The management of malaria in a district hospital: what drugs?—A reply

236

TRANSACTIONS OF THE ROYAL SOCIETY OF TROPICAL MEDICINEANDHYGIENE(1993) 87, CORRESPONDENCE

1Correspondence

/

Estimating the prevalence of schistosomiasis We read with interest the article by Dr R. C. Ratard and colleagues (1992: Transactions, 86, 274276), which described a method of estimating the prevalence of schistosomiasis infection for an entire nation based on the authors’ own surveys of schoolchildren, the results of which were then extrapolated to the entire country using the ratio of the prevalence of infection in the general population to that in schoolchildren as previously found in other published studies. While the overall approach is reasonable, it would have seemed preferable to us, from a methodological point of view, to have weighted the general population/children ratios found by others according to the size of the samples used. Surveys carried out on children aged 10-19 years have the advantage of focusing on a group with a high prevalence of infection which is often easy to reach through school-based surveys. However, our experience shows that in certain socio-epidemiological situations studies limited to this target group are not always feasible and may underestimateinfection rates. Between 1986 and 1989. bv means of a census and cluster surveys carried out according to the field methodology popularized by the World Health Organization, we conducted over 25 studies in villages bordering a large artificial lake (Lac de Lagdo) in the Northern Province of Cameroon. From 1149 stool specimens examined for Schistosoma mansoni, we found a population/children ratio of 0.62, and with 1263 subiects surveved for S. haematobium we found a ratio of 0.79, which;s appreciably higher than that reported in the studies cited by Ratard et al. (lot. cit.). These results most probably reflect the influence of parasite importation into a recently flooded area coupled with conditions which favoured- transmission among adults whose chief means of livelihood was fishing. Large rural irrigation projects can completely modify rhe local socio-demographic situation (Robert, C. F. 1989: Tropical Medicine and Parasitology, 40, 153). In such circumstances establishment of the educational structure is often delayed which makes evaluation of the epidemiologv of schistosomiasis difficult using classical methods. Hywever, it is precisely in these pol&lations, where the epidemiological situation is in such flux, that trends need to be monitored closely. We regret that Ratard et al. (lot. cit.) did not take into account all data published concerning the epidemiological situation in Northern Province. This would have provided information on S. mansoni, which is lacking in the cited reference concerning this region (Ndamkou. C. N.. 1989: DSc Thesis. Facultv of Science, University of Yaounde, Cameroun).’ _ A. Rougemont C. F. Robert Unite de Sante Communautaire et Medecine‘Tropicale Centre’ Medical Universitaire 1211 Geneve 4, Switzerland 23 September 1992

Estimating reply

the prevalence

of schistosomiasis:

a

The comments by Drs Rougemont and Robert [above] underline the difficulty in extrapolating from a special age group to the entire population. The ratio cited by them is slightly higher (b.79) than the 2 highest ratios cited in our paper (0.77) (Ratard et al., 1992: Transactions, 86,27&276). ks explained by Rougemont and Robert, their high ratio was due to transmission among adults in a recently flooded area. We agree that large irrigation projects are of special epidemiological interest and need to be monitored closely with methods that have to be adapted to the special circumstances. The large majority of the population affected by schistosomiasis in Cameroon lives in areas with neither irrigation nor water

development. The most intense transmission occurs in heavily populated areas with only temporary bodies of water (Ratard et al., 1990: American Journal of Tropical Medicine and Hygiene, 42, 561-572). In estimating numbers of cases on the scale of an entire country, we attempted to look at the most typical situations and provided a wide confidence interval to take into account special situations. Raoult Ratard clo Saudi Aramco P.O. Box 10010 Dharhan 3131 I Saudi Arabia 22 October 1992

The management of malaria in a district hospital: what drugs? The article bv D. G. C. Emerton (1992: Transactions. 86,476-478) is an excellent example of audit in a tropical hospital which could have a crucial role in improving patient care. Some aspects of the antimalarial treatment merit comments. Repetition of chloroquine treatment in patients with a positive blood slide the day after a 3 d chloroquine treatment is not really satisfactory. Some of these cases may have an infection which is fully sensitive to chloroquine, and will eventually clear with or without extra chloroquine. Most of them, however, will have a chloroquine resistant infection, which should be given an alternative treatment, more likely to afford a radical cure. It is understandable that the doctors wanted to assess the response of their patients as early as day 3, before resistance could be proven; but there should be no reluctance to give sulfadoxine-pyrimethamine to hospital patients in whom the ability of chloroquine to clear parasitaemia is uncertain. To go further, it may be questioned whether, in areas where resistance to chloroquine is common, hospital patients should be routinely treated with chloroquine as the first-line drug. Negative blood slides during the first week are no guarantee that the parasites will not recrudesce later, and parasitological follow-up may prolong hospital stay beyond what is warranted by the patient’s condition. It could well be argued that hospital patients must be considered a risk group in whom a radical cure should be ensured from the onset. Allan Schapira Malaria Unit Division of Control of Tropical Diseases World Health Organization CH-1211 Geneva 27 Switzerland 10 November 1992

The management of malaria in a district hospital: what drugs?-A reply During the one year of audit of malarial treatment referred to in my paper (1992: Transactions, 86, 476478) there were 129 readmissions of patients who had been previously treated for malaria earlier in the year. Their previous in-patient treatment is shown in the Table. Table. Details of treatment of patients Murgwanza hospital, Tanzania

Drugs given One chloroquine course Two chloroquine courses Quinine and pyrimethaminesulfadoxitv? One chloroquine course and pyrimethamine-sulfadoxine Two chloroquine coursesand pyrimethamine-sulfadoxine

re-admitted

with malaria;

Patients Interval since No. who No. previous survived re-admitted admission (d) 849 279 238

79 (9.3%) 25 (8.9%) 18 (7.6%)

32

4

20

3

“Mean; ranges in parentheses. %cluding patients given chloroquine initially.

65 (5-263) 82 (6331) 107 (29-312) 124(79-244) 76 (39-102)

237 Similar percentages of patients were readmitted after one or 2 chloroquine courses and the time intervals were comparable. This suggests that, although a second chloroquine course did not achieve a more lasting cure than a single course, there was no widespread chloroquine resistance in the group given a second chloroquine course. Such a hypothesis cannot be proven, as the majority of patients who were admitted for malaria were lost to follow-up and significant numbers of children may have relapsed or even died at home. I agree with Dr Schapira’s advice that there should be no reluctance to give sulfadoxine-pyrimethamine to hospital patients in whom the ability of chloroquine to clear parasitaemia is uncertain. One restraint on the wider use of quinine for in-patients was a shortage of suitable oral quinine for children. The smallest tablet available was (and still is) the unscored 125 mg tablet. As 33% of the admissions were under one year old and a significant number of children over one year were marasmic no suitable oral quinine preparation was available. At Murgwanza hospital it is hoped that a quinine syrup will soon be prepared regularly in the dispensary. However, as quinine powder costs &lo0 per kg plus the cost of transport and any duties, it is only possible for hospitals with access to foreign currency to pay for it. As the problem of chloroquine resistence increases there is a need for 50 mg and 100 mg tablets to be available in developing countries at prices which can be afforded. David Emerton Accident and Emergency Department Morriston Hospital Morriston Swansea, SA6 6NL, UK 2Januay 1993

Vaccines view

against

schistosomes:

an alternative

In his recent leading article (1992: Transactions, 86, l2) Butterworth considered the need for a vaccine against schistosomiasis. He reviewed work on immunity in animal models and on defined recombinant antigens, concluding that in humans, in contrast to the mouse, a vaccine will need to induce responses regulated only by the Th2 subset of T-helper cells. His analysis was based on the observations that expression of immunity in human populations may involve antibodies of the immunoglobulin (Ig) E and IgA classes, both likely to be favoured by production of Th2 cytokines. Whilst agreeing that there is a strong case for a schistosome vaccine, we feel that it is too early to limit research to one particular strategy and suggest that there are equally viable alternatives. In several laboratory models of schistosomiasis a chronic infection results in an apparent immunity which has a basis in pathology, rather than acquired responses directed against the parasites (Wilson, 1991: Parasitology Today, 6, 354-358). If the integrity of the hepatic portal vascular bed is modified by the occurrence of porta-caval anastomoses, then schistosomula are able to escape and fail to establish. Since such anastomoses are a feature of the pathology in human schistosomiasis mansoni, and might persist after curative chemotherapy, it is important to show that they do not contribute to the age-related immunity so well documented by Butterworth and colleagues (1988: Philosophical Transactions of the Royal Society, series B, 321, 495-511). The advent of ultrasound techniques for assessing portal fibrosis may provide the requisite, non-invasive method to answer the question. A vaccine based on IgA responses to schistosomes is an attractive proposition but more information is needed before we can evaluate the potential contribution of such resuonses to nrotection in man. A vaccine based on the induction of *IgE responses seems more problematical. The Th2 cvtokine. interleukin 4 (IL-4). controls the switching of the immunolgobulin heavy chain gene expression from y to E in B lymphocytes. Regimes which induce IL-4 production may have some harmful sequelae _I

\

,,

since, in addition to inducing allergic responses, the cytokine has strong down-regulatory effects on cell-mediated immune function (Sher et al:, 1992: Immunological Reviews. 127. 183-204’1. It will thus be necessarv to use a vaccination regime in’which the induction of IgE responses is tightly controlled, if side effects are to be avoided. The logic underlying Butterworth’s approach to a schistosome vaccine is that it should mimic as closely as possible the protective mechanisms which develop in chronically infected humans. We suggest that an alternative strategy, employing immune mechanisms not elicited by exposure to normal parasites, is equally applicable. The irradiated cercariae vaccine, which is highly protective in mice and other laboratory hosts, provides such an example. One reason for its success is that optimally attenuated parasites induce more prolonged and intense immune responses than normal- parasites (Pemberton et al.. 1991: Zmmunologv. 73. 327-333X The vaccine elicits a iopulation of T c%s w&h Thl characteristics (Caulada-Benedetti et al., 1991: Journal of Immunology, 146, 1655-1660; Pearce & Sher, 1991: Experimental Parasitology, 73, 110-116) and these cells are recruited to the lungs. There they participate in the immune effector response which eliminates migrating schistosomula. The Thl cytokine interferon y plays a key role in the urocess (Smvthies et al.. 1992: 7ournal of Immunology, 149, 3654-3658). Such responses are not ;>rominent in mice infected with normal parasites and there is no reason to expect their occurrence in humans with schistosomiasis. Nevertheless, if they could be induced in man they might prove just as effective as they are in vaccinated mice. Thus, whilst the irradiated vaccine is unsuitable for use in the human population it nevertheless provides a useful paradigm for the development of a recombinant antigen vaccine. Further work is needed on the identification and cloning of the larval antigens which mediate protection in the irradiated vaccine model, and on vehicles for their delivery to elicit the appropriate Thl responses, before the approach can be brought to fruition. Human responses to the relevant larval antigens will also need to be characterized-a largely uncharted territory. The data outlined in the preceding paragraph provide sound evidence in an animal model for protective immunity based on Thl-mediated mechanisms. In contrast, the evidence for protection in man via ThZ-mediated mechanisms is based largely on correlative observations. At this stage in development, different strategies to reach the goal of an effective vaccine deserve equal merit. Indeed, it may prove possible to elicit both Thl and Th2 responses against different parasite stages. If these acted in a synergistic manner, then a vaccine inducing a very high level of protection might be within our grasp. R. Alan Wilson Department of Biology University of York York, YOI SDD, UK Alan Sher Laboratory ofParasitic Diseases National institute of Allergy and Infectious Disease National Institutes of Health Building 4, Room 126 Bethesda, Maryland 20892 USA 11 August I992