Helicobacter pylori eradication can induce platelet recovery in idiopathic thrombocytopenic purpura

Helicobacter pylori has been clearly recognized as the main cause of gastritis and most cases of peptic ulcer,1 and it has also been implicated in the pathogenesis of gastric adenocarcinoma and mucosa-associated lymphoid tissue lymphoma.2 It has been hypothesized that the host immunologic response against H pylori plays a main role in determining gastric mucosal injury, through the release of cytokines and the action of autoantibodies against H⁺/K⁺-adenosine triphosphatase of gastric epithelial cells.3,4 Moreover, there is increasing evidence that H pylori strains are highly diverse and that strain diversity associated with variability in host immune response may contribute to the clinical outcome in infected patients.5 In the last few years, H pylori has been implicated in the pathogenesis of some autoimmune diseases, such as rheumatoid arthritis,6 autoimmune thyroid diseases,7 and idiopathic thrombocytopenic purpura (ITP).8 Relevant to this, Gasbarrini et al8recently showed a high prevalence of H pylori infection in patients with ITP and reported a good response to bacterium eradication in most of them. This finding has been confirmed in a further anecdotal ITP case.9 Thus, we performed a prospective open study to verify the prevalence of H pylori infection and the effects of its eradication in a larger proportion of patients with chronic ITP.

Thirty patients with ITP were evaluated, including 13 males and 17 females; median age was 50.3 years (range 17-89). ITP was defined by idiopathic thrombocytopenia (platelets < 100 × 10⁹/L) when other causes had been excluded, megakaryocytic hyperplasia in the bone marrow, shortened platelet survival (measured using ¹¹¹In-labeled platelets), and increased autoantibodies against platelets (platelet-associated immunoglobulin G [PAIgG]) in the serum.

H pylori infection was assessed by ¹³C urea breath test (Helicobacter test, Infai, Bochum, Germany), by the detection of serum antibodies (indirect immunofluorescence) and, whenever possible, by histologic examination (Giemsa stain) of specimens obtained by an upper gastrointestinal endoscopy.

The presence of serum antibodies against hepatitis C virus was also evaluated by the enzyme-linked immunosorbent assay and confirmed by the recombinant-based immunoblot assay. All immunosuppressive treatments were withdrawn 1 month before eradication except in 1 patient.H pylori eradication was performed with amoxicillin (1000 mg twice daily), clarithromycin (250 mg 3 times daily), and pantoprazole (40 mg twice daily) for 7 days. Eradication was assessed by urea breath test 4 weeks after treatment withdrawal. Platelet counts were monitored every 2 weeks and assessed at 3 and 6 months, after the end of treatment. In statistical analysis, data were expressed as the mean (SD) or median (range) as appropriate and were analyzed by using the t test. Percentages were compared by χ² test (Fisher exact test for values ≤ 5). AP value < .05 was considered statistically significant.

Approval for this study was obtained from the Institutional Review Board of the University of Modena, and informed consent was provided according to the Declaration of Helsinki.

H pylori infection was found in 13 of 30 ITP patients (43.3%) by ¹³C urea breath test. A prevalence of infected males was noted (8 infected vs 5 uninfected; 5 infected females vs 12 uninfected); median age was higher in infected patients (63.2 [SD 14.6] vs 47 [SD 20.8] years, P < .02). Platelet count was similar in infected and uninfected patients (53.1 × 10⁹/L [SD 28.5] vs 41.7 × 10⁹/L [SD 14.8], P < .16). Eight of 13 H pylori–infected patients showed, at diagnosis, a severe form of ITP (platelets < 30 × 10⁹/L) and needed immunosuppressive treatment (one also needed splenectomy) before eradication. All patients showed a shortened platelet survival (2-4 days vs 8-9 days, as normal value), except for the patients no. 9 and 21 (Table 1). In 10 patients, the diagnosis of H pylori infection was also confirmed by histologic examination, while in 9 the presence of serum antibodies against the bacterium was also detected. No patient was found to be positive for serum antibodies against hepatitis C virus. The bacterium eradication was obtained in 12 of 13 H pylori–positive patients (92.3%). The follow-up, performed for a median of 8.33 months (range 6-12), showed that 6 of the 12 eradicated patients (50%) had a significant increase in platelet count after 3 and 6 months (P < .007) (Table 1); 4 patients (33.3%) achieved a complete response (CR) (P < .03), while 2 patients showed a partial response (platelet count not higher than 120 × 10⁹/L) (P < .16) (Figure1A). The outcome over time of platelet counts of H pylori–infected and treated patients compared with noninfected patients looked similar (Figure 1B). Four of 30 ITP patients were PAIgG-positive; 2 of these were infected withH pylori. Among the responsive patients, 2 who achieved CR were PAIgG-positive, while the remaining 4 were PAIgG-negative. The only patient in whom H pylori eradication failed (no. 1) has been unresponsive. Only 1 patient (no. 7), with a platelet count of 25 × 10⁹/L, required steroid treatment for 1 month after eradication; subsequently, the platelet count increased and maintained in normal range without therapy. In 5 of the 6 responsive patients followed for more than 6 months, the platelet count did not change after H pylori eradication. In 1 patient (no. 10), the ITP relapsed 7 months after eradication, although the breath test remained negative.

Our data lend further support to a relationship betweenH pylori infection and ITP. The bacterium might induce, at least in some cases, the formation of platelet autoantibodies by means of a chronic immunologic stimulus or a cross mimicry between platelets and itself.8,9 In respect to the only nonanecdotal report of the literature by Gasbarrini et al,8 we observed, in our series of ITP patients, a lower prevalence of H pyloriinfection (43.33% vs 61.11%, P < .37) and a lower frequency of platelet response to the same eradication treatment (46.14% vs 72.72%, P < .40). Moreover, only 33.33% of our patients achieved a CR, versus 100% of eradicated patients in the above-mentioned report (P < .27), even though eradication has been obtained in a similar percentage of patients (92.3% vs 100%). We followed our patients for a median of 8.33 months—versus a maximum of 4 months in the study by Gasbarrini et al.8This allowed us   to document a relapse of ITP 7 months after H pylori eradication in 1 PAIgG-negative patient. Finally, variety in the frequency of H pylori–associated ITP and in the rate of platelet response to bacterium eradication may be related either to the variability of host immune response to H pylori or to the bacterium's high genetic diversity, ie, to the existence of different H pylori strains with possibly different pathogenic potential.5 

In conclusion, even though the pathogenetic mechanisms of H pylori–induced thrombocytopenia remain obscure, the search forH pylori infection and an attempt to eradicate bacterium in positive cases seem appropriate in ITP patients at diagnosis. This approach may be a new good option for a nonimmunosuppressive treatment, at least in some ITP patients. Further investigations on a larger number of patients, with a long follow-up, might allow a better definition of the true prevalence of H pylori infection and the duration of the effect of its eradication in ITP patients.