Hematopathology / β-THALASSEMIA TRAIT
Usefulness of Cell Counter–Based Parameters and
Formulas in Detection of β -Thalassemia Trait in Areas
of High Prevalence
Dinesh A. Rathod, MD,1 Amarjeet Kaur, MD,2 Vinod Patel, MSc,2 Kunjal Patel, MD,2
Raviraj Kabrawala, MD,2 Viral Patel, MD,2 Mohak Patel, MSc,2 and Pranay Shah, MD3
Key Words: β-Thalassemia trait; High-performance liquid chromatography; HPLC; Cell counter; Iron deficiency anemia; Hemoglobin A2
DOI: 10.1309/R1YL4B4BT2WCQDGV
Abstract
The present study aimed to retrospectively evaluate
the usefulness of cell counter–based parameters and
formulas in β-thalassemia trait (BTT) detection. The
study included 170 BTT cases (hemoglobin [Hb]A2
>4.0% [0.04]) and 30 non-BTT cases (HbA2, 2.3%-
3.5% [0.02-0.04]). Depending on the hemoglobin level
and iron deficiency, the BTT group was further
classified into classic BTT (n = 112) and BTT with iron
deficiency anemia (n = 58). The RBC count, MCH,
MCV, RDW, and Shine and Lal, Mentzler, Srivastava,
England and Fraser, Ricerca, and Green indexes were
applied. For the first time in the population of India,
these 10 cell counter parameters and manual formulas
were compared with high-performance liquid
chromatography–derived HbA2 levels for deriving a
cost-effective alternative method; and receiver
operating characteristic curves were applied. We found
that the Shine and Lal, Srivastava, and Mentzler
indexes, MCV, and MCH have better discriminative
function than the RBC count and red cell distribution
width and their related formulas.
© American Society for Clinical Pathology
Downloaded from https://academic.oup.com/ajcp/article/128/4/585/1760184 by guest on 22 March 2024
Thalassemia is a genetic disorder affecting synthesis of
the globin moiety of the hemoglobin molecule. β-Thalassemia
is prevalent in a broad belt extending from the Mediterranean
basin to Southeast Asia. It is estimated that 1.5% of the
world’s population carries β-thalassemia, ie, at least 80 to 90
million people with an estimated 60,000 new carriers born
each year. The Southeast Asian region (which includes India,
Thailand, and Indonesia) accounts for about 50% of the
world’s carriers, ie, around 40 million people and almost half
of homozygous births, whereas in the developed world,
Europe and the Americas jointly account for just 10% to 13%
of the world’s carriers.1 The prevalence of β-thalassemia trait
(BTT) is about 3.3% in India. In various parts of India, the
prevalence of BTT is different: 6.5% in Punjab, 8.4% in
Tamilnadu, 4.3% in south India, and 3.5% in Bengal. β-
Thalassemia has a high prevalence in some communities, such
as Sindhi, Luvana, Tribes, and Rajputs. The incidence of BTT
in Gujarat is 10% to 15% in these communities, whereas the
incidence in the general population is 2% to 3%.2
Because a majority of people with BTT are asympto-
matic, they may not be aware of their carrier state. BTT is
associated with mild or no anemia but with reduced mean cor-
puscular volume (MCV) mean corpuscular hemoglobin
(MCH) values and an elevated hemoglobin A2 (HbA2) level.
The prevention of this homozygous condition can be
achieved through the detection and education of heterozygous
carriers. One way of achieving this goal is to screen the popu-
lation at risk.3 The screening for BTT can be done by naked-
eye single-tube red cell osmotic fragility (NESTROF) test,
cell counter–based formulas, HbA2 by microcolumn chro-
matography, elution following cellulose acetate electrophore-
sis, and high-performance liquid chromatography (HPLC).
Am J Clin Pathol 2007;128:585-589 585
585 DOI: 10.1309/R1YL4B4BT2WCQDGV 585
Rathod et al / β-THALASSEMIA TRAIT
The NESTROF test has poor sensitivity and specificity and is
difficult to standardize. Microcolumn chromatography is time-
consuming and, therefore, not suitable for large-scale screening.
Elution following cellulose acetate electrophoresis is not reli-
able or accurate. HPLC is considered a standard method of
screening, but it is costly and not available routinely. Screening
by cell counter–based parameters and formulas is rapid, auto-
mated, inexpensive, and technically sound. At present, cell
counters are widely used in routine practice, so screening can be
done without additional costs to medical systems.1,3 Successful
prevention programs for BTT in Greece and Italy have relied on
screening by RBC indices and HbA2 concentration.4
Materials and Methods
We retrospectively analyzed 200 cases that were sent to
Greencross Voluntary Blood Bank, Pathology and RIA
Laboratory, Ahmedabad, India, for HbA2 estimation by HPLC
during January 2005 to March 2006. For the study, 3 mL of
venous blood was collected in EDTA tubes. The CBC count
was done on the Cell-Dyn 3200 (Abbott, Santa Clara, CA) fully
automated 5-part differential cell counter, working on the prin-
ciple of light scattering. The 2-level controls were run every day
in the cell counter, and the counter was maintained according to
the manufacturer’s instructions. The HbA2 by HPLC was done
on the D-10 VARIANT (Bio-Rad Laboratories, Hercules, CA).
It is a fully automated cation exchange HPLC system. The
instrument was maintained according to the manufacturer’s
instructions, and the controls were run every day.
Subjects with an HbA2 level of 4.0% (0.04) or more were
classified in the BTT group (n = 170), and subjects with an
HbA2 level between 2.3% and 3.5% (0.02-0.04) in the non-
BTT group (n = 30). Patients with evidence of an acute or a
chronic inflammatory disorder, malignancy, recent surgery,
pregnancy, or an abnormal hemoglobin level on HPLC were
excluded from the study. Based on transferrin saturation (100
× serum iron concentration/total iron binding capacity) and the
serum ferritin level, the BTT group was further subdivided into
❚Table 1❚
Laboratory Values in Subjects With or Without BTT*
Parameter All BTT (n = 170)
Hemoglobin (g/L) 103.4 ± 0.15 (33-141)
RBC count (×1012/L) 5.52 ± 0.07 (1.44-7.57)
MCV (fL) 62.99 ± 0.52 (42.1-98.5)
MCH (pg) 18.83 ± 0.20 (10-33.8)
RDW (%) 15.58 ± 0.28 (11.1-31.0)
Hemoglobin A2 (%) 5.84 ± 0.06 (4.0-7.9)
BTT, β-thalassemia trait; IDA, iron deficiency anemia; MCH, mean corpuscular hemoglobin; MCV, mean corpuscular volume; RDW, red cell distribution width.
* Data are given as mean ± SE (range) in Système International (SI) units for hemoglobin, RBC count, MCV, and MCH; hemoglobin A is given in conventional units.
Conversions to conventional units are as follows: hemoglobin (g/dL), divide by 10.0; RBC count (×106/µL), divide by 1.0; and MCV (µm3), divide by 1.0; conventional and SI
units for MCH are the same. To convert hemoglobin A2 values to SI units (proportion of 1.0), multiply by 0.01.
586 Am J Clin Pathol 2007;128:585-589
586 DOI: 10.1309/R1YL4B4BT2WCQDGV
classic BTT (n = 112) and BTT with iron deficiency anemia
(IDA) (n = 58). The criteria to diagnose iron deficiency were
a transferrin saturation less than 16% and/or a serum ferritin
level of less than 16 ng/mL.
Serum iron level,5 total iron binding capacity,6 and serum
ferritin level7 were done in each case. We applied the follow-
ing cell counter–based formulas: F1, Shine and Lal Index:
MCV × MCV × MCH/1008; F2, Mentzler Index: MCV/RBC
count9; F3, Srivastava Index: MCH/RBC count10; F4, England
and Fraser Index: MCV– (5 × Hemoglobin) – RBC11; F5,
Ricerca Index: RDW/RBC count12; and F6, Green Index:
MCV × MCV × RDW/(Hemoglobin × 100).13 The RBC
Downloaded from https://academic.oup.com/ajcp/article/128/4/585/1760184 by guest on 22 March 2024
count, MCH, RDW, and MCV were applied in each case.
Sensitivity and specificity were calculated for each parameter
and formula. Receiver operating characteristic (ROC) curves
were plotted for each parameter and formula to check discrim-
inative efficacy, for deriving new cutoffs, and to select the best
parameter and formula for BTT detection.
Results
The mean and SE were computed in all groups for hemo-
globin, RBC count, MCV, MCH, RDW, and HbA2 ❚Table
1❚.The Student t test showed significant differences in RBC
count (P < .0001), hemoglobin (P < .0001), and RDW (P <
.0001) in classic BTT vs BTT with IDA. HbA2 showed a mar-
ginal difference (P = .035), whereas MCH and MCV did not
show significant differences (P = .16 and P = .99, respective-
ly). In all BTT cases and the non-BTT group, all parameters
except RDW showed significant differences (P < .0001). In
the classic BTT and non-BTT groups, the RBC count, MCV,
and MCH showed significant differences (P < .0001). In cases
of BTT with IDA and the non-BTT group, all parameters
(hemoglobin, MCV, MCH, HbA2, and RDW) showed signifi-
cant differences (P < .0001), except the RBC count (P = .08).
As depicted in ❚Table 2❚, F1, F2, and F3 formulas, MCV,
and MCH showed good specificity and sensitivity in both
groups of BTT. ROC curves were plotted to obtain new cutoff
Classic BTT (n = 112) BTT With IDA (n = 58) Non-BTT (n = 30)
115.4 ± 1 (97-148) 81.9 ± 0.2 (33-99.3) 119 ± 0.28 (97-148)
5.63 ± 0.07 (3.15-7.57) 4.62 ± 0.12 (1.44-6.23) 4.2 ± 0.1 (5.67-4.28)
67.72 ± 0.89 (42.1-96.1) 62.75 ± 1.16 (43.5-98.5) 85.3 ± 1.07 (73.7-96.1)
20.93 ± 0.36 (10-33.9) 18.39 ± 0.48 (10.2-33.8) 27.9 ± 0.45 (22.7-33.9)
14.47 ± 0.28 (9.75-33.2) 17.65 ± 0.52 (13.2-31.0) 14.3 ± 0.85 (9.7-33.2)
5.42 ± 0.12 (4.0-7.9) 5.76 ± 0.11 (4.2-7.7) 3.02 ± 0.06 (2.3-3.5)
2
© American Society for Clinical Pathology
 Hematopathology / ORIGINAL ARTICLE

❚Table 2❚
True-Positive Cases in the Classic BTT and BTT With IDA Groups and False-Positive Cases in the Non-BTT Group*

Parameters and Formulas (Cutoffs) Classic BTT (n = 112) BTT With IDA (n = 56) Non-BTT (n = 30)

RBC count (>5 × 1012/L) 107 (95.5) 24 (41) 1 (3)

MCV (<76.5 fL) 107 (95.5) 55 (95) 8 (27)
MCH (<27 pg) 111 (99.1) 55 (95) 2 (7)
RDW(<13.6%) 56 (50.0) 2 (3) 17 (57)
F1 (<1,530) 111 (99.1) 57 (98) 2 (7)
F2 (<13.0) 102 (91.1) 30 (52) 0 (0)
F3 (<3.8) 99 (88.4) 29 (50) 0 (0)
F4 (<0) 82 (73.2) 5 (9) 3 (10)
F5 (<3.3) 104 (92.9) 25 (43) 18 (60)
F6 (<72) 104 (92.9) 32 (55) 9 (30)

BTT, β-thalassemia trait; IDA, iron deficiency anemia; MCH, mean corpuscular hemoglobin; MCV, mean corpuscular volume; RDW, red cell distribution width.

Downloaded from https://academic.oup.com/ajcp/article/128/4/585/1760184 by guest on 22 March 2024

* Data are given as number (percentage). For conversions from Système International units for RBC count, MCV, and MCH, see the footnote for Table 1. The formulas are as

follows: F1, Shine and Lal Index: MCV × MCV × MCH/1008; F2, Mentzler Index: MCV/RBC count9; F3, Srivastava Index: MCH/RBC count10; F4, England and Fraser Index:
MCV– (5 × Hemoglobin) – RBC11; F5, Ricerca Index: RDW/RBC count12; and F6, Green Index: MCV × MCV × RDW/(Hemoglobin × 100).

points with higher sensitivity and specificity for every formu-
la and parameter, as depicted in ❚Table 3❚. After computing
multiple ROC curves and considering the area under the curve
near to 1 as the best, we derived the best parameter and formu-
la for the detection of BTT.
Discussion
In the population of India, IDA and BTT are common caus-
es of a microcytic hypochromic blood picture. In the present
study, an elevated RBC count with mild anemia was indicative
of BTT. An RBC count of more than 5.0 × 106/µL (5.0 ×
1012/L) was observed in more than 95% of cases in the classic
BTT group, 41% of cases in the BTT with IDA group, and
only 1 case in the non-BTT group. The mean ± SE RBC count
in the BTT with IDA group was 4.62 ± 0.12 × 106/µL (4.62 ±
0.12 × 1012/L). If we apply correction in the RBC count by a
multiplication factor derived by comparing individual hemo-
globin levels with mean hemoglobin levels, the sensitivity of
the RBC count is increased for detecting BTT. An increased
RBC count despite a low hemoglobin concentration in BTT
has been reported.14,15

❚Table 3❚
Sensitivity and Specificity of Various Parameters and Formulas in Classic BTT and BTT With IDA Groups*

Classic BTT BTT With IDA

Parameters
and Formulas Cutoff Sensitivity (%) Specificity (%) Cutoff Sensitivity (%) Specificity (%)

RBC (× 1012/L) >5 95.5 90.0 >5 41.4 90.0

>5.28† 91.1 96.7 >4.81† 48.3 86.7
MCV (fL) <76 93.7 96.6 <76 94.82 93.33
<71.3† 93.7 100 <76.6† 96.6 93.33
MCH (pg) <27 97.3 100 <27 96.6 66.7
<22.5† 99.1 66.7 <22.2† 94.8 100
RDW (%) <13.6 46.4 56.7 <13.6 3.4 43.33
<12.4† 83.0 50.0 >12.5† 100 53.3
F1 <1,530 99.01 93.33 <1,530 94.8 70
<1,204† 96.4 100 <1,232† 94.8 100
F2 <13 92 100 <13 53 100
<15.2† 96.4 96.7 <16.4† 79.3 90.0
F3 <3.8 89.3 100 <3.8 51.7 100
<4.2† 95.5 100 <5.5† 82.8 86.7
F4 <0 72.3 96.7 <0 10.3 96.7
<7.08† 92.9 93.3 <18.8† 77.6 60.0
F5 <3.3 92.9 36.7 <3.3 41.6 40.0
<6.0† 86.6 66.7 >3.72† 43.1 80.0
F6 <72 92.85 70.0 <72 55.17 40.0
<62.8† 91.1 90.0 <71.9† 55.2 70.0

BTT, β-thalassemia trait; IDA, iron deficiency anemia; MCH, mean corpuscular hemoglobin; MCV, mean corpuscular volume; RDW, red cell distribution width.
* For conversions from Système International units for RBC count, MCV, and MCH, see the footnote for Table 1. The formulas are as follows: F1, Shine and Lal Index: MCV ×

MCV × MCH/1008; F2, Mentzler Index: MCV/RBC count9; F3, Srivastava Index: MCH/RBC count10; F4, England and Fraser Index: MCV– (5 × Hemoglobin) – RBC11; F5,
Ricerca Index: RDW/RBC count12; and F6, Green Index: MCV × MCV × RDW/(Hemoglobin × 100).13
† Revised cutoff as per receiver operating characteristic curve.

© American Society for Clinical Pathology Am J Clin Pathol 2007;128:585-589 587

587 DOI: 10.1309/R1YL4B4BT2WCQDGV 587
Rathod et al / β-THALASSEMIA TRAIT
The BTT group had significantly lower values for MCV
and MCH (P < .001) compared with non-BTT cases.
However, in the classic BTT and BTT with IDA groups, the
MCV and MCH values did not show significant differences
(P = .99 and P = .16, respectively). In the present study, an
MCV of less than 76.5 µm3 (76.5 fL) was reported in 95.5%
of the classic BTT cases, 95% of BTT with IDA cases, and
27% of non-BTT cases. An MCH value of less than 27 pg
was observed in 99.1% of classic BTT cases, 95% of BTT
with IDA cases, and 7% of non-BTT cases. Similar results
have been reported earlier.16-18 The BTT cases showed
reductions in MCV and MCH values. However, the values
did not correlate with the degree of anemia.16,17 MCV and
MCH values were thus found to be important parameters for
detecting BTT.
Owing to a high prevalence of IDA in our population,
lower cutoff values should be considered to increase the sen-
sitivity and specificity of BTT detection. The proposed cut-
offs based on ROC curve analysis were 71.3 µm3 (71.3 fL)
for MCV and 22.5 pg for MCH. Lafferty et al19 have pro-
posed a similar revised cutoff for MCV, which is based on
ROC curve results.
Six formulas were calculated in each case. The best for-
mula was that of Shine and Lal,8 which identified 99.1% of
cases of classic BTT and 95% of cases of BTT with IDA. The
Mentzler,9 Srivastava,10 England and Fraser,11 Ricerca,12 and
Green13 indexes identified 91.1%, 88.4%, 73.2%, 92.9%, and
92.9% of classic BTT cases and 52%, 50%, 9%, 43%, and
55% of BTT with IDA cases, respectively. The poor sensitiv-
ity of formulas in the BTT with IDA cases may be due to sig-
nificant differences noted in the RBC count, hemoglobin
level, and RDW in the classic BTT and BTT with IDA
groups (P < .0001 for all). Madan et al18 observed 97.9%,
88.7%, 89%, and 83.4% sensitivity in classic BTT cases and
99.2%, 90.5%, 89.7%, and 80.1% sensitivity in BTT with
IDA cases for the Shine and Lal,8 Mentzler,9 Srivastava,10
and England and Fraser11 indexes, respectively. This differ-
ence may be due to different cutoffs used by Madan et al,18
which were 14 or less for the Mentzler Index, 4.4 or less for
the Srivastava Index, and 8.4 or less for the England and
Fraser Index. If the cutoffs are changed, the results are com-
parable with those of Madan et al.18 In the present study, the
Ricerca and Green indexes detected 92.9% and 92.9% of
classic BTT cases and 43% and 55% in BTT with IDA cases,
respectively. d’Onofrio et al20 observed 85.6% and 90.0%
sensitivity for Ricerca and Green indexes, respectively.
In present study, the Shine and Lal Index8 was the best
method for detecting classic BTT status, followed by MCV,
MCH, the Srivastava Index,10 the Mentzler Index,9 RBC
count, the England and Fraser Index,11 the Green Index,13
the Ricerca Index,12 and RDW, in decreasing order of pref-
erence. Similar observations were made by Lafferty et al19
588 Am J Clin Pathol 2007;128:585-589
588 DOI: 10.1309/R1YL4B4BT2WCQDGV
for the Shine and Lal Index,8 the Mentzler Index,9 MCV, the
England and Fraser Index,11 RBC, and RDW. Madan et al18
also noted superior discriminative ability of the Shine and
Lal Index8 over the Srivastava Index,10 the Mentzler Index,9
and the England and Fraser Index.11 The Shine and Lal
Index8 is also a reliable method for detecting BTT status in
the presence of IDA, followed by MCV, MCH, the
Srivastava Index,10 the Mentzler Index,9 RDW, the England
and Fraser Index,11 RBC, the Ricerca Index,12 and the Green
Index.13 Similar observations were reported by Madan et al18
for the Shine and Lal Index,8 MCV, MCH, the Srivastava
Index,10 the Mentzler Index,9 and the England and Fraser
Downloaded from https://academic.oup.com/ajcp/article/128/4/585/1760184 by guest on 22 March 2024
Index11 in BTT with IDA cases. MCV and MCH values and
their related formulas showed better discriminative function
than RBC and RDW values and their related formulas.
A higher prevalence of iron, folic acid, and vitamin B12
deficiencies along with chronic disorders in the population
of India affects the blood parameters and indexes. Therefore,
the cutoffs should be revised (in the population of India)
using ROC curve analysis. The cutoffs also depend on the
principle of the instrument on which the relevant analysis
was done. Kotwal et al21 have reported similar revised cut-
offs in the population of India for MCV (≤76 µm3 [76 fL]),
RBC count (≥4.9 × 106/µL [4.9 × 1012/L]), and RDW
(≥18%). Lafferty et al19 also proposed a revised cutoff for
MCV (<72.0 µm3 [72.0 fL]) by using ROC curve analysis. In
the present study, revised cutoffs for classic BTT and BTT
with IDA, respectively, were as follows: MCV, less than 71.3
µm3 (71.3 fL) and less than 76.6 µm3 (76.6 fL); RBC count,
5.28 × 106/µL (5.28 × 1012/L) or more and more than 4.81 ×
106/µL (4.81 × 1012/L); and RDW, less than 12.4% and more
than 12.5%.
None of the formulas and parameters can be 100% sen-
sitive and specific, maybe due to atypical β-thalassemia car-
riers, characterized by normal MCV and MCH values (α-
and β-thalassemia interaction) and a normal HbA2 level
(coinheritance of δ- and β-thalassemia, some mild β-gene
mutations, and γδβ-thalassemia). Cell counter–based formu-
las and parameters are, therefore, important only for screen-
ing in areas of high prevalence of BTT, and these cases
should undergo further confirmatory tests. The order of pref-
erence for detection of BTT using formulas and cell counter
parameters are proposed ❚Table 4❚, with F1, MCV, and MCH
leading the list.
Conclusions
From the present study, we concluded that automated
cell counter–based parameters and formulas are technically
good, rapid, cheaper, easily available, and reliable methods for
BTT detection. Cell counter–based parameters and formulas,
© American Society for Clinical Pathology

❚Table 4❚
Order of Preference for Detection of BTT Using Formulas and
Cell Counter Parameters*
BTT BTT With IDA
F1 F1
MCV MCV
MCH MCH
F3 F3
F2 F2
RBC RDW
F4 F4
F6 RBC
F5 F5
RDW F6
BTT, β-thalassemia trait; IDA, iron deficiency anemia; MCH, mean corpuscular
hemoglobin; MCV, mean corpuscular volume; RDW, red cell distribution width.
* The formulas are as follows: F1, Shine and Lal Index: MCV × MCV × MCH/1008;
F2, Mentzler Index: MCV/RBC count9; F3, Srivastava Index: MCH/RBC count10;
F4, England and Fraser Index: MCV– (5 × Hemoglobin) – RBC11; F5, Ricerca
Index: RDW/RBC count12; and F6, Green Index: MCV × MCV ×
RDW/(Hemoglobin × 100).13
particularly MCV and MCH and their related formulas (Shine
and Lal, Srivastava, and Mentzler indexes), have good discrim-
inative function compared with RBC and RDW and their relat-
ed formulas (Green and Ricerca indexes) for the diagnosis of
BTT status. However, the cutoffs for these parameters and for-
mulas should be revised to increase the diagnostic sensitivity
and specificity in the population of India.
From the Departments of Pathology, 1Smt NHL Municipal Medical
College, Ahmedabad, India; 2Greencross Voluntary Blood Bank,
Pathology and RIA Laboratory, Ahmedabad; and 3B.J. Medical
College, Ahmedabad.
Address reprint requests to Dr Rathod: D-502 SUPATH-II
Apt, near Oldvadaj AMTS bus terminus, Ashram Rd, Oldvadaj,
Ahmedabad-13, Gujarat, India.
Acknowledgments: We acknowledge the expert efforts of Rakesh
Raval, MD, Mina Dalal, Alpa Patel, Ankita Gandhi, Dipavali
Bhatt, Adarshjeet Singh, MD, Bipin Patel, MD, N Jesalpura, MD,
Nita Ninama, MD, Imran Shaikh, MD, and Minesh Gandhi, MD.
References
1. Angastinotis M. Epidemiology. In: Galanello R, Eleftheriou
A, Traeger-Synodinos J, et al, eds. Prevention of Thalassemias
and Other Haemoglobin Disorders. Vol I. Nicosia, Cyprus:
Thalassemia International Federation Publication; 2005:10-13.
2. Ambek SS, Phadke MA, Mokashi GD, et al. Pattern of
hemoglobinopathies in western Maharastra. Indian Pediatr.
2001;38:530-535.
3. Shalev O, Yehezkel E, Rachmilewitz EA. Inadequate
utilization of routine electronic RBC counts to identify beta
thalassemia carriers. Am J Public Health. 1993;78:1476-1477.
© American Society for Clinical Pathology
Hematopathology / ORIGINAL ARTICLE
4. Tan GB, Aw TC, Dunstan RA, et al. Evaluation of high
performance liquid chromatography for routine estimation of
hemoglobins A2 and F. J Clin Pathol. 1993;46:852-856.
5. Tietz NW. Clinical Guide to Laboratory Tests. 3rd ed.
Philadelphia, PA: Saunders; 1995:374-375.
6. Baadenhuinjsen H. Modification in Ramsay’s method for
correct measurement of TIBC. Clin Chem. 1998;175:9-16.
7. Franco CD. Ferritin. In: Kaplan LA, Pesce AJ, eds. Methods in
Clinical Chemistry. 2nd ed. St Louis, MO: Mosby; 1987:1240-
1242.
8. Shine I, Lal S. A strategy to detect beta thalassemia minor.
Lancet. 1977;1:692-694.
9. Mentzler WC Jr. Differentiation of iron deficiency from
thalassaemia trait. Lancet. 1973;1:882.
Downloaded from https://academic.oup.com/ajcp/article/128/4/585/1760184 by guest on 22 March 2024
10. Srivastava PC. Differentiation of thalassaemia minor from iron
deficiency [letter]. Lancet. 1973;2:154.
11. England JM, Fraser PM. Differentiation of iron deficiency from
thalassaemia trait by routine blood-count. Lancet. 1973;1:449-
452.
12. Ricerca BM, Stortis, d’Onofrio G, et al. Differentiation of iron
deficiency from thalassemia trait: a new approach.
Haematologica. 1987;72:409-413.
13. Green R, King R. A new red cell discriminant incorporating
volume dispersion for differentiating iron deficiency anaemia
from thalassemia minor. Blood Cells. 1989;68:506-513.
14. Lec GR. Microcytosis and the anemias associated with
impaired hemoglobin synthesis. In: Greer JP, Foerster J, Lukens
JN, eds. Wintrobe’s Clinical Hematology. 9th ed. Philadelphia,
PA: Lippincott Williams & Wilkins; 1993;1:791-807.
15. McDonagh KT, Nienhuis AW. The thalassaemias. In: Nathan
DG, Oski FA, eds. Hematology of Infancy and Childhood. 4th
ed. Philadelphia, PA: Saunders; 1993:783-880.
16. Beutler E, Lichtmann MA, Coller BS, et al. Iron deficiency.
In: Williams Hematology. 5th ed. New York, NY: McGraw-Hill;
1995:490-511.
17. Klee GG, Fairbanks VF, Pierre RV, et al. Routine erythrocyte
measurements in diagnosis of iron-deficiency anemia and
thalassemia minor. Am J Clin Pathol. 1976;66:870-877.
18. Madan N, Sikka M, Sharma S, et al. Red cell indices and
discriminant function in the detection of beta thalassaemia
trait in a population with high prevalence of iron deficiency
anaemia. Indian J Pathol Microbiol. 1999;42:55-61.
19. Lafferty JD, Crowther MA, Ali MA, et al. The evaluation
of various mathematical RBC indices & their efficacy in
discriminating between thalassemic and non-thalassemic
microcytosis. Am J Clin Pathol. 1996;106:201-205.
20. d’Onofrio G, Zini G, Bianca M, et al. Automated
measurement of red blood cell microcytosis and hypochromia
in iron deficiency and β-thalassemia trait. Arch Pathol Lab
Med. 1992;116:84-89.
21. Kotwal J, Saxena R, Choudhry VP, et al. Erythrocyte indices
for discriminating thalassaemic and non-thalassaemic
microcytosis in Indians. Natl Med J India. 1999;12:266-267.
Am J Clin Pathol 2007;128:585-589 589
589 DOI: 10.1309/R1YL4B4BT2WCQDGV 589