Impact of Helicobacter pylori eradication timing on the risk of thromboembolism events in patients with peptic ulcer disease: a population-based cohort study.

OBJECTIVES: To evaluate the impact of Helicobacter pylori eradication on venous thromboembolism (VTE) events, and the differences between early and late treatment timing.
DESIGN: A population-based cohort study.
SETTING: Taiwan's National Health Insurance Research Database. PARTICIPANTS: A total of 6736 patients who received H. pylori eradication therapy from 2000 to 2010 were identified. We randomly selected 26 944 subjects matching in gender, age and baseline year as comparison cohort. PRIMARY AND SECONDARY OUTCOME MEASURES: The incidence rate ratios of VTE in the H. pylori eradication cohorts to that of the control cohort were examined. Multivariable Cox proportional hazard regression analysis was used to estimate the relative HRs and 95% CI of VTE development.
RESULTS: The total incidence rate of VTE was observed in the late H. pylori eradication cohort, the early H. pylori eradication cohort and the control cohort (15.2, 3.04 and 2.91 per 1000 person-years, respectively). An age-specific trend was found in the late H. pylori eradication cohort, with a greater rate of VTE in the 50-65 years and more than 65 years age groups (adjusted HR 5.44; 95% CI 4.21 to 7.03 and 3.13; 95% CI 2.46 to 3.99). With comorbidities, the late H. pylori eradication cohort seemed to have the highest VTE incidence rate and adjusted HR (4.48, 95% CI 3.78 to 5.30).
CONCLUSIONS: Late H. pylori eradication was associated with a significantly increased risk of VTE, and there was a significantly greater risk of VTE in patients with female gender, age more than 50 years and with comorbidities. © Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

This study has a large number of subjects selected from national population database to investigate the effect of Helicobacter pylori eradication on venous thromboembolism events.

Multivariate Cox proportional hazard regression was used to control the confounding factors and elucidate the influence of early or late H. pylori treatment on venous thromboembolism events.

The National Health Insurance database lacked some important information of H. pylori treatment, such as the compliance and success or failure of H. pylori eradication.

Introduction

Helicobacter pylori infection is one of the most important infections in humans. Chronic H. pylori infection causes chronic gastritis, atrophic gastritis, intestinal metaplasia, dysplasia and gastric cancer.1 Besides gastric diseases, non-gastric events are also associated with H. pylori infection, including cardiovascular diseases, lung diseases, haematological diseases, eye and skin diseases, hepatobiliary diseases, diabetes mellitus and neurological disorders.2 3 H. pylori eradication treatment for diagnosed H. pylori infection is the standard of care. In recent decades, the clinical guidelines have seen significant scientific advances regarding the management of H. pylori infection.4

The issue of post-H. pylori eradication requires further investigation. H. pylori eradication could improve gastric mucosa atrophy and intestinal metaplasia, decrease peptic ulcer and lower gastric cancer risk, but it exacerbates gastro-oesophageal reflux due to acid.5 The most promising long-term effect of H. pylori eradication could decrease the risk of gastric cancer.6 H. pylori could trigger an inflammatory state and induce molecular mechanisms by expressing virulence peptides. Moreover, H. pylori infection interferes with the host absorbance of different nutrients, and potentially influences the host’s health outside the gastrointestinal tract. Growing evidence has shown that a variety of conditions and disorders are caused by H. pylori infection, such as idiopathic thrombocytopenic purpura, unexplained iron deficiency anaemia and vitamin B12 deficiency. However, the associations of H. pylori with certain cardiovascular diseases, neurodegenerative diseases and disorders of the digestive system outside the stomach have yet to be fully elucidated.

There are several effects and related issues following H. pylori eradication that have yet to be investigated. An early study demonstrated that H. pylori eradication could improve symptoms.7 There were significant long-term improvements in gastric histology following H. pylori eradication when compared with patients with persistent infection.8 It has been clearly established that H. pylori eradication can protect against peptic ulcer recurrence and gastric cancer occurrence.6 9 10 Besides gastric disorders, H. pylori eradication could also cause systemic effects, and several clinical events were found to be associated with H. pylori eradication, regardless of whether or not the H. pylori pathogenic role was defined.11–16

Venous thromboembolism (VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE), is a common and potentially lethal acute medical event with high morbidity and mortality.17 18 H. pylori infection may be a risk factor for stroke of atherothrombotic origin or cardiovascular diseases in animals and humans.7 8 A recent large study of H. pylori eradication demonstrated that eradication of H. pylori could provide metabolic benefits, improving insulin resistance and lipid profile.19 H. pylori eradication provided not only H. pylori-associated pathological events, but also systemic effects, such as thromboembolic events. However, there are few studies on the association of H. pylori infection or H. pylori eradication with VTE. Moreover, the timely eradication of H. pylori infection leads to a better overall health status and better clinical outcomes, as well as the prevention of peptic ulcers and gastric cancer.20–22 Thus, we aimed to investigate H. pylori eradication and the timely effect on VTE events due to the systemic effect of treatment using population-based data.

Materials and methods

Data source

The National Health Insurance (NHI) programme was launched on 1 March 1995 and currently covers almost 99% of Taiwan’s 23 million residents. Data from the Longitudinal Health Insurance Database 2000 were provided by Taiwan’s National Health Research Institutes, and included information on outpatient and ambulatory visits, hospital inpatient care and dental services.23 The diagnostic codes in the current study were based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM).

Sampled participants

Recruited patients were aged 20 years or above with a history of peptic ulcer disease (ICD-9-CM codes 531, 532 and 533), which was subsequently treated with H. pylori eradication therapy. According to the insurance payment rules of the NHI system, the diagnosis of most patients who received H. pylori-related treatment was confirmed by upper gastrointestinal endoscopy examination. The treatment of H. pylori eradication by triple or quadruple therapy was defined as involving multiple medications as follows: a proton pump inhibitor or H2 receptor blocker, clarithromycin or metronidazole, and amoxicillin or tetracycline, with or without bismuth (details of all eligible H. pylori eradication regimens were reported previously).22 These drug combinations were prescribed in the same order, and the duration of therapy was from 7 to 14 days. The early H. pylori eradication cohort was defined as patients receiving H. pylori eradication therapy within 1 year of diagnosis of peptic ulcer disease, and the late H. pylori eradication cohort was defined as patients receiving H. pylori eradication therapy more than 1 year after peptic ulcer diseases were diagnosed. The index date for patients was set as the date that they first received H. pylori eradication therapy. The exclusion criteria were missing data regarding date of birth, sex or history of VTE (ICD-9-CM 415.1, 453.8 except 415.11) prior to the index date. For comparison, controls were randomly selected from the pool of participants without peptic ulcer disease who did not receive H. pylori eradication therapy. According to age (every 5-year span), sex and year of receiving H. pylori eradication therapy, controls were 1:4 frequency matched to each H. pylori eradication case.

Patient and public involvement

Taiwan NHI programme is a general population healthcare insurance system covering more than 99% population. NHI was initiated on 1 March 1995 in Taiwan, and is still working well now. The de-identified database included medical records of enrollees registered in the National Health Insurance Research Database (NHIRD). A subset of the NHIRD contains data of 1 million randomly selected enrollees from the NHI programme and was used in this study. The claims data were totally encrypted identification information. The disease codes of the database were identified according to the ICD-9-CM. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Variables of interest

The general diagnostic process of the NHI programme was based on physical examination and quantitative ultrasound. All patients were followed up until a diagnosis of VTE was made or they were censored for loss to follow-up, withdrawal from the NHI programme or 31 December 2011, whichever occurred first. Hypertension, diabetes, hyperlipidaemia, coronary artery disease (CAD), heart failure, cancer, stroke, and lower leg fracture or surgery were considered as covariates.

Statistical analysis

The Χ2 test and Student’s t-test were used to determine differences in categorical and continuous variables between H. pylori eradication (including early and late H. pylori eradication therapy) and control cohorts. The incidence density rate of VTE was calculated for each instance of early H. pylori eradication therapy, late H. pylori eradication therapy and for the control cohort. The incidence rate ratios of VTE in the H. pylori eradication cohorts to that of the control cohort and the 95% CI were estimated using a Poisson regression model. The relative HRs and 95% CIs of VTE development for the H. pylori eradication cohorts were estimated with multivariable Cox proportional hazard regression analysis adjusted for age, sex, and comorbidities of hypertension, diabetes, hyperlipidaemia, CAD, heart failure, cancer, stroke, and lower leg fracture or surgery. All data analyses were performed using the SAS statistical package (V.9.4 for Windows; SAS Institute). A two-tailed p<0.05 indicated statistical significance.

Results

Demographic data

From 2000 to 2010, a total of 6736 subjects received H. pylori eradication treatment, including 3062 patients who received early H. pylori eradication (early eradication cohort), and 3674 patients who received late H. pylori eradication (late eradication cohort). There were 26 944 subjects who served as the control cohort for comparison. The demographic data are shown in table 1. There were differences in age among the three cohorts: early eradication, late eradication and control (49.5±14.1, 56.0±14.3, 52.6±14.9; p=0.01). There were no differences in gender ratios among the three cohorts. The late H. pylori eradication cohort had more comorbidities compared with the early eradication and control cohorts.

Table 1

Comparison of demographics between gastric disease with Helicobacter pylori eradication and controls

	Control	Early eradication	Late eradication	Eradication, overall
	n=26 944 (%)	n=3062 (%)	n=3674 (%)	n=6736 (%)	P value
Age, years					0.99
≤49	11 432 (42.4)	1616 (52.8)	1242 (33.8)	2858 (42.4)
50–65	9544 (35.4)	986 (32.2)	1400 (38.1)	2386 (35.4)
≥65	5968 (22.2)	460 (15.0)	1032 (28.1)	1492 (22.2)
Mean (SD)*	52.6 (14.9)	49.5 (14.1)	56.0 (14.3)	53.0 (14.6)	0.01
Sex					0.99
Female	12 128 (45.0)	1243 (40.6)	1789 (48.7)	3032 (45.0)
Male	14 816 (55.0)	1819 (59.4)	1885 (51.3)	3704 (55.0)
Comorbidity
Hypertension	7890 (29.3)	868 (28.4)	1721 (46.8)	2589 (38.4)	<0.001
Diabetes	2358 (8.75)	298 (9.73)	541 (14.7)	839 (12.5)	<0.001
Hyperlipidaemia	5354 (19.9)	656 (21.4)	1450 (39.5)	2106 (31.3)	<0.001
CAD	3573 (13.3)	434 (14.2)	1224 (33.3)	1658 (24.6)	<0.001
Heart failure	728 (2.70)	56 (1.83)	217 (5.91)	273 (4.05)	<0.001
Cancer	640 (2.38)	71 (2.32)	156 (4.25)	227 (3.37)	<0.001
Stroke	855 (3.17)	77 (2.51)	222 (6.04)	299 (4.44)	<0.001
Low leg fracture or surgery	580 (2.15)	48 (1.57)	126 (3.43)	174 (2.58)	0.03

*Two-sample t-test.

CAD, coronary artery disease.

HR and incidence rate of VTE

In table 2, it can be seen that there was no significant difference in risk of VTE between the early H. pylori eradication cohort and control cohort. The total incidence rates of VTE were determined for the late H. pylori eradication cohort, the early H. pylori eradication cohort and the control cohort (15.2, 3.04 and 2.91 per 1000 person-years, respectively). With respect to gender, VTE incidence was higher in the late H. pylori eradication cohort in men and women, and female gender in the late H. pylori eradication cohort had the highest incidence rate (20.5 per 1000 person-years). The age-specific trend in the late H. pylori eradication cohort was found to be greater in the 50–65 years and more than 65 years age groups (adjusted HR 5.44; 95% CI 4.21 to 7.03 and 3.13; 95% CI 2.46 to 3.99). The VTE incidence was lower in the early eradication cohort with more than 65 years old (adjusted HR 0.21; 95% CI 0.08 to 0.56). The early H. pylori eradication cohort seemed to have a lower risk of VTE compared with the late H. pylori eradication cohort. Regarding comorbidities, the late H. pylori eradication cohort seemed to have the highest VTE incidence rate and adjusted HR (4.48, 95% CI 3.78 to 5.30).

Table 2

HRs of venous thromboembolism between control subjects and early or late Helicobacter pylori eradication

	Control(N=26 944)		Early eradication (N=3062)		IRR (95% CI)	Adjusted HR	Late eradication (N=3674)		IRR (95% CI)	Adjusted HR†
	Event	Rate*	Event	Rate*		(95% CI)	Event	Rate*		(95% CI)
All	414	2.91	54	3.04	1.04 (0.93 to 1.18)	1.12 (0.84 to 1.49)	272	15.2	5.22 (4.90 to 5.57)‡	3.91 (3.33 to 4.59)‡
Gender
Female	220	3.46	23	3.14	0.91 (0.75 to 1.10)	0.96 (0.62 to 1.48)	171	20.5	5.92 (5.40 to 6.48)‡	4.39 (3.54 to 5.43)‡
Male	194	2.47	31	2.97	1.20 (1.04 to 1.39)‡	1.29 (0.88 to 1.89)	101	10.6	4.29 (3.91 to 4.71)‡	3.13 (2.43 to 4.03)‡
Age
≤49	83	1.27	28	2.82	2.23 (1.96 to 2.53)‡	2.07 (1.34 to 3.18)‡	12	1.83	1.45 (1.21 to 1.73)‡	1.29 (0.69 to 2.41)
50–65	138	2.83	22	4.10	1.45 (1.19 to 1.76)‡	1.43 (0.91 to 2.24)	145	22.1	7.78 (7.04 to 8.61)‡	5.44 (4.21 to 7.03)‡
≥65	193	6.92	4	1.62	0.23 (0.14 to 0.40)‡	0.21 (0.08 to 0.56)‡	115	24.2	3.50 (3.09 to 3.96)‡	3.13 (2.46 to 3.99)‡
Comorbidity
No	133	1.51	26	2.42	1.61 (1.42 to 1.82)‡	1.78 (1.17 to 2.72)§	13	2.01	1.33 (1.12 to 1.58)‡	1.49 (0.84 to 2.64)
Yes	281	5.20	28	3.98	0.77 (0.63 to 0.94)§	0.82 (0.55 to 1.21)	259	22.7	4.36 (4.00 to 4.76)‡	4.48 (3.78 to 5.30)‡

*Incidence rate per 1000 person-years.

†Multiple analyses including age, sex, and comorbidities of hypertension, diabetes, hyperlipidaemia, coronary artery disease, heart failure, cancer, stroke, and lower leg fracture or surgery.

‡P<0.001.

§P<0.01.

IRR, incidence rate ratio.

HRs of late and early H. pylori eradication

Table 3 shows that the risk of VTE was significantly higher in the late H. pylori eradication cohort than in the early H. pylori eradication cohort (adjusted HR 3.56, 95% CI 2.63 to 4.81). The risk of VTE was significantly greater for both genders, and women had a significantly higher risk compared with men. The age-specific adjusted HR of VTE in the late H. pylori eradication cohort compared with the early H. pylori eradication cohort was not significant for the younger group (adjusted HR 0.84; 95% CI 0.43 to 1.67). Without comorbidities, the difference of adjusted HR between the late H. pylori eradication cohort and the early eradication cohort was not significant (0.83, 95% CI 0.42 to 1.62).

Table 3

HRs of venous thromboembolism between early and later Helicobacter pylori eradication

	Early eradication		Late eradication
	IRR (95% CI)	Adjusted HR*	IRR (95% CI)	Adjusted HR*
		(95% CI)		(95% CI)
All	1 (reference)	1 (reference)	5.00 (4.25 to 5.90)†	3.56 (2.63 to 4.81)†
Gender
Female	1 (reference)	1 (reference)	6.52 (4.98 to 8.54)†	4.05 (2.57 to 6.38)†
Male	1 (reference)	1 (reference)	3.57 (2.91 to 4.38)†	2.83 (1.86 to 4.31)†
Age
≤49	1 (reference)	1 (reference)	0.65 (0.51 to 0.82)†	0.84 (0.43 to 1.67)
50–65	1 (reference)	1 (reference)	5.37 (4.03 to 7.17)†	3.50 (2.18 to 5.60)†
≥65	1 (reference)	1 (reference)	14.9 (7.77 to 28.7)†	16.5 (6.05 to 45.3)†
Comorbidity
No	1 (reference)	1 (reference)	0.83 (0.66 to 1.05)	0.83 (0.42 to 1.62)
Yes	1 (reference)	1 (reference)	5.70 (4.41 to 7.37)†	5.38 (3.64 to 7.97)†

*Multiple analyses including age, sex, and comorbidities of hypertension, diabetes, hyperlipidaemia, coronary artery disease, heart failure, cancer, stroke, and lower leg fracture or surgery.

†P<0.00.

IRR, incidence rate ratio.

Incidences and adjusted HRs of DVT and PE, according to follow-up period, in the H. pylori eradication patients compared with the control patients

Furthermore, we compared incidence densities and HRs of VTE among three cohorts by follow-up duration (table 4). The incidence of VTE was greater in the late H. pylori eradication cohort when follow-up duration ≤5 years (25.4 per 1000 person-years). The highest risk occurred during the >5 years of follow-up period (adjusted HR 7.92, 95% CI 4.27 to 14.7) in the late H. pylori eradication cohort compared with the early eradication cohort.

Table 4

Trends of venous thromboembolism risks by stratified follow-up years

	Control(N=26 944)		Early eradication(N=3062)		IRR(95% CI)	Adjusted HR†	Late eradication (N=3674)		IRR(95% CI)	Adjusted HR†
Follow-up time, years	Case	Rate*	Case	Rate*		(95% CI)	Case	Rate*		(95% CI)
≤5	216	2.08	42	3.41	1.64 (1.46 to 1.84)‡	1.92 (1.38 to 2.68)‡	165	12.1	5.81 (5.41 to 6.24)‡	4.21 (3.41 to 5.20)‡
>5	198	5.19	12	2.21	0.43 (0.33 to 0.54)‡	0.44 (0.24 to 0.78)‡	107	25.4	4.90 (4.45 to 5.40)‡	3.45 (2.69 to 4.43)‡
≤5					1 (reference)	1 (reference)			3.54 (3.00 to 4.18)‡	2.41 (1.70 to 3.41)‡
>5					1 (reference)	1 (reference)			11.5 (8.43 to 15.8)‡	7.92 (4.27 to 14.7)‡

*Incidence rate per 1000 person-years.

†Multiple analyses including age, sex, and comorbidities of hypertension, diabetes, hyperlipidaemia, coronary artery disease, heart failure, cancer, stroke, and lower leg fracture or surgery.

‡P<0.001.

IRR, incidence rate ratio.

Discussion

Our results showed that the timing of H. pylori eradication significantly affected risk of VTE. There was a significantly greater risk of late H. pylori eradication for VTE for female gender, age over 50 years and with comorbidities.

It is possible that H. pylori initiates an inflammatory process in the stomach and may even generate a systemic response throughout the host. H. pylori eradication therapy could effectively prevent the progression of the pathological process in the gastric mucosa. The patient may also experience systemic effects, beyond the stomach, from H. pylori eradication. The association between H. pylori and VTE may not be strong, which would explain why our results showed no significant difference in risk of VTE between the early H. pylori eradication and control cohorts. However, late H. pylori eradication, particularly with comorbidities, showed a significantly increased risk of VTE compared with early H. pylori eradication. The timely eradication of the infection would likely lead to better health status in an H. pylori-infected population, not only with respect to prevention of peptic ulcers and gastric cancer, but also in terms of risk of VTE, as shown in the present study.

H. pylori may be an independent risk factor for atherothrombotic events in animals and humans.24 25 Infection with H. pylori did not appear to be associated with DVT in a specific disease.26 H. pylori eradication therapy could cause more extensive beneficial effects, regardless of any relationship with H. pylori. Eradicating H. pylori infection is currently the standard of care, and could ultimately cause more good than harm. Beyond peptic ulcer recurrence and cancer occurrence, H. pylori eradication could cause systemic effects, such as metabolic and inflammatory processes.14 15 While the association of H. pylori and clinical conditions is not directly correlated, H. pylori eradication may cause systemic changes within the host that are not immediately apparent. Previous studies were primarily concerned with how H. pylori could affect systems within the body, including cardiovascular diseases, neurological disorders, diabetes mellitus, ear and eye diseases, immunological and haematological disorders, liver and bile tract diseases, and gynaecological and respiratory tract diseases.27 28 Prophylactic eradication therapy is thus considered the standard of care even for extra-gastric diseases. Our results demonstrated that H. pylori eradication treatment was associated with VTE, and delayed H. pylori eradication was a risk for VTE. H. pylori infection may not directly cause VTE events, but late H. pylori eradication could be a risk factor for VTE. The processes underlying this phenomenon are unclear, but several possible mechanisms have been suggested. H. pylori eradication therapy could have beneficial systemic effects. In addition, a recent study reported that H. pylori eradication could lead to changes in the gut microbiome.29

In the present study, there was no significant difference in risk of VTE between the early H. pylori eradication and control cohort. Early H. pylori eradication is associated with decreased risk of VTE especially in the aged patients more than 65 years. After 8 weeks of H. pylori eradication, some metabolic benefits on improvement of insulin resistance and lipid profile were reported. Improved metabolic parameter may decrease the risk of VTE. Thus, early eradication of H. pylori is recommended.

In the present study, our findings demonstrated differences in VTE risk after early or late H. pylori eradication. However, there were limitations in this study. First, the compliance of H. pylori eradication for each patient in this study was unknown. We can therefore only assume that patients receiving a prescribed H. pylori eradication regimen were fully compliant and completed the course of medications. It is known that the treatment regimen for H. pylori eradication may cause mild or severe adverse effects, and therefore, some patients may not have completed the course of treatment. Second, the dataset used in this study did not contain information on the success or failure of H. pylori eradication. As such, it was only possible to estimate the general success rate of H. pylori eradication. Third, the adherence of post-treatment follow-up and the re-infection rate could not be demonstrated in the database. Finally, the database could not provide the information about usage of other medications, such as hormone therapy or anticoagulant. Those medications might affect the occurrence of VTE.

Besides gastric diseases and cancers, H. pylori infection could cause non-gastric systemic effects in humans. H. pylori eradication is currently the standard of treatment for preventing recurrence of peptic ulcer diseases and occurrence of cancer. There may be unexpected effects resulting from H. pylori eradication. Further study is needed to gain a more comprehensive understanding of the systemic insults and molecular mechanisms involved in H. pylori eradication.

Conclusions

Late H. pylori eradication was associated with a significantly increased risk of VTE. The risk of VTE was significantly greater in patients with female gender, age more than 50 years and with comorbidities. Therefore, we recommended administering H. pylori eradication therapy early.