Skip Navigation

Severe Bacterial Infection in Transfusion-Dependent Patients with Thalassemia Major

  1. Shih-Chung Wang1,
  2. Kai-Hsin Lin2,
  3. Jimmy P. S. Chern3,
  4. Meng-Yao Lu2,
  5. Shiann-Tarng Jou2,
  6. Dong-Tsamn Lin2, and
  7. Kuo-Sin Lin2
  1. 1Department of Pediatrics, Chang-Hua Christian Hospital, Chang-Hua
  2. 2Department of Pediatrics, College of Medicine, National Taiwan University Hospital, Taipei
  3. 3Department of Medicine, Tao-Yuan Hospital, Department of Health, Tao-Yuan, Taiwan
  1. Reprints or correspondence: Dr. Kai-Hsin Lin, Dept. of Pediatrics, National Taiwan University Hospital, No. 7 Chungshan S. Rd., Taipei, Taiwan 100 (link{at}ha.mc.ntu.edu.tw).
 
Next Section

Abstract

The incidence and clinical spectrum of severe bacterial infection were studied in 89 patients with thalassemia major that was diagnosed between January 1971 and March 2002. There were 20 patients with 24 episodes of severe bacterial infection, resulting in an incidence of 1.6 infections per 100 patient-years. The clinical spectrum included liver abscess (6 cases), septicemia (6 cases), soft-tissue infection (2 cases), osteomyelitis (2 cases), corneal ulcer (1 case), enteritis (1 case), and abscesses of the lung, kidney, intra-abdominal region, retropharynx, gums, and buttocks (1 case each). The leading causal microorganisms were gram-negative bacilli, especially Klebsiella pneumoniae (10 of 20 isolates). Other responsible pathogens were Pseudomonas aeruginosa (2/20), Vibrio vulnificus (2/20), Acinetobacter baumanii (1/20), Streptococcus intermidius (1/20), Yersinia enterocolitica (1/20), Staphylococcus aureus (1/20), Escherichia coli (1/20), and Salmonella species (1/20). Splenectomy and delays in the start of iron-chelating therapy were 2 independent risk factors.

Thalassemia is one of the major health concerns in Taiwan. The rate of carriage of β-thalassemia is ⩾1.1% [1, 2]. With receipt of regular blood transfusions and iron-chelating therapy, the prognosis of the disease improves [3]. Unfortunately, the prevalence of severe complications due to iron overload is still high. Excluding heart failure, infection is the predominant cause of death; it is directly responsible for 12%–46% of patient deaths [3, 4]. Early recognition with prompt treatment and prevention of predisposing factors are essential in the treatment of patients with thalassemia and infection. Nonetheless, data relevant to the incidence of infection and the spectrum of causal organism are limited in Taiwan [5, 6]. In the past 9 years, we have observed 104 transfusion-dependent patients treated at National Taiwan University Hospital (NTUH; Taipei) and Tao-Yuan General Hospital (TYGH; Tao-Yuan), Taiwan's 2 major thalassemia clinics. The incidence of infection and the spectrum of clinical presentations were reviewed retrospectively.

Patients and methods. During the period of June 1994 through October 2002, a total of 104 transfusion-dependent patients received regular transfusions at 2 different teaching hospitals in Taiwan (NTUH and TYGH). Among patients with a definite diagnosis of β-thalassemia major (based on findings of hemoglobin electrophoresis and/or molecular defect), only those with complete and reliable clinical records were included in this study. A cohort of 89 of 104 patients met the criteria. The distribution of patient population by calendar period of diagnosis is as follows: 1971–1979, 17 cases; 1980–1989, 40 cases; 1990–1999, 31 cases; and 2000–2002, 1 case. Patient-years of observation were calculated as the sum of the years of medical history available from 1971 through 2002. The compliance with iron-chelating therapy was considered to be “good” if ⩾5 deferoxamine infusions per week were observed.

Patients were considered for enrollment in this study if a severe infection involving major organs was found, if hospitalization was required for administration of parenteral antibiotics, and/or if surgical intervention was necessary. The diagnosis was based on the clinical presentations and evidence from physical examinations and imaging studies. Bacterial infection was confirmed by isolation of pathogens from blood, urine, stool, pus, or body fluid specimens. Data regarding clinical presentation, investigations, potential risk factors, and causal organisms were reviewed together. The serum ferritin level determined at the very last examination before the onset of infection was used as an estimate of the severity of iron burden, because the level was further elevated during inflammation.

Differences between patients with and patients without bacterial infections were analyzed using the independent t test, χ2 test, and Fisher's exact test, as appropriate. Multivariate logistic regression analysis was performed to select independent risk factors. A 2-tailed P value of <.05 was considered to be statistically significant.

Results. Of 89 patients with β-thalassemia major during the 30-year period (total no. of patient-years, 1466.6 ), 20 patients (22.5%) had 24 episodes of bacterial infections, resulting in an incidence of 1.6 infections per 100 patient-years. The median duration of follow-up was 16 years (range, 7 months to 29.5 years). The clinical characteristics of these patients are shown in table 1. The mean age (±SD) at diagnosis of bacterial infection was 17.5 ± 6.0 years. All of the recorded infections were distributed from 1985 through 2002: there were 2 cases during 1985–1989 and 11 cases each during 1990–1999 and 2000–2002. The mean serum ferritin concentration (±SD) determined at the very last examination before the onset of bacterial infection was 5375.6 ± 2982.7 µg/L, indicating moderate iron overload. All 20 patients were offered deferoxamine therapy. Compliance was poor for 11 patients and good for 9. Thirteen patients underwent splenectomy, 10 had cases complicated by diabetes mellitus, and 9 had chronic hepatitis C virus (HCV) infection. As reported in previous studies [7–9], we also found overlaps between diabetes mellitus and chronic hepatitis C (in 5 patients). Klebsiella pneumoniae was the most common causative organism in this patient population (10 of 20 isolates). Other responsible pathogens were Pseudomonas aeruginosa (2 of 20 isolates), Vibrio vulnificus (2 isolates), Acinetobacter baumanii (1 isolate), Streptococcus intermidius (1 isolate), Yersinia enterocolitica (1 isolate), Staphylococcus aureus (1 isolate), Escherichia coli (1 isolate), and Salmonella species (1 isolate). Four cases for which pathogens were not isolates were also included in the infection group, because there were the obvious symptoms of infection and improvement of clinical status after receipt of antibiotic therapy.

Table 1
View larger version:
Table 1

Clinical characteristics of 20 patients with β-thalassemia major and severe bacterial infection.

Two of these 20 patients died of sepsis, which was caused by E. coli in one patient and K. pneumoniae in the other. It is noteworthy that the patient who died of K. pneumoniae sepsis had an infection that was complicated by congestive heart failure, despite his relatively low serum ferritin level and good compliance with iron-chelating therapy.

The risk factors associated with bacterial infection were assessed. Detailed data are shown in table 2. Univariate analysis demonstrated that the factors that affected bacterial infections were age at the start of iron-chelating therapy (P < .001), previous splenectomy (P < .001), diabetes mellitus (P = .001), and positive results of tests for anti-HCV antibody (P = .010). On logistic regression analysis, previous splenectomy and age at the start of iron-chelating therapy were identified as independent risk factors for bacterial infection.

Table 2
View larger version:
Table 2

Characteristics of 89 patients who had β-thalassemia, with or without severe bacterial infection.

Discussion. The treatment of thalassemia major has changed in many ways in the past 3 decades. In Taiwan, patients with thalassemia rarely received adequate blood transfusions from 1957 to 1974 [10]. From 1979 to 1988, the transfusion regimen evolved from correction of symptomatic anemia to the so-called “hypertransfusion,” which aimed to maintain a minimum hemoglobin level of 9–10 g/dL. However, only few patients received regular, daily iron-chelating therapy [11]. Alternatively, allogeneic bone marrow transplantation has been used for treatment of thalassemia in Taiwan since 1984 [12].

Since 1995, the cost of thalassemia treatment has been covered by national health insurance. Today, because they benefit from this insurance, nearly all patients with thalassemia major can receive adequate blood transfusions and iron-chelating therapy. As the outlook for patients with transfusion-dependent thalassemia improves, information about complications becomes particularly important. However, publications on the incidence and the spectrum of bacterial infections in patients with thalassemia major have been limited. Current information about severe infection in persons with thalassemia comes primarily from retrospective studies and case record analyses [4, 13, 14].

The incidence of severe bacterial infections in the present study was 1.6 infections per 100 patient-years. As mentioned above, few patients received adequate treatment for thalassemia in the 1970s, and only the minority of patients treated before the 1990s had available long-term follow-up data. The true incidence of severe bacterial infection might be underestimated among these patients. However, physicians have become aware of infectious complications in patients with thalassemia, and sonography and/or CT have been used to detect the possible existence of infection foci in the past 2 decades. These may be the reasons why there have been more diagnoses of infectious complications in the cohort of patients with thalassemia since the 1990s.

The major causative organisms were gram-negative bacilli, especially K. pneumoniae. Only 10% of the causative agents were gram-positive bacteria. These results were similar to those reported from Thailand [13]. According to a review by Wanachiwanawin [13], one-half of the major organisms responsible for severe infection in patients with thalassemia in Thailand were gram-negative bacilli, such as E. coli (26% of organisms) and K. pneumoniae (23% of organisms). Much like Thailand, in Taiwan, antibiotics are not strictly controlled and can be easily obtained without a prescription. This may explain the high prevalence of severe infections with gram-negative bacilli in patients with thalassemia in these 2 countries.

The pattern of bacterial infections in our study is also significant. Primary K. pneumoniae-associated liver abscess is a well-known complication in diabetic patients in Taiwan [15–18]. However, it is rarely reported as a complication of β-thalassemia major [5]. Of interest, none of the 6 patients in our study had biliary tract infection. This finding is consistent with the findings of previous reports on primary K. pneumoniae-associated liver abscess in Taiwan [16, 17]. It has been suggested that the cirrhotic liver might have permitted bacteria to bypass the liver's reticuloendothelial system and allow the microorganism to disseminate via the peripheral blood [19]. Nonetheless, it remains unclear why patients with β-thalassemia in Taiwan have a greater frequency of K. pneumoniae-associated liver abscess than do those who live in the other regions of the world [20].

Y. enterocolitica seems to have been a less important cause of infection in our study than in studies from Western countries [21, 22]. In Taiwan, only a few cases of Y. enterocolitica infection have been reported in patients with thalassemia [23]. This is also evident in Thailand [13]. Geographical variability could play a role for the difference.

In the present study, iron overload was noted in the infection group. It is clear that many organisms, such as Y. enterocolitica, Klebsiella species, E. coli, Streptococcus pneumonia, P. aeruginosa and Listeria monocytogenes, have been shown to increase in virulence in the presence of excess iron in vitro [24–26]. However, 3 patients with relatively low levels of serum ferritin (1040 µg/L, 1467 µg/L, and 1032 µg/L) developed severe infection, and 1 of them died of K. pneumoniae infection.

Patients with thalassemia major have a high risk of developing postsplenectomy sepsis [27]. Thirteen (65%) of our 20 patients who had severe bacterial infection had previously undergone splenectomy (P < .01). This suggests that splenectomy is a significant predisposing factor for serious bacterial infection among transfusion-dependent patients with thalassemia. However, other predisposing factors among patients with thalassemia major, such as alterations in complement activation, abnormal levels of immunoglobulins, cardiopulmonary disease, and hemochromatosis, also increase the predisposition.

In summary, clinicians should be aware of any potential infections when taking care of patients with thalassemia major, especially those whose cases are complicated by a history of splenectomy and/or diabetes mellitus. Once fever or infection is observed, aggressive investigation and proper treatment should commence as soon as possible. The predisposing factors for infection, such as anemia and iron overload, should be eliminated by regular administration of blood transfusion and proper iron-chelating therapy. Moreover, patients with thalassemia major have to be educated to seek early medical care when fever or other signs of infection develop. In Taiwan, for the treatment of patients with thalassemia who have bacterial infection, we recommend use of antibiotics with activity against gram-negative bacteria, especially K. pneumoniae.

  • Received January 6, 2003.
  • Accepted June 5, 2003.
Previous Section
 

References

  1. ↵
    1. Ko TM,
    2. Hsu PM,
    3. Chen CJ,
    4. Hsieh FJ,
    5. Hsieh CY,
    6. Lee TY
    . Incidence study of heterozygous β-thalassemia in northern Taiwa. Taiwan Yi Xue Hui Za Zhi 1989;88:678-81.
    Medline
  2. ↵
    1. Lin KH,
    2. Lin LL,
    3. Su S,
    4. Lin KS
    . The incidence of β-thalassemia trait in Taiwa. Int J Pediatr Oncol 1994;1:383-88.
  3. ↵
    1. Borgna-Pignatti C,
    2. Rugolotto S,
    3. DeStefano P,
    4. et al
    . Survival and disease complications in thalassemia majo. Ann N Y Acad Sci 1998;850:227-31.
    CrossRefMedlineWeb of Science
  4. ↵
    1. Fucharoen S,
    2. Piankijagum A,
    3. Wasi P
    . Deaths in β-thalassemia/HbE patients secondary to infection. Birth Defects Orig Artic Ser 1987;23:495-500.
    Medline
  5. ↵
    1. Hwang CF,
    2. Lee CY,
    3. Lee PI,
    4. et al
    . Pyogenic liver abscess in β-thalassemia major—report of two case. Acta Paediatr Sin 1994;35:466-70.
    Medline
  6. ↵
    1. Peng CT,
    2. Tsai CH,
    3. Wang JH,
    4. Chiu CF,
    5. Chow KC
    . Bacterial infection in patients with transfusion-dependent β-thalassemia in central Taiwa. Acta Paediatr Taiwan 2000;41:318-21.
    Medline
  7. ↵
    1. Chern Jimmy PS,
    2. Lin KH,
    3. Lu MY,
    4. et al
    . Abnormal glucose tolerance in transfusion-dependent β-thalassemic patient. Diabetes Care 2001;24:850-4.
    Abstract/FREE Full Text
    1. De Sanctis V,
    2. Zurlo MG,
    3. Senesi E,
    4. et al
    . Insulin dependent diabetes in thalassemi. Arch Dis Child 1988;63:58-62.
    Abstract/FREE Full Text
  9. ↵
    1. Labropoulou-Karatza C,
    2. Goritsas C,
    3. Fragopanagou H,
    4. et al
    . High prevalence of diabetes mellitus among adult β-thalassemic patients with chronic hepatitis. Eur J Gastroenterol Hepatol 1999;11:1033-6.
    CrossRefMedlineWeb of Science
  10. ↵
    1. Tsai MD,
    2. Lin KS,
    3. Lee TC
    . Clinical observation of β-thalassemia in infants and childre. Acta Paediatr Sin 1975;16:119-36.
  11. ↵
    1. Lin KH,
    2. Lin KS
    . Results of therapy for β-thalassemia majo. J Formosan Med Assoc 1992;91:126-30.
    Medline
  12. ↵
    1. Lin KH,
    2. Lin KS
    . Allogeneic bone marrow transplantation for thalassemia in Taiwan: factors associated with graft failur. Am J Pediatr Hematol Oncol 1989;11:417-23.
    MedlineWeb of Science
  13. ↵
    1. Wanachiwanawin W
    . Infections in E-β thalassemi. J Pediatr Hematol Oncol 2000;22:581-7.
    CrossRefMedlineWeb of Science
  14. ↵
    1. Issaragrisil S,
    2. Wanachiwanawin W,
    3. Bhuripanyo K,
    4. et al
    . Infection in thalassemia: a retrospective study of 1018 patients with β-thalassemia/HbE diseas. Birth Defects Orig Artic Ser 1987;23:505-11.
    Medline
  15. ↵
    1. Cheng DL,
    2. Liu YC,
    3. Yen MY,
    4. Liu CY,
    5. Shi FW,
    6. Wang LS
    . Causal bacteria of pyogenic liver absces. Taiwan Yi Xue Hui Za Zhi 1989;88:1008-11.
    Medline
  16. ↵
    1. Chang FY,
    2. Chou MY
    . Comparison of pyogenic liver abscesses caused by Klebsiella pneumoniae and non-K. pneumoniae pathogen. J Formos Med Assoc 1995;94:232-7.
    Medline
  17. ↵
    1. Wang JH,
    2. Liu YC,
    3. Lee Susan SJ,
    4. et al
    . Primary liver abscess due to Klebsiella pneumoniae in Taiwa. Clin Infect Dis 1998;26:1434-8.
    Abstract/FREE Full Text
  18. ↵
    1. Chang SC,
    2. Fang CT,
    3. Hsueh PR,
    4. Chen YC,
    5. Luh KT
    . Klebsiella pneumoniae isolates causing liver abscess in Taiwa. Diagn Microbiol Infect Dis 2000;37:279-84.
    CrossRefMedlineWeb of Science
  19. ↵
    1. Conn HO
    . Spontaneous peritonitis and bacteremia in Laennec's cirrhosis caused by enteric organism. Ann Intern Med 1964;60:568-80.
    Abstract/FREE Full Text
  20. ↵
    1. Chung BHY,
    2. Ha SY,
    3. Chan GCF,
    4. et al
    . Klebsiella infection in patients with thalassemi. Clin Infect Dis 2003;36:575-9.
    Abstract/FREE Full Text
  21. ↵
    1. Cherchi GB,
    2. Pacifico L,
    3. Cossellu S,
    4. et al
    . Prospective study of Yersinia enterocolitica infection in thalassemic patient. Pediatr Infect Dis J 1995;14:579-84.
    MedlineWeb of Science
  22. ↵
    1. Adamkiewicz TV,
    2. Berkovitch M,
    3. Krishnan C,
    4. Polsinelli C,
    5. Kermack D,
    6. Olivieri NF
    . Infection due to Yersinia enterocolitica in a series of patients with β-thalassemia: incidence and predisposing factor. Clin Infect Dis 1998;27:1362-6.
    Abstract/FREE Full Text
  23. ↵
    1. Yang CP,
    2. Hung IJ
    . Infectious complications in Cooley's anemia—report of 3 case. Acta Pediatr Sin 1997;38(Suppl B):165.
  24. ↵
    1. Bullen JJ
    . The significance of iron in infectio. Rev Infect Dis 1981;3:1127-38.
    MedlineWeb of Science
    1. Miles AA,
    2. Khimji PL,
    3. Maskell J
    . The variable response of bacteria to excess ferric iron in host tissue. J Med Microbiol 1979;12:17-28.
    Abstract/FREE Full Text
  26. ↵
    1. Polk HC Jr,
    2. Miles AA
    . Enhancement of bacterial infection by ferric iron: kinetics, mechanisms, and surgical significanc. Surgery 1971;70:71-7.
    MedlineWeb of Science
  27. ↵
    1. Bisharat N,
    2. Omari H,
    3. Lavi I,
    4. Raz R
    . Risk of infection and death among post-splenectomy patient. J Infect 2001;43:182-6.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. Clin Infect Dis. (2003) 37 (7): 984-988. doi: 10.1086/378062
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 54 (5)
  1. Clinical Infectious Diseases
  1. Alert me to new issues

Most