Highlights
No Benefit to High-Energy Loading
Early administration of 30 kcal/kg/day in patients with acute pancreatitis (AP) showed no significant reduction in mortality or the incidence of severe acute pancreatitis compared to a gradually increasing caloric strategy.
Potential for Clinical Harm
Patients receiving high-energy nutrition experienced higher rates of organ failure (16.7% vs. 9.1%) and pain relapse (27.1% vs. 19.0%) in uncorrected analyses, suggesting that aggressive early feeding may exacerbate the inflammatory response.
Trial Terminated Early
Due to the lack of benefit and potential safety signals, the GOULASH trial was halted following an interim analysis of 636 patients, concluding that reaching the high-energy target was unlikely to yield superiority even with a larger sample size.
The Nutritional Paradox in Acute Pancreatitis
Acute pancreatitis (AP) remains a significant challenge for gastroenterologists and critical care specialists. As one of the most frequent causes of gastrointestinal-related hospitalizations globally, its management has historically fluctuated between “bowel rest” and early aggressive hydration. In recent decades, the paradigm shifted toward early enteral nutrition (EN), which has been shown to reduce infectious complications and maintain the intestinal mucosal barrier. However, the optimal caloric dose during the hypermetabolic early phase of the disease has remained a subject of intense debate.
Clinicians have long grappled with the question: Should we meet full caloric requirements immediately to counteract the systemic inflammatory response syndrome (SIRS), or should we adopt a trophic, gradual approach? The GOULASH (High versus gradually increasing energy nutrition in the early phase of acute pancreatitis) trial sought to provide a definitive answer to this clinical dilemma.
Trial Architecture: The GOULASH Study Design
This multicentre, double-blind, randomized clinical trial was conducted across several high-volume centers between January 2017 and April 2023. The study enrolled 636 patients diagnosed with AP, regardless of the predicted severity at admission, ensuring a pragmatic and generalizable patient population.
Interventions and Comparators
Patients were randomized into two distinct nutritional arms:
1. High Energy (HE) Group: Targeted 30 kcal/kg/day starting immediately upon initiation of enteral support.
2. Low Energy (LE) Group: A gradually increasing strategy starting at 0 kcal/kg/day (trophic/minimal) and increasing to the 30 kcal/kg/day target over a period of four days.
Primary and Secondary Endpoints
The primary outcome was a composite of mortality and the development of severe acute pancreatitis, as defined by the Revised Atlanta Criteria. Secondary outcomes included the specific rates of organ failure, local complications (such as pancreatic necrosis), infection rates, and the frequency of pain relapse, which often complicates early refeeding.
Analysis of Results: Primary and Secondary Outcomes
An interim analysis was pre-planned after 50% of the target enrollment was achieved. Following the recruitment of 636 patients, the data monitoring committee and investigators performed a modified intention-to-treat (mITT) analysis that led to the trial’s termination.
Primary Outcome Findings
In the mITT population, the primary composite outcome occurred in 28 of 312 patients (9.0%) in the HE group compared to 18 of 307 patients (5.7%) in the LE group. While the percentage was numerically higher in the high-energy group, the difference did not reach statistical significance (p-corrected = 0.42).
Secondary Outcomes and Safety Signals
Significant findings emerged within the secondary endpoints when viewed through the lens of uncorrected p-values. The HE group demonstrated a substantially higher rate of organ failure (16.7% vs. 9.1%, p=0.007) and pain relapse (27.1% vs. 19.0%, p=0.03). Although these findings did not maintain statistical significance after the Benjamini-Hochberg correction for multiple testing (p=0.13 and p=0.23, respectively), they presented a concerning clinical trend that influenced the decision to stop the trial.
Statistical Nuances: The Impact of Multiplicity Correction
One of the most critical aspects of the GOULASH trial is the interpretation of its statistical results. The use of the Benjamini-Hochberg false discovery rate (FDR) method is a rigorous approach to prevent Type I errors (false positives) in studies with multiple endpoints.
While the corrected p-values suggest that the differences in organ failure and pain relapse could be due to chance, the uncorrected data (p=0.007 for organ failure) is difficult for clinicians to ignore. In clinical practice, a doubling of the organ failure rate is a signal of high clinical relevance. The investigators correctly identified that the likelihood of the HE strategy eventually proving superior was negligible, thus rendering further enrollment unethical.
Clinical Interpretation: Why High Energy Might Fail
Several biological mechanisms may explain why an aggressive caloric load in the early phase of AP could be detrimental:
1. Metabolic Stress: During the peak of the systemic inflammatory response, the body’s ability to process exogenous nutrients may be impaired. Overfeeding can lead to metabolic complications, including hyperglycemia and hepatic steatosis, which may worsen the inflammatory state.
2. Cholecystokinin (CCK) Stimulation: High-energy enteral feeding can stimulate the secretion of CCK and other pancreatic secretagogues. In the context of an already inflamed pancreas, this “overstimulation” may trigger further enzyme activation and exacerbate local tissue damage, potentially explaining the higher pain relapse rates.
3. The Refeeding Phenomenon: Even in non-malnourished patients, the rapid introduction of high calories during acute illness can lead to shifts in electrolytes and fluid balance that may contribute to organ dysfunction.
Expert Commentary and Practical Applications
For the practicing clinician, the GOULASH trial provides a clear directive: the “start low, go slow” approach to nutrition in acute pancreatitis is not only sufficient but likely safer. While early enteral nutrition remains a cornerstone of therapy to prevent gut-derived sepsis, there is no clinical mandate to hit full caloric targets within the first 24 to 48 hours.
Guidelines from organizations like ESPEN and ASPEN already suggest that enteral nutrition should be started early, but the GOULASH data suggests that we should be cautious about the speed of titration. A gradual ramp-up over 3 to 4 days aligns with the physiological recovery of the patient and minimizes the risk of triggering recurrent pain or systemic organ stress.
Study Limitations
It is important to note that the trial was stopped early, which can sometimes overestimate treatment effects or leave secondary questions unanswered. Additionally, the heterogeneity of AP (ranging from mild interstitial to severe necrotizing) means that while the overall trend favors gradual feeding, individual patient needs may still vary based on baseline nutritional status.
Conclusion
The GOULASH trial serves as a vital reminder in clinical medicine that “more” is not synonymous with “better.” In the early, volatile phase of acute pancreatitis, a high-energy nutritional strategy (30 kcal/kg/day) fails to improve survival or disease severity and may potentially increase the risk of organ failure and pain recurrence. Clinicians should prioritize the initiation of enteral feeding but adopt a gradual caloric titration strategy to ensure patient safety and optimize recovery.
Funding and Registration
This study was registered under ISRCTN63827758. The trial was supported by various academic and clinical research grants as detailed in the primary publication in Gut (2026). The authors declare no significant conflicts of interest related to the nutritional interventions used in this study.
References
1. Márta K, et al. High versus gradually increasing energy nutrition in the early phase of acute pancreatitis (GOULASH): a multicentre double-blind randomised clinical trial. Gut. 2026. PMID: 41786585.
2. Banks PA, et al. Classification of acute pancreatitis–2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013;62(1):102-111.
3. Arvanitakis M, et al. ESPEN guideline on clinical nutrition in acute and chronic pancreatitis. Clin Nutr. 2020;39(3):612-631.
