 |
|
Brief Reports Indian Pediatrics 2002; 39:743-746 |
||||||||||||||||||||||||||||||||||||||||||
|
Comparison of HemoCue Method with Cyanmethemoglobin Method for Estimation of Hemoglobin |
||||||||||||||||||||||||||||||||||||||||||
S.K. Kapoor* Umesh Kapil Sada Nand Dwivedi** K. Anand* Priyali Pathak Preeti Singh
Iron deficiency anemia is an important public health problem resulting in considerable morbidity and mortality. It is a problem of serious public health significance, given its impact on psychological and physical development, behaviour and work performance(1). Hemoglobin measurement is the primary method for anemia screening and to study the prevalence in populations(2,3). Estimate of the prevalence of anemia in turn depends upon the methods used for assessing hemoglobin concentration and the cut-off point applied(4,5). There are various methods which have been recommended for estimation of hemoglobin for assessment of anemia. Each method has its advantages and limitations. In India, the most widely used quantitative methods are Sahli’s method and Cynamethemoglobin methods(6). The HemoCue method is currently extensively used for estimating the concentration of hemoglobin in capillary blood in field situations(7). HemoCue method was used for estimation of hemoglobin in recently conducted nationwide National Family Health Survey (NFHS)-II. A pilot study conducted earlier documented that HemoCue method provided higher estimates of hemoglobin by 0.5 g than the blood cell counter (BCC) method in the venous blood samples(8). A recent study from Indonesia also reported that the HemoCue method overestimated hemoglobin against the standard Cyanmethemoglobin method(9). Concerns have been raised on the prevalence of anemia amongst children and pregnant mothers as documented through the NFHS-II survey utilizing the HemoCue method. The present study was conducted to compare the estimation of hemoglobin obtained between standard Cyanmethemoglobin and HemoCue methods. Subjects and Methods A hospital based study was conducted amongst adult patients attending the outdoor patient clinic at a rural hospital in Ballabhgarh which is the rural field practice area of All India Institute of Medical Sciences (AIIMS), New Delhi for various diseases requiring hematological investigations. The study was conducted in two phases. The subjects included in phase I and II of the study were independent of each other. A total of 151 subjects were included in Phase I of the study. Finger prick was made in each of the subjects utilizing sterilized needles. The first two blood drops were discarded. The third drop was taken for hemoglobin estimation by standard Cyanmethemoglobin method utilizing the Hemoglobin pipette. The fourth drop was taken for hemoglobin estimation by HemoCue method(9). In Phase II, 335 subjects were included. The blood was directly collected by venous puncture in EDTA vials. In each blood sample the hemoglobin estimation were conducted first by the standard Cyanmethemoglobin method and subsequently by HemoCue method as done in Phase I. For hemoglobin estimation by standard Cyanmethemoglobin method, 20 mL of blood was taken and mixed in 5 mL of Drabkin’s solution. After 4-5 hours the solution was read at 540 nm in a spectrophotometer to estimate the hemoglobin concentration(9). For estimation of Hemoglobin by HemoCue method, a new microcuvette container was opened for the estimation. The microcuvette was applied to the blood drop that filled automatically by capillary action. The cavity of microcuvette is coated with chemical reagents on its inner wall. Surplus blood on the tip of the microcuvette was wiped off by using the clean end of the sterile gauze. It was ensured that no blood is sucked out of the microcuvette when the surplus blood was dried. If there was any sign of air bubbles, the microcuvette was discarded and a new microcuvette was used. The microcuvette was placed in the holder in the HemoCue machine and then gently pushed into the photometer. The microcuvette container was resealed immediately after use. Hemoglobin level was displayed after 15-45 seconds on the screen(6). The standard universal precautions were adopted while collecting the blood samples. Results Tables I - II depict the distribution of hemoglobin levels as obtained by standard Cyanmethemoglobin and HemoCue methods in capillary and venous blood samples. It was found that for the different levels of hemoglobin i.e. 10 g/dL and more, 7 to less than 10 g/dL and < 7 g/dL, HemoCue method overestimated hemoglobin level as compared to the standard Cyanmethemoglobin method. Table I- Comparison of Hemoglobin Levels Obtained by Cyanmethemoglobin and HemoCue Methods in Capillary Blood Samples
Method 1 = Cyanmethemoglobin method; Method 2 = HemoCue method. Table II-Comparison of Hemoglobin Levels Obtained by Cyanmethemoglobin and HemoCue Methods in Venous Blood Samples
Method 1 = Cyanmethemoglobin method; Method 2 = HemoCue method HemoCue method overestimated the hemoglobin levels by 1.5 g/dL in capillary blood samples and by 1.2 g/dL in venous blood samples, as compared to the standard Cyanmethemoglobin method. Discussion The results of the present study revealed that HemoCue method estimated higher mean hemoglobin levels as compared to the standard Cyanmethemoglobin method in capillary and venous blood samples. A recent study published from Indonesia(9) has also reported similar findings. The two important factors identified in this study which influenced the hemoglobin estimations obtained by HemoCue method were (i) the time interval between opening of the microcuvette container and their use, and (ii) humid climatic conditions. The earlier study conducted at AIIMS in venous blood samples also documented that HemoCue method overestimated hemoglobin as compared to the blood cell counter method by 0.5 g/dL(8). The comparison of results between the standard Cyanmethemoglobin method and the HemoCue (HQ) method using finger prick and venous samples at the various levels of hemoglobin as observed by the standard method (< 7.0, 7-10.0 & ³ 10g/dL) reveal that HQ method significantly provides higher values in each category (P < 0.001). This analysis on the pooled results also reveals significantly higher values under the HQ method (P < 0.001). In capillary samples, the mean differences at the various levels of hemoglobin were 0.81, 1.54, and 1.55 g/dL. The mean difference for the pooled data was observed as 1.49 g/dL. On account of the fact that the first sub-group has a sample of only 11 subjects and last two subgroups reveal mean differences closer to that observed in case of total data set, it appears more appropriate to suggest a correction factor of 1.5 g/dL, which is also more practical. In venous samples, the mean differences at the various levels of hemoglobin are as 0.66, 1.09 and 1.33 g/dL. The mean difference for the pooled data was observed as 1.2 g/dL. As explained above, again it seems more appropriate to suggest a correction factor of 1.2 g/dL for the venous samples. The findings of the present study indicate that a correction factor of minus 1.5 g/dL and minus 1.2 g/dL should be applied for hemoglobin estimation obtained from capillary and venous blood samples by HemoCue method, respectively. For research studies, Cyanmethemoglobin method should be used. HemoCue method has been found to be simple in operation in field conditions hence, in areas where Cyanmethemoglobin method cannot be used, HemoCue may be used and results should be interpreted after applying the correction factor. Contributors: SKK and KA coordinated the study along with UK who also drafted the paper and will act as the guarantor of the study. SND helped in statistical analysis. PP and PS participated in data collection, analysis and helped in drafting the manuscript. Funding: None. Competing interests: None stated.
| ||||||||||||||||||||||||||||||||||||||||||
|
References | ||||||||||||||||||||||||||||||||||||||||||
|
