Impact of the 2004 mass drug administration for the ...

Annals of Tropical Medicine & Parasitology, Vol. 101, No. 4, 335–341 (2007)

Impact of the 2004 mass drug administration for the control of lymphatic filariasis, in urban and rural areas of the Western province of Sri Lanka G. S. A. GUNAWARDENA*, M. M. ISMAIL*, M. H. BRADLEY{ and N. D. KARUNAWEERA* *

Department of Parasitology, Faculty of Medicine, University of Colombo, P.O. Box 271, Kynsey Road, Colombo 8, Sri Lanka { GlaxoSmithKline Services Unlimited, 980 Great West Road, Brentford TW8 9GS, U.K. Received 11 May 2006, Revised 29 August 2006, Accepted 1 September 2006

Lymphatic filariasis is targeted to be eliminated globally, at least as a public-health problem, by 2020. The comprehensive strategy for achieving this goal includes the interruption of the transmission of the causative parasites, by drastically reducing the prevalences of microfilaraemia in at-risk communities. In a descriptive, comparative, cross-sectional and community-based study, the impact of the 2004 mass drug administration (MDA) on filarial infection, in selected areas of the Western province of Sri Lanka, has now been assessed 1–2 and 11 months after the administration of the diethylcarbamazine–albendazole combination employed. Using the cluster-sampling method, urban study populations were selected in the Colombo districts and rural ones were selected in the Gampaha district. After obtaining informed written consent, 2 ml venous blood were collected, between 20.00 and 24.00 hours, from each subject. Personal details and drug compliance in the 2004 MDA were recorded. The samples of ‘night’ blood were checked for microfilariae, using membrane filtration, and for filarial antigenaemia, using commercial (NOWH) immunochromatographic test kits. Eighty-four (4.10%) of the 2034 subjects examined 1–2 months after the 2004 MDA but only four (0.20%) of the 1974 subjects checked 11 months after the MDA were found antigenaemic and/or microfilaraemic (P,0.001). Between the two follow-ups, the overall prevalences of both antigenaemia (4.03% v. 0.15%; P,0.001) and microfilaraemia (0.20% v. 0.05%; P50.38) fell, although only the reduction in antigenaemia was statistically significant. The prevalence of infection (as indicated by antigenaemia and/or microfilaraemia) fell significantly within each of the two study districts (P,0.001). Although, when the prevalence of infection was high, drug compliance appeared to be an important determinant of the reduction of antigenaemia (P50.04), the 20% difference in compliance between urban and rural areas had no apparent effect on the corresponding prevalences of infection recorded at either follow-up.

Lymphatic filariasis (LF) is endemic in 83 countries, and about 40 million of the 120 million people infected by one of the causative parasites are incapacitated and disfigured by the disease (WHO, 2005a, b). In 1997, the 50th World Health Assembly resolved that lymphatic filariasis should be eliminated as a public-health Reprint requests to: G. S. A. Gunawardena. E-mail: [email protected]; fax: z94 11 269 9284. # 2007 The Liverpool School of Tropical Medicine DOI: 10.1179/136485907X176364

problem. The World Health Organization subsequently proposed a comprehensive strategy for achieving this goal, which included interrupting the transmission of the causative parasites, by drastically reducing the prevalences of microfilaraemia in endemic communities, and the prevention or management of LF-related disability in those already infected (WHO, 2005a). The Global Programme to Eliminate Lymphatic Filariasis commenced its first campaign based on mass drug administrations

336

GUNAWARDENA ET AL.

(MDA) in 1999, in Samoa (WHO, 2005a). By 2003, more than 70 million people at risk in 36 endemic countries were covered by MDA campaigns (WHO, 2005b). The Sri Lankan programme of MDA against LF commenced in 1999, with the distribution of a single dose of diethylcarbamazine citrate (DEC) given twice a year. Since 2001, albendazole has been added to the DEC and the MDA have been annual. In 2002, 86% of the at-risk population in Sri Lanka, then estimated to be 9 million, received treatment in the MDA (WHO, 2003). Further rounds of MDA were conducted in August 2004 and July 2005. The objective in Sri Lanka is to interrupt the transmission of the parasites causing LF by the annual administration of single doses of the DEC–albendazole combination to all eligible individuals in endemic areas for at least five consecutive years. When assessing the effects of MDA on transmission it is important to explore treatment compliance and the impact of the MDA on the prevalence of microfilaraemia (WHO, 2005a). The aim of the present study, which was descriptive, comparative, cross-sectional and communitybased, was to assess the impact of the 2004 Sri Lankan MDA on filarial infection 1–2 and 11 months after the drug distribution, in both urban and rural areas of the Western province.

SUBJECTS AND METHODS Study Areas and Population The urban subjects came from six medicalofficer-of-health (MOH) areas (Colombo, Kolonnawa, Nugegoda, Thimbirigasyaya, Dehiwela and Moratuwa) randomly selected from those forming the Colombo district of the Western province. The rural subjects came from six MOH areas (Katana, Wattala, Gampaha, Biyagama, Kelaniya and Mahara) randomly selected from those forming the Gampaha district of the same province. A public-health-inspector (PHI)

area was then randomly selected from those forming each of the selected MOH areas. Five main roads in each chosen PHI area were then selected at random, using area maps, and the first 35 houses in each selected road were visited. A volunteer from each visited household was enrolled in the study. Using this cluster-sampling technique, approximately 1000 subjects from each district were selected for blood sampling 1–2 months after the 2004 MDA. The subjects for the second follow-up, conducted 11 months after the MDA, were mostly those investigated in the first. If, however, a subject from the first follow-up declined to participate in (or was unavailable for) the second follow-up, he or she was replaced in the second follow-up by a volunteer from the same household or, if this was not possible, by a volunteer from another household nearby. At each follow-up, the personal details of each subject and his or her compliance in the 2004 MDA were recorded. Subjects found to be microfilaraemic and/ or harbouring circulating filarial antigen (CFA) during the first follow-up were advised to take a course of DEC, as recommended by the World Health Organization (WHO, 1984), and were excluded from the analysis of the data from the second follow-up. Collection and Processing of Blood Samples In each follow-up, a 2-ml sample of venous blood was collected from each subject between 20.00 and 24.00 hours and then stored at 4uC until it could be checked, at the University of Colombo’s Department of Parasitology, for microfilariae (mff) and CFA. All the blood samples were processed within 72 h. Half of each sample was passed through a 3-mm-pore membrane filter (NucleporeH; Whatman International, Maidstone, U.K.) and any mff trapped on the filter were then counted. A further 100 ml of each sample were checked for CFA, using

IMPACT OF MASS DRUG ADMINISTRATION IN SRI LANKA

a commercial, immunochromatographic test (NOWH Filariasis Test; Binax, Scarborough, ME).

Data Analysis Data were analysed using version 10.0 of the SPSS for Windows software package (SPSS Inc., Chicago, IL). Prevalences of filarial infection, as revealed by microfilaraemia and/or antigenaemia, were compared, between follow-ups and study areas, using x2 tests. The reported compliance in the MDA was also considered, in relation to the prevalences of filarial infection at each follow-up and in each study area, using x2 tests or, whenever the expected figure in a cell was ,5, Fisher’s exact tests.

Ethical Clearance Written informed consent was obtained from each adult subject and from a parent or guardian of each child investigated. The study protocol was approved by the Ethical Review Committee of the University of Colombo’s Faculty of Medicine.

RESULTS There were no significant differences in age, gender or other demographic characteristics between the subjects of the first follow-up and those of the second or between the subjects from each district, overall or at each follow-up (data not shown).

Prevalence of Filarial Infection Blood was collected from 2034 subjects 1–2 months after the 2004 MDA and from 1974 subjects 11 months after the MDA. The ages of the subjects ranged between 5 and 85 years (median537 years). The prevalences of microfilaraemia and antigenaemia recorded in each MOH area at each followup are shown in Table 1.

337

Impact of the 2004 MDA The prevalence of infection, as indicated by microfilaraemia and/or antigenaemia, was significantly higher 1–2 months after the 2004 MDA than 9–10 months later (4.10% v. 0.20%; x2571.95; P,0.001). Eighty-two subjects in the first follow-up (4.03%) but only three (0.15%) in the second were found positive for CFA (x2572.63; P,0.001). The prevalence of microfilaraemia also fell between the follow-ups but, probably because microfilaraemia was quite rare at both follow-ups, the difference was not statistically significant. Four (0.20%) of the subjects of the first follow-up but just one (0.05%) of those investigated during the second follow-up were found microfilaraemic (P50.38). The prevalences of infection recorded in the urban Colombo district 1–2 and 11 months after the 2004 MDA were 4.60% and 0.31%, respectively (x2537.85; P,0.001). The corresponding prevalences of infection in the rural Gampaha district also showed a significant reduction, from 3.70% to 0.05% (x2534.29; P,0.001). Although the reduction in the prevalence of antigenaemia between the two follow-ups was highly significant in Colombo district (x2535.84; P,0.001) and Gampaha district (x2537.19; P,0.001), the corresponding reductions in the prevalences of microfilaraemia were not large enough to reach statistical significance, either in Colombo district (P50.13) or Gampaha (P50.49). Prevalence of Infection: Urban v. Rural One to two months after the 2004 MDA, the prevalence of antigenaemia recorded in the urban study areas of Colombo district was similar to that recorded in the rural study areas of Gampaha district (4.40% v. 3.70%; x250.66; P50.42). Although the prevalences of microfilaraemia recorded in the first follow-up were higher in the urban areas than in the rural, the difference did not quite reach statistical significance (0.40% v. 0.0%; P50.06).

Colombo Colombo (Aluthmawatha) Thimbirigasyaya (Pamankade East) Dehiwela (Dehiwela West) Moratuwa (Moratumulla) Kolonnawa (Kolonnawa Town) Nugegoda (Nawala East) All investigated Gampaha Gampaha (Ihala Yagoda) Kelaniya (Waragoda) Wattala (Nugape) Mahara (Ranmuthugala) Biyagama (Mabima) Katana (Demanhandiya) All investigated

District

Medical-officer-of-health and (public-health-inspector) area

11 months after MDA

344 154 130 124 127 123 1002 176 188 167 163 182 156 1032

0 0 0 4 0 0 4 0 0 0 0 0 0 0

(0.0) (0.0) (0.0) (3.2) (0.0) (0.0) (0.4) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)

7 7 12 12 4 2 44 2 14 7 4 3 8 38

(2.0) (4.5) (9.2) (9.7) (3.1) (1.6) (4.4) (1.1) (7.4) (4.2) (2.5) (1.6) (5.1) (3.7)

147 200 200 183 57 196 983 151 200 167 201 85 187 991

0 0 0 0 0 0 0 0 1 0 0 0 0 1

(0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.5) (0.0) (0.0) (0.0) (0.0) (0.1)

0 1 0 0 0 2 3 0 0 0 0 0 0 0

(0.0) (0.5) (0.0) (0.0) (0.0) (1.0) (0.3) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0) (0.0)

Investigated Found microfilaraemic Found antigenaemic Investigated Found microfilaraemic Found antigenaemic

1–2 months after MDA

No. and (%) of subjects:

TABLE 1. Prevalences of filarial infection, as indicated by microfilaraemia and/or antigenaemia, 1–2 and 11 months after the 2004 mass drug administrations (MDA) in the Colombo and Gampaha districts of Sri Lanka

338 GUNAWARDENA ET AL.

339


Similarly, 11 months after the 2004 MDA, the between-district differences in the prevalences of antigenaemia (0.31% in Colombo v. 0.0% in Gampaha; P50.12) and microfilaraemia (0.0% in Colombo v. 0.10% in Gampaha; P51.00) were not statistically significant.

compliers but 0.10% (just one) of the noncompliers was found to be infected (x250.19) but no significant relationships were observed between drug compliance and the prevalence of infection (P51.00), microfilaraemia (P.0.05) or antigenaemia (P.0.05).

Drug Compliance and Filarial Infection The number of subjects who reported that they had or had not ingested tablets distributed in the 2004 MDA, and the corresponding prevalences of infection at each follow-up, are shown in Table 2. Almost 60% of the subjects who were examined 1–2 months after the MDA reported that they had taken the drugs. At the first follow-up, the prevalence of infection in these ‘compliers’ (3.5%) was lower than that recorded among the ‘noncompliers’ (5.1%) — the subjects who said that they had not taken any tablets in the MDA — but the difference did not quite reach statistical significance (x253.34; P50.07). The prevalences of microfilaraemia recorded at the first follow-up were similar in the compliers and non-compliers (x250.39; P50.65) but the prevalence of antigenaemia was significantly lower in the compliers than in the other subjects (3.3% v. 5.1%, x254.14; P50.04). Of the 1974 persons examined 11 months after the MDA, 64.5% reported that they had taken the drugs offered them in the MDA. At this follow-up, 0.20% of the

Drug Compliance: Urban v. Rural Compliance during the 2004 MDA appeared to be significantly lower in urban Colombo district than in rural Gampaha district (Table 3), whether measured in the subjects of the first follow-up (46.4% v. 72.4%; x25142.46; P,0.001) or those of the second follow-up (55.8% v. 73.1%; x2563.81; P,0.001).

DISCUSSION Monitoring the implementation of an MDA programme is a vital element that enables the success of the strategy to be assessed (WHO, 2005a). Although every step of a programme’s implementation can be monitored, the selective monitoring of a few critical aspects is generally sufficient and more cost-effective (WHO, 2005a). In the present study, only treatment compliance and the prevalences, at two time-points after an MDA, of microfilaraemia and antigenaemia were investigated in detail. As recommended by the World Health Organization (2005a), samples of ‘night’ blood were

TABLE 2. Drug compliance and the numbers of microfilaraemics and antigenaemics recorded 1–2 and 11 months after the 2004 mass drug administration (MDA) No. of subjects found: Time after 2004 MDA (months)

Compliance

1–2

Drugs Drugs Any Drugs Drugs Any

11

taken not taken taken not taken

Microfilaraemic

Antigenaemic

3 1 4 1 0 1

40 42 82 2 1 3

340

GUNAWARDENA ET AL.

TABLE 3. Drug compliance in the 2004 mass drug administration (MDA), as recorded, in the urban study areas of Colombo district and the rural study areas of Gampaha district, in the first follow-up (1–2 months after the MDA) and the second (11 months after the MDA) No. and (%) of subjects: In first follow-up, from: Compliance Drugs taken Drugs not taken Any

In second follow-up, from:

Colombo

Gampaha

Colombo

465 (46.4) 537 (53.6) 1002 (100)

747 (72.4) 285 (27.6) 1032 (100)

549 (55.8) 434 (44.2) 983 (100)

checked for mff and an immunochromatographic antigen-detection test was used to check each subject for CFA. The MDA carried out on 8 August 2004 was the fourth to be implemented in Sri Lanka. The subsequent ‘mopping-up’ operations, with drug distribution in areas missed by the main MDA, took 2–3 weeks. The first follow-up of the present study was conducted in September–October 2004 (1– 2 months after 8 August and soon after the mopping-up had been completed). The second follow-up immediately preceded the fifth Sri Lankan MDA, which began on 24 July 2005. The general trend in the prevalence of microfilaraemia has certainly been downward since the first Sri Lankan MDA, although changes in the methods used to detect microfilaraemia over this period make valid multi-year comparisons difficult. The present results, which show the prevalence of microfilaraemia falling from 0.20% to just 0.05% between the fourth and fifth MDA, are very encouraging. In surveys conducted, before and 6 months after the 2003 MDA, by the Anti-Filariasis Campaign of Sri Lanka — using the relatively insensitive method of the examination of smears of fingerprick samples of ‘night’ blood to detect mff — the recorded prevalences of microfilaraemia fell from 0.21%–0.82% overall to 0.04%–0.17% in Colombo district and 0.01%–0.69% in Gampaha district (WHO, 2005b). Even using the membrane filtration of a 1-ml sample of ‘night’ blood

Gampaha 724 (73.1) 267 (26.9) 991 (100)

(which is a more sensitive method than bloodsmear examination), the prevalences of microfilaraemia recorded in the present study were generally lower at the first follow-up, immediately after the 2004 MDA had been completed, than those recorded 6 months after the 2003 MDA, and were even lower 9–10 months later. The three subjects found antigenaemic in the second follow-up had not been examined during the first follow-up. There was thus no evidence of the acquisition of antigenaemia, and thus no evidence of transmission, after the 2004 MDA. As repeated treatment with DEC–albendazole combinations is known to have an important effect on the adult filarial worms as well as the mff (Ismail et al., 2001), it should reduce the prevalences of antigenaemia over time. Unfortunately, as no attempt was made to estimate the prevalence of filarial antigenaemia prior to the first Sri Lankan MDA, the effects of four MDA on filarial antigenaemia cannot be fully evaluated. In the present study, however, there is clear evidence of antigenaemia becoming increasingly rare with time after the fourth MDA. Drug compliance in the 2004 MDA ranged between 60% and 65%. Although it was markedly higher in the rural areas of Gampaha than in the urban areas of Colombo (73% v. 51%; P,0.001), this difference was not reflected in any significant difference in the prevalence of infection at the second follow-up. Compliance did, however, significantly reduce a subject’s


chance of being antigenaemic at the first follow-up (P50.04), perhaps because the prevalence of infection at the time of the MDA needs to be relatively high for treatment to have a statistically significant effect on infection. Larger sample sizes, however, may have overcome the lack of apparent effect at low levels of prevalence. Several marked differences seen in the prevalences recorded in the present study probably did not reach statistical significance because, given the generally low prevalences, the sample size was too small. In 2003, the World Health Organization received reports on mid-term assessments of MDA in sentinel sites in 13 countries in the African, American, Mediterranean and Asian regions. The results revealed significant reductions in the prevalence microfilaraemia in most of the sentinel sites (WHO, 2005b). In India, the prevalences of microfilaraemia have been reduced from 13.2% to 1.9% by six rounds of DEC (with 54%–75% compliance) and from 14.5% to 4.1% by MDA based on ivermectin (Ramaiah et al., 2002; Ramaiah and Das, 2004). Apart, however, from surveys in Haiti, which indicated that the first MDA reduced the prevalence of antigenaemia from 10.4% to 6.3% (Washington et al., 2004), there appear to be no published reports on postMDA reductions in filarial antigenaemia, or the relationship between compliance and prevalence of infection, to corroborate the findings of the present study. The authors are grateful to GlaxoSmithKline and the Sri Lankan National Science Foundation (RG/ 2005/HS/08) for providing financial support. They thank Dr T. Liyanage (director of the Anti-Filariasis Campaign in Sri Lanka) and her staff, and the medical

ACKNOWLEDGEMENTS.

341

officers of health in the study areas and their staff, for their co-operation. The assistance of H. Gunatilaka and the other staff members of the University of Colombo’s Department of Parasitology is also gratefully acknowledged.

REFERENCES Ismail, M. M., Jayakody, R. L., Weil, G. J., Fernando, D., de Silva, M. S. G., de Silva, G. A. C. & Balasooriya, W. K. (2001). Long term efficacy of single-dose combinations of albendazole, ivermectin and diethylcarbamazine for the treatment of bancroftian filariasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 95, 332–335. Ramaiah, K. D. & Das, P. K. (2004). Mass drug administration to eliminate lymphatic filariasis in India. Trends in Parasitology, 20, 499–502 Ramaiah, K. D., Vanamail, P., Pani, S. P., Yuvaraj, J. & Das, P. K. (2002). The effect of six rounds of single dose mass treatment with diethylcarbamazine or ivermectin on Wuchereria bancrofti infection and its implications for lymphatic filariasis elimination. Tropical Medicine and International Health, 7, 767–774 Washington, C. H., Radday, J., Streit, T. G., Boyd, H. A., Beach, M. J., Addiss, D. G., Lovince, R., Lovegrove, M. C., Lafontant, J. G., Lammie, P. J. & Hightower, A. W. (2004). Spatial clustering of filarial transmission before and after a mass drug administration in a setting of low infection prevalence. Filaria Journal, 3, 3. World Health Organization (1984). Lymphatic Filariasis. Fourth Report of the WHO Expert Committee on Filariasis. Techinical Report Series No. 702. Geneva: WHO. World Health Organization (2003). Annual Report on Lymphatic Filariasis 2002. Document WHO/CDS/ CPE/CEE/2003.38. Geneva: WHO. World Health Organization (2005a). Monitoring and Epidemiological Assessment of the Programme to Eliminate Lymphatic Filariasis at Implementation Unit Level. Document WHO/CDS/CPE/CEE/2005.50. Geneva: WHO. World Health Organization (2005b). Annual Report on Lymphatic Filariasis 2003. Document WHO/CDS/ CPE/CEE/2005.52. Geneva: WHO.