Comparison of five methods of malaria detection in the outpatient setting.

In eastern Africa where 90% of the malaria is due to Plasmodium falciparum, the accuracy of malaria diagnosis at the outpatient level is becoming increasingly important due to problems of drug resistance and use of alternative, costly antimalarial drugs. The quantitative buffy coat (QBC) technique, acridine orange staining with an interference filter system, and the ParaSight-F test have been introduced as alternative methods to conventional microscopy for the diagnosis of malaria. Two hundred thirteen outpatients were tested using these alternative methods and conventional microscopy by five experienced technologists; two were randomly allocated to read the results of each test. Paired results showed the highest level of agreement with the ParaSight-F test (99%), followed by Field stain (92%). The results of the QBC technique showed the least agreement (73%). Using conventional microscopy as the reference standard, the ParaSight-F test had a sensitivity range of 90-92% and a specificity of 99%, staining with acridine orange had a sensitivity range of 77-96% and a specificity range of 81-98% and the QBC technique had a sensitivity range of 88-98% and a specificity range of 58-90%. All microscopic tests showed lower sensitivities (as low as 20% using staining with acridine orange) in detecting low parasitemias (< or = 320/microl) than the ParaSight-F test (70%). Due to the high cost of the ParaSight-F test, Field-stained blood films remain the most appropriate method for diagnosis of P. falciparum in eastern Africa. The ParaSight-F test may be used in situations where no trained microscopists are available, or where malaria is strongly suspected and the results of microscopy are negative.

Prevalence of anaemia in patients attending an outpatient clinic in western Rift Valley in Kenya during a low malaria season.

OBJECTIVE: To assess the prevalence of anaemia in outpatients attending a rural health clinic in an area of seasonal malaria, during the low transmission season. METHODS: Haemoglobin estimation and blood slide examination for malaria parasites were performed on 280 consecutive patients attending outpatient curative services at Entasopia Health Centre, Kajiado District, Kenya, between April-May 1996. Anaemia was defined according to World Health Organisation guidelines for age, sex and pregnancy status. RESULTS: In all groups except adult males, more than half of the patients tested had haemoglobin values below the lower reference limits, suggesting that anaemia is widely present in this population even during the low malaria season. Only 5% of patients were positive for Plasmodium falciparum malaria. Peripheral blood film examination suggested iron deficiency as the major cause of anaemia. CONCLUSIONS: Further studies to define the underlying causes of anaemia and to develop community strategies to prevent anaemia are required. The association between fever and anaemia and the use of pallor to diagnose anaemia, are discussed.

Antibiotic resistance pattern of Vibrio cholerae and Shigella causing diarrhoea outbreaks in the eastern Africa region: 1994-1996.

Between March 1994 and December 1996, 1797 rectal swabs were transported to the AMREF laboratory from sites in six countries in the eastern Africa region: 1749 were cultured for Vibrio cholerae and 48 for Shigella/Salmonella. Culture, isolation, identification and antibiotic susceptibility testing were performed using standardized techniques. The isolates were categorized as sensitive or resistant based on standardized zones of inhibition. The rate of isolation of V. cholerae from rectal swabs increased progressively from less than 20% to more than 45% between 1994 and 1996, 80-100% of isolates of V. cholerae from Kenya and south Sudan, and 65-90% from Somalia were sensitive to tetracycline, although in 1995 isolates from Mogadishu showed only 44% sensitivity. All isolates from Tanzania and Rwanda were 100% resistant to tetracycline. In Kenya and Somalia, the percentage of isolates sensitive to chloramphenicol and cotrimoxazole reduced markedly from 85% in 1994 to < 10% in 1996. 100% of isolates from Rwanda and Tanzania were resistant to chloramphenicol and cotrimoxazole while in south Sudan > 70% of isolates were sensitive. Nalidixic acid and erythromycin retained > 75% sensitivity in all areas. Shigella dysenteriae and Shigella flexneri were recovered from dysentery specimens in northern Kenya. Both species showed similar antibiotic sensitivity patterns and were sensitive only to nalidixic acid and furazolidone. Due to variations of resistance patterns within countries in the region, antibiotic sensitivity testing should be performed at the start of an outbreak, and antibiotic use should be restricted to severe cases of V. cholerae and Shigella infection.

Evaluation of the alkaline haematin D-575 method for haemoglobin estimation in east Africa.

In many health facilities in east Africa, haemoglobin estimation is performed using visual colour comparison methods. Efforts to establish colorimetric methods face numerous constraints, including the unavailability of standards for quality control. In contrast, the alkaline haematin D-575 method for haemoglobin estimation is a colorimetric method that uses primary standards prepared from pure, crystalline chlorohaemin. There is no significant difference in the accuracy of the alkaline haematin D-575 method and that of the reference haemiglobincyanide method (P > 0.05), and the response of the alkaline haematin D-575 method is linear for serially diluted blood samples over the haemoglobin concentration range 19.6-3.3 g/dl (r = 0.994, y = 1.01 x - 0.3). The method has a precision of +/-0.3 g/dl (coefficient of variation = (1.8%) for whole blood, and is suitable for use with fixed-wavelength haemglobinometers (lambda = 565 nm) or with colorimeters at lambda = 580 nm. Stable quality control standards could be prepared at provincial, zonal, or reference laboratories and distributed regularly to outlying laboratories.