- Email: [email protected]
Eradication of Helicobacter pylori and improvement of hereditary angioneurotic oedema
RESEARCH LETTERS
Research letters
Eradication of Helicobacter pylori and improvement of hereditary angioneurotic oedema Henriette Farkas, George Füst, Béla Fekete, István Karádi, Lilian Varga Helicobacter pylori infection is thought to be a causal factor in various dermatological disorders. We assessed the frequency of H pylori infection in 65 patients with hereditary angioneurotic oedema. We measured the serum concentration of antibodies against H pylori and did the carbon-14-urease breath test in patients with positive H pylori serology. 19 of 65 patients had H pylori infection. All patients with infection, and 11 of 46 without infection, had a history of recurrent episodes of acute abdominal pain. We successfully eradicated H pylori infection in 18 patients. The frequency of abdominal symptoms was significantly higher in the infected group (p=0·002 after adjustment for age). In nine of 19 patients with dyspepsia, the frequency of oedematous episodes decreased from 100 over 10 months before eradication to 19 during the 10-month follow-up period. Screening for, and eradication of, H pylori infection seems to be justified in patients with hereditary angioneurotic oedema.
Lancet 2001; 358: 1695–96
Hereditary angioneurotic oedema results from deficiency of C1-esterase inhibitor. This autosomal dominant disorder is characterised by recurrent oedema in the subcutis, submucosa, or both. Oedema in the submucosa can have life-threatening consequences. Laryngeal oedema can cause suffocation, and oedematous swelling of the intestinal wall causes acute abdominal pain.
Infected with H pylori CagA-positive CagA-negative Not infected with H pylori
Number of patients (male)
Age (years, mean [SE])
19 (8) 16 (8) 3 (0) 46 (20)
38·4 (4·1) 38·7 (4·0) 36·7 (17·2) 26·3 (2·0)
Oedematous episodes can be triggered by various factors, such as mechanical trauma, mental stress, surgery, hormonal effects, drugs (eg, angiotensin-converting-enzyme inhibitors, oral contraceptives), and infection.1 Wedi and colleagues2 suggested that Helicobacter pylori infection could be a causal factor for various dermatological disorders, such as rosacea, urticaria, angioneurotic oedema, and atopic dermatitis. In 1989, Collen and colleagues3 concluded, “basal acid output and upper gastrointestinal endoscopy are important determinants when evaluating abdominal pain in patients with hereditary angioedema that fails to respond to standard therapy.” Rais and colleagues4 suggested that “H pylori infection may be an important factor in refractory exacerbations of [hereditary angioneurotic oedema]”. We have reported a case of acquired C1-esterase inhibitor deficiency associated with H pylori infection, in which oedematous episodes completely disappeared after eradication.5 We designed a screening programme to assess the frequency of H pylori infection in patients with hereditary angioneurotic oedema and to explore any relation between this infection and oedematous attacks. We also looked at the effect of infection on clinical manifestations and laboratory abnormalities associated with hereditary angioneurotic oedema. The ethics committee of Semmelweis University approved the study, and all patients gave informed consent. Lifetime episodes of acute abdominal pain* None
1–5
6–10
11–30
31–50
0 0 0 35
3 2 1 8
1 0 1 1
5 4 1 2
3 3 0 0
>50 7 7 0 0
*H pylori infected versus non-infected. p<0·0001.
Table 1: Episodes of acute abdominal pain in patients with hereditary angioneurotic oedema Pretreatment observation
Pre-treatment oedematous episodes
Post-eradication oedematous episodes
GI
SC
L
Total
GI
SC
L
Total
12 8 4 11 3 10 6 8 7
0 0 1 0 1 0 1 0 0
13 11 6 14 6 15 13 14 8
0 0 0 1 0 0 0 0 0
1 2 1 0 2 3 2 4 2
0 0 0 0 1 0 0 0 0
1 2 1 1 3 3 2 4 2
10 10 8 14 9 25 11 13 9
1·00 0 0·05 0·10 0·10 0·10 0·15 0·05 0
13 (7–14)
0 (0–0)*
2 (1–2·5)*
0 (0–0)
2 (1–3)*
10 (9–13·5)
··
0·97 (0·26)
0·01 (0·02)
0·17 (0·09)
··
0·19 (0·09)
Patient number 1 2 3 4 5 6 7 8 9
10 9 8 13 9 25 11 13 9
1 3 1 3 2 5 6 6 1
Median (IQR)
10 (9–13)
3 (1–5·5)
7 0 (5–10·5) (0–1)
0·26 (0·16)
0·66 (0·28)
Frequency ·· of attacks (months, mean [SD])
··
Post-eradication Reduction follow-up in drug dose (months) (g/day)
··
··
All patients were treated with danazol, except patient 1, who received tranexamic acid. GI=gastrointestinal; SC=subcutaneous; L=laryngeal. *p=0·0039 compared with the frequency of attacks before eradication, by Wilcoxon matched pairs test.
Table 2: The effect of eradication on frequency of clinical manifestations and drug dose
THE LANCET • Vol 358 • November 17, 2001
1695
For personal use. Only reproduce with permission from The Lancet Publishing Group.
RESEARCH LETTERS
We measured the serum concentration of IgG antibodies against H pylori in 65 patients (29 men, 36 women) with hereditary angioneurotic oedema. All patients are followed up at our institution, which is the only centre for hereditary angioneurotic oedema in Hungary. We did the carbon-14urease breath test on every patient with positive H pylori serology. 19 patients had a positive 14C-urease breath test, and these patients were judged to have H pylori infection. Furthermore, 16 of 19 patients were positive for CagA, the cytotoxic protein of H pylori. All patients infected with H pylori, but only 11 of 46 not infected, had a history of acute abdominal pain (table 1). The frequency of acute abdominal pain was substantially higher in patients with H pylori infection than in those not infected. 15 of 19 patients with infection, but only two of 46 patients not infected, had more than ten occurrences of acute abdominal pain before eradication of H pylori. Patients infected with H pylori were significantly older than non-infected patients; the difference in frequency of acute abdominal pain between infected and non-infected patients was significant (p=0·002) after adjustment for age by multiple logistic regression analysis. Episodes of acute abdominal pain were not very frequent in the three patients infected with CagA-negative H pylori (table 1). H pylori eradication was done in 18 of 19 infected patients. Treatment failed in one patient because of poor compliance. A 1-week course of triple combination therapy (amoxicillin 1000 mg, clarithromycin 500 mg, and omeprazole 20 mg twice daily) eradicated H pylori infection in all 18 patients, and eradication was confirmed by a negative 14C-urease breath test. Episodes of acute abdominal pain occurred in 16 of 18 patients over the year before eradication, whereas such symptoms were seen in only one patient during follow-up of 2–25 months (median 8 months) after eradication. Nine patients were selected for detailed analysis because they had dyspepsia, and because continuous drug prophylaxis with danazol or tranexamic acid failed to prevent angioedemic episodes. Endoscopy, done in four patients with the most severe symptoms, showed chronic antral gastritis. Episodes of acute abdominal pain in these nine patients decreased strikingly, from 28 to one after successful eradication, and the frequency of other oedematous manifestations also decreased (table 2). Furthermore, patients’ dyspeptic symptoms completely disappeared. Drug doses for long-term prophylaxis were reduced in seven patients (table 2). After successful eradication of H pylori there was significant elevation of C4 (p=0·016) and C1-esterase inhibitor (p=0·008) concentrations. We postulate that exacerbations of acute hereditary angioneurotic oedema might be triggered by activation of humoral immune responses against H pylori, leading to depletion of C1-esterase inhibitor. This process might result in the uncontrolled activation of complement and other plasma enzyme systems. Elimination of potential triggering factors is the cornerstone of treatment for hereditary angioneurotic oedema; this measure can reduce frequency of oedematous episodes and improve patients’ quality of life. Screening for H pylori infection and eradication of the pathogen seems justified in patients with hereditary angioneurotic oedema. 1
2
3
4
Agostoni A, Cicardi M. Hereditary and acquired C1 inhibitor deficiency: biological and clinical characteristics in 235 patients. Medicine 1992; 71: 206–15. Wedi B, Wagner S, Werfel T, Manns MP, Kapp A. Prevalence of Helicobacter pylori-associated gastritis in chronic urticaria. Int Arch Allergy Immunol 1998; 116: 288–94. Collen MJ, Lewis JH, Deschner WK, et al. Abdominal pain in hereditary angioedema: the role of acid hypersecretion. Am J Gastroenterol 1989; 84: 873–77. Rais M, Unzeitig J, Grant JA. Refractory exacerbations of hereditary angioedema with associated Helicobacter pylori infection. J Allergy Clin Immunol 1999; 103: 713–14.
1696
5
Farkas H, Gyeney L, Majthényi P, Füst G, Varga L. Angioedema due to acquired C1-esterase inhibitor deficiency in a patient with Helicobacter pylori infection. Z Gastroenterol 1999; 37: 513–18.
Semmelweis University, Kútvölgyi Clinical Centre, Budapest, Hungary (H Farkas MD, Prof I Karádi MD); Semmelweis University, 3rd Department of Internal Medicine, Budapest (Prof G Füst MD, Prof B Fekete MD); Research Group of Metabolism, Genetics, and Immunology, Hungarian Academy of Sciences, Budapest (Prof G Füst); and National Institute of Haematology and Immunology, Complement Laboratory, Budapest (L Varga PhD) Correspondence to: Dr Henriette Farkas, Semmelweis University, Kútvölgyi Clinical Centre, Allergology and Angioedema Outpatient Clinic, H-1125 Budapest, Kútvölgyi út 4, Hungary (e-mail: [email protected])
Analysis of hereditary component of cancer by use of a familial index by site Bernt Lindelöf, Gunnar Eklund To quantify hereditary factors in the risk of cancer, we matched 1 283 047 cancer patients listed in the Swedish Cancer Registry with healthy controls from a national database, and identified the number of individuals who were first-degree relatives of other indivduals in the same cohort. Division of the number of relatives in the patient cohort by the number in the control cohort yielded a “familial index”. The following cancers had high familial indices: eye 16·5 (95% CI 2·5–666), testis 9·0 (3·2–35), Hodgkin’s disease 6·5 (2·3–26·0), and thyroid 6·2 (3·7–12). Overall, however, familial factors made only a minor contribution to susceptibility to cancer.
Lancet 2001; 358: 1696–98
Clustering of cancers within families can be due to chance alone, to genetic factors, or to exposure to common environmental factors. The most common method of identifying genetic factors1–5 is the clinical detection of probands and other affected individuals in multigeneration pedigrees.1 The disadvantages of this method are the paucity of cases and the risk of bias. Twin studies make it possible to estimate the overall contribution of inherited genes, but because of the rareness of twinning, even large twin registers allow estimation of the risk of only the commonest cancers.2 A third method involves calculation of the ratio between observed and expected cases. This technique can be used in region-wide, population-based studies with access to linkage between cancer cases and family relations, and its advantage is the large numbers of individuals studied. The disadvantages are that genetic and environmental factors are difficult to separate, and few geographic areas have the infrastructure necessary for this type of study. However, some large-scale studies of many cancer sites have been done in the Mormon population of Utah, USA,3 and in Scandinavia.4 We have now done a nationwide study in Sweden using data from the obligatory national cancer registry (1958–97). We classified patients on this register by cancer site (using criteria from revision seven of the International Classification of Diseases [ICD-7]), and matched each cancer case by year of birth, sex, and year of death with one person without cancer (control). Patients with bone marrow cancers were further subdivided by histological subtype. Living controls were identified from the Total Population Registry maintained by Statistics Sweden, and the dead controls were identified from the Swedish Causes of Death Registry maintained by the National Board of Health and Welfare, Stockholm. By using a unique personal identification number, we linked the cancer cohorts and the control cohorts with the Total Population
THE LANCET • Vol 358 • November 17, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.
Research letters
Eradication of Helicobacter pylori and improvement of hereditary angioneurotic oedema Henriette Farkas, George Füst, Béla Fekete, István Karádi, Lilian Varga Helicobacter pylori infection is thought to be a causal factor in various dermatological disorders. We assessed the frequency of H pylori infection in 65 patients with hereditary angioneurotic oedema. We measured the serum concentration of antibodies against H pylori and did the carbon-14-urease breath test in patients with positive H pylori serology. 19 of 65 patients had H pylori infection. All patients with infection, and 11 of 46 without infection, had a history of recurrent episodes of acute abdominal pain. We successfully eradicated H pylori infection in 18 patients. The frequency of abdominal symptoms was significantly higher in the infected group (p=0·002 after adjustment for age). In nine of 19 patients with dyspepsia, the frequency of oedematous episodes decreased from 100 over 10 months before eradication to 19 during the 10-month follow-up period. Screening for, and eradication of, H pylori infection seems to be justified in patients with hereditary angioneurotic oedema.
Lancet 2001; 358: 1695–96
Hereditary angioneurotic oedema results from deficiency of C1-esterase inhibitor. This autosomal dominant disorder is characterised by recurrent oedema in the subcutis, submucosa, or both. Oedema in the submucosa can have life-threatening consequences. Laryngeal oedema can cause suffocation, and oedematous swelling of the intestinal wall causes acute abdominal pain.
Infected with H pylori CagA-positive CagA-negative Not infected with H pylori
Number of patients (male)
Age (years, mean [SE])
19 (8) 16 (8) 3 (0) 46 (20)
38·4 (4·1) 38·7 (4·0) 36·7 (17·2) 26·3 (2·0)
Oedematous episodes can be triggered by various factors, such as mechanical trauma, mental stress, surgery, hormonal effects, drugs (eg, angiotensin-converting-enzyme inhibitors, oral contraceptives), and infection.1 Wedi and colleagues2 suggested that Helicobacter pylori infection could be a causal factor for various dermatological disorders, such as rosacea, urticaria, angioneurotic oedema, and atopic dermatitis. In 1989, Collen and colleagues3 concluded, “basal acid output and upper gastrointestinal endoscopy are important determinants when evaluating abdominal pain in patients with hereditary angioedema that fails to respond to standard therapy.” Rais and colleagues4 suggested that “H pylori infection may be an important factor in refractory exacerbations of [hereditary angioneurotic oedema]”. We have reported a case of acquired C1-esterase inhibitor deficiency associated with H pylori infection, in which oedematous episodes completely disappeared after eradication.5 We designed a screening programme to assess the frequency of H pylori infection in patients with hereditary angioneurotic oedema and to explore any relation between this infection and oedematous attacks. We also looked at the effect of infection on clinical manifestations and laboratory abnormalities associated with hereditary angioneurotic oedema. The ethics committee of Semmelweis University approved the study, and all patients gave informed consent. Lifetime episodes of acute abdominal pain* None
1–5
6–10
11–30
31–50
0 0 0 35
3 2 1 8
1 0 1 1
5 4 1 2
3 3 0 0
>50 7 7 0 0
*H pylori infected versus non-infected. p<0·0001.
Table 1: Episodes of acute abdominal pain in patients with hereditary angioneurotic oedema Pretreatment observation
Pre-treatment oedematous episodes
Post-eradication oedematous episodes
GI
SC
L
Total
GI
SC
L
Total
12 8 4 11 3 10 6 8 7
0 0 1 0 1 0 1 0 0
13 11 6 14 6 15 13 14 8
0 0 0 1 0 0 0 0 0
1 2 1 0 2 3 2 4 2
0 0 0 0 1 0 0 0 0
1 2 1 1 3 3 2 4 2
10 10 8 14 9 25 11 13 9
1·00 0 0·05 0·10 0·10 0·10 0·15 0·05 0
13 (7–14)
0 (0–0)*
2 (1–2·5)*
0 (0–0)
2 (1–3)*
10 (9–13·5)
··
0·97 (0·26)
0·01 (0·02)
0·17 (0·09)
··
0·19 (0·09)
Patient number 1 2 3 4 5 6 7 8 9
10 9 8 13 9 25 11 13 9
1 3 1 3 2 5 6 6 1
Median (IQR)
10 (9–13)
3 (1–5·5)
7 0 (5–10·5) (0–1)
0·26 (0·16)
0·66 (0·28)
Frequency ·· of attacks (months, mean [SD])
··
Post-eradication Reduction follow-up in drug dose (months) (g/day)
··
··
All patients were treated with danazol, except patient 1, who received tranexamic acid. GI=gastrointestinal; SC=subcutaneous; L=laryngeal. *p=0·0039 compared with the frequency of attacks before eradication, by Wilcoxon matched pairs test.
Table 2: The effect of eradication on frequency of clinical manifestations and drug dose
THE LANCET • Vol 358 • November 17, 2001
1695
For personal use. Only reproduce with permission from The Lancet Publishing Group.
RESEARCH LETTERS
We measured the serum concentration of IgG antibodies against H pylori in 65 patients (29 men, 36 women) with hereditary angioneurotic oedema. All patients are followed up at our institution, which is the only centre for hereditary angioneurotic oedema in Hungary. We did the carbon-14urease breath test on every patient with positive H pylori serology. 19 patients had a positive 14C-urease breath test, and these patients were judged to have H pylori infection. Furthermore, 16 of 19 patients were positive for CagA, the cytotoxic protein of H pylori. All patients infected with H pylori, but only 11 of 46 not infected, had a history of acute abdominal pain (table 1). The frequency of acute abdominal pain was substantially higher in patients with H pylori infection than in those not infected. 15 of 19 patients with infection, but only two of 46 patients not infected, had more than ten occurrences of acute abdominal pain before eradication of H pylori. Patients infected with H pylori were significantly older than non-infected patients; the difference in frequency of acute abdominal pain between infected and non-infected patients was significant (p=0·002) after adjustment for age by multiple logistic regression analysis. Episodes of acute abdominal pain were not very frequent in the three patients infected with CagA-negative H pylori (table 1). H pylori eradication was done in 18 of 19 infected patients. Treatment failed in one patient because of poor compliance. A 1-week course of triple combination therapy (amoxicillin 1000 mg, clarithromycin 500 mg, and omeprazole 20 mg twice daily) eradicated H pylori infection in all 18 patients, and eradication was confirmed by a negative 14C-urease breath test. Episodes of acute abdominal pain occurred in 16 of 18 patients over the year before eradication, whereas such symptoms were seen in only one patient during follow-up of 2–25 months (median 8 months) after eradication. Nine patients were selected for detailed analysis because they had dyspepsia, and because continuous drug prophylaxis with danazol or tranexamic acid failed to prevent angioedemic episodes. Endoscopy, done in four patients with the most severe symptoms, showed chronic antral gastritis. Episodes of acute abdominal pain in these nine patients decreased strikingly, from 28 to one after successful eradication, and the frequency of other oedematous manifestations also decreased (table 2). Furthermore, patients’ dyspeptic symptoms completely disappeared. Drug doses for long-term prophylaxis were reduced in seven patients (table 2). After successful eradication of H pylori there was significant elevation of C4 (p=0·016) and C1-esterase inhibitor (p=0·008) concentrations. We postulate that exacerbations of acute hereditary angioneurotic oedema might be triggered by activation of humoral immune responses against H pylori, leading to depletion of C1-esterase inhibitor. This process might result in the uncontrolled activation of complement and other plasma enzyme systems. Elimination of potential triggering factors is the cornerstone of treatment for hereditary angioneurotic oedema; this measure can reduce frequency of oedematous episodes and improve patients’ quality of life. Screening for H pylori infection and eradication of the pathogen seems justified in patients with hereditary angioneurotic oedema. 1
2
3
4
Agostoni A, Cicardi M. Hereditary and acquired C1 inhibitor deficiency: biological and clinical characteristics in 235 patients. Medicine 1992; 71: 206–15. Wedi B, Wagner S, Werfel T, Manns MP, Kapp A. Prevalence of Helicobacter pylori-associated gastritis in chronic urticaria. Int Arch Allergy Immunol 1998; 116: 288–94. Collen MJ, Lewis JH, Deschner WK, et al. Abdominal pain in hereditary angioedema: the role of acid hypersecretion. Am J Gastroenterol 1989; 84: 873–77. Rais M, Unzeitig J, Grant JA. Refractory exacerbations of hereditary angioedema with associated Helicobacter pylori infection. J Allergy Clin Immunol 1999; 103: 713–14.
1696
5
Farkas H, Gyeney L, Majthényi P, Füst G, Varga L. Angioedema due to acquired C1-esterase inhibitor deficiency in a patient with Helicobacter pylori infection. Z Gastroenterol 1999; 37: 513–18.
Semmelweis University, Kútvölgyi Clinical Centre, Budapest, Hungary (H Farkas MD, Prof I Karádi MD); Semmelweis University, 3rd Department of Internal Medicine, Budapest (Prof G Füst MD, Prof B Fekete MD); Research Group of Metabolism, Genetics, and Immunology, Hungarian Academy of Sciences, Budapest (Prof G Füst); and National Institute of Haematology and Immunology, Complement Laboratory, Budapest (L Varga PhD) Correspondence to: Dr Henriette Farkas, Semmelweis University, Kútvölgyi Clinical Centre, Allergology and Angioedema Outpatient Clinic, H-1125 Budapest, Kútvölgyi út 4, Hungary (e-mail: [email protected])
Analysis of hereditary component of cancer by use of a familial index by site Bernt Lindelöf, Gunnar Eklund To quantify hereditary factors in the risk of cancer, we matched 1 283 047 cancer patients listed in the Swedish Cancer Registry with healthy controls from a national database, and identified the number of individuals who were first-degree relatives of other indivduals in the same cohort. Division of the number of relatives in the patient cohort by the number in the control cohort yielded a “familial index”. The following cancers had high familial indices: eye 16·5 (95% CI 2·5–666), testis 9·0 (3·2–35), Hodgkin’s disease 6·5 (2·3–26·0), and thyroid 6·2 (3·7–12). Overall, however, familial factors made only a minor contribution to susceptibility to cancer.
Lancet 2001; 358: 1696–98
Clustering of cancers within families can be due to chance alone, to genetic factors, or to exposure to common environmental factors. The most common method of identifying genetic factors1–5 is the clinical detection of probands and other affected individuals in multigeneration pedigrees.1 The disadvantages of this method are the paucity of cases and the risk of bias. Twin studies make it possible to estimate the overall contribution of inherited genes, but because of the rareness of twinning, even large twin registers allow estimation of the risk of only the commonest cancers.2 A third method involves calculation of the ratio between observed and expected cases. This technique can be used in region-wide, population-based studies with access to linkage between cancer cases and family relations, and its advantage is the large numbers of individuals studied. The disadvantages are that genetic and environmental factors are difficult to separate, and few geographic areas have the infrastructure necessary for this type of study. However, some large-scale studies of many cancer sites have been done in the Mormon population of Utah, USA,3 and in Scandinavia.4 We have now done a nationwide study in Sweden using data from the obligatory national cancer registry (1958–97). We classified patients on this register by cancer site (using criteria from revision seven of the International Classification of Diseases [ICD-7]), and matched each cancer case by year of birth, sex, and year of death with one person without cancer (control). Patients with bone marrow cancers were further subdivided by histological subtype. Living controls were identified from the Total Population Registry maintained by Statistics Sweden, and the dead controls were identified from the Swedish Causes of Death Registry maintained by the National Board of Health and Welfare, Stockholm. By using a unique personal identification number, we linked the cancer cohorts and the control cohorts with the Total Population
THE LANCET • Vol 358 • November 17, 2001
For personal use. Only reproduce with permission from The Lancet Publishing Group.