Helicobacter pylori eradication and the prevention of peptic ulcer bleeding and cardiovascular disease progression in chronic aspirin users

Table 1 Baseline characteristics of study participants before Helicobacter pylori treatment, n (%)

Category	Variable	H. pylori-positive (n = 262)	H. pylori-negative (n = 138)	Nominal P value	Adjusted q value
Sociodemographic	Age (years)	60.2 ± 9.2	60.3 ± 7.3	0.938	0.938
	Male	138 (52.7)	78 (56.5)	0.604	0.672
	Female	124 (47.3)	60 (43.5)	0.604	0.672
	Diabetes mellitus	72 (27.5)	43 (31.2)	0.452	0.603
Biochemical	ALT (IU/L), median (interquartile range)	40.4 (22.6)	36.3 (28.8)	0.269	0.403
	AST (IU/L), median (interquartile range)	42.7 (23.2)	37.9 (20.6)	0.150	0.300
	Albumin (g/dL)	3.5 ± 0.4b	3.9 ± 0.6	< 0.001	< 0.001
	Urea (mg/dL)	29.3 ± 6.4b	23.5 ± 4.6	< 0.001	< 0.001
	Sodium (mmol/L)	137.8 ± 2.9b	135.3 ± 3.5	< 0.001	< 0.001
	Potassium (mmol/L)	4.1 ± 0.9a	3.8 ± 0.6a	0.014	0.028
	Cholesterol (mg/dL)	191.0 ± 29.7a	177.0 ± 35.9a	0.004	0.012
	Triglycerides (mg/dL), median (interquartile range)	94.6 (58.6)	87.6 (55.2)	0.414	0.518
	HDL (mg/dL), median (interquartile range)	43.1 (41.2)	53.5 (50.2)	0.118	0.236
	LDL (mg/dL), median (interquartile range)	156.3 (88.9)	140.6 (81.5)	0.224	0.374
	Random blood sugar (mg/dL), median (interquartile range)	128.5 (89.2)	130.1 (92.3)	0.850	0.850
	HbA1c (%)	6.8 ± 1.2	6.9 ± 1.3	0.587	0.671
Digestive symptoms (GSRS)	Total GSRS score	2.9 ± 0.8b	1.7 ± 0.6	< 0.001	< 0.001
	Abdominal pain	90 (34.4)a	30 (21.7)	0.012	0.024
	Bloating	60 (22.9)	20 (14.5)	0.045	0.068
	Nausea	50 (19.1)a	10 (7.2)	0.003	0.009

^aP < 0.05 vs Helicobacter pylori (H. pylori)-negative group, after false discovery rate (FDR) adjustment where applicable.

^bP < 0.001 vs H. pylori-negative group, after FDR adjustment where applicable.

Data are presented as mean ± SD for normally distributed continuous variables and were compared using Student’s t-test. Continuous variables presented as median (interquartile range) were compared using Mann-Whitney U test. Categorical variables (n, %) were compared using Pearson’s χ² test. Exploratory comparisons were adjusted for multiple testing using the Benjamini-Hochberg false discovery rate procedure; adjusted q-values are reported. H. pylori: Helicobacter pylori; ALT: Alanine transaminase; AST: Aspartate aminotransferase; HDL: High density lipoprotein; LDL: Low-density lipoprotein; HbA1c: Glycated hemoglobin; GSRS: Gastrointestinal Symptom Rating Scale.

Table 2 Endoscopic and histopathological findings in study participants, n (%)

Category	Variable	H. pylori-positive (n = 262)	H. pylori-negative (n = 138)	P value
Endoscopic	Normal	122 (46.6)	136 (98.6)	< 0.001
	Atrophic gastritis	62 (23.7)	0 (0.0)	< 0.001
	Gastric erosion	34 (12.2)	0 (0.0)	< 0.001
	Gastric ulcer	28 (10.7)	0 (0.0)	< 0.001
	Antral erosion	6 (2.3)	0 (0.0)	0.378
	Duodenal ulcer	6 (2.3)	0 (0.0)	0.378
	Duodenitis	6 (2.3)	0 (0.0)	0.378
Histopathological	H. pylori colonization	262 (100)	0 (0.0)	< 0.001
	Inflammation severity
	Mild	80 (30.5)	40 (29.0)	0.75
	Moderate	120 (45.8)	30 (21.7)	< 0.001
	Severe	50 (19.1)	0 (0.0)	< 0.001
	Atrophy	40 (15.3)	0 (0.0)	< 0.001
	Intestinal metaplasia	20 (7.6)	0 (0.0)	0.002

Gastric ulcers and duodenal ulcers are subtypes of peptic ulcers, differentiated by their location in the stomach and duodenum, respectively. A P value < 0.05 was considered statistically significant. H. pylori: Helicobacter pylori.

Table 3 Prevalence of cardiovascular diseases in study participants before Helicobacter pylori treatment, n (%)

Disease	H. pylori positive	H. pylori negative	P value
IHD	142 (54.3)	46 (33.3)	< 0.001
PCI	23 (8.7)	17 (12.5)	0.012
IHD-PCI	17 (6.5)	17 (12.5)	0.018
CABG	46 (17.4)	17 (12.5)	0.045
CSA	6 (2.1)	0 (0.0)	0.078
Post CABG	11 (4.2)	17 (12.5)	0.064
Post PCI	6 (2.1)	0 (0.0)	0.210
Inferior MI	6 (2.1)	6 (4.2)	0.320
Prosthetic MV	6 (2.1)	6 (4.2)	0.320
Post PCS	6 (2.1)	0 (0.0)	0.210

Comparisons between groups were performed using the χ² test. A P value < 0.05 was considered statistically significant. IHD: Ischemic heart disease; CSA: Chronic stable angina; PCI: Percutaneous coronary intervention; CABG: Coronary artery bypass graft; MI: Myocardial infarction; MV: Mitral valve; PCS: Post-cardiac surgery; H. pylori: Helicobacter pylori.

Table 4 Association between peptic ulcer bleeding and specific cardiovascular diseases

Cardiovascular disease	PUB incidence	No PUB incidence	Adjusted odds ratio (95%CI)	P value
Category	54.3%	33.3%	2.5 (1.3-4.8)	0.006
Endoscopic	17.4%	12.5%	1.8 (1.2-2.7)	0.045
Histopathological	8.7%	6.0%	1.2 (0.8-1.8)	0.120
Category	2.1%	1.5%	1.1 (0.9-1.3)	0.320

PUB: Peptic ulcer bleeding; CI: Confidence interval.

Table 5 Incidence of peptic ulcer bleeding and cardiovascular events by type and number of additional antithrombotic/anticoagulant medications

Medication group	PUB incidence n/N (%)	Cardiovascular events n/N (%)	Adjusted odds ratio for PUB	95%CI for PUB	P value for PUB
By type of additional medication
Aspirin + clopidogrel	42/400 (10.5)	101/400 (25.3)	1.2	0.8-1.8	0.120
Aspirin + warfarin	39/400 (9.8)	99/400 (24.7)	1.1	0.7-1.6	0.150
Aspirin + DOACs	45/400 (11.2)	104/400 (26.1)	1.3	0.9-1.9	0.200
By number of medications
Dual therapy (aspirin + 1 drug)	35/400 (8.7)	94/400 (23.5)	1.0 (reference)	-	-
Triple therapy (aspirin + 2 drugs)	61/400 (15.2)	116/400 (28.9)	1.8	1.2-2.7	0.004

Medication-type categories (aspirin + clopidogrel/warfarin/direct oral anticoagulants) are not mutually exclusive. Patients receiving triple antithrombotic therapy contributed to all relevant medication-type groups as well as to the “triple therapy” group. All percentages are calculated using the total study cohort (n = 400) as the denominator. PUB: Peptic ulcer bleeding; CI: Confidence interval; DOACs: Direct oral anticoagulants.

Table 6 Prevalence of gastric lesions, cardiovascular disease status, and adverse events: Comparison among Helicobacter pylori eradicated, persistent (6-week and 6-month), and Helicobacter pylori-negative groups, n (%)

Category	Variable	H. pylori-positive group			H. pylori-negative group (n = 138)	P value (eradicated vs persistent 6-week post-treatment)	P value (persistent 6-month vs negative)
		Eradicated (n = 136)	Persistent 6-week post-treatment (n = 126)	Persistent 6-month follow-up (n = 126)
Gastric	PUB	5 (3.7)	26 (20.6)	15 (11.9)	0 (0.0)	< 0.001	< 0.001
Outcomes	Gastric erosions	8 (5.9)	26 (20.6)	12 (9.5)	0 (0.0)	< 0.001	< 0.001
	Gastric ulcer	6 (4.4)	22 (17.5)	10 (7.9)	0 (0.0)	< 0.001	< 0.001
	Antral erosion	2 (1.5)	6 (4.8)	2 (1.6)	0 (0.0)	0.02	0.06
	Duodenal ulcer	3 (2.2)	8 (6.3)	4 (3.2)	0 (0.0)	0.03	0.08
CV disease progression status1	CV disease progression	40 (29.4)	96 (76.2)	92 (73.0)	44 (32.0)	< 0.001	< 0.001
	PCI	10 (7.4)	11 (8.7)	10 (7.9)	15 (11.0)	0.05	0.06
	IHD-PCI	16 (11.8)	22 (17.5)	20 (15.9)	6 (4.3)	0.01	0.02
	CABG	12 (8.8)	30 (23.8)	28 (22.2)	17 (12.0)	0.002	0.005
	CSA	4 (2.9)	6 (4.8)	3 (2.4)	0 (0.0)	0.03	0.07
	Post CABG	5 (3.7)	12 (9.5)	10 (7.9)	4 (2.9)	0.07	0.08
	Post PCI	5 (3.7)	6 (4.8)	5 (4.0)	0 (0.0)	0.22	0.33
	Inferior MI	4 (2.9)	6 (4.8)	4 (3.2)	0 (0.0)	0.33	0.34
	Prosthetic MV	4 (2.9)	6 (4.8)	4 (3.2)	0 (0.0)	0.33	0.34
	Post PCS	4 (2.9)	6 (4.8)	4 (3.2)	0 (0.0)	0.22	0.33
Adverse events	Fatigue	4 (2.9)	14 (11.1)	10 (7.9)	8 (5.8)	0.01	0.02
	Dizziness	3 (2.2)	10 (7.9)	6 (4.8)	5 (3.6)	0.02	0.03
	Skin Rash	2 (1.5)	6 (4.8)	4 (3.2)	2 (1.4)	0.02	0.04

¹CV disease progression status at 6 months represent the presence of the condition (either pre-existing or newly progressed) at the 6-month follow-up assessment.

Comparisons between groups were performed using the χ² test. A P value < 0.05 was considered statistically significant. H. pylori: Helicobacter pylori; PUB: Peptic ulcer bleeding; IHD: Ischemic heart disease; PCI: Percutaneous coronary intervention; CABG: Coronary artery bypass graft; CSA: Chronic stable angina; MI: Myocardial infarction; MV: Mitral valve; PCS: Post-cardiac surgery; CV: Cardiovascular.

Table 7 Kaplan-Meier analysis of time to first peptic ulcer bleed and time to first documented incident cardiovascular disease progression

Outcome	Time point	H. pylori-positive (persistent), incidence (95%CI)	H. pylori-positive (eradicated), incidence (95%CI)	H. pylori-negative, incidence (95%CI)	P value
PUB	6 weeks	8.7% (5%-13%)	1.5% (0.5%-4%)	0% (0%-2%)	< 0.001
	6 months	11.9% (8%-17%)	3.7% (2%-7%)	0% (0%-2%)	< 0.001
Incident CV disease progression	6 weeks	8.0% (5%-12%)	3.7% (2%-7%)	4.3% (2%-8%)	0.045
	6 months	14.3% (10%-20%)	5.9% (3%-10%)	7.2% (4%-12%)	0.002
New CABG procedure	6 weeks	2.4% (1%-6%)	0.7% (0.2%-3%)	1.4% (0.5%-4%)	0.210
	6 months	4.0% (2%-8%)	1.5% (0.5%-4%)	2.9% (1%-6%)	0.120
New PCI procedure	6 weeks	3.2% (1%-7%)	1.5% (0.5%-4%)	2.2% (1%-5%)	0.320
	6 months	5.6% (3%-10%)	2.9% (1%-6%)	4.3% (2%-8%)	0.180

Cardiovascular disease progression is the pre-specified composite endpoint defined as the first occurrence of: New coronary revascularization (percutaneous coronary intervention or coronary artery bypass graft), hospitalization for unstable angina, or documented worsening of Canadian Cardiovascular Society angina class during follow-up. Only incident events occurring after study enrollment are included. A P value < 0.05 was considered statistically significant. H. pylori: Helicobacter pylori; PUB: Peptic ulcer bleeding; CABG: Coronary artery bypass graft; PCI: Percutaneous coronary intervention; CV: Cardiovascular; CI: Confidence interval.

Table 8 Cox regression analysis for predictors of time-to-first-peptic ulcer bleeding and time-to-first-cardiovascular-disease-progression

Variable	Adjusted hazard ratio	95%CI	P value
Predictors of PUB
H. pylori infection	4.2	2.1-8.3	< 0.001
Low albumin levels	2.8	1.5-5.2	0.002
Diabetes mellitus	1.3	0.8-2.0	0.28
Age	1.1	0.9-1.3	0.32
Sex (male)	1.2	0.8-1.8	0.45
Predictors of CV disease progression
H. pylori infection	3.5	1.8-6.7	< 0.001
Multivessel coronary disease at baseline	2.8	1.5-5.2	0.001
Diabetes mellitus	1.6	1.0-2.5	0.045
Age	1	0.9-1.2	0.55
Sex (male)	1.1	0.8-1.5	0.6

Adjusted for age, sex, diabetes mellitus, baseline albumin, adherence, and antithrombotic regimen intensity. Cardiovascular (CV) progression models also adjusted for baseline multivessel coronary disease. Hazard ratios (HRs) were calculated using Cox proportional hazards regression. Models for peptic ulcer bleeding (PUB) were adjusted for age, sex, diabetes mellitus, baseline albumin levels, adherence to treatment, and antithrombotic regimen intensity (dual vs triple therapy). Models for CV disease progression were adjusted for the same covariates plus baseline multivessel coronary disease. Proportional hazards assumptions were verified using Schoenfeld residuals (all P > 0.05). Complete-case analysis was used (n = 400 for PUB models; n = 188 for CV disease progression models, restricted to patients with baseline CV disease). The adjusted HR for Helicobacter pylori infection predicting PUB (4.2) was coincidentally identical to the adjusted odds ratio from the logistic regression model (Table 9), as both models were adjusted for the same set of covariates. CV: Cardiovascular; H. pylori: Helicobacter pylori; PUB: Peptic ulcer bleeding; CI: Confidence interval.

Table 9 Multivariable logistic regression for predictors of Helicobacter pylori eradication failure and for the composite endpoint of peptic ulcer bleeding or cardiovascular disease progression at 6 months

Variable	Adjusted odds ratio	95%CI	P value
Predictors of PUB
H. pylori infection	4.2	2.1-8.3	< 0.001
Low albumin levels	2.8	1.5-5.2	0.002
Diabetes mellitus	1.2	0.8-1.9	0.35
Poor adherence	3.1	1.8-5.4	< 0.001
Predictors of CV disease progression
H. pylori infection	3.2	1.7-6.0	< 0.001
Multivessel coronary disease at baseline	2.8	1.5-5.2	0.001
Diabetes mellitus	1.5	1.0-2.4	0.04
Poor adherence	2.1	1.1-3.9	0.012

Adjusted for age, sex, diabetes mellitus, baseline albumin, adherence, antithrombotic regimen intensity, and for cardiovascular (CV) progression models, baseline multivessel coronary disease. Adjusted odds ratios (aORs) were calculated using multivariate logistic regression. Models for peptic ulcer bleeding (PUB) were adjusted for age, sex, diabetes mellitus, baseline albumin levels, adherence to treatment, and antithrombotic regimen intensity (dual vs triple therapy). Models for cardiovascular disease progression were adjusted for the same covariates plus baseline multivessel coronary disease. Complete-case analysis was used (n = 400 for PUB models; n = 188 for CV disease progression models). The aOR for Helicobacter pylori infection predicting PUB (4.2) was coincidentally identical to the adjusted hazard ratio from the Cox regression model (Table 8), as both models were adjusted for the same set of covariates. CV: Cardiovascular; H. pylori: Helicobacter pylori; PUB: Peptic ulcer bleeding; CI: Confidence interval.

Table 10 Diagnostic accuracy of Helicobacter pylori detection methods

Diagnostic method	Sensitivity (%)	95%CI	Specificity (%)	95%CI	Kappa statistic
Stool antigen test	90	85-94	92	88-95	0.85
Urea breath test	88	83-92	90	86-93	0.82
Histopathology	95	91-98	98	95-99	1.00

CI: Confidence interval.

Table 11 Adverse events and their impact on treatment adherence

Adverse event	Frequency (%)	OR (95%CI)	P value
Fatigue	11.1	2.5 (1.3-4.8)	0.006
Dizziness	7.9	2.2 (1.1-4.2)	0.020
Skin rash	4.8	1.8 (1.0-3.2)	0.040
Nausea	5.6	2.0 (1.1-3.6)	0.030
Diarrhea	3.2	1.5 (0.8-2.8)	0.200

OR: Odds ratio; CI: Confidence interval.

Table 12 Subgroup analysis of risk factors for pub and cardiovascular disease progression in low-dose aspirin users

Subgroup	Outcome	HR	95%CI	Nominal P value	Adjusted q value
Age	Risk of PUB (≥ 65 years vs < 65 years)	2	1.4-2.9	0.002	0.003
	CV events (≥ 65 years vs < 65 years)	1.8	1.2-2.5	0.008	0.009
Sex	Risk of PUB (males vs females)	1.7	1.2-2.3	0.011	0.013
	CV events (males vs females)	1.5	1.1-2.1	0.042	0.042
Comorbidities	Risk of PUB (IHD vs no IHD)	2.8	1.6-5.1	0.004	0.005
	Treatment failure (IHD vs no IHD)	2.4	1.3-4.3	0.006	0.007
Diabetes status	Risk of PUB (DM vs no DM)	1.4	0.9-2.1	0.12	0.12
	CV event (DM vs no DM)	1.7	1.1-2.5	0.015	0.018
Eradication status	PUB risk (persistent vs eradicated H. pylori)	3.5	2.1-5.9	< 0.001	< 0.001
	CV events (persistent vs eradicated)	3.2	2.0-5.0	< 0.001	< 0.001
Adherence levels	PUB risk (low vs high adherence)	2.6	1.5-4.5	0.003	0.004
	Treatment failure (low vs high adherence)	3.1	1.8-5.4	< 0.001	< 0.001
Biochemical markers	PUB risk (hypoalbuminemia vs normal albumin)	2.5	1.4-4.2	0.005	0.006
	CV events (high urea vs normal)	1.9	1.2-3.1	0.02	0.022

P values were adjusted using the Benjamini-Hochberg false discovery rate procedure. q < 0.05 indicates statistical significance after correction. DM: Diabetes mellitus; H. pylori: Helicobacter pylori; HR: Hazard ratio; IHD: Ischemic heart disease; PUB: Peptic ulcer bleeding; CV: Cardiovascular; CI: Confidence interval.

Table 13 Subgroup analysis of cardiovascular conditions

CV disease	Outcome	HR	95%CI	Nominal P value	Adjusted q value
IHD	Risk of PUB	2.8	1.6-5.1	0.004	0.012
	CV events	2.5	1.3-4.8	0.006	0.012
PCI	Risk of PUB	1.8	1.2-2.7	0.045	0.068
	CV events	1.5	1.1-2.1	0.042	0.068
CABG	Risk of PUB	2	1.4-2.9	0.002	0.012
	CV events	1.8	1.2-2.5	0.008	0.024

P values were adjusted using the Benjamini-Hochberg false discovery rate procedure. q < 0.05 indicates statistical significance after correction. HR: Hazard ratio; IHD: Ischemic heart disease; PCI: Percutaneous coronary intervention; CABG: Coronary artery bypass graft; PUB: Peptic ulcer bleeding; CI: Confidence interval; CV: Cardiovascular.

Table 14 Sensitivity analysis for unmeasured confounding using E-values

Outcome	Adjusted HR for H. pylori	95%CI	E-value (point estimate)	E-value for CI lower bound	Interpretation
PUB	4.2	2.1-8.3	5.0	3.1	To fully explain the observed HR of 4.2, an unmeasured confounder would need to be associated with both H. pylori and PUB by hazard ratios of ≥ 5.0 each, beyond the measured covariates
CV disease progression	3.5	1.8-6.7	4.2	2.6	To fully explain the observed HR of 3.5, an unmeasured confounder would need to be associated with both H. pylori and CV progression by hazard ratios of ≥ 4.2 each

E-value: The minimum strength of association (on the risk-ratio scale) an unmeasured confounder must have with both exposure and outcome, conditional on measured covariates, to explain away the observed association. PUB: Peptic ulcer bleeding; H. pylori: Helicobacter pylori; CI: Confidence interval; HR: Hazard ratio; CV: Cardiovascular.