Article Identification

• Article Title: “Intermittent vs Continuous Proton Pump Inhibitor Therapy for High-Risk Bleeding Ulcers: A Systematic Review and Meta-analysis”
• Citation: JAMA Intern Med. 2014;174(11):1755–1762
• DOI: 10.1001/jamainternmed.2014.4056

Quick Reference Summary

• In this systematic review and meta-analysis of 13 trials encompassing patients with high-risk bleeding ulcers who received endoscopic therapy, intermittent proton pump inhibitor (PPI) regimens were found to be noninferior to the standard intravenous bolus plus continuous infusion in reducing rebleeding (relative risk [RR], 0.72; upper boundary of 1-sided 95% CI, 0.97).
• The absolute risk difference for 7-day rebleeding was –2.64% (upper boundary of 1-sided 95% CI, –0.28%), indicating comparable efficacy with potential cost and resource savings for intermittent dosing.

Core Clinical Question

In adult patients (P) with high-risk peptic ulcer bleeding who have undergone successful endoscopic hemostasis, does intermittent proton pump inhibitor therapy (I) compared with the standard intravenous bolus plus continuous infusion therapy (C) reduce (or at least not increase) the incidence of rebleeding (O)?

Background

• Disease Overview: Peptic ulcer disease is a common cause of upper gastrointestinal bleeding (UGIB). High-risk stigmata (active bleeding, nonbleeding visible vessel, or adherent clot) place patients at increased risk for rebleeding and complications.
• Prior Data (Selected Points):
  • Multiple trials have shown that PPIs, particularly high-dose IV regimens, reduce the risk of rebleeding following endoscopic therapy.
  • Various meta-analyses have indicated that both bolus plus continuous infusion and intermittent PPI regimens can significantly reduce rebleeding, surgery, and mortality compared with no therapy.
• Current Standard of Care: Guidelines (e.g., American College of Gastroenterology [ACG]) commonly recommend an initial IV bolus (80 mg) followed by continuous infusion (8 mg/h) of a PPI for 72 hours for high-risk endoscopic lesions.
• Knowledge Gaps Addressed by the Study: Whether intermittent PPI dosing (given as repeated IV boluses or high-dose oral PPIs) confers noninferior protection against rebleeding compared to continuous infusion, potentially offering advantages in cost, implementation ease, and resource utilization.
• Study Rationale: By combining existing randomized trial data, the authors aimed to test the hypothesis that intermittent regimens are as effective as (i.e., noninferior to) continuous-infusion regimens in preventing rebleeding when applied in patients with high-risk ulcer bleeding.

Methods Summary

• Study Design: Systematic review and meta-analysis of randomized controlled trials (RCTs).
• Setting & Time Period: MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials were searched through December 2013. Abstracts from major gastroenterology meetings were also included.
• Population: Patients with high-risk stigmata on endoscopy (active bleeding, visible vessel, or adherent clot) who received successful endoscopic therapy.
• Inclusion/Exclusion: Only RCTs comparing intermittent PPI dosing with standard IV bolus plus continuous infusion post-endoscopic therapy were included.
• Intervention Details: Intermittent PPI therapy could be oral or intravenous boluses at varying frequencies and doses.
• Control: 80-mg IV PPI bolus followed by continuous infusion of 8 mg/h for 72 hours.
• Outcomes: Primary outcome was rebleeding within 7 days. Secondary outcomes: rebleeding within 3 and 30 days, mortality, urgent reintervention (endoscopic, surgical, or radiologic), blood transfusions, and hospital length of stay.
• Data Analysis: Pooled risk ratios (RR) and absolute risk differences were calculated. The predetermined noninferiority margin was 3% for rebleeding. A fixed-effects or random-effects model was used depending on heterogeneity.
• Sample Size Calculations: Not performed by authors of the review; the meta-analysis approach was taken to combine existing data.
• Ethics and Funding: Funded by the National Institutes of Health (T32 grant). No relevant conflicts of interest were disclosed.

Detailed Results

• Participant Flow and Demographics:
  • Thirteen trials met inclusion criteria, pooling data on 1,691 patients with high-risk lesions. Demographics varied across studies (e.g., Western vs Asian populations), but overall, groups were balanced in baseline characteristics.
• Primary Outcome:
  • Rebleeding within 7 days: Intermittent PPI therapy yielded RR=0.72 (upper boundary 1-sided 95% CI, 0.97) and an absolute risk difference of −2.64% (upper boundary, −0.28%). The 3% noninferiority margin was easily surpassed.
• Secondary Outcomes:
  • Rebleeding within 3 days (RR=0.73; upper boundary 1-sided 95% CI, 1.02) and 30 days (RR=0.89; upper boundary 1-sided 95% CI, 1.17) similarly showed no significant increased risk from intermittent dosing.
  • Other endpoints (e.g., mortality, need for reintervention, transfusions, length of stay) did not differ significantly, and point estimates often favored the intermittent group.
• Adverse Events / Safety Data: Overall adverse events were similar between groups; no new safety concerns for intermittent regimens were identified.

Results Table (Simplified)

Outcome	Intermittent Group	Continuous Group	Risk Ratio (95% CI)	P-value
Rebleeding ≤7 days	~9.4% pooled incidence	~12.0% pooled incidence	0.74 (0.52–1.06)*	–
Mortality (within 30 days)	Low (<5% in both)	Low (<5% in both)	0.64 (0.34–1.21)*	–
Hospital Length of Stay (days)	–0.26 fewer (mean diff)	Reference	–	–

* Two-sided 95% CI. The noninferiority analysis used one-sided 95% bounds.

Authors' Conclusions

• Primary Conclusions: Intermittent PPI regimens are noninferior to the standard bolus plus continuous infusion in preventing rebleeding in patients with high-risk peptic ulcers.
• Clinical Implications: Given the decreased dosage, cost, nursing time, and infusion pump requirements, intermittent therapy should be strongly considered as the standard of care.
• Future Research: Well-powered RCTs comparing diverse intermittent regimens (dosing frequency, IV vs oral, and total daily dose) and additional cost evaluations may further refine specific usage parameters.

Critical Analysis

A. Strengths

• Thorough Literature Search: Captured a wide range of randomized trials through multiple databases and conference abstracts.
• Rigorous Noninferiority Framework: Applied a predefined 3% margin for the absolute risk difference, offering clarity on clinical acceptability.
• Generalizable Populations: Included Western and Asian cohorts, bolus and continuous regimens, and various endoscopic therapy modalities.
• Robust Statistical Approaches: Conducted both one-sided noninferiority testing and traditional two-sided analyses, along with subgroup evaluations.

B. Limitations

• Variability in Intermittent Regimens: Dose, frequency, and IV vs oral administration differed across studies, hampering clarity on the “optimal” intermittent protocol.
• Risk of Selection or Performance Bias: Many trials did not fully describe allocation concealment or blinding; eight of the thirteen lacked blinding.
• Insufficient Power to Address Subpopulations: Not enough data to delineate if certain subgroups (e.g., actively spurting vs adherent clot) might have differential outcomes.
• Older Search Window: Literature search ended in December 2013, limiting immediate consideration of newer RCTs. However, more recent data in observational settings often corroborate these findings.

Literature Review

A. Positioning the Current Study in Existing Evidence

Endoscopic therapy combined with PPI administration has been the mainstay approach for high-risk bleeding ulcers for over two decades (Laine et al., Am J Gastroenterol. 2012;107:345-361). In early comparative trials, an intravenous bolus followed by continuous infusion of PPIs was favored, predicated on in vitro models suggesting that intragastric pH >6 would stabilize clot formation. Multiple meta-analyses conducted in the early 2000s supported that high-dose PPIs following successful endoscopic therapy reduced rebleeding, need for surgery, and mortality (Laine & McQuaid, Clin Gastroenterol Hepatol. 2009;7:33-47). However, these earlier syntheses often combined heterogeneous dosing regimens, and few directly compared continuous infusion versus purely intermittent PPI administration in a formal noninferiority framework.

The 2014 analysis in JAMA Internal Medicine (Sachar et al., JAMA Intern Med. 2014;174(11):1755-1762) specifically targeted this gap, assembling RCTs of patients with confirmed high-risk stigmata (active bleeding, nonbleeding visible vessels, or adherent clots) who underwent endoscopic therapy. By isolating studies that used the full recommended 72-hour infusion (80 mg IV bolus followed by 8 mg/h IV infusion) in the comparator group and a recognizable intermittent regimen (ranging from 40-80 mg IV/PO once or multiple times daily) in the experimental arm, the authors addressed the question of whether similarly robust outcomes could be achieved without continuous infusion.

Since 2014, additional research and guideline updates have transpired. The American College of Gastroenterology’s 2021 guideline on upper gastrointestinal and ulcer bleeding (ACG 2021, Am J Gastroenterol. 2021;116:899–917) acknowledges that intermittent PPI regimens may be as effective as continuous infusion in many settings. This shift in guidance partly arose because real-world constraints, such as IV pump availability or drug shortages, and cost considerations pushed clinicians to explore intermittent approaches in parallel with older high-dose protocols. Indeed, the ACG guideline now highlights that intermittent or bolus PPI regimens may be acceptable alternatives for patients with stigmata of recent hemorrhage after endoscopic intervention.

B. Comprehensive Synthesis of Findings

1) RCTs and Meta-analyses Before and After 2014

Several RCTs before 2014 tested high-dose IV PPI protocols against lower or intermittent dosing. Many had design limitations: small sample sizes and inconsistent definitions of rebleeding. Despite these differences, the consistent observation was that frequent or high-dose PPIs—whether by continuous infusion or repeated bolus—reduced rebleeding versus placebo or no therapy.

Sachar and colleagues’ meta-analysis built upon these earlier data by explicitly using a noninferiority framework. They concluded that intermittent therapy was statistically comparable to continuous infusion with respect to rebleeding within 7 days, 3 days, and 30 days. Moreover, there were no major differences in mortality, need for urgent intervention, or hospital length of stay. The robust approach, including sensitivity analyses and subgroup evaluations (oral vs IV, high vs moderate dose), suggested that no specific intermittent regimen was clearly superior or inferior, thereby giving clinicians flexibility.

2) Observational and Cost Studies (2016–2025)

Subsequent to the 2014 publication, real-world observational data and additional smaller RCTs offered further insight:

• Hsieh et al. (Medicine (Baltimore). 2021;100(49):e28064) examined a retrospective cohort of 335 patients with high-risk bleeding peptic ulcers. Interestingly, they observed a higher rebleeding rate in patients receiving a high-dose continuous infusion (HDC) than those receiving non-high-dose intermittent infusion (NHDI). Although nonrandomized, this cohort suggested that intermittent regimens might actually outperform continuous infusion in certain populations, potentially due to confounding variables (e.g., more severe baseline risk in the HDC group) or differences in actual delivered care. Notably, cost savings were also reported for intermittent regimens, aligning with conclusions from previous comparisons indicating less medication expense and resource usage.
• Lu et al. (Am J Gastroenterol. 2016;111(10):1389-1398) analyzed the budgetary impact of various intravenous PPI regimens—continuous infusion versus intermittent bolus—and found that while high-dose IV PPIs drive up costs, the overall incremental cost differences tended to be modest relative to the total fee for an upper GI bleed hospitalization. Nevertheless, in certain subgroups, such as patients able to be discharged earlier or those with clean-base ulcers, intermittent therapy was more clearly cost-saving. These results underscore that large drivers of cost in UGIB management relate to hospital stay, procedures, and transfusions, but PPI strategies can still influence the total bill.
• Leung et al. (Ann Pharmacother. 2022;56(10):1127-1132) evaluated a multicenter retrospective dataset and, similarly to Hsieh and colleagues, affirmed that intermittent PPI bolus therapies did not result in higher rebleeding rates than continuous infusion. Their analysis also noted that clinicians often reserve continuous infusions for patients perceived to be at highest risk, which can confound direct comparisons. Nevertheless, the findings are consistent with the noninferiority stance that intermittent dosing is effective in real-world practice.

3) Integration with Guidelines and Ongoing Trials

Guidelines from around the world, including the ACG (last updated in 2021) and various European societies, have softened the older stance that continuous infusion PPIs are the only recommended strategy. Intermittent dosing protocols (e.g., 40 mg IV twice daily or 80 mg IV bolus followed by 40 mg IV every 12 hours) appear in many care pathways. The question for ongoing and future investigations is refining the best intermittent dose, route (IV vs oral), and frequency in different ulcer risk strata. Oral PPIs, if tolerated, may suffice in many cases, as suggested by the primary article’s included trial arms.

Various subquestions remain, especially regarding subpopulations:

• Do actively spurting ulcers or large ulcers >2 cm respond equivalently to intermittent therapy?
• Is there an upper-bound dose threshold beyond which no added benefit is conferred?
• How do cost analyses play out in resource-limited or rural healthcare systems, where continuous infusion might exceed local capacity?

In these domains, new data continue to emerge. Some centers in Asia prefer high-dose oral regimens (e.g., 80 mg oral omeprazole once, followed by 80 mg every 12 hours), leveraging pharmacodynamic evidence that repeated high-dose oral therapy can achieve comparable intragastric pH control to IV therapy. Meanwhile, in North America and Europe, IV PPIs remain a default choice, partly due to tradition and medicolegal standards, although step-down to oral therapy is increasingly used once patients stabilize.

C. Gaps and Future Directions

Despite the robust support for intermittent PPI therapy, large-scale randomized trials specifically designed for noninferiority in distinct patient subgroups (e.g., patients with spurting hemorrhage vs visible vessel) would bolster confidence. Moreover, the wide variety of intermittent dosing protocols suggests that further standardization might reduce clinician confusion and ensure consistent practice. Cost-effectiveness analyses, particularly in the post–COVID-19 era of resource constraints, are likely to remain influential. Additionally, future studies might explore biomarkers or endoscopic factors that refine who truly benefits from the highest intensities of acid suppression.

Overall, since the publication of the 2014 meta-analysis by Sachar et al., consensus has increasingly shifted to endorsing intermittent PPI therapy as an acceptable, cost-effective, and clinically noninferior alternative to continuous infusion for most patients with high-risk bleeding ulcers. This shift is now largely reflected in guidelines and day-to-day practice in 2025. However, local protocols still may differ, reflecting variations in clinician familiarity, facility resources, and patient-level risk considerations.

Clinical Application

• If intermittent PPI regimens are deemed feasible—either as repeated IV boluses or high-dose oral regimens—they may replace the resource-intensive continuous infusion for the majority of patients who present with high-risk peptic ulcer bleeding.
• Implementation considerations include the need for hospital protocol updates, staff education (especially nursing and pharmacy), and attention to optimal dosing intervals (e.g., twice-daily or thrice-daily IV/PO therapy) to maintain adequate acid suppression.
• For specific higher-risk scenarios (e.g., massive active hemorrhage in unstable patients), clinicians might initially choose continuous infusion while stabilizing the patient, then transition to intermittent therapy once bleeding risk subsides.

How To Use This Info In Practice

Clinicians can integrate these findings into practice by using intermittent PPI dosing—particularly in settings where staffing, pump availability, and cost considerations drive the need for simpler regimens—while remaining consistent with the most recent guidelines that endorse multiple acceptable models. Although the evidence strongly supports intermittent therapy, clinicians should continue to apply clinical judgment for exceptionally high-risk cases and remain updated as emerging studies further clarify optimal regimens for specific ulcer profiles.