The ultimate goal of thalassaemia screening is to prevent severe forms of the disorder, namely haemoglobin (Hb) Barts hydrops fetalis and b-thalassaemia major (Cooley's anaemia). This has been achieved in Hong Kong by an antenatal screening program for pregnant women that incorporates carrier detection, genetic counselling and prenatal diagnosis [1]. However, new cases of hydrops fetalis [2] and Cooley's anaemia [3] are still encountered, attributable to various factors: absence of antenatal follow up (e.g. immigrant mothers), presentation too late in pregnancy, thalassaemia screening not performed, and missed diagnosis of carrier state by laboratory. There are merits therefore to develop a screening program in the community [4] for carriers to be given both the time and genetic information of reproductive significance on which to plan their future family. Needless to say, public education is also an important element in ensuring success of a thalassaemia preventive program.

Laboratory haematologists and clinical pathologists have a major role to play in laying down the screening strategy. The initial screen involves determination of the red cell indices, the mean corpuscular volume (MCV) and mean corpuscular Hb (MCH). Clinical guidelines and practices vary from place to place [5,6]. For example, even in Hong Kong, different cutoff values based on MCV < 75 fL [1] and < 80 fL [4,7] have been described.

We have the opportunity to correlate thalassaemia genotypes from our archive with respect to MCV and MCH. The following observations are made:

1. A cutoff value of MCV < 80 fL or MCH < 27 pg is able to detect all heterozygous carriers of (--^SEA) a-thalassaemia (SEA) deletion and seven most common b-thalassaemia alleles at our locality, and should be recommended for screening in the Chinese population. Any MCV cutoff lower than 80 fL as well as MCH cutoff of below 27 pg will miss SEA deletion carriers, so that hydrops fetalis may not be completely prevented. A MCV cutoff of 75 fL will also miss b-thalassaemia carriers.

2. Subjects with less common thalassaemia carrier states harbouring the SEA deletion which may potentially be associated with high MCV and MCH values, including SEA deletion in association with triplicated a-globin gene (aaa/--^SEA) [8], and concurrent SEA deletion and b-thalassaemia trait [9], also have MCV < 80 fL and MCH < 27 pg.

The need for adequate quality assurance of automated cell counters to produce accurate and precise MCV and MCH results cannot be over-emphasized. There is a reliance on MCH value in the United Kingdom due to its better stability on specimen storage, but we do not have the problem of sending samples over long distances in Hong Kong. While we feel that either MCV < 80 fL or MCH < 27 pg is suitable, the former is more commonly employed.

The cutoff of MCV < 80 fL, however, is not able to detect:

1. Around 75% of subject carrying single a-globin gene deletion (-a^3.7 and -a^4.2) and non-deletional a-globin gene mutations (Hb Constant Spring, Hb Quong Sze, and a2 codon 30 D GAG) [7].

2. Around 50% of Hb E carriers [7].

3. Subjects that carry the rare silent b-thalassaemia alleles in the Chinese, including nt+8 (C®T) mutation at the 5'-UTR of the b-globin gene [10] and the AATAAA®CATAAA mutation at the polyadenylation site [11].

Furthermore, it should be noted that the MCV cutoff is intended for thalassaemia screening and not detection of haemoglobinopathy.

When a sample for thalassaemia screening arrives at the laboratory, the following tests are indicated:

1. Determination of red cell indices MCV and MCH (see above).

2. For subjects with MCV < 80 fL, the iron status should be determined. The reasons are firstly to diagnose iron deficiency and explain the microcytosis, and secondly to avoid missing a concurrent b-thalassaemia trait due to spurious lowering of HbA₂ by iron deficiency. Having said that, unless the iron deficiency is very severe, the HbA₂ will usually still be elevated if the subject is a b-thalassaemia carrier [12].

3. Screening for thalassaemia in subjects with MCV < 80 fL involve Hb analysis by chromatographic methods and searching for HbH inclusion bodies. In Sardinia [6], Hb chromatography is performed in all subjects irrespective of the MCV and MCH to detect concurrent a (-a/-a configuration) and b thalassaemia (i.e. normal MCV and MCH, but high HbA₂). However, based on absence of SEA deletion and b-thalassaemia allele in subjects with MCV 80 - 85 fL, and clinical experience of concurrent a and b thalassaemia in the Chinese [9], this should not be a concern here.

4. There are occasional cases in which the red cell indices are suggestive of thalassaemia trait and iron deficiency is absent, but with normal HbA₂ and no HbH inclusion bodies. For these cases a PCR assay should be employed to look for the SEA deletion [13], which is the most clinically relevant mutation to exclude in the context of thalassaemia prevention program. Apart from this, mutation detection techniques are not indicated for routine thalassaemia screening.

There are three issues that are not considered routine but may warrant further discussion:

1. Screening for SEA deletion in b-thalassaemia carriers. Carriers of concurrent a and b thalassaemia are identical in phenotype to b-thalassaemia trait, since HbH inclusion bodies are usually absent [9]. Currently, as part of the antenatal screening program, the a-gene map is also determined in b-thalassaemia for couples who are discordant carriers of a and b thalassaemia to fully prevent hydrops fetalis. As the genetic information is of reproductive significance, there may be a point for detection of SEA deletion in b-thalassaemia carriers by PCR in the setting of community screening.

2. Half of the patients with HbE/b-thalassaemia have a severe phenotype consistent with b-thalassaemia major. Therefore all partners of known b-thalassaemia carriers should be screened for HbE irrespective of MCV and MCH to predict for the risk of HbE/b-thalassaemia. This is achievable by Hb chromatography.

3. It has been documented that HbH disease due to non-deletional a-globin gene mutations has a more severe clinical course than deletional HbH disease [14]. Therefore there may be a case for screening partners of known SEA deletion carriers for non-deletional a-globin gene mutations to predict for more severe forms of HbH disease, so that appropriate genetic counselling may be offered. This can only be achieved through molecular genetic techniques for mutation detection.

References

1. Sin SY, Ghosh A, Tang LCH, Chan V. Ten years' experience of antenatal mean corpuscular volume screening and prenatal diagnosis for thalassaemias in Hong Kong. J Obstet Gynaecol Res 26: 230 - 208, 2000.
2. Feng CS, Tsoi WC. A survey of pregnancies that ended in haemoglobin Bart's hydrops foetalis and Cooley's anaemia. HK Med J 1 (1): 22 - 26, 1995.
3. Lee ACW, Wong KW, So KT, Cheng MY. Why are thalassaemia patients born when prenatal screening is available? HK Med J 4 (2): 121 - 124, 1998.
4. Lau YL, Chan LC, Chan AYY, Ha SY, Yeung CY, Waye JS, Chui DHK. Prevalence and genotypes of a- and b-thalassemia carriers in Hong Kong - implications for population screening. N Engl J Med 336: 1298 - 1301, 1997.
5. Working Party of the General Haematology Task Force of the British Committee for Standards in Haematology. The laboratory diagnosis of haemoglobinopathies. Br J Haematol 101: 783 - 792, 1998.
6. Cao A, Galanello R, Rosatelli MC. Prenatal diagnosis and screening of the haemoglobinopathies. Bailliere Clin Haematol 11: 215 - 238, 1998.
7. Chan LC, Ma SK, Chan AYY, et al. Should we screen for globin gene mutations in blood samples with mean corpuscular volume (MCV) greater than 80 fL? J Clin Pathol 54: 317 - 320, 2001.
8. Ma SK, Chan AYY, Chan LC, Chui DHK, Waye JS. Compound heterozygosity for triplicated a-globin gene and (--^SEA) a-globin gene deletion: implication for thalassaemia screening. Br J Haematol 110: 498 - 499, 2000.
9. Ma SK, Chan AYY, Ha SY, Chan GCF, Lau YL, Chan LC. Screening for (--^SEA) a-globin gene deletion in b-thalassaemia carriers and prevention of hydrops fetalis. Haematologica 85: 991 - 993, 2000.
10. Ma SK, Ha SY, Chan AYY, Chan GCF, Lau YL, Chan LC. Two novel b-thalassemia alleles in the Chinese: the IVS-II-2 (-T) and nucleotide +8 (C®T) b-globin gene mutations. Hemoglobin 24: 327 - 332, 2000.
11. Ma SK, Lee ACW, Chan AYY, Chan LC. A novel AATAAA®CATAAA mutation at the polyadenylation site of the b-globin gene. Br J Haematol (in press).
12. Steinberg MH. Case report: Effects of iron deficiency and the -88 C®T mutation on HbA₂ levels in b-thalassemia. Am J Med Sci 305: 312 - 313, 1993.
13. Chan AYY, So CKC, Chan LC. Comparison of the HbH inclusion test and a PCR test in routine screening for a-thalassaemia in Hong Kong. J Clin Pathol 96: 411 - 413, 1996.
14. Chan FE, Ooi C, Ha SY, Cheung BMY, Todd D, Liang R, Chan TK, Chan V. Genetic and clinical features of hemoglobin H disease in Chinese patients. N Engl J Med 343: 544 - 550, 2000.