An evaluation of the HemoCue for measuring haemoglobin in field studies in Jamaica.

An evaluation of the HemoCue for measuring haemoglobin in field studies in Jamaica M. Hudson-Thomas,1 K.C. Bingham,2 & W.K. Simmons3 The HemoCue system utilizes the principle of oxidation of haemoglobin to hemiglobin by sodium nitrite and the subsequent conversion of hemiglobin to hemiglobinazide by sodium azide. The reagents for these reactions are contained within a small disposable microcuvette of approximately 10 pl in volume. A venous or capillary sample is introduced into the microcuvette by capillary action and, after reaction with the reagents, the absorbance is read in the HemoCue photometer at 565 and 880 nm. The haemo- globin concentration is then displayed as a digital reading, in either gldl or mmol/l in 15-45 seconds. We compared haemoglobin values obtained by the HemoCue system with those from the Coulter Counter S-Plus IV in 366 pregnant women in urban Jamaica, and found a highly significant correlation (r = 0.78, P <0.01). However, because of the convenience and ease of use of this instrument and con- sidering the relatively high cost, we recommend it for use only as a research tool in field studies where accurate and rapid haemoglobin determinations are required. Introduction Anaemia has been shown to be a public health prob- lem in the English-speaking Caribbean, with preva- lences averaging 27-75% and 11-75% in pregnant women and preschool-age children, respectively. Severe anaemia, with a haemoglobin concentration below 8.0 g/dl (5 mmol/l), is found in approximately 6% of pregnant and lactating women and 11% of preschool children in some of these countries (1, 2).a The problem of measuring this anaemia in the field remains a major obstacle in some Caribbean countries. This paper describes the field testing of the HemoCue system (from HemoCue AB, Helsing- borg, Sweden) and discusses its advantages and limi- tations as a research tool for determining the haemo- globin level in the region. The evaluation was done as part of a larger study designed to investigate the efficacy and side-effects of various iron formulations during pregnancy. The HemoCue system The HemoCue blood haemoglobin testing system consists of a direct reading haemoglobinometer and I Biochemist, Caribbean Food and Nutrition Institute (CFNI), P.O. Box 140, Kingston 7, Jamaica. Requests for reprints should be sent to this address. 2 Systems Analyst, University of the West Indies, Kingston, Jamaica. 3 Public Health Nutritionist, CFNI. a Simmons WK. Anaemia in the Caribbean: its prevalence and causes (unpublished document CFNI-J-10-83, 1983). Reprint No. 5493 disposable microcuvettes which contain the reagents. Each microcuvette is approximately 10pl in volume, and contains deoxycholate to haemolyse the erythro- cytes, sodium nitrite to oxidize haemoglobin to hemiglobin (methaemoglobin, Hi), and sodium azide to convert hemiglobin to hemiglobinazide (HiN3). The blood sample, which can be obtained either by venepuncture or by a finger prick, is drained into the cuvette by capillary action and mixes with the reagents. The absorbance is then read in the Hemo- Cue photometer, at 565 and 880 nm. The latter meas- urement compensates for any turbidity in the sample due to lipids, high leukocyte count, etc. The pho- tometer calculates the haemoglogin concentration in mmol/l or g/dl and displays the results as a digital reading in 15-45 seconds. All readings must be done within 10 minutes of filling the cuvette. The unit is powered either by connecting to the mains electrical supply or via a rechargeable nickel-cadmium battery pack. Materials and methods Haemoglobin values, measured on pregnant women in the health centres using the HemoCue system, were compared with those measured on the Coulter Counter S-Plus IV. Data were collected in eight maternal and child health centres in Kingston and St. Andrew, and Spanish Town, Jamaica, over a period of approximately one year by two teams, under the direction of two field supervisors. The haemoglobin estimations were carried out by members of the trained team. Bulletin of the World Health Organization, 1994, 72 (3): 423-426 © World Health Organization 1994 423 M. Hudson-Thomas et al. The criteria for inclusion of pregnant women in the larger study, under which the HemoCue system was evaluated, were: - the mother should be attending the antenatal clin- ic for the first time; - gestation should be between 14 and 22 weeks; - the mother's age should be between 16 and 35 years; - she should not currently be on iron supplements; - haemoglobin as measured by the HemoCue system should fall in the range of 8.0-12.0 g/dl. From women who fulfilled these criteria on entry, 3 ml of venous blood was collected in freeze- dried EDTA and kept on ice, to be later transferred to the Iron Monitoring Laboratory at the Caribbean Food and Nutrition Institute, Kingston, Jamaica. From here all samples were taken to the Sickle Cell Research Laboratory where haemoglobin and other haematological determinations were performed on the Coulter Counter. As far as possible, all readings were done within twenty four hours of the venepunc- ture. The two sets of haemoglobin measurements taken on entry to the antenatal clinics were com- pared. Results Results were obtained for 366 women. A correlated sample t-test yielded a small difference in means (t = 4.97, P <0.01, 365 df) with slightly wider stan- dard deviations from the Coulter Counter (Table 1). The two sets of haemoglobins correlated highly (r = 0.78, P <0.01, n = 366) and the scattergram is shown in Fig. 1. Discussion The measurement of haemoglobin constitutes the first step in the investigation of anaemia. In clinical laboratories, the cyanomethaemoglobin method remains one of the most widely used for its determination (3), but fully automated systems such as the Coulter S- Plus III or the Coulter S-880 have greatly enhanced the ease of determination. Table 1: Mean and standard deviation values for haemoglobin from pregnant women No. of women Mean (g/dl) S.D. HemoCue 366 10.3 0.84 Coulter Counter 366 10.1 1.03 Fig. 1. Scattergram of haemoglobin values (g/dl) in 366 pregnant women, as determined by HemoCue and by Coulter Counter. 14- 12- * 0 310 _____ 8- - r*_ 0 to.~~~~~ 6 6- I v- | I ~I 6 7 8 9 10 11 12 13 14 HemoCue In Third World countries, where Coulter Coun- ters, spectrophotometers and other sophisticated laboratory equipment are not always available, there needs to be simple, inexpensive methods that can reliably determine haemoglobin in field studies as well as in the health centres. Many different methods have been used with some degree of success in the English-speaking Caribbean. These include the Tall- quist, copper sulfate, Spencer and Sahli (4) tech- niques. At present, the Spencer method is the most widely accepted. However, in one island with an exceptionally heavy clinic load (one or two nurses or midwives with 60-70 patients), it is felt that haemo- lysing the blood samples is far too time-consuming and puts excessive pressure on the limited staff. The correlation reported here is slightly lower than those reported by other investigators who com- pared the HemoCue with the Coulter Counter and the cyanomethaemoglobin methods (5, 6).] This may be explained by the fact that we selected a limited range of haemoglobin values (8-12 g/dl), whereas other researchers considered a wider range. These other studies were performed under controlled labor- b Bridges N, Parvin RM, Van Assendelft OW. Evaluation of a new system for haemoglobin measurement. Reprinted from American Clinical Products Review, 1987. 424 WHO Bulletin OMS. Vol 72 1994 HemoCue for measuring haemoglobin atory conditions, but our venepunctures and haemo- globin determinations on the HemoCue were carried out in the health centres and we estimated each hae- moglobin on the Coulter Counter within 24 hours of blood withdrawal. It is unlikely that this delay would have affected the Coulter values since samples were stored at 4 'C. The HemoCue was found to be convenient and easy to use by field workers in our study. This is in agreement with conclusions from another field study conducted in Peru where the HemoCue was found to be simple to use even by unskilled workers (7). In fact, when we introduced the HemoCue in another Caribbean country for use in an island-wide anaemia survey, the enthusiasm for the instrument was great. In this study the HemoCue was tested by field workers especially trained in its use, and not by reg- ular clinic staff under normal clinic conditions. Because of this, coupled with the relatively high cost of the instrument (US$ 425) and the cuvettes (US$ 0.80 each), we are unable to recommend its use routinely in our clinics. We do, however, recommend its use as a research tool in field studies where an accurate and rapid measurement of the haemogobin level is needed. Acknowledgements We thank Mrs Maria Jackson and Mrs Jean Jackson who worked as field supervisors on the parent project, Dr Jim Cook and Dr Sam Kahn who served as advisers, Dr Graham Serjeant and his staff at the Sickle Cell Re- search Laboratory, and Miss Karlene Russell for secretarial assistance. The parent project was funded by the Inter- national Center for Research on Women, USAID, and AID cooperative agreement DAN-5115-A-00-7908-00. Resume Evaluation de l'HemoCue pour le dosage de I'hdmoglobine lors d'6tudes sur le terrain c la Jamaique L'anemie est depuis longtemps un probleme majeur de sante publique dans les Caraibes anglophones. Une an6mie severe (h6moglobine <8,0 g/dl) s'observe chez environ 6% des femmes enceintes et allaitantes et, dans certains pays, chez 11% des enfants d'age pr6scolaire. La diffi- culte de mesurer exactement l'hemoglobine sur le terrain constitue un obstacle dans certains de ces pays. Nous avons 6valu6 le systeme HemoCue (produit par HemoCue AB, Helsingborg, Suede) comme moyen de proc6der a un dosage rapide et exact de l'hemoglobine lors d'etudes sur le terrain dans les pays en developpement. Le systeme HemoCue utilise le principe de l'oxydation de l'hemoglobine en m6themoglobine par le nitrite de sodium, puis la conversion de la m6themoglobine en azoture de methemoglobine par I'azoture de sodium. Les reactifs sont conte- nus dans une microcuve jetable d'environ 10 pl de volume. Un 6chantillon de sang veineux ou capil- laire est introduit dans la microcuve par capillarit6 et, apres mise en pr6sence des reactifs, I'absor- bance est lue sur le photometre a 565 et 880 nm. La concentration en hemoglobine est alors affi- chee au bout de 15 a 45 secondes, soit en g/dl soit en mmol/dl. Nous avons compar6 les valeurs de I'hemo- globine obtenue par le systeme HemoCue avec les valeurs obtenues sur Coulter Counter S-Plus IV chez 366 femmes enceintes provenant de zones urbaines de la Jamaique. Un prelevement de sang veineux a ete recueilli sur EDTA lyophili- se chez des consultantes de dispensaires de sant6 maternelle et infantile situ6s dans un rayon d'une vingtaine de kilometres du laboratoire de recherche. Le dosage de l'hemoglobine a Wte r6a- lise au dispensaire avec le systeme HemoCue, et le reste de l'echantillon a et maintenu sur de la glace et exp6di6 au laboratoire de recherche pour dosage de l'hemoglobine sur Coulter Counter S- Plus IV. Cette derniere d6termination a et6 effec- tuee dans les 24 heures suivant la mesure sur HemoCue. On a observe une corr6lation haute- ment significative (r= 0,78, P<0,01) entre les deux s6ries de mesures. Cette corr6lation 6levee, associ6e a la com- modit6 et a la facilit6 d'emploi de l'HemoCue, autorise a recommander son emploi comme outil de recherche lors d'6tudes sur le terrain ou il est n6cessaire d'obtenir une mesure exacte et rapide de l'h6moglobine. Nous ne pouvons toutefois le recommander pour l'utilisation en routine en rai- son de son coOt relativement 6leve (US$ 425 pour l'appareil et US$ 0,80 pour chaque microcuve). References 1. Simmons WK et al. A survey of the anaemia status of pre-school age children and pregnant and lacta- ting women in Jamaica. American journal of clinical nutrition, 1982, 35: 319-326. 2. Simmons WK, Gurney JM Nutritional anaemia in the English-speaking Caribbean and Suriname. Amer- ican journal of clinical nutrition, 1982, 35: 327-337. 3. Hainline A jr. Haemoglobin. In: Seligson D, ed. Standards methods of clinical chemistry, Vol. 2. New York, Academic Press, 1958: 45-51. WHO Bulletin OMS. Vol 72 1994 425 M. Hudson-Thomas et al. 4. Stone JE et al. An evaluation of methods of screening for anaemia. Bulletin of the World Health Organization, 1984, 62: 115-120. 5. Von Schenck H, Falkensson M, Lundberg B. Evaluation of "HemoCue", a new device for deter- mining haemoglobin. Clinical chemistry, 1986, 32: 526-529. 6. Cohen AR, Seidl-Friedman J. HemoCue system for haemoglobin measurement: evaluation in anaem- ic and non-anaemic children. American journal of clinical pathology, 1988, 90: 302-305. 7. Johns WL, Lewis SM. Primary health screening by haemoglobinometry in a tropical community. Bulletin of the World Health Organization, 1989, 67: 627-633. 426 WHO Bulletin OMS. Vol 72 1994