Safety assessment of omeprazole use: a review

ABSTRACT BACKGROUND: Risks regarding hospital admission due to adverse drug reactions and drug interactions from use of omeprazole have been reported. The question guiding the present review was “Which adverse events occur in patients using omeprazole in a Food and Drug Administration-approved and/or off-label manner?” It was also proposed to evaluate the safety of use of omeprazole. DESIGN AND SETTING: Qualitative narrative review with critical evaluation, in a public university. METHODS: The PubMed, SCOPUS, LILACS, SciELO, EMBASE and EBSCO databases were searched on July 31, 2018. Studies evaluating adverse events were screened. RESULTS: 72 articles were included, among which 58 reported on adverse drug events (47, adverse drug reactions; 5, drug interactions; and 6, situations of ineffectiveness). 28 adverse drug reactions not described in compendia and drug leaflets were described in these studies: myocardial infarction (6); stroke (2); spontaneous abortion (1); proliferative changes (1); chills (1); heart failure (1); thrombosis (2); and dementia (1), among others. Severe adverse reactions, for instance cardiac problems, Steven-Johnson syndrome and proliferative changes, were identified. The antiplatelet effects of drugs such as clopidogrel, in patients who underwent heart-related surgery, increased the risk of developing cardiac problems, such as cardiovascular death, myocardial infarction and stroke. In newly transplanted patients, decreased absorption of mycophenolate mofetil occurred, thus leading to rejection of transplanted organs. CONCLUSION: Use of omeprazole should be monitored primarily in patients with heart disorders using antiplatelet agents concomitantly, and in newly transplanted patients using mycophenolic acid, in order to avoid serious adverse reactions.

[27] These off-label indications are for long-term use and are widespread and commonly prescribed in some countries, 28 such as Brazil.

OBJECTIVE
The purpose of this review was to evaluate the adverse outcomes relating to omeprazole use in clinical practice.

Study design
We conducted a qualitative narrative review with critical evaluation, to answer the following guiding question: "Which adverse events occur in patients using omeprazole in a Food and Drug Administration (FDA)-approved and/or off-label manner?"Thus, we aimed to gather, organize and critically review articles on these topics, to include the highest level of scientific evidence.

Search of the literature and inclusion criteria
The search for studies was performed using the MEDLINE (via PubMed), LILACS, EMBASE (via Ovid), SciELO and SCOPUS databases and was conducted on July 31, 2018.During the search and selection process, there was no limitation on the time when articles were published.The languages were restricted to Portuguese, English and Spanish.
The following search strategies were used: ("Omeprazole" OR "Proton Pump Inhibitors") AND ("Adverse Drug Reaction Reporting Systems" OR "Pharmacovigilance" OR "Drug-Related Side Effects and Adverse Reactions" OR "Risk Assessment" OR "Treatment Outcome" OR "Off-Label Use").All descriptors used in these search strategies are Medical Subject Headings (MeSH terms).We included randomized clinical trials, phases I and II clinical trials, case-control studies, cohort studies, cross-sectional and quasi-experimental studies (clinical trials in which there was no comparator group for the intervention) evaluating adverse events from therapeutic or prophylactic use of omeprazole among individuals in all age groups whose health status was well defined and who were using omeprazole in an FDA-approved and/or off-label manner.
We excluded review articles, dissertations and theses, case reports, abstracts published in annals of events, editorials, letters to the editor, news and comments.

Types of participant
The target population comprised patients of any kind whose health status was well defined and who were using omeprazole in an FDA-approved and/or off-label manner.There was no age limitation.

Types of intervention
The interventions considered comprised use of omeprazole from the outset of treatment to clinical outcome, without restrictions on doses, therapeutic regimens or duration of use.In addition, it was proposed to include both preventive use and therapeutic use.

Types of outcome
The outcomes considered comprised any safety-related outcome, including adverse events, withdrawal due to adverse events, mortality and therapeutic ineffectiveness, i.e. adverse events in which the medicine used did not present any therapeutic response or its therapeutic response was lower than expected.Safety-related outcomes of all causes and omeprazole-related causes were considered.
After selecting potential articles in the databases, the titles and abstracts were reviewed by verifying patient exposure to omeprazole.The following variables were defined during the screening of articles: indication of use; study design; patient's clinical condition; clinical outcomes, including all types of adverse events relating to use of omeprazole; recommendations; author; and year of publication.
The severity of adverse events was classified as described by the World Health Organization.In this definition, severe adverse reactions are harmful effects that occur during drug treatment and which can result in death, be life-threatening or lead to persistent or significant disability, congenital anomaly, clinically important effects, hospitalization or prolongation of hospitalization.Non-serious adverse reactions also fall within the concept of severe adverse reactions. 29e search for studies, selection of studies and extraction of data were performed by three authors, in triplicate independently, to avoid the presence of bias in the selection and exclusion of articles.In addition, the kappa function was applied to analyze the agreement rate.

Risk of bias assessment
For randomized clinical trials, risk of bias was evaluated using the Cochrane collaboration tool (RoB 1), 30 which is based on seven domains: random sequence generation, concealment of allocation, blinding of participants and professionals, blinding of outcome assessors, outcome completeness, selective reporting of outcomes and other sources of bias.Each domain is judged as presenting low risk of bias, uncertain risk of bias or high risk of bias.
For case-control and cohort studies, we used the Newcastle-Ottawa tool.This provides evaluations in three domains: selection, comparability and outcome for cohort studies; and selection, comparability and exposure for case-control studies.Each item that is identified as presenting low risk of bias is given a "'star".There is a maximum of one "star" for each item within the "selection" and "exposure/outcome" categories; and a maximum of two "stars" for "comparability".Therefore, each study can be classified with a maximum of nine "stars", which corresponds to a low risk of bias. 31e cross-sectional and quasi-experimental studies included in this review were not evaluated, since there are no validated tools for analysis on these study designs.

RESULTS
A total of 5,500 potentially relevant studies were identified.After reading the titles and/or abstracts, 4,746 studies were excluded because they did not meet the inclusion criteria.Another 218 were duplicates, and thus 536 studies were examined further.
It was not possible to access 2 of these 546 studies, because one of them is no longer indexed in the database and the other does not provide for the option to purchase and access the article.Our attempts to contact the authors of these two studies were unsuccessful.After screening the remaining articles, 191 studies were found to be eligible for complete text reading.After reading in full, 119 were excluded because they did not meet the inclusion criteria.Thus, 72 articles were considered eligible for the safety assessment on use of omeprazole, since they included all the variables that were being analyzed (Figure 1).
The proportion of overall agreement (kappa) observed in relation to making final decisions (inclusion and exclusion) from the database that included the screened articles was 0.807 (confidence interval, CI: 0.658-0.957).
Among these 72 studies, 58 reported on adverse drug events (ADEs): 47 studies on adverse drug reactions (ADRs), 5 studies on drug interactions (DIs) and 6 studies on therapeutic ineffectiveness (Table 1). 13,14,32-48, 60, 65-67The duration of use of omeprazole ranged from 5 days to 11 years in these studies.Only one study evaluated the off-label use of omeprazole. 41relationship was observed between use of omeprazole and increased risk of severe adverse events, such as development of coronary disorders that might lead to death. 32,39,43,62,88,94,96,101garding the clinical outcomes of the studies, the safety (ADRs and DIs) and the therapeutic ineffectiveness can be correlated.
1][42] Among the 40 clinical trials included in the review, after riskof-bias analysis, it was found that eight were classified as presenting low risk of bias, 14 as having high risk of bias and 17 as having uncertain risk of bias.][39]

DISCUSSION
This review allowed us to identify and update the most severe and prevalent ADEs relating to use of omeprazole, and our findings corroborate similar results found in other studies. 3,4Severe ADEs occurred in patients who underwent heart-related surgery or drug interventions, such as in situations of acute coronary syndromes or percutaneous coronary intervention, 75,78,96 or in cases of concomitant use of such medications. 76These events were associated with concomitant use of omeprazole and clopidogrel, which caused inhibition of the antiplatelet effect of omeprazole, 83 due to competitive inhibition of CYP2C19. 32][34][35][36][37] Uncommon reaction (≥ 0.1% and < 1%) Eczematous eruption, insomnia, somnolence, urticaria, urticaria vasculitis and vertigo. 38re reaction (≥ 0.01% and < 0.1%) Angioedema, arthralgia, muscle pain, erythema multiforme, weakness, metallic taste in the mouth, allergic reaction, Steven-Johnson's syndrome and thirst. 38

Post-marketing experience
Unstable angina, increased risk of fractures, cancer, cystitis, ulcerative colitis, stomatitis, abnormal renal function, hypergastrinemia, decreased levels of vitamin B12, increased creatinine levels, hypomagnesemia.Omeprazole and clopidogrel: cardiovascular death, myocardial infarction, inhibition of the effect of clopidogrel, increased leukocyte and platelet levels and increased brain adverse events. 75,76meprazole and acenocoumarol: increased anticoagulant effect of acenocoumarol. 77meprazole and mycophenolate mofetil: reduced absorption of mycophenolic acid. 49ficacy TI (n = 5) Some patients did not respond to omeprazole therapy and continued with colitis symptoms and gastrointestinal discomforts.Omeprazole failed to control the gastric acidity of some patients. 13ble 1.Frequencies of adverse events resulting from indication of omeprazole that were reported in studies published up to 2016 ADE = adverse drug event; ADR = adverse drug reaction; DI = drug interaction; TI = therapeutic ineffectiveness.The frequency of adverse reactions was classified according to the leaflet of the reference drug product, except for the 28 studies for which there was no information on the leaflet.

Cross-sectional studies (2)
Newborns with hypospadias born to mothers who had used proton-pump inhibitors during pregnancy (430,569) Erichsen et al., 2014 98 Patients with stage 5 chronic kidney disease (CKD) on hemodialysis therapy and chronic use of proton pump inhibitors (37)   Several drug interactions relating to omeprazole, especially with antiplatelet agents, are known. 78,94The non-serious events that have been described are diarrhea, headache and somnolence, relating to use of omeprazole concomitantly with the antiretroviral drug etravirine. 32The severe adverse events that have been described comprise inhibition of the antiplatelet effects of drugs such as clopidogrel, which increases the risk of developing heart problems that may lead to death; and decreased absorption of mycophenolic acid, which leads to rejection of transplanted organs. 49[96] In two studies in which omeprazole was added to dual antiplatelet therapy (a combination of clopidogrel and acetylsalicylic acid), it reduced the stomach pain resulting from this therapy and no risk was found in this combination. 95Nonetheless, it is always necessary to monitor potentially dangerous drug combinations between omeprazole and clopidogrel, acetylsalicylic acid and mycophenolate mofetil, among others.
Regarding drug interactions, all patients may be exposed to their effects, regardless of age or clinical condition.However, some patients are more susceptible, such as those who already have some type of heart disease or the elderly, who commonly use polypharmacy.
Only 12 studies included elderly patients, and these studies reported occurrences of severe adverse events such as dementia, myocardial infarction, cardiovascular death, stroke and pulmonary embolism, among others.In the non-elderly population, the severe adverse events reported included myocardial infarction, stroke, death and pulmonary embolism, but no relationship between the severity or the frequency of events and the patients' age was observed from use of omeprazole.However, other authors have suggested that age is a factor that influences occurrences of adverse events.Varallo et al. 24 observed in a cross-sectional study that the elderly population had fewer ADEs than adults did, probably because doctors provide greater care and attention regarding pharmacotherapeutic management for patients of this age group, since there are other factors that increase the likelihood of ADEs, such as polypharmacy.Beijer and de Blaey 100 reported that the chances that elderly individuals would need to be hospitalized due to adverse drug reactions (ADRs) were four times higher than those of younger people (16.6% versus 4.1%).Additionally, in 2015, the American Geriatrics Society advised through the Beers criteria that unjustified use of PPIs among the elderly for more than eight weeks should be avoided, since exposure to such drugs increases the risks of infection by Clostridium difficile, bone loss and fractures. 13,16,101other factor that may have influenced the appearance of adverse events is the duration of use of omeprazole.34][35][36]74 Severe adverse events have been reported among patients who used omeprazole for longer times, i.e. more than one month. 35-37, 54,57,60,63,65-67,70,71,74-76,70,95, only one of the studies analyzed here was omeprazole prescribed for off-label use. 41However, off-label prescription of omeprazole is widespread in many countries and there is a need to assess the safety of this use.We take the view that the duration of exposure is likely to increase the likelihood of adverse events, since polypharmacy alone is a risk factor for occurrences of adverse events. 24tcomes of therapeutic ineffectiveness and symptom worsening were identified.It was noted that some patients did not respond to omeprazole treatment 13,32,64,70 and that for others, their symptoms worsened. 62The most likely reason for such events would be high concentrations of acid in the stomach, which could cause gastroparesis, decrease absorption and, consequently, decrease the therapeutic effect of omeprazole.
7]102 Because the associations between these adverse events and use of omeprazole are not fully understood, there is a need to carry out further studies to investigate the relationships between omeprazole and these events.If such associations are verified, they should be described in the package leaflet.
In addition, more recent studies have identified other adverse events, such as decreased vitamin B12 levels, 40 increased levels of creatinine 41 and hypomagnesia. 42e of omeprazole is considered safe in the following situations: when it is not combined with antiplatelet drugs; when it is administered to replace H2 receptor antagonists in patients who are resistant to treatment with drugs of this class; when the most appropriate posology and dosage is established for each condition and patient; and when omeprazole is used in conjunction with a combination of antibiotics to eradicate H. pylori and to treat esophagitis, among other situations. 94,102mitations of the present study No a priori design was provided for this review and the languages were restricted to Portuguese, English and Spanish.
Gray literature was not included.However, its inclusion would be unviable and probably would not add to the results found, since this type of literature is characterized by incomplete and poorly constructed data.
No methods were used to assess the homogeneity or heterogeneity between the studies, and the risk of publication bias among the studies included was not assessed.Furthermore, no information regarding potential conflicts of interest in the primary studies included was available.
All the outcomes evaluated related to approved indications for use of omeprazole.Therefore, the data confirm that there is no evidence of clinical outcomes (safety and effectiveness) resulting from unapproved use of omeprazole, such as polypharmacy (although polypharmacy is commonly used).The duration of use of omeprazole influenced occurrences of adverse events.Severe adverse events, such as death, stroke and myocardial infarctions occurred during prolonged treatments (more than one month).Non-serious adverse events occurred over short periods (from a few days to a maximum of two weeks).Use of omeprazole needs to be monitored primarily in patients with heart disorders who are using antiplatelet agents and omeprazole concomitantly and in newly transplanted patients who are using mycophenolic acid as a suppressive agent, in order to avoid severe adverse reactions such as organ transplant rejection, death, stroke and myocardial infarction.

CONCLUSION
Therefore, use of omeprazole can be considered safe in the following situations: when it is not combined with antiplatelet drugs; when it is administered to replace H2 receptor antagonists in patients who are resistant to treatment with drugs of this class; when the posology is well established for each condition and type of patient; and when omeprazole is used to eradicate H. pylori, among others.
Most of the trials included in this review presented uncertain risk or high risk of bias, which indicates that there is a need for betterdesigned studies.The high risk of bias related mainly to the blinding of the participants and outcome assessors.It should be noted that if patients and professionals believe that omeprazole is a gastric protector and is risk-free, this may lead to bias in the analysis and to under identification and underreporting of adverse events relating to omeprazole.This may suggest that the existing studies may have underestimated the adverse events.

Figure 1 .
Figure 1.Flowchart of the stages of selection, skim-reading and full reading of the eligible articles.

Table 2 .
Adverse events from approved use of omeprazole that were reported in the studies analyzed, published from 1994 to July 2018 ADR: Diarrhea, nausea, headache, cold, vomiting and fever Patients with gastroesophageal reflux without erosive esophagitis (359) Richter et al., 2000 59 Continue... ADR: Increased risk of fibrosis progression, cirrhosis, hepatic decompensation and development of hepatocellular carcinoma Patients with hepatitis C virus (HCV) infection.Li et al., 2018 48

Table 3 .
Assessment of risk of bias in clinical trials using the RoB 1.0 tool and evaluation of quality of cohort and control case studies using the Newcastle-Ottawa scale