Abstract
Abstract
To discriminate iron deficiency anemia (IDA) from β thalassemia trait (βTT), several indices obtained from modern blood count analyzers have been reported. Discrimination power of seven indices to differentiate between IDA and βTT, such as Green and King Index (GKI), RDW Index (RDWI), Srivastava Index (SRI), Mentzer Index (MI), Shine and Lal Index (SLI), Ehsani Index (EI), and Sirdah Index (SI), were evaluated. These indices were applied on 47 patients with βTT and on 289 patients with IDA, as confirmed by gold standard tests. Sensitivity, specificity, positive and negative predictive values, efficiency, area under receiver-operating characteristics curve (AUC), and Youden's Index (YI) were calculated. GKI and RDWI showed the highest reliability, as they had the largest AUCs (0.919, 0.912, respectively) and Youden's Index (70.4, 74.6, respectively). Conversely, SLI presented a less satisfactory performance (AUC = 0.786 and YI = 6.6). Data taken together suggest the superiority of GKI and RDWI to discriminate between IDA and βTT.
Introduction
Microcytic and hypochromic anemia can result from genetic defects in the synthesis of globin chains, alterations in iron metabolism, or decrease in its supply to the bone marrow. As a consequence, the production of hemoglobin by erythroid precursors is hampered, resulting in less red blood cell production with low hemoglobin concentration, except in thalassemia the number of red blood cells (RBCs) being typically increased. Several diseases result in the production of microcytic hypochromic red blood cells, such as iron deficiency anemia (IDA), thalassemia trait (TT), lead poisoning, chronic inflammation, sideroblastic anemia,1–3 among others.
IDA is one of the most common diseases in the world affecting approximately one billion people.4,5 On the other hand, thalassemia is the most common monogenic disorder in the world6 characterized by a partial or total lack of production of globin chains, mainly alpha and beta chains.7,8 The thalassemias are common in people whose ancestors came from the countries of the Mediterranean to Southeast Asia.9 In Southeast Brazil, a frequency of 1.3% of β thalassemia trait (βTT) was reported as being found in the general population,10 while α thalassemia trait (αTT), determined by a 3.7-kb DNA deletion, varied from 20.0 to 25.0% in a black population.11 According to Ferreira et al.,12 the overall prevalence of IDA was 5.6% in Acre, Brazil.
The differential diagnosis of microcytic anemia is of great clinical importance. Firstly, while iron deficiency is easily corrected with iron supplementation, such treatment should be totally avoided in thalassemia patients unless iron deficiency is also documented.13 Moreover, the lack of a correct diagnosis of a hemoglobinopathy decreases the effectiveness of genetic counseling programs to prevent the occurrence of homozygosity in one's offspring. In addition, once the diagnosis of TT has been established, further investigation searching for possible sites of blood loss or even other causes including malabsorption and inadequate intake, which would be essential in patients with IDA, becomes unnecessary.
The differential diagnosis of microcytic anemia, however, is often complex, and its laboratory investigation has significant costs. Microscopic analysis of the blood film, a lower-cost procedure, cannot discriminate between such anemias, due to a great morphological similarity among the diseases. Thus, complementary exams of blood count are required. Currently, the accepted protocols for the diagnostic confirmation of IDA include an assessment of iron metabolism, such as laboratory measurements of serum iron, total iron binding capacity, as well as serum ferritin. In the case of βTT, it is necessary to perform hemoglobin electrophoresis and measure both HbA2 and HbF to reach a proper diagnosis. Despite their great utility, these gold standard tests for the diagnosis of microcytic and hypochromic anemias involve time-consuming methodologies and are much more expensive and often inaccessible to a great number of people.
Thus, the development of a simpler and less costly index to screen for microcytic and hypochromic anemias would be extremely useful. To streamline the diagnostic approach, prior literature has proposed the use of indices derived from parameters commonly available in modern automatic counters. These may provide evidence of the presence of either IDA or TT. The most common indices include the Green and King,14 RDW,15 Srivastava,16 Mentzer,17 Shine and Lal,18 Ehsani,19 and Sirdah20 Indices.
The present study aimed to evaluate the efficiency of these seven previously cited indices in discriminating patients with IDA from those with βTT, whose diagnosis has been previously confirmed by gold standard tests.
Methods
Ethical statement
The present study was approved by the Ethics Committees from Federal University of Minas Gerais (UFMG), under protocol no. 344/09, and from the Governor Israel Pinheiro Hospital in Belo Horizonte, Brazil, under protocol no. 361/09.
Patients were informed of the research objectives and were required to sign the Informed Consent Form before sampling and data collection.
Study design
This cross-sectional study was conducted from 2005 to 2011. Once the blood samples had been collected, all tests were processed. Samples were collected from 289 patients enrolled specifically for this study (251 women and 38 men with a mean ± SD age of 49.2 ± 14.8 years, ranging from 18 to 89 years of age), who had been admitted to the Clinical Pathology Service of the Governor Israel Pinheiro Hospital. Patients with hemoglobin of less than 12.0 g/dL for women and 13.0 g/dL for men,21 and a mean corpuscular volume (MCV) and mean corpuscular hemoglobin (MCH) below 80 fL and 27 pg for both sexes, respectively,22 indicating a diagnosis of microcytic and hypochromic anemia, were deemed eligible for the study. The hematological evaluation was performed in a Bayer® Advia® 120 automatic counter (Bayer Corporation, Newbury, Berks, UK). Serum ferritin levels were determined by the Immulite® DPC® unit, using the chemiluminescent and the Immulite 2000 DPC® kits (Diagnostic Products Corporation, Los Angeles, CA, USA) whose reference values were between 28 and 397 ng/mL for men and between 6 and 159 ng/mL for women. All patients who presented ferritin values under the reference values were classified as having IDA.
A second group of 47 patients (34 women and 13 men with a mean ± SD age of 52.5 ± 17.0 years, ranging from 18 to 81 years of age) from the Clinical Pathology Service of the Governor Israel Pinheiro Hospital and from the UFMG Clinical Hospital in Belo Horizonte, Brazil, were also enrolled. Hematological evaluation was performed using a Bayer® Advia® 120 automatic counter. The use of controls at two levels and attention paid in the pre-analytical, analytical, and post-analytical steps ensured the quality of the results in the hematological counter. These patients also presented an MCV and an MCH below 80 fL and 27 pg, respectively,22 indicating microcytosis and hypochromia. Hemoglobin electrophoresis was performed using cellulose acetate strips at an alkaline pH of 8.6, using a Cuba Source Electrophoresis TECNOW® 7000 (Tecnow Scientific Instruments, São Paulo, Brazil). In the case of suspicion of increased HbA2, the dosage was performed by applying the method of elution by using a cellulose acetate strip. Serum ferritin was measured in the unit of DPC® Immulite® in both laboratories. All these patients were classified as having the βTT due to the presence of normal serum ferritin levels, but increased levels of hemoglobin A2 (more than 3.5%).23 For further characterization, the data from various hematological parameters and serum ferritin of patients in both groups are presented in Table 1.
Table 1. Hematological parameters in patients with IDA or β thalassemia trait βTT. Data are shown as medians, with the first (Q1) and third quartiles (Q3) of each group in parentheses
Subsequently, seven indices were calculated for all patients with either IDA or βTT using the following formulas:
Green and King Index (GKI): (MCV)2 × RDW/100 × Hb,
RDW Index (RDWI): MCV × RDW/RBC,
Srivastava Index (SRI): MCH/RBC,
Mentzer Index (MI): MCV/RBC,
Shine and Lal Index (SLI): MCV2 × MCH × 0.01,
Ehsani Index (EI): MCV − 10 × RBC,
Sirdah Index (SI): MCV − RBC − 3 × Hb
where MCV = mean corpuscular volume, RDW = red cell distribution width, Hb = hemoglobin, RBC = red blood cells, and MCH = mean corpuscular hemoglobin.
The cutoff points used were 73 (GKI), 220 (RDWI), 3.8 (SRI), 13 (MI), 1530 (SLI), 15 (EI), and 27 (SI). Values below the respective cutoffs suggested the presence of βTT.
The choice of these formulas was based on the frequency and performance reported in the literature, ranging from indices developed since the advent of automation in hematology to the present day.
For each index, sensitivity, specificity, positive and negative predictive values, and efficiency were calculated. Positive and negative predictive values were calculated based on the prevalence of 1.3 and 5.6% of βTT and IDA, respectively.10,12 The area under the receiver-operating characteristics (ROC) curve (AUC) and Youden's Index were also calculated. While efficiency means proportion of accurately classified individuals, the largest AUC indicates the index most likely to correctly discriminate patients with one of these anemias. The AUCs were obtained using the Prism software version 5 (Prism software, Irvine, CA, USA) and statistical comparison between AUCs was performed using Stata software version 11 (Stata Corp., College Station, TX, USA). Youden's Index provides an appropriate measure of the validity of a technique. A value for this rate close to zero suggests that the method is not better than any other technique used randomly. Youden's Index was calculated by applying the following formula: Youden's Index (sensitivity + specificity)–100.
Results
Median values and interquartile values of some hematological parameters for the two groups, as well as the difference between these groups, are presented in Table 1.
Table 2 illustrates the number and percentage of patients correctly identified using the tested indices. The values of sensitivity, specificity, positive and negative predictive values, efficiency, AUC, and Youden's Index are presented in Tables 3 and 4.
Table 2. Patients (in absolute numbers and percentages) correctly diagnosed as having iron deficiency anemia (IDA) or β thalassemia trait (βTT) by applying the indices below
Table 3. Sensitivity, specificity, positive predictive (PPV) and negative predictive (NPV) values, and efficiency of some indices to discriminate between IDA and βTT
Table 4. Values of area under the ROC curve and Youden's Index for the discriminant functions tested
Among the seven tested indices, SLI correctly diagnosed the greatest number of patients with βTT. However, this index failed to diagnose IDA patients, classifying 270 patients in this group as having βTT, which shows a high sensitivity (100.0%) and a very low specificity (6.6%) for the diagnosis of βTT. In addition, three patients with IDA showed values of all formulas compatible with βTT. In the group of IDA, the index that correctly identified a greater number of patients was the SI (272 patients correctly identified).
In terms of sensitivity and specificity, the highest values of these parameters in the diagnosis of βTT were observed for both SLI and SI, respectively. The highest efficiency was achieved for RDWI (Table 3).
Due to the low prevalence of IDA and βTT used as references low values of PPVs were obtained for all indices tested.
Based on the higher values of AUC (Fig. 1 and Table 4) and Youden's Index (Table 4), parameters that provide appropriate measures of the validity and accuracy of a technique, the most reliable indices were GKI and RDWI, while SLI presented the lowest accuracy.
Published online:
18 July 2013![]({"type":"image","src":"/na101/home/literatum/publisher/tandf/journals/content/yhem20/2013/yhem20.v018.i03/1607845412y.0000000054/20181031/images/medium/yhem_a_11748394_f0001_b.jpg"})
Figure 1. ROC curves of GKI, SI, RDWI, MI, EI, SRI, and SLI.
Discussion
Several studies have been conducted to verify and compare the efficiency of indices for the differential diagnoses of IDA and TT.3,24–27 These results proved to be widely varied. This study tested seven indices: GKI, MI, SRI, RDWI, SLI, SI, and EI.
Table 1 shows some results of this study. Significant differences could be observed among all hematological parameters of the βTT and IDA groups, which could aid in differentiating between the two anemias.
The highest value of efficiency was observed for RDWI (92.0), which was much higher than those found in studies by Ntaios et al.27 and Ferrara et al.28 In terms of sensitivity for the diagnosis of βTT, the highest value was found for SLI (100.0), which is consistent with data reported by Shen et al.29 and Demir et al.25 However, this formula showed the lowest specificity for the diagnosis of βTT, which is also in agreement with findings from Shen et al.29 and Demir et al.25 in which the specificity value was zero. The highest specificity was obtained by applying the SI (94.1), which was higher than that observed by Sirdah et al.20 and similar to that found by Shen et al.29 However, in our study, the sensitivity of this index was relatively low when compared with most other tested indices.
By determining the AUC curve and Youden's Index, which provide a measure of the accuracy and reliability of a technique, the indices that showed a better performance in correctly differentiating cases of IDA from βTT were GKI and RDWI. In the original study carried out by Green and King,14 the formula developed by these authors correctly identified 100% of patients with IDA and 100% of patients with βTT. These authors also tested their formula for some patients with other hematological disorders, such as polycythemia vera, α thalassemia, and chronic disease anemia, which produced conflicting results. A result similar to the Green and King Index was found by Lima et al.30 in which this index has correctly identified in 97% of the patients with IDA and the same percentage of βTT, when using a cutoff value of 80. Ntaios et al.27 observed that the Green and King Index proved to be the most reliable index among the discriminant functions evaluated by presenting the highest sensitivity, efficiency, and Youden's Index in detecting βTT. In addition, the value of Youden's Index was similar to that obtained in this study. Shen et al.29 also found that, according to the AUC, the GKI was the best discriminant formula. In contrast to these studies, the results reported by Alfadhli et al.,3 based on the value of Youden's Index, was that the GKI proved to be the formula with the third-lowest discriminant performance in differentiating between βTT and IDA, with a Youden's Index of 47.6. Nesa et al.,31 who conducted the latest study on discriminant formulas, concluded that the RDWI is a reliable and useful formula in discriminating between these two microcytic anemias.
The function with the lowest performance in discriminating between IDA and βTT was the SLI, which presented an extremely low value for Youden's Index (6.6), a result consistent with recent findings from Demir et al.,25 Alfadhli et al.,3 and Shen et al.29 who found a value for this parameter of 0.5, 0, and −0.8, respectively. It is important to note that a value for Youden's Index close to zero suggests that the technique is not superior to any method used at random. In the study carried out by Shen et al.,29 in addition to a low Youden's Index, the lowest AUC for the Shine and Lal formula could also be observed. In contrast to these studies, Okan et al.32 found the highest Youden's Index for SLI. Ethnical and methodological differences most likely explain these conflicting results.
The other indices presented Youden's Index and the ROC curve area values situated within the values of GKI, RDWI, and SLI. Considering only the ROC curve, the index with the second best performance was the SI, whose value for this parameter was 0.916, which is very similar to that found by Sirdah et al.20 and Shen et al.29 As previously described RDWI also presented a good performance having showed the third biggest AUC (0.912). In fourth and fifth places were the MI (0.894) and EI (0.893). Shen et al.29 found values similar to those observed in this study for both indices. In the study conducted by Sirdah et al.,20 values were 0.902 and 0.899 for MI and EI, respectively. SRI presented the second lowest value for the ROC curve area (0.860), which is also similar to that observed by Sirdah et al.20
One limitation to the present study refers to the 289 patients classified as IDA who were not assessed for α nor β thalassemia trait. However, probability for β thalassemia trait in the general Brazilian population is 1.3%.10 Therefore, it is likely that four out of all those patients are also β-thalassemia heterozygotes. This number is quite small to affect the statistical results. It is also important to highlight that HbA2 levels are influenced by the presence of IDA considering that its levels are significantly reduced in patients carrying both conditions simultaneously. Concerning αTT, which demands molecular methods for its confirmation, it is noteworthy to mention that these specialized methods are not available in conventional clinical analyses laboratories.
Another limitation is that a smaller number of patients in the group with the βTT could be observed. This may well be due to the fact that the frequency of the βTT is much higher than the number of people looking for a specialized service, since most cases present no symptoms to justify the search for medical assistance. Thus, despite the extended time searching for new cases of βTT, the number of patients with βTT included in this study was limited.
The use of indices does not replace the ‘gold standard’ exams for the diagnosis of microcytic and hypochromic anemias. However, it should be noted that under certain conditions of concomitant diseases (e.g. IDA and anemia of chronic disease) even the results of gold standard tests may suffer interference from intercurrent diseases, which render the proper diagnosis more difficult. It is well known that the diagnosis of IDA is particularly altered in patients with acute or chronic inflammatory conditions, given that most of the biochemical markers of iron metabolism are affected by acute phase reaction.5 The interference in the gold standard exams of a concomitant illness can be exemplified by the increase in ferritin in patients with chronic disease anemia which could mask an IDA, or also possibly decrease the levels of HbA2 in a patient with TT when IDA is a simultaneous finding.33,34 The dosage of the transferrin receptor could confirm the diagnosis of IDA in cases of concomitant chronic diseases, since this dosage is not affected by such diseases. However, the measurement of transferrin receptor levels is not yet available in most clinical labs.
Conclusion
Although one cannot reach a definitive diagnosis of IDA or βTT based merely on the discriminant functions, these simple calculations are potentially useful in screening patients with microcytic anemia. These indices are a useful tool in the doctor's guidance about the initial approach to be adopted, but do not relieve patient monitoring that may eventually require confirmatory tests to elucidate the strong suspicion initially raised by the application of these simple indices. Furthermore, these formulas can be the only differential tool in situations where other specific confirmatory tests are not available.
Even though there are several indices with the potential to be applied in the clinical laboratory, the comparative analysis in this study showed the superiority of Green and King and RDW Indices in discriminating between IDA and βTT.
| Parameters | IDA (n = 289) | βTT (n = 47) | P |
|---|---|---|---|
| RBC (106/μL) | 4.50 (4.17–4.77) | 5.47 (5.14–5.87) | <0.0001 |
| Hb (g/dL) | 10.1 (9.3–10.7) | 11.5 (10.5–12.3) | <0.0001 |
| Ht (%) | 32.1 (30.3–34.0) | 35.7 (33.2–38.0) | <0.0001 |
| MCV (fL) | 73.0 (67.8–76.0) | 63.7 (61.7–69.0) | <0.0001 |
| MCH (pg) | 22.8 (20.8–24.1) | 20.4 (19.6–21.7) | <0.0001 |
| MCHC (g/dL) | 31.3 (30.5–32.1) | 31.8 (31.3–32.7) | 0.001 |
| RDW (%) | 17.4 (16.0–19.2) | 15.8 (15.2–16.7) | <0.0001 |
| Pla (103/μL) | 319 (257–383) | 235 (196–271) | <0.0001 |
| Ferritin (ng/mL) | 3.80 (2.90–5.15) | 129.0 (48.8–275.0) | <0.0001 |
RBC, red blood cells; Hb, hemoglobin; Ht, hematocrit; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; RDW, red blood cell distribution width; Pla, platelets.
| Index | IDA (n = 289) | βTT (n = 47) |
|---|---|---|
| GKI | 259 (89.6%) | 38 (80.9%) |
| SI | 272 (94.1%) | 30 (63.8%) |
| RDWI | 271 (93.8%) | 38 (80.9%) |
| MI | 254 (87.9%) | 36 (76.6%) |
| EI | 256 (88.6%) | 34 (72.3%) |
| SRI | 263 (91.0%) | 26 (55.3%) |
| SLI | 19 (6.6%) | 47 (100.0%) |
GKI, Green and King Index; SRI, Srivastava Index; RDWI, RDW Index; SLI, Shine and Lal Index; MI, Mentzer Index; SI, Sirdah Index; EI, Ehsani Index.
| Index | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | Efficiency (%) |
|---|---|---|---|---|---|
| GKI | |||||
| IDA | 89.6 | 80.9 | 21.7 | 99.2 | 88.4 |
| βTT | 80.9 | 89.6 | 9.3 | 99.7 | |
| SI | |||||
| IDA | 94.1 | 63.8 | 13.4 | 99.4 | 88.6 |
| βTT | 63.8 | 94.1 | 12.5 | 99.5 | |
| RDWI | |||||
| IDA | 93.8 | 80.9 | 22.5 | 99.5 | 92.0 |
| βTT | 80.9 | 93.8 | 14.6 | 99.7 | |
| MI | |||||
| IDA | 87.9 | 76.6 | 18.2 | 99.1 | 86.3 |
| βTT | 76.6 | 87.9 | 7.7 | 99.6 | |
| EI | |||||
| IDA | 88.6 | 72.3 | 15.9 | 99.1 | 86.3 |
| βTT | 72.3 | 88.6 | 7.7 | 99.6 | |
| SRI | |||||
| IDA | 91.0 | 55.3 | 10.8 | 99.0 | 86.0 |
| βTT | 55.3 | 91.0 | 7.5 | 99.4 | |
| SLI | |||||
| IDA | 6.6 | 100.0 | 100.0 | 94.7 | 19.6 |
| βTT | 100.0 | 6.6 | 1.4 | 100.0 | |
GKI,Green and King Index; SRI, Srivastava Index; RDWI, RDW Index; SLI, Shine and Lal Index; MI, Mentzer Index; SI, Sirdah Index; EI, Ehsani Index.
| Index | Area under ROC curve | Youden Index |
|---|---|---|
| GKI | 0.919 (0.865, 0.973)*,** | 70.4 |
| SI | 0.916 (0.869, 0.962)*,** | 57.9 |
| RDWI | 0.912 (0.855, 0.970)*,** | 74.6 |
| MI | 0.894 (0.843, 0.945)*,** | 64.5 |
| EI | 0.893 (0.843, 0.943)*,** | 60.9 |
| SRI | 0.860 (0.803, 0.918)** | 46.3 |
| SLI | 0.786 (0.719, 0.852) | 6.6 |
GKI, Green and King Index; SRI, Srivastava Index; RDWI, RDW Index; SLI, Shine and Lal Index; MI, Mentzer Index; SI, Sirdah Index; EI, Ehsani Index
*Difference with SRI.
**Difference with SLI.
Acknowledgements
We would like to express our thanks to the Laboratory of Clinical Pathology of the Governor Israel Pinheiro Hospital and UFMG Clinical Hospital for their assistance.
References
- Oliveira RAG, Neto AP. Anemias e Leucemias – Conceitos Básicos e Diagnóstico por técnicas laboratoriais. São Paulo: Roca; 2004. [Google Scholar]
- Zago MA, Falcão RP, Pasquini R. Hematologia: Fundamentos e prática. São Paulo: Atheneu; 2005. [Google Scholar]
- Alfadhli SM, Al-Awadhi AM, Alkhaldi D. Validity assessment of nine discriminant functions used for the differentiation between iron deficiency anemia and thalassemia minor. J Trop Pediatr 2007;53:93–7. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Nobili B, Perrotta S, Matarese SMR, Conte ML, Giudice EM. Evaluation of body iron status in Italian carriers of beta-thalassemia trait. Nutr Res 2001;21:55–60. [Crossref], [Web of Science ®], [Google Scholar]
- Brugnara C. Iron deficiency and erythropoiesis: new diagnostic approaches. Clin Chem 2003;49:1573–8. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Angastiniotis M, Modell B. Global epidemiology of hemoglobin disorders. Ann N Y Acad Sci 1998;850:251–69. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Hilliard LM, Berkow RL. The thalassemia syndromes. Prim Care Update Ob Gyns 1996;3:157–62. [Crossref], [Google Scholar]
- Sarnaik SA. Thalassemia and related hemoglobinopathies. Indian J Pediatr 2005;72:319–24. [Crossref], [PubMed], [Google Scholar]
- Joshi DD, Nickerson HJ, Mcmanus MJ. Hydrops fetalis caused by homozygous α-thalassemia and Rh antigen alloimmunization. Clin Med Res 2004;2:228–32. [Crossref], [PubMed], [Google Scholar]
- Ramalho AS, Silva RBP, Teixeira RC, Compri MB. Hemoglobin screening: response of a Brazilian community to optional programs. Cad Saude Publica 1999;15:591–5. [Crossref], [PubMed], [Google Scholar]
- Sonati MF, Farah SB, Ramalho AS, Costa FF. High prevalence of alpha-thalassemia in a black population of Brazil. Hemoglobin 1991;15:309–11. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
- Ferreira MU, da Silva-Nunes M, Bertolino CN, Malafronte RS, Muniz PT, Cardoso MA. Anemia and iron deficiency in school children, adolescents, and adults: A community-based study in rural Amazonia. Am J Public Health 2007;97:237–9. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Yermiahu T, Ben-Shalom M, Porath A, Vardi H, Boantza A, Mazor D, et al. Quantitative determinations of microcytic-hypochromic red blood cell population and glycerol permeability in iron deficiency anemia and β thalassemia minor. Ann Hematol 1999;78:468–71. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Green R, King R. A new red cell discriminant incorporating volume dispersion for differentiating iron deficiency anemia from thalassemia minor. Blood Cells 1989;15:481–95. [PubMed], [Google Scholar]
- Jayabose S, Giavanelli J, Levendoglu-Tugal O, Sandoval C, Ozkaynak F, Visintainer P. Differentiating iron deficiency anemia from thalassemia minor by using a RDW-based index. J Pediatr Hematol Oncol 1999;21:314. [Crossref], [Google Scholar]
- Srivastava PC. Differentiation of thalassaemia minor from iron deficiency. Lancet 1973;2:154–5. [PubMed], [Web of Science ®], [Google Scholar]
- Mentzer WCJ. Differentiation of iron deficiency from thalassaemia trait. Lancet 1973;1:882. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Shine I, Lal S. A strategy to detect beta-thalassaemia minor. Lancet 1977;1:692–4. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Ehsani M, Darvish A, Aslani A, Seighali F. A new formula for differentiation of iron deficiency anemia (IDA) and thalassemia trait (TT). Turkish J Hematol (Supplement) 2005;22:268. [Google Scholar]
- Sirdah M, Tarazi I, Al Najjar E, Al Haddad R. Evaluation of the diagnostic reliability of different RBC indices and formulas in the differentiation of the beta-thalassaemia minor from iron deficiency in Palestinian population. Int J Lab Hematol 2008;30:324–30. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- World Health Organization – WHO. Guidelines for the use of iron supplements to prevent and treat iron deficiency anemia. Available from: http://www.who.int/nutrition/publications/micronutrients/anaemia_iron_deficiency/1-57881-020-5/en/ [Google Scholar]
- World Health Organization – WHO. Iron deficiency anaemia: assesment, prevention and control. A guide for programme managers, 2001. Available from: http://www.who.int/nutrition/publications/micronutrients/anaemia_iron_deficiency/WHO_NHD_01.3/en/ [Google Scholar]
- Rodak BF, Fritsma GA, Keohane EM. Hematology: clinical principles and applications. 4th ed. Pennsylvania: Elsevier Saunders; 2011. [Google Scholar]
- Eldibany MM, Totonchi KF, Joseph NJ, Rhone D. Usefulness of certain red blood cell indices in diagnosing and differentiating thalassemia trait from iron deficiency anemia. Am J Clin Pathol 1999;111:676–82. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Demir A, Yarali N, Fisgin T, Duru F, Kara A. Most reliable indices in differentiation between thalassemia trait and iron deficiency anemia. Pediatr Int 2002;44:612–6. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Beyan C, Kaptan K, Ifran A. Predictive value of discrimination indices in differential diagnosis of iron deficiency anemia and beta-thalassemia trait. Eur J Haematol 2007;78:524–6. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Ntaios G, Chatzinikolaou A, Saouli Z, Girtovitis F, Tsapanidou M, Kaiafa G, et al. Discrimination indices as screening tests for β-thalassemic trait. Ann Hematol 2007;86:487–91. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Ferrara M, Capozzi L, Russo R, Bertocco F, Ferrara D. Reliability of red blood cell indices and formulas to discriminate between β thalassemia trait and iron deficiency in children. Hematology 2010;15:112–5. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
- Shen C, Jiang YM, Shi H, Liu JH, Zhou WJ, Dai QK, et al. Evaluation of indices in differentiation between iron deficiency anemia and beta-thalassemia trait for Chinese children. J Pediatr Hematol Oncol 2010;32:e218–222. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- Lima CSP, Reis ARC, Grotto HZW, Saad STO, Costa FF. Comparison of red cell distribution width and a red cell discriminant function incorporating volume dispersion for distinguishing iron deficiency from beta thalassemia trait in patients with microcytosis. São Paulo Med J 1996;114:1265–9. [Crossref], [PubMed], [Google Scholar]
- Nesa A, Munir SF, Sultana T, Rahman MQ, Ahmed AN. Role of discrimination indices in differentiation of beta thalassaemia trait and iron deficiency anaemia. Mymensingh Med J 2011;20:110–4. [PubMed], [Google Scholar]
- Okan V, Cigiloglu A, Cifci S, Yilmaz M, Pehlivan M. Red cell indices and functions differentiating patients with the beta-thalassaemia trait from those with iron deficiency anaemia. J Int Med Res 2009;37:25–30. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
- El-Agouza I, Abu Shahla A, Sirdah M. The effect of iron deficiency anaemia on the levels of haemoglobin subtypes: possible consequences for clinical diagnosis. Clin Lab Haematol 2002;24:285–9. [Crossref], [PubMed], [Google Scholar]
- Weiss G. Pathogenesis and treatment of anaemia of chronic disease. Blood Rev 2002;16:87–96. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]