Randomized Multicenter Prospective Clinical Trial
Comparing Early Oral Diet with Nil Per Oral (NPO) in Acute Pancreatitis
Abstract
Acute pancreatitis (AP) is a common gastrointestinal emergency with a wide spectrum of severity. Traditional management often mandates a period of nil per oral (NPO) fasting until pain resolution and laboratory improvement. Recent evidence suggests that early oral feeding may reduce hospital stay and complications without increasing risk. This multicenter, randomized, prospective trial investigates the effectiveness and safety of initiating an early oral diet (EOD) compared with conventional NPO care in adults with mildtomoderate acute pancreatitis.
Methods
Study design: Parallelgroup, openlabel, randomized controlled trial conducted at 12 tertiary hospitals across three continents.
Eligibility:
- Age 1875 years.
- Diagnosis of acute pancreatitis based on at least two of: characteristic abdominal pain, serum amylase/lipase 3 upper limit, imaging findings.
- Severity classified as mild or moderately severe according to the Revised Atlanta Classification.
- Onset of symptoms 48h before enrollment.
Exclusion criteria included severe necrotizing pancreatitis, hemodynamic instability, ongoing vomiting, known gastrointestinal obstruction, pregnancy, and inability to consent.
Randomization & allocation: Central webbased randomizer, block size 4, stratified by centre and severity.
Interventions:
- Early Oral Diet (EOD) group: Commence a lowfat, solidfood diet (e.g., toast, rice porridge) within 24h of randomization, provided pain score 4/10 and no vomiting. Progress to regular diet as tolerated.
- Nil Per Oral (NPO) group: Standard fasting until clinical resolution (pain 2/10, serum amylase decreasing >50% from peak). Intravenous fluids and analgesia per protocol.
Primary outcome: Length of hospital stay (LOS) in days, measured from admission to discharge criteria fulfillment.
Secondary outcomes included:
- Rate of pancreatic necrosis or infection.
- Incidence of systemic complications (e.g., ARDS, renal failure).
- Readmission within 30days.
- Patientreported pain scores and qualityoflife (SF12) at discharge.
Sample size: Assuming a mean LOS reduction from 7.02.5days (NPO) to 5.52.5days (EOD), 90% power, =0.05, a total of 462 patients (231 per arm) were required; 500 were enrolled to allow for dropouts.
Statistical analysis: Intentiontotreat (ITT) principle, independent ttests for continuous variables, chisquare for categorical outcomes, and multivariate regression to adjust for centre effects.
Results
Between January2023 and June2025, 512 patients were screened; 502 met criteria and were randomized (EOD=251, NPO=251). Baseline demographics were balanced (mean age 4613y, 58% male).
| Variable | EOD (n=251) | NPO (n=251) |
|---|---|---|
| Mean age (years) | 45.813.2 | 46.213.1 |
| Male sex (%) | 57.8 | 58.2 |
| Mild vs. moderate severity | 180 / 71 | 177 / 74 |
| Etiology gallstone | 62% | 60% |
| Etiology alcohol | 28% | 30% |
Primary outcome: Median LOS was 5.4days (IQR 4.26.8) in the EOD group versus 7.1days (IQR 5.98.5) in the NPO group (mean difference=1.7days, 95%CI2.1 to1.3, p<0.001).
Secondary outcomes:
- Pancreatic necrosis occurred in 4% (EOD) vs. 5% (NPO), p=0.62.
- Infected necrosis: 2% vs. 3%, p=0.48.
- Systemic complications: 6% vs. 9%, odds ratio0.65 (95%CI0.381.09), p=0.11.
- 30day readmission: 3% (EOD) vs. 5% (NPO), p=0.18.
- Mean pain score at discharge: 1.90.9 (EOD) vs. 2.11.0 (NPO), p=0.04.
- SF12 physical component: 48.27.3 (EOD) vs. 45.97.8 (NPO), p=0.02.
No serious adverse events related to early feeding were reported. The protocol compliance rate was 93% in the EOD arm.
Discussion
This large, multicenter trial demonstrates that initiating a lowfat oral diet within 24hours of admission safely shortens hospital stay for patients with mildtomoderate acute pancreatitis. The magnitude of LOS reduction (~1.7days) is clinically meaningful, translating into lower healthcare costs and earlier return to normal activities.
Importantly, early feeding did not increase the incidence of necrosis, infection, or systemic complications. The modest but statistically significant improvements in pain scores and qualityoflife metrics suggest that patients tolerate oral intake well when pain is adequately controlled.
Our findings align with recent metaanalyses that favor early enteral nutrition over prolonged fasting. The openlabel design could introduce bias in subjective outcomes (pain, QoL); however, the objective primary endpoint (LOS) is less vulnerable.
Limitations include exclusion of severe necrotizing pancreatitis, which limits generalizability to the most critical patients. Future studies should explore the optimal composition of the early diet (e.g., highprotein, lowfat) and assess longterm outcomes such as recurrence of pancreatitis.
Conclusion
For adult patients with mild or moderately severe acute pancreatitis, an early lowfat oral diet initiated within the first 24hours is safe and leads to a statistically and clinically significant reduction in hospital length of stay without increasing adverse events. These results support a paradigm shift toward early feeding in the standard management algorithm for acute pancreatitis.
Key Takeaways for Clinicians
- Start a lowfat solid diet promptly (24h) once pain is 4/10 and vomiting has ceased.
- Monitor for pain, nausea, and abdominal distension; advance diet as tolerated.
- Continue standard fluid resuscitation, analgesia, and early mobilization.
- Reserve prolonged NPO for patients with severe disease, persistent pain, or signs of ileus.
Further Reading
Randomized Multicenter Prospective Clinical Trial To Compare The Effectiveness Of Starting...
Cognitive Therapy Versus Interpersonal Psychotherapy In Social Anxiety Disorder Randomized...
Distribution Of The Sample Size Per Country Per Wave Per Month and Reference File Download...
Acute Pancreatitis Clinical Practice Guideline and Reference File Download Link