Keywords
pregnancy - anemia - reticulocyte hemoglobin content - immature reticulocyte fraction
Introduction
Pregnancy has always been a puzzle for both pathologists and clinicians in interpreting
any hematological or biochemical parameters due to marked physiological and associated
plasma volume changes. According to World Health Organization (WHO) 40.1% of total
pregnant females are anemic worldwide, and the burden lies mostly on developing and
under developed countries.[1] Out of these, approximately 50% of anemia is due to iron deficiency with Southeast
Asian and African countries contributing about half of global maternal deaths. Therefore,
it is essential to diagnose iron deficiency anemia (IDA) as early as possible in pregnancy
so as to prevent complications to both mother and the baby. It is also important that
the required investigations should not only be simple, safe, feasible but also cost
effective for early diagnosis of IDA. Although the gold standard for diagnosing IDA
is to evaluate bone marrow iron stores and serum ferritin but it has its own limitations
in pregnancy. Bone marrow iron stores should not usually be evaluated in pregnancy
because it is an invasive procedure while serum ferritin shows variation in thresholds
depending on various physiological changes in pregnancy.[2]
[3] Still, serum ferritin has been used routinely and threshold level of less than 15
ng/mL is usually used to define IDA in pregnancy.[4]
[5] Recently, different reticulocyte parameters are being studied in the diagnosis of
IDA in various age groups but their analysis exclusively in pregnancy is rarely reported.[6]
[7] The major advantage of these reticulocyte parameters is that it can be evaluated
easily with recent automated hematology analyzers using routine blood samples. The
feasibility and the ease of use of reticulocyte parameters may also prove beneficial
in evaluation of IDA in pregnancy. The present study was therefore conducted to evaluate
reticulocyte parameters including reticulocyte hemoglobin content (Ret-He), immature
reticulocyte fraction (IRF), and reticulocyte count (RC) in diagnosing IDA in pregnancy.
In addition, it was also intended to compare these parameters with serum ferritin,
mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) in assessment
of IDA in pregnant females.
Material and Methods
Prospective study was undertaken in the hematology section of the institute which
included first trimester pregnant females who came for routine obstetrical examination
followed by laboratory investigations. The blood samples were collected in ethylene
diamine tetra-acetic acid (EDTA) anticoagulant tubes and plain vials after written
informed consent. Complete clinical details were taken from all the females at the
time of phlebotomy. The EDTA sample was subjected for complete blood count analysis,
using Sysmex-XN 1000 Automatic Full Digital Cell Counter (Sysmex Corporation). Hemoglobin
level (Hb), MCV, MCH, RC, IRF, and Ret-He that were available on automated analyzer
were recorded for every patient. The serum was used for estimation of serum ferritin
using ADVIA Centaur XP semi-automated analyzer ( Siemens Healthineers) (immunoassay
via chemiluminescence). All the samples were subjected to cell counter within 6 hours
of collection, but not at a fixed time. Cases that were on hematinic were excluded
from the study. According to WHO, females having hemoglobin less than 11 g/dL were
considered anemic and according to United Kingdom guidelines on the management of
iron deficiency in pregnancy threshold of serum ferritin less than 15 ng/mL was considered
significant for IDA.[4]
[8] All the data were entered in a spreadsheet and statistical analysis was done using
R software. The comparison of the measured reticulocyte parameters was accomplished
by boxplots and Kruskal–Wallis test (with post-hoc Mann-Whitney U test). p-Value less than 0.05 was considered statistically significant. The study was approved
by Institutional Research and Ethics Committee.
Results
The study included total 155 first trimester pregnant females with mean age group
of 26.7 years. All the females were categorized according to Hb, MCV, and MCH in 3,
4, and 2 categories, respectively as shown in [Table 1]. It shows that females with Hb less than 11 g/dL were 37, MCV less than 80 fl were
22, and MCH less than 27 pg were 46. It was observed that median serum ferritin level
was 18 ng/mL and 64 females had serum ferritin level less than 15 ng/mL. [Table 2] shows reticulocyte parameters and serum ferritin levels according to different Hb
categories. It was observed that there was statistical difference in Ret-He between
all the categories of Hb (p < 0.05) while IRF was statistically different between categories 2 and 1 and category
3 and 1 (p < 0.05). Serum ferritin was statistically higher in category 3 of Hb in comparison
to category 1 and 2 (p < 0.05). [Table 3] shows reticulocyte parameters and serum ferritin levels according to different MCV
categories. It was observed that there was statistically significant difference in
Ret-He between all the categories of MCV (p < 0.05) except between category 3 and 4. IRF did not show any statistical difference
between the different MCV categories. [Fig. 1] shows the boxplot analysis of Ret-He in different categories of Hb, MCV, and MCH.
[Table 4] shows reticulocyte parameters and serum ferritin levels according to different MCH
categories. It shows that Ret-He, IRF, and serum ferritin are statistically significant
in differentiating hypochromic and normochromic category. It was also observed that
RC was not statistically significantly different in any of the category of all analyzed
hematological parameters. The analysis of ROC curve using MCV < 83 fl and MCH < 27
pg shows that area under curve for Ret-He at cut-off of 27.8 ng/mL is 0.93 (95% confidence
interval [CI] 0.90–0.98, sensitivity 93%, specificity 83%) and for serum ferritin
at cutoff of 11 ng/mL is 0.76 (95% CI 0.66–0.87, sensitivity 76%, specificity 69%)
([Fig. 2]). The cutoff of 29.4 ng Ret-He for diagnosis of IDA (serum ferritin < 15 ng/mL)
has specificity of 87%, sensitivity 52%, positive predictive value 0.73, negative
predictive value 0.72, and likelihood ratio of 3.9. The cutoff of 27.2 ng/mL of Ret-He
for the diagnosis of IDA (serum ferritin < 15 ng/mL and Hb < 11 g/dL) as determined
by ROC curve had specificity of 86% and sensitivity of 71%. Further analysis yielded
positive predictive value 0.39, negative predictive value 0.96, and likelihood ratio
of 5.1 for diagnosing IDA.
Table 1
Categorical classification of hematological parameters
|
Parameters
|
Category I
|
Category II
|
Category III
|
Category IV
|
|
Hemoglobin (gm/dL)
|
Anemia (0–11)
Number of cases: 37
|
Borderline anemia (11.01–12)
Number of cases: 48
|
Normal (> 12.01)
Number of cases: 70
|
|
|
Mean corpuscular Volume (fl)
|
Severe microcytosis (0–70)
Number of cases: 7
|
Mild to moderate microcytosis (70.01–80)
Number of cases: 15
|
Normocytosis (80.01–100)
Number of cases: 130
|
Macrocytosis (> 100.01)
Number of cases: 3
|
|
Mean corpuscular hemoglobin (%)
|
Hypochromic (0–27)
Number of cases: 46
|
Normochromic (> 27.01)
Number of cases: 109
|
|
|
Table 2
Serum ferritin and reticulocyte parameters in different categories of hemoglobin
|
Parameters (Median value)
|
Category I
|
Category II
|
Category III
|
|
Abbreviations: IRF, immature reticulocyte fraction; RC, reticulocyte fraction; Ret-He,
reticulocyte hemoglobin.
|
|
Serum ferritin (ng/mL)
|
10.7
|
13.8
|
23.5
|
|
Ret-He (ng/mL)
|
28.6
|
30.3
|
31.7
|
|
IRF
|
13.5
|
11.2
|
9.5
|
|
RC
|
0.06
|
0.07
|
0.06
|
Table 3
Serum ferritin and reticulocyte parameters in different categories of mean corpuscular
volume
|
Parameters (Median value)
|
Category I
|
Category II
|
Category III
|
Category IV
|
|
Abbreviations: IRF, immature reticulocyte fraction; RC, reticulocyte fraction; Ret-He,
reticulocyte hemoglobin.
|
|
Serum ferritin (ng/mL)
|
7.8
|
8
|
20.3
|
62
|
|
Ret-He (ng/mL)
|
21.1
|
25.8
|
31.2
|
31.7
|
|
IRF
|
17.7
|
12.2
|
10.2
|
13.7
|
|
RC
|
0.07
|
0.06
|
0.06
|
0.08
|
Fig. 1 (a) Box plot analysis of Ret-He in different categories of hemoglobin; (b) Box plot analysis of Ret-He in different categories of MCV; (c) Box plot analysis of Ret-He in different categories of MCH. MCH, mean corpuscular
hemoglobin; MCV, mean corpuscular volume; Ret-He, reticulocyte hemoglobin.
Table 4
Serum ferritin and reticulocyte parameters in different categories of mean corpuscular
hemoglobin
|
Parameters (Median value)
|
Category I
|
Category II
|
p-Value
|
|
Abbreviations: IRF, immature reticulocyte fraction; RC, reticulocyte fraction; Ret-He,
reticulocyte hemoglobin.
|
|
Serum ferritin (ng/mL)
|
10.6
|
21.3
|
< 0.05
|
|
Ret-He (ng/mL)
|
26.3
|
31.6
|
< 0.05
|
|
IRF
|
12.5
|
10.2
|
< 0.05
|
|
RC
|
0.06
|
0.067
|
0.6
|
Fig. 2 Receptor operating characteristics curve for (a) serum ferritin and (b) Ret-He at MCV < 83fl and MCH < 27 pg. MCH, mean corpuscular hemoglobin; MCV, mean
corpuscular volume; Ret-He, reticulocyte hemoglobin.
Discussion
Ret-He is the measure of the functional iron which is available in hemoglobin of reticulocytes
for new red blood cell production while IRF is the proportion of young reticulocytes
with the highest RNA content.[8] These reticulocyte parameters have not only been studied in IDA but also in response
to therapy, hereditary spherocytosis, and pyruvate kinase deficiency anemia screening.[9]
[10] Serum ferritin, an important parameter has a varied cutoff for diagnosing IDA in
pregnancy ranging from 12 to 30 ng/mL.[5]
[11] This variation in serum ferritin in pregnancy may be attributed to different physiological
changes occurring in pregnancy that may include second trimester plasma volume expansion,
reactive acute phase proteins rise, or third trimester changes in inflammatory measures.[2]
[3]
[12] Therefore the evaluation of easily available reticulocyte parameters may be helpful
in diagnosing IDA of pregnancy. It was observed in the present study that Ret-He and
IRF were able to differentiate between categories of normal, borderline, or anemic
pregnant females according to Hb while serum ferritin observed no statistical difference
between borderline (category II) and anemic (category I) patients. The differentiation
between frank anemia and borderline anemia in pregnancy is essential to initiate an
early therapy to prevent the maternal and fetal complications and Ret-He and IRF may
be helpful in this regard. Previously, it was reported that Ret-He content is an early
and accurate predictor of hematological response to oral iron therapy in children.[13] Few recent studies have also evaluated Ret-He content to assess latent iron deficiency
and microcytic red blood cells in adult populations.[14]
[15] Chaipokam et al concluded that Ret-He content combined with reticulocyte and blood
count is useful in evaluating microcytic red blood cells in Thai patients while Tiwari
et al observed that Ret-He may be used in routine screening of latent iron deficiency
in blood donors.[14]
[15] Although evaluation of Ret-He in pregnant females has been rarely studied but it
is observed in the present study that Ret-He and IRF is able to statistically differentiate
between normochromic and hypochromic blood picture ([Table 4]) while Ret-He was also able to differentiate between severely microcytic, microcytic,
and normocytic red blood cells. This observation further suggests that Ret-He may
be helpful to differentiate between microcytic hypochromic and normocytic normochromic
anemia in pregnant females. Rabindrakumar et al also evaluated hematological indices
MCV, MCH, MCHC, and Hb in predicting early iron deficiency among pregnant females
but reticulocyte parameters were not studied.[16]
Another interesting finding observed in the present study was that cutoff of 29.4
ng Ret-He was specific (87%) for the diagnosis of IDA in pregnant females with positive
predictive value of 73%. Previous studies have observed variable cutoffs of Ret-He
in IDA ranging from 26 to 30.9 pg.[7]
[11] It has also been reported that there is an agreement between the measurements of
Ret-He on Sysmex XE 2100 and Bayer ADVIA autoanalyzers and thus eliminating the difference
based on instruments.[17]
However, an important limitation of the present study is that females with concomitant
thalassemia minor or anemia of chronic disorder (ACD) were not analyzed and therefore
the role of reticulocyte parameters in these patients cannot be predicted. Therefore,
the authors suggest that further larger studies in pregnant females may be done to
assess the role of Ret-He and IRF in diagnosing IDA coexisting with thalassemia minor
or ACD.
Conclusion
Ret-He is a feasible and easily available parameter which may be helpful in differentiating
between microcytic hypochromic and normocytic normochromic anemia along with frank
and borderline anemia in first trimester pregnant females. This in turn is essential
to initiate an early therapy so as to prevent the maternal and fetal complications.
