|
ORIGINAL ARTICLE |
 |
|
|
| Ahead of print
publication |
|
| Red blood cell indices in different hemoglobinopathies: A cross-sectional study in eastern India |
|
|
Ayandip Nandi1, Manas Talukdar1, Saikat Bhattacharya2, Susruta Sen3, Supratik Biswas4, Kuntal Roy5
1 Department of Pathology, Medical College and Hospital, Kolkata, West Bengal, India
2 Department of Community Medicine, NRS Medical College, Kolkata, West Bengal, India
3 Department of Laboratory Medicine, CK Birla Hospital, Kolkata, West Bengal, India
4 Department of Biochemistry, One Diagnostic Laboratory Kolkata, West Bengal, India
5 Department of Pathology, One Diagnostic Laboratory Kolkata, West Bengal, India
Click here for correspondence address and email
| Date of Submission | 01-Nov-2021 |
| Date of Decision | 20-May-2022 |
| Date of Acceptance | 23-May-2022 |
| Date of Web Publication | 18-Oct-2022 |
|
|
 Abstract | | |
Introduction: Beta thalassemia and hemoglobin (HbE)-related hemoglobinopathies are common public health problems in developing countries. High-performance liquid chromatography (HPLC) is currently the diagnostic test of choice for carrier detection, but it is costly. Hence, some initial screening and complementary tests are required, which can be affordable. Aims: To find out the distribution of different red blood cell (RBC) indices in beta thalassemia trait (BTT) and HbE-related hemoglobinopathies and to determine their significance as screening tests to distinguish between these hemoglobinopathies. Study Settings and Design: This observational cross-sectional study has been carried out at an NABL (National Accreditation Board for Testing and Calibration Laboratories)-accredited Laboratory of Eastern India with approval from the concerned Institutional Ethics Committee from January 2021 to March 2021. Methods and Material: HPLC tests and complete hemograms were performed on 2247 ethylenediaminetetraacetic acid anti-coagulated blood samples over 3 months. Patients <1 year of age or having a history of blood transfusion within the past 06 months were excluded. Statistical Analysis: One-way analysis of variance along with Bonferroni post-hoc test was performed to find out significant differences of means of mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), hemoglobin%, red blood cell (RBC) count, and red cell distribution width (RDW-CV) among concerned hemoglobinopathies. Results: The results show a significant difference of total RBC count, RDW, MCV, MCH, and MCHC between BTT and E-trait. No significant difference of mean was found between HbE homozygous and E-beta. E-trait differs from both HbE homozygous and E-beta significantly in three parameters, namely, RDW, MCV and MCH. A value of MCV at ≤73.8 fl and MCH at ≤21.9 pg may be a clue of diagnosis for BTT rather than E-trait with >90% sensitivity and >80% specificity. Conclusion: RBC indices vary significantly between BTT and other HbE-related hemoglobinopathies. They can specially be utilized to differentiate BTT and E-trait as supportive tests in addition to the gold standard test of HPLC.
Keywords: Erythrocyte indices, hemoglobinopathies, thalassemia
How to cite this URL:
Nandi A, Talukdar M, Bhattacharya S, Sen S, Biswas S, Roy K. Red blood cell indices in different hemoglobinopathies: A cross-sectional study in eastern India. Indian J Pathol Microbiol [Epub ahead of print] [cited 2023 Sep 30]. Available from: https://www.ijpmonline.org/preprintarticle.asp?id=358846 |
| Introduction | |  |
Hemoglobinopathies and thalassaemia are genetic disorders creating a major public health problem throughout the world including the South East Asian region.[1],[2] Among them, beta thalassemia is considered to be the most common, followed by hemoglobin E (HbE)-related disorders.[3]
Detection of carriers by high-performance liquid chromatography (HPLC) is the method of choice to control symptomatic hemogloinopathies, but being an expensive test for developing countries such as India, some cost-effective screening method is necessary. Previous studies mainly from northern India have already found significant correlation between red blood cell (RBC) indices and hemoglobinopathies.[4] As an automated hematology analyzer is widely available in most parts of the country, RBC indices and count may be evaluated as an initial screening method to complement HPLC findings.
The current study is aimed primarily to find out the distribution of different RBC indices in beta thalassemia trait (BTT) and HbE-related hemoglobinopathies and also to determine their significance to distinguish between these hemoglobinopathies.
| Methodology | | |
This observational cross-sectional study has been carried out at an NABL (National Accreditation Board for Testing and Calibration Laboratories)-accredited Laboratory of Eastern India with approval from the concerned Institutional Ethics Committee from January 2021 to March 2021.
Sample size was calculated using the formula for sensitivity and specificity testing. For calculation purposes, the prevalence of BTT was considered as its prevalence is lower than the prevalence of E-trait. A prevalence of 3.6%, a type I error of 0.05, a power of 80%, null hypothesis values (H0) for sensitivity and specificity of 50%, and alternative hypothesis values (H1) of 80% for sensitivity and 60% for specificity were considered values for calculation.[5] As the aim of study is to find out a suitable screening test, sensitivity was given priority compared to specificity. After calculation, the sample size for BTT was found to be 556 (NCSS-PASS 11 software). As we are interested in four hemoglobinopathies, namely, BTT, E-trait, E-homozygous (EE), and E-beta thalassemia (EB), this calculated sample size was multiplied by 4, and the calculated final minimum sample size was 2224.
Patients >1 year of age who were referred by their consultant physician to the laboratory for a complete hemogram and screening of abnormal hemoglobinopathy by the HPLC method were included, whereas those having hematological malignancy or on chemotherapy were excluded. Patients who had a history of blood transfusion within the past 06 months were also not included.
After obtaining informed consent, blood samples were collected in the ethylenediaminetetraacetic acid anti-coagulant and were run using a Horiba YUMIZEN H500 6 part hematology auto-analyzer for a complete hemogram. The HPLC test was performed by VARIANT 2 BETA THALASSEMIA SHORT PROGRAM (BIO RAD Laboratories, Hercules, CA, USA) within 3 hours of blood collection. With every batch of samples in BIO-RAD HPLC, two levels of controls (normal: HbF 1–2%, HbA2 1.8–3.2% and abnormal: HbF 5–10%, HbA2 4–6%) and the Hb A2/F calibrator were run. The RBC morphology was evaluated for correlation by examining the blood films stained by Leishman's stain under a binocular light microscope (Olympus CH20i).
HbA2 is probably the most problematic parameter in HPLC because it runs in a continuous spectrum between normal (AA), borderline-elevated HbA2 (3.6–4.5%), BTT (HbA2 4.6–10%), Hb lepore (11–15%), HbE trait (15–40%), HbE homozygous (HbA2 ≥70%), or E-beta, that is, high HbA2 (50–80%) with high HbF (15–50%) [BIORAD variant 2 kit insert]. Because of this significant overlap and borderline areas, alongside the hemogram and peripheral blood pictures, the transfusion history becomes necessary to distinguish between them.[6] Hence, in this study, we emphasized to find any relation between RBC indices and hemoglobinopathies with elevated A2 values. HbE trait, BTT, EBeta, and HbE homozygous (EE) are the four types of hemoglobinopathies that were considered for analysis for the purpose of the present study.
The results were tabulated, and data were analyzed in statistical packages for social sciences (SPSS) software version 11. Data are represented descriptively as mean for central tendency and standard deviation and the range for measures of dispersion. One-way analysis of variance (ANOVA) along with Bonferroni post-hoc test was performed to find out the significant differences of means of mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), hemoglobin (Hb)%, RBC count, and red cell distribution width (RDW-CV) among the above-mentioned hemoglobinopathies. Significance was considered at 95% confidence level (p < 0.05). A receiver operating characteristic (ROC) curve was drawn for all these blood parameters taking the HPLC test result as the gold standard. For the curves occupying more than 50% of the total area, a cutoff value was determined for each parameter which is optimum for sensitivity and specificity.
| Result | | |
A total of 2247 samples were run for HPLC in the study period, of which 2033 were of a normal pattern. Twelve abnormal variants such as beta thalassemia major, sickle cell trait, sickle cell homozygous, SD thalassemia, HbQ trait, and HbD traits were found. The rest 202 samples were of HbE trait (AE), BTT, E Beta (EB), and HbE homozygous (EE), and these subjects were included in the study [Figure 1]. Among these 202 subjects, patients suffering from BTT and HbE trait were 90 each. Fifteen subjects were suffering from HbE homozygous, and seven were suffering from E-beta. | Figure 1: HPLC graph showing (a) HbE homozygous, (b) HbE trait, (c) Beta thalassemia trait, and (d) E-beta thalassemia
Click here to view |
[Table 1] shows the mean of the blood indices for all four groups along with minimum and maximum values and standard deviation. From this, it can be concluded that the mean hemoglobin is the highest for HbE trait (10.55 g/dl) and the minimum for E-beta (7.62 g/dl). Therefore, clinically, most HbE traits will be asymptomatic and E-beta will have the highest degree of anemia. The standard deviation is similar for all four groups with E-beta being the minimum (1.11%). In the case of total RBC count, the highest count is for BTT (4.81 million/μl) and the lowest is again for E-beta (3.78 million/μl). This is also clinically apparent as all BTT shows compensatory erythroid hyperplasia. The mean RDW-CV is the highest among the E-beta group (20.35%) and the lowest among the HbE-Trait group (16.49%), thus showing that anisocytosis is the highest among the E-beta group. | Table 1: Mean and measures of dispersion of different blood indices among the subjects with different types of hemoglobinopathies (n=202)
Click here to view |
The mean value of MCV is the highest for HbE trait (80.27 fl), and although the mean values of other three groups are similar, the lowest is for HBE homozygous (68.49 fl), corresponding to the fact that all hemoglobinopathies except HbE trait have microcytic indices. For MCH and MCHC, the highest mean values were for the groups of HbE trait and HbE homozygous, respectively. For the above-mentioned parameters, the lowest mean values were found for beta thalassemia trait (20.14) and E-beta (29.17), respectively [Table 1].
To find out any significant difference of the mean among these four groups, one-way ANOVA along with Bonferroni post-hoc test was conducted. No significant difference of mean of Hb% was found among these four groups. The results show a significant difference of total RBC count (p = 0.003), RDW (p < 0.001), MCV (p < 0.001), MCH (p < 0.001), and MCHC (p = 0.003) between BTT and HbE trait. No significant difference of mean was found between HbE homozygous and E-beta patients. BTT differs significantly from E-beta regarding total RBC count (p = 0.018). HbE trait differs from both HbE homozygous and E-beta significantly in three parameters, namely, RDW, MCV, and MCH.
To determine the cutoff value for the diagnosis of E-trait compared to BTT, the ROC curve was drawn. [Figure 2] Four variables, namely, MCV, MCH, hemoglobin %, and MCHC, had more than 50% area under the curve, denoting that they can be used to find out the cutoff value. Among these, MCV has 93% and MCH has 92% area under the curve, making them very important indices for use as cutoff values between these two hemoglobinopathies. A value of MCV at 73.8 (fl) or more gives 91% sensitivity and 86.7% specificity of the diagnosis for E-trait. Similarly, the value of MCH (pg) at 21.9 or more gives 93.3% sensitivity and 83.3% specificity of diagnosis for E-trait. A hemoglobin cutoff value of 9.78 (g/dl) can be used with 70.8% sensitivity [Table 2]. | Figure 2: ROC curve showing the area under the curve for parameters such as hemoglobin %, MCV, MCH, and MCHC for the diagnosis of E-trait compared to beta thalassemia trait (n = 180)
Click here to view |
 | Table 2: Cutoff point and related sensitivity and specificity to make differentiate in diagnosis between HbE-trait and beta thalassemia trait (n=180)
Click here to view |
Positive and negative likelihood ratios were also calculated and presented in [Table 2]. The results suggest that an MCV value of 73.8 or more is 6.84 times more likely to be present in a patient of HbE trait than in a patient of BTT. Similarly, an MCH value of 21.9 or more is 5.59 times more likely in HbE traits. The negative likelihood ratio is 0.55 for HbE-trait compared to BTT if the cutoff hemoglobin value is taken as 9.78.
The relative prevalence of HbE-trait compared to BTT was calculated from a previous study conducted in Kolkata by Mohanty D et al.[5] With the help of the data, PPV and NPV were also calculated for all four parameter cutoffs for which sensitivity and specificity were calculated [Table 2].
To determine the cutoff value for the diagnosis of HbE homozygous compared to E-beta, another ROC curve was drawn [Figure 3]. Three variables, namely, total RBC count, hemoglobin %, and MCHC, had more than 50% area under the curve, denoting that they can be used to find out the cutoff value. Among these, RBC count has 72% and Hb% has 69% area under the curve. A value of RBC count at 4.11 million/μl or more gives 60% sensitivity and 75% specificity of the diagnosis for HbE homozygous. Similarly, the value of Hb% at 8.34 or more gives 60% sensitivity and 62.5% specificity of the diagnosis for HbE homozygous [Table 3]. | Figure 3: ROC curve showing the area under the curve for parameters such as hemoglobin %, RBC count, and MCHC for the diagnosis of E homozygous compared to E-beta thalassemia (n = 22)
Click here to view |
 | Table 3: Cutoff point and related sensitivity and specificity to make differentiate in diagnosis between HBE homozygous and E beta (n=22)
Click here to view |
Again, positive and negative likelihood ratios were calculated and are presented in [Table 3] for HbE homozygous compared to E-beta. The results suggest that an RBC count of 4.11 or more is 2.4 times more likely to be present in a patient of HbE-homozygous than in a patient of E-beta. The negative likelihood ratio (0.64) for HbE-homozygous compared to E-beta is the highest for a hemoglobin value of 8.34 or more.
The relative prevalence of HbE-homozygous compared to E-beta was calculated from a previous study conducted in Kolkata by Mohanty D et al.[5] With the help of the data, PPV and NPV were also calculated for all three parameter cutoffs for which sensitivity and specificity were calculated [Table 3].
| Discussion | | |
Thalassemia and other hemoglobinopathies are one of the major public health problems in countries such as India, where the total population size is quite large. The estimated number of persons with hemoglobinopathies is around 25 million in countries such as India.[7] In addition to the commonly encountered beta thalassemia group of patients, hemoglobin E (HbE) is another common and important mutation which is very much prevalent in the North East region of India. It is important to differentiate HbE disorders diagnostically because of marked differences in clinical course among various genotypes.[8] HbE disorders may be found in heterozygous, homozygous, and compound heterozygous states (e.g., HbE with other abnormal hemoglobins or thalassemias) with widely variable clinical characteristics.[9] Patients with double heterozygous for HbE trait and BTT interactions have a significant contribution to morbidity and mortality having severe anemia as is found in ß-thalassemia major.[3],[5],[10]
According to a multi-centric study by Mohanty D, et al.[5] in collaboration with the Indian Council of Medical Research, the prevalences of different hemoglobinopathies were detected on average as 2.78%, 3.63%, 0.67%, and 0.19% for BTT, E-trait, EE, and E-beta, respectively.[5] According to the same study, the prevalences of BTT, E-trait, EE, and E-beta were 3.64%, 3.92%, 0.12%, and 0.09%, respectively, in Kolkata.[5] In our study based on the study population of Kolkata, the prevalences of BTT, E-trait, EE, and E-beta are calculated as 4%, 4%, 0.66%, and 0.31%, respectively, which closely resemble the data of a previous study.
Regarding RBC indices, from a study by Sharma A, et al.,[4] the mean Hb for E-trait, EE, and E-beta was 11, 8.97, and 7.28 g/dl, respectively. These findings are quite similar to our study results where the mean Hb for E-trait, EE, and E-beta was 10.55, 9.31, and 7.62 g/dl, respectively. However, there was no statistically significant difference found between these groups in respect of Hb values. Regarding mean MCV, the lowest value was determined in HbE homozygous cases by Sharma A, et al.,[4] which is again similar to our finding.
Regarding significance of RBC indices, Sadiya S, et al.[11] concluded that RBC indices can be utilized for screening of thalassemia and other hemoglobinopathies and are to be confirmed by the HPLC method. However, they compared RBC indices of normal subjects with different hemoglobinopathies, and they found most of the RBC indices having a significant difference. In contradiction to their study, Khondaker T, et al.[12] found no diagnostic significant role of RBC indices in HbE-related hemoglobinopathies. In comparison to these two studies, the current study has approached in a slight different way, where the authors tried to find out the significance of RBC indices in between these hemoglobinpathies rather than comparing with normal subjects. The diagnosis of E-trait, EE, E-beta, and BTT mainly depends on HbA2 values because HbA2 and HbE co-elute at the same region with the same retention time by the beta thalassemia short program of the Bio-Rad Variant 2 system and thus always having a chance of the carryover phenomenon causing confusion in diagnosis sometimes by HPLC.[13] Hence, our study mainly focussed on differentiating BTT versus E-trait and EE versus E-beta, where both clinical and HPLC findings may be overlapping. Accordingly, it is found that RBC count and indices such as MCV, MCH, MCHC, and RDW vary significantly between BTT and E-trait. MCV and MCHC are statistically significantly higher in the case of E-trait than in BTT. The authors in this study also tried to set a cutoff value to differentiate between BTT and E-trait, and it was found that MCV and MCHC values may be utilized with high sensitivity to differentiate BTT and E-trait. On the other hand, none of the RBC indices were found to be statistically significant to discriminate E beta and E homozygous.
To the best of our knowledge, this is the first study in the eastern Indian population to determine any cutoff value of RBC indices to differentiate BTT and E-trait. Most of the previous authors have studied cutoff values of different RBC indices either between BTT and normal subjects or with an iron deficiency status.[14] A few previous studies also demonstrated different formulas to differentiate the iron deficiency status and BTT as a screening method in microcytic hypochromic cases.[15],[16]
However, this study has a few limitations. Confounding factors such as iron deficiency, pregnancy, or thyroid disorders have not been taken into account. These confounding factors, especially nutritional factors, are distributed randomly in all the concerned study groups, and hence, it can be considered that these factors will not affect the study results significantly. Second, the sample size of E homozygous and E-beta are considerably low, but this is attributed to the overall low prevalence in the population. Also, the RBC indices may vary in different ethnic groups of the population. Hence, these cutoff values may not be generalized to other ethnic groups.
| Conclusion | | |
This study demonstrated that RBC indices such as MCV, MCH, MCHC, and RDW vary significantly between BTT and other HbE-related hemoglobinopathies. They can specially be utilized to differentiate BTT and E-trait as supportive tests in addition to the gold standard test of HPLC. However, further studies in different ethnic populations with larger sample sizes may be required to determine cutoff values of RBC indices with high specificity and sensitivity to differentiate these hemoglobinopathies.
Acknowledgements
- Mr Harshit Parolia, CEO, One Diagnostics Laboratory, 5A/1A Lord Sinha road, Amar sudha building, Kolkata-700071.
- Mr Akash Chatterjee, Senior lab technician, One diagnostics Laboratory.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Weatherall DJ, Clegg JB. Thalassemia--A global public health problem. Nat Med 1996;2:847-9.  |
| 2. | Weatherall DJ, Clegg JB. Inherited haemoglobin disorders: An increasing global health problem. Bull World Health Organ 2001;79:704-12. |
| 3. | Das R, Datta S, Kaviraj A, Sanyal SN, Nielsen P, Nielsen I, et al. A decision support scheme for beta thalassemia and HbE carrier screening. J Adv Res 2020;24:183-90. |
| 4. | Sharma A, Marwah S, Buxi G, Yadav R. Hemoglobin e syndromes: Emerging diagnostic challenge in north India. Indian J Hematol Blood Transfus 2013;29:21-5. |
| 5. | Mohanty D, Colah RB, Gorakshakar AC, Patel RZ, Master DC, Mahanta J, et al. Prevalence of β-thalassemia and other haemoglobinopathies in six cities in India: A multicentre study. J Community Genet 2013;4:33-42. |
| 6. | Mishra DP, Triya L, Silal S. HPLC: An ideal methodology and screening tool for diagnosis of haemoglobinopathies in paediatric age group in southern Odisha. Sch J App Med Sci 2018;6:2316-32. |
| 7. | Saxena A, Phadke SR. Thalassaemia control by carrier screening: The Indian scenario. Curr Sci 2002;83:291–5. |
| 8. | Vichinsky E. Hemoglobin e syndromes. Hematology Am Soc Hematol Educ Program 2007;2007:79-83. |
| 9. | Patne SC, Shukla J. Hemoglobin E disorders in eastern Uttar Pradesh. Indian J Pathol Microbiol 2009;52:110-2. [ PUBMED] [Full text] |
| 10. | Kiran SS, Aithal S, Belagavi CS. Hemoglobin E hemoglobinopathy in an adult from assam with unusual presentation: A diagnostic dilemma. J Lab Physicians 2016;8:116-9. [ PUBMED] [Full text] |
| 11. | Sadiya S, Khan WA, Banu B, Akhter N, Sarwardi G, Mia B. Role of RBC indices in screening of thalassaemia and other haemoglobinopathies in children. DS (Child) H J 2015;31:101-8. |
| 12. | Khondaker T, Ahmed AK, Amin S. Diagnostic significance of haemogram parameters and RBC indices in haemoglobin e trait. Anwer Khan Mod Med Coll J 2017;8:94-9. |
| 13. | Chopra P, Bhardwaj S, Negi P, Arora A. Comparison of two high-pressure liquid chromatography instruments bio-rad variant-ii and tosoh HLC-723G11 in the evaluation of hemoglobinopathies. Indian J Hematol Blood Transfus 2020;36:725-32. |
| 14. | Surjawan Y, Tan HL, Setiabudy RD, Rositawati W. Early screening of hemoglobinopathy in Indonesia using erythrocyte indices. Indones Biomed J 2017;9:99-105. |
| 15. | Srivastava PC, Bevington JM. Iron deficiency and-or thalassaemia trait. Lancet 1973;1:832. |
| 16. | Mentzer WC Jr. Differentiation of iron deficiency from thalassaemia trait. Lancet 1973;1:882. |
Correspondence Address:
Manas Talukdar,
10/1, Girish Ghosh Street, Kolkata - 700108, West Bengal
India
 Source of Support: None, Conflict of Interest: None DOI: 10.4103/ijpm.ijpm_1071_21
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3] |
|
|
|
|
|
|
|
|
|
|
|
|
|
| Article Access Statistics | | | Viewed | 2059 | | | PDF Downloaded | 60 |
|
