Acute pancreatitis (AP) refers to the activation of pancreatic exocrine enzymes due to various etiologies [1,2]. Its diagnosis in children requires at least two of the following: abdominal pain consistent with AP, serum amylase and/or lipase levels ≥ 3 times the upper limit of normal, or imaging findings indicative of AP.(2) As a common acute abdominal condition, AP is clinically categorized by severity into mild, moderately severe, and severe forms [3]. Mild AP involves no organ failure or complications, and typically resolves within one week; moderately severe AP (SAP) is defined by either transient organ failure or dysfunction (resolving within 48 h) or the presence of local or systemic complications; SAP refers to pancreatitis with persistence of organ failure lasting beyond 48 h [3]. In recent decades, there has been a notable increase in the incidence of AP among pediatric populations [4,5].

Enteral nutrition (EN) is the preferred method of clinical nutritional support for children and is indicated for the management of conditions such as pediatric AP and pediatric Crohn's disease [2,6]. EN offers clear benefits for AP patients, shorter hospital stays, lower intensive care unit (ICU) admission, and reduced SAP incidence versus parenteral nutrition [7,8]. Inappropriate nutritional interventions may compromise children's immune response and exacerbate complications, thereby impeding recovery. Therefore, judicious nutritional support is paramount for managing pediatric AP.

Several guidelines recommend early EN as the primary therapy for pediatric AP, suggesting oral feeding over nasogastric (NG) feeding in mild-to-moderate cases [2]. However, these are predominantly based on evidence from adults and limited pediatric research [8,9]. Our previous randomized controlled trial confirmed that NG feeding is safe and effective in pediatric AP;[10] whereas in adults with high-risk AP, early nasoenteric feeding was not superior to an oral diet initiated after 72 h [11]. This prospective study compared oral and NG feeding, based on the hypothesis that oral feeding would be as safe and effective as or non-inferior to NG feeding.

A single-center, prospective, randomized trial was conducted at a tertiary center, the Children’s Hospital, Zhejiang University School of Medicine in China, from September 2021 to August 2024. The participants were patients who were diagnosed with AP and required hospitalization. The study was approved by the Ethics Committee of the Children’s Hospital, Zhejiang University School of Medicine, and was carried out in accordance with the Declaration of Helsinki. The purpose of the study was explained clearly and accurately to all patients and guardians, and their informed written consent was signed. The trial was registered at the Chinese Clinical Trial Registry (ChiCTR2100050915).

Figure 1 shows the CONSORT flowchart of this study. A total of 56 patients were recruited, with 26 (46.4 %) participants randomly allocated to the ORAL group and 30 (53.5 %) to the NG group. Two patients from the ORAL group either withdrew from the study or were lost to follow-up, whereas 6 patients in the NG group experienced similar outcomes. Forty-eight participants successfully completed the protocol, with an equal distribution of 24 patients in each group. The ORAL group included 2 patients with moderately SAP and 22 with mild AP, compared to 4 and 20 patients, respectively, in the NG group (p = 0.382). There was no difference between the 2 groups in demographic and anthropometric characteristics or laboratory and imaging examinations at presentation (Table 1). With respect to clinical presentation, all patients experienced abdominal pain prior to admission, with an average duration of approximately 1.5 (0.6 to 3.8) days for the ORAL group and approximately 1.0 (0.5 to 3.8) days for the NG group (p = 0.975). The most prevalent accompanying symptom was vomiting, which occurred in 45.8 % of the ORAL group and 66.7 % of the NG group (p = 0.146, Table 2). In terms of etiology, anatomical abnormalities were observed as the primary cause, followed by idiopathic factors. There was no difference in etiology between the groups (Table 3).

Fig. 1.

Recruitment, randomization, and follow-up in clinical trials.

Secondary outcomes

In the ORAL group, one patient experienced recurrent abdominal pain, which improved following a reduction in feeding. The child subsequently showed increased tolerance for further feeding with a gradual increase in quantity. No instances of recurrent abdominal pain were noted in the NG group (p = 1.000). Upon discharge, no significant differences in nutritional parameters, including hemoglobin, albumin, and prealbumin levels, were observed between the two groups (Table 4). With respect to the changes in pancreatic enzyme levels, a significant median reduction in lipase levels was observed in the NG group compared with the ORAL group on day 3 of feeding (p = 0.019). Additionally, during the oral/NG feeding period and at the time of discharge, no significant differences were observed in the serum amylase and lipase levels or their changes between the two groups of children (Table 4, Figure 2). In terms of complications, there were 4 patients in the NG group, with 3 patients presenting with fever and abdominal pain accompanied by diarrhea, and 1 patient with unexpected tube removal; in contrast, the ORAL group had no cases of complications (p = 0.109, Table 4). The first 3 patients were clinically diagnosed with gastroenteritis; however, no viral infection was detected.

Fig. 2.

Changes of serum amylase and lipase on different time points. Day 1, the first day of EN. Day 3, the third of EN.

Patients in both groups underwent US at baseline and at two follow-up time points. Compliance with the scheduled scans was 100 % in the ORAL group and 87.5 % in the NG group (p = 0.234). At the 1-week follow-up, mild pancreatic changes (e.g., swelling, abnormal echogenicity) were observed in 5 children (20.8 %) in the ORAL group, compared with 3 children (14.2 %) in the NG group (p = 0.705). At the 5-week follow-up, all children in the ORAL group exhibited normal findings, while 1 child (4.7 %) in the NG group still presented with focally increased echogenicity (p = 0.467). Throughout the follow-up period, US assessments revealed no local pancreatic complications in either cohort of patients.

In the NG group, two patients were readmitted to the hospital within 5 weeks post-discharge, whereas no patients in the ORAL group experienced readmission. The first patient presented with 2 heterozygous mutations in SPINK1, whereas the second patient exhibited incomplete pancreatic division; both patients were readmitted due to recurrent pancreatitis despite complete resolution of abdominal pain and normal amylase levels at discharge.

Although guidelines emphasize the value of early oral feeding in pediatric AP, supporting evidence from pediatric RCTs remains limited. Therefore, the authors conducted an RCT specifically in children, comparing oral with NG tube feeding. The present trial demonstrated no significant differences between the two strategies in length of hospital stay, feeding tolerance, abdominal pain duration, readmission rates, or complications in pediatric mild-to-moderate AP. Notably, weight-gain trends did not differ significantly between groups, corroborating earlier findings [15]. Hence, based on the present findings, the authors propose that oral feeding should be the preferred nutritional strategy for children with mild AP to moderately SAP, which aligns with the established adult data [11].

Although patients with AP are often advised to follow a low-fat diet, accumulating evidence questions its necessity. Studies indicated full-fat diets are safe during early EN in children with AP [15], and animal models suggested the inflamed pancreas may be refractory to further stimulation [16]. Consistently, a retrospective analysis found that average dietary fat intake did not significantly affect hospital stay duration or lipase levels in mild pediatric AP [9]. In the present study, employing a polymeric formula (35.5 % fat) and advancing to a full-fat oral diet after reaching nutritional targets yielded favorable recovery outcomes without strict low-fat restriction (< 30 % of calories from fat) [17], lending further support to the safety of full-fat diets in this population.

The EN (oral, gastric, or jejunal) can maintain the integrity of the intestinal mucosa, safeguard against the decline in intestinal immune function, and restricts bacterial translocation [18,19]. NG feeding has been shown to be well tolerated in both adult and pediatric populations with AP [10,14,20], and serves as a suitable nutritional strategy for patients with severe illness, mostly in the ICU or with distinctly decreased oral feeding [2]. All of these participants initiated EN within 48 h, yet two patients in the NG group exhibited symptoms of intolerance. Further examination revealed that they had choledochal cysts and underwent choledochocystectomy. The previous study on tube feeding revealed similar findings [10], which suggested a possible association between choledochal cysts and feeding intolerance; however, further validation is needed.

The literature indicates that prevalent methods of EN access, including nasoenteric tubes, gastrostomy, and jejunostomy, carry risks of mechanical, gastrointestinal, infectious, and metabolic complications [21,22]. The earlier studies also revealed the possibility of accidental tube removal [10]. In a large cohort of 925 critically ill patients with SAP, any form of EN was associated with infections, while oral feeding correlated with the lowest infection and mortality rates [23]. It is therefore understandable that the complication profiles in the present study differed.

Although a serum lipase level greater than 7 times the upper limit of normal within 24 h post-onset could be a suggested predictor for pediatric SAP [24], and although lipase levels were elevated on admission in 2 cohorts, no patient in the present study developed SAP. The inflammatory process in AP is characterized by intra-pancreatic enzyme activation and failure of cellular protective mechanisms [1]. Given the potential for recurrence and atypical presentation, serial monitoring of lipase and amylase was conducted during follow-up, alongside pancreatic ultrasonography and clinical assessments.

The present study has several limitations. First, the etiologies of pediatric AP are varied and distinct from those in adults, as are the severity and overall prognosis associated with pediatric AP [25]. The predominant etiological factors in both cohorts were anatomical anomalies and idiopathic origins, which were similar to the findings of a previous study [25]. Owing to the experimental design, cases resulting from trauma, systemic diseases and pharmacological agents, as well as patients with pre-existing comorbidities, were excluded; therefore, it remains uncertain whether the findings can be generalized to other types of pancreatitis. Second, the authors recruited as many eligible participants as possible, but the sample size was limited (71 % of the intended sample size). Furthermore, given the high risk of life-threatening complications, this study does not address the implementation of early nutritional therapy for SAP and may not be able to comprehensively reflect all AP cases. Additionally, the present study focused on differences in feeding patterns; thus, caloric intake was increased every day until the target amount was achieved. However, more details of feeding, such as how to advance feeding and how much increase in calories and fat, still need to be explored.

Pediatric AP not only impacts the physical health of children but also imposes additional burdens on their families. Improving treatment strategies for this condition holds long-term significance for both individuals and society. The present research demonstrated that early oral feeding can effectively replace NG feeding in children with mild-to-moderate AP, thereby providing critical support for early nutritional recommendations to minimize unnecessary invasive procedures in young patients. Moving forward, it remains essential to continue in-depth research on early nutritional strategies for severe pediatric AP or AP caused by different causes, and to conduct similar multicenter studies, as the potential impact could be even more profound.

YH, HZ were involved in conceptualization, data curation, formal analysis and writing-original draft. LCF, XYJ, XD were involved in data curation, formal analysis and visualization. LQZ was involved in software and visualization. JDY and JC were involved in methodology, resources and supervision. JC and JGL were involved in project administration, validation and writing-review & editing. All authors agree to be accountable for all aspects of the work.

There is no sponsorship/funding in the preparation of this article. All project funding was obtained through internal self-funding sources.

The trial was registered at the Chinese Clinical Trial Registry (https://www.chictr.org.cn, ChiCTR2100050915).

Deidentified individual participant data (including data dictionaries) will be made available, in addition to study protocols, the statistical analysis plan, and the informed consent form. The data will be made available upon publication to researchers who provide a methodologically sound proposal for use in achieving the goals of the approved proposal. Proposals should be submitted to jingan@zju.edu.cn.