SEARCH

SEARCH BY CITATION

Keywords:

  • haemoglobin H disease;
  • anaemia;
  • heart function;
  • iron overload;
  • pregnancy

Summary

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References

In this study, 251 Sardinian patients (187 adults and 64 children) with haemoglobin (Hb) H disease were evaluated. Two-hundred and sixteen patients (86%) had the deletional type (- -/-α) and 36 (14%) patients had the non-deletional type (- -/αNDα). A clear genotype–phenotype correlation was found, with the non-deletional type more severe than the deletional type. Diagnosis of Hb H disease was incidental in about 60% of cases. Aplastic crises due to B19 parvovirus infection were found in five patients (2·1%), while 23 patients (9·6%) experienced one or more haemolytic crises. Nineteen patients with Hb H received sporadic red blood cell transfusions and three patients were repeatedly transfused. Forty-seven of 61 married women (77%) had 82 pregnancies. In children, mean serum ferritin was 87 ±92 μg/l and in adults, was 192 ± 180 μg/l in females and 363 ± 303 μg/l in males. For the 98 male patients, a significant correlation was found between ferritin values and age (r2 = 0·33, P < 0·0001). In the Sardinian population, Hb H disease needs regular monitoring for early detection and treatment of possible complications, such as worsening of anaemia that may require red cell transfusion, cholelithiasis and iron overload.

Deletion or inactivation of three of the four α-globin genes located on chromosome 16, gives rise to haemoglobin (Hb) H disease, an intermediate clinical form of α-thalassaemia characterised by microcytic, hypochromic haemolytic anaemia that is frequently associated with hepatosplenomegaly and mild jaundice (Higgs & Bowden, 2001). Hb H disease is relatively common in south-east Asian, Middle Eastern and Mediterranean populations and shows a wide spectrum of clinical phenotypes, ranging from asymptomatic to severe forms, requiring occasional or even regular transfusions and, exceptionally, to Hb H-hydrops foetalis syndrome. This clinical variability depends on a considerable genetic heterogeneity.

In Hb H disease, the severely reduced production of α-globin chains results in a relative excess of β-chains which aggregate into β4 tetramers (Hb H). Patients with significant amounts of Hb H have a defect in the oxygen-carrying capacity that is more severe than expected on the basis of the Hb concentration. Red cells containing Hb H are sensitive to oxidative stress and thus may be more susceptible to haemolysis when oxidant drugs, such as sulphonamides, are administered. Ageing erythrocytes contain more precipitated Hb H than younger erythrocytes and are removed from the circulation prematurely. Thus, Hb H disease is primarily a haemolytic disorder, while ineffective erythropoiesis, which is the main cause of anaemia in β-thalassaemia, plays a secondary role (Wasi et al, 1974; Papassotirious et al, 1998; Pootrakul et al, 2000).

Two main types of Hb H disease, deletional and non-deletional, are characterised. The deletional Hb H disease, caused by a deletion removing both α-globin genes on one chromosome 16 and a deletion of a single α-globin gene on the other chromosome (- -/-α), is the most common type. Less frequently, the deletion of both α-globin genes on one chromosome is associated with a point mutation, or an oligonucleotide insertion/deletion in the α1 or α2-globin gene on the other chromosome (non-deletional Hb H disease, - -/αNDα) (Higgs, 2001). Patients with non-deletional Hb H disease have a more severe phenotype when compared with those with the deletional type (Galanello et al, 1983; Chen et al, 2000; Kanavakis et al, 2000). An Hb H disease-like phenotype of variable severity can be sometimes due to homozygosity or compound heterozygosity for non-deletional α-thalassaemia mutations in α2-globin genes, which normally account for the majority of α-chain production (Liebhaber et al, 1986). Several authors have investigated the correlation between haematological, biochemical and clinical findings and α-globin genotypes in different populations, but systematic studies of the natural history of Hb H disease are lacking (Galanello et al, 1983; Chen et al, 2000; Kanavakis et al, 2000; Chui et al, 2003). This study aimed to define the natural history of Hb H disease in a Mediterranean population, with special attention to children and pregnancies in affected women.

Patients and methods

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References

Two-hundred and fifty-one Sardinian patients with Hb H disease were evaluated. One hundred and eighty-seven patients were adults (aged 18–84 years) and 64 were children (aged 0–18 years).

Haematological and biochemical assessment

Haemoglobin, red blood cell indices and reticulocyte counts were measured by an electronic cell counter (Gene S or LH700-Beckman Coulter, Inc., Fullerton, CA, USA). Glucose 6-phosphate dehydrogenase (G6PD) deficiency was detected by a potentiometric fully automated method using the G6PD/6-phosphogluconate dehydrogenase ratio (Mosca et al, 1996). Unconjugated bilirubin and lactate dehydrogenase (LDH) were determined with standard methods. Serum ferritin was measured using an automated immunoassay system (Immulite 2000®, Medical System, Genova, Italy) and chemiluminescence was used to determine serum erythropoietin.

Molecular analysis

DNA from peripheral blood leucocytes was amplified by polymerase chain reaction (PCR) to detect -α3·7, -α4·2, –MED and - -(α)20·5 deletions (Galanello et al, 1998). NcoI and HphI restriction enzymes were used to detect the α1 and α2ATG [RIGHTWARDS ARROW] ACG initiation codon mutation and the IVS-I splice junction pentanucleotide deletion (-TGAGG), respectively, which are the only non-deletion mutations so far described in the Sardinian population (Orkin et al, 1981; Pirastu et al, 1984). The –TA-insertion in the –TATAA-promoter element of the uridine diphosphate glucuronyl transferase (UGT1A1) gene (UGT1A1), responsible for bilirubin glucuronidation, was detected by PCR and automated sequencing (Bosma et al, 1995).

Clinical assessment

The clinical records of 251 patients with Hb H disease that were regularly followed-up at Ospedale Regionale Microcitemie (Cagliari, Italy) were evaluated. Gestational age, birth weight and longitudinal growth assessment were available for 53 patients. Z-score (the deviance from the median reference value expressed in units of SD) was used to assess weight and length. As currently recommended by the World Health Organization (WHO), a Z-score of −2 SD was taken as the cutoff value for growth deficiency (World Health Organization Working Group, 1986). Sudden worsening of anaemia with reticulocytopenia with or without leucopenia and thrombocytopenia was defined as ‘aplastic crisis’, while an episode of acute haemolysis characterised by low Hb value with reticulocytosis, increase of serum bilirubin and urobilinuria was defined as ‘haemolytic crisis’. Abdominal ultrasound scans, performed to evaluate the prevalence of liver and spleen enlargement and of gallstones, were available for 188 patients, electrocardiogram and echocardiography for 153 patients. Liver and/or spleen enlargement have been defined according to age and body height (Konus et al, 1998). Liver biopsy was performed in four patients with abnormal liver function and/or elevated ferritin values. Liver iron concentration (LIC) was measured by histological grading of haemosiderotic granules and was semiquantitatively evaluated in one patient and determined by atomic absorption spectroscopy in two patients (Brissot et al, 1981; Desmet et al, 1994). Echocardiographic studies were performed with a Hewlett Packard Agilent Sonos 5500 phase-array scanner (Agilent, Andover, MA, USA). Left ventricular ejection fraction (LVEF) was calculated from left ventricule volumes using the modified Simpson's method (Schiller et al, 1989). A LVEF >55% was considered normal. Pulsed Doppler analysis of mitral inflow included measurement of the mitral valve early peak filling velocity (E), the late peak filling velocity (A) and the E to A ratio (E/A). The assessment of diastolic dysfunction was performed according to the basic patterns described by Kitabatake et al (1982). Left ventricle septal and posterior wall thickness, cavity dimensions and volumes, mass, and left atrial size were determined from two-dimensional images (Schiller et al, 1989; Park et al, 1996). The pulmonary artery pressure was indirectly estimated by Doppler and two-dimensional echocardiography according to Schiller (1990).

Statistical evaluation

Statistical evaluation of haematological and other continuous data were carried out using the Student's t-test and the chi-squared test was used for categorical variables. The correlation between variables was expressed by means of Spearman's correlation coefficient. P < 0·05 was considered significant.

Results

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References

Molecular analysis

The genotypes identified are reported in Table I. Two-hundred and sixteen patients (86%) had the deletional type (- -/-α) and 36 (14%) had the non-deletional type (- -/αNDα). The gene deletion (- -/) was of the Mediterranean type in all but two patients who had the - -(α)20·5 deletion. Moreover, one previously described patient (Romao et al, 1992), had α-thalassaemia associated with a truncation of the short arm of chromosome 16 that removes the locus control region and inactivates the adjacent intact α-globin genes (Romao et al, 1992). The - -/-α3·7 genotype was the most common (82·4%), followed in decreasing order of incidence by the combination of deletion of α0-thalassaemia with the initiation codon mutation (ATG [RIGHTWARDS ARROW] ACG) of the α2 gene, recognised by the NcoI restriction enzyme (- -/αNcoIα) (7·9%), or with the α2 IVS-I pentanucleotide deletion recognised by the HphI restriction enzyme (- -/αHphIα) (5·6%), and by other deletional types - -/-α4·2 (2%) and - -(α)20·5/-α (0·8%). In two sisters, the deletion α0-thalassaemia was combined with the codon mutation (ATG [RIGHTWARDS ARROW] GTG) in the α1-globin gene (Moi et al, 1987).

Table I. α Globin genotype in Sardinian patients with haemoglobin H disease.
α-Globin genotypeNumber of patients (%)
- -/-α3·7207 (82·5)
- -/-α4·25 (1·9)
-(α)20·5/-α2 (0·8)
- -/αNcoIα20 (8·0)
- -/ααNcoI2 (0·8)
- -/αHphIα14 (5·6)
(αα)IdF/-α1 (0·4)

From the analysis of –TATAA-promoter element of UGT1A1 in 183 patients, 101 (55%) were found to be homozygous for the wild-type motif (TA)6, 13 (7%) were homozygous for the variant motif associated with Gilbert's syndrome [(TA)7/(TA)7], and 69 (38%) were heterozygotes [(TA)6/(TA)7].

Haematological evaluation

Haematological data are shown in Table II. Hb values were reduced in all patients (both children and adults), and were significantly lower in patients with the non-deletional genotype, while mean corpuscular volume (MCV), reticulocytes and bilirubin were significantly higher (P < 0·001). No difference was found in mean corpuscular haemoglobin (MCH) between the patients with the deletional and the non-deletional genotypes. Haematological data at birth were available for four patients (Table III). A significant neonatal anaemia, with severe reduction of MCV and MCH, was present and Hb Bart's was markedly increased in all newborns. No wide fluctuations of Hb values were found during the follow up of paediatric patients (Fig 1), and patients with the non-deletional type had consistently lower levels. Among males, 15% were hemizygous for G6PD deficiency and among females, 6% were homozygotes and 22% heterozygotes. No difference in Hb level was detected between hemizygote male patients and those with normal G6PD activity (96 ± 8 g/l vs. 97 ± 10 g/l, P = 0·4). Serum LDH levels were higher in the non-deletional type (430 ± 119 U/l vs. 802 ± 343 U/l, P < 0·0001), while erythropoietin was inconstantly and only slightly increased, both in patients with deletional and non-deletional type (30·3 ± 12·9 U/l in the deletional type and 39·6 ± 12·3 U/l in the non-deletional type, P = 0·2).

Table II.   Haematological data of haemoglobin H disease patients, according to α-globin genotype.
 Children (≤18 years)P-valueAdults (>18 years)P-value
Deletional (50)Non-deletional (14)Deletional (159)Non-deletional (28)
Haemoglobin (g/l)91 ± 8·086 ± 7·0<0·001103 ± 8·090 ± 7·0<0·001
Mean corpuscular volume (fl)56 ± 460 ± 4<0·00161 ± 464 ± 6<0·001
Mean corpuscular haemoglobin (pg)18 ± 118 ± 10·119 ± 119 ± 10·3
Reticulocytes (%)2·9 ± 1·34·5 ± 2·0<0·0013·5 ± 1·44·6 ± 1·40·04
Unconjugated bilirubin (μmol/l)13·7 ± 10·318·8 ± 8·60·0920·5 ± 17·130·8 ± 17·10·002
Table III.   Haematological data at birth in four patients with haemoglobin (Hb) H disease.
 - -/-α- -/-α- -/-α- -/αHphIα
Hb (g/l)88113128107
Mean corpuscular volume (fl)61·672·87762·5
Mean corpuscular haemoglobin (pg)19·422·3NA19·7
Hb Bart's (%)20·5222025
image

Figure 1.  Haemoglobin follow-up (mean ± 2 SD) in paediatric patients with haemoglobin H disease according to α-globin genotype.

Download figure to PowerPoint

Clinical findings

Mean age at diagnosis was 11 ± 12 years (range 0–73) in patients with deletional Hb H disease and 13 ± 17 years (range 0–59) in patients with other genotypes (P > 0·5). Forty-one per cent of the patients had symptoms at presentation, with no differences between deletional and non-deletional types (Table IV). In all other patients diagnosis was incidental during routine examination. Among 53 patients for whom information was available, four (7·5%) were born preterm and six (11%) were small for gestational age. Seven of them (13·2%) had growth impairment (length Z-score < 2). A mild thalassaemia-like facies (prominent facial bones) was found in 5% of patients with deletional type of Hb H disease and in 73% of patients with the non-deletional type (P < 0·0001). In children, liver enlargement was present in 17·5% of patients with deletional type and in 50% with non-deletional type (P = 0·03). In adults, a mild hepatomegaly was present in 48% of patients with deletional type and in 30·4% with non-deletional type (P = 0·1). A mild spleen enlargement was found in 47·5% of children with deletional and 60% with non-deletional type (P = 0·5) and in 76% and 100% of the adult patients respectively (P = 0·007). In most patients, a mild jaundice was present. Unconjugated bilirubin was significantly higher in patients with homozygosity for (TA)7 repeats in the –TATAA-promoter element UGT1A1, which is associated with Gilbert's syndrome [15·4 ± 8·5 μmol/l in patients with (TA)6/(TA)6 genotype, 18·8 ± 15·4 μmol/l in (TA)6/(TA)7 and 47·9 ± 34·2 μmol/l in (TA)7/(TA)7, P < 0·00001]. Thirty-five of 188 patients (18·6%) experienced gallstones, four of them in childhood. The mean age of patients when gallstones was diagnosed was 33·6 ± 15·5 years (range 5–73). No correlation was found between risk of gallstones and Hb H genotype (P = 0·16). Although the prevalence of colelithiasis was slightly higher in adult patients with homozygosity for (TA)7 repeats, the correlation between UGT1A1 genotype and gallstones was not statistically significant (P = 0·4). Aplastic crises due to B19 parvovirus infection were found in five patients (2·1%), while 23 patients (9·6%) experienced one or more haemolytic crises, with an incidence of 3·3% in childhood and 8·9% in adults. Causes of haemolytic crises were drugs in seven patients, infections in two and unknown in 14 patients. There was a correlation between the genotype and the occurrence of haemolytic crises: 7·8% of patients with the deletional type and 19·4% of patients with non-deletional type had haemolytic crises (P = 0·03). Ten patients who manifested haemolytic crises had G6PD deficiency (P = 0·4). Only one patient underwent splenectomy for hypersplenism, at the age of 43 years.

Table IV.   Reasons for presentation of haemoglobin (Hb) H disease.
 %
Anaemia, paleness18·8
Haemolytic crisis8·3
Neonatal anaemia3·5
Asthenia1·7
Jaundice1·2
Hb H disease family history1·2
Aplastic crisis0·6
Cardiac murmur0·6
Incidental59·7

Heart function

Electrocardiogram was normal in all the examined patients. Mild heart dilatation was found in 30% of the patients, independently of genotype (P = 0·11). Heart dilatation appeared at a median age of 36 ± 18 years (range 5–64), with an incidence of 6·6% in childhood and 32% in adulthood. LVEF was normal in all but one patient with a congenital dilatative cardiopathy (mean value 62 ± 4%). Eleven adults patients (mean age 56 ± 11 years) (7·2% of the examined patients) had a diastolic dysfunction grade 1 (abnormal relaxation pattern) and 2 a diastolic dysfunction grade 2 (pattern 2). Mean ferritin in patients with diastolic function was 342 ± 225 μg/l and only five of 13 patients had ferritin values above normal. A mild increase of pulmonary artery resistance was found in only one young patient at pulsed Doppler echocardiography. Incidentally, 11 patients were found to have an aortic valve root diameter larger than normal values (mean 40 ± 2·5 mm, normal values 20–37 mm).

Transfusions

Nineteen patients with Hb H received sporadic red blood cell transfusions (1–8 units) and three patients were repeatedly transfused (10–24 units) because of persistent low Hb values (<70 g/l). Excluding pregnancy, 14·2% of patients with the deletional genotype and 31·4% of patients with other genotypes (P = 0·01) have been transfused. Five per cent of the patients were transfused in childhood, 17% in adulthood. In children, the reasons for transfusion were neonatal anaemia (n = 4), aplastic anaemia due to B19 parvovirus (n = 2), haemolytic crisis (n = 3), surgery (n = 1) and anaemia without signs of infection (n = 1). Among adults, the reasons for transfusions were haemolytic crisis (n = 11), pregnancy (n = 10), anaemia without signs of infection (n = 6), surgery (n = 5), infection-associated anaemia (n = 3) and aplastic anaemia due to B19 parvovirus (n = 2).

Iron overload

Mean serum ferritin in children was 87 ± 92 μg/l (range 11–666 μg/l). Ferritin values were above the normal reference ranges for the paediatric population (22–107 μg/l) in 19·7% of patients. Children with deletional Hb H disease had a significantly lower mean serum ferritin than patients with other genotypes (78 ± 85 and 154 ± 112 μg/l, P = 0·008). In adults, excluding the repeatedly transfused patients, mean ferritin was 192 ± 180 μg/l in females (range 18–960 μg/l) and 363 ± 303 μg/l in males (range 13–1390 μg/l). Fifty-nine per cent of males and 26% of females had ferritin values above the normal reference ranges (25–280 μg/l in males and 4–254 μg/l in females respectively). A significant correlation was found between ferritin values in 98 male patients and age (r2 = 0·33, P < 0·0001) (Fig 2). Four patients underwent liver biopsy for a persistent increase in alanine aminotransferase activity (more than twice normal values for at least 6 months). One of these patients was hepatitis C virus (HCV) positive. In two cases, the LIC measured by atomic absorption spectroscopy was 8·3 and 8·4 mg/g dry weight, with a ferritin value of 1180 and 1000 μg/l respectively. In a third patient, LIC measured by histological grading of haemosiderotic granules was equivalent to a grade III by Sciot classification (Desmet et al, 1994). In the fourth patient, chronic hepatitis by HCV was identified, but the LIC was not measured. Chelation therapy with subcutaneous desferrioxamine was prescribed in five patients with ferritin values around 1000 μg/l. Two of them were repeatedly transfused because of low Hb values.

image

Figure 2.  Correlation between serum ferritin levels and age in 98 male patients with haemoglobin H disease.

Download figure to PowerPoint

Pregnancy

Forty-seven of 61 married women (77%) had pregnancies. In 32% of cases their husbands were α-thalassaemia carriers (-α/αα and -α/-α genotype) and 13% of newborns inherited Hb H disease. Detailed data on 82 pregnancies from 31 women were available. There were nine miscarriages (11%) and two extrauterine pregnancies (2·4%). Eight women (10·9%) experienced a threatened abortion, four (5·5%) had pre-eclampsia, one (1·4%) had biliary colics, one developed an acute B hepatitis due to transfusions, one had polyhydramnios and one had gestational diabetes. Six children (8·2%) were born prematurely and four (5·5%), all born at term, had birth weight <2·5 kg. Caesarean delivery was performed in 22 cases (30·1%). Reasons for Caesarean section were a previous Caesarean or mother preference (42%), maternal dystocia (36%), fetal hypoxia (13·6%), pre-eclampsia (9%), gestational diabetes (4·5%) and twin birth (4·5%). During pregnancy, the Hb fell to 60 g/l in 10 women (14%) and one or more transfusions were necessary. During pregnancy, the transfusional need was similar in women with deletional and non-deletional genotype (P = 0·7).

Discussion

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References

We studied a large group of patients with Hb H disease from the Mediterranean area, evaluating the prevalence of clinical and haematological findings according to the genotype and age. The deletion of three structural α-genes is the commonest defect producing Hb H disease (86% of cases) in this group of patients and a similar prevalence of deletional and non-deletional genotypes has been reported in patients with Hb H disease from California, Hong Kong and Ontario (Chen et al, 2000; Lorey et al, 2001; Waye et al, 2001), while non-deletional genotypes were more common (43%) in the Greek population (Kanavakis et al, 2000). A clear phenotype–genotype correlation was found in terms of Hb and reticulocyte values, MCV, bilirubin in children, LDH and splenomegaly, while erythropoietin values were normal or only mildly increased also in patients with non-deletional genotypes, due to the secondary role of ineffective erythropoiesis in the genesis of anaemia in Hb H disease. A significant microcytic hypochromic anaemia with a marked increase of Hb Bart's was present at birth in all four patients for whom data at birth were available. Diagnosis of Hb H disease was incidental in most of the patients, independently from the genotype and, although the mean age at diagnosis was about 10 years, many patients were recognised in adulthood or old age. This diagnostic delay could be due to the relatively benign clinical phenotype of Hb H in Sardinia, or most probably, to an inadequate knowledge of this condition among paediatricians and family doctors or parental neglect.

Chen et al (2000) reported that growth was impaired in 13% of children with Hb H disease from Hong Kong and we found a similar percentage (13·7%). Complications, such as gallstones and heart dilatation, although more frequent in adults, also appeared in childhood. Although the correlation between serum unconjugated bilirubin levels and UGT1A1 genotype was significant, the coexistence of the (TA)7/(TA)7 genotype associated with Gilbert's syndrome does not seem a risk factor for developing gallstones, probably because the number of patients with the (TA)7/(TA)7 genotype was small. Transfusion need, especially for haemolytic crises, was higher in patients with non-deletion genotypes and in adults. A possible explanation of this could be the greater use of therapeutic drugs in adults when compared with children. Aplastic crises, less common than haemolytic crises, were due to B19 parvovirus infection in all cases and had a similar incidence in adults and children.

Serum ferritin was in the normal range in most of the children, while it was increased in more than 50% of adult males, with a significant age correlation. The reason for the high serum ferritin levels could be increased iron absorption secondary to chronic anaemia. The blood loss by menses could be the reason for the less frequent high ferritin values in women. Serum ferritin levels in the normal range have been reported in non-transfused patients with Hb H disease from Greece (Kanavakis et al, 2000), while raised ferritin values were found by Hsu et al (1990); Lin et al (1990) and, more recently, by Chen et al (2000) in Asian patients. However, Chen et al (2000) reported more marked iron overload, with no difference between males and females. Ferritin levels were not associated with the α-globin genotype, Hb level or reticulocyte count. LIC was higher than expected on the basis of ferritin levels in three evaluated patients, suggesting that ferritin levels may underestimate the degree of iron overload, as in other conditions, such as thalassaemia intermedia (Fiorelli et al, 1990).

Chen et al (2000) found three cases of heart failure secondary to iron overload and suggested a correlation between abnormal diastolic function and increasing serum ferritin. However, the systolic function was normal in our patients and the abnormal relaxation pattern found in a small group of patients can be a common pattern in elderly and may represent the earliest manifestation of diastolic dysfunction in younger individuals. This apparent mild cardiac involvement in Sardinian patients with Hb H disease could be due to the relatively young age of this patient cohort and needs to be confirmed by a longer follow-up. Further studies are also necessary to understand if the dilatation of the aortic root is clinically relevant in patients with Hb H disease and to investigate its cause. Haemolytic disease is now known to be an important causative factor in pulmonary hypertension, because the high levels of Hb in the plasma result in the consumption of nitric oxide, a critical regulator of smooth muscle relaxation, endothelial adhesion molecule expression and platelet activation and aggregation (Rother et al, 2005). If the absence of pulmonary hypertension is confirmed by a longer follow-up, it could be explained by the low grade of intravascular haemolysis in Hb H disease.

During pregnancy, worsening of anaemia, with occasional need of red cell transfusions, has been associated with the expansion of blood volume (Vaeusorn et al, 1988; Lin et al, 1990). Vaeusorn et al (1998) reported pre-eclampsia in 18% and congestive heart failure in 9% of 34 pregnancies among 29 Thai women, while in our group only 5·5% had pre-eclampsia and none had heart failure. An 11% risk of miscarriage was found in our population. This percentage is similar to that of the general Italian population, even when threatened abortions, preterm newborns, and Caesarean deliveries are taken into account (Sabbadini & Sebastiani, 2002). In Italy, in the general population, the prevalence of neonates born at term with a birth weight <2·5 kg is only 2% while in newborns of mothers with Hb H disease it was 5·5%. Thus it is possible that maternal anaemia may compromise the fetal growth, even if only rarely. On the other hand, pregnancy is confirmed as one of the most significant causes of transfusion need in adulthood, regardless of genotype.

From our findings we conclude that Hb H disease is usually more benign in the Sardinian population than in south-east Asia and the Middle East. It needs regular monitoring for early detection and treatment of possible complications, such as worsening of anaemia that may require red cell transfusion, cholelithiasis and iron overload. For this reason, a better knowledge of Hb H among paediatricians will enable earlier diagnosis, which should be suspected in patients with microcytic hypochromic anaemia with reticulocytosis. Screening the partners of women with Hb H disease, to detect their α-thalassaemia carrier status, is needed to provide these couples with the appropriate genetic counselling before planning a pregnancy. A longer follow-up is required to monitor the clinical manifestations of the disease, to detect unknown complications that may manifest later in life and to better evaluate the quality of life in these patients.

Acknowledgements

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References

We are indebted to Prof. Antonio Cao for his continuous support and advice. We thank Franca Rosa Demartis for editorial assistance.

Research grants and any other financial support

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References

This study was supported by a grant from Regione Sardegna (L.R. 11, 1998). Enerca (European Network for Rare Congenital Anaemias), Ithanet (Electronic Infrastructure for Thalassemia Research Network).

References

  1. Top of page
  2. Summary
  3. Patients and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. Research grants and any other financial support
  8. References
  • Bosma, P.J., Chowdhury, J.R., Bakker, C., Gantla, S., De Boer, A., Oostra, B.A., Lindhout, D., Tytgat, G.N.J., Jansen, P.L.M., Elferink, R.P.J.O. & Chowdhury, N.R. (1995) The genetic basis of the reduced expression of bilirubin UDP-glucuronosyltransferase 1 in Gilbert's syndrome. The New England Journal of Medicine, 333, 11711175.
  • Brissot, P., Bourel, M., Herry, D., Verger, J.P., Messner, M., Beaumont, C. Regnouard, F., Ferrand, B. & Simon, M. (1981) Assessment of liver iron content in 271 patients: a re-evaluation of direct and indirect methods. Gastroenterology, 80, 557565.
  • Chen, F.E., Ooi, C., Ha, S.Y., Cheung, B.M., Todd, D., Liang, R., Chan, T.K. & Chan, V. (2000) Genetic and clinical features of haemoglobin H disease in Chinese patients. The New England Journal of Medicine, 343, 544550.
  • Chui, D.H.K., Fucharoen, S. & Chan, V. (2003) Haemoglobin H disease: not necessarily a benign disorder. Blood, 101, 791800.
  • Desmet, V.J., Gerber, M., Hoofnagle, J.H. & Scheuer, P.J. (1994) Classification of chronic hepatitis: diagnosis, grading and staging. Hepatology, 19, 15131520.
  • Fiorelli, G., Fargion, S., Piperno, A., Battafarano, N. & Cappellini, M.D. (1990) Iron metabolism in thlassaemia intermedia. Haematologica, 75, 8995.
  • Galanello, R., Pirastu, M., Melis, M.A., Paglietti, E., Moi, P. & Cao, A. (1983) Phenotype–genotype correlation in haemoglobin H disease in childhood. Journal of Medical Genetics, 20, 425429.
  • Galanello, R., Sollaino, C., Paglietti, E., Barella, S., Perra, C., Doneddu, I., Pirrone, M.G., Maccioni. L. & Cao, A. (1998) α-Thlassaemia carrier identification by DNA analysis in the screening for thalassaemia. American Journal of Hematology, 59, 273278.
  • Higgs, D.R. (2001) Molecular mechanisms of α-thlassaemia. In: Disorders of Haemoglobin; Genetics, Pathophysiology, and Clinical Management (ed. by M.H.Steinberg, B.G.Forget, D.R.Higgs & R.L.Nagel), pp. 405430. Cambridge University Press, Cambridge, UK.
  • Higgs, D.R. & Bowden, D.K. (2001) Clinical and laboratory features of the α-thlassaemia syndromes. In: Disorders of Haemoglobin; Genetics, Pathophysiology, and Clinical Management (ed. by M.H.Steimberg, B.G.Forget, D.R.Higgs & R.L.Nagel), pp. 431469. Cambridge University Press, Cambrige, UK.
  • Hsu, H.-C., Lin, C.-K., Tsay, S.H., Tse, E., Ho, C.H., Chow, M.P., Yung, C.H. & Peng, H.W. (1990) Iron overload in Chinese patients with haemoglobin H disease. American Journal of Hematology, 34, 287290.
  • Kanavakis, E., Papassotiriou, I., Karagiorga, M., Vrettou, C., Metaxotou-Mavrommati, A., Stamoulakatou, A., Kattamis, C. & Traeger-Synodinos, J. (2000) Phenotypic and molecular diversity of haemoglobin H disease: a Greek experience. British Journal of Haematology, 111, 915923.
  • Kitabatake, A., Inoue, M., Asao, M., Tanouchi, J., Masuyama, T., Abe, H., Morita, H., Senda, S. & Matsuo, H. (1982) Transmitral blood flow reflecting diastolic behavior of the left ventricle in health and disease: a study by pulsed Doppler technique. Japanese Circulation Journal, 40, 92102.
  • Konus, O.L., Ozdemir, A., Akkaya, A., Erbas, G., Celik, H. & Isik, S. (1998) Normal liver, spleen, and kidney dimensions in neonates, infants, and children: evaluation with sonography. American Journal of Roentgenology, 171, 16931698.
  • Liebhaber, S.A., Cash, F.E. & Ballas, S.K. (1986) Human α-globin gene expression. The dominant role of the α2-locus in mRNA and protein synthesis. Journal of Biological Chemistry, 261, 1532715333.
  • Lin, C.K., Peng, H.W., Ho, C.H. & Yung, C.H. (1990) Iron overload in untransfused patients with haemoglobin H disease. Acta Haematogica, 83, 137139.
  • Lorey, F., Cunningham, G., Vichinsky, E.P., Lubin, B.H., Witkowska, H.E., Matsunaga, A., Azimi, M., Sherwin, J., Eastman, J., Farina, F., Waye, J.S. & Chui, D.H. (2001) Universal newborn screening for Hb H disease in California. Genetic Testing, 5, 93100.
  • Moi, P., Cash, F.E., Liebhaber, S.A., Cao, A., Pirastu, M. (1987) An initiation codon mutation (AUG [RIGHTWARDS ARROW] GUG) of the human α1-globin gene. Journal of Clinical Investigation, 80, 14161421.
  • Mosca, A., Paleari, R., Rosti, E., Luzzana, M., Barella, S., Sollaino, C. & Galanello, R. (1996) Simultaneous automated determination of glucose 6-phosphate dehydrogenase and 6-phosphogluconate dehydrogenase activities in whole blood. European Journal of Clinical Chemistry and Clinical Biochemistry, 34, 431438.
  • Orkin, S.H., Goff, S.C. & Hechtman, R.L. (1981) Mutation in a interverting sequence splice junction in man. Proceedings of the National Academy of Sciences of the United States of America, 78, 50415045.
  • Papassotirious, I., Traeger-Syndinos, J., Kanavakis, E., Karagiorga, M., Stamoulakatou, A. & Kattamis, C. (1998) Erythroid marrow activity and haemoglobin H levels in haemoglobin H disease. Journal of Pediatric Hematology/Oncology, 20, 539544.
  • Park, S.H., Shub, C., Nobrega, T.P., Bailey, K.R. & Seward, J.B. (1996) Two-dimensional echocardiographic calculation of left ventricular mass as recommended by the American Society of Echocardiography: correlation with autopsy and M-mode echocardiography. Journal of the American Society of Echocardiography, 2, 119128.
  • Pirastu, M., Saglio, G., Chang, J.C., Cao, A. & Kan, Y.W. (1984) Initiation codon mutation as a cause of α-thlassaemia. Journal of Biological Chemistry, 259, 12351237.
  • Pootrakul, P., Sirankapracha, P., Hemsorach, S., Moungsub, W., Kumbunlue, R., Piangitjagum, A., Wasi, P., Ma, L. & Schrier, S.L. (2000) A correlation of erythrokinetics, ineffective erythropoiesis, and erythroid precursor apoptosis in Thai patients with thlassaemia. Blood, 96, 26062612.
  • Romao, L., Cash, F., Weiss, I., Liebhaber, S., Pirastu, M., Galanello, R., Loi, A., Paglietti, E., Ioannou, P. & Cao, A. (1992) Human alpha-globin gene expression is silenced by terminal truncation of chromosome 16p beginning immediately 3′ of the zeta-globin gene. Human Genetics, 89, 323328.
  • Rother, R.P., Bell, L., Hillmen, P. & Gladwin, M.T. (2005) The clinical sequelae of intravascular hemolysis and extracellular plasma haemoglobin. The Journal of the American Medical Association, 293, 16531662.
  • Sabbadini, L. & Sebastiani, G. (2002) Il Percorso della Maternità: Gravidanza, parto e Allattamento al Seno, p. 23. ISTAT (Istituto Nazionale di Statistica), Rome, Italy.
  • Schiller, N.B. (1990) Pulmonary artery pressure estimation by Doppler and two-dimensional echocardiography. Cardiology Clinics, 8, 277287.
  • Schiller, N.B., Shah, P.M., Crawford, M., DeMaria, A., Devereux, R., Feigenbaum, H., Gutgesell, H., Reichek, N., Sahn, D. & Schnittger, I. (1989) Recommendations for quantitation of the left ventricle by two-dimensional echocardiography: American Society of Echocardiography Committee on Standards, Subcommittee on Quantitation of Two-Dimensional Echocardiograms. Journal of the American Society of Echocardiography, 2, 358367.
  • Vaeusorn, O., Fucharoen, S. & Wasi, P. (1988) A study of thlassaemia associated with pregnancy. Birth Defects Original Article Series, 23, 295299.
  • Wasi, P., Na-Nakorn, S. & Pootrakul, S.-N. (1974) The α thlassaemias. Clinics in Haematology, 3, 383410.
  • Waye, J.S., Eng, B., Patterson, M., Waye, J.S., Eng, B., Patterson, M., Carcao, M.D., Olivieri, N.F. & Chui, D.H.K. (2001) Haemoglobin H (Hb H) disease in Canada: molecular diagnosis and review of 116 cases. American Journal of Hematology, 68, 1115.
  • World Health Organization Working Group (1986) Use and interpretation of anthropometric indicators of nutritional status. Bulletin of the World Health Organization, 643, 929941.