Risk factors for enteral nutrition intolerance and its impact on prognosis in patients with severe acute pancreatitis

Abstract

BACKGROUND

Severe acute pancreatitis (SAP) affects 10%-20% of acute pancreatitis patients, with mortality rates of up to 30%, requiring early enteral nutrition (EN) within 24-48 hours per the European Society for Clinical Nutrition and Metabolism guidelines. However, intolerance to EN occurs in 25%-40% of SAP patients due to gastrointestinal dysfunction, leading to prolonged intensive care unit stays, increased infections, and compromised survival outcomes.

AIM

To investigate the risk factors for EN intolerance in patients with SAP and its impact on prognosis, providing evidence for clinical nutrition therapy strategies.

METHODS

A retrospective cohort study design was adopted, and the clinical data of 195 SAP patients of our hospital were collected from January 2019 to June 2025. Patients were divided into an intolerance group (76 patients) and a tolerance group (119 patients) according to their EN intolerance status. Multivariate logistic regression analysis was used to identify independent risk factors for EN intolerance, and a Cox proportional hazards regression model was used to analyze independent factors affecting 28-day mortality.

RESULTS

The incidence of EN intolerance in SAP patients was 39.0%. Multivariate logistic regression analysis revealed that the Acute Physiology and Chronic Health Evaluation II score [odds ratio (OR) = 1.124, 95% confidence interval (CI): 1.042-1.213, P = 0.002], serum albumin level (OR = 0.879, 95%CI: 0.801-0.965, P = 0.006), and intra-abdominal pressure (OR = 1.152, 95%CI: 1.063-1.248, P = 0.001) were independent risk factors for EN intolerance. The 28-day mortality rate in the intolerance group was significantly greater than that in the tolerance group (25.0% vs 12.6%, P = 0.025). Cox regression analysis revealed that EN intolerance (hazard ratio = 2.164, 95%CI: 1.127-4.156, P = 0.020) was an independent risk factor for 28-day mortality

CONCLUSION

The incidence of EN intolerance in SAP patients is high, with the Acute Physiology and Chronic Health Evaluation II score, serum albumin level, and intra-abdominal pressure being independent risk factors. EN intolerance significantly increases the risk of mortality in patients and is an independent risk factor affecting prognosis. Early identification of high-risk factors and the development of individualized nutrition therapy strategies are crucial for improving patient outcomes.

Key Words: Severe acute pancreatitis; Enteral nutrition; Intolerance; Risk factors; Prognosis

Core Tip: The incidence, risk factors, and prognostic significance of enteral nutrition (EN) intolerance in patients with severe acute pancreatitis were examined in this retrospective cohort study. The rate of EN intolerance among 195 patients was 39.0%. Multivariate analysis revealed that the Acute Physiology and Chronic Health Evaluation II score, serum albumin concentration, and intra-abdominal pressure were independent predictors. Patients with EN intolerance had significantly greater 28-day mortality. In addition to offering evidence for the optimization of EN therapy in critical care settings, these findings emphasize the significance of early risk assessment and customized nutritional measures to enhance outcomes for severe acute pancreatitis patients.

INTRODUCTION

Acute pancreatitis is one of the most common diseases of the digestive system, and its incidence has shown a continuous upward trend in recent years, making it the leading cause of gastrointestinal-related hospitalizations in the United States[1]. Approximately 10%-20% of acute pancreatitis patients develop severe acute pancreatitis (SAP), which is characterized by pancreatic necrosis, a systemic inflammatory response, and organ failure, with mortality rates of up to 30%[2,3].

An essential part of a thorough SAP treatment plan is nutrition therapy. In order to preserve intestinal barrier integrity, lower the risk of infection complications, and boost immunological function, SAP patients should begin enteral nutrition (EN) therapy as soon as possible after admission, according to the European Society for Clinical Nutrition and Metabolism’s (ESPEN) guidelines[4]. However, EN intolerance is common in SAP patients, occurring at a rate of 25% to 40% due to gastrointestinal dysfunction brought on by pancreatic inflammation[5,6].

Enteral nutritional intolerance not only makes it difficult to achieve nutritional goals but also may prolong mechanical ventilation time, increase infection risk, prolong the intensive care unit (ICU) stay, and even affect patient survival prognosis[7]. Thus, the formulation of customized nutrition therapy plans and the early identification of risk factors for EN intolerance are crucial clinically for enhancing SAP patient outcomes.

Currently, research on risk factors for EN intolerance in SAP patients is relatively limited, with inconsistent results[8,9]. Previous studies have focused mainly on gastrointestinal symptoms and physiological indicators, with less comprehensive analyses of disease severity scores and organ function status[10]. Additionally, prospective studies on the impact of EN intolerance on SAP patient prognosis are relatively rare.

This study aimed to systematically analyze the clinical characteristics and risk factors for EN intolerance in SAP patients through a retrospective cohort study, evaluate its impact on patient prognosis, and provide scientific evidence for the development of more precise clinical nutritional therapy strategies.

MATERIALS AND METHODS

Study design

This study adopted a retrospective cohort study design in which the clinical data of SAP patients of the First Affiliated Hospital of Anhui Medical University from January 2019 to June 2025 were collected. The research protocol was approved by the hospital ethics committee, and informed consent was waived because of the retrospective nature of the analysis.

Sample size calculation

Preliminary experimental data revealed that the incidence of EN intolerance was approximately 40%, with an expected odds ratio (OR) of 2.5, α = 0.05, and β = 0.20. Using PASS 15.0 software, the required sample size was calculated to be 180 cases. Considering a 10% incomplete data rate and patients not meeting the inclusion/exclusion criteria, 200 patients were planned for inclusion, with a final effective sample size of 195 cases.

Study subjects

Inclusion criteria: (1) Age ≥ 18 years; (2) Meeting SAP diagnostic criteria[11]: Clinical manifestations of acute pancreatitis with serum amylase and/or lipase elevation ≥ 3 times the upper limit of normal; imaging findings suggestive of pancreatic inflammation; and concurrent organ failure and/or local complications; (3) After hospitalization, once intestinal function recovers, EN begins; and (4) Complete clinical data.

Exclusion criteria: (1) Acute pancreatitis associated with pregnancy; (2) Incapacity to undergo EN therapy for any other reason; and (3) Hospitalization for less than seven days.

EN protocol

A consistent enteral feeding strategy was used in this study to guarantee that nutritional therapy was uniform and comparable for every patient. The dietary routes were chosen in accordance with known clinical practice recommendations and the principles of evidence-based medicine. Gastric feeding using nasogastric tubes is the preferred method of EN; jejunal feeding through nasojejunal tubes was chosen as a backup strategy in cases of severe gastrointestinal dysfunction, chronic vomiting, or gastric retention.

Standard whole protein EN formulas were utilized with an energy density of 1.0-1.5 kcal/mL, protein content of 16%-20%, fat content of 20%-35%, and carbohydrate content of 45%-65%. Individualized principles guided the start and progression of nutrition therapy, guaranteeing patient safety and effectively reaching nutritional objectives. Enteral nourishment was started as soon as bowel function recovered as part of the nutrition implementation program. In order to obtain the desired volume, EN therapy was started with an infusion rate of 20-25 mL/hour and increased by 10-20 mL/hour every 12-24 hours based on patient tolerance. With all calculations based on actual body weight rather than ideal body weight, the target energy was set at 25-30 kcal/(kg × day) and the target protein energy was set at 1.2-1.5 g/(kg × day). This method was chosen to conform with current ESPEN guidelines for acute pancreatitis and conventional ICU practice standards. To guarantee internal consistency and direct relevance to standard clinical practice, all weight-based nutritional parameters were regularly computed using actual body weight throughout the trial. When clinically required, intermittent infusion is used in conjunction with continuous pump infusion techniques for nutrition administration.

Observation indicators

This study established a comprehensive indicator observation system covering the definition of EN intolerance, patient baseline characteristics, disease severity, laboratory tests, gastrointestinal function assessment, nutrition therapy effectiveness, treatment measures, complications, and prognosis across multiple dimensions to ensure the scientific validity and clinical value of the research results.

Definition of EN intolerance: The criteria for determining EN intolerance were based on international authoritative guidelines to ensure the objectivity and consistency of the diagnosis. According to the ESPEN guidelines, EN intolerance was defined as any of the following conditions persisting for more than 24 hours[12]: (1) Gastric residual volume > 500 mL (measured twice/day at 4-6 hours intervals using a 50 mL syringe to aspirate gastric contents before infusion); (2) Worsening of abdominal distension and pain with decreased or absent bowel sounds; (3) Nausea and vomiting ≥ 3 times/day; (4) Diarrhea > 3 times/day or stool volume > 500 mL/day; (5) Intra-abdominal pressure > 20 mmHg (measured by the bladder method every 24 hours); and (6) Need to stop EN or reduce to < 50% of target intake due to gastrointestinal symptoms for ≥ 24 hours.

Baseline data: The collection of patient baseline data helps analyze the impact of different characteristics on EN intolerance, providing basic data for risk factor analysis. Patient age, sex, body mass index, medical history (diabetes, hypertension, cardiovascular and cerebrovascular diseases), etiology of acute pancreatitis (biliary, alcoholic, hyperlipidemic, idiopathic), time from onset to hospital admission, and time from admission to ICU transfer were recorded.

Disease severity assessment: Determining the prognosis of patients and directing treatment depend heavily on an accurate assessment of the severity of the condition. Multiple grading systems were utilized in this study to ensure a thorough review: (1) Acute Physiology and Chronic Health Evaluation (APACHE) II score (worst value within 24 hours of ICU admission)[12]; (2) Sequential organ failure assessment score (at ICU admission); (3) Modified computed tomography severity index (within 72 hours of admission); and (4) Number and types of organ failure (respiratory, circulatory, renal, hepatic, hematologic, or neurologic).

Laboratory parameters: The pathophysiological condition and response to treatment of patients can be objectively reflected by dynamic monitoring of laboratory markers. Within 24 hours following ICU admission, the worst values for each laboratory parameter were used for analysis: (1) Inflammatory markers: White blood cell count, neutrophil percentage, C-reactive protein, and procalcitonin; (2) Pancreatic enzymes: Serum amylase and lipase; (3) Nutritional markers: Serum albumin (ALB), prealbumin, and hemoglobin; (4) Organ function markers: Liver function (alanine aminotransferase, aspartate aminotransferase, and total bilirubin), renal function (serum creatinine and blood urea nitrogen), coagulation function (prothrombin time, activated partial thromboplastin time, and international normalized ratio); and (5) Metabolic markers: Blood glucose and electrolytes (sodium, potassium, chloride), and lactate.

Gastrointestinal function assessment: Gastrointestinal function status is a core indicator for determining EN tolerance and requires standardized assessment methods[13]: (1) Intra-abdominal pressure monitoring (bladder method, monitored every 24 hours); (2) Gastric residual volume measurement (measured before each infusion); and (3) Gastrointestinal symptoms, such as abdominal pain and distension severity, bowel sounds (auscultation assessment), and bowel movements (abnormal bowel movements defined as constipation > 3 days or diarrhea > 3 times/day).

Nutrition therapy-related indicators: Assessment of nutritional therapy effectiveness helps elucidate the specific impact of EN intolerance on nutritional intake: (1) Nutrition route and tube type; (2) Nutrition initiation time; (3) Time to achieve the target energy and achievement rate; (4) Daily actual energy and protein intake; (5) Number and reasons for nutrient interruption; and (6) Parenteral nutrition supplementation ratio.

Treatment measures: Detailed recording of various treatment measures helps analyze the disease severity and treatment needs of patients with EN intolerance: (1) Mechanical ventilation duration and parameters; (2) Vasoactive drug use; (3) Gastrointestinal motility drug use; (4) Antibiotic therapy; (5) Continuous renal replacement therapy; and (6) Surgical interventions (percutaneous drainage, necrosectomy, etc.).

Complication definitions: The occurrence of complications is an important indicator for assessing patient prognosis and requires standardized diagnostic criteria: (1) Infectious complications: Infected pancreatic necrosis (imaging + clinical manifestations + pathogen evidence), abdominal infection, bloodstream infection (positive blood culture), and pulmonary infection; (2) Noninfectious complications: Pancreatic fistula (international study group definition), gastrointestinal bleeding (hemoglobin decrease > 20 g/L with melena or hematemesis), and thromboembolism; and (3) Nutrition-related complications: Aspiration (radiologically confirmed), electrolyte imbalance (sodium < 135 mmol/L or > 145 mmol/L, potassium < 3.5 mmol/L or > 5.5 mmol/L), and hyperglycemia (blood glucose mmol/L > 11.1 mmol/L).

Prognostic indicators: The setting of prognostic indicators clarified the primary and secondary endpoints of the study, providing a basis for statistical analysis: (1) Primary endpoint: 28-day all-cause mortality; and (2) Secondary endpoints: ICU length of stay and total hospital length of stay.

Data collection methods

To ensure the accuracy and completeness of data collection, this study established standardized data collection procedures and quality control measures. Trained researchers extracted relevant data from the hospital information system, laboratory information system, and critical care information system via standardized questionnaires. All the data were independently entered and verified by two researchers, with discrepancies resolved through discussion or third-party consultation. Double data entry was used to ensure data quality. Missing data < 5% were analyzed via complete case analysis, and 5%-20% of the missing data were handled via multiple imputation methods.

Statistical analysis

This study adopted rigorous scientific statistical methods to ensure the reliability and validity of the analysis results. The statistical analysis process included descriptive analysis, univariate analysis, and multivariate analysis.

Statistical analysis was performed via SPSS 26.0 software. Continuous variables were tested for normality via the Shapiro-Wilk test. Normally distributed data are expressed as the means ± SD and were compared between groups via independent samples t tests. Nonnormally distributed data are expressed as medians and interquartile ranges [median (Q1, Q3)] and were compared via the Mann-Whitney U test. Categorical variables are expressed as numbers and percentages, n (%) and were compared via the χ² test or Fisher’s exact test.

Multiple comparisons were controlled for the false discovery rate via the Benjamini-Hochberg method. Univariate analysis was used to screen for factors influencing EN intolerance, with variables with P < 0.10 included in multivariate logistic regression analysis. Multicollinearity diagnosis was performed before regression analysis (variance inflation factor < 5), and forward selection was used to determine independent risk factors. The bootstrap method (1000 resamples) was used for internal validation. The Kaplan-Meier method was used to plot survival curves, and the log-rank test was used to compare survival differences. A Cox proportional hazards regression model was used to analyze independent factors affecting prognosis. The proportional hazards assumption was tested before model construction, and hazard ratios (HRs) with 95% confidence intervals were reported. Model goodness of fit was evaluated via the concordance index. The significance level was set at α = 0.05, and P < 0.05 after false discovery rate correction was considered statistically significant.

RESULTS

Patient baseline characteristics

This study included 195 SAP patients, including 128 males (65.6%) and 67 females (34.4%), with a mean age of 52.3 ± 14.7 years. Patients were divided into an intolerance group (76 patients, 39.0%) and a tolerance group (119 patients, 61.0%). There were no statistically significant differences between the two groups in terms of age, sex, body mass index, or other baseline characteristics. The etiology composition was not significantly different between the groups (P = 0.279). The time from onset to hospital admission (P = 0.012) and the time from admission to ICU transfer (P = 0.001) were significantly longer in the intolerance group than in the tolerance group (Table 1).

Table 1 Comparison of baseline characteristics between two groups, n (%).

Characteristics	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
Age (years), mean ± SD	54.1 ± 15.2	51.2 ± 14.3	1.328	0.186
Gender			0.234	0.629
Male	48 (63.2)	80 (67.2)
Female	28 (36.8)	39 (32.8)
BMI (kg/m²), mean ± SD	24.8 ± 3.6	24.2 ± 3.4	1.142	0.255
Medical history
Diabetes	18 (23.7)	25 (21.0)	0.186	0.666
Hypertension	32 (42.1)	45 (37.8)	0.355	0.551
Cardiovascular disease	14 (18.4)	19 (16.0)	0.189	0.663
Etiology			3.847	0.279
Biliary	38 (50.0)	51 (42.9)
Alcoholic	18 (23.7)	34 (28.6)
Hyperlipidemic	14 (18.4)	17 (14.3)
Idiopathic	6 (7.9)	17 (14.3)
Time from onset to admission (hour), mean ± SD	16.8 ± 7.2	14.2 ± 6.1	2.541	0.012
Time from admission to ICU (hour), mean ± SD	11.4 ± 4.8	9.1 ± 3.7	3.658	0.001

BMI: Body mass index; ICU: Intensive care unit.

Disease severity assessment

The intolerance group had significantly greater disease severity than did the tolerance group. The APACHE II score (P < 0.001), sequential organ failure assessment score (P < 0.001), and modified computed tomography severity index score (P < 0.001) were significantly elevated. The number of organ failures was significantly greater in the intolerance group than in the tolerance group (P = 0.002). Among the organ failure types, respiratory (P = 0.010), circulatory (P = 0.020), and renal failure (P = 0.023) were more common in the intolerance group (Table 2).

Table 2 Comparison of disease severity between two groups, n (%).

Indicators	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
APACHE II score, mean ± SD	21.8 ± 6.4	18.2 ± 5.7	4.083	< 0.001
SOFA score, mean ± SD	8.6 ± 3.2	6.9 ± 2.8	3.884	< 0.001
MCTSI score, mean ± SD	7.4 ± 2.1	6.1 ± 2.3	3.946	< 0.001
Number of organ failures, mean ± SD	2.3 ± 1.1	1.8 ± 0.9	3.203	0.002
Types of organ failure
Respiratory	68 (89.5)	89 (74.8)	6.648	0.010
Circulatory	52 (68.4)	61 (51.3)	5.428	0.020
Renal	41 (53.9)	44 (37.0)	5.135	0.023
Hepatic	29 (38.2)	32 (26.9)	2.637	0.104
Hematologic	18 (23.7)	19 (16.0)	1.813	0.178
Neurologic	12 (15.8)	14 (11.8)	0.694	0.405

APACHE: Acute Physiology and Chronic Health Evaluation; SOFA: Sequential organ failure assessment; MCTSI: Modified computed tomography severity index.

Laboratory parameters

Compared with the tolerance group, the intolerance group presented significantly greater inflammatory marker levels, including white blood cell counts (P = 0.002), neutrophil percentages (P = 0.008), C-reactive protein levels (P < 0.001), and procalcitonin levels (P = 0.001), than did the tolerance group. There were no significant differences in pancreatic enzyme markers between the groups. Nutritional markers, including serum ALB (P = 0.001) and prealbumin (P < 0.001), were significantly lower in the intolerance group. Organ function indicators, including liver function, renal function, and coagulation function, were significantly worse in the intolerance group than in the tolerance group (P < 0.05). Compared with those in the tolerant group, the levels of metabolic markers, including blood glucose (P = 0.001) and lactate (P = 0.001), were significantly greater, whereas the level of serum sodium (P = 0.048) was significantly lower (Table 3).

Table 3 Comparison of laboratory parameters between two groups, mean ± SD.

Indicators	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
Inflammatory markers
WBC count (× 109/L)	16.8 ± 6.2	14.2 ± 5.4	3.074	0.002
Neutrophil (%)	82.4 ± 8.6	78.9 ± 9.2	2.664	0.008
CRP (mg/L)1	186.4 (98.2, 274.6)	142.6 (76.8, 208.4)	-4.247	< 0.001
PCT (μg/L)1	3.8 (1.6, 8.2)	2.1 (0.8, 4.9)	-3.247	0.001
Pancreatic enzymes
Serum amylase (U/L)1	428 (186, 862)	384 (168, 726)	-1.428	0.153
Lipase (U/L)1	1246 (542, 2184)	1087 (468, 1896)	-1.723	0.085
Nutritional markers
Serum albumin (g/L)	26.8 ± 4.2	29.1 ± 4.8	-3.384	0.001
Prealbumin (μg/mL)	148.6 ± 32.4	168.2 ± 38.7	-3.689	< 0.001
Hemoglobin (g/L)	108.4 ± 18.6	112.8 ± 20.2	-1.534	0.126
Organ function markers
ALT (U/L)1	89.6 (42.8, 156.4)	68.2 (35.4, 128.6)	-2.146	0.032
AST (U/L)1	106.8 (58.2, 178.4)	84.6 (46.8, 142.2)	-2.218	0.026
Total bilirubin (μmol/L)1	48.6 (22.4, 86.8)	36.4 (18.2, 64.2)	-2.364	0.018
Serum creatinine (μmol/L)1	142.8 (86.4, 208.6)	108.2 (68.4, 156.8)	-2.968	0.003
BUN (mmol/L)1	12.4 (7.8, 18.6)	9.2 (5.6, 14.8)	-2.642	0.008
PT (second)	16.8 ± 4.2	14.6 ± 3.8	3.724	0.001
APTT (second)	42.6 ± 8.4	38.2 ± 7.6	3.698	0.001
INR1	1.52 (1.18, 1.96)	1.24 (1.06, 1.58)	-3.842	< 0.001
Metabolic markers
Blood glucose (mmol/L)	12.8 ± 4.6	10.4 ± 3.8	3.847	0.001
Serum sodium (mmol/L)	138.4 ± 6.8	140.2 ± 5.4	-1.984	0.048
Serum potassium (mmol/L)	3.8 ± 0.8	3.9 ± 0.7	-0.924	0.357
Serum chloride (mmol/L)	102.4 ± 8.2	104.1 ± 7.6	-1.468	0.144
Lactate (mmol/L)	2.8 ± 1.4	2.2 ± 1.1	3.242	0.001

¹Non-normally distributed data, expressed as median (interquartile range).

WBC: White blood cell; CRP: C-reactive protein; PCT: Procalcitonin; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; BUN: Blood urea nitrogen; PT: Prothrombin time; APTT: Activated partial thromboplastin time; INR: International normalized ratio.

Nutrition therapy-related indicators

Compared with the tolerant group, the intolerance group had a significantly greater proportion of jejunal feeding (P < 0.001). Compared with the tolerance group, the intolerance group had a significantly delayed nutrition initiation time (P = 0.001), prolonged time to achieve the target energy (P < 0.001), significantly reduced target energy achievement rate (P < 0.001), significantly decreased daily actual energy and protein intake (P < 0.001), significantly increased nutrition interruption frequency (P < 0.001), and a significantly greater parenteral nutrition supplementation ratio (P < 0.001) (Table 4).

Table 4 Comparison of nutrition therapy-related indicators between two groups, n (%).

Indicators	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
Nutrition route			17.428	< 0.001
Gastric feeding	34 (44.7)	89 (74.8)
Jejunal feeding	42 (55.3)	30 (25.2)
Tube type			21.846	< 0.001
Nasogastric tube	32 (42.1)	78 (65.5)
Nasojejunal tube	38 (50.0)	35 (29.4)
PEG	6 (7.9)	6 (5.0)
Nutrition initiation time (hour), mean ± SD	38.4 ± 12.8	32.6 ± 10.4	3.368	0.001
Time to achieve target energy (day), mean ± SD	8.4 ± 3.2	4.6 ± 2.1	9.548	< 0.001
Target energy achievement rate (%), mean ± SD	68.4 ± 18.6	86.2 ± 14.8	-7.084	< 0.001
Daily actual energy intake (kcal/kg), mean ± SD	18.2 ± 6.4	24.8 ± 5.2	-7.652	< 0.001
Daily protein intake (g/kg), mean ± SD	0.9 ± 0.4	1.3 ± 0.3	-8.043	< 0.001
Nutrition interruption frequency, mean ± SD	3.8 ± 2.4	1.2 ± 1.6	8.426	< 0.001
Parenteral nutrition supplementation ratio (%), mean ± SD	42.6 ± 18.4	18.2 ± 12.6	10.254	< 0.001

PEG: Percutaneous endoscopic gastrostomy.

Treatment measures

The intolerance group required more organ support therapy. The mechanical ventilation utilization rate (P = 0.012), positive end-expiratory pressure setting (P = 0.001), fraction of inspired oxygen setting (P < 0.001), vasoactive drug utilization rate (P = 0.005), duration (P < 0.001), gastrointestinal motility drug utilization rate (P < 0.001), antibiotic therapy duration (P < 0.001), and continuous renal replacement therapy utilization rate (P = 0.022) were significantly greater in the tolerance group than in the tolerance group. There was no significant difference in surgical intervention rates between the groups (P = 0.116) (Table 5).

Table 5 Comparison of treatment measures between two groups, n (%).

Indicators	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
Respiratory support
Mechanical ventilation	68 (89.5)	88 (73.9)	6.298	0.012
PEEP (cmH2O), mean ± SD	8.4 ± 2.6	7.2 ± 2.1	3.468	0.001
FiO2 (%), mean ± SD	52.4 ± 18.2	42.8 ± 14.6	4.227	< 0.001
Circulatory support
Vasoactive drug use	60 (78.9)	70 (58.8)	7.941	0.005
Vasoactive drug duration (day), mean ± SD	8.6 ± 4.2	5.9 ± 3.1	4.873	< 0.001
Gastrointestinal therapy
Gastrointestinal motility drug use	68 (89.5)	70 (58.8)	19.586	< 0.001
Other therapies
Antibiotic therapy duration (day), mean ± SD	14.8 ± 6.2	11.2 ± 4.8	4.284	< 0.001
Continuous renal replacement therapy	32 (42.1)	31 (26.1)	5.259	0.022
Surgical intervention	28 (36.8)	31 (26.1)	2.472	0.116

PEEP: Positive end expiratory pressure.

Complications

The total incidence of infectious complications in the intolerance group was significantly greater than that in the tolerance group (P = 0.002), with statistically significant differences in the incidence of infected pancreatic necrosis (P = 0.020) and bloodstream infections (P = 0.021). There were no significant differences in noninfectious complications between the groups. Among nutrition-related complications, the intolerance group had significantly greater incidences of aspiration (P = 0.024), electrolyte imbalance (P = 0.006), and hyperglycemia (P = 0.013) than did the tolerance group (Table 6).

Table 6 Comparison of complications between two groups, n (%).

Complications	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
Infectious complications
Total incidence	50 (65.8)	51 (42.9)	9.644	0.002
Infected pancreatic necrosis	26 (34.2)	23 (19.3)	5.399	0.020
Abdominal infection	18 (23.7)	21 (17.6)	1.023	0.312
Bloodstream infection	22 (28.9)	18 (15.1)	5.330	0.021
Pulmonary infection	35 (46.1)	42 (35.3)	2.206	0.137
Non-infectious complications
Pancreatic fistula	16 (21.1)	16 (13.4)	1.943	0.164
Gastrointestinal bleeding	12 (15.8)	13 (10.9)	0.946	0.331
Thromboembolism	8 (10.5)	9 (7.6)	0.474	0.491
Nutrition-related complications
Aspiration	14 (18.4)	9 (7.6)	5.119	0.024
Electrolyte imbalance	40 (52.6)	38 (31.9)	7.668	0.006
Hyperglycemia	54 (71.1)	63 (52.9)	6.129	0.013

Prognostic indicators

The 28-day mortality rate in the intolerance group was significantly greater than that in the tolerance group (P = 0.025). The ICU length of stay (P < 0.001) and total hospital length of stay (P < 0.001) were both significantly longer in the intolerance group than in the tolerance group (Table 7).

Table 7 Comparison of prognostic indicators between two groups, n (%).

Indicators	Intolerance group (n = 76)	Tolerance group (n = 119)	Test statistic	P value
Primary endpoint
28-day mortality	19 (25.0)	15 (12.6)	5.019	0.025
Secondary endpoints, mean ± SD
ICU length of stay (day)	18.6 ± 8.4	13.2 ± 6.1	4.989	< 0.001
Total hospital length of stay (day)	32.4 ± 12.8	25.7 ± 9.6	4.067	< 0.001

ICU: Intensive care unit.

Multivariate analysis of risk factors for EN intolerance

Eighteen candidate variables were screened through univariate analysis for multicollinearity diagnosis. After excluding variables with multicollinearity, 16 variables were included in the multivariate logistic regression analysis. According to Table 8, the final model showed that the intra-abdominal pressure (P = 0.001), serum ALB concentration (P = 0.006), and APACHE II score (P = 0.002) were independent risk variables for EN intolerance.

Table 8 Multivariate logistic regression analysis of enteral nutrition intolerance.

Variables	β	SE	Wald χ²	P value	OR	95%CI
APACHE II score	0.117	0.038	9.424	0.002	1.124	1.042-1.213
Serum albumin	-0.129	0.047	7.532	0.006	0.879	0.801-0.965
Intra-abdominal pressure	0.141	0.041	11.847	0.001	1.152	1.063-1.248
Constant	-0.268	1.824	0.022	0.882	0.765	-

Model goodness of fit: Hosmer-Lemeshow test P = 0.553; concordance index = 0.812 (95% confidence interval: 0.752-0.871); Bootstrap validation concordance index = 0.798 (95% confidence interval: 0.741-0.855). OR: Odds ratio; CI: Confidence interval; APACHE: Acute Physiology and Chronic Health Evaluation.

Survival analysis

Kaplan-Meier survival analysis revealed that patients in the intolerance group had lower 28-day survival rates than those in the tolerance group did (P = 0.016). Cox proportional hazards regression analysis revealed that EN intolerance (P = 0.020), the APACHE II score (P = 0.007), and the number of organ failures (P = 0.003) were independent risk factors for 28-day mortality (Table 9).

Table 9 Cox regression analysis of factors affecting 28-day mortality.

Variables	β	SE	Wald χ²	P value	HR	95%CI
Enteral nutrition intolerance	0.771	0.333	5.366	0.020	2.164	1.127-4.156
APACHE II score	0.083	0.031	7.305	0.007	1.086	1.023-1.153
Number of organ failures	0.485	0.162	8.954	0.003	1.624	1.186-2.225

Cox regression model concordance index = 0.778 (95% confidence interval: 0.701-0.855); Bootstrap validation concordance index = 0.765 (95% confidence interval: 0.698-0.832); proportional hazards assumption test P = 0.276. HR: Hazard ratio; CI: Confidence interval; APACHE: Acute Physiology and Chronic Health Evaluation.

DISCUSSION

This study systematically analyzed clinical data from 195 SAP patients and reported that the incidence of EN intolerance was 39.0%, which is consistent with previously reported rates of 25%-40%[14,15]. Multivariate analysis revealed that the APACHE II score, serum ALB level, and intra-abdominal pressure were independent risk factors for EN intolerance. More importantly, this study confirmed that EN intolerance significantly increases 28-day mortality in patients and is an independent risk factor affecting prognosis.

This study revealed that the APACHE II score was an independent risk factor for EN intolerance (OR = 1.124, P = 0.002), which is consistent with the findings of multiple recent studies[16,17]. The APACHE II score, as a comprehensive indicator reflecting disease severity, includes not only physiological parameters but also chronic health status and can better predict the prognosis of SAP patients[18]. Higher scores indicate more severe patient conditions, stronger systemic inflammatory responses, and more severe gastrointestinal dysfunction, increasing the likelihood of EN intolerance.

The study revealed that the number of organ failures in the intolerance group was significantly greater than that in the tolerance group (2.3 ± 1.1 vs 1.8 ± 0.9, P = 0.002), with significantly greater incidences of respiratory, circulatory, and renal failure. This phenomenon may be related to the pathophysiological mechanism of SAP: Severe pancreatic inflammation leads to massive release of inflammatory mediators, causing systemic inflammatory response syndrome, which subsequently leads to multiple organ dysfunction syndrome[19]. The gastrointestinal tract, one of the most vulnerable organs in the body, is often the first organ affected in the pathological process of multiple organ failure, manifesting as decreased gastrointestinal motility and impaired intestinal barrier function, ultimately leading to EN intolerance.

The gastrointestinal tract's functioning state is directly reflected in delayed stomach emptying, altered bowel sounds, and abdominal distension[20]. SAP patients often experience gastrointestinal dysfunction due to pancreatic inflammation, which manifests as delayed gastric emptying, slowed intestinal peristalsis, and increased intestinal permeability[21].

This study’s data support this viewpoint: The gastric residual volume in the intolerance group was significantly greater than that in the tolerance group (368 ± 156 mL vs 186 ± 98 mL, P < 0.001), and the abdominal distension was also significantly greater. These clinical manifestations directly make EN implementation difficult, not only affecting nutrient infusion but also potentially increasing the risk of serious complications such as aspiration.

Elevated intra-abdominal pressure was another independent risk factor identified in this study (OR = 1.152, P = 0.001). The mean intra-abdominal pressure in the intolerance group was significantly greater than that in the tolerance group (18.6 ± 4.8 mmHg vs 14.2 ± 3.6 mmHg, P < 0.001). Elevated intra-abdominal pressure is relatively common in SAP and is caused mainly by pancreatic and surrounding tissue edema, abdominal fluid accumulation, and bowel distension[22].

Elevated intra-abdominal pressure has multiple adverse effects on gastrointestinal function. First, elevated intra-abdominal pressure compresses the gastrointestinal tract, affecting gastrointestinal motility and blood perfusion, leading to delayed gastric emptying and intestinal dysfunction[23]. Second, elevated intra-abdominal pressure may also affect diaphragmatic movement, exacerbating respiratory dysfunction and further deteriorating the patient’s overall condition[24]. ESPEN guidelines clearly recommend that when intra-abdominal pressure > 15 mmHg, EN should be implemented cautiously, and when intra-abdominal pressure > 20 mmHg, consideration should be given to temporarily stopping EN[25]. The results of this study further confirm the important value of intra-abdominal pressure monitoring in SAP nutrition management.

Our analysis showed that a lower blood ALB concentration was a significant predictor of EN intolerance risk (OR = 0.879, P = 0.006), suggesting that hypoalbuminemia is a risk factor in and of itself with major clinical consequences. Reduced hepatic protein synthesis, increased protein catabolism, and increased vascular permeability due to systemic inflammatory responses are the three pathophysiological mechanisms that combine to cause hypoalbuminemia, a common clinical symptom in SAP patients[26]. Gastrointestinal dysfunction and ALB deficit are related through a number of interrelated routes. ALB depletion, the main factor influencing plasma colloid osmotic pressure, causes tissue fluid to accumulate, especially in the walls of the gastrointestinal tract, which impairs proper peristaltic activity[27]. Additionally, the integrity of the intestinal barrier depends on sufficient ALB levels; a lack of these can lead to increased mucosal permeability, which can allow bacterial translocation and increase vulnerability to infectious consequences[28]. This leads to a vicious cycle whereby gastrointestinal dysfunction is made worse by malnutrition, which further deteriorates nutritional status. Our investigation’s most important clinical finding may have to do with the serious prognostic implications of EN intolerance. According to our Cox regression modeling, patients with feeding intolerance had a 116.4% higher risk of dying after 28 days (hazard ratio = 2.164, P = 0.020), supporting other studies that connected dietary issues to unfavorable clinical outcomes[29,30]. This mortality association is caused by intricate and multifaceted pathophysiological mechanisms. First and foremost, eating intolerance significantly hinders the delivery of sufficient calories and protein. Our data showed significant differences in nutritional performance between groups: Compared to patients tolerating EN, patients with intolerance received significantly less protein (0.9 ± 0.4 g/kg vs 1.3 ± 0.3 g/kg, P < 0.001) and energy (18.2 ± 6.4 kcal/kg vs 24.8 ± 5.2 kcal/kg, P < 0.001). Cascade physiological effects, such as weakened immunological responses, poor tissue repair processes, and gradual skeletal muscle wasting, are brought on by such nutritional deficiencies[31].

Second, intolerance to EN often indicates severely impaired gastrointestinal function, which itself is a manifestation of disease severity. Gastrointestinal dysfunction not only affects nutritional absorption but also may lead to impaired intestinal barrier function and bacterial translocation, increasing infection risk. This study revealed that the incidence of infectious complications in the intolerance group was significantly greater than that in the tolerance group (65.8% vs 42.9%, P = 0.002), further confirming the important role of intestinal function in maintaining body homeostasis.

Limitations

This study has several limitations. First, as a single-center retrospective study with a relatively small sample size, selection bias may exist. Second, although the definition of EN intolerance referenced international guidelines, some symptom assessments remain subjective. Third, this study did not analyze factors such as intestinal microecology and gastrointestinal hormones that may affect EN tolerance. Finally, the lack of long-term follow-up data prevents assessment of the impact of EN intolerance on patients’ long-term prognosis.

Future research directions

On the basis of the results of this study, future research can be developed in the following directions: First, multicenter prospective studies with larger sample sizes should be conducted to validate the universality of the results of this study; second, the roles of intestinal microecology and gastrointestinal hormones in EN intolerance mechanisms should be explored in detail; third, prediction models for EN intolerance should be developed and validated to provide more precise tools for clinical decision-making; and finally, the effects of individualized nutrition intervention strategies for improving SAP patient outcomes should be studied.

CONCLUSION

The incidence of EN intolerance in SAP patients is high (39.0%), with the APACHE II score, serum ALB level, and intra-abdominal pressure being independent risk factors. Enteral nutritional intolerance significantly increases patient 28-day mortality and is an independent risk factor affecting prognosis. Clinicians should establish risk assessment systems for EN intolerance, identify high-risk patients early, and develop individualized nutritional therapy strategies to improve patient outcomes.