5 (23.8)

a

At least one toe or finger affected Of 21 patients with lesions at the hands

b

Fig. 1 A Right heel of a 12-year-old girl with a cluster of lesions in different stages of development. B Fissures on the plantar side on the metacarpophalangeal joint of the foot of an 8-year-old boy. C Loss of a toenail after severe superinfection of lesions on the fifth toe of a 2-year-old girl. D Multiple lesions on the hands of a 9-yearold boy

curred at the feet, 1,593 of these (64.1% of all lesions) were located at the toes. Only one patient had no lesions at the toes. Other areas of the feet, such as the sole and the heel, were also frequently affected. Lesions located at ectopic sites (hand, gluteal area) represented only 2.8%. However, 21 patients (24.4%) had at least one lesion at the hand, usually at the fingertip. The clinical pathology is depicted in Table 3. Intense pain and itching were perceived as the most irritating symptoms, and prevented patients from sleeping soundly. Disability and/or sequels were observed in 76% of the patients. The typical clinical pathology is illustrated in Fig. 1. When the pathology score was applied to assess the degree of clinical pathology semi-quantitatively, it became clear that the number of embedded fleas significantly correlated with the degree of pathology (rho=0.36, P<0.01; Table 4). The number of lesions was particularly high in patients with signs of generalised inflammation (median 131 lesions), ulcers (median 31 lesions) and difficulty in walking (median 12 lesions).

Histopathological findings Of the 109 biopsies examined, 33 (30.3%) showed a normal epidermis and 76 (69.7%) demonstrated histo-

279 Table 4 Relationship between the number of lesions present and the degree of pathology observed Pathology score

Number of patients

Number of lesions: median

1 2 3

31 29 26

9 14 30

epidermal hyperplasia developed, surrounding the area where the flea had been embedded.

Discussion

logical abnormalities (Table 5). Hyperplasia, parakeratosis, hyperkeratosis and, less often, spongiosis were the most common pathological alterations of the texture of the epidermis (Fig. 2A). In two cases, a pseudoepitheliomatous hyperplasia was observed. Vascularisation of the keratin layer was seen in four biopsies, and occurred together with an inflammatory infiltrate. Neutrophils were the predominant cell type in these infiltrates. They tended to concentrate beside the cuticle of the parasite, and eventually led to the formation of micro-abscesses (Fig. 2B). After the parasite had died, its carcass became filled with neutrophils which later transformed to pus (Fig. 2C). Sixty of the 109 biopsies were sufficiently deep to allow examination of the dermis below the embedded parasites. Sectioning showed mild to severe inflammatory infiltrates with or without the presence of dilated blood vessels. These infiltrates were composed in decreasing order of frequency by lymphocytes (mainly plasma cells), eosinophils, neutrophils, mast cells and histiocytes. Dilated blood vessels and extravasation of erythrocytes were observed mainly in the vicinity of the proboscis of the flea, penetrating through the basal membrane of the epidermis (Fig. 2D). Once the dead parasite was sloughed from the epidermis, a circular Table 5 Histopathological findings in tungiasis patients Type Epidermis (109 biopsies) No alteration Abnormal histological texture Hyperplasia Parakeratosis Hyperkeratosis Spongiosis Pseudoepitheliomatous hyperplasia Indicators of inflammation and/or superinfection Inflammatory infiltrateb Micro-abscess, pus Dermis (57 biopsies) No alteration Abnormal histological texture Inflammatory infiltratec Vasodilation/neovascularisation a

n

(%)

33 76 43 34 16 7 2

30.3 69.7 56.6a 44.7a 21.1a 9.2a 2.6a

26 28

34.2a 36.8a

3 57 57 4

5.0 95.0 100d 7.0d

Of epidermal sections with abnormal findings Frequency of cells within infiltrate in decreasing order: neutrophils, eosinophils, plasma cells, giant cells c Frequency of cells within infiltrate: lymphocytes (mainly plasma cells)>eosinophils>neutrophils>mast cells>histiocytes d Of dermal sections with abnormal findings b

With a length of less than 1 mm, Tunga penetrans is the smallest flea species known (Connor 1976). True to its name, the female burrows into the epidermis of its host. There it remains for a period of up to 5 weeks, during which it matures, produces and releases eggs, and eventually dies (Eisele et al. 2003). Microbiologically, the embedded flea behaves as a foreign body with a continuously enlarging surface (Feldmeier et al. 2002). The protruding posterior end produces a sore through which the surface of the skin is connected to the epidermis and—through the flea’s proboscis—also to the dermis. Translocation of the normal micro-flora is expected to occur when the skin’s surface is linked with underlying tissue compartments by means of a foreign body (Feldmeier et al. 2002). Therefore, although the infection with T. penetrans is a self-limiting process, the risk for infectious complications is obvious. The group of patients with tungiasis we have identified at a PHCC serving the population of a favela can be considered representative for this community. They sought care at the PHCC for reasons other than tungiasis, and therefore a selection bias towards patients with severe pathology seems to be unlikely. The median number of embedded fleas was high (14.5), and 90% of the lesions had been left untreated. A remarkable finding in these patients was that the majority of lesions occurred in clusters (Fig. 1A), with a highly significant correlation between the total number of embedded fleas and those located in clusters. The topographic distribution of lesions confirmed previous findings of aggregation of embedded fleas at few anatomic sites (Hoeppli 1963). The aggregation of lesions was particularly high at the periungual area of the toes, where 36% of all ectoparasites were located. Consequently, it is not surprising that nail deformation or loss of nails, complications known to be sequels of severe tungiasis (Feldmeier et al. 2003), were observed in more than half of the patients. Signs of superinfection of the lesions were present in about one third of the patients, and corresponded to the presence of microabscesses or pus around or within the exoskeleton of the parasite in histological sections (Table 5, Fig. 2). These findings confirm the results of a bacteriological study showing pathogenic aerobic or anaerobic micro-organisms in almost all tungiasis lesions (Feldmeier et al. 2002). The presence of ulcers at sites where fleas were previously embedded was a common finding and was associated with high parasite load. In the pre-antibiotic era, ulcers seem to have been very common and were frequently of the phagodenic type (Decle 1900; Joyeux and Sice 1937).

280

Fig. 2 A Tangential cut through a fully developed, gravid flea embedded in the stratum corneum of the epidermis. The flea’s head and thorax are enfolded in the hypertrophic anterior abdominal segments. The epidermis is hyperplastic and shows papillomatosis, parakeratosis and hyperkeratosis. B Tangential cut through the posterior abdominal segments of an embedded sand flea. Next to the chitinous cuticle, a micro-abscess has formed. C Dead parasite; the exoskeleton of posterior abdominal segment has remained intact; the cuticle has disintegrated at the epidermal–dermal interface. The carcass is infiltrated by neutrophils, and pus has formed. D The head of the flea is located at the epidermal–dermal interface, has penetrated the basal membrane, and is surrounded by many erythrocytes, presumably having leaked from a blood vessel. The abdomen of the parasite is separated from host tissue by a thick, chitinous cuticle

Fissures as a sequel of prolonged inflammation due to tungiasis have not yet been described. The fissures observed in our patients were located in the interdigital area or at the plantar side of the toes, usually under the metacarpophalangeal joint (Fig. 1B). We assume that the development of fissures at these sites of intense

movements is linked to the extraordinary hyperkeratosis of the epidermis which surrounds embedded fleas. Similar to ulcers, fissures mainly occurred in patients with a high parasite load. Difficult walking was suffered by almost half of our patients. In colonial British East Africa, physical disability due to tungiasis was also very common. Decle (1900) observed in a garrison in Kenya that in 102 out of 230 (44%) local Askari soldiers, walking was so impaired that they were considered unfit for military service. A decade before, a vivid description was given by Waterton (1973): ‘‘A knowing eye may always perceive when the feet of the negroes are the abode of the Chigoe. They dare not place their feet firmly on the ground, on account of the pain which such a position would give them; but they hobble along with their toes turned up.’’ So far, histopathological findings are based on single observations made in travellers (Smith and Procop 2002; Franck et al. 2003). As the number of embedded fleas

281

was usually one, and signs of inflammation were absent or weak, this may explain why in these patients histological sectioning rarely yielded abnormal results. In our study, a pathological texture of the epidermis was observed in 70% of the biopsies, and inflammatory infiltrates in the dermis were observed in virtually all biopsies. There was a clear correlation between macroscopic findings and microscopic alterations. For example, the presence of oedema coincided with spongiosis, and that of fissures with hyperkeratosis. Itching was frequently associated with parakeratosis, and incontentia pigmenti—an indicator of frequent scratching—and shiny skin coincided with remodelling of the dermal stroma (data not shown). We have recently discussed the various mechanisms by which embedded fleas may induce pathological alterations of the skin in the early stage of flea development (Feldmeier et al. 2003). Supposedly, the intense inflammation around older lesions, particularly around clusters, is due to a combination of tissue damage and bacterial superinfection. As immediately after penetration the lesion becomes itchy, repetitive scratching should favour the entry of bacteria through the sore existing in the epidermis. Bacterial superinfection also results from inappropriate manipulation of lesions with non-sterile instruments by the patient or his/her guardian. The formation of pustules and suppuration (Fig. 2C, D), particularly when the fleas had died, was associated with an intense infiltration of the lesions by leukocytes. When fleas were involuting or dead (stage IV), histological sections showed that the outer surface of the parasites became densely covered with neutrophils, eosinophils and lymphocytes, which eventually invaded the carcass of the flea (Fig. 2C). The remarkable desquamation of the skin observed around such late-stage lesions had its histopathological correlates in hyperkeratosis and parakeratosis of the stratum corneum. It is conceivable that two major factors determine the degree of inflammation and subsequent pathology: the number of newly embedding fleas per unit of time, and the aggregation of lesions in clusters. This is corroborated by the observation that clinical pathology was particularly severe at topographic sites bearing the highest number of lesions, such as the toes, the heel and the sole of the foot, and that the pathology score, a semiquantitative measure of clinical pathology, was significantly related to the parasite load (Table 4). These findings explain why in Western travellers tungiasis is associated with only insignificant clinical pathology—usually tourists harbour only few sand fleas, and the ectoparasites are usually taken out during an early stage of development (Franck et al. 2003). It is not known why sand fleas tend to aggregate on human skin. We suggest that cluster formation reflects an evolutionary advantage for the propagation of the ectoparasite. The oedema surrounding an existing lesion presumably facilitates the penetration process. In addition, as

embedded females are thought to become fertilised in situ by males exploring the skin (Geigy and Suter 1960), a male looking for female partners would have a better chance of fertilising several females within a short period of time—faecal material is released from embedded female fleas and spread in the dermal papillae, thus attracting males to a site where females have penetrated (Eisele et al. 2003). The question arises why our patients allowed lesions to develop, rather than extracting fleas as soon as they penetrated. First of all, taking out an embedded flea with a sharp instrument is painful by itself and requires skills, a steady hand and—in the early stage, when the lesion measures a few millimetres—a good eyesight as well as good light. Consequently, this cannot be done by children, the elderly, physically handicapped, or after sundown. It also needs an experienced and patient caretaker who is not over-committed to other daily routines. We have recently shown that in a similar environment, a caretaker would need about 30 min/day to take out all newly embedded fleas from the average 3.5 children of a household (Heukelbach et al. 2004). Secondly, help from another person is required when lesions are located at topographic sites difficult to reach, such as the sole of the foot or the heel. Thirdly, when fleas are taken out by the non-sterile instruments typically used, i.e. needles, razor blades or thorns, superinfection is very likely, causing even more harm. Tungiasis is the epitome of a parasitic skin disease which has been completely neglected by both the scientific community and professional health care providers. This is reflected by the fact that 500 years after the first publication, the epidemiology of the ectoparasitosis is still not understood, means for treatment and prevention are not at hand, and that the spectrum of clinical histopathology is described here for the first time. Our data convincingly demonstrate that tungiasis is associated with significant pathology in a representative group of patients from the endemic area. In an impoverished population, like in our study area, tungiasis is clearly not merely a nuisance but an important disease which merits attention on the individual as well on the public health level. Acknowledgments This study was supported in part by the A¨rztekomittee fu¨r die Dritte Welt, Frankfurt (Germany), the World Health Organization, Geneva (Switzerland), the DAAD/CAPES PROBRAL program (Germany/Brazil), and donations from the following individuals: Michael Peitz, Hamburg, Gitta and Christian Hertz, Reinbek, and Kimio Powils-Okano, Wiesbaden, all Germany. We are indebted to the ‘‘Associac¸a˜o dos Moradores do Morroda Sandra’s’’ for supporting the study, and we are grateful to Vania Santos de Andrade and Walter Antoˆnio da Silva for skilful assistance. The data are part of a medical thesis by M.E. The secretarial assistance of Michi Feldmeier is gratefully acknowledged.

References Andrade RN (1952) Tungiasis. Gac Med Mex 82:11–15 Arene FO (1984) The prevalence of sand flea (Tunga penetrans) among primary and postprimary school pupils in Choba area of the Niger Delta. Public Health 98:282–283

282 Balfour A (1928) Health lessons from Bermuda. Br Med J 1:447– 448 Bezerra SM (1994) Tungiasis—an unusual case of severe infestation. Int J Dermatol 33:725 Bruce CO, Knigin TD, Yolles SF (1942) A discussion of the chigoe (Tunga penetrans) based on experiences in British Guiana. Milit Surg 82:446–452 Cardoso A (1981) Generalized tungiasis treated with thiabendazole. Arch Dermatol 117:127 Carvalho RW, Almeida AB, Barbosa-Silva SC, Amorim M, Ribeiro PC, Serra-Freire NM (2003) The patterns of tungiasis in Araruama township, state of Rio de Janeiro, Brazil. Mem Inst Oswaldo Cruz 98:31–36 Connor DH (1976) Diseases caused by arthropods—tungiasis. In: Binford CH, Connor DH (eds) Pathology of tropical and extraordinary diseases. Armed Forces Institute of Pathology, Washington, DC, pp 610–614 Decle L (1900) Three years in savage Africa, new edn. Methuen, London Eisele M, Heukelbach J, van Marck E, Mehlhorn H, Meckes O, Franck S et al. (2003) Investigations on the biology, epidemiology, pathology and control of Tunga penetrans in Brazil: I. Natural history of tungiasis in man. Parasitol Res 90:87–99 Family Health Program (1999) UBASF Aı´ da Santos e Silva. Relato´rio de territorializac¸a˜o. Municipal Health Council of Fortaleza, Fortaleza Feldmeier H, Heukelbach J, Eisele M, Sousa AQ, Barbosa LM, Carvalho CB (2002) Bacterial superinfection in human tungiasis. Trop Med Int Health 7:559–564 Feldmeier H, Eisele M, Sabo´ia Moura RC, Heukelbach J (2003) Severe tungiasis in underprivileged communities: case series from Brazil. Emerg Infect Dis 9:949–955 Franck S, Feldmeier H, Heukelbach J (2003) Tungiasis: more than an exotic nuisance. Travel Med Infect Dis 1:159–166 Geigy R, Suter P (1960) Zur Copulation der Flo¨he. Rev Suisse Zool 67:206–210 Gordon RM (1941) The jigger flea. Lancet 2:47–49 Henning G (1904) Zur Geschichte des Sandflohs (Sarcopsylla penetrans L.) in Afrika. Naturwiss Wochenschr 20:310–312 Heukelbach J, de Oliveira FA, Hesse G, Feldmeier H (2001) Tungiasis: a neglected health problem of poor communities. Trop Med Int Health 6:267–272

Heukelbach J, Mencke N, Feldmeier H (2002a) Cutaneous larva migrans and tungiasis: the challenge to control zoonotic ectoparasitoses associated with poverty. Trop Med Int Health 7:907–910 Heukelbach J, Wilcke T, Eisele M, Feldmeier H (2002b) Ectopic localization of tungiasis. Am J Trop Med Hyg 67:214–216 Heukelbach J, Franck S, Feldmeier H (2004) High attack rate of Tunga penetrans infestation in an impoverished Brazilian community. Trans R Soc Trop Med Hyg 98:431–434 Hoeppli R (1963) Early references to the occurrence of Tunga penetrans in tropical Africa. Acta Trop 20:143–152 Ibanez-Bernal S, Velasco-Castrejon O (1996) New records of human tungiasis in Mexico (Siphonaptera: Tungidae). J Med Entomol 33:988–989 Joyeux C, Sice´ A (1937) Pre´cis de me´decine coloniale, 2nd edn. Masson, Paris Macias PC, Sashida PM (2000) Cutaneous infestation by Tunga penetrans. Int J Dermatol 39:296–298 Muehlen M, Heukelbach J, Wilcke T, Winter B, Mehlhorn H, Feldmeier H (2003) Investigations on the biology, epidemiology, pathology and control of Tunga penetrans in Brazil II. Prevalence, parasite load and topographic distribution of lesions in the population of a traditional fishing village. Parasitol Res 90:449–455 Nte AR, Eke FU (1995) Jigger infestation in children in a rural area of Rivers State of Nigeria. West Afr J Med 14:56–58 Smith MD, Procop GW (2002) Typical histologic features of Tunga penetrans in skin biopsies. Arch Pathol Lab Med 126:714–716 Valenc¸a ZO, Cardoso AEC, Cardoso AS (1972) Tunguiase generalizada: relato de dois casos tratados com thiabendazol. Dermatol Ibero-Latino-Am 3:375–378 Waterton C (1973) Wanderings in South America, the north-west of the United States and the Antilles, in the years 1812, 1816, 1820 and 1824 with original instruction for the perfect preservation of birds and for cabinets of natural history. Oxford University Press, Oxford, pp 108–109 Wilcke T, Heukelbach J, Cesar Saboia MR, Regina SK-P, Feldmeier H (2002) High prevalence of tungiasis in a poor neighbourhood in Fortaleza, Northeast Brazil. Acta Trop 83:255–258