Erythrocyte Indices and Hemoglobin Analysis for α-Thalassemia Screening in an Area with High Carrying Rate

Abstract

Carriers of α-thalassemia exhibit hypochromic microcytosis with mean corpuscular volume (MCV) < 80 fL, mean corpuscular hemoglobin (MCH) < 27 pg, and reduced hemoglobin A₂ (HbA₂). We studied the distribution and diagnostic efficiencies of these indicators and their combinations in patients with and without alpha-thalassemia. Based on genetic diagnosis, 10,883 participants were divided into alpha-thalassemia group (n = 1655) and negative-for-alpha-thalassemia group (n = 9228). Erythrocyte parameters and hemoglobin analysis of the groups were analyzed. Moreover, we compared the four screening schemes (MCV/MCH, MCV/MCH/HbA₂, MCV + MCH, MCV + MCH + HbA₂) to find the best for α-thalassemia screening. The genotypes of --^SEA/αα, and -α^3.7/αα are the most prevalent with 54.9% and 27.6% in Fujian Province, China. There were significant differences in the distribution of MCV, MCH, and HbA₂ in the two groups. Among the three, MCH exhibited the highest sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy. Although the four screening schemes have their advantages, there are significant differences in their sensitivity and specificity. MCV + MCH had the best diagnostic performance (72.6% sensitivity, 89.0% specificity) as well as the highest Youden index (61.59%). Our results showed that MCH could be used to screen α-thalassemia instead of MCV and HbA₂. However, it is recommended that MCV/MCH/HbA₂ screening be used in areas with high α-thalassemia incidence to increased sensitivity.

Introduction

Thalassemia is one of the most common autosomal recessive hemoglobin (Hb) disorders in the world [1]. Two major forms of the disease are described as α- and β-thalassemia [2]. It is widespread throughout the world, especially in Southeast Asian countries [3–5]. In China, thalassemia is most frequent in Guangxi, Guangdong, Yunnan, Fujian, and Hainan providences [6, 7], and the incidence of α-thalassemia is higher than β-thalassemia [8–10]. While the prevalence of α-thalassemia is 3.17% in Fujian province, China, which is higher than β-thalassemia too [11]. Furthermore, α-thalassemia can lead to anemia, Hemoglobin H (HbH) disease, and hydrops fetalis syndrome [12], which creates an enormous economic burden on the family and society. Therefore, it is important to screen the carriers of α-thalassemia.

Each region has a different screening strategy because of the variety in α-thalassemia genotypes and different ethnic populations. Clinical criteria for suspected α-thalassemia was determined by the hypochromic microcytic anemia, including decreasing mean cell volume (MCV) and/or mean cell hemoglobin (MCH), and/or decreasing hemoglobin A₂ (HbA₂) measured by a robust method such as high performance liquid chromatography (HPLC) or capillary electrophoresis (CE) [13]. Genetic testing is not available in most laboratories and poses cost constraints for population screening. Hence, only couples suspected to be carriers of the disease were confirmed by a genetic test. In the present study, we performed a large-scale survey to study the distribution of MCV, MCH and HbA₂ in different groups and discover the efficiency of different combined screening programs.

Materials and Methods

Subjects and Hematological Indices Detection

This study was approved by the Medical Ethics Committee of Fujian Maternity and Child Health Hospital, affiliated hospital of Fujian Medical University (NO.2016-101). All procedures were carried out in accordance with ethical guidelines for human subject research. The population investigated in this study comprised of 10,883 individuals who were admitted to our hospital from May 2016 to July 2019. Peripheral blood samples from subjects were collected in EDTA containing tubes for analyzing hematological indices and hemoglobin content, and performing molecular testing. 9228 participants who tested negative for α- and β-thalassemia genes were categorized as normal (including 2165 men, 7063 women; median age 29.5 years). Whereas 1655 participants who tested positive for α-thalassemia and negative for β-thalassemia genes were designated as α-thalassemia carrier. All participants had no blood transfutions for one year. MCV and MCH were measured on the automated analyzer (XN3000; Sysmex, Japan).

Capillary Electrophoresis

Hemoglobin analysis was performed using the automated flex piercing hemoglobin analyzer (Capillarys2™; Sebia, France). This was used to measure the percentages of HbA and HbA₂, as well as any variants, including hemoglobin E (HbE), hemoglobin H (HbH), hemoglobin constant spring (HbCS), and hemoglobin Barts (Hb Barts).

Common Genotype Test

DNA was extracted using the DNA Blood Extraction Kit (Yaneng Biosciences, Shenzhen, China) according to the manufacturer’s instructions. The α-thalassemia deletions (--^SEA/, -α^4.2/, and -α^3.7/), the three α-thalassemia non-deletional mutations (α^csα/, α^Qsα/, and α^wsα/) and seventeen β-thalassemia mutations (including β^{CD41/42(−TCTT)}, β^{IVS-2-654(c→T)}, β^CD17(A→T), β^−28(A→G), β^CD26(G→A), β^CD71/72(+A), β^CD43(G→T), β^−29(A→G), β^{InitiationCD(ATG→AGG)}, β^CD14/15(+G), β^CD27/28(+C), β^−32(C→A), β^−30(T→C), β^{IVS-1-1(G→T)}, β^{IVS-1-5(G→C)}, β^CD31(−C), β^{CAP+40--+43(−AAAC)}) commonly found in Chinese populations were analyzed by PCR-reverse dot-blot assay using commercial kits (Yaneng Biosciences, Shenzhen, China) as described [14].

--^THAI and HKαα Genotype Test

The genotyping for --^THAI was performed by gap polymerase chain reaction (gap-PCR) assay, whereas the genotyping for HKαα was performed by single PCR and nested PCR assay as described in a previous study [15, 16].

Statistical Analysis

Data was analyzed using the statistical package for social science (SPSS) version 24 (IBM Inc, Chicago, USA). Continuous data was checked for normality using the Kolmogorov–Smirnov test. Medians and percentiles were used to describe non-normally distributed data. The box plots of MCV, MCH and HbA₂ were generated using the SPSS. Mann–Whitney U test was used to compare non-parametric variables, including MCV, MCH, and HbA₂ between the alpha-thalassemia group and the negative-for-alpha-thalassemia group. McNemar’s test was used to compare the sensitivity and specificity of the four combined screening schemes. Results with P < 0.05 were considered statistically significant.

Results

From May 2016 to July 2019, blood specimens were collected from 10,883 subjects referred to our center for thalassemia investigation. Based on the genetic analysis, 1655 subjects were found to be α-thalassemia carriers. Sixteen different genotypes were identified and are listed in Table 1.

Table 1.

Genotypes identified among α-thalassemia carriers

Genotype	Subjects (n)	Frequency (%)
--SEA/αα	909	54.9
-α3.7/αα	457	27.6
-α4.2/αα	122	7.4
αQSα/αα	51	3.1
--THAI/αα	34	2.1
α csα/αα	30	1.8
α wsα/αα	29	1.8
--SEA/-α3.7	7	0.4
--SEA/-α4.2	5	0.3
-α3.7/-α3.7	3	0.18
--SEA/αwsα	2	0.12
-α3.7/αQSα	2	0.12
-α3.7/-α4.2	1	0.06
-α3.7/α wsα	1	0.06
--SEA/α csα	1	0.06
--SEA/HKαα	1	0.06
Total	1655	100

The Kolmogorov–Smirnov test was carried out in the alpha-thalassemia group and the negative-for-alpha-thalassemia group and the results indicated that the distribution of each indicator in both groups was not normal (P < 0.05). The distributions of MCV, MCH, and HbA₂ in the α-thalassemia and the negative-for-alpha-thalassemia groups are shown in Table 2. These indices were significantly lower in the α-thalassemia group when compared with the negative-for-alpha-thalassemia group. The P values from the Mann–Whitney U test used to compare MCV, MCH and HbA₂ between the alpha-thalassemia group and the negative-for-alpha-thalassemia group are also shown in Table 2. The specific distributions of these indices are represented by the box plots shown in Fig. 1.

Table 2.

Comparison of various parameters between two groups

Indices/groups	Alpha-thalassemia group	Negative-for-alpha-thalassemia group	P
MCV (fl)	71.1 (66.9, 79.8)	84.7 (80.9, 88.1)	< 0.01
MCH (pg)	22.7 (21.2, 26.6)	29.2 (27.9, 30.2)	< 0.01
HbA2 (%)	2.4 (2.3, 2.6)	2.7 (2.5, 2.8)	< 0.01

The distribution of indices is described in the way of median (quartile)

Fig. 1.

Comparisons of MCV (a), MCH (b), and HbA₂ (c) between the alpha-thalassemia group and the negative-for-alpha-thalassemia group. The heavy lines and boxes indicate the median and interquartile range, respectively. Values below the 25th percentile and above the 75th percentile represent 50% of the data

Referred to the requirements of technical service specifications for thalassaemia prevention and control pilot projects issued by the General Office of the National Health and Family Planning Commission of China, we adopted the following judgment criteria with MCV < 80 fL, MCH < 27 pg, and HbA₂ < 2.5%. The sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy of MCV, MCH and HbA₂ are shown in Table 3. The Youden indices of the three indicators were 57.5%, 64.8% and 40.6%.

Table 3.

Screening efficiencies of various parameters and their combinations

Category	Se (%)	Sp (%)	PPV (%)	NPV (%)	Da (%)
MCV	76.1 (1260/1655)	81.4 (7510/9228)	42.3 (1260/2978)	95.0 (7510/7905)	80.6 (8770/10883)
MCH	79.9 (1323/1655)	84.9 (7830/9228)	48.6 (1323/2721)	95.9 (7830/8162)	84.1 (9153/10883)
HbA2	58.9 (974/1655)	81.7 (7536/9228)	36.5 (974/2666)	91.7 (7536/8217)	78.2 (8510/10883)
MCV/MCH	83.4	77.3	39.7	96.3	78.2
	(1381/1655)	(7131/9228)	(1381/3478)	(7131/7405)	(8512/10883)
MCV/MCH/HbA2	88.6	67.1	32.6	91.7	70.3
	(1467/1655)	(6189/9228)	(1467/4506)	(6189/6377)	(7656/10883)
MCV + MCH	72.6	89.0	54.1	94.8	86.5
	(1202/1655)	(8209/9228)	(1202/2221)	(8209/8662)	(9411/10883)
MCV + MCH + HbA2	49.8	94.3	61.0	91.3	87.5
	(825/1655)	(8701/9228)	(825/1352)	(8701/9531)	(9526/10883)

Se, sensitivity, TP /( TP + FN) × 100%; Sp, specificity, TN /( FP + TN) × 100%; PPV, positive predictive value, TP/(TP + FP) × 100%; NPV, negative predictive value, TN/(FN + TN) × 100%, Da, diagnostic accuracy, (TP + TN)/(TP + FP + FN + TN) × 100%. The positive cases of alpha-thalassemia group were expressed with TP and the negative cases of alpha-thalassemia group were expressed with FN. The positive cases of negative-for-alpha-thalassemia group were expressed with FP and the negative cases of negative-for-alpha-thalassemia group were expressed with TN. MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; HbA₂: hemoglobin A₂

Decision criteria of MCV/MCH primary screening was positive for one item or more of MCV and MCH, and MCV/MCH/HbA₂ primary screening was positive for one item or more of MCV, MCH and HbA₂. MCV + MCH screening was positive for both MCV and MCH, and MCV + MCH + HbA₂ screening was positive for all of the three indices. The comparison results of the four schemes are shown in Table 3. McNemar’s test was used to compare the sensitivity and specificity of these screening schemes. The sensitivity of all schemes significantly differed from each other (P < 0.05), MCV/MCH/HbA₂ > MCV/MCH > MCV + MCH > MCV + MCH + HbA₂. The specificity of all schemes also were different from each other (P < 0.05), MCV + MCH + HbA₂ > MCV + MCH > MCV/MCH > MCV/MCH/HbA₂. The Youden indices of the four screening schemes were 60.72%, 55.71%, 61.59%, and 44.14% respectively.

Discussion

Despite being the most routinely reported hereditary hematological disorder in Fujian Province, little was known about the distribution of α-thalassemia in this area [17]. The genotype of --^SEA/αα was predominant in the adjacent high incidence provinces such as Guangdong Province [18], Guangxi Zhuang Autonomous Region [9], and Hainan Province [10]. A previous study also found --^SEA/αα as the most prevalent genotype in Fujian province [14]. Our study confirms these findings but found different carrier proportions. In this study, --^SEA/αα, and -α^3.7/αα were the most popular genotypes with 54.9% and 27.6% prevalence, respectively. We speculate that the difference may be due to the larger sample size and greater number of genotypes examined than in the previous study.

Further, in our study, MCV, MCH, and HbA₂ were lower in the α-thalassemia group than in the negative-for-alpha-thalassemia group. The three indicators were significantly different between the two groups (P < 0.05). Compared to MCV, MCH showed higher sensitivity and specificity. This was consistent with the previously reported findings of β-thalassemia [19]. Also, it has been noted that MCH is much more stable than MCV during storage of blood specimens [20]. Moreover, MCH is less influenced by age [13]. Hemoglobin electrophoresis is one of the main screening methods for thalassemia. There have been reports that most of β-thalassemia carrier have elevated HbA₂ levels [21, 22]. It has high sensitivity and resolution in β-thalassaemia. But in our study, only 58.9% of α-thalassemia carriers showed HbA₂ reduction which indicated that it may not be a perfect index for α-thalassemia. Our study concluded that MCH was the best of the three indices in terms of sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy. However, it is worth noting that other factors, such as iron deficiency [23, 24], reticulocytosis, megaloblastic anemia, drugs and liver disease [25] may affect these indices.

There are two types of HbH disease; deletional and non-deletional HbH [26, 27]. The non- deletional type often shows more severe clinical symptoms. In our study, 13 deletional and 3 non-deletional HbH disease cases were found. One of the non-deletional HbH(--^SEA/α ^csα) participants suffered from severe anemia and required irregular blood transfusions. In the hemoglobin analysis of HbH disease, not only the decrease of HbA₂, but also the peak of HbH and HbBarts may appear in the electrophoresis pattern. In this study, 10 cases of deletion HbH disease and 1 case of non-deletion HbH disease showed these two peaks. While they were not found in both --^SEA/α^wsα cases. It was speculated that which was caused by the different degree of imbalance between alpha and beta chains.

In the present study, we mainly used the index of HbA₂ to conduct the statistical analysis of sensitivity and specificity of capillary electrophoresis, but some genotypes associated with α-thalassemia, such as α^csα/αα, may show an abnormal bands (CS band appeared in zone 2), which also suggested α-thalassemia. For example, of the 30 participant carrying α^csα/αα, only 27 of them showed a decrease in HbA₂, whereas the other three showed abnormal CS bands, which did not affect the diagnosis. Furthermore, when thalassemia is combined with a hemoglobin variant, the condition becomes more complex. Hemoglobin analysis can also detect hemoglobin variants while predicting thalassemia. Therefore, although the sensitivity of capillary electrophoresis to predict α-thalassemia is not very high, it is still an important detection method at present.

Due to the lack of genetic counseling and standardized prevention systems, China still has a high incidence of α-thalassemia. The silent and mild types are usually asymptomatic microcytosis or with normal hematologic findings and normal growth development [28]. However, the middle and severe types may have severe anemia or fetal death in utero. If both partners are carriers of silent or mild types of α-thalassemia, the children with middle or severe types may be born. Hence, screening for α-thalassemia is very important. Although treatments for controlling and recovering from thalassemia, such as routine blood transfusion (followed by the iron chelation therapy) [29–32], bone marrow transplantation, and gene therapy are available [33–36], carrier screening and preventing the birth of severely affected individuals are considered the best preventive measures [1, 37]. Furthermore, Since common genetic tests may miss rare thalassaemia genotypes, screening tests are particularly important [38]. In our study, we found that the sensitivity of MCV + MCH + HbA₂ is only 49.8%, which means that more than half of the α-thalassemia carriers will be missed due to this scheme. Therefore, in areas with high incidence of α-thalassemia, especially for pregnant women and their spouses, it is recommended to choose a highly sensitive screening program to avoid missed diagnosis, which may lead to the birth of children with intermediate or severe thalassemia. At the same time, we also focus on the contrasts of the several strategies, the positive predictive value of MCV/MCH/HbA₂ is 32.6%, which means that only about a third of the participants with positive results are really the carriers. If the scheme is adopted to mass screening, it will cause the waste of social medical resources. In our study, the Youden indexes of MCV + MCH was higher than other combined screening programs, which indicated that it is the most suitable for screening of α-thalassaemia. In a word, we recommend that the most appropriate screening program can be selected according to the different screening purpose and disease carrying rate in the regions, and it also based on the medical resources and economic conditions of the regions.

Our study lasted approximately three years. Some participants refused to complete all the tests and had to be excluded from the study. Due to the insufficient cases of α-thalassemia, it was impossible to groups patients according to genotype. The number of participants is expected to be expanded in a future study.

Conclusions

Our study outcomes are a reliable reference resource for performing laboratories screening for α-thalassemia disease. They offer meaningful insights for future research in this field and will be useful for the clinical geneticists working with α-thalassemia patients.

Funding

The National Natural Science Foundation of China (ID 81970170).

Declarations

Conflict of interest

The authors declare no potential conflicts of interest.

Ethical Approval

The present study was approved by the Protection of Human Ethics Committee of Fujian Maternity and Child Health Hospital, affiliated hospital of Fujian Medical University (NO.2016-101).

Patient Consent

Written consent taken from each patient and/or their legal guardians at the entry into the study.