Chapter l of the thesis gives a brier introduction to lymphatic filariasis, its discovery, the distribution of the infection, the associated symptoms, and recommended treatment. The parasites causing lymphatic filariasis (Wuchereria bancrofti, Brugia malayi and B. timori) and their lire cycle were discovered during the latter half of the 18th century. It is widely distributed throughout the tropical world, with an estimated one billion people at risk of infection and 120 million people being infected or having disease manifestations. The most common clinical manifestations include acute episodes of adenolymphangitis (ADL), and chronic manifestations of hydrocoele, lymphoedema of arm, leg, breast or genitals, chyluria and tropical pulmonary eosinophilia. The chapter also contains a short introduction to the research area in general and to the endemic situation of lymphatic filariasis in particular.

Chapter 2 describes a longitudinal surveillance undertaken in Magoda and Mpapayu villages (approximately 2,400 inhabitants) with daily monitoring for two years for patients with episodes of ADL. In the first year of surveillance, the incidence of ADL was 127 per 1,000 individuals, with 7.8% of the population being affected. Episodes were more frequent in adults with chronic lymphoedema (odds ratio 12.5) and hydrocoele (odds ratio 2.4) than in adults without chronic lymphoedema, with no correlation to sex or microfilaraemia. After one year, mass treatment with a monthly low doge of the antifilarial drug diethylcarbamazine (DEC) was given for one year, and the surveillance was continued for one more year. In the intervention year the incidence was reduced by 40% overall, but by 69% from the first year to the second half of the second year, implying a delayed effect of treatment. Distribution of insecticide treated mosquito nets in the latter half of the intervention year in one village showed a significant added effect of reducing ADL incidence compared to the village without mosquito nets.

Microfilaraemia prevalence data for the two villages were obtained before starting surveillance, before starting intervention, and after surveillance had ended. DEC mass treatment resulted in a 46% reduction of mf prevalence (from 24% to 13%), and a 77% reduction of geometric mean intensity (all individuals).

The results suggest that transmission of lymphatic filariasis is of major significance for the incidence of ADL, and that this effect is more important in individuals without lymphoedema than individuals with lymphoedema, since the frequency of episodes in the latter group was not reduced to the same extent after treatment. The results also suggest a potential for community mass treatment with DEC to reduce the incidence of ADL, an effect which in the context of mass treatment as part of a lymphatic filariasis elimination strategy is a more immediate and tangible benefit to a larger part of the population than the expected more long term effect of reducing the population prevalence of chronic manifestations.

It has been proposed that bacterial infections play a major role in precipitating ADL episodes and in the development and progression of chronic lymphoedema in lymphatic filariasis-endemic areas. Chapter 3 reports the examination of individuals with ADL, chronic lymphoedema, abscesses and endemic controls for presence of bacteria in cultures from skin biopsies, skin swabs, blood, lymph node aspirates, ulcers and abscess material. Bacterial growth was significantly more 1ffion in blood from ADL patients with lymphoedema than in patients without lymphoedema, and in skin swabs from chronic lymphoedema patients compared to endemic controls. Pathogenic bacteria were found to occur significantly more often than commensal species in skin biopsies from lymphoedema patients with ADL n in lymphoedema patients without ADL, and in skin swabs from lymphoedema patients compared to endemic controls, but there was no significant difference between non-lymphoedematous ADL patients and endemic controls. Antibody titres against the streptococcal antigens AST, ASH and ASDB were significantly higher in individuals with chronic lymphoedema compared to individuals without lymphoedema, and showed a significant reduction in chronic lymphoedema patients after three months of intense hygiene, but there was no difference between )L patients without chronic manifestations and endemic controls.

‘Interpretation of these results suggests that secondary bacterial infection may ay a role in the pathogenesis of ADL, especially in patients with chronic lymphoedema, but that other factors also are significant, particularly in individuals without pre-existing clinical disease manifestations.

Chapter 4 also addresses the pathogenesis of ADL. Forty-seven skin punch biopsies from the legs of individuals from a W. bancrofti endemic area (five biopsies from individuals with ADL but no chronic lymphoedema, 15 from lymphoedema patients with ADL, 17 from lymphoedema patients without ADL, ld 10 from endemic control s) were examined by histopathology and immunohistochemistry. Seven lymphoedema patients were biopsied before and after three months of a treatment that included washing and hygiene. All biopsies from patients with lymphoedema showed epidermal hyperplasia, hyperkeratosis, chantosis, papil1omatosis, and dermal fibrosis and oedema. The majority of biopsies from individuals with acute and/or chronic manifestations showed only mild inflammation of a predominantly chronic type, and neutrophilic leukocytes were not a prominent finding. Inflammatory cells were mainly T -lymphocytes, macrophages and plasma cells. Endemic control s all displayed a mild, chronic inflammation.

Inflammation score was significantly higher in patients with clinical manifestations (acute or chronic) than in controls. There was no significant change n the inflammation score of lymphoedema patients before and after treatment, or between paired samples taken in acute and chronic phase, respectively. No bacteria, fungi or parasites were seen in any of the samples. The results indicate that factors other than bacterial infections are important for the development of ADL. These factors need to be identified and studied further to optimise the treatment for acute and chronic lymphoedema and the prevention of ADL.

Chapter 5 addresses the treatment of chronic filarial lymphoedema. The WHO strategy for elimination of lymphatic filariasis recommends hygienic measures for lymphoedema, elephantiasis and ADL, but there is a lack of data on the effectiveness of this approach.

In Teule Designated District Hospital in Muheza a lymphoedema treatment clinic was established. Patients were allocated to one of two treatment regimes consisting of either washing and disinfection combined with exercises (133 patients) or decongestive physiotherapy (51 patients), and then followed every three months with leg volume measurements, clinical examination and information on symptom development. Three months after treatment start, the effect of washing on lymphoedema volume was a statistically significant reduction (in percent of pre-treatment leg size), with an average reduction of 0.13 litres (range -1.3 to 2.3 litres), but the effect was not sustained for the subsequent follow-ups. In the physiotherapy group there was a statistically significant volume reduction sustained up to 15 months after treatment start, with hest results coming after six months, showing an average reduction of 1.0 litres (range -0.2 to 9.7 litres). Both treatment regimes showed a statistically significant reduction of hyperkeratosis, cutaneous ulcers, fungal infection and pain. Patients generally expressed a positive attitude towards the effect of treatment, and at the 12 months' follow-up reported decreased pain, itching, cutaneous ulcers and paraesthesia, and reduced frequency and severity of ADL episodes.

Although washing did not show a long-lasting volume-reducing effect in this trial, we endorse hygiene as a base line approach to the treatment of filarial lymphoedema, and recommend decongestive physiotherapy only for severe cases due to high cost and low durability of materials, and high educational requirements for therapists. More research is needed to optimise treatment regimes for filarial lymphoedema patients.

Chapter 6 describes a trial of the effect of DEC on hydrocoele of filarial origin, based on previous community studies suggesting that DEC might reduce the size of a hydrocoele. To investigate this, a double-blind, placebo-controlled trial was carried out in Magoda and Mpapayu villages, where 100 adult male volunteers with chronic hydrocoele of all grades were randomised to DEC 300 mg per day in three divided doses for 12 days, or placebo. Circumferential and ultrasonographic size measurements of the scrotum and a serum sample for filarial antigen were obtained at entry and after three, six and 12 months. A scrotal size index and a hydrocoele fluid volume index was calculated. After three and six months there was a statistically significant reduction of the mean scrotal size index in both the DEC and the placebo group, and after 12 months in the placebo group. Mean hydrocoele fluid volume index remained constant in the DEC group throughout the study and in the placebo group after three and six months, but showed a significant increase in the placebo group after 12 months. There was no difference between the DEC and the placebo group for any volumetric measurement at any of the follows, either overall or if stratified by antigen status or hydrocoele size. DEC significantly reduced geometric mean intensities of circulating W bancrofti antigen in antigen positive individuals compared to placebo.
In conclusion, DEC is not effective for individual treatment of hydrocoele of filarial origin. Other easy and cheap interventions should be sought after, preferentially such which could be employed in the lymphatic filariasis elimination strategy.
 
Chapter 7 attempts to describe the importance of lymphatic filariasis for the development of chronic genital manifestations and for reproductive health in women. A cross-sectional study of Magoda and Mpapayu was conducted including interview and gynaecological examination of adult women, focusing primarily on productive history and genital health. Five hundred and thirty women completed interview, and 404 of these took a gynaecological examination.

The presence of microfilaraemia did not appear to influence fertility or fertility-related variables including age at menarche, parity, or experience of spontaneous abortion, stillbirth, caesarean section, premature labour, or infertility. Mf status was correlated with abnormal menstruation pattern in the 30+ years' age group. Cervical, vaginal and vulval pathology was unrelated to mf status. Two women, 46 d 77 years old, had vulval oedema of possible filarial origin, both were mf negative. Forty-nine percent of women were found to be circumcised (clitoridectomy).

Overall, microfilaraemia appeared to have no relationship with genital disease reproductive health, and chronic manifestations of lymphatic filariasis of the genitals do not appear to be a substantial problem in women.

Chapter 8 reports five cases of ultrasonographic detection of live adult worms hi biting the so-called characteristic 'filarial dance sign' from various anatomical locations, including the spermatic cord, breast and epitrochlear lymph node, and in various clinical circumstances including asymptomatic infection, acute episode and chronic filarial disease. Ultrasonographic detection of adult W bancrofti was first scribed in the spermatic cord in 1994 from Brazil, and has since been reported from Egypt and India.

Ultrasound investigation of adult worms is a valuable tool and has helped expand our knowledge of the lire cycle of human filariases and of the efficacy of various treatment regimens. Further ultrasonographic studies are recommended to increase our understanding of the parasite and the pathogenesis of the different clinical manifestations of lymphatic filariasis.

Chapter 9 gives a brier status and perspectives of filariasis research with special emphasis on pathogenesis and treatment of clinical manifestations.

Involved research institutions
The Ph.D. study was performed at the Panum Institute and Copenhagen University Hospital (Rigshospitalet), linked to Centre for Medical Parasitology.

Supervisor(s)
Ib C. Bygbjerg, Department of International Health, Institute of Public Health, University of Copenhagen
Pascal Magnussen, Institute for Health Research and Development

Correspondence