The American Journal of Medicine
Volume 116, Issue 9 , Pages 601-605, 1 May 2004

Effect of helicobacter pylori infection on the risk of upper gastrointestinal bleeding in users of nonsteroidal anti-inflammatory drugs☆

  • George V Papatheodoridis, MD

      Affiliations

    • Academic Department of Medicine, Henry Dunant Hospital, Athens, Greece
    • Corresponding Author InformationRequests for reprints should be addressed to George V. Papatheodoridis, MD, Academic Department of Medicine, Hippokration General Hospital of Athens, 114 Vas. Sophias Avenue, 115 27 Athens, Greece
    • ,
  • Despina Papadelli, MD

      Affiliations

    • Academic Department of Medicine, Henry Dunant Hospital, Athens, Greece
    • Despina Papadelli received a scholarship from the Hellenic Gastroenterology Association.
    • ,
  • Evangelos Cholongitas, MD

      Affiliations

    • Academic Department of Medicine, Henry Dunant Hospital, Athens, Greece
    • ,
  • Dimitrios Vassilopoulos, MD

      Affiliations

    • Academic Department of Medicine, Henry Dunant Hospital, Athens, Greece
    • ,
  • Andreas Mentis, MD

      Affiliations

    • Hippokration General Hospital of Athens; Hellenic Pasteur Institute, Henry Dunant Hospital, Athens, Greece
    • ,
  • Stephanos J Hadziyannis, MD

      Affiliations

    • Department of Medicine and Hepatology, Henry Dunant Hospital, Athens, Greece

Received 25 February 2003; received in revised form 18 September 2003; accepted 18 September 2003.

Article Outline

Abstract 

Purpose

We evaluated whether infection with Helicobacter pylori, including specific cytotoxic-associated antigen (CagA)-positive strains, increase the risk of upper gastrointestinal bleeding in users of nonsteroidal anti-inflammatory drugs (NSAIDs).

Methods

Cases with upper gastrointestinal bleeding and recent NSAID use, including aspirin, who were admitted during 2001, were compared with age- and sex-matched outpatient controls who had recent NSAID use. H. pylori infection was diagnosed by serum antibodies or the 13C-urea breath test; and CagA seropositivity was diagnosed by enzyme-linked immunoassay.

Results

H. pylori was detected significantly more frequently in cases of bleeding than controls (79% [63/80] vs. 56% [45/80], P = 0.004). Cases of bleeding were more likely than controls to have a history of peptic ulcer (34% [n = 27] vs. 13% [n = 10], P = 0.003), previous upper gastrointestinal bleeding (19% [n = 15] vs. 6% [n = 5], P = 0.03), recent dyspepsia (29% [n = 23] vs. 15% [n = 12], P = 0.06), and <3 months of NSAID use (58% [n = 46] vs. 40% [n = 32], P = 0.04). CagA positivity was not associated with gastrointestinal bleeding. In a multivariate analysis, H. pylori infection was the only significant risk factor for upper gastrointestinal bleeding (odds ratio = 1.7; 95% confidence interval: 1.2 to 2.5; P = 0.004).

Conclusion

H. pylori infection almost doubles the risk of upper gastrointestinal bleeding among users of NSAIDs.

 

Gastrointestinal complications, particularly peptic ulcer and erosions, are the most common adverse events among users of nonsteroidal anti-inflammatory drugs (NSAIDs) 1, 2, 3, whereas Helicobacter pylori infection is the most frequent cause of peptic ulcer disease in patients not taking NSAIDs (4). However, the effects of the simultaneous presence of both H. pylori infection and NSAID use on the risks of peptic ulcer and its complications are uncertain 5, 6, 7, 8, 9, 10, 11, 12, 13.

Although infection with H. pylori causes gastric inflammation in all persons who are infected with the organism (4), the host response and the clinical manifestations of infection appear to be associated with genetic differences in the bacterial strain 4, 14, 15. For example, the risk of development of peptic ulcer disease is higher in persons infected with cytotoxic-associated antigen (CagA)-positive strains than with CagA-negative strains (16). (CagA is a bacterial antigen, to which antibodies develop.)

We sought to assess the effect of infection with H. pylori, and CagA status, on the risk of upper gastrointestinal bleeding in users of NSAIDs.

Back to Article Outline

Methods 

All patients with acute upper gastrointestinal bleeding (cases of bleeding) admitted to our department between January and December 2001 were included in this prospective, case-control study if they were current users of NSAIDs, including aspirin (defined as use of at least one dose within the last 7 days before the episode of bleeding), and had endoscopically verified gastric or duodenal ulcer or hemorrhagic/erosive gastritis as the most probable source of bleeding. All endoscopies were performed within 24 hours after admission. Controls were recruited from the outpatient medicine clinics. They were current (within the last 7 days) users of NSAIDs, including aspirin, without signs of gastrointestinal bleeding. Controls were matched to patients at a 1:1 ratio by sex and by age in 5-year bands (17). All subjects gave informed consent to participate in the study. There were no exclusion criteria, but 3 NSAID users admitted with upper gastrointestinal bleeding were excluded: 2 with uncertain sources of bleeding and 1 who died within a few hours of admission.

A structured questionnaire was completed for all cases and controls. The duration of NSAID/aspirin use was considered to be short if the drug had been taken for ≤3 months. NSAID dose was expressed as the number of defined daily doses (18) consumed within the last 7 days. For example, one defined daily dose of aspirin was considered to be 150 mg, whereas low-dose aspirin was considered to be the consumption of seven or fewer defined daily doses (1050 mg within the last 7 days).

H. pylori testing 

H. pylori infection was diagnosed by serum immunoglobulin (Ig) G antibodies or the 13C-urea breath test. IgG antibodies against H. pylori were measured in duplicate by an in-house enzyme immunoassay (19) using rabbit anti-human IgG (IDEIA Helicobacter pylori IgG; Dako, Copenhagen, Denmark). Antibodies against H. pylori were determined in cases on the day after admission, and in controls on the day of completion of the study questionnaire. After at least an 8-hour fasting period, 13C-urea breath tests were performed after the patients drank 75 mg of 13C-labeled urea dissolved in 50 mL of water and 200 mL of orange juice. Duplicate breath samples were taken before and 30 minutes after the ingestion of urea and analyzed by infrared spectroscopy. The breath test was considered to be positive for H. pylori if the increase in the 13C/12C ratio was >4%. A 13C-urea breath test was performed in 65 of the cases as soon as oral intake was allowed, as well as in 51 controls. H. pylori status was considered to be positive if either the serological test or the urea breath test was positive.

CagA status was determined by serum antibodies against CagA protein using a commercially available enzyme immunoassay (Helicobacter pylori CagA IgG EIA WELL; Radim, Liege, Belgium) according to the manufacturer's instructions.

Statistical analysis 

All data were analyzed using SPSS, version 10.1 (SPSS Inc., Chicago, Illinois). Proportions were compared with the chi-squared test or Fisher exact test. Continuous variables were compared with the Student t test. Multivariate analysis was performed using logistic regression models. Variables associated with the dependent variable (at P ≤0.10) were entered in the multivariate models. A two-tailed P value <0.05 was considered to be significant.

Back to Article Outline

Results 

We enrolled 80 cases of bleeding and 80 controls. Age, sex, and mean dose of NSAID used were similar between cases and controls (Table 1), but cases were more likely to have used NSAIDs for <3 months (46 [58%] cases vs. 32 [40%] controls, P = 0.04). Aspirin was the most common NSAID used within the last 7 days in both cases and controls (Table 2).

Table 1. Characteristics of Cases of Gastrointestinal Bleeding and Controls*
CharacteristicCases of Bleeding (n = 80)Controls (n = 80)P Value
Number (%) or Mean ± SD
Age (years)67 ± 1566 ± 150.63
Male sex53 (66)52 (65)1.00
History of peptic ulcer27 (34)10 (13)0.003
History of upper gastrointestinal bleeding15 (19)5 (6)0.03
Dyspepsia within the last 2 weeks23 (29)12 (15)0.06
Current smoking18 (23)17 (21)1.00
Alcohol >20 g/wk17 (21)9 (11)0.13
Type of NSAID 0.21
Nonaspirin NSAID28 (35)26 (33)
Aspirin47 (59)53 (66)
Both5 (6)1 (1)
NSAID duration 0.16
<1 week21 (26)16 (20)
1 week–3 months25 (31)16 (20)
3–12 months12 (15)19 (24)
>12 months22 (28)29 (36)
Dose of NSAID 0.15
<3.5 defined daily doses/wk18 (23)9 (11)
3.5–7 defined daily doses/wk33 (41)40 (50)
>7 defined daily doses/wk29 (36)31 (39)
Corticosteroids3 (4)3 (4)1.00
Anticoagulants2 (3)1 (1)1.00
Other drugs57 (71)60 (75)0.72
Comorbid diseases 0.15
None/other35 (44)29 (36)
Heart disease30 (38)39 (49)
Cerebrovascular disease5 (6)3 (4)
Hypertension9 (11)4 (5)
Rheumatoid arthritis1 (1)5 (6)
Helicobacter pylori infection63 (79)45 (56)0.004
CagA seropositivity48 (60)37 (46)0.11
CagA seropositivity in H. pylori-positive cases, n/N (%)48/63 (76)37/45 (82)0.61

CagA = cytotoxin-associated antigen; NSAID = nonsteroidal anti-inflammatory drug.

* All cases and controls were users of NSAIDs.

Table 2. Type of Nonsteroidal Anti-inflammatory Drugs Used by Cases (within the 7 Days before the Bleeding Episode) and by Controls (within the 7 Days before Completion of Questionnaire)
Type of NSAIDCases of Bleeding (n = 80)Controls (n = 80)
Number
Nonaspirin NSAID2826
Aceclofenac2*5
Diclofenac4†, ‡, §6
Ibuprofen11
Indomethacin10
Ketoprofen20
Mefenamic acid23∥
Meloxicam3‡0
Naproxen22
Niflamol10
Nimesulid10*, †11∥
Piroxicam30
Tenoxicam2§0
Aspirin4753
Aspirin2513
Salospir2139
Egicalm11
Nonaspirin NSAID + aspirin51

NSAID = nonsteroidal anti-inflammatory drug.

* Both aceclofenac and nimesulid were used by 1 case of bleeding.

† Both diclofenac and nimesulid were used by 2 cases of bleeding.

‡ Both diclofenac and meloxicam were used by 1 case of bleeding.

§ Both diclofenac and tenoxicam were used by 1 case of bleeding.

∥ Both mefenamic acid and nimesulid were used by 2 controls.

Upper gastrointestinal bleeding was associated with gastric ulcer(s) only in 24 patients (30%), duodenal ulcer only in 21 patients (26%), both gastric and duodenal ulcers in 12 patients (15%), hemorrhagic gastritis only in 8 patients (10%), gastric ulcer and hemorrhagic gastritis in 6 patients (8%), duodenal ulcer and hemorrhagic gastritis in 6 patients (8%), and gastric and duodenal ulcers and hemorrhagic gastritis in 3 patients (4%).

A previous history of peptic ulcer disease, upper gastrointestinal bleeding, and perhaps dyspepsia within the last 2 weeks were more common in cases than controls (Table 1). In contrast, there was no difference between the two groups in smoking habits, alcohol consumption, recent use of corticosteroids, use of anticoagulants, or comorbid diseases (Table 1).

Association of h. pylori infection with bleeding 

H. pylori infection was detected in 63 cases (79%) and 45 controls (56%) (P = 0.004; Table 1). Two cases and 1 control had discordant results between the serologic test and the 13C-urea breath test; exclusion of these patients did not affect the results. Similarly, among patients who underwent the 13C-urea breath test, H. pylori infection was present in 78% (51/65) of the cases of bleeding and 57% (29/51) of the controls (P = 0.02).

There were no significant differences in CagA positivity either between the cases and controls, or between the H. pylori–positive cases and controls (Table 1).

In a multivariate logistic regression analysis, H. pylori infection was the only significant risk factor for upper gastrointestinal bleeding (odds ratio = 1.7; 95% confidence interval: 1.2 to 2.5; P = 0.004), although both a history of peptic ulcer disease and a short duration of NSAID use were more common in cases (Table 3). An analysis to investigate whether the effect of H. pylori on the risk of bleeding differed by the type of NSAID (aspirin vs. other) found no evidence of a statistically significant interaction (P = 0.22).

Table 3. Factors Associated with Upper Gastrointestinal Bleeding in a Multivariate Analysis
FactorOdds Ratio (95% Confidence Interval)*P Value
Helicobacter pylori infection1.7 (1.2–2.5)0.004
History of peptic ulcer1.8 (0.97–3.2)0.06
History of upper gastrointestinal bleeding1.2 (0.5–2.5)0.73
Dyspepsia within the last 2 weeks1.4 (0.9–2.1)0.15
NSAID duration of use ≤3 months†1.4 (0.9–2.1)0.10

NSAID = nonsteroidal anti-inflammatory drug.

* Also adjusted for the type of NSAID used (nonaspirin only, aspirin only, or both).

† Compared with >3 months.

Back to Article Outline

Discussion 

We found that H. pylori infection was an independent risk factor for upper gastrointestinal bleeding in users of NSAIDs, including aspirin, consistent with the results of a recent meta-analysis (20). In our study, H. pylori infection almost doubled the risk of upper gastrointestinal bleeding among NSAID users after adjustment for other risk factors for bleeding. Although the reported effects of H. pylori infection on the risk of upper gastrointestinal bleeding vary widely, from a 50% decrease 13, 21 to a threefold increase (22), differences in the inclusion criteria for cases of bleeding and for controls may be responsible for many of the discrepancies 12, 13, 23, 24. Our sample consisted exclusively of current NSAID users and our findings are in agreement with those from studies with similar design and inclusion criteria 11, 25.

CagA status of H. pylori did not affect the risk of upper gastrointestinal bleeding among NSAID users in our study. More than 75% of H. pylori–positive cases and controls were CagA positive. Although CagA seropositivity has been associated with a higher risk of ulcer bleeding (21), it was not associated with the risk of bleeding in a study that included low-dose aspirin users (26).

We found no differences between aspirin and other NSAIDs in terms of their effects on the risk of upper gastrointestinal bleeding, as has been seen in other case-control studies 11, 25. However, the results of a recent randomized trial suggested that the effect of H. pylori infection on the risk of bleeding may differ between users of aspirin and other NSAIDs (27).

Short duration (≤3 months) of NSAID use was more common in cases of bleeding than in controls, consistent with observations that the risk of gastrointestinal complications is higher during the first months of NSAID use 1, 28. In addition, previous peptic ulcer disease or upper gastrointestinal bleeding, and recent dyspepsia, were more common among cases 17, 26, 28, 29.

In conclusion, H. pylori infection almost doubled the risk of upper gastrointestinal bleeding among users of NSAIDs, whereas CagA status was not associated with the risk of bleeding. Thus, as has been suggested by others 11, 25, 27, H. pylori eradication may reduce the risk of upper gastrointestinal bleeding among NSAID users. Whether H. pylori eradication alone is sufficient, or whether additional long-term use of antisecretory drugs should be advised in high-risk subgroups of NSAID users (30), should be evaluated in prospective clinical trials.

Back to Article Outline

References 

  1. Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious gastrointestinal complications related to use of nonsteroidal anti-inflammatory drugs (a meta-analysis). Ann Intern Med. 1991;115:787–796
  2. Wolfe M, Lichtestein D, Singh G. Gastrointestinal toxicity of non-steroidal anti-inflammatory drugs. N Engl J Med. 1999;340:1888–1899
  3. Lauer MS. Aspirin for primary prevention of coronary events. N Engl J Med. 2002;346:1468–1474
  4. Suerbaum S, Michetti P. Helicobacter pylori infection. N Engl J Med. 2002;347:1175–1186
  5. Barkin J. The relation between Helicobacter pylori and nonsteroidal anti-inflammatory drugs. Am J Med. 1998;105(suppl):22S–27S
  6. Laine L. Review article (the effect of Helicobacter pylori infection on nonsteroidal anti-inflammatory drug-induced upper gastrointestinal tract injury). Aliment Pharmacol Ther. 2002;16(suppl 1):34–39
  7. Chan FK. Helicobacter pylori, NSAIDs and gastrointestinal haemorrhage. Eur J Gastroenterol Hepatol. 2002;14:1–3
  8. Laine L, Marin-Sorenson M, Weistein WM. Nonsteroidal anti-inflammatory drug associated gastric ulcers do not require Helicobacter pylori for their development. Am J Gastroenterol. 1992;87:1398–1402
  9. Hawkey CJ, Tulassay Z, Szczepanski L, et al.  Randomised controlled trial of Helicobacter pylori eradication in patients on non-steroidal anti-inflammatory drugs (HELP NSAIDs study. Helicobacter Eradication for Lesion Prevention). Lancet. 1998;352:1016–1021
  10. Loeb DS, Talley NJ, Ahlquist DA, et al.  Long-term nonsteroidal anti-inflammatory drug use and gastroduodenal injury (the role of Helicobacter pylori infection). Gastroenterology. 1992;102:1899–1905
  11. Aalykke C, Lauritsen JM, Hallas J, et al.  Helicobacter pylori and risk of ulcer bleeding among users of nonsteroidal anti-inflammatory drugs (a case-control study). Gastroenterology. 1999;116:1305–1309
  12. Cullen DJ, Hawkey GM, Greenwood DC, et al.  Peptic ulcer bleeding in the elderly (relative roles of Helicobacter pylori and non-steroidal anti-inflammatory drugs). Gut. 1997;41:459–462
  13. Pilotto A, Leandro G, Di Mario F, et al.  Role of Helicobacter pylori infection on upper gastrointestinal bleeding in the elderly (a case-control study). Dig Dis Sci. 1997;42:586–591
  14. Covacci A, Censini S, Bugnoli M, et al.  Molecular characterization of the 128-kDa immunodominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc Natl Acad Sci USA. 1993;90:5791–5795
  15. Li L, Kelly LK, Ayub K, et al.  Genotypes of Helicobacter pylori obtained from gastric ulcer patients taking or not taking NSAIDs. Am J Gastroenterol. 1999;94:1502–1507
  16. van Doorn LJ, Figueiredo C, Sanna R, et al.  Clinical relevance of the cagA, vacA, and iceA status of Helicobacter pylori. Gastroenterology. 1998;115:58–66
  17. Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual non-steroidal anti-inflammatory drugs. Lancet. 1994;343:769–772
  18. WHO Collaborating Centre for Drug Statistics and Methodology . Anatomical Therapeutic Chemical (ATC) Classification Index Including Defined Daily Doses (DDDs) for Plain Substances. Oslo, Norway: World Health Organization; 1995;
  19. Pateraki E, Mentis A, Spiliadis C, et al.  Seroepidemiology of Helicobacter pylori infection in Greece. FEMS Microbiol Immunol. 1990;64:129–136
  20. Huang J-Q, Sridhar S, Hunt RH. Role of Helicobacter pylori infection and non-steroidal anti-inflammatory drugs in peptic-ulcer disease (a meta-analysis). Lancet. 2002;359:14–22
  21. Stack WA, Atherton JC, Hawkey GM, et al.  Interactions between Helicobacter pylori and other risk factors for peptic ulcer bleeding. Aliment Pharmacol Ther. 2002;16:497–506
  22. Labenz J, Peitz U, Kohl H, et al.  Helicobacter pylori increases the risk of peptic ulcer bleeding (a case-control study). Ital J Gastroenterol Hepatol. 1999;31:110–115
  23. Santolaria S, Lanas A, Benito R, et al.  Helicobacter pylori infection is a protective factor for bleeding gastric ulcers but not for bleeding duodenal ulcers in NSAID users. Aliment Pharmacol Ther. 1999;13:1511–1518
  24. Ng TM, Fock KM, Khor JL, et al.  Non-steroidal anti-inflammatory drugs, Helicobacter pylori and bleeding gastric ulcer. Aliment Pharmacol Ther. 2000;14:203–209
  25. Hawkey CJ. Risk of ulcer bleeding in patients infected with Helicobacter pylori taking non-steroidal anti-inflammatory drugs. Gut. 2000;46:310–311
  26. Lanas A, Fuentes J, Benito R, et al.  Helicobacter pylori increases the risk of upper gastrointestinal bleeding in patients taking low-dose aspirin. Aliment Pharmacol Ther. 2002;16:779–786
  27. Chan FK, Chung SC, Suen BY, et al.  Preventing recurrent upper gastrointestinal bleeding in patients with Helicobacter pylori infection who are taking low-dose aspirin or naproxen. N Engl J Med. 2001;344:967–973
  28. Laine L. Approaches to nonsteroidal anti-inflammatory drugs use in the high-risk patient. Gastroenterology. 2001;120:594–606
  29. Hansen JM, Hallas J, Lauritsen JM, Bytzer P. Non-steroidal anti-inflammatory drugs and complications (a risk factor analysis for clinical decision-making). Scand J Gastroenterol. 1996;31:126–130
  30. Lai KC, Lam SK, Chu KM, et al.  Lansoprazole for the prevention of recurrences of ulcer complications from long-term low-dose aspirin use. N Engl J Med. 2002;346:2033–2038

☆ Despina Papadelli received a scholarship from the Hellenic Gastroenterology Association.

PII: S0002-9343(04)00024-5

doi:10.1016/j.amjmed.2003.10.040

The American Journal of Medicine
Volume 116, Issue 9 , Pages 601-605, 1 May 2004

Article Tools