A randomized, placebo-controlled trial of intermittent preventive treatment with sulphadoxine-pyrimethamine in Gambian multigravidae

We investigated the ability of intermittent preventive treatment in pregnancy (IPTp) with sulphadoxine/pyrimethamine to prevent anaemia and low birthweight in Gambian multigravidae. Between July 2002 and February 2004, 2688 multigravidae living in a rural area of The Gambia received SP (1346 women) or placebo (1342 women) up to four times during pregnancy and were followed until 6-weeks post-partum. Shortly after delivery, 10.7% of women in the intervention group and 8.8% in the control group were severely anaemic [Hb < 7 g/dl, risk difference = 0.02 (95% CI -0.01, 0.04), P = 0.17]. The overall mean birthweight of infants born to women who had received SP (3103 g) was very similar to that observed in infants born to women in the control group [3075 g; difference = 28 g (95% CI -11 g, 67 g), P = 0.16]. However, among women who did not use a bednet (either insecticide treated or untreated), infants born to women who had received SP weighed more than infants born to women in the control group [3147 g vs. 3044 g; difference 143 g (95% CI 53 g, 232 g), interaction test P < 0.001]. This study did not show that IPTp with SP benefited Gambian multigravidae overall but that it may benefit a sub-group of women who do not use a bednet. In areas such as The Gambia, provision of insecticide-treated bednets to multigravidae may provide an adequate means of protection against malaria in pregnancy without the need for additional IPTp.

Integrated systems to reduce length of stay for knee and hip joint replacement surgeries

To reduce length of stay while maintaining quality of care, St. Joseph's Hospital, Hamilton, ON implemented a care path with three discharge options. Two of these discharge options were early discharge to integrated community services. Patients meeting early discharge criteria are discharged home five days post-operatively with follow-up by home care nursing and physiotherapy. Otherwise, patients are discharged on day four to a multi-disciplinary rehabilitation unit at a separate facility. Patients requiring acute medical services for complications or co-morbidity stay in the acute care hospital. A prospective cohort evaluation showed no difference in complications and similar functional outcomes for the three discharge options.

Why do some African children develop severe malaria?

Malaria is still a major cause of death and severe illness among children in many parts of tropical Africa, but only a small proportion of children, perhaps 1-2%, who become ill with malaria develop severe disease. Why only, some children experience a severe or fatal attack is not understood clearly. In this article, Brian Greenwood, Kevin Marsh and Robert Snow review some of those characteristics of the parasite and the host that may influence the outcome of a malaria infection. Identification of the relative importance of the many factors likely to be involved is needed in order to develop rational strategies for the prevention of deaths from malaria among children in malaria-endemic areas.

Risk factors for gametocyte carriage in Gambian children

A widespread reduction in Plasmodium falciparum gametocyte prevalence could reduce malaria transmission. After infection with P. falciparum, a variable proportion of people are found to be gametocytemic. We analyzed risk factors associated with gametocytemia at presentation and 7 days later. We enrolled 1,198 children in 2 antimalarial drug trials between September and December 1998. The children were assigned to 1 of 4 treatment groups: chloroquine only; pyrimethamine-sulfadoxine (PSD) only; PSD combined with 1 dose of artesunate; and PSD combined with 3 doses of artesunate. By the time of enrollment, 200 (17%) of 1,198 children were gametocyte carriers. Three independent risk factors were associated with gametocytemia at enrollment. Children with anemia were more likely to carry gametocytes, whereas children with fever (> 37.4 degrees C) or high parasite densities (> 100,000 parasites/microL) were less frequently gametocyte carriers. Children with at least 2 of the risk factors were 4 times more likely to be gametocytemic than children with < 2 risk factors (odds ratio [OR], 4.4; 95% confidence interval [CI], 2.7-7.1). Seven days after the start of treatment, 355 (37%) of 466 assessable children were found to be gametocyte carriers. Children treated with PSD alone had a significantly higher risk of being gametocytemic by Day 7 compared with children in the other 3 treatment groups. In the subgroup of children who had no detectable gametocytes on enrollment, the effect of treatment with PSD + 3 doses of artesunate was most marked. Nineteen (10%) of 198 children treated with PSD + 3 doses of artesunate became gametocytemic, in contrast to 184 (57%) of 321 children treated with PSD alone (OR, 12.7; 95% CI, 7.3-22.1). Early treatment with highly effective antimalarial therapy has the greatest chance of preventing gametocytemia. The choice of a first-line antimalarial drug for uncomplicated malaria should not only take into consideration the ablation asexual parasitemia but also the suppression of gametocytemia.

A clinical algorithm for the diagnosis of malaria: results of an evaluation in an area of low endemicity

We conducted a study of 1945 children and 2885 adults who presented with fever to a hospital outpatients clinic in an urban area of India order to develop and evaluate a clinical algorithm for the diagnosis of malaria. Only 139 (7%) children and 349 (12%) adults had microscopically confirmed malaria. None of the symptoms or signs elicited from the respondents were good predictors of clinical malaria. Simple scores were derived through combining clinical features which were associated with slide positivity or were judged by clinicians to be important. The best-performing algorithms were a score of 4 clinical features in children (sensitivity 60.0% and specificity 61.2%) and a score of 5 in adults (sensitivity 54.6% and specificity 57.5%). The clinical features differed and algorithm performances were poorer than in previous studies in highly endemic areas. The conclusion is that malaria diagnosis in areas of low endemicity requires microscopy to be accurate.

Community perceptions of a mass administration of an antimalarial drug combination in The Gambia

To test the hypothesis that widespread treatment with artemisinin derivatives can reduce malaria transmission, a mass drug administration (MDA) campaign was undertaken in an area of The Gambia in 1999. Coverage of 85% of the target population was achieved, but the intervention did not reduce overall malaria transmission. We studied the perceptions, knowledge and attitudes of the community to the MDA campaign. A validated questionnaire was administered to randomly selected MDA participants (n = 90) and MDA refusers (n = 71). Individuals who believed in the importance of the MDA (adjusted OR 58.3%; 95% CI 17.4-195.8) and those who were aware that a high level of participation was needed for the MDA to be successful (adjusted OR 28.1; 95% CI 10.3-75.9) were more likely to participate. Understanding that the purpose of the MDA was to reduce malaria (adjusted OR 13.9; 95% CI 5.5-35.1) and knowledge of the fact that malaria is transmitted by mosquitoes and of the clinical signs of malaria (adjusted OR 3.4; 95% CI 3.1-9.0) were associated with participation. Individuals who discussed the MDA with other villagers (adjusted OR 5.5; 95% CI 2.2-13.5) and those who attended the sensitization meeting (adjusted OR 2.6; 95% CI 1.1-6.0) were also more likely to participate. Women were significantly more likely to participate in the MDA than men (adjusted OR 3.1; 95% CI 1.5-6.2). Individuals who refused to participate were unlikely to plan participation in future MDAs. One of the most difficult challenges in the implementation of a malaria control strategy such as an MDA is to convince villagers to participate and to make them aware that a high level of participation by the community is needed for success. We found that our sensitization meetings could be improved by giving more information on how the MDA works and finding means to generate small group discussions after the meeting.

Immunologic memory 5 years after meningococcal A/C conjugate vaccination in infancy

Infant vaccination with meningococcal conjugates may provide long-term protection against disease. Antibody levels and immunologic memory were assessed in 5-year-old Gambian children who received meningococcal A/C conjugate vaccination (MenA/C) in infancy. At 2 years, they were randomized to receive a booster of MenA/C (conjugate group), meningococcal A/C polysaccharide (MPS group), or inactivated polio vaccine (IPV group). All groups were revaccinated with 10 microg MPS at 5 years of age, as were 39 previously unvaccinated age-matched control subjects. Before revaccination, titers were higher in the conjugate and MPS groups than in control subjects (P<.001); titers for the IPV group were similar to those for control subjects. Ten days after revaccination, the conjugate and IPV groups had similar serogroup C serum bactericidal antibody titers (3421 vs. 2790, respectively). These levels were significantly higher than those in the MPS (426) and control (485) groups (P<.001). Thus, immunologic memory was sustained for > or =5 years; however, MPS challenge at 2 years interfered with a subsequent memory response.

Serogroup W135 meningococcal disease in Hajj pilgrims

An outbreak of W135 meningococcal disease occurred in the spring of 2000 among pilgrims returning from Saudi Arabia and their contacts. Clinical isolates from England and France were examined and compared with reference strains from other countries. Characterisation of isolates by a range of typing methods showed them to be of clonal origin (ET-37) and closely related to other meningococci with an established propensity to cause disease clusters. A reappraisal of vaccination strategies for travellers is required.

Salivary antibody response to vaccination with meningococcal A/C polysaccharide vaccine in previously vaccinated and unvaccinated Gambian children

Development of salivary antibodies at the age of 4 or 5 years to group A and C meningococcal polysaccharides (MenA/C PS) was studied among Gambian children, who had received MenA/C conjugate or PS vaccine in infancy. There was also a control group of 64 age matched children. IgG, IgA, and secretory Ig concentrations were measured by enzyme immuno assay. MenA/C PS vaccine induced antibodies both in previously vaccinated and unvaccinated children. The previous vaccination had not induced long lasting IgA-mediated memory. IgA antibodies were secretory, and most of IgG was serum derived. The IgG salivary response seen was similar to the serum response.

Impact of asymptomatic malaria parasitaemia on cognitive function and school achievement of schoolchildren in the Yemen Republic

Asymptomatic malaria parasitaemia is prevalent among schoolchildren in many countries. The relationship between asymptomatic parasitaemia and children's cognitive functions was examined in a case control study and then in a natural experiment. A group (n = 445) of asymptomatic parasitaemic boys were compared with a group of non-parasitaemic boys (n = 142) matched for grade and school on their performance on a battery of cognitive tests. Two weeks later the parasitaemic children were re-screened and 150 children of those who remained parasitaemic were matched for grade and school with 150 children who were no longer parasitaemic. These children were then re-tested and their cognitive function compared. Initially, after controlling for age, socio-economic background and nutritional status the parasitaemic children performed worse than the non-parasitaemic children in fine motor function tests. There was no difference in change in cognitive test scores between those who became non-parasitaemic and those who remained parasitaemic. However, children who initially had the highest parasite density improved the most in 2 fine motor tests and a picture memory test. We were unable to show a benefit from losing parasitaemia over a 2 week period, but it remains possible that parasitaemia may affect cognition and longer term trials should be conducted.

Carriage of serogroup W-135, ET-37 meningococci in The Gambia: implications for immunisation policy?

We found high levels of symptomless carriage of a hyperinvasive Neisseria meningitidis strain (electrophoretic type 37 [ET-37], serogroup W-135) during a vaccine trial in Gambian children in 1996. Serogroup C, ET-37 complex meningococci cause 30-40% of meningococcal disease in countries such as the UK, and have a point prevalence of 0.5-1.0%. The recent Haj-associated spread of serogroup W-135, ET-37 complex meningococci, which has been accompanied by numerous secondary cases, might be explained by the apparently raised carriage rates reported here.

Season of birth is not associated with delayed childhood mortality in Upper River Division, The Gambia

There is evidence that season of birth may predict adult mortality from infectious diseases in rural Gambia. Using data collected over a five-year period from the rural, eastern region of the Gambia, we examined whether the season of birth influences mortality in childhood. 26 894 births and 3776 deaths among children under the age of five years were recorded in this region during the period 1989-1993. The estimated 1-4 year population was 95 355. In children aged 1-4 years, the mortality rate per 1000 per year was 16.1 (95% CI 14.9, 17.2) for those born in the 'harvest' season (January to June), which was not significantly different from the rate of 17.9 (95% CI 16.7, 19.0) recorded for those born in the 'hungry' season (July to December) (age-stratified Mantel-Haenszel mortality ratio 0. 91, 95% CI 0.83, 1.01; p = 0.08). Nearly all deaths of 1-4 year olds were attributed to infectious diseases, with malaria accounting for over 40%. None of the cause-specific child mortality rates differed significantly according to the season of birth. These data suggest that beyond infancy, when it is easier to separate the effect of season on cause of death from that of the season of birth, there is no marked difference in the rate of death between Gambian children born in the harvest season and those born in the hungry season.

Effect of fever on the serum antibody response of Gambian children to Haemophilus influenzae type b conjugate vaccine

BACKGROUND

Acute malaria is a major pediatric problem in developing countries and it is known to be immunosuppressive.

METHODS

The serum antibody response to Haemophilus influenzae type b (Hib) conjugate vaccine was investigated in children ages 12 to 30 months with fever associated with malaria, fever associated with other causes or no fever. Groups of 57 children with malaria, 57 children with fever without malaria and 60 healthy children were bled and vaccinated with a single dose of H. influenzae type b capsular polysaccharide-tetanus protein conjugate vaccine. Of these 137 were bled again 1 to 2 months after vaccination.

RESULTS

The median antibody titers at baseline were low and similar in the three groups; 77, 65 and 57% of children in the malaria, febrile and healthy groups, respectively, had prevaccination titers of anti-polyribosylribitol phosphate antibodies below 0.15 microg/ml. The median antibody titers after vaccination were 6.3, 7.5 and 23 microg/ml in the malaria, febrile and healthy groups, respectively (P < 0.001, healthy group vs. the two febrile groups). All the healthy children had protective titers (>0.15 microg/ml) after vaccination, but 11% of the children with malaria and 4% of the other febrile children did not have protective titers.

CONCLUSIONS

Anti-polyribosylribitol phosphate titers after Hib vaccination were lower in children with malaria or other febrile illnesses at the time of vaccination than in controls. Fever associated with malaria or other acute illnesses is associated with a diminished response to Hib conjugate vaccine. These findings raise questions about the vaccination of febrile children and indicate the need for further studies in this area.

HPV16 E6 oncogene variants in women with cervical intraepithelial neoplasia

Human papillomaviruses (HPVs) are strongly associated with the development of high grade cervical intraepithelial neoplasia (CIN) and cervical carcinoma, with between 40-80% of patients with cervical carcinoma being attributed to a single HPV type, HPV16 depending on the methods used and geographical location of the particular study [van den Brule et al., 1996]. An HPV16 E6 variant has been described which is strongly associated with high grade CIN [Ellis et al., 1997] and with the human leukocyte antigen (HLA)-B7 genotype in women with cervical carcinoma where HLA-B7 positive patients were demonstrated to have a significantly poorer clinical outcome [Ellis et al., 1995]. To determine whether this HPV16 E6 variant might play a significant role in the pathogenesis of cervical disease, 174 HPV16 positive women were selected from those attending the colposcopy clinics of Guy's and St Thomas' Hospital Trust following polymerase chain reaction (PCR) amplification of HPV16 L1 or E5 DNAs from cervical brush swabs or fixed biopsy tissue. HPV16 E6 DNA was amplified by PCR and the variant sequence was identified by Msp 1 restriction enzyme digestion, as the nucleotide substitution creates an additional unique Msp 1 site. The study group comprised 29 normal controls, 7 women with borderline cytology, 123 women with cervical dysplasia and 12 women with cervical cancer. 101/174 (58%) of these women had amplifiable E6 DNA and restriction enzyme digestion was performed on 95 of these. The variant E6 sequence was identified in 3/95 (3%) individuals, two of whom had normal histology and one had a CIN II lesion. Wild type E6 sequence was identified in the remaining 92/95 (97%) individuals. These data suggest that this particular E6 variant does not play a major role in the pathogenesis of HPV16 related cervical disease in women living in the South London area.

Efficacy of artesunate plus pyrimethamine-sulphadoxine for uncomplicated malaria in Gambian children: a double-blind, randomised, controlled trial

BACKGROUND

Resistance to cheap effective antimalarial drugs, especially to pyrimethaminesulphadoxine (Fansidar), is likely to have a striking impact on childhood mortality in sub-Sharan Africa. The use of artesunate (artesunic acid) [corrected] in combination with pyrimethamine-sulphadoxine may delay or prevent resistance. We investigated the efficacy, safety, and tolerability of this combined treatment.

METHODS

We did a double-blind, randomised, placebo-controlled trial in The Gambia. 600 children with acute uncomplicated Plasmodium falciparum malaria, aged 6 months to 10 years, at five health centres were randomly assigned pyrimethaminesulphadoxine (25 mg/500 mg) with placebo; pyrimethamine-sulphadoxine plus one dose of artesunate (4mg/kg bodyweight); or pyrimethamine-sulphadoxine plus one dose 4 mg/kg bodyweight artesunate daily for 3 days. Children were visited at home each day after the start of treatment until parasitaemia had cleared.

FINDINGS

The combined treatment was well tolerated. No adverse reactions attributable to treatment were recorded. By day 1, only 178 (47%) of 381 children treated with artesunate were still parasitaemic, compared with 157 (81%) of 195 children in the pyrimethamine-sulphadoxine alone group (relative risk 1.7 [95% CI 1.5-2.0], p<0.001). Treatment-failure rates at day 14 were 3.1% in the pyrimethamine sulphadoxine alone group, and 3.7% in the one-dose artesunate group (risk difference -0.6% [-4.2 to 3.0]) and 1.6% in the three-dose group (1.5 [1.5-4.5], p=0.048). Symptoms resolved faster in children who received artesunate, but there was no additional benefit for three doses of artesunate over one dose. Children given artesunate were less likely to be gametocytaemic after treatment.

INTERPRETATION

The combined treatment was safe, well tolerated, and effective. The addition of artesunate to malaria treatment regimens in Africa results in lower gametocyte rates and may lower transmission rates.

What can the residents of malaria endemic countries do to protect themselves against malaria?

The incidence of malaria may vary substantially between adjacent communities and within an individual community, even in areas of high malaria transmission. Analysis of the factors responsible for these local variations in the incidence of malaria may identify potential control measures. Factors shown to be associated with local protection against malaria in some situations include house position, house design, the use of insect repellents and mechanical barriers such as bednets and curtains. The efficacy of insecticide treated nets and curtains in preventing mortality and morbidity from malaria, at least in the short-term, has been demonstrated convincingly. However, other measures of personal protection have not been evaluated in large trials which have clinical malaria as their endpoint. Such trials are needed to see if new malaria control tools can be identified that will assist current international efforts to improve malaria control, especially in Africa. The millions of non-immune travellers who visit malaria endemic areas each year need to protect themselves against malaria and the ways in which they can do this most effectively have been studied extensively. However, less attention has been paid to the local population of malaria endemic areas. What steps can they adopt to provide personal protection against malaria and how effective are these measures? Clues to which measures might be effective can come from study of the reasons for local variations in the incidence of malaria.

Immune response to revaccination with meningococcal A and C polysaccharides in Gambian children following repeated immunisation during early childhood

Forty-two Gambian children randomised to receive two doses of meningococcal A/C polysaccharide vaccine (MPS) in infancy and either MPS (n = 15), meningococcal A/C conjugate (n = 13) or inactivated polio vaccine (IPV n = 14) at 2 years, were revaccinated with MPS at 5 years of age along with 39 matched control children. Meningococcal A and C polysaccharide antibodies were analysed by ELISA and bactericidal assay (SBA) in sera taken before and 10 days after revaccination. The geometric mean group SBA titre in the MPS group following revaccination was about half that of the unvaccinated controls (0.51 95%CI: 0.28, 0.90) for group A and less than half that of the controls for group C (0.41, 95%CI: 0.16, 1.03 P = 0.06). The group C SBA response in the conjugate group was 14-fold higher than in the MPS group (P < 0.001). Multiple doses of meningococcal polysaccharide in childhood may therefore attenuate the SBA response to both group A and group C polysaccharides. In contrast, vaccination with meningococcal A/C conjugate after MPS in infancy gives immunological memory to N. meningitidis group C.

Predicting malaria infection in Gambian children from satellite data and bed net use surveys: the importance of spatial correlation in the interpretation of results

In line with the renewed World Health Organization Global Malaria Control Strategy, we have advocated the use of satellite imagery by control services to provide environmental information for malaria stratification, monitoring, and early warning. To achieve this operationally, appropriate methodologies must be developed for integrating environmental and epidemiologic data into models that can be used by decision-makers for improved resource allocation. Using methodologies developed for the Famine Early Warning Systems and spatial statistics, we show a significant association between age related malaria infection in Gambian children and the amount of seasonal environmental greenness as measured using the normalized difference vegetation index derived from satellite data. The resulting model is used to predict changes in malaria prevalence rates in children resulting from different bed net control scenarios.

Manson Lecture. Meningococcal meningitis in Africa

This review covers the history of meningococcal meningitis in Africa since epidemics of the infection were first described around 100 years ago. It is possible that an epidemic strain of the meningococcus was introduced into West Africa from the Sudan by pilgrims returning from the Haj around the turn of the century. Since 1905 major epidemics of meningococcal meningitis have occurred in countries of the Sahel and sub-Sahel every few years, culminating in a massive epidemic in which nearly 200,000 cases were reported in 1996. Attempts to control epidemic meningococcal meningitis in Africa by vaccination with meningococcal polysaccharide vaccines have met with only modest success because epidemics can progress with great rapidity and vaccination is often started too late. This situation should be improved as a result of a recent initiative, the International Coordinating Group (ICG), which is contributing to better surveillance in countries at risk and ensuring that vaccine is available when needed. However, in the medium term, the best prospect for the control of meningococcal meningitis in Africa lies in the recent development of polysaccharide-protein conjugate vaccines which, unlike polysaccharide vaccines, are immunogenic in the very young, induce immunological memory and are likely to give long-lasting protection.

The epidemiology of pneumococcal infection in children in the developing world

Pneumonia causes about three million deaths a year in young children, nearly all of which are in developing countries. Streptococcus pneumoniae (the pneumococcus) is the most important bacterial cause of pneumonia in young children and so is likely to be responsible for a high proportion of these deaths. The pneumococcus is also responsible for a substantial proportion of the 100,000-500,000 deaths that occur from meningitis in children each year. The incidence of invasive pneumococcal disease in children in the developing world is several times higher than in industrialized countries. This discrepancy may, in part, be due to socio-economic differences but genetic factors may also play a role. Children with sickle cell disease have a substantially increased risk of invasive pneumococcal infection and a search is being made for other possible genetic risk factors. Infection with human immunodeficiency virus (HIV) also predisposes to invasive pneumococcal disease and so the incidence of this disease in young children is expected to rise as increasing numbers of African and Asian children are born with a perinatally acquired HIV infection. Until recently, pneumococcal infections could be treated effectively with penicillin, a cheap and safe antibiotic. However, pneumococci that are resistant to penicillin are becoming prevalent in many countries, necessitating a change to more costly antibiotics which may be beyond the reach of the health services of poor, developing countries. The spread of antibiotic resistance has provided an added stimulus to the development of vaccines that might be able to prevent pneumococcal disease in infants. Recently developed polysaccharide-protein conjugate vaccines show promise and are now undergoing field trials. How deployment of these vaccines will influence the balance between invasive pneumococcal infections and asymptomatic nasopharyngeal carriage of pneumococci is uncertain.

Reduction in the mean number of Plasmodium falciparum genotypes in Gambian children immunized with the malaria vaccine SPf66

SPf66, a synthetic peptide Plasmodium falciparum vaccine, did not protect young Gambian children against clinical attacks of malaria. Nevertheless, Gambian children who had been vaccinated with SPf66 and who were parasitaemic at the end of the first malaria transmission season after vaccination had significantly fewer detectable P. falciparum genotypes than control children, as determined by polymerase chain reaction analysis of 3 polymorphic loci--the msp1 block 2 repeat region, the msp2 repeat region, and the R11 region of the glutamate-rich protein gene (glurp). Geometric mean numbers of genotypes were 1.66 vs. 1.87, 1.95 vs. 2.43, and 1.21 vs. 1.50 for msp1, msp2 and glurp, respectively (P = 0.31, P = 0.04 and P < 0.01). Differences between groups became a little more marked for msp1 and msp2 when children with symptomatic malaria were excluded. No significant difference was found between parasites obtained from SPf66-vaccinated or control children in the prevalences of amino acid alleles at positions 44 and 47 in the 11 amino acid sequence of the merozoite surface protein 1 molecule, which is present in SPf66. The reduction in the number of genotypes observed could not be explained by a difference in parasite densities between SPf66-vaccinated and control children, as geometric mean parasite densities were almost identical in the 2 groups. These observations suggest that SPf66 vaccine may have induced an immune response which reduced the incidence of new infections in immunized children or accelerated the rate of clearance of parasites of individual genotypes. However, no reduction in the prevalence or density of parasitaemia was recorded in SPf66-vaccinated children, suggesting the existence of some kind of density-dependent mechanism for controlling low levels of malaria parasitaemia.

Clinical predictors of hypoxaemia in Gambian children with acute lower respiratory tract infection: prospective cohort study

OBJECTIVES

To determine clinical correlates and outcome of hypoxaemia in children admitted to hospital with an acute lower respiratory tract infection.

DESIGN

Prospective cohort study.

SETTING

Paediatric wards of the Royal Victoria Hospital and the hospital of the Medical Research Council's hospital in Banjul, the Gambia.

SUBJECTS

1072 of 42 848 children, aged 2 to 33 months, who were enrolled in a randomised trial of a Haemophilus influenzae type b vaccine in the western region of the Gambia, and who were admitted with an acute lower respiratory tract infection to two of three hospitals.

MAIN OUTCOME MEASURES

Prevalence of hypoxaemia, defined as an arterial oxygen saturation <90% recorded by pulse oximetry, and the relation between hypoxaemia and aetiological agents.

RESULTS

1072 children aged 2-33 months were enrolled. Sixty three (5.9%) had an arterial oxygen saturation <90%. A logistic regression model showed that cyanosis, a rapid respiratory rate, grunting, head nodding, an absence of a history of fever, and no spontaneous movement during examination were the best independent predictors of hypoxaemia. The presence of an inability to cry, head nodding, or a respiratory rate >/= 90 breaths/min formed the best predictors of hypoxaemia (sensitivity 70%, specificity 79%). Hypoxaemic children were five times more likely to die than non-hypoxaemic children. The presence of malaria parasitaemia had no effect on the prevalence of hypoxaemia or on its association with respiratory rate.

CONCLUSION

In children with an acute lower respiratory tract infection, simple physical signs that require minimal expertise to recognise can be used to determine oxygen therapy and to aid in screening for referral. The association between hypoxaemia and death highlights the need for early recognition of the condition and the potential benefit of treatment.