Use this URL to cite or link to this record in EThOS:
Title: The inter-relationship between alcohol consumption and liver disease on the prevalence and disease manifestations of Helicobacter pylori : a clinical, histological and in vitro study
Author: Marshall, Jonathan Charles Walter
Awarding Body: University of London
Current Institution: University College London (University of London)
Date of Award: 2002
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Peptic ulceration of the upper gastrointestinal tract, typically secondary to infection with H.pylori, is a significant cause of morbidity and mortality in the general population. Patients with liver disease have a higher prevalence of peptic ulceration than that observed in the normal population, yet the association between it, and H.pylori infection, is much less clearly defined. In chapter 1, following a discussion about the clinical significance of peptic ulcer disease, the metabolism of alcohol and its effects on the upper gastrointestinal tract are reviewed. Then the discovery, epidemiology, diagnostic techniques and effects of H.pylori on the upper gastrointestinal tract are outlined. The role of H.pylori in gastric alcohol metabolism introduces a review of how alcohol and H.pylori might interact to cause upper gastrointestinal pathology in man. The current literature in relation to H.pylori prevalence in liver disease in general, and alcohol related liver disease in particular, is conflicting. Studies to date generally fail because of small numbers, inadequate demographic data, inadequate or inappropriate control comparisons and failure to accurately characterize the level of liver injury or drinking history. A detailed review of the current state of knowledge in this area is presented which attempts to analyze the prevalence of H.pylori after controlling for the important variables of degree of liver injury, alcohol consumption and demography, where possible. Studies suggesting that, in the otherwise normal population, the prevalence of H.pylori declines with increasing alcohol consumption and that this effect might be partially beverage dependent are discussed. In patients with liver disease in general, and cirrhosis in particular, no firm conclusion about the prevalence of H.pylori in these patient populations can be drawn and the reasons for this are discussed in detail. The presented literature review highlights areas for further work concerning H.pylori prevalence in patients misusing alcohol both with and without significant liver disease. The need to carefully control for extent of liver injury, demography and alcohol consumption is emphasized. Chapter 2 examines the seroprevalence of H.pylori in four carefully defined populations of healthy volunteers (n=l42), hospital inpatients (n=55) and patients with non-alcoholic (n=l 51 ) and alcoholic liver disease (n=l 76). The overall H.pylori seroprevalence in the four groups was healthy volunteers ( 18.3%), hospital patients (47.3%), non-alcoholic liver disease (40.4%) and alcohol misusers (35.2%). The lower overall seroprevalence observed in healthy volunteers was due to this group being significantly younger than the three patient groups and this apparent effect was lost following correction for age. No significant differences existed in overall H.pylori seroprevalence between the three patient groups. In healthy volunteers, the seroprevalence of H.pylori increased by 7%/year of increasing age and in hospital patients by 11 %/year of increasing age. This age-related increase in H.pylori seroprevalence is in broad agreement with other published work. A unique finding was however a 5%/year of increasing age decline in H.pylori seroprevalence in both liver injury groups. In minimal injury, non-alcoholic liver disease, the H.pylori seroprevalence increased by 3%/year of increasing age. But, in cirrhotic non-alcoholic liver disease, the seroprevalence declined by approximately 20%/year of increasing age. The net effect being the 5%/year of increasing age decline in H.pylori seroprevalence observed overall for non-alcoholic liver disease. In alcohol related minimal liver injury, H.pylori seroprevalence declined by 10%/year of increasing age. But, in significant alcohol related liver injury, the seroprevalence remained constant for all decades. Again, combining these two seroprevalence rates for H.pylori gave the overall 5%/year of increasing age decline in seroprevalence in the alcohol related injury group. Putative explanations for the observed differences in age related seroprevalence between alcoholic and non-alcoholic liver injury are discussed including differences in mechanism of liver injury and alcohol consumption. Chapter 3 prospectively assesses the prevalence of H.pylori in 100 consecutively recruited alcohol misusers: These individuals were carefully assessed in terms of extent of liver injury (92 had diagnostic liver biopsy), drinking history, recent drug ingestion and socioeconomic circumstances. Daily median (range) ethanol consumption was 196 (48-800) g ethanol/day, with a lifetime dose of 1138 (40-7884) Kg ethanol. Men consumed significantly higher daily ethanol than women (224 vs 120 g ethanol/day: p<0.001). But no significant difference between men and women was observed for age of first drink, onset of alcohol misuse or years of misuse. 80 patients had minimal injury (normal or fatty change, 72 based on biopsy) and 20 significant liver injury (alcoholic hepatitis and/or cirrhosis-all on biopsy). The overall H.pylori prevalence (sensitivity; specificity) by diagnostic modality was serology 43.6% (64.4%; 75.5%), UBT 66.7% (92.9%; 90.0%), urease 25.5% (52.2%; 98.2%) and histology 33.0% (71.7%; 100.0%). The ‘gold-standard’ used for determining sensitivity and specificity was either histology positive or histology negative but at least two non-histological tests positive. Using this gold-standard, the overall H.pylori prevalence was 46.0%. No significant association between H.pylori status and smoking, caffeine consumption, recent drug ingestion in the previous three months or level of alcohol misuse could be determined. Similarly, no significant difference in H.pylori prevalence could be detected using any diagnostic modality between significant and minimal liver injury groups (50.0% vs 45.0% respectively, with gold-standard) The H.pylori prevalence by age, using the gold-standard was 56.3% (<40yr), 40.5% (41- 50yr), 52.2% (51-60yr) and 66.6% (>61yr). These differences between decades were not significant. It is the lack of an apparent age related increase in H.pylori prevalence that emphasizes the unique features of this patient population. Broadly similar results were obtained with other diagnostic modalities except urease testing. Here, H.pylori prevalence was significantly lower at the extremes of age when compared with the gold standard (21.0% vs 56.3% (<40yr: p=0.03); 31.3% vs 66.6% (>61yr: p=0.04)). It is hypothesized that the apparently lower H.pylori prevalence obtained with urease testing might be accounted for by a direct alcohol toxicity effect on H.pylori itself. This would be analogous to the false negative urease tests produced following recent PPI or antibiotic ingestion. Chapter 4 examines the relationship between symptoms, endoscopic findings, H.pylori status, alcohol consumption, and gastroduodenal pathology in the 100 prospectively assessed alcohol misusers: 36% had no dyspeptic symptoms but 24% had symptoms scored as ‘severe.’ No significant relationship existed between dyspepsia scores and daily alcohol consumption, period of abstinence, any drug usage (PPI, HI blockers, antibiotics or other), current smoking or H.pylori status. Similarly, following initial classification of the 100 alcohol misusers by H.pylori status (33 H.pylori +) no relationship between either presence or severity of symptoms and any other risk factors for dyspepsia could be determined. Of the 100 alcohol misusers, 60% had ‘any abnormality’ at endoscopy. Within the sub-set of patients with dyspepsia, 59.4% had an abnormality at endoscopy and 61.1% of the asymptomatic patients also had an endoscopic abnormality. 58.8% of the H.pylori positive alcohol misusers had dyspepsia (66.7% H.pylori negative). Exactly half had an abnormality at endoscopy (63.6% H.pylori negative) and in the 41.2% of the H.pylori positive patients who were asymptomatic (33.3% H.pylori negative), 46.1% had an endoscopic abnormality (68.2% H.pylori negative). None of these differences either within, or between groups was significant. Histological changes in the gastric mucosa of the 100 alcohol misusers were classified according to the Sydney system as either antral (A) or body (B): normal (13%A; 22%B), H.pylori associated gastritis (31%A; 29%B), acute gastritis (3%A; 3%B), reactive gastritis (7%; 3%B), intestinal metaplasia/atrophy (4%A; 7%B). The commonest histological abnormality was chronic gastritis (41% A; 35%B). None of these differences of histological abnormalities between sites was significant. The mean gastric inflammatory scores for the 100 alcohol misusers was 1.07 and significantly higher scores were observed in H.pylori positive alcohol misusers when compared with negative misusers (Hp+; 2.37, Hp-; 0.411 : p<0.0001). Comparison between 50 age and sex matched pairs of alcohol misusers and controls were undertaken: A non-significantly lower prevalence of H.pylori was found in the antrum and body of the alcohol misusers in comparison with matched controls. A significantly lower prevalence of chemical gastritis and antral intestinal metaplasia was observed in the alcohol misusers (6% vs 22%: p=0.021). Explanations for this finding, apparently in conflict with the Sydney system predictions, are discussed.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available