Helicobacter pylori infection an underestimated trigger of bullous pemphigoid?

Abstract

Dear Editors,

Bullous pemphigoid (BP) belongs to the group of autoimmune bullous diseases. It is characterized by subepidermal blistering and the presence of autoantibodies against hemidesmosomal proteins.^1-3 In the literature, BP is associated with advanced age, neurological disease, medication, paraneoplastic syndrome, genetic predisposition, and ultraviolet light.^{4, 5} Some reports found an association between BP and bacterial pathogens.⁶ To our knowledge, we present the first two cases to report complete remission of BP after Helicobacter (H.) pylori eradication.

An 83-year-old female patient presented with new-onset tense bullae and severe pruritus all over her body. No medications were taken by the patient and no previous medical diseases were reported. Laboratory tests showed leukocytosis, eosinophilia, elevated LDH and CRP. Chronic infectious diseases (HIV, hepatitis, Lyme disease, syphilis, toxoplasmosis) were excluded. Histological examination of biopsies from lesions revealed subepidermal, non-acantholytic bullae with an eosinophilic infiltrate. Direct immunofluorescence (DIF) showed linear deposits of C3 at the dermo-epidermal junction. Indirect immunofluorescence (IIF; on salt-split skin) showed linear deposition of antibodies in the bladder roof. IIF diagnostics using the BIOCHIP (Dermatology Mosaic 7, EUROIMMUN, Lübeck, Germany) showed reactivity for BP180 and BP230. After clinical-pathological correlation, we diagnosed bullous pemphigoid. Due to heartburn and occult blood in the stool, a gastroscopy and colonoscopy were performed. The gastroscopy revealed mild reflux esophagitis and an atrophic gastric mucosa. H. pylori was detected in the biopsied gastric mucosa and verified by a fecal antigen test. The colonoscopy was unremarkable.

For the acute exacerbation of BP, we started topical therapy with clobetasol (1–2 times daily) and systemic therapy with prednisolone (initially with a dose of 1 mg/kg body weight [BW]). Prednisolone was gradually reduced by 25% each week after the first week. After achieving a dose of less than 20 mg per day, the dose was gradually reduced every 2 weeks. In addition, we started eradication therapy against H. pylori with amoxicillin (1,000 mg 2 x daily), pantoprazole (40 mg 2 x daily), and clarithromycin (500 mg 2 x daily) for 14 days. The skin improved significantly with this therapy. Successful eradication of H. pylori was confirmed by a negative stool antigen test for H. pylori after 6 weeks. Complete remission was achieved after 12 weeks. Prednisolone and clobetasol were discontinued at the patient's request. There was no recurrence of BP for 24 months, and no local or systemic therapy was required for BP.

In the second case, an 89-year-old female patient presented with an exacerbation of pre-existing BP that was treated for 6 years with two relapses per year. Based on the typical skin lesions and findings (DIF and IIF) at the current presentation, the diagnosis of BP was reconfirmed. DIF showed a linear deposition of IgG and C3 at the dermo-epidermal junction. In the IIF, there was a linear deposition of antibodies in the bladder roof (substrate: salt-split-skin). The BIOCHIP Mosaic detected anti-BP180 and anti-BP230 antibodies in the serum. Her medical history included arterial hypertension, heart failure and mild dementia. Laboratory chemistry also showed eosinophilia. Many medications for BP (e.g., azathioprine, dapsone, doxycycline) were ineffective. Treatment with topical (clobetasol 1–2 times daily) and systemic glucocorticoids (prednisolone initially with a dose of 1 mg/kg BW) in the past resulted in a temporary improvement, so we restarted this therapy. In this case, a positive stool test for the H. pylori antigen and a 13C-urea breath test also indicated active H.-pylori-infection. Helicobacter pylori was also successfully eradicated in the second patient using amoxicillin (1,000 mg 2 x daily), pantoprazole (40 mg 2 x daily) and clarithromycin (500 mg 2 x daily) for 14 days. A negative stool test for H. pylori antigen confirmed successful eradication after 4 weeks. After 16 weeks in the second case, a complete remission was achieved, and prednisolone was discontinued. Clobetasol should only be used if new skin changes occur. A follow-up after 15 months also showed complete remission. There was no need for local or systemic therapy during this time.

The exact pathogenesis of BP is not yet fully understood. In both cases, BP was under control after H. pylori eradication, so no long-term BP treatment was necessary. This temporal relationship suggests that H. pylori should be discussed as a triggering factor for BP.

Chronic H. pylori infection is usually asymptomatic and thus goes undetected for a long time. H. pylori antigens can activate cross-reactive T cells and induce the production of autoantibodies.⁷ An ongoing interaction between bacterial antigens and the immune system could explain the autoimmune link between BP and H. pylori.⁸ In both our cases, the probably long-standing H. pylori infection may have resulted in the production of BP180 and BP230 antibodies, leading to the subepidermal blisters in the skin. However, the exact autoimmune mechanism is still hypothetical and requires further investigation. Our hypothesis is also supported by the significantly increased prevalence of antibodies against HBV, HCV, H. pylori, Toxoplasma gondii and CMV in patients with pemphigus or bullous pemphigoid. The increased antibodies could be the result of a permanent activation of the immune system, regardless of the bacterial or viral pathogen. In the study by Sagi et al., patients with bullous autoimmune disease had an increased prevalence of elevated IgG titers against H. pylori compared to controls (90% vs. 31%), which may indicate an important immunological effect of H. pylori in bullous autoimmune diseases.⁹ An association between H. pylori and bullous autoimmune dermatoses (BAID) was also described in pemphigus vulgaris and linear bullous IgA dermatosis.^{10, 11} These data suggest that our observation may also be of interest in relation to other BAID. Mortazavi et al. also found a significantly higher prevalence of elevated IgG antibodies against H. pylori in untreated pemphigus vulgaris compared to healthy controls (79.3% vs. 59.5%).¹¹ Matsuo et al. reported a linear bullous IgA dermatosis of the sublamina dense type that disappeared completely after eradication of H. pylori.¹⁰ The potential impact of infections on various bullous autoimmune diseases is inconsistent. One retrospective study showed an association between BP and leucocytosis.¹² Leukocytosis is most associated with acute or chronic infections. Leukocytosis is significantly more common in chronic H. pylori infection than in H. pylori negative controls.^{13, 14} Whether this observed association between leukocytosis, H. pylori and BP is causal cannot be conclusively assessed. The important role of eosinophil granulocytes in BP is generally accepted.^{15, 16} Eosinophilia was also found in the tissues and blood of both of our patients.

Meta-analyses showed that the prevalence of H. pylori infection is highest in developing countries. For example, the prevalence of H. pylori infection in Africa is 70.1%, twice that of Western Europe.¹⁷ Differences in hygiene standards may be an important cause. We could not find any data in the literature on the increased prevalence of BP in Africa. Lu et al. reported a lower prevalence of BP in Africa compared to Europe, which contradicts our hypothesis.¹⁸ A possible reason for this discrepancy could be underdiagnosis of BP in Africa. This discrepancy may also reflect the fact that BP is a multifactorial disease and other environmental, lifestyle and genetic factors may be involved. We estimate that the under-reporting of H. pylori infection in BP is higher than expected. Larger studies are required to confirm the association between BP and H. pylori infection.

In conclusion, the temporal relationship between eradication of H. pylori and stabilization of BP is an important observation. Both cases showed long-term remission, so that no long-term BP therapy was required. These both cases show that H. pylori infection should be discussed as a trigger factor for BP. Further observational studies and case series are necessary in this area.

None.