Efficacy and Safety of Omeprazole for the Treatment of Acid Peptic Disorders: A Systematic Review and Meta-Analysis

,

Standard treatments for APD include the use of PPIs, adjunctive treatments (histamine H2 receptor antagonists, prokinetics, and alginate), surgery, life-style changes, and dietary considerations [1,11,12].Te World Gastroenterology Organization, Chinese, Korean, and Japanese guidelines recommend the frst line of treatment to be a PPI given over 4-6 weeks, but there is no consensus of which type of PPI is more efective [1,[13][14][15][16][17][18][19][20].PPIs difer in their pKa, bioavailability, peak plasma levels, route of excretion, recommended doses, and level of efcacy [9,21].Te choice of which individual PPI drug is more efective and safe is still controversial.In addition, whether the efcacy of PPIs difers for the diferent APD syndromes is rarely directly compared.
Our aim in this study is to identify and analyze data from randomized, controlled trials (RCTs) to determine efcacy and safety for the most commonly used PPI (omeprazole) compared to the other types of PPIs or placebo and to determine the efcacy separately for NERD, EO, and erosive ulcers.

Protocol and Registration.
Te project and protocol for this meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines [22].Te PRISMA checklist is provided in Supporting Information Table 1.Te project and protocol were prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO CRD 415397 (April 7, 2023), https://www.crd.york.ac.uk/PROSPERO/).

Search Strategy.
PubMed, Google Scholar, and the China National Knowledge Infrastructure (CNKI) databases were searched (from database inception to March 30, 2023) to identify prospective RCTs or cross-over trials comparing omeprazole to other types of PPIs or placebo.Te search strategy for PubMed was: ((((GERD) AND (omeprazole) AND (randomized controlled trial OR cross-over) AND (efcacy) AND NOT (prokinetics) OR NOT (alginate)))).Secondary searches of grey literature included reference lists, authors, reviews, meeting abstracts websites, and https:// clinicaltrials.govfor unpublished trials.Tere were no language restrictions and articles in languages other than English were translated and reviewed.

Study Selection.
Inclusion criteria included randomized, controlled clinical trials (RCTs) with prospective parallel groups or cross-over design with a minimum of two weeks washout period in children or adult subjects with diagnosis of acid/peptic disorder including either GERD, NERD, OE, upper gastrointestinal (GI) ulcers, GI bleeding, and presence or absence of H. pylori.Included interventions are as follows: omeprazole (oral or IV) for at least 4 weeks (later amended to at least one week) compared to other types of PPIs (esomeprazole (ESO), ilaprazole (ILA), lansoprazole (LAN), pantoprazole (PAN), rabeprazole (RAB), or placebo).Study outcome includes a measure of improvement of APD symptoms and/or ulcer/erosion healing.
Exclusion criteria included nonhuman studies, case reports or case series, early phase 1 (safety) or 2 (mechanism of action, dose ranging, formulation, kinetics) studies, validation of measurement tools for APD, no control group, intervention not well-described, no relevant outcomes provided, not a comparison of interest, reviews, metaanalysis, duplicate reports, presence of other disorders with similar symptoms (organic, metabolic, or druginduced, chronic cough or asthma, simple laryngitis, Zollinger-Ellison syndrome, primary motility disorder, esophageal stricture, Barrett's esophagus, upper GI malignancy or other severe comorbidity), PPI treatment less than 1 week, only non-PPI comparison group (prokinetics, H2 receptor antagonists, surgery, alginates, or potassiumcompetitive acid blockers), or did not contain original quantitative data.

Data Extraction.
Two reviewers (LV and PM) independently screened titles and abstracts of studies identifed by the search strategies.Data from all full-text articles were extracted and reviewed independently by two reviewers using a predesigned data extraction form following the standard methods for systematic reviews and meta-analysis [22,23].Any disagreements were discussed until resolved.
Te data extracted included PICO data: (1) population (age range and country), ( 2) intervention (type of PPI or controls used, daily doses, formulation, duration, and follow-up times), (3) comparisons (type of control group either placebo or open and unblinded), (4) outcomes, including improvement in APD symptoms, improvement in symptoms scores (frequency scale for symptoms of GERD (FSSG), dyspepsia symptom scores, heartburn scores, symptom index, etc.) and/or ulcer or erosion healing rates, time to ulcer healing, pH > 4 for 24 hours by treatment end, or percent remaining in remission.

Primary Outcomes
2.5.1.Improvement in APD Symptoms for NERD.Tis outcome was measured as either "overall improvement/ cure" and by improvement of specifc APD symptoms (heartburn, pain, and nausea).Other potential outcome measures included frequency remaining in remission, pH > 4 for 24 hours at the end of the treatment period, prepost improvement of esophageal pH, improvements in symptom scores (dyspepsia or heartburn and composite laryngeal score), or other visual analogue scales for symptom severity.[26].Bayesian random efects models were used for the meta-analyses if signifcant heterogeneity was detected (overall I 2 ≥ 50%); otherwise, fxed-efect models were used [27].Dichotomous outcomes were assessed using relative risks (RRs) and 95% confdence intervals (CI) and continuous outcomes were assessed using standardized mean diference (SMD) and 95% CI using standard methods [28].Outcomes were analyzed separately by the type of APD as follows: nonerosive syndrome (NERD) or erosive syndrome (EO or ulcers).Te signifcance level was set at p value ≤0.05.Heterogeneity across trials was evaluated using the I 2 statistic [26].To assess sources of heterogeneity or inconsistencies and their infuence on efcacy, the following data on potential confounding factors were collected: study design (double blinded or open), study quality, setting (inpatient or outpatient), and H. pylori status.For data missing from the published article, we attempted to contact the author.Publication bias was assessed using funnel plots and Egger's test [26].Subgroup analysis was used to explore sources of heterogeneity (geographic region, PPI dose, length of PPI treatment, and study quality), and assessed with the Cochrane Q test (X 2 test statistic) [28].Sequential sensitivity analysis was done to explore the extent outcomes were dependent upon a particular trial, but none were found.

Study Participant Characteristics.
Te characteristics of the trials and study participants are provided in Table 1.All trials were conducted in adults (range: 16-85 years old).Patients had erosive ulcers (n � 19, 61%) or NERD (n � 7, 23%), and only fve (16%) were in patients with EO.Most trials did not describe if the patients were outpatients or inpatients (n � 22, 71%), seven trials (23%) were done in outpatients, and two trials enrolled both inpatients and outpatients.
International Journal of Clinical Practice Most trials did not follow patients after the PPI study was stopped (n � 25, 81%), but six trials did follow patients for varied durations after the study PPI was discontinued as follows: 2-6 weeks [46,53], or 6 months [39,40,64], or 18 months [41].

Efcacy of Primary
Outcomes.Te most consistently reported outcome for improved APD symptoms was for heartburn relief.Improvements of other specifc symptoms (nausea, regurgitation, belching, bloating, and pain) did not have sufcient numbers of trials within PPI control groups to be analyzed.Only one trial reported the frequency of ulcer remissions [41] and only one trial reported pH levels over 24 hours during the trial [52].

Heartburn Relief.
Efcacy for heartburn was assessed in 6 trials (11 arms) in patients with NERD, 4 trials (5 arms) in patients with EO, and 5 trials (5 arms) in patients with ulcers (Supporting Information Table 3).Signifcant publication bias was found, Egger's test = 4.99, and p < 0.001 (Supporting Information Figure 1) due to studies comparing omeprazole and placebo, which were all strongly positive.Our meta-analysis using a random efects model of omeprazole compared to placebo or other PPIs revealed found efcacy signifcantly varied by the type of control (X 2 = 91.07,p < 0.001).Omeprazole signifcantly reduced heartburn by 2.5 times compared to placebo (RR = 2.47, p < 0.001), as shown in Figure 2. Te efcacy of omeprazole to reduce heartburn was equivalent to the other three types of PPI (Figure 2): pantoprazole (RR = 1.04, p � 0.26), lansoprazole (RR = 1.02, p � 0.51), and rabeprazole (RR = 1.0, p � 0.62).Omeprazole was slightly less efective than esomeprazole (RR = 0.95, p � 0.02) for heartburn relief.As only one RCT with ilaprazole assessed heartburn relief, it could not be analyzed.

Overall Symptom Improvement.
Trials reporting cure of all APD symptoms were assessed in patients with NERD and had similar results when the outcome was focused on  International Journal of Clinical Practice      3).Te challenge in assessing ulcer and erosion healing was that trials used diferent defnitions of healing (Supporting Information Table 6).For our meta-analysis, we standardized the defnition of "ulcer healing" as healing of the ulcer with or without infammation and/or scarring stage seen by endoscopy and "erosion healing" as complete epithelialization or no mucosal breaks seen in patients with EO after treatment.Using a fxed-efects model, omeprazole had equivalent healing rates (Figure 3) compared to pantoprazole (RR � 1.04 and p � 0.12), ilaprazole (RR � 0.96 and p � 0.26) and rabeprazole (RR � 1.00 and P � 0.99).Omeprazole showed a trend for better ulcer/ erosion healing compared to lansoprazole (RR � 0.94, 95% CI: 0.89 and 0.98, I 2 � 0%, and p � 0.09).Esomeprazole had signifcantly better healing rates of lesions compared to omeprazole in the three treatment arms of patients with EO (RR � 0.89, 95% CI: 0.84 and 0.93, I 2 � 13.6%, and p < 0.001).Subgroup analysis indicated signifcant diferences were found depending upon the type of PPI control (X 2 � 16.2 and p � 0.002).When trials assessing ulcer healing in patients with PU/DU were analyzed separately from EO, no signifcant diferences from those above were found.No trials using placebo controls were found for patients with ulcers or EO.

Efcacy of Secondary Outcomes.
Sufcient data were available to analyze the safety and cost efectiveness of omeprazole.Other secondary outcomes were not reported consistently in trials and when subgrouped by the type of PPI controls and had insufcient trials to be assessed (Supporting Information Table 7).Only two trials reported patient satisfaction [36,52], four reported night/day time acid break-through [36,38,45,49], and fve trials reported the use of rescue medications [38,39,42,46,55].
3.6.1.Safety.Adverse events or safety data were not reported in 4 (13%) trials; only a statement that "no adverse events were found" was reported in two trials (6%), no overall adverse event rate was reported in 3 trials (10%), and the types of adverse events by the treatment group was reported in 22 (71%) of trials (Supporting Information Table 8).Te incidence of at least one reported adverse event was signifcantly lower in patients treated with omeprazole (11/100) compared to other PPIs (range: 17-31/100), Table 2. Ilaprazole reported a low but signifcantly higher incidence of serious adverse events (1.5%) when compared to omeprazole (0.4%).Omeprazole also reported signifcantly less reported nausea compared to pantoprazole (3.6% versus 8.4%, respectively).Our meta-analysis using a fxed-efects model found that the risk of adverse events was equivalent when omeprazole was compared to placebo and the other fve types of PPI (X 2 � 2.09 and p � 0.84, as seen in Supporting Information Figure 4).Tere was no publication bias found for trials reporting adverse event data (Supporting Information Figure 5) (Egger's test � 0.87 and p � 0.4).

Direct Costs of Treatments.
Direct costs of PPIs/patient based on standard recommended doses and a 4-week duration of treatment found that omeprazole (20 mg/d) was the most cost-efective PPI (89.6 Indian rupees), followed by esomeprazole (305.2 rupees), then pantoprazole (322 rupees), followed by ilaprazole (450.8 rupees), rabeprazole (498.4 rupees), and lansoprazole (876.4 rupees) (Supporting Information Table 9).Even when failure rates and retreatment costs were included, omeprazole remained the most cost-efective PPI in India.

Subgroup Analyses.
Sufcient data were available to analyze efcacy of omeprazole for geographic region, degree of blinding, and study quality.Tere was a trend (X 2 � 9.36, p � 0.096) that omeprazole was more efective when trials were conducted in European countries compared with Asian countries, especially when compared against pantoprazole (RR � 1.07, 95% CI: 1.00-1.13,and p � 0.035 and RR � 0.97, 95% CI: 0.89, 1.06, and p � 0.52, respectively), as shown in Supporting Information Figure 6.Most trials used 20 mg/ d and the limited number of trials assessing diferent doses of PPIs did not allow analysis of other doses.Te degree of blinding (double blinded versus open) did not signifcantly change efcacy outcomes by the type of PPI control, but the placebo control (which was double-blinded) was signifcantly diferent (X 2 � 91.1, p < 0.001), as shown in Supporting Information Figure 7.

Study Quality.
Of the 31 RCTs (21, 68%) were ranked overall as low risk of bias (Supporting Information Table 10) and 10 (32%) had an overall high risk of bias.Domains of high-risk included the randomization process not well described (10 trials) and due to nonblinded study designs (12 open trials).When high risk trials were excluded, omeprazole still showed signifcantly better heartburn resolution compared to placebo (RR � 2.47 and p < 0.001, Supporting Information Figure 8) and for ulcer/erosion, healing compared to pantoprazole (RR � 1.05 and p � 0.03, Supporting Information Figure 9).Esomeprazole was signifcantly better than omeprazole for ulcer/erosion healing in low-risk trials (RR � 0.88 and p � 0.001).

Terapeutic Efects of Omeprazole.
A total of 4606 patients were treated with omeprazole in the 31 included trials.
Comparing the use of omeprazole in patients with NERD to those with erosive disease (Table 3); more patients responded to 20 mg/d of omeprazole if they had erosive disease (77% healed) compared to 59% with NERD.Resolution of symptoms was signifcantly higher for patients with erosive disease at both two and four weeks (74% and 92.3%, respectively) compared to patients with NERD (44% and 65%, respectively).More patients with erosive disease reported no nocturnal acid breakthroughs compared to NERD (57% and 32%, respectively).[36,52].Use of rescue medications was less frequent in patients treated with omeprazole compared to other PPIs or standard treatments [38,39,55].Use of omeprazole was also well tolerated (Table 3) but more patients with erosive disease reported mild-moderate adverse events (16%, p < 0.001) compared to patients with NERD (6%).

Discussion
Our systematic review and meta-analysis found that omeprazole has maintained its role as an efective treatment for the healing of heartburn and ulcers/erosions in patients with APD.In India, omeprazole is the only oral PPI listed in the National List of Essential Medicines [67].Omeprazole was signifcantly more efective for heartburn relief when compared to placebo and was equivalent to the other four types of PPIs.Omeprazole also was signifcantly more effective to placebo for the overall improvement in APD symptoms and was equivalent to the other types of PPIs.For the resolution of ulcers or erosions, omeprazole was comparable to most of the other types of PPIs but had a trend for better healing when compared to lansoprazole.Esomeprazole was signifcantly better than omeprazole for ulcer healing and heartburn relief.Edwards et al. reported higher efcacy of esomeprazole (40 mg) over omeprazole (20 mg) in 12 trials with patients with severe RO (OR = 1.84, 95% CI: 1.5 and 2.2) [68].A dose of the single S-enantiomer (esomeprazole) results in a greater body exposure when compared to an equal mg dose of the racemate, omeprazole.Hence, it is not surprising in all studies comparing esomeprazole to omeprazole, a higher efcacy has been observed with esomeprazole.
Other meta-analyses have confrmed the efectiveness of omeprazole compared to other types of treatments for APD.Dean et al. conducted a network meta-analysis (62 RCTs) focused on duodenal ulcer healing and concluded that PPIs were superior to H2RAs or placebo [69].Barberio et al. conducted a network meta-analysis (23 RCTs) focused on NERD and concluded that omeprazole ranked frst for the relief of symptoms, with esomeprazole ranked second [70].International Journal of Clinical Practice Omeprazole (20 mg/day) has also been found to provide quick relief and symptom control in long-term nonsteroidal anti-infammatory drugs (NSAIDs) users [71].Omeprazole and other PPIs can prevent bleeding associated with NSAIDs [30,31,71].Zhang et al. reported a meta-analysis in patients with duodenal ulcers and concluded that ilaprazole was more efective in trials done in China, but the outcome was largely infuenced by one Chinese trial with a low study quality [43] and when this trial was excluded, no signifcant impact by country was found [9].Tis was similar to our fndings, which did not fnd signifcant diferences in efcacy whether the trials were done in Asia, Europe, USA, or in mixed geographic regions, except when omeprazole was more efective in trials done in Europe compared with Asia when pantoprazole was the control.
Omeprazole was found to be the most cost-efective treatment for patients with APD based on direct costs in India (89.6 rupees/patient).Omeprazole was also found to be the most cost-efective PPI in a study of Chinese patients with duodenal ulcers (USD $5.30/patient) [9].
Strengths of our meta-analysis include an extensive literature search done independently by two reviewers; detection of 12 articles not previously included in previously published PPI meta-analyses due to translation or journal access issues and use of intent-to-treat analysis data and use of only RCTs to assess efcacy.Another strength is the use of a standardized Data Extraction Form (Supporting Information Figure 10).Unlike most meta-analyses that have focused on only one type of APD [43,[69][70][71], another strength was the comprehensive inclusion and separate analysis by the diferent types of APDs in our study.Use of meta-analysis subgroups allowed omeprazole to be compared to each type of PPI or placebo separately.
Limitations of our meta-analysis are related to the exclusion of trials that did not share common outcomes.While most trials reported common outcomes (heartburn symptom relief and/or ulcer/erosion healing), some trials had uncommon outcome measures (recurrences of bleeding ulcers or pH levels or overall symptom scores).Some trials with a high risk of bias were hampered by the lack of blinding (PPI controls had diferent formulation from omeprazole).Another limitation was the varied defnitions used for "relief of symptoms" or "ulcer healing."For example, defnitions of ulcer/erosion healed included "epithelium healed," "no ulcer crater," "only residual scar," "no mucosal break," or healing categorized in four levels (ulcer healed with no infammation, ulcer healed but infammation present, ulcer size reduced by less than 50%, or no change in ulcer size).We attempted to minimize this by using standardized defnitions across the studies based on data reported in each trial.Many outcomes for APD which may be clinically important (for example, night-time relief of symptoms or monitoring pH levels) could not be assessed as these outcomes were not commonly reported in the trials.No phase 3 RCTs could be found that were published after 2017.As no trials were done in children, these results may not be generalizable to the pediatric population.
Omeprazole use was well tolerated and had a low rate of adverse events, with only 11% of the patients reporting mild symptoms, most commonly headache or diarrhea.Omeprazole has been the most extensively studied and used, with more than 1200 clinical trials and 400 million patient treatment courses worldwide [72].Omeprazole has an extensively documented long-term safety profle for over 30 years, is approved as treatment for most acidrelated indications, and is efective for the treatment of dyspepsia as well as healing and prevention of NSAID-associated duodenal and gastric ulcers [71].We were also not able to analyze the safety in high-risk populations (diabetic, chronic kidney disease, etc.), as none of the included trials were done in these subpopulations, but several studies did not report any increase in adverse events when used in diabetic patients with chronic kidney disease [73] or patients with cardiovascular disease [74].Concerns with drug interactions between clopidogrel, an anticlotting medication given to cardiac patients, and proton-pump inhibitors have been raised [75].Observational studies do not indicate an increased cardiovascular risk while combining the two drugs despite the theoretical risk of reduced availability of the active moiety of clopidogrel due to the competitive sage of CYP 2C19 by the PPI.

Conclusions
Omeprazole is an efective and safe treatment for acid peptic disorders, including the rapid resolution of GERD symptoms and resolution of erosions and ulcers.Omeprazole was the most cost-efective type of PPI in India.Omeprazole's therapeutic role for patients with acid peptic disorders remains strong.

4
Duration, patients unresponsive to initial duration of treatment had extended 2-4 weeks of treatment (outcome was determined at original time of treatment not extended time); b

Table 1 :
Study population characteristics of 31 included trials (41 treatment arms) for the treatment of acid peptic disease (APD).

Table 2 :
Incidence rate (per 100) of adverse events by the type of PPI and the placebo group.Signifcantly diferent compared to omeprazole (p < 0.05), AE, adverse events.a Data from 23 trials with overall AE rate data reported.b Data from 15 trials with reported SAE data.c Data from 19 trials with type of AE data.

Table 3 :
Terapeutic efects of omeprazole in 4606 patients with nonerosive or erosive acid peptic disorders.
d, day; mg, milligram; nr, not reported; ns, not signifcant.a Pain: NERD (dysphagia) or EO/ulcers (pain associated with ulcer or erosion).b Response time: NERD (time to heartburn symptom resolution) or EO/ulcers (time to pain resolution), measured at two time periods during treatment.c Safety: number reporting at least one adverse event (mild moderate).