Thalassemias in Hong Kong
Vivian Chan, TK Chan
Department of Medicine, Queen Mary Hospital, The University of Hong Kong

SY Ha, *Anita Li,
Department of Paediatrics, Queen Mary Hospital, The University of Hong Kong

Mary Tang,
Department of Obstetrics & Gynaecology, Tsan Yuk Hospital, The University of Hong Kong

MK Shing,
Department of Paediatrics, Prince of Wales Hospital, The Chinese University of Hong Kong

* Dr. Anita Li, pioneer in thalassemia management in Hong Kong, passed away in December 1999.

The α and β thalassemias (thal) are common genetic disorders, where deficiency of α or β globin chain respectively gives rise to anemia. In Hong Kong, the incidence of the various forms of α thal is 4.5%, with 2.2% α thal 1 (deletion of 2 α genes in cis) and 2.3% α thal 2 (deletion of 1 α gene) respectively,1 while the heterozygous state for β thal affects 3-6% of the local population.2,3 Since couples who are both α thal 1 or both β thal minor run a 25% risk of having a homozygous severely affected child with each pregnancy, screening for those at risk and prenatal diagnosis are advocated. The strategy, carried out at the antenatal clinic of all major hospitals and all maternal and child health centres throughout Hong Kong, is based on low MCV (< 80 fl) of the pregnant woman and her husband. They are assessed for α or β thal trait by the presence of H inclusion, z gene map and elevated HbA2 respectively. Since 1990, screening is extended to couples attending the Family Planning Association clinics (FPA) throughout the territory for premarital counselling. If both partners are found to have the same type of thal trait, they are counselled regarding the mode of inheritance and management of these congenital disorders. Thus they would be forewarned of the need for prenatal diagnosis before marriage and well before pregnancy. Prenatal testing for α and β thal has been established since 1982 with approximately 100 cases of α thal and 50 cases of β thal per annum. Traditionally, diagnosis of α thal is performed by detection of α genes in fetal DNA extracted from chorionic villi (CV) or amniocytes4 depending on the time of presentation of the pregnant women. Since 1997, non-invasive ultrasound diagnosis based on measurement of cardiothoracic ratio of the fetus5 is also used, thereby reducing the number of DNA based diagnosis. A recent study of 114 HbH patients revealed that 23.2% of cases are due to compound heterzygosity of α thal 1 and a non-deletion α thal gene. These cases of non-deletion HbH disease have a more severe phenotype compared to the deletional form, with more marked anemia, splenomegaly and increased iron overload6 (Table 1). In addition, 2 cases of HbH hydrops fetalis have been noted when associated with HbH-codon 30 DGAG and HbH-codon 59 (G®A) respectively.7 Thus identification of non-deletion α thal parents and prenatal diagnosis of these severe HbH hydrops fetalis are required.

With β thal, 12 known β thal mutations have been identified in the Chinese, ten of which account for 95.7% of cases. Direct detection of the β thal mutation in the fetus using 'reverse-dot-blot' ybridization with allele-specific oligonucleotide probes is performed.8 It is possible to screen for all 12 known Chinese β thal mutations simultaneously on one strip and obtain a result within six hours of fetal sampling.

To date 1448 diagnoses have been performed and 326 affected pregnancies identified. Over 294 females have returned for multiple prenatal testing, with five mothers having undergone their sixth and seventh testing respectively. This serves as an indication of the widespread acceptability of the prenatal diagnostic procedure in Hong Kong.

Management of Transfusion-dependent β-thalassemia

There are currently 330 transfusion-dependent thal patients managed at eight major hospitals in the region. In Hong Kong, approximately 85% of the population attends public hospitals where medical care is heavily subsidized by public funds (HK$68 per day for inpatient and/or $44 for outpatient consultation, inclusive of drug costs). The remainder are cared for by practitioners in private hospitals where all costs would be borne by the patients. Thus most patients with chronic illnesses, which require long-term follow-up and treatment would tend to go to public hospitals. Figure 1 shows the patient number according to their birth year. Even with the implementation of the prenatal diagnosis programme in 1984, the decreasing trend of β thal major births is only evident from 1991 onwards. This underscores the need for public education regarding congenital diseases and more widespread screening of at risk couples. A survey of the 34 new thal major cases seen at two of the major hospitals (TMH & QMH) between 1990-1996 revealed that 23.5% (8 cases) were either due to late or no antenatal visits because these were cross-border delivery of temporary visitors from Mainland China. 41.2% (14 cases) and 20.6% (7 cases) were due to lack of maternal or paternal screening respectively, 8.8% (3 cases) were couples' refusal for prenatal testing and 5.9% (2 cases) offered no reason.

Figure 1

Legend to figure
No. of transfusion - dependent β thalassemia patients in Hong Kong according to the year of birth. Data from eight of the major hospitals are included. Hatched histograms show cases born before the establishment of DNA-based prenatal diagnosis programme.


Table 2 shows the landmarks in the management of β thal in Hong Kong. The homozygous β thal child usually presents with symptoms of anaemia by 6-9 months of age, when the gamma (γ) genes switch off and the β gene is expected to go into full expression. The current trend is a high-transfusion regime to maintain the child's haemoglobin at 9-10 g/dl. This would most likely be at 6 weekly interval and chelation to remove excess iron would be needed within 12-18 months of starting regular blood transfusion or when serum ferritin reaches 1000 mg/l. Parents are taught the use of the infusion pump and would administer the drug to their young children. Patients with more severe iron overload are additionally given intravenous desferral during blood transfusions. Although in many instances, this does not fully chelate all excess iron, it helps to reduce the level. The establishment of the Cooley's Anemia Association as a patient-parent support group in 1980 provides a venue for the thalassemia families to share their experience and encourages better communication between medical care providers, the patients and the parents.

Apart from chelation therapy, endocrine complications such as retarded growth and disturbed pubertal development are also treated, although to-date, none of the β thal patients have produced their own offsprings.

Viral Infections

The testing of donor blood for hepatitis B and C and HIV has only been performed since 1985. None of the thalassemic patients are HIV positive, but the incidence of HCV infection amongst the patient group is significantly higher compared to that of the general population (34% versus 0.5%). Of the 65 β thal patients managed at Queen Mary Hospital, 21 (i.e. 32%) are HCV positive. It is of interest to note that these are mainly patients in the older age group, no one born after 1985 has been infected. This group of patients are symptom-free, with silent infection but had significantly elevated transaminases (ALT 91±82 iu/L, AST 67±38 iu/L) compared to the HCV negative group (p = 0.0018 and p = 0.0142 respectively). With time, most would develop liver fibrosis and 20% cirrhosis subsequently. HCV positive patients can be treated with α-interferon (3 mega-unit three times per week) for one year and oral ribavirin supplemented for those who do not respond well to α interferon alone.


Haemopoietic stem cell transplant (HSCT) has been available at two centres since 1991. Initially, bone marrow transplant (BMT) from HLA-matched sibs were performed. At the Prince of Wales Hospital, in 44 BMT cases there were five treatment-related mortality and two rejections. The overall disease free survival was 84%9 (Table 3). At Queen Mary Hospital, the first 7 patients using busulphan and cyclophosphamide alone had 3 rejections but subsequently with the addition of antithymocyte globulin (ATG) in the conditioning regime, the results have improved and in 18 cases the disease free survival at 5 years (DFS) was 61%. The recent use of cord blood stem cell (CBST) from compatible sibs (identified through the prenatal diagnosis programme),10 had a 100% favourable transplant outcome11 (Table 3). The strategy now is for known β thal patients to await for a HLA-matched-sibling donor using the prenatal diagnosis programme. The fetal DNA of the mother's current pregnancy which had been diagnosed as normal or β thal trait would be typed for HLA-DR antigens by polymerase chain reaction technique, and if suitably matched to that of the affected sib (5 out of 6 antigens match would suffice), cord blood would be collected at delivery and processed. All six cases of HLA-matched sibling CBSCT done at Queen Mary Hospital were successful, and the longest follow-up was 7.4 years (median 3.2 years). This combined allogeneic CBSCT-prenatal diagnosis programme avoids the delay of HLA-typing till after birth, allows the paediatricians and family adequate time to prepare for transplantation and reduces the storage of incompatible cord blood.

Management cost

Table 4 lists the cost requirement of different strategies in the management of β thal. Obviously the cost of stem cell transplant is far less compared to the conventional life-long transfusion-chelation therapy. Whilst transplant carries with it a small but definite risk of mortality, this is further reduced with the use of allogeneic CBSCT. Most economical is the practice of PREVENTION. In order to do so effectively, apart from the prenatal diagnosis programme which we already have in operation for almost 20 years, is the need of public education, counselling and more widespread screening programme to detect couples at risk.


The authors gratefully acknowledge the Wideland Foundation for the initial funds to establish a DNA laboratory at the Department of Medicine, Queen Mary Hospital, the Mrs. Wu-Chung Prenatal Diagnostic Laboratory for ultrasound facilities and the Croucher Foundation for research support.


  1. Chan V, Chan TK, Cheng MY, Kan YW, Todd D. Organization of z-α genes in the Chinese. Br J Haematol 1986; 64:97-105.

  2. Ghosh A, Woo JSK, Wan CW, MacHenry C, Wong V, Ma HK, Chan V, Chan TK. Evaluation of a prenatal screening procedure for β-thalassemia carriers in a Chinese population based on the mean corpuscular volume (MCV). Prenatal Diag 1985; 5:59-65.

  3. Lau Y-L, Chan L-C, Chan Y-YA, et al. Prevalence and genotypes of α- and β-thalassemia carriers in Hong Kong - implications for population screening. New Engl J Med 1997; 336:1298-1301.

  4. Chan V, Ghosh A, Chan TK, Wong V, Todd D. Prenatal diagnosis of homozygous α thalassemia by direct DNA analysis of uncultured amniotic fluid cells. BMJ 1984; 288:1327-1329.

  5. Lam YH, Ghosh A, Tang MH, Lee CP, Sin SY. Early ultrasound prediction of pregnancies affected by homozygous α-thalassemia 1. Prenatal Diag 1997; 4:327-332.

  6. Chen 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. New Engl J Med 2000; 343:544-550.

  7. Chan V, Chan VWY, Tang M, Lau K, Todd, Chan TK. Molecular defects in HbH hydrops fetalis. Br J Haematol 1997; 96:224-228.

  8. Chan V, Tang M, Chan TK. Prenatal diagnosis of common genetic diseases in Hong Kong. HK J Paediatr (new series) 1996; 1:184-188.

  9. Li CK, Shing MMK, Chik KW, Lee V, Leung TF, Cheung AYK, Yuen MP. Haematopoietic stem cell transplantation for thalassaemia major in Hong Kong: prognostic factors and outcome. Bone Marrow Transpl 2002; 29:101-105.

  10. Lau Y-L, Ma ESK, Ha SY, Chan GCF, Chiu D, Tong M, Hawkins BR, Chan V, Liang RHS. Sibling HLA-matched cord blood transplant for β-thalassemia: report of two cases, expression of fetal hemoglobin, and review of the literature. J Pediatr Hematol/Oncol 1998; 20:477-481.

  11. Ha SY. Personal Communications 2002.

Table 1 Clinical phenotypes in different types of HbH disease (n=114)

      n   Hb
(g / dl)
(mg / L)
(mean ± SD)

A) Deletion   87   9.5 ± 1.3   1.52 ± 1.99   536 ± 649    
B) Non-deletion   26          
    Hb H-CS   15   8.2 ± 1.9   5.1 ± 4.1   882 ± 899    
    Hb H-QS   5   8.2 ± 0.4   6.9 ± 5.2   654 ± 891    
    Hb H-Codon 30   4   8.8 ± 1.7   2.7 ± 2.5   1437 ± 1848   Hydrops*
    Hb H-Codon 31   2   9.4 ± 0.4   1   154 ± 68    
    Hb H-Codon 59   1   6.6   -   ND   Hydrops

* 1 case of HbH hydrops detected with z-α thal 1 and non-deletion α thal 2 (D codon 30) genotype.


Table 2 Landmarks in management of thalassemia major in Hong Kong

1967 - Regular transfusion
1979 - Chelation - desferrioxamine
1980 - Cooley's Anaemia Association
1982 - Prenatal diagnosis
1991 - Bone marrow transplant
1994 - Sibling cord blood stem cell transplant


Table 3 Haemopoietic stem cell transplant (HSCT) for transfusion-dependent thalassemia (1991-2002)

                Outcome - Number (%)
Team No. Donor Stem Cell Source*
  Treatment Related Mortality
Rejection Disease-free Survival
(5 years)

PWH 44** HLA-matched-Sib (43) BM (38)   4 (veno-occlusive) 2 37 (84%)
    HLA-matched-Father (1) PBSC (3)   1 (interstitial pneumonitis)    
        UCB (2)        
        BM + UCB (1)        
QMH 18 HLA-matched-Sib     BM (18)   2 (grafts vs host disease) 3   11 (61.1%)
  6*** HLA-matched-Sib     UCB (6)   0 0   6 (100%)

* BM = Bone marrow
PBSC = Peripheral blood stem cells
UCB = Umbilical cord blood  
** Includes 2 homozygous a thal 1  
*** Includes 1 homozygous a thal 1 (HLA-mismatched)  
      CK Li et al, 2002; SY Ha et al, 2002


Table 4 Unit cost of medical management cost for thalassemia major in Hong Kong

A. Child    
  1. Transfusion   HK$ 30,420 / yr
  2. Chelation   HK$ 20,280 / yr
B. Adult    
  1. Transfusion   HK$ 34,320 / yr
  2. Chelation   HK$ 31,980 / yr
  Total expenses per patient for 45 yrs   HK$ 760,500 + HK$1,989,000
    = HK$ 2,749,500
C. Stem cell transplant   HK$ 400,000
D. Prenatal diagnosis HK$ 6,250


Table 5 Number of patients with multiple prenatal diagnosis (1982-2002)

Type of diagnosis

Number of diagnosis/patient
2x 3x 4x 5x 6x 7x
α thal 119 48 13 3 2 2
β thal 83 13 6 2 1
α+β thal 1 1
Total 203 62 19 5 3 2