|Year : 2019 | Volume
| Issue : 8 | Page : 1078-1082
Assessing agreement of hemoglobin and three- fold conversion of hematocrit as methods for detecting anemia in children living in malaria-endemic areas of Calabar, Nigeria
OA Oduwole1, S Ameh2, EB Esu3, CM Oringanje4, JT Meremikwu5, MM Meremikwu6
1 Department of Medical Laboratory Science, College of Natural and Applied Sciences, Achievers University, Owo, Ondo State, Nigeria
2 Department of Community Medicine, Faculty of Medicine, College of Medical Sciences, University of Calabar, Calabar, Nigeria; Department of Global Health and Population, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, United States of America
3 Department of Public Health, College of Medical Sciences, University of Calabar, Calabar, Nigeria
4 Department of Entomology, University of Arizona, Tucson, Arizona, United States of America
5 Institute of Tropical Diseases Research and Prevention, University of Calabar Teaching Hospital, Calabar, Nigeria
6 Department of Paediatrics, College of Medical Sciences, University of Calabar, Calabar, Nigeria
|Date of Acceptance||27-May-2019|
|Date of Web Publication||14-Aug-2019|
Dr. O A Oduwole
Department of Medical Laboratory Science, Achievers' University, Owo, Ondo State
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Background: One of the major causes of anemia, defined as the reduction in the level of hemoglobin or red blood cells (RBCs) in the blood, in children in sub-Saharan Africa is malaria. Anemia is diagnosed by using either the hematocrit method or by measuring the hemoglobin concentration. Aims: To evaluate the relationship and agreement between hemoglobin and three-fold conversion of hematocrit results of participants in a clinical trial. Materials and Methods: This is a cross-sectional study that obtained data from a multi-center clinical trial that took place from 2007 to 2008 in public health facilities in Calabar, Nigeria. The hemoglobin and hematocrit results of 494 children who had ≥2000 parasite density recruited were pooled to evaluate the relationship and agreement between the two methods. The difference between the measures against the mean of the two measures was plotted according to the theory of Bland and Altman. Results: The mean age of the children was 34 months, with approximately equal number of boys and girls. The measured hemoglobin was lower than the calculated hemoglobin in 84.5% of the children. The result showed that lower the hemoglobin concentration, the higher the chances that the three-fold hematocrit conversion overestimates hemoglobin levels in the participants. Conclusions: The three-fold hematocrit conversion of hemoglobin estimation is a less reliable method than the measured hemoglobin in anemic children in the study setting.
Keywords: Anemia, hematocrit, hemoglobin, malaria, packed cell volume
|How to cite this article:|
Oduwole O A, Ameh S, Esu E B, Oringanje C M, Meremikwu J T, Meremikwu M M. Assessing agreement of hemoglobin and three- fold conversion of hematocrit as methods for detecting anemia in children living in malaria-endemic areas of Calabar, Nigeria. Niger J Clin Pract 2019;22:1078-82
|How to cite this URL:|
Oduwole O A, Ameh S, Esu E B, Oringanje C M, Meremikwu J T, Meremikwu M M. Assessing agreement of hemoglobin and three- fold conversion of hematocrit as methods for detecting anemia in children living in malaria-endemic areas of Calabar, Nigeria. Niger J Clin Pract [serial online] 2019 [cited 2019 Aug 24];22:1078-82. Available from: http://www.njcponline.com/text.asp?2019/22/8/1078/264421
| Introduction|| |
Anemia is defined as a reduction in the level of hemoglobin or red blood cells (RBCs) in the blood. This reduction may be owing to excessive blood loss from trauma, breakdown of the red blood cells due to parasitic diseases or hemoglobinopathies, and from malnutrition.
One of the major causes of anemia in sub-Saharan is malaria. It occurs as a result of lysis of uninfected and malaria-infected RBCs, increase splenic sequestration of RBCs, and transient reduction in erythropoiesis attributed to effect malaria infection on the bone marrow., In addition, some of the other factors responsible for anemia, especially among children are malnutrition, helminthiasis, and sickle cell disease.
Anemia can be diagnosed by using either the hematocrit method, also known as packed cell volume (PCV), or by measuring the haemoglobin concentration.,, The two measurements are often used inter-changeably for determining anemia either for surveys or clinical trials.
It has been argued that using these two methods interchangeably may give false hemoglobin values., This is as a result of an overestimation of the hemoglobin concentration by the hematocrit method, which in turn leads to an underestimation of the prevalence of anemia. Haemo-concentration and increased erythrocyte rigidity during malaria infection can also alter the accuracy of the three-fold conversion method of hemoglobin determination.
Owing to inadequate resources and the need to give prompt treatment to children living in malaria-endemic areas, most of which are resource-limited, the hemoglobin concentration is derived by dividing the hematocrit value by three. This conversion method is refered to as the three-fold conversion.
There are concerns about the three-fold conversion method as a means of determining if a person is anemic or not due to lack of confidence in the hematocrit. In Nigeria, it is common for clinicians to divide hematocrit by three to derive hemoglobin value. This is largely due to a lack of point of care hemoglobinometer and the burden of the cost of these tests on the patients who are mostly poor. It is important to determine the accuracy of the three-fold conversion method for the measurement of hemoglobin particularly in Nigeria, a malaria endemic area. This is because anemia, in addition to parasite density, is used to monitor patients' response to antimalarial treatment both clinically and programmatically.
| Aims|| |
The aim of this study was to evaluate the agreement between hemoglobin and hematocrit results of participants in a clinical trial.
| Materials and Methods|| |
Compliance with ethical standard
Ethical approval for the original study  was obtained from the Ethical Review Committee of the Cross River State Ministry of Health and Ethics committee of the University of Calabar Teaching Hospital.
Informed consent was obtained from participants in the original study.
The study took place from November 2007 to December 2008 (a period covering both the wet and dry seasons) in public health facilities in Calabar, South–East, Nigeria.
This was a cross-sectional study that obtained data from two sites in Cross River State that were part of a multi-center clinical trial. (ClinicalTrials.gov NCT00393679; Pan African Clinical Trials Registry PACTR2009010000911750). The first health facility was the Ikot Ansa Primary Health Center, and the second was the Pamol clinic located in a rubber plantation. However, all sites were located in the semi-urban community with holoendemic P. falciparum malaria transmission areas, hence their suitability for the trials. The results of hemoglobin and hematocrit of 494 children under 5 years of age who had ≥2000 parasite density recruited in a multi-center clinical trial  were used to evaluate the accuracy of the three-fold conversion method as described by Rodríguez-Morales et al.
The definition of different levels of anemia in this study is according to the following cut-off points [Table 1] recommended by the WHO for children aged six to 59 months.
In the original study, children with the following criteria were included 6–59 months old with suspected uncomplicated clinical malaria attending the health facility where the study was carried out with the following, Weight >5 kg; mono-infection with P. falciparum, and a parasitemia of 2000–200,000 asexual parasites per ul, fever >37.5 degree, or history of fever in the preceding 24 h, hemoglobin value >7.0d/l, and signed informed consent by the parents or guardian.
Participants were excluded if they have at least one of the following criteria: Participation in any other investigational drug study (antimalarial or others) during the previous 30 days, known hypersensitivity to the study drugs, danger signs: not able to drink or breast-feed, vomiting (>twice in 24 h), recent history of convulsions (>1 in 24 h), unconscious state, and unable to sit or stand. Others were the presence of intercurrent illness or any condition (cardiac, renal, and hepatic diseases), which would place the subject at undue risk or interfere with the results of the study, including known G6PD deficiency severe malnutrition (defined as weight for height <70% of the median NCHS/WHO reference) and ongoing prophylaxis with drugs having antimalarial activity such as cotrimoxazole for the prevention of Pneumocisticarini pneumonia in children born to HIV+ women. Children infected with other malaria species and those with mixed malaria infection.
Children 6–59 months old with suspected uncomplicated clinical malaria attending the health facility were recruited into the study and randomized to one of the study treatments if they fulfilled the inclusion criteria. The details of the trial are described elsewhere.
Sample collection and laboratory procedure
Samples for hematocrit were obtained simultaneously with that of malaria microscopy. Hematocrit measurement for each participant was obtained by centrifuging sample collected in heparinized capillary tube using Hawksley micro-hematocrit tube and centrifuge (Hawksley and Sons Ltd, Sussex, UK) at 11,000 revolutions per minute (rpm) for 5 min. Venous blood samples were put into bottles containing EDTA to determine hemoglobin and other hematologicalparameters. Hemoglobin concentration was measured with hemophotometry (ERMA INC. particle counter, model PCE-210).
Statistical analysis used
We used the method described by Rodríguez-Morales  according to the theory of Bland and Altman, which states that “two methods designed to measure the same thing will inevitably give a positive linear correlation, and the most useful comparison is gained by plotting the difference between the measures against the mean of the two measures.” Thus, the hemoglobin (g/dl) measurement and hematocrit (%) divided by three was compared on the same scale (grams of hemoglobin per deciliter) using this method. The difference between actual hemoglobin measurement and hematocrit divided by three (Hb - Hct/3) and their average (Hb + Hct/3)/2) for each participant were calculated. The value obtained was used to plot the linear regression curve to determine the correlation between the two methods and accuracy of the three-fold conversion (Hct/3) for hemoglobin estimation using the Bland-Altman method. The 95% limits of agreement were calculated as +/– 1.96 standard deviations.
| Results|| |
This study obtained the actual hemoglobin and hematocrit measurement of 494 children with malaria who were included in a clinical trial.
[Table 2] shows the mean age of the children was 34 months, with approximately equal number of boys and girls. The mean hematocrit was 28.5%. The mean measured hemoglobin and calculated hemoglobin (Hct/3) were 9.1 g/dl and 9.5 g/dl, respectively.
|Table 2: Demographic characteristics and hematologic parameters among children under 5 years in Calabar, 2010|
Click here to view
[Table 3] shows the measured hemoglobin was lower than the calculated hemoglobin in 84.5% of the children. There were no statistically significant differences between measured hemoglobin and calculated hemoglobin with respect to anaemia status.
|Table 3: Anemia status by measured and calculated hemoglobin in the study population|
Click here to view
[Figure 1] shows a statistically significant direct correlation between measured hemoglobin and calculated hemoglobin (Hct/3) (r = 0.326, P < 0.001), both were expressed as g/dl.
|Figure 1: Scatter plot of the correlation between measured hemoglobin and calculated hemoglobin (Hct/3) in Calabar|
Click here to view
The correlation matrix in [Figure 2] shows a statistically significant direct correlation between the difference in measured and calculated hemoglobin (Hct/3) and their average values measured in g/dl in both axes (r = 0.326, P < 0.001).
|Figure 2: Scatter plot of difference against the average of measured hemoglobin and calculated hemoglobin in Calabar|
Click here to view
The Bland and Altman plot in [Figure 3] shows the differences between measured and calculated hemoglobin. The mean difference differed significantly from the null value of 0 (-0.424, 95% CI: –0.465 to –0.383); hence, a systematic difference between the measured and calculated hemoglobin. There was a direct correlation between the differences and mean of measured and calculated hemoglobin (r = 0.269, P < 0,001), with the limits of agreement ranging from –1.346 to 0.49.
|Figure 3: Bland and Altman graph of the difference in measured and calculated hemoglobin (Hct/3), by their average values in Calabar|
Click here to view
[Figure 4] shows the receiver operation characteristic curve of hematocrit and diagnosis of anemia in children under 5 years, adjusted for age and sex. The area under the curve is 66%; hence, the hematocrit value below 35% correctly identifies 66% of children with hemoglobin less than 11.0 g/dl.
|Figure 4: ROC curve of hematocrit and diagnosis of anemia in children under 5 years, adjusted for age and sex in Calabar|
Click here to view
| Discussion|| |
The results showed a correlation between hemoglobin and hematocrit/3 and lower mean hemoglobin than mean hematocrit/3. A plot of the difference between hemoglobin and hematocrit/3 against the mean of the two measures showed a systematic difference. These results support Bland and Altman's argument that the most useful test of agreement between two methods designed to measure the same thing is to plot the difference between the measures against the mean of the measures. Hence, a test of the relationship between two methods measuring the same thing inevitably shows a correlation, which is not necessarily indicative of agreement.
One of the probable reasons for the systematic difference in the plot of the difference between hemoglobin and hematocrit/3 against the mean of these same measures could be due to hemoconcentration in anemia, which could result in overestimation of hematocrit measurement,, often attributable to dehydration in an underlying infection such as malaria. Hence, malaria endemicity may have accounted for Bland and Altman's phenomenon of a systematic difference between hemoglobin and hematocrit/3 measurements in our study conducted among children in Calabar, a malaria-endemic region.
However, a study in Mato Grosso, Brazil showed there was no difference in hemoglobin and hematocrit/3 methods. This may be owing to the fact that Mato Grosso is not a malaria-endemic region.
This result also showed that lower the hemoglobin concentration, the higher the chances that the three-fold hematocrit conversion overestimates hemoglobin levels, this finding is similar to studies from other malaria-endemic regions., The implication of this is that the three-fold hematocrit conversion is misleading and a less reliable method of estimating hemoglobin levels in children with malaria, especially in malaria endemic regions.
This study concludes that the three-fold hematocrit conversion of hemoglobin estimation is a less reliable method than the measured hemoglobin in anemic children in the study setting.
Study strength and limitation
The main strength of this study was the use of data in children infected with malaria to assess the correlation and agreement between hemoglobin and hematocrit/3 measurements in assessing anemia in children. The limitation of this research was that we could not assess the agreement between the two measurements after parasite clearance or on the follow-up days (Day 7 and Day 14) post-treatment because data were only available for the hemoglobin measurements at baseline.
Friday Odey, Vivian Asiegbu, Nnanke Eno for assisting with data collection during the original trial.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
WHO. Haemoglobin concentrations for the diagnosis of anaemia and assessment of severity Vitamin and Mineral Nutrition Information System [Internet]. 2011 06 August 2018.
Perkins DJ, Were T, Davenport GC, Kempaiah P, Hittner JB, Ong'echa JM. Severe malarial anemia: Innate immunity and pathogenesis. Int J Biol Sci 2011;7:1427-42.
Zwang J, D'Alessandro U, Ndiaye J-L, Djimdé AA, Dorsey G, Mårtensson AA, et al
. Haemoglobin changes and risk of anaemia following treatment for uncomplicated falciparum malaria in sub-Saharan Africa. BMC Infect Dis 2017;17:443.
Ehrhardt S, Burchard GD, Mantel C, Cramer JP, Kaiser S, Kubo M, et al
. Malaria, anemia, and malnutrition in African children—defining intervention priorities. JInfect Dis 2006;194:108-14.
Billett HH. Hemoglobin and hematocrit. 1990.
Lee SJ, Stepniewska K, Anstey N, Ashley E, Barnes K, Binh TQ, et al
. The relationship between the haemoglobin concentration and the haematocrit in Plasmodium falciparum malaria. MalarJ 2008;7:149.
Skikne BS, Flowers CH, Cook JD. Serum transferrin receptor: A quantitative measure of tissue iron deficiency. Blood 1990;75:1870-6.
Carneiro IA, Drakeley CJ, Owusu-Agyei S, Mmbando B, Chandramohan D. Haemoglobin and haematocrit: Is the threefold conversion valid for assessing anaemia in malaria-endemic settings? MalarJ 2007;6:67.
Rodríguez-Morales AJ, Sánchez E, Arria M, Vargas M, Piccolo C, Colina R, et al
. Haemoglobin and haematocrit: The threefold conversion is also non valid for assessing anaemia in Plasmodium vivax malaria-endemic settings. MalarJ 2007;6:166.
Quintó L, Aponte JJ, Menéndez C, Sacarlal J, Aide P, Espasa M, et al
. Relationship between haemoglobin and haematocrit in the definition of anaemia. Trop Med Int Health 2006;11:1295-302.
Lewis SM, Barbara JB, Bates I. Dacie and Lewis Practical Haematology. Dacie Lewis Practical Haematology, Churchill Livingstone/Elsevier; 2006.p. 722.
McNerney R. Diagnostics for developing countries. Diagnostics 2015;5:200-9.
Meremikwu M, Odey F, Oringanje C, Oyo-Ita A, Udoh E, Eyong K, et al.
Effectiveness of a 6-dose regimen of Artemether-Lumefantrine for unsupervised treatment of uncomplicated childhood malaria in Calabar, Nigeria. Niger J Paediatr 2013;40:145-9.
The Four Artemisinin-Based Combinations Study G. A Head-to-Head Comparison of Four Artemisinin-Based Combinations for Treating Uncomplicated Malaria in African Children: A Randomized Trial. PLoS Medicine 2011;8:e1001119.
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10.
Schellenberg D, Menendez C, Kahigwa E, Font F, Galindo C, Acosta C, et al
. African children with malaria in an area of intense Plasmodium falciparum transmission: Features on admission to the hospital and risk factors for death. AmJTrop MedHyg 1999;61:431-8.
Wilson D. Clinical Veterinary Advisor-E-Book: The Horse. Elsevier Health Sciences; 2010.
Brunken GS, França GVAd, Luiz RR, Szarfarc SC. Agreement assessment between hemoglobin and hematocrit to detect anemia prevalence in children less than 5 years old. Cad Saúde Colet 2016;24:118-23.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3]