Helicobacter pylori eradication: a novel therapeutic option in chronic immune thrombocytopenic purpura

Immune thrombocytopenic purpura (ITP) is the most common autoimmune-mediated haematological disorder, characterised by antibodies against platelet surface antigens.1 Its aetiology, pathogenesis and molecular receptor targets remain unclear. It has been associated with HIV infection and hepatitis C.1

Helicobacter pylori is a causative agent in a spectrum of gastrointestinal diseases, including peptic ulcer disease, gastric cancer and MALToma (mucosa-associated lymphoid tissue lymphoma). There is growing evidence of an association between H. pylori eradication and platelet recovery in patients with ITP.2-6 The evidence is strongest in Japan, where H. pylori eradication is now recommended as an initial treatment for infected ITP patients.7 Screening and eradication may be a simpler and safer therapeutic option than immunosuppression or splenectomy in certain ITP patients.8,9 Most studies of H. pylori and ITP are from Japan, Spain and Italy; however, the British Society for Haematology now recommends H. pylori screening and eradication as a treatment in ITP (evidence level, III).10 To date, there have been no published reports of such therapy in Australian patients. Local experience is important, given the potential regional variation in strains of H. pylori.

Sixteen consecutive patients (age range, 24–79 years) with ITP were referred to a single gastrointestinal surgeon (R B W) for laparoscopic splenectomy at a tertiary hospital between August 2005 and November 2007. All patients had been diagnosed with ITP following investigations by a haematologist, including abdominal computed tomography scan, bone marrow biopsy, HIV and hepatitis serological testing, and antibody testing for other autoimmune disorders such as rheumatoid arthritis, antiphospholipid syndrome and systemic lupus erythematosis.

We began routine testing for H. pylori in patients with ITP referred for splenectomy in August 2005. Testing was via gastroscopy, serological testing or urea breath test.11 Patients were treated with standard triple therapy, comprising clarithromycin 500 mg, amoxycillin 1000 mg and oral esomeprazole 20 mg, all twice daily for 1 week. One patient was treated with metronidazole instead of amoxycillin due to penicillin allergy. Eradication was confirmed by urea breath test, with titres of isotope 14C < 50 disintegrations/min 4 weeks after completion of triple therapy indicating a negative result. Patients who had previously undergone successful H. pylori eradication therapy were not treated again.

Platelet counts were measured at least 3-monthly after H. pylori eradication. ITP patients who were H. pylori-negative or whose platelet counts did not rise after H. pylori eradication underwent splenectomy.

Clinical characteristics of the patients and their responses to treatment are summarised in Box 1.

Nine of 16 patients tested positive for H. pylori and all had successful eradication therapy. Of these, five patients showed an initial platelet response, and three of these showed a long-term (> 12 months) response and did not require immunosuppression or splenectomy (Box 2). Only one of these patients had initial severe ITP (platelet count, 4 × 109/L [reference range, 140–450 × 109/L]). Two of three patients who tested positive for anticardiolipin or antinuclear antibodies also tested positive for H. pylori and did not have a sustained platelet response to H. pylori eradication.

Five patients who were H. pylori-positive (and the seven who were H.pylori-negative) had laparoscopic splenectomies between November 2005 and December 2007. There were no conversions to open surgery. Two patients who underwent splenectomy later relapsed, requiring recommencement of immunosuppression; both had achieved successful H. pylori eradication before splenectomy.

This study found that H. pylori eradication was of long-term benefit in three of nine patients with ITP who were H. pylori-positive, leading to a sustained platelet response. Two other patients had a short-term response. H. pylori eradication appeared to be least successful in the context of severe ITP that had already been extensively treated. To our knowledge, this is the first published study of H. pylori eradication therapy in Australian ITP patients.

The association between ITP and H. pylori was first reported in 1998.12 Possible aetiological links include molecular mimicry and cross-reactivity of bacterial and platelet antigens, in association with host factors such as specific human leukocyte antigen (HLA) class II alleles.13,14 A recent Italian study found that the HLA-DRB1*11, -DRB1*14 and -DQB1*03 alleles occur significantly more often in H. pylori-infected ITP patients and indicate a higher probability of platelet response to eradication therapy.15 ITP patients who had successful H. pylori eradication showed significantly decreased serum platelet-associated IgG levels, suggesting that removal of the H. pylori antigenic stimulus leads to platelet recovery.9

However, response rates have varied. A literature review of 13 major studies published between 1998 and 2004 revealed an overall platelet response rate of 52% following H. pylori eradication therapy when used as an adjunct to standard ITP therapy or in otherwise untreated cases.5 Another review concluded that 58% of 482 patients with ITP were H. pylori-positive and, of these, more than 50% showed a significant platelet response to successful H. pylori eradication.13 Long-term follow-up studies, including one with a median follow-up of 31 months, have also shown good results.2,16,17 A large retrospective analysis in Japan yielded a platelet response rate as high as 63%, with remission sustained after 12 months, even among patients whose ITP was refractory to splenectomy.7 A small randomised controlled trial, also in Japan, showed a significant increase in platelet count in the group who had successful eradication therapy.3 Some studies showed that platelet recovery was better in less severe or recent-onset ITP,18,19 or in the absence of concurrent steroid therapy.20

In contrast, several studies have refuted a significant association between H. pylori and ITP.21-23 For example, a prospective study in the United States found that only one of 14 patients who responded to H. pylori eradication had a rise in platelet count.21

There is substantial variability in ITP response to H. pylori eradication in different countries.5,18,20,24 This may be due to variations in bacterial virulence factors, such as CagA expression or host immunological class II HLA factors.2,24 Most Japanese H. pylori strains are CagA-positive, unlike most American strains.9 Patients infected with CagA-positive strains (as measured by anti-CagA IgG antibodies) are thought to have increased platelet response to H. pylori eradication compared with those who are CagA-negative or who have low serum titres of CagA antibodies.3,5 Serum levels of anti-CagA IgG may be used as a predictor of response to H. pylori treatment.2,25

The prevalence of H. pylori infection in ITP patients varies from 22% in a US study21 to 47% in a study from the United Kingdom and Italy19 to 71% in a Spanish study.23 In a comparison of H. pylori and ITP data from the UK and Italy, the prevalence of H. pylori infection in patients with ITP appeared to be similar to that in the general population.19 The prevalence of H. pylori infection in the general Australian population is 25%–38%,26 but its prevalence in Australians with ITP is yet to be determined. In our series, 56% of ITP patients were H. pylori-positive.

The results in this small series support a role for H. pylori eradication in the treatment of patients with ITP in Australia. H. pylori screening and eradication should be considered in patients presenting with ITP, given the low cost and potential for avoidance of immunosuppression and splenectomy. Larger prospective studies are needed to fully ascertain the role of H. pylori in Australian patients with ITP.

1 Clinical characteristics of patients with immune thrombocytopenic purpura (ITP) referred for splenectomy, 2005–2007

2 Pattern of platelet response to successful Helicobacter pylori eradication in five patients with immune thrombocytopenic purpura who responded (reference range, 140–450 × 109/L)