Original Article
.auto-style1 {
background-color: #FFFF00;
}
Reliability of Different RBC Indices to Differentiate Between Beta Thalassemia Trait and Iron Deficiency Anemia During Antenatal Screening
*Reema Bhushan, * Shailaja Shukla * Divyanshu Singh, *SS Trivedi, * Sunita Sharma
- * Pathology, Lady Hardinge Medical College, New Delhi, India
- Submitted:Friday, March 2, 2018
- Accepted: Sunday, May27, 2018
- Published:Monday, June 4, 2018
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited
Abstract
Thalassemia and related hemoglobinopathies affect the quantity and quality of haemoglobin. High performance liquid chromatography (HPLC) is considered the gold standard for diagnosis of hemoglobinopathies. This test is not available at many peripheral centers. Many formulae or indices based on complete blood counts are available that can aid in the diagnosis. The aim of the study was to analyze the reliability of these indices in distinguishing between the most common causes of microcytic hypochromic anemia i.e. iron deficiency anemia (IDA) and hemoglobinopathies. A retrospective study was performed on 2000 women who underwent antenatal screening for beta thalassemia trait (b - TT) and related hemoglobinopathies. Nine indices were calculated among both the groups using complete blood count parameters. Sensitivity and specificity of these indices to detect beta -TT versus IDA was compared. Likelihood ratio, accuracy and youden’s index for these indices were calculated. Sensitivity of Shine and Lal index was highest (98.4%) followed by Kerman 1 index (66.7%). Mentzer index, England and Fraser index, RDW index and Ehsani index each, showed high specificity (99.66%). Kerman 1 index could point towards the correct diagnosis in 97.5% of the patients. Youden index was highest for Kerman 1 index (65.4) followed by RDW index (63.1). None of the index was 100% sensitive or specific.
Keywords
Haemoglobin,RBC indices,High performance liquid chromatography,antenatal screening
Introduction
Thalassemias and related hemoglobinopathies are caused by genetic mutations of the hemoglobin (Hb) genes resulting in reduced production or total absence of one or more globin chains, hence affecting the quantity and the quality of haemoglobin [1]. Ten percent of the total world thalassemics are born in India every year [2].Certain communities in India, like Sindhis, Gujratis, Punjabis, and Bengalis, are more commonly affected with beta thalassemia, the incidence varying from 1 to 17% [3].Diagnosis of beta- thalassaemia trait (b-TT) often becomes difficult due to its overlapping and similar features with iron deficiency anaemia (IDA). There are similarities in red cell indices between the two disorders such as reduced Hb, MCV and MCH. Various electrophoretic approaches, including cellulose acetate electrophoresis, isoelectric focusing in polyacrylamide or agarose gel, as well as high-performance liquid chromatography (HPLC), immunological assays, structural analysis and genotype methods, are used to investigate hemoglobinopathies [4].
High performance liquid chromatography (HPLC) forms an important gold standard tool for screening and detection of various hemoglobinopathies with rapid, reproducible and precise results [5]. However, this test is not available at many peripheral centers. Electronic cell counters have been used to determine red cell indices as a first indicator of b-TT. Since 1970, a number of complete blood count indices have been proposed as simple and inexpensive tools to determine whether a blood sample is more suggestive of b-TT or IDA [6]. An ideal discrimination index has high sensitivity and specificity; that is, it can detect the maximum number of patients with b-TT (high sensitivity) while eliminating patients with IDA (high specificity). In this study, we compared the ability of ten indices to distinguish b-TT from IDA by calculating their sensitivity, specificity, and Youden’s index values.
Material and Methods
A retrospective study was performed on 2000 women who underwent antenatal screening for b thalassemia trait (b- TT) and related hemoglobinopathies. Erythrocyte microcytosis and hypochromia was assessed using automated blood cell counter Sysmex KX-21 and by peripheral smear. All cases with normal MCV and MCH were not tested further (N=739). All samples which were microcytic (MCV<80fl) and hypochromic (MCH≤ 27pg) were tested for the serum ferritin levels by ELISA method. The cases with reduced level of serum ferritin were labelled as iron deficiency anemia (N=1194) while those with normal or increased serum ferritin level (N=67) were subjected to HPLC examination. On HPLC examination, there were 59 cases of b- thalassemia trait. Ten indices (Table 1) were calculated among both the groups i.e. b-thalassemia trait (N=59) and iron deficiency anemia (N=1194) using complete blood count parameters. Sensitivity and specificity of these indices to detect b-TT versus IDA was compared. Likelihood ratio, accuracy and Youden’s index for these indices were also calculated. Youden’s index was suggested by W.J. Youden and is a way of summarising the performance of a diagnostic test and is calculated by substracting 100 from the sum of sensitivity and specificity [6, 7].
Table 1: Different indices which were studied
| INDICES |
FORMULAE |
β- TT |
IDA |
| RDW |
β- TT=<13 |
<13 |
>13 |
| MENTZERS |
(MCV/RBC) |
<13 |
>13 |
| EFI |
MCV-(5XHb)-RBC-3.4 |
<0 |
>0 |
| SLI |
MCVXMCVXMCH/100 |
<1530 |
>1530 |
| GKI |
MCVXMCVXRDW/HbX100 |
<65 |
>65 |
| SRIVASTAVA |
MCH/RBC |
<3.8 |
>3.8 |
| RDW Index |
MCVXRDW/RBC |
<220 |
>220 |
| SIRDAH |
MCV-RBC-(3Hb) |
<27 |
>27 |
| EHSANI |
MCV-(10XRBC) |
<15 |
>15 |
| |
MCVXMCH/RBC |
<300 |
>300 |
Results
Hemoglobin concentration in the β-TT group was 10.88+2.35g/dl, and that in the IDA group was 8.84 +1.14 (Ρ < 0.05). MCV was 67.02+ 5.46fl and MCH was 20.96+ 1.99pg in the β-TT group. These values were lower than those in the IDA group (73.83+ 3.04 and 24.48+ 2.52, respectively; Ρ < 0.05). Thus, the amount of microcytosis and hypochromia was more in β-TT than IDA. The RDW was increased in both groups: 17.2 + 3.16 in the IDA group and 16.94 + 1.57 in patients with β-TT (Ρ > 0.05). RBC count reduced in IDA (3.7 + 0.27) while it was mildly elevated in β-TT (5.6 + 1.08) and this difference was statistically significant (Ρ < 0.05) (Table 2).
Table 2: Different parameters among the cases of thalassemia and iron deficiency anemia
| Parameters |
AGE(yr) |
Hb |
RBC |
MCV |
MCH |
RDW |
| THALASSEMIA CASES |
N |
63 |
63 |
63 |
63 |
63 |
63 |
| Mean |
24.22 |
10.886 |
5.625 |
67.02 |
20.96 |
17.2 |
| Std. Deviation |
2.562 |
2.3528 |
1.0842 |
5.461 |
1.992 |
3.163 |
| Median |
25 |
10.6 |
5.07 |
67.5 |
21.2 |
16.6 |
| Minimum |
19 |
5.3 |
3.2 |
50 |
17 |
13 |
| Maximum |
34 |
15.9 |
7 |
77 |
26 |
30 |
| IRON DEFICIENCY ANEMIA CASES |
N |
1187 |
1187 |
1187 |
1187 |
1187 |
1187 |
| Mean |
23.89 |
8.843 |
3.703 |
73.83 |
24.48 |
16.94 |
| Std. Deviation |
2.596 |
1.1497 |
0.2748 |
3.049 |
2.525 |
1.572 |
| Median |
24 |
8.7 |
3.7 |
74.3 |
25 |
17 |
| Minimum |
18 |
6.2 |
3 |
60 |
2 |
12 |
| Maximum |
34 |
13.9 |
4.8 |
80 |
27 |
24 |
| p value |
0.324 |
<0.001 |
<0.001 |
<0.001 |
<0.001 |
0.517 |
None of the indices studied showed 100% sensitivity or specificity. However, Shine and Lal index demonstrated highest sensitivity (98.4%) to detect β thalassemia trait but low specificity (8.46%). Kerman 1 index showed second highest sensitivity (66.7%). Mentzer index, England and Fraser Index, RDW index and Ehsani index each, showed high specificity (99.66% each). Moreover, RDW index has highest positive predictive value (90.91%), while Green and King Index have highest negative predictive value (98.87%) (Table 3). Kerman 1 index gave the correct diagnosis in 97.5% of the patients (Table 4). Youden’s index showed following ranking with respect to the indices’ ability to distinguish between β thalassemia trait and iron deficiency anemia- highest for Kerman 1 index (65.4%) ≥ RDW index (63.1%) ≥ Green and King index(56.3%) ≥ Mentzer index(42.5%) ≥ Ehsani index(36.1%) > Sirdah index (36%)≥Srivastava index (35.5%) > England and Fraser index (25%) ≥ Shine and Lal index (6.8%) > RDW (0.23%) (Table 3). .
Table 3: Sensitiv ity, specificity and other parameters of different indices for detection of b- thalassemia trait
| INDEX |
Senstivity |
Specificity |
PPV |
NPV |
LR+ |
LR- |
Accuracy |
Youden’s index |
| RDW |
7.94% |
92.29% |
5.15% |
95.00% |
1.03 |
0.99 |
88 |
0.23 |
MENTZERS
(MCV/RBC)
|
42.86% |
99.66% |
87.10% |
97.06% |
127.93 |
0.57 |
96 |
42.5 |
EFI
MCV-(5XHb)-RBC-3.4
|
25.40% |
99.66% |
80.00% |
96.20% |
75.81 |
0.18 |
95 |
25 |
SLI
MCVXMCVXMCH/100
|
98.41% |
8.46% |
5.37% |
98.02% |
1.0751 |
0.18 |
12 |
6.8 |
GKI
MCVXMCVXRDW/HbX100
|
60.32% |
96.06% |
44.71% |
98.87% |
15.323 |
0.41 |
94 |
56.3 |
SRIVASTAVA
MCH/RBC
|
36.51% |
98.99% |
65.71% |
96.73% |
36.325 |
0.64 |
99 |
35.5 |
RDW Index
MCVXRDW/RBC
|
63.49% |
99.66% |
90.91% |
98.10% |
189.52 |
0.36 |
97 |
63.1 |
SIRDAH
MCV-RBC-(3Hb)
|
36.51% |
99.58% |
82.14% |
96.75% |
87.181 |
0.63 |
96 |
36 |
EHSANI
MCV-(10XRBC)
|
36.51% |
99.66% |
85.19% |
96.75% |
108.98 |
0.63 |
96 |
36.1 |
KERMAN1
MCVXMCH/RBC
|
66.67% |
98.83% |
75.00% |
98.25% |
56.857 |
0.33 |
97 |
65.4 |
Table 4: Table depicting the correctly diagnosed cases by different RBC indices
| INDEX |
TT |
IDA |
CORRECTLY DIAGNOSED (%) |
MENTZER
β-TT<13
IDA>13
|
27
32
|
4
1190
|
1217 (97.1%)
|
RDWI
β-TT<220
IDA>220
|
38
21
|
12
1182
|
1217 (97.1%)
|
SHINE AND LAL INDEX
β-TT<1530
IDA >1530
|
58
1
|
1089
105
|
163 (13%)
|
SRIVASTAVA INDEX
β-TT<3.8
IDA>3.8
|
23
36
|
12
1182
|
1205 (96.1%) |
GREEN AND KING INDEX
β-TT<65
IDA>65
|
38
21
|
47
1147
|
1185 (94.5%) |
SIRDAH INDEX
β-TT<27
IDA>27
|
23
36
|
5
1189
|
1212 (96.7%) |
EHSANI
β-TT<15
IDA>15
|
23
36
|
4
1190
|
1213 (96.8%) |
ENGLAND FRASER INDEX
β-TT<0
IDA>0
|
16
43
|
4
1190
|
1206 (96.2%) |
RDW
β-TT<13
IDA>13
|
5
54
|
92
1102
|
1107 (88.3%) |
KERMAN1
β-TT<300
IDA>300
|
42
17
|
14
1180
|
1222 (97.5%) |
Discussion
Both β-TT and IDA have an entirely different cause, prognosis, and treatment. Hence, distinguishing them has important clinical implications. However, no single marker or any combination of tests has been found to be optimal for this discrimination [8].
Shine and Lal index demonstrated highest sensitivity in the present study (98.4%) (Table 3). This was similar to the study done by Boardbar et al [9] (87.6%) and Batebi et al [10] (83.1%). Bain (1988) reported that Shine and Lal index successfully identified 57 of 58 index pregnancies in patients with beta thalassaemia trait [11]. However, the specificity was low in the present study (8.46%).
Mentzer index in this study showed sensitivity of 42.86% with highest specificity i.e. 99.66% (Table 3). Batebi et al [10] reported sensitivity and specificity of Mentzer index as 86.3% and 85.4% respectively. Some have reported a lower sensitivity of 67% in Mentzer’s index while other studies have shown higher sensitivity with this index (82–95%) [11-14].This study showed sensitivity and specificity of RDW as 7.94% and 92.29% respectively (Table 3). Garg et al [8] showed specificity of this index as 94.8%.
Another study by Demir et al (2002) showed that Youden’s indices of RBC count and RDW were the highest, with values of 82% and 80% respectively [12]. However, in present study, Youden index was highest in Kerman 1 (65.4%) and RDW index (63.1%) (Table 3). AlFadhli et al showed that the England and Fraser index had the highest Youden’s index value (98%) for correctly differentiating b-TT and IDA, whereas the Shine and Lal index was ineffective [13]. Vehapoglu A et al [6] showed that Mentzer index had the highest Youden’s index for correctly distinguishing b-TT and IDA (81%).
The England and Fraser and the Shine and Lal indices had the lowest Youden’s index values of 51% and 10.2% respectively, in their study. In 2009, Ehsani et al showed that the best discrimination index according to Youden’s criteria was the Mentzer index (90%), followed by the Ehsani index (85%) [14].
In this study, Mentzer and Ehsani indices were able to correctly diagnose 97.1% and 96.8% cases respectively (Table 4). Ehsani et al showed that the Mentzer and Ehsani indices were able to correctly diagnose 94.7% and 92.9% of cases respectively [14].In a study by Madan et al (1999), a MCV below 80 fL and a MCH value below 27 pg were found to be very sensitive markers in the detection of β-TT, even in the presence of iron deficiency (P ≤ 0.0001) [15].
Present study showed sensitivity and specificity of Green and King index as 60.32% and 96.06% (Youden index, 56.3%) (Table 3). Ntaios et al (2007) reported that the Green and King index was the most reliable index, as it had the highest sensitivity (75.06%) and Youden’s index (70.86%) for detecting β-TT [16]. A similar result for the Green and King index (Youden’s index, 80.9%) was found by Urrechaga et al [17].
Ferrara et al (2010) demonstrated that RDWI had the highest sensitivity (78.9%), England and Fraser index had the highest specificity and highest Youden’s index (0.99 and 0.64, respectively) [18]. The differences in various studies can be due to the inter-population differences in the effectiveness of various RBC indices for discriminating β-TT from IDA and could be attributed to differences in the mutation spectrum of the thalassemia disease in different populations [19].
The ability to discriminate IDA from b-TT by different indices also depends on the age of the patient [20] and may be influenced by the pregnant state. The present study was done in pregnant women. Hence, the values may not be representative of the general population and not all indices being used worldwide were included in the present study.
The basic hematological analysis in present study was done on Sysmex KX 21 three part analyzer which cost generally between two to four lakh compared to the HPLC analyzer which costs more than 8 to 9 lakh .Moreover, due to set up issues and technical staff needed, it is practically not possible as for now to have HPLC analysis being available at many of the peripheral centers in a resource poor country. Whereas, three part analyzer is the basic analyzer needed for nearly all baseline hematological assessment of all the patient and all the hematological indices can be calculated using it using designated formulae.
Thus, considering, non availability of automation at the grass root level where most screening is being done, thalassemia indices can be used for screening the patients with low MCV and MCH values. Though HPLC definitely is gold standard for detection of thalassemia as well as several different hemoglobinopathies and especially it can detect abnormal hemoglobin variants with normal blood cell indices which can be missed by routine basic hematological analyzer, it should be made available at the ground level, for screening as well as to make early diagnosis so as to prevent any loss of life, of mother or fetus and its impact on lowering maternal mortality as well as mortality and morbidity in the general population.
Conclusion
In the present study, Shine and Lal index demonstrated highest sensitivity (98.4%). Mentzer index, England and Fraser index, RDW index and Ehsani index each, showed highest specificity (99.66% each). RDW index has highest positive predictive value (90.91%), while Green and King index has highest negative predictive value (98.87%). Youden index which tells about the overall performance of a diagnostic test was highest for Kerman 1 index (65.4%).
According to our results, the percentage of correctly diagnosed patients was highest with Kerman 1 index (97.5%) followed by RDW index (97.3%) and Mentzer index (97.1%). From the present study, we can emphasize that with the easier availability of basic automation in haematology, red cell indices have become fairly sensitive, specific, reproducible and precise and can be reliably used in the peripheral health centers where expensive investigations like electrophoresis and HPLC are not available. Moreover, it can minimize the expenditure during mass screening of β -TT. However, it should be kept in mind that several studies have shown that none of the red cell indices carry 100% sensitivity and specificity in diagnosing β -TT, but adequate utilization of combination of these indices can facilitate identification of the majority of β -TT cases at no additional cost to the healthcare system which is important in the developing country like India till HPLC or chromatography is available to all at the grass root level.
Authors' Contribution
None
Conflict of Interests
None
Ethical Considerations
None
Ethical Considerations
None
References
[1]Soliman AR, Kamal G, Elsalakawy Walaa
A, Sallam Mohamed TH. Blood indices to differentiate between β-thalassemia trait
and iron deficiency anemia in adult healthy Egyptian blood donors. The Egyptian
Society of Haematology 2014; 39(2): 91-97.http://www.ehj.eg.net/text.asp?2014/39/3/91/148223
[2]Gupta A, Hattori Y, Gupta UR. Molecular genetic testing of beta thalassemia patients of Indian origin and a novel 8bp deletion mutation at codons 36/37/38/39. Genet Test 2003; 7: 163-8 [PubMed]
[3]Christianson A, Howson CP, Modell B.
March Dimes global of report on birth defects: The Hidden Toll of Dying and
Disabled Children. March of Dimes Birth Defects Foundation 2006.http://discovery.ucl.ac.uk/id/eprint/7223
[4]Adlekha S, Chadha T, Jaiswal RM, Singla
A. Screening of beta-thalassaemia trait by means of red cell indices and derived
formulae.Medical Journal of Dr. D.Y. Patil University 2013; 6(1): 71-74.
http://www.mjdrdypu.org/text.asp?2013/6/1/71/108649
[5]Sachdev R, Dam AR, Tyagi G. Detection of Hb variants and hemoglobinopathies in Indian population using HPLC: Report of 2600 cases Indian J Pathol Microbiol 2010;53:57-62.[PubMed] [Free Full Text]
[6]. Vehapoglu A, Ozgurhan G, Demir AD, Uzuner S, Nursoy MA, Serdar Turkmen S, Kacan A. Hematological Indices for Differential Diagnosis of Beta Thalassemia Trait and Iron Deficiency Anemia. Anemia 2014; 1-7.[PubMed][Free Full Text]
[7].Youden WJ. Index for rating diagnostic tests. Cancer 1950; 3:32-35.[PubMed]
[8]Garg S, Srivastava A, Singh S, Jaiswal R, Singh YK. Role of Hematological Indices in the Screening of Β-Thalassemia Minor (Trait) and Iron Deficiency Anaemia. American Research Journal of Hematology 2016:1-5. https://www.arjonline.org/papers/arjh/v1-i1/1.pdf
[9].Bordbar E, Taghipour M, Zucconi BE. Reliability of Different RBC Indices and Formulas in Discriminating between β-Thalassemia Minor and other Microcytic Hypochromic Cases. Mediterr J Hematol Infect Dis 2015;7(1): e2015022.[PubMed][Free Full Text]
[10].Batebi A, Esmailian R. Discrimination of beta-thalassemia minor and iron deficiency anemia by screening test for red blood cell indices. Turk J Med Sci 2012. 42(2): 275- 280.
http://dergipark.gov.tr/download/article-file/128546
[11]Bain BJ. Screening of antenatal patients in a multiethnic community for thalassaemia trait. J Clin Pathol 1988; 41:481-485.[PubMed][free Full Text]
[12].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 Dec; 44(6):612-6.[PubMed]
[13]AlFadhli SM, Al-Awadhi AM, AlKhaldi D. Validity assessment of nine discriminant functions used for the differentiation between Iron deficiency anemia and thalassemia minor. Journal of Tropical Pediatrics 2007;53(2):93–97.[PubMed]
[14].Ehsani MA, Shahgholi E, Rahiminejad MS, Seighali F, Rashidi A. A new index for discrimination between iron deficiency anemia and beta-thalassemia minor: results in 284 patients. Pak J Biol Sci. 2009 Mar 1; 12(5):473-5.[PubMed]
[15].Madan N, Beachler L, Konstantinopoulos P, Worley S, Sun Z, Latson LA. Red cell indices and discriminant functions in the detection of beta-thalassaemia trait in a population with high prevalence of iron deficiency anaemia. Indian J Pathol Microbiol 1999; 42(1):55-61.[PubMed]
[16]Ntaios G, Chatzinikolaou A, Saouli Z. Discrimination indices as screening tests for beta-thalassemic trait. Annals of Hematology 2007; 86(7): 487–491.[PubMed
[17].Urrechaga E, Borque L, Escanero JF. The role of automated measurement of RBC subpopulations in differential diagnosis of microcytic anemia and β-thalassemia screening. American Journal of Clinical Pathology 2011; 135(3): 374- 379.[PubMed]
[18]. 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(2):112–115.[PubMed]
[19]Rosatelli C, Leoni GB, Tuveri T. Heterozygous β- thalassemia: relationship between the hematological phenotype and the type of β-thalassemia mutation. American Journal of Hematology 1992(39): 1–4.[Pubmed]
[20]. Rahim F, Keikhaei B. Better differential diagnosis of iron deficiency anemia from beta thalassemia trait. Turk J Hematol 2009; 26:138-45.[PubMed]
