TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINE AND HYGIENE (1996) 90,69-73 69

Mass diethylcarbamazine chemotherapy for control of bancroftian filariasis: comparative efficacy of standard treatment and two semi-annual single-dose treatments

D. W. Meyrowitschl, P. E. Simonsenl and W. H. Makunde 2 ‘Danish Bilharziasis Laboratory Jaegersborg Alle lo,2920 Charlottenlund, Denmark; ‘Amani Medical Research Centre, l? 0. Box 4, Amani, Tanzania

Abstract The efficacy of 2 strategies for control of bancroftian filariasis using mass diethylcarbamazine (DEC) che- motherapy was evaluated and compared in 2 endemic communities in Tanzania with pre-treatment micro- filarial (mf) prevalences of 285% and 17.7%, and mf geometric mean intensities (GMI) of 588 mf/mL and 251 mf/mL, respectively. All individuals in the first community were offered DEC treatment with 6 mg/kg body weight given daily for 12 d (standard treatment). The second community was offered DEC treatment with 2 single doses of 6 mg/kg body weight given with an interval of 6 months (semi-annual single-dose treatment). Among those who were microfilaraemic before treatment, the mf clearance rates were 51.2% and 36.0%, and the mf GMIs were reduced by 98.6% and 92.2% one year after the start of the standard and the semi-annual regimens, respectively. At community level, the standard strategy and the semi-annual strategy reduced the mf prevalences to 15.1% and 11.6% (reductions of 47.0% and 34.5%) and the mf GMIs to 112 mf/mL and 102 mf/mL (reductions of 81.0% and 59.4%, respectively) one year after start of treat- ment. Both regimens resulted in remarkable improvements in small hydroceles among males presenting this condition before treatment. The lower efficacy of the semi-annual single-dose treatment in relation to the standard treatment in reducing microfilaraemias might be compensated for by continuing semi-annual treatments for a slightly longer period of time. Considering that the semi-annual treatment is easy to ad- minister and more acceptable to the treated individuals, it may in the long run be a more feasible strategy for mass DEC chemotherapy than the standard treatment.

Keywords: Wuchereria bancrojii, filariasis, mass chemotherapy, diethylcarbamazine, comparison of regimens

Introduction Diethylcarbamazine (DEC) is presently the only avail-

able drug for treatment of filariasis due to infection with Wuchereria bancrofi. It is highly effective in reducing mi- crofilaraemia in infected individuals (OTTESEN, 1985; WHO, 1992) and there is indirect evidence for a macro: filaricidal effect (RICHARDS et al.. 1991: WEIL et al.. 1991: KAZURA et al., 1993; DREYER et il., 1994). The therapeu: tic efficiency combined with low cost, low toxicity, and the fact that it induces only mild and transient adverse reactions (DREYER et al., 1994) makes DEC an appropri- ate drug for use in large scale control programmes.

The standard dose of DEC for treatment of ban- croftian filariasis is 6 mg/kg body weight given daily for 12 consecutive days, but the many daily dosages are dif- ficult to administer on a large scale. Thus, there is a need to develop alternative regimens for control pro- grammes, which are easy to administer, effective in re- ducing the parasite burdens and acceptable to the popu- lation (WHO. 1992). Various studies have shown that DEC given in low, ‘single or spaced dosages can effec- tively reduce microfilaraemia (PARTONO et al., 1984; EBERHARH et al., 1991; CARTEL et al., 1992; KIMURA et al.. 1992: ADDISS et al.. 1993: KAZURA et al.. 1993). Al- though such strategies ‘may offer feasible alternatives to the standard regimen, it is generally difficult to compare results from the different trials since study designs, diag- nostic techniques and evaluation criteria have varied.

A series of community based trials have therefore been carried out in an endemic area of Tanzania, with the aim of comparing the efficacy of different strategies of DEC chemotherapy in bancroftian filariasis. The first study was based on selective treatment given to micro- lilaraemic individuals only (SIMONSEN et al., 1995). Based on the results, it was argued that mass DEC che- motherapy would be superior to selective treatment, both with respect to its cost and its effect on micro- filaraemia and clinical manifestations. The present study therefore evaluated and compared the efficacy of the standard dose regimen and 2 semi-annual single doses (6 mg/kg body weight) administered as mass treatment to

Address for correspondence: Dr Paul E. Simonsen, Danish Bil- harziasis Laboratory, Jaegersborg Alle lD, 2920 Charlotten- lurid, Denmark.

all individuals, irrespective of infection status, in 2 en- demic communities in north-eastern Tanzania. The treated individuals were followed for one year, and the efficacy of treatment was assessed both among individu- als who were microfilaraemic or had clinical manifesta- tions, and at the community level.

Materials and Methods Study areas and populations

The study took place in 2 coastal villages, Tawalani and Kwale, 38 and 27 km north of Tanga in Tanzania. They were fishing communities, but some subsistence farming was also practised. Most houses had mud walls and roofs thatched with dried coconut leaves. Electricity was not available. Medical treatment was most often pro- vided by hospitals in Tanga. In recent years, no control measure against bancroftian filariasis has been imple- mented in this part of Tanzania, and DEC has not been readily available.

Study design Initial village meetings were held to explain the pur-

pose and procedure of the study. The villagers were reg- istered durine house-to-house visits, and all individuals aged one yea; and above at the beginning of the study were included. Following a pre-treatment cross-sectional clinical and parasitological survey, all individuals were offered treatment with DEC according to the strategy adopted for the village. The intake of tablets was closely supervised by the survey team. All individuals who were microfilaraemic during the pre-treatment survey were re-examined parasitologically 3 and 6 months after the start of treatment. One year after the start of treatment, a full clinical and parasitological survey, covering all vil- lagers examined in the pre-treatment survey, was carried out.

Clinical and parasitological examinations Males and females were examined for clinical manifes-

tations of lymphatic filariasis in legs and arms, and males were examined for genital manifestations. The manifestations were graded as previously described (MEYROWITSCH et al., 1995). but the grading has been omitted in this presentation. Instead,lhydrocele is de- fined as a swelling in the scrotum s6 cm with fluid ac-

70 D.W.MEYROWITSCH ETAL.

cumulation and leg elephantiasis is defined as loss of contour due to swelling of the affected part (including both lymphoedema and elephantiasis). During the clini- cal examination, males were asked whether hydrocelec- tomy had been performed. Only a few cases of arm and scrotal elephantiasis were seen and records of these con- ditions have been omitted.

Blood sampling for parasitological examinations was conducted between 21:00 and 01:OO; 100 PL of finger prick blood were collected in heparinized capillary tubes and the samples were transferred to 1 mL of 3% acetic acid. The microfilariae (rnf) in the specimens were counted using the counting chamber technique (MC MAHON et al., 1979), and the number of mf per mL was calculated.

Chemotherapy DEC tablets (50 mg; Pharmamed Ltd, Malta) were

obtained from UNICEF, UNIPAC, Copenhagen, Denmark. In Tawalani, all individuals received the standard dose (WHO, 1992), i.e. 6 mg/kg body weight per day for 12 d, given as single daily doses. Each individual received the full amount of tablets in a container and was carefully instructed on how to take them. The daily intake of tab- lets was closely supervised by the survey team. In Kwale, all individuals received a single dose of 6 mg/kg body weight under supervision at the start of the treatment. Six months later, they received another similar single dose under supervision. The second dose was given shortly after the 6 months parasitological follow-up ex- amination had been conducted. Only a few irregularities or refusals of drug intake were recorded. Individuals complaining of side effects were given paracetamol.

Data analysis The geometric mean mf intensity (GMI) was calcu-

lated as antilog [C(log x+1)/n], with x being the number of mf/mL in the blood of microfilaraemic individuals and n the number of microfilaraemic individuals exam- ined. When analysing the effect of DEC treatment in the originally microfilaraemic cases, those reverting to ami- crofilaraemic status were also included.

Student’s t test was used for simple comparisons of GMIs. The effect of DEC treatment on mf intensities was examined by analysis of variance (ANOVA) on log- transformed values, with age, sex and treatment strategy as independent variables. The pre-treatment intensities were used as a covariate in these analyses in order to compensate for differences in the level of pre-treatment mf intensities.

Pearson’s x2 test was used for simple comparisons of mf prevalences. The effect of DEC treatment on mf prevalences was examined by logistic regression analysis. In all tests, sex, age and pre-treatment levels of mf inten-

Table 1. Overall results from the pre-treatment sur- veys for bancroftian filariasis in Tawalani and Kwale, Tanzania

No. of inhabitants Survey coverage Micro’rilarial prevalence Microfilarial intensitya

Range GMIb

Hydrocele prevalenceb Leg elephantiasis prevalenced

a!Y. bancrofti microfilariae/mL. bGeometric mean intensity. CIn males aged 220 years. din individuals aged 320 years.

Tawalani 502

97.2%

285%

Kwale 839

94.8%

17.7%

lo-12880 588

37.9% 6.8%

10-7170

251

30.2%

4.3%

Table 2. The effect of DEC treatment on microfilarae- mia in individuals from Tawalani (receiving the standard 12 d treatment) and Kwale (receiving 2 semi- annual single-dose treatments) who were micro- filaraemic during the pre-treatment survey

Tawalani Kwale Before treatment

No. of individual@ 120 (64156) 113 (56/57) Mean age in years (range) 34.8 (5-70) 30.9 (3-73)

Microlilarial intensityb Range lo-12880 10-7170 GMIc 645 257

One year after start of treatment Microlilarial clearance rate 51.2% 36.0%

Microfilarial intensityb Range O-1210 O-1250 GMIc 9 20

Reduction in GM1 98.6% 92.2%

aOnly individuals completing the treatments and follow- up examinations satisfactorily are included (numbers of males/females in parentheses).

bI#‘. bancrofii microfilariaeiml. CGeometric mean intensity.

sities were used as confounding variables. This allowed for a comparison of the effect of DEC treatment in com- munities with unbalanced age, sex and pre-treatment levels of mf intensity. For analysis of the effect of age on the outcome of treatment, the examined individuals were divided into 7 groups aged l-10, 11-19, 20-29, 30-39,40-49, 50-59, and a60 years. Factors were elimi- nated from the complete model on the basis of the Wald statistic, which indicated whether the variables had a significant influence on the mf prevalence, when adjust- ments for confounding by the other variables were made. The level of significance of the Wald statistic is indicated in the text.

Results Filariasis in the villages before treatment

The results of the pre-treatment surveys are summa- rized in Table 1. A more detailed description and analy- sis of the pre-treatment survey results is given by MEY- ROWITSCH et al. (1995). Before treatment, the mf prevalence and GM1 wet-f significantly higher in Tawalani than in Kwale (x test, P<O.OOl and t test, PcO.001, respectively). In the following analyses, the difference in pre-treatment mf GM1 between these 2 vil- lages was compensated for by the statistical methods used.

The effect of DEC treatment on microjilaraemia in individu- als who were microfilaraemic in the pre-treament survey

In order to analyse the effect of the 2 treatment regi- mens on microfilaraemia among individuals who were microfilaraemic at the pre-treatment survey, those who did not complete the full treatments were excluded. Those who missed the 12 months follow-up examination or who missed more than one of the 3 or 6 months fol- low-up examinations were also excluded. These criteria allowed 120 microfilaraemic individuals from Tawalani and 113 microfilaraemic individuals from Kwale to be included in the analvsis (Table 2). In Tawalani and Kwale, the 2 study groups comprised 88.2% and 82.5%, respectively, of those who were microfilaraemic in the pre-treatment survey.

The effect of the 2 treatment regimens on micro- filaraemia is shown in Table 2 and Fig. 1. The standard

MASS DEC CHEMOTHERAPY FOR BANCROFTIAN FILARIASIS: COMPARISON OF TWO REGIMENS 71

0 3 6 9 12

Months after start of treatment

Fig. 1. The effect of DEC chemotherapy on the prevalence of IV. bancrofn’ microfilaraemia in individuals who were microfilaraemic during the pre- treatment survey: n , standard dose treatment given in Tawalani; A, 2 semi-annual single treatments given in Kwale. Only individuals com- pleting the treatments and follow-up surveys are included.

12 d treatment resulted in mf clearance rates of 28.6%, 40.7% and 51.2% after 3, 6 and 12 months, respectively. The semi-annual treatment regimen resulted in a mf clearance rate of 19.4% 3 months after start of treatment. Six months later (shortly before the second dose was given), the clearance rate had increased to 30.8%, and 12 months after the start of treatment (i.e. 6 months after the second dose), 36.0% had reverted to an amicrofilarae- mic status. The continued decrease in the mf prevalence observed in both study groups between 3 and 6 months and between 6 and 12 months after the start of treatment was statistically significant (x2 test, P<O.OOl for all tests). Logistic regression analysis showed that the standard 12 d treatment was significantly more effective in reducing the prevalence of microfilaraemia than the semi-annual treatment during the entire follow-up pe- riod (P<O.Ol at all times). No significant effect of age or sex on the clearance rate of microfilaraemia was ob- served (JYO.05 at all times).

Both treatment regimens had a marked effect on the mf GM1 in the study groups (Table 2 and Fig. 2). In the Tawalani group, the mf GM1 was reduced by 96.2%, 97.5% and 98.6% at 3, 6 and 12 months after the start of

0 3 6 9 12

Months after start of treatment

Fig. 2. The effect of DEC chemotherapy on the geometric mean intensity of W. bancrofti microfilaraemia in individuals who were microfilaraemic during the pre-treatment survey: n , standard treatment given in Tawalani; A, 2 semi-annual single treatments given in Kwale. For each village, the intensity is indicated as a percentage of the pre-treatment level. Only individuals completing the treatments and follow-up surveys are included.

treatment. In the Kwale group, the mf GM1 was reduced by 70.0%, 86.3% and 92.2% at 3, 6 and 12 months after starting treatment. In both groups, the mf GM1 de- creased significantly between the pre-treatment and 3 months, between thk 3 and 6 montfis, and between the 6 and 12 months follow-un examinations U’<O.OOl for all periods). Analysis of var;ance, with mf pie-treatment in- tensities as a co-variate, showed that the standard regi- men reduced the mf GMIs significantly more than the semi-annual regimen at all times during the follow-up period (P<O.OOl at all times).

The effect of DEC treatment on clinical manifestations The effect of the 2 DEC treatment regimens on the

clinical manifestations was examined. fhose who did not complete the full treatments and those who did not attend the 12 months follow-up examination were ex- cluded from the analysis. Males who claimed that they had had hvdrocelectomies nerformed were also excluded from anal-ysis. These critkria allowed 19 males from

Table 3. The effect of DEC treatment on hydrocele in males from Tawalani (receiving the standard 12 d treatment) and Kwale (receiving 2 semi-annual single- dose treatments)

No. of males No. of males one year after

with start of treatment showing

hydrocele at Reduction in size Complete

pre-treatment of hydrocele, but disappearance

Village survey not disappearance of hydrocele

Tawalani 19 4 (21.0%) 7 (36.8%)

Kwale 22 5 (22.7%) 10 (45.5%)

aOnly males completing the treatment and the one year follow- up examinations, and who had not had a hydrocelectomy, are included.

Tawalani and 22 males from Kwale (43.2% and 52.3% of those having hydroceles during the pre-treatment sur- veys in Tawalani and Kwale, respectively) to be in- cluded. Both treatment regimens had a marked effect on the hydroceles (Table 3). One year after the start of treat- ment, 57.9% and 68.2% of the males with hydrocele dur- ing the pre-treatment surveys in Tawalani and Kwale, respectively, showed improvement (either reduction in size of the hydrocele or its complete disappearance). Some males whose hydrocele had disappeared presented a slightly extended scrotal sac, but the fluid accumula- tion had vanished. Complete disappearance of the hy- drocele was seen only in males who had had small hy- droceles (6-8 cm) during the pre-treatment survey.

One year after the start of DEC treatment, no clear ef- fect was observed on leg elephantiasis. Among 12 indi- viduals in Tawalani who had leg elephantiasis during the pre-treatment survey, one had developed a more ad- vanced stage, whereas the condition had been reduced to an earlier developmental stage in one individual, and in another it had disanueared. Among 5 individuals in Kwale who had leg kiephantiasis diring the pre-treat- ment survey, the condition remained unchanged in 4 and was reduced to an earlier stage in one.

Filariasis in the villages one year after the start of treatment Nobody in the villages refused to complete the treat-

ments. However, a few individuals failed to receive the full treatment due to travelling. One year after initiation of mass treatment, a clinical and parasitological survey was carried out in both villages (Table 4). Compared to the pre-treatment surveys, the burden of filariasis in the villages had been reduced, but microfilaraemia was still prevalent in both villages.

One new case of microfilaraemia was found in Tawalani and 4 new cases were found in Kwale among

D. W. MEYROWITSCH ETAL.

Table 4. Overall results from the one year po$t-treat- ment surveys for bancroftian filariasis in Tawalani (receiving the standard 12 d treatment) and Kwale (receiving 2 semi-annual single-dose treatments)

Tawalani Kwale No. examined 415 617

Coverage 82.7% 73.5% Microfilarial prevalence 15.1% 11.6%

Reduction 47.0% 34.5% Microfilarial intensitya

Range 10-5540 lo-1250 GM1 112 102

Reduction 81.0% 59.4% Hydrocele prevalenceb 27.0% 22.5%

Leg elephantiasis prevalencec 3.7% 2.2%

New microfilaraemic casesd 1 (0.3%) 4 (0.7%)

alFS: bancrofi microfilariae/mL. bin males aged>20 years. CIn individuals aged a20 years. dMicrofilaraemic individuals who were amicrofilaraemic one year previously (incidence in parentheses).

the group of individuals who were amicrofilaraemic dur- ing the pre-treatment survey (Table 4). Two of the indi- viduals in Kwale had not completed their treatment.

Two new cases of hydrocele developed in Tawalani, and one in Kwale, over the one year follow-up period, giving annual incidences of 3.2% and l.l%, respectively, in males aged 20 years and above; one individual in Tawalani had not completed the treatment. No new case of leg elephantiasis was observed in either village over the one year follow-up period.

Discussion The present study evaluated and compared the effi-

cacy of mass DEC chemotherapy using the standard treatment and a semi-annual treatment regimen in 2 en- demic communities. Among individuals who were mi- crofilaraemic in the pre-treatment survey, both regimens resulted in dramatic reductions in mf GMIs one year af- ter the start of treatment; the mf clearance rates were lower. A proportion of treated individuals thus remained microfilaraemic, but they had much lower mf intensi- ties. Overall, the standard treatment was significantly more effective than the semi-annual regimen in reduc- ing mf prevalences and GMIs. This was also seen in a previous study comparing the same regimens in a selec- tive treatment strategy (SIMONSEN et al., 1995). However, considering that the first group had received 6 times as much DEC as the latter group, the differences between the efficacv of the 2 regimens anoeared to be small. One or 2 mori semi-annuil sing&lose treatments would probably have resulted in a larger reduction in micro- filaraemia than the standard dose treatment.

It was surprising to observe that both treatment regi- mens continGed to-reduce the prevalence and intensity-of microfilaraemia long after the administration of DEC. DEC and its metabolites have been reported to be com- pletely excreted from the body within 48 h after intake (MACKENZIE & KRON. 1985). The mode of action of DEC is poorly understood, but thk prolonged effect of the treat- ments could be accounted for by an indirect action of the drug, e.g. by activating the host’s immune system or by modifying the parasite surface (MACKENZIE & IRON, 1985; OTTESEN, 1985), thereby promoting immune attack. It is also possible that killing or sterilization of adult worms by DEC may have caused the slow but steady decrease of microfilaraemia observed after treatment.

In addition to reducing microfilaraemia, the DEC

treatment resulted in a remarkable improvement of the small hydroceles in males in both villages. It is likely that this effect was due to killing of adult worms, thereby easing the lymphatic drainage of the scrotum. A suppressive effect of DEC on host responses associated with obstruction of the lymphatics may also have con- tributed to this. A previous study based on selective che- motherapy also suggested that DEC treatment had a beneficial effect on hydroceles (SIMONSEN et aZ., 1995), but the number of cases treated was small since only a few hydrocele patients had microfilaraemia. The present study thus stresses the importance of mass treatment, in- cluding amicrofilaraemic individuals, in order to obtain the maximum benefit for patients with hydroceles in the community. DEC treatment has previously been shown to result in an improvement of lymphoedema and ele- phantiasis (PARTONO et al., 1981; CASLEY-SMITH et al., 1993) but, since only a few patients with these condi- tions were present in the study villages, no clear impres- sion of this effect was obtained during the present study.

Although considerable reductions in microfilarial loads were observed after treatment of individuals who were microfilaraemic before treatment, a smaller effect was observed at the community level. This was partly due to individuals failing to complete the treatments and partly due to new cases of microfilaraemia which devel- oped over the study period. However, a more significant contributor to this difference was the fact that individu- als reverting from being microfilaraemic before treat- ment to being amicrofilaraemic one year after the start of treatment were included when calculating the mf GM1 in the microfilaraemic study population, but not at the community level. These are standard ways of calcu- lating the outcome of DEC trials in microfilaraemic in- dividuals and at community level, but the present study emphasizes that one should be cautious when extrapolat- ing the results from one type of study to the other. One year after the start of treatment, the mf prevalences and the GMIs in the treated communities were comparable to those reported from several untreated communities in other endemic areas such as India (RAJAGOPALAN et aZ., 1989), Indonesia (MAHFUDIN et al., 1977) and Kenya (ESTAMBALE et al., 1994). Thus, despite the dramatic re- ductions in the microfilarial burden in both treated communities, transmission was probably still occurring one year after the start of treatment, although at a much reduced level.

Remarkable differences were observed in the inci- dence of new cases of microfilaraemia between the pre- sent study of mass DEC chemotherapy and a previous studv in the same area based on selective DEC chemo- therapy (SIMONSEN et al., 1995). Thus, annual incidences in the present study were 0.3% and 0.7% following treat- ment with the standard and the semi-annual single-dose strategies, respectively, whereas in the study on selective treatment the figures were 4.8% and 6.5% following treatment with the same dosages. The lower incidences in the present study were probably due to an effect of the mass treatment on prepatent infections, which appeared to be amicrofilaraemic at the pre-treatment survey. It is also possible that mass DEC treatment prevented estab- lishment of new infections (WHO, 1992).

With respect to mass DEC chemotherapy, the present study indicated that, given the disadvantages of the standard dose regimen, semi-annual single-dose treat- ment offers a useful and effective alternative. The lower efficacy of the semi-annual regimen in reducing micro- filaraemia could probably be compensated for by con- tinuing semi-annual treatments for a slightly longer pe- riod. The frequency and severity of side effects and adverse reactions after treatment were not recorded, but it was obvious to the survey team that these would have had a much bigger influence on compliance among vil- lagers receiving the standard treatment than among those receiving the semi-annual single-dose treatments, if the survey team had not been present. Semi-annual

MASS DEC CHEMOTHERAPY FOR BANCROFTIAN FILARIASIS: COMPARISON OF TWO REGIMENS 73

single-dose treatment was clearly more acceptable to the community than the standard regimen. Furthermore, considering that semi-annual treatment has the advan- tage of being easy to administer, it may in the long run be a more feasible strategy for mass DEC chemotherapy than the standard treatment. Additional follow-up sur- veys will be carried out in these 2 communities to moni- tor the effect of the mass DEC chemotherapy over a longer period.

Acknowledgements The inhabitants and village assistants of Tawalani and Kwale

are thanked for their co-operation. We are also grateful to the survev team from Bombo Field Station (Z. Ali. M. Baeabuie. C. Guzoand J. Kivugo), for their valuable assistauce. Pr~fessdr’W. L. Kilama (Director General, National Institute for Medical Research, Tanzania), the late Dr S. G. Irare (former Director, Amani Medical Research Centre, Tanzania) and Mr Y. G. Ma- tola (Director, Amani Medical Research Centre, Tanzania) are thanked for their support. Further, we thank Dr H. Madsen (Danish Bilharziasis Laboratory) for helpful advice on statis- tics, and Drs E. A. Ottesen and C. P. Ramachandran (WHO, Geneva) for valuable discussions. The study received financial support from the Council for Development Research (DANIDA), Denmark.

References Addiss, D. G., Eberhard, M. L., Lammie, P. J., McNeeley, M.

B.. Lee. S. H.. McNeelev, D. F. & Suencer. H. C. (1993). Comparative efficacy of clearing-dose .and single high-dose ivermectin and diethylcarbamazine against Wuchereriu bun- crofi microfilaremia. American Journal of i"ropical Medicine and Hygiene, 48,178-185.

Cartel, J. L., Spiegel, A., Ngnoc, L. N., Cardines, R., Plichart, R., Martin, P. M. V., Roux, J. F. & Moulia-Pelat, J. P. (1992). Compared efficacy of repeated annual and semi-annual doses of ivermectin and diethylcarbamazine for prevention of Wucheretiu bancrofii filariasis in French Polvnesia. Final evaluation. Tropical Medicine and Parasitology, 43; 91-94.

Casley-Smith, J. R., Jamal, S. & Casley-Smith, J. R. (1993). Re- duction of lilaritic lymphoedema and elephantiasis by 5,6 benzo-cc-pyrone (coumarin), and the effects of diethylcar- bamazine (DEC). Annals of Tropical Medicine and Parasitology, 87,247-258.

Dreyer, G., Pires, M. L., de Andrade, L. D., Lopes, E., Medei- ros, Z., Tenorio, J., Coutinho, A., Noroes, J. & Figueredo- Silva, J. (1994). Tolerance of diethylcarbamazine by microtilaraemic and amicrofilaraemic individuals in an en- demic area of Bancroftian filariasis, Recife, Brazil. Transac- tions of the Royal Society of Tropical Medicine and Hygiene, 88, 232-236.

Eberhard, M. L., Dickerson, J. W., Hightower, A. W. & Lam- mie, P. J. (1991). Bancroftian tilariasis: long-term effects of treatment with diethylcarbamazine in a Haitian population. American Journal of Tropical Medicine and Hygiene, 45, 728-733.

Estambale, B. B. A., Simonsen, P. E., Knight, R. & Bwayo, J. J. (1994). Bancroftian lilariasis in Kwale District of Kenya. I. Clinical and parasitological survey in an endemic commu- nitv. Annals of Troaical Medicine and Parasitologv. 88. 145-15 1.

Kazura, J., Greenbeig, J., Perry, R., Weil, G., Da”;; K.‘& Alpers, M. (1993). Comparison of single-dose diethylcarbamazine and ivermectin for treatment of bancroftian filariasis in

Papua New Guinea. American Journal of Tropical Medicine and Hygiene, 49,804-g 11.

Kimura, E., Spears, G. F. S., Singh, K. I., Samarawickrema, W. A., Penaia, L., Sone, P. F., Pelenatu, S., Faaiuaso, S. T., Self, L. S. & Dazo, B. C. (1992). Long-term efficacy of single-dose mass treatment with diethvlcarbamazine citrate against diur- nally subperiodic Wucher&a bancrofti: eight years’experience in Samoa. Bulletin of the World Health Organization, 70, 76%77h I__ .,-.

Mackenzie, C. D. & Kron, M. A. (1985). Diethylcarbamazine: a review of its action in onchocerciasis, lymphatic tilariasis and inflammation. Tropical Diseases Bulletin, 82, l-37.

Mahfudin, H,., Partono, F., Purnomo & Idris, K. N. (1977). Parasitologtcal and clinical survey of bancroftian filariasis in Kepu District, Central Jakarta, Indonesia. Southeast Asian Journal of Tropical Medicine and Public Health, 8, 173-178.

McMahon, J. E., Marshall, T. F. de C., Vaughan, J. P. & Abaru, D. E. (1979). Bancroftian filariasis: a comparison of micro- lilariae counting techniques using counting chamber, stand- ard slide and membrane (Nuclepore) filtration. Annuls of Tropical Medicine and Parasitology, 73,457-464.

Meyrowitsch, D. W., Simonsen, P. E. & Makunde, W. H. (1995). Bancroftian filariasis: analysis of infection and dis- ease in five endemic communities of north-eastern Tanzania. Annals of Tropical Medicine and Parasitology, 89,653-663.

Ottesen, E. A. (1985). Efficacy of diethylcarbamazine in eradi- cating infection with lymphatic-dwelling filariae in humans. Reviews of Infectious Diseases, 7, 341-356.

Partono, F., Purnomo, Oemijati, S. & Soewarta, A. (1981). The long term effects of repeated diethylcarbamazine administra- tion with special reference to microfilaraemia and elephantia- sis. Acta Tropica, 38,217-225.

Partono, F., Purnomo, Soewarta, A. & Oemijati, S. (1984). Low dosage diethylcarbamazine administered by villagers for the control of timorian filariasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 78, 370-372.

Rajagopalan, P. J., Das, P. K., Subramanian, S., Vanamail, P. & Ramaiah, K. D. (1989). Bancroftian filariasis in Pondicherry, South India: 1. Pre-control epidemiological observations. Epidemiology and Infection, 103,685-692.

Richards, F. O., Eberhard, M. L., Bryan, R. T., McNeeley, D. F., Lammie, P. J., McNeeley, M. B., Bernhard, Y., Hightower, A. W. & Spencer, H. C. (1991). Comparison of high dose ivemectin and dtethylcarbamazine for activity against bancroftian filariasis in Haiti. American Journal of Tronical Medicine and HvPiene. 44.3-10.

Simonsen, P. E., Meyrow&h, b. W., Makunde, W. H. & Mag- nussen, P. (1995). Selective DEC chemotherapy for control of bancroftian filariasis in two communities of-Tanzania: com- pared efficacy of a standard dose treatment and two semi-an- nual single dose treatments. American Journal of Tropical Medicine and Hygiene, 53,267-272.

Weil, G. J., Lammie, P. J., Richards, F. 0. & Eberhard, M. L. (1991). Changes in circulating parasite antigen levels after treatment of bancroftian tilariasis with diethylcarbamazine and ivermectin.3oumal OfInfectious Diseases, 164,814-816.

WHO (1992). Lymphatic filariasis: the disease and its control. Fifth report of the WHO Expert Committee on Filariasis. Geneva: World Health Organization, Technical Report Series, no. 821.

Received 22 June 1995; revised 18 August 1995; accepted for publication 22 August 1995