The effect of statins on the occurrence of peptic ulcer.

Original Article

The effect of statins on the occurrence of peptic ulcer

Anning Feng

, Eric Chuang

, Szu-HsienWu

, Jia-ChiWang

, Shih-Ni Chang

,

Cheng-Li Lin

, Chia-Hung Kao

⁎

1. Introduction

Conglomerated and known as “statin,” the hydroxymethylglutaryl coenzymeA (HMG-CoA) reductase inhibitors act to obstruct the synthesis

of cholesterol in the liver [1,2]. Since mevalonic acid, the end product

of said enzymatic reaction, is one of the foundational baseline substances

to various isoprenoid metabolites [1,3], it could potentially

lead to pleiotrophic effects of the aforementioned drugs. Also, growing evidence reveals that some other benefcial infuence of these agents may have a signifcant effect on lowering plasma cholesterol levels

[4–8]. As it appears, the positive effect surpasses the cholesterollowering result of statins through its protective effects on atherosclerotic

plague stabilization and endothelial nitric oxide bioactivity [9,10].

Moreover, evidences point to statins affecting angiogenesis therapeutically; endothelial functions and angiogenesis are impaired by hypercholesterolemia, therefore, by reducing cholesterol and strengthening

endothelial functions, statins could potentially ameliorate angiogenesis

[11].

Peptic ulcer disease (PUD) is one of themost prevalent causes of gastrointestinal bleeding,which leads to hospitalization and/ormortalities.

It is known that nonsteroidal anti-infammatory drugs (NSAIDs) and Helicobactor pylori (H. pylori) are what cause peptic ulcers. While the cases of H. pylori have been decreasing over the last decade due to improved cleanliness and eradication therapy, the rate of PUD has not decreased dramatically in some countries. This leads us to conclude that the usage of NSAIDs and aspirin in the elderly, then, is the reason

why PUD is still common [12–14]. Despite that, an increase in statin related side effects such as

gastrointestinal

(GI) symptoms and GI hemorrhage has been reported

[15–17]. Contrarily, much research on gastroprotective effects and the preservation of vascular integrity are on animal subjects, and it appears that statin use actually lowers the rate of peptic ulcer development

when administering antithrombotic treatment and NSAID to a patient

[15,26,27]. Consequently, it is unknown how relevant it is to human

models. Though experiments suggest that an imbalance between gastrointestinal toxic agents and protective mechanisms provokes infammation

leading to peptic mucosa injury [24,25]. Our study was one of

the leads to investigate the effects of statin on the occurrence of PUD. Sincemany patients are treatedwith statins, and because epidemiological evidence for a connection between statin use and the risk of PUD is minimal, we committed ourselves to the study in Taiwan to deduce whether statin use can reduce PUD incidences, and to ascertain the dosage-dependent effects of statins.

2. Methods 2.1. Source of data

In March, 1995, the Taiwanese government commenced its staterunNational Health Insurance (NHI) program,which provides universal

health insurance coverage to 99% of Taiwan's entire population and has contractedwith 97% of the hospitals. For research purpose, the National Health Research Institute (NHRI) compiled all medical claims in NHI programand released the database annually to the public. The National Health Insurance Research Database (NHIRD) contains medical information including inpatient and outpatient care facilities, drug prescriptions, insurant's sex and the date of birth, date of visit or hospitalization, and diagnoses coded in the ICD-9-CM format (International Classifcation of Disease, 9th Revision, Clinical Modifcation).

The NHRI established several randomly selected claim databases representative of the whole population. This cohort study used the randomly selected insurance claim data of one million insurants. Sets of information available for the database cover all medical services received by each enrollee from 1996 to 2011.

The NHIRD encrypts the patients' personal information for privacy protection and provides researchers with anonymous identifcation numbers associated with the relevant claim information, which includes the patient's sex, date of birth, registry of medical services, and medication prescriptions. Patient's consent is not required for accessing the NHIRD. This study was approved by the Institutional Review Board of China Medical University in central Taiwan (CMU-REC-101-012). 2.2. Study population

(ICD-9-CM 272.xx) during the period of 1998 to 2011 and who were divided into two groups based on their use of statin. The statin cohort was defned as patients who underwent statin treatment for at least 90 days. The index date for each patent was the frst treatment date of statin. Patients were excluded if they were younger than 20 years or with peptic ulcer diseases (ICD-9-CM 531.xx-535.xx) history before the index date. The non-statin cohort (without statin treatment) were 1:1 frequency matched with sex, age, year of diagnosis of hyperlipidemia and index-year to the statin cohort. The same exclusion criteria

were also applied to non-statin cohort.

2.3. Outcome measurement, comorbidities and medications We identifed the frst diagnosis of PUD from outpatient claims

and/or hospitalization records from 1998 to 2011 as the study endpoint. All of the study subjects were followed from the index date

to occurrence of endpoint, withdrawal from the database, the end of 2011, whichever date came frst.

We also incorporated inpatient and outpatient diagnosis records to ascertain the baseline comorbidities, including hypertension (ICD-9-CM 401-405), stroke (ICD-9-CM 430-438), coronary artery disease (ICD-9-CM 410-414), and end stage renal disease (ESRD) (ICD-9-CM 585). In addition, concomitant uses of NSAIDs and/or aspirin were also analyzed.

2.4. Statistical analysis

The Chi-square test and Mann–Whitney U test were used for comparison between the statin and non-statin cohorts. The incidence density rate (IR) stratifed by baseline characteristics was estimated in both cohorts. The relative risk of patients with and without statin treatment

on the occurrence of PUD was analyzed using the univariable and multivariable Cox proportional hazards regression model. Themultivariable

models were adjusted for age, gender, and comorbidities of hypertension, stroke, coronary artery disease, end stage renal disease, and the

use of aspirin and/or NSAID. The hazard ratio (HR) and 95% confdence

interval (CI) were estimated by Cox model. The defned daily dose, recommended by the World Health Organization, is the assumed average

maintenance dose per day per drug. The formula of cumulative defned daily dose (DDD) equals to the sums of defned daily dose of any statin in the duration of follow-up (years) of patients.We classifed the cumulative DDDof statin use by the third quartile (b575, ≧575 DDD). Further,

we estimated the dose–response of statin use on the risk of developing PUD based on cumulative DDD. A two-tailed p value of b0.05 was considered statistically signifcant. All analyses were performed with SAS

statistical software (version 9.2 for Windows; SAS Institute, Inc., Cary, NC, USA).

3. Results

The baseline characteristics of the patients in the two cohorts are

presented in Table 1. Of the statin cohort, 54% were men and 46%

werewomen. Themedian age of the statin cohort and non-statin cohort were 55.2 (range=20.2–99.9) and 55.6 (range=20.0–99.0) years, respectively. No signifcant differences were noted in sex (p= 0.30) and

age (p = 0.36). Statin cohort tended to have hypertension (65.4% vs. 38.2%, p b 0.0001), stroke (15.4% vs. 8.25%, p b 0.0001), coronary artery disease (26.0% vs. 13.0%, p b 0.0001), end stage renal disease (0.60% vs. 0.20%, p b 0.0001), and accompanying use of aspirin (36.9% vs. 23.0%,

p b 0.0001) or NSAID (27.8% vs. 23.4%, p b 0.0001). Among 24,423 patients who underwent statin treatment at entry,

4145 developed peptic ulcer disease. The crude incidence of PUD was

33.5 per 1000 person-years (Table 2). Among non-statin cohort, 5026

developed PUD, with a crude incidence of 45.0 per 1000 person-years. Statin cohort had a signifcantly lower incidence of PUD than nonstatin cohort with an adjusted HR of 0.73 (95% CI= 0.69–0.75). Stratifed for sex, the statin cohort was found to be associated with a decreased risk of PUD in female (adjusted HR = 0.72, 95% CI = 0.68–

0.77) and male patients (adjusted HR = 0.72, 95% CI = 0.68–0.76). The incidence of PUD increased with age in both cohorts and female had a higher occurrence rate than male in both cohorts. Compared with the non-statin cohort, the statin cohort was associated with a signifcant lower risk of PUD for all age group. The concomitant use of aspirin and/or NSAIDs had higher incidence of PUD than those without in both cohorts.

The effects of cumulative DDD of statin on PUD showed a dose–

response relationship in Table 3. Compared with non-statin users,

the adjusted HR decreased from 0.96 (95% CI = 0.92–1.01) for those had taken b575 DDD of statin to 0.21 (95% CI = 0.19–0.24) for those had taken ≧575 DDD of statin. Compared with the nonstatin cohort, the statin cohort had a signifcantly lower risk of PUD

(Table 4). 4. Discussion

Other than its known lipid lowering effects, statins are still in the testing stage in other medical felds. Despite this, there is an increasing amount of potential clinical applications underway in multiple pathological

processes, including stroke protection [27,28], Alzheimer's

disease [29,30], and rheumatoid arthritis [31,32], that have been effectively

using statins. Thus far, statins are recognized for their potential gastroprotective effects through inhibition of neutrophil activity, and the reduction of oxidative stress and maintenance of vascular integrity. There are four major fndings in this study: (1) patients treatedwith

statin had a signifcantly higher incidence of comorbidities such as hypertension, coronary artery disease, stroke, end stage renal disease,

and accompanying use of aspirin and NSAIDs than patients not treated with statin; (2) the incidence of PUD increased with age and female had a higher occurrence rate than male in both cohorts. The aspirin and/or NSAIDs users had higher incidence of PUD than those without in both cohorts; (3) compared with the non-statin cohort, the statin cohort had signifcant lower risk of PUD at different locations of alimentary tract except gastric ulcer regardless of variables in age and gender; and (4) the effects of statin were dosage-dependent.

Substantial research has shown that statin treatment can effectively

prevent cardiovascular disease [2–4] and stroke [27,28]. As a result,

numerous patients receiving statin,NSAIDs, and low-dose aspirin are elderly, or have comorbid conditions, or both. However, as anticipated,

the patients treated with statin usually have statistically more comorbidities than patients not treated with statin.

The pathogenesis of peptic ulcers is multifaceted. Although many

patientswho had PUD bore H. pylori infection, a large majority of the patients in Taiwan do not agree to endoscopic biopsy for histopathological

diagnosis or there is limited access to endoscopic services in some

clinics. Because of this, many patients with PUD are treated empirically [33]. In the last decade

though, the causes of PUD have changed from

H. pylori to NSAIDs or idiopathic ulcers. This is due to the high population of elderly patients because of their high usage of NSAIDS and lowdose aspirin for relief of intense musculoskeletal disorders and prevention of arteriosclerotic disease. This explains (1) the increased frequency of PUD after age 60, and (2) the unchanged hazards ratio in gastric ulcer

after treatmentwith statin, since patients were older and more likely to

have past usage of NSAIDs and aspirin [12,13,15,34,35].

It is thought that suffcient gastrointestinal mucosal blood fow is important for inhibiting the back-diffusion of gastric acid andmaintaining

gastrointestinal mucosal integrity [25]. Nonetheless, in vivo data is

still insuffcient regarding peptic ulcers' reactions ensuing the excessive generation of oxygen-derived free radicals and pro-infammatory mediators

[18–20]. In spite of that, other evidences present that statins may

actually decrease the acidity and volume of gastric secretions, furthermore, by way of optimal regulation of NO production, prostaglandins

and other endogenous chemicalmediators inmucosa, statins play an essential role in modulating gastrointestinal circulation in recovery from

peptic ulcers [3,21–23,25]. The fndings in this study support the

hypothesis—signifcant dosage-dependent effects in high and low DDD were noted concerning the rate of occurrence of peptic ulcers. This study was subject to limitations, which must be mentioned. First, the NHIRD does not provide detailed information on patients on factors such as their lifestyle, habits, body mass index (BMI), physical activity, socioeconomic status, and family history. All of these are possible confounding factors in this study. Second, the evidence derived from an observation study is generally of lower methodological quality than that from randomized trials because a cohort study design is subject to many biases related to the necessary adjustments for confounding factors. Despite our meticulous study design with an adequate control of the confounding factors, a key limitation is that biases could remain because of possibly unmeasured or unknown confounders. Third, the registries in the NHI claims are primarily used for administrative billing and are not verifed for scientifc purposes. Moreover, we were unable to contact the patients directly to obtain additional information on

their real use of drugs because of the anonymity assured by the identifcation numbers. However, the data on the diagnoses of PUD and the

orders for statins' uses were highly reliable. Finally, additional large population-based unbiased randomized trials are required because it is essential to confrm our current fndings before drawing any frm conclusions.

5. Conclusion

The results of the present study indicated that statinsmight be associated with the protection of peptic ulcer in a dose-respondent manner.