Frequency and predictors of small intestinal bacterial overgrowth in acute pancreatitis

Abstract

BACKGROUND

Acute pancreatitis (AP) is associated with intestinal dysmotility, barrier dysfunction, and dysbiosis, which may increase the risk of small intestinal bacterial overgrowth (SIBO). Evidence on the prevalence of SIBO in AP and its clinical correlates is limited.

AIM

To evaluate the frequency of SIBO in patients with AP and identify the clinical, laboratory, and imaging predictors of SIBO.

METHODS

This hospital-based case-control study was conducted at a tertiary gastroenterology unit in New Delhi, India, from December 2022 to June 2024. Consecutive adults with AP were enrolled as cases, and age- and sex-matched healthy controls (HCs) were included. The glucose hydrogen breath test was used for the diagnosis of SIBO. The predictors of SIBO were examined.

RESULTS

In total, 30 cases and 60 HC were included in the study. SIBO was detected in 12 (40.0%) cases, significantly higher than in HC (1.7%) (P < 0.001). SIBO occurred more frequently in severe AP [4 (100%)] than in moderate-severe AP [6 (75%)] and mild AP [2 (11.1%)] (P = 0.001). Predominant methane producers were more common among AP cases than among HC [12 (40.0%) vs 8 (13.3%); P = 0.004]. Univariate analysis revealed that factors such as abdominal bloating, obstipation, severe pancreatitis, ileus, systemic inflammatory response syndrome, acute necrotic collection, higher computed tomography severity index score, higher neutrophil-lymphocyte ratio, higher creatinine, and elevated high-sensitivity C-reactive protein were associated with SIBO.

CONCLUSION

SIBO was frequent in AP and was strongly associated with disease severity. Larger prospective studies are required to determine whether identifying and treating SIBO can improve clinical outcomes in AP.

Key Words: Acute pancreatitis; Small intestinal bacterial overgrowth; Glucose hydrogen breath test; Methane; Gut dysbiosis

Core Tip: Small intestinal bacterial overgrowth (SIBO) is an under-recognized disease entity in acute pancreatitis (AP). In this case-control study, SIBO was significantly more prevalent among patients with AP than among healthy controls, with a strong graded association with disease severity. SIBO was more common in those with systemic inflammatory response, ileus, and pancreatic necrosis. The predominant methane-producer status was more frequent in patients with AP than in healthy controls. These findings reinforce the clinical relevance of small intestinal dysbiosis in AP and provide a therapeutic window for early management to decrease the morbidity and mortality.

INTRODUCTION

Acute pancreatitis (AP), an inflammatory disease of the pancreas, is one of the most common causes of hospital admission for gastrointestinal conditions worldwide[1,2]. The clinical spectrum of AP ranges from a mild, self-limiting illness to severe pancreatitis complicated by persistent organ failure and infected necrosis[3]. AP is complicated by ileus, splanchnic hypoperfusion, systemic inflammatory response syndrome (SIRS), and the use of opioids and other supportive therapies, all of which can impair intestinal motility and permeability and disrupt the gut-pancreas axis[4].

In AP, abnormal trypsin secretion and pancreatic structural destruction lead to irregular pancreatic secretion, disrupting intestinal homeostasis and altering the composition of the intestinal flora[5,6]. Microcirculatory injury and hypovolemia, which cause gut mucosal ischemia and reperfusion injury in AP, result in loss of gut barrier integrity, translocation of gut flora, and local and systemic infections. Intestinal permeability is altered early in the course of AP, and this alteration may be associated with disease severity[7-10].

Infectious complications in patients with AP result from a series of events, including disturbances in the gut microbiota, imbalances in immune homeostasis, failure of the mucosal barrier, and proinflammatory responses. These factors collectively facilitate the translocation of intestinal bacteria, such as Escherichia coli and Enterococcus faecalis[11]. The severity of AP is linked to proportional changes in the intestinal microbiota, which lead to increased bacterial translocation, in turn resulting in systemic inflammation, infected pancreatic necrosis, and poor clinical outcomes, creating a vicious cycle that further deteriorates the condition. The mortality rate in patients with infected pancreatic necrosis and organ failure is approximately twice that in those with sterile pancreatic necrosis and organ failure[12]. In a meta-analysis by Wu et al[13], 59% of patients with AP exhibited impaired gut barrier function, and bacterial translocation was significantly linked to disease severity.

Animal studies have reported that necrotizing pancreatitis can disrupt gastrointestinal motility and lead to small intestinal bacterial overgrowth (SIBO)[14]. This condition is marked by gut microbiota dysregulation and develops when the homeostatic mechanisms controlling enteric bacterial populations are disrupted[15]. In AP, these mechanisms may be altered, which predisposes patients to SIBO. In addition, various pancreatic disorders, particularly chronic pancreatitis (CP) and pancreatic cancer, are associated with SIBO[16-18]. Nonetheless, the burden of SIBO in patients with AP, especially during the acute phase of the disease, remains poorly defined. Comprehending the link between SIBO and AP may provide insights into the pathophysiology of the disease and open avenues for targeted therapeutic interventions that modulate the gut microbiome.

Existing data on SIBO in AP are limited, heterogeneous, and largely derived from small studies that have used varied diagnostic methods[19-21]. Furthermore, the clinical predictors of SIBO in AP and their association with disease severity and systemic complications have not been comprehensively examined. Therefore, this study aimed to assess the prevalence of SIBO in patients with AP compared with healthy controls (HCs) and to identify clinical, biochemical, and radiographic predictors of the condition in patients with AP (Figure 1).

Figure 1 Study protocol. SIBO: Small intestinal bacterial overgrowth.

MATERIALS AND METHODS

Study design and setting

This case-control study was conducted in the Department of Gastroenterology at the Govind Ballabh Pant Institute of Postgraduate Medical Education and Research (GIPMER), New Delhi, a tertiary center for gastroenterology, from December 2022 to June 2024, spanning 18 months. The study was approved by the Institutional Ethics Committee (No. F.1/IEC/MAMC/94/06/2022/No. 484). Written informed consent was obtained from all participants.

Study participants and inclusion criteria

Consecutive adults aged 18-60 years who were admitted with AP were recruited as cases. Age- and sex-matched healthy adults without gastrointestinal symptoms or systemic illness were enrolled as controls.

Exclusion criteria

Individuals with conditions known to affect intestinal motility or predispose them to SIBO, such as impaired gastrointestinal motility (e.g., systemic sclerosis, diabetes mellitus, hypothyroidism, and intestinal pseudo-obstruction), were excluded. Moreover, patients with a history of prior gastrointestinal surgery or vagotomy, hepatobiliary disease, CP, gastrointestinal disease, pancreatic or gastrointestinal malignancy, or recent gastrointestinal infection were excluded. In addition, patients with a nasogastric tube; septic shock or acute respiratory distress syndrome; gastrointestinal bleeding; developmental pancreatic anomalies (e.g., pancreatic divisum and annular pancreas); anatomical conditions promoting intestinal stasis (e.g., small-bowel diverticula and blind loop); antibiotic, probiotic, or lactulose therapy within 1 month; chronic obstructive pulmonary disease or active smoking; and those with human immunodeficiency virus infection were excluded.

Study definitions

AP: AP was diagnosed based on the presence of two or more of the following: (1) Acute onset of persistent, severe, epigastric pain often radiating to the back; (2) Elevation of serum lipase or amylase to ≥ 3 times the upper limit of normal; and (3) Characteristic findings of AP on imaging (contrast-enhanced computed tomography, magnetic resonance imaging, or transabdominal ultrasonography)[3]. AP is divided into the following[3].

SIBO: SIBO can be broadly defined as a clinical syndrome of gastrointestinal symptoms caused by the excessive presence of bacteria in the small intestine[22]. According to the recent North American Consensus, a bacterial colony count of ≥ 10³ colony-forming units per milliliter in a duodenal/jejunal aspirate is considered diagnostic of SIBO. An indirect method for diagnosing SIBO is a rise in exhaled hydrogen or methane above baseline after oral ingestion of either glucose or lactulose[23].

Clinical and laboratory assessment

All patients underwent a complete hemogram, renal and liver function tests, prothrombin time-international normalized ratio, routine stool examination and microscopy, blood or other relevant body fluid cultures, routine workup for AP, chest and abdominal radiographs, ultrasonography of the whole abdomen and pelvis, and computed tomography of the abdomen to assess the severity of AP after 72 hours. All controls underwent clinical screening and a glucose hydrogen breath test (GHBT).

SIBO was assessed using a GHBT with an SC Microlyser (Quintron, Milwaukee, WI, United States). Participants were advised to avoid cigarette smoking or physical exercise for 2 hours before and during the test, as it may cause hyperventilation and consequent changes in breath hydrogen content. Furthermore, they were advised to brush their teeth and rinse their mouth with chlorhexidine mouthwash prior to the test. Following a 12-hour fast, a basal end-expiratory breath sample was collected before the test meal was administered. Each participant underwent a glucose breath test using 100 g of glucose (dissolved in 200 mL of water). Breath hydrogen and methane concentrations were measured at baseline (i.e., before glucose administration), and the average of four values was taken as the basal hydrogen/methane level. Participants were instructed to consume 100 g of glucose dissolved in 200 mL of water. Subsequently, breath hydrogen and methane levels were measured every 15 minutes for the next 3 hours[24]. A persistent rise in breath hydrogen or methane > 12 parts per million (ppm) above the basal level (at least two readings) was considered diagnostic of SIBO[25]. Subjects with a fasting methane level of > 10 ppm were labeled as predominant methane producers (PMP)[23,26-28]. Although various definitions are available for a PMP, > 10 ppm was used as the PMP in this study.

Sample size calculation

The sample size was calculated in Epi-Info using a 1:2 patient-to-control ratio, 80% power, and a two-sided 95% confidence interval. The prevalence rates of SIBO in patients with AP and HCs were 17.78% and 1.3%, respectively (19), resulting in sample sizes of 28 patients and 56 HCs. Consequently, 30 consecutive patients with AP and 60 HCs were included.

Statistical analyses

Continuous variables were reported as median and range, whereas categorical variables were expressed as proportions or percentages. The Mann-Whitney U test was used to compare continuous variables between the two groups, whereas Fisher’s exact test was used to compare categorical variables between the two groups. Univariate and multivariate analyses were performed to identify predictors of SIBO in patients with AP. SPSS version 25 was used for statistical analyses. A P-value of < 0.05 was considered statistically significant.

RESULTS

A total of 79 patients with AP were screened during the study period. Finally, 30 eligible participants were enrolled as cases and underwent the GHBT (Figure 2). In addition, 60 HCs were recruited, who also underwent the GHBT. The cases and controls were comparable in terms of age [33.5 (24-60) vs 37 (18-60); P = 0.687] and sex [female: 13 (43.3%) vs 22 (36.7%); P = 0.541]. All patients included in the study experienced abdominal pain (100%) (Table 1); vomiting (83.3%), fever (33.3%), and obstipation/constipation (70%) were also common. Other systemic features included jaundice (20.0%), breathlessness (16.7%), hiccups (6.7%), and oliguria (6.7%). In addition to symptoms characteristic of AP, those suggestive of SIBO were noted, including bloating (43.3%), abdominal distension (40.0%), belching (13.3%), and flatulence (10.0%).

Figure 2 Study flow chart. GHBT: Glucose hydrogen breath test; TB: Tuberculosis; IBD: Inflammatory bowel disease; CD: Crohn’s disease.

Table 1 Etiology, disease severity, clinical features, laboratory parameters, and computed tomography findings among patients with acute pancreatitis, n (%).

1a: Etiology of acute pancreatitis, n = 30
1	Ethanol	10 (33.3)
2	Biliary	15 (50.0)
3	Idiopathic	3 (10.0)
4	Ethanol plus biliary	2 (6.7)
1b: Severity of acute pancreatitis, n = 30
1	Mild	18 (60.0)
2	Moderate	8 (26.7)
3	Severe	4 (13.3)
1c: Local complications, n = 30
1	APFC	3 (10.0)
2	ANC	9 (30.0)
3	SVT	5 (16.7)
4	Ileus	11 (36.7)
1d: Systemic complications, n = 30
1	AKI	2 (6.7)
2	Sepsis	2 (6.7)
3	ALI	7 (23.3)
1e: Symptoms among cases, n = 30
1	Pain abdomen	30 (100)
2	Vomiting	25 (83.3)
3	Obstipation/constipation	21 (70)
4	Bloating	13 (43.3)
5	Abdominal distension	12 (40.0)
6	Fever	10 (33.3)
7	Jaundice	6 (20.0)
8	Breathlessness	5 (16.7)
1f: Baseline clinical and laboratory parameters among acute pancreatitis cases, median (range)
1	Height in cm	167 (152-175)
2	Weight in kg	64 (52-91)
3	BMI (kg/m2)	23.68 (18.59-36.26)
4	Hb (gm %)	12.6 (7.8-16.4)
5	TLC (cells/mm3)	12050 (5400-23500)
6	Neutrophil (%)	82 (57-94)
7	Lymphocytes (%)	12 (4-33)
8	NLR	6.83 (1.75-23.5)
9	Platelets (lac/mm3)	2.025 (0.53-6.29)
10	HsCRP (mg/dL)	34.25 (1.6-338)
11	Serum procalcitonin (ng/mL)	0.1475 (0.02-2.45)
12	Serum LDH (U/L)	270 (186-830)
13	Serum amylase (U/L)	1237 (233-7235)
14	Serum lipase (U/L)	987 (310-8076)
15	PT (seconds)	10 (1-17)
16	INR	1.03 (0.8-1.5)
17	Urea (mg/dL)	26.1 (9-66)
18	Creatinine (mg/dL)	0.8 (0.4-1.6)
19	Sodium (mEq/L)	139 (129-151)
20	Potassium (mEq/L)	4.1 (3.3-5.5)
21	Serum Calcium (mg/dL)	8.9 (7.7-11.6)
22	Serum Phosphorous (mg/dL)	2.55 (1.1-4.9)
23	Total Bilirubin (mg/dL)	1.3 (0.2-9.6)
24	SGOT (IU/L)	65 (15-1040)
25	SGPT (IU/L)	64.5 (10-790)
26	ALP (IU/L)	182.5 (31-512)
27	Total protein (g/dL)	6.95 (4.6-8.5)
28	Albumin (g/dL)	3.55 (2.3-5.4)
29	Globulin (g/dL)	3 (1.9-4.6)
30	TG (mg/dL)	141 (23-272)
31	HDL (mg/dL)	36 (17-73)
32	Cholesterol (mg/dL)	154.5 (77-289)
1 g: CT based parameters in AP
1	No necrosis	19 (63.3)
2	0-30% necrosis	3 (10.0)
3	> 30% necrosis	8 (26.6)
4	CTSI score	2 (1-10)

AKI: Acute kidney injury; ALI: Acute lung injury; ALP: Alkaline phosphatase; ANC: Acute necrotic collection; AP: Acute pancreatitis; APFC: Acute peripancreatic fluid collection; BMI: Body mass index; CT: Computed tomography; CTSI: Computed tomography severity index; Hb: Hemoglobin; HDL: High-density lipoprotein; HsCRP: High-sensitivity C-reactive protein; INR: International normalized ratio; LDH: Lactate dehydrogenase; NLR: Neutrophil-to-lymphocyte ratio; PT: Prothrombin time; SGOT: Serum Glutamic-oxaloacetic transaminase; SGPT: Serum Glutamic-pyruvic transaminase; SVT: Splanchnic venous thrombosis; TLC: Total leukocyte count; TG: Triglycerides.

Biliary pancreatitis was the most common cause of AP [15 (50.0%)], followed by alcohol-related pancreatitis [10 (33.3%)], idiopathic pancreatitis [3 (10.0%)], and mixed ethanol and biliary etiology [2 (6.7%)] (Table 1). Most patients exhibited mild AP (18; 60.0%), whereas eight (26.7%) and four (13.3%) had moderately severe and severe disease, respectively. The median duration of AP was 5 days (range, 1-23). Local and systemic complications were observed in 36.7% and 23.3% of patients, respectively, whereas 23.3% experienced both. Acute necrotic collection (ANC) was the most common local complication, noted in nine patients (30.0%), followed by splanchnic venous thrombosis in five patients (16.7%) and acute peripancreatic fluid collection in three patients (10.0%). In addition, ileus was evident in eleven (36.7%) patients. Acute lung injury was present in seven (23.3%) patients, whereas acute kidney injury and sepsis affected two (6.7%) patients each (Table 1).

The baseline inflammatory and biochemical parameters are presented in Table 1. Among patients with AP, contrast-enhanced computed tomography revealed an edematous pancreas without necrosis in nineteen patients (63.3%), whereas eight (26.6%) had necrosis exceeding 30%, with a median computed tomography severity index (CTSI) of 2 (1-10). SIBO was detected in twelve (40.0%) patients with AP compared with one (1.7%) HC (P < 0.001) (Table 1).

Breath test results

There was a significant difference in the prevalence of SIBO between cases and controls (40% vs 1.7%, P < 0.001) (Figure 3). SIBO was observed in all patients with severe AP [4 (100%)], followed by those with moderately severe AP [6 (75%)] and mild AP [2 (11.1%)], and the difference was statistically significant (P < 0.05) (Figure 4). Regarding other characteristics of the GHBT, there were no significant differences in basal or peak hydrogen levels between cases and controls, but there were significant differences in basal methane, peak methane, and the proportion of PMPs between the groups (P < 0.05; Table 2).

Figure 3 Frequency of small intestinal bacterial overgrowth in patients with acute pancreatitis vs those with healthy controls. ^aP < 0.001, categorical variables were compared with Fisher’s exact test (OR = 39.3; 95%CI: 4.8-324.6). SIBO: Small intestinal bacterial overgrowth; OR: Odds ratio; CI: Confidence interval.

Figure 4 Frequency of small intestinal bacterial overgrowth in patients with acute pancreatitis based on the severity of acute pancreatitis. The prevalence of SIBO increased with disease severity, being 11.1% (95%CI: 1.4%-34.7%) in mild acute pancreatitis, 75.0% (95%CI: 34.9%-96.8%) in moderate disease, and 100% (95%CI: 39.8%-100%) in severe disease (Fisher’s exact test, P < 0.001). SIBO: Small intestinal bacterial overgrowth; CI: Confidence interval.

Table 2 Characteristics of glucose hydrogen breath test among cases and controls.

No	Values	Cases (n = 30)	Control (n = 60)	P value
1	H2 - baseline	3 (0-6)	3 (0-10)	0.282
2	H2 - peak	5 (3-131)	5 (1-64)	0.294
3	Predominant methane producers	12 (40.0%)	8 (13.3%)	0.004
4	Methane - baseline	5 (3-22)	4 (0-12)	0.019
5	Methane - peak	6 (4-19)	5 (1-15)	0.001
6	Symptoms during GHBT	9 (30.0%)	0 (0%)	0.001

GHBT: Glucose hydrogen breath test.

Predictors of SIBO in AP

A univariate analysis of demographic, clinical, and laboratory parameters was conducted to identify possible predictors of SIBO in AP. Abdominal bloating, obstipation, ileus, severe pancreatitis, presence of SIRS, laboratory parameters such as neutrophil lymphocyte ratio (NLR), creatinine, high-sensitivity C-reactive protein (hsCRP), and imaging parameters such as ANC and CTSI scores differed significantly between the SIBO-positive and SIBO-negative patients (P < 0.05) (Table 3).

Table 3 Univariate analysis of predictors of small intestinal bacterial overgrowth among acute pancreatitis cases.

No	Parameters	SIBO present (n = 12)	SIBO absent (n = 18)	P value
Demographic parameters and symptoms
1	Age in years	32.5 (24-60)	35 (27-60)	1.000
2	Gender (M/F)	9/3	8/10	0.141
3	Abdominal bloating	12 (100.0%)	1 (5.6%)	< 0.001
4	Flatulence	3 (25.0%)	0 (0.0%)	0.054
5	Belching	3 (25.0%)	1 (5.6%)	0.274
6	Vomiting	10 (83.3%)	15 (83.3%)	1.000
7	Hiccups	1 (8.3%)	1 (5.6%)	0.765
8	Obstipation	8 (66.7%)	3 (16.7%)	0.009
9	Jaundice	4 (33.3%)	2 (11.1%)	0.184
10	Oliguria	2 (16.7%)	0 (0.0%)	0.152
Etiology, duration, severity, and complications of pancreatitis
1	Etiology of pancreatitis (ethanol/biliary/idiopathic/ethanol + biliary)	4/5/1/2	6/10/2/0	0.343
2	Duration of pancreatitis (days)	6.5 (3-23)	5 (1-15)	0.134
3	Severe pancreatitis	4 (33.3%)	0 (0.0%)	0.018
4	APFC	1 (8.3%)	2 (11.1%)	0.804
5	ANC	9 (75.0%)	0 (0.0%)	< 0.001
6	Ileus	9 (75.0%)	2 (11.1%)	0.001
7	AKI	2 (16.7%)	0 (0.0%)	0.152
8	Sepsis	2 (16.7%)	0 (0.0%)	0.152
9	Presence of SIRS	10 (83.3%)	3 (16.7%)	0.001
Laboratory parameters
1	NLR	11.71 (2.28-23.5)	4.95 (1.75-8.8)	0.001
2	Urea (mg/dL)	36.5 (12-50)	24 (9-66)	0.113
3	Creatinine (mg/dL)	0.9 (0.5-1.6)	0.65 (0.4-1.4)	0.012
4	Amylase (U/L)	1327.5 (371-4009)	1184.5 (233-7235)	0.819
5	Lipase (U/L)	947.5 (310-6800)	1162.5 (383-8076)	0.267
6	Total bilirubin (mg/dL)	2.2 (0.2-9.6)	0.95 (0.2-3.8)	0.325
7	Total protein (g/dL)	6.45 (4.6-8.5)	7 (5.2-8)	0.573
8	Albumin (g/dL)	3.6 (2.3-5.4)	3.5 (2.7-4.7)	0.917
9	TG (mg/dL)	147.5 (75-272)	135 (23-218)	0.491
10	Cholesterol (mg/dL)	165.5 (77-289)	151 (113-231)	0.884
11	hsCRP (mg/dL)	61 (2.8-338)	21.5 (1.6-204)	0.004
12	Procalcitonin (ng/mL)	0.332 (0.02-2.45)	0.136 (0.02-1.28)	0.305
13	CTSI score	8 (2-10)	2 (1-6)	< 0.001
14	Basal hydrogen (ppm)	2.5 (1-5)	3 (0-6)	0.285
15	Basal methane (ppm)	5 (3-11)	5 (3-22)	0.632
16	Predominant methane producer	4 (33.3%)	8 (44.4%)	0.709

AKI: Acute kidney injury; ANC: Acute necrotic collection; APFC: Acute peripancreatic fluid collection; CTSI: Computed tomography severity index; SIBO: Small intestinal bacterial overgrowth; SIRS: Systemic inflammatory response syndrome; hsCRP: High-sensitivity C-reactive protein; NLR: Neutrophil-to-lymphocyte ratio; TG: Triglyceride; M: Male; F: Female.

In addition, collinearity diagnostics were conducted to assess multicollinearity, which identified substantial collinearity among candidate predictors, particularly between the CTSI score and SIRS, as well as among inflammatory markers (NLR, ANC, and hsCRP). Multivariable logistic regression was not performed owing to the low event-to-variable ratio (12 events across 10 candidate variables) and the presence of significant multicollinearity. Accordingly, results were presented as univariate associations, supported by collinearity diagnostics and least absolute shrinkage and selection operator-based exploratory variable selection. These findings were interpreted as hypothesis-generating rather than confirmatory.

Least absolute shrinkage and selection operator logistic regression indicated that CTSI score, SIRS, abdominal bloating, ANC, NLR, ileus, obstipation, and creatinine were the potential predictors of SIBO, whereas severe pancreatitis and hsCRP were shrunk to zero. SIRS and ANC were excluded based on collinearity diagnostics for variable clusters. In addition, a Bonferroni correction was applied for multiple testing across the six selected variables, and the corrected univariate analyses are presented in Table 4.

Table 4 Univariate association of variables with small intestinal bacterial overgrowth after least absolute shrinkage and selection operator logistic regression-based selection and collinearity exclusion.

No	Variable	GHBT positive (n = 12)	GHBT negative (n = 18)	P value
1	Abdominal bloating	12 (100.0%)	1 (5.6%)	< 0.001a
2	Ileus	9 (75.0%)	2 (11.1%)	0.001a
3	NLR	11.71 (2.28-23.5)	4.95 (1.75-8.8)	0.001a
4	CTSI score	8 (2-10)	2 (1-6)	< 0.001a
5	Obstipation	8 (66.7%)	3 (16.7%)	0.009
6	Creatinine (mg/dL)	0.9 (0.5-1.6)	0.65 (0.4-1.4)	0.012

^aP < 0.001, statistically significant after Bonferroni correction (α = 0.05/6 = 0.0083).

GHBT: Glucose hydrogen breath test; CTSI: Computed tomography severity index; NLR: Neutrophil-to-lymphocyte ratio.

DISCUSSION

In this case-control study, a significantly higher prevalence of SIBO was observed in patients with AP than in HCs (40% vs 1.7%). To the best of our knowledge, this study is the first from India to assess the SIBO burden in patients with AP. Zhang et al[19] from China reported the prevalence of SIBO to be 17.78% in patients with AP. However, a comparison with HCs was not performed, and the lactulose hydrogen breath test was used to identify SIBO. In contrast, this study utilized the GHBT, which is preferred owing to its higher sensitivity and specificity, which may explain the higher SIBO positivity observed. Historically, wide variations of 1%-40% have been observed in the prevalence of SIBO among HCs[24,29-32]. In this study, SIBO was detected in only 1.7% of the controls, which is substantially lower and may reflect differences in inclusion or exclusion criteria, diagnostic methodology, population characteristics, or breath-test criteria.

Another study by Kim et al[33] identified that total breath H₂ or CH₄ levels in patients with AP were significantly higher than those in controls. Nonetheless, this investigation did not observe a significant difference in GHBT positivity between patients with AP and controls (12.0% vs 13.3%, P = 0.85). Kim et al[33] used a lower glucose dose of 75 g for the GHBT than the standard 100 g used in this study. Variations in testing methodologies, as well as breath test substrates, dosages, and diagnostic criteria, may account for the perceived differences in SIBO prevalence across studies. Furthermore, in this study, the prevalence of methanogenic SIBO, also called intestinal methanogen overgrowth, was higher in the AP group than in the HC group (40% vs 13.3%, P = 0.004). Methane-producing bacteria, such as Methanobrevibacter smithii, are often more resistant to antibiotics, emphasizing the need for their identification[34].

The median duration of AP was 5 (1-23) days and did not differ between SIBO-positive and SIBO-negative patients. Zhang et al[19] included patients within 72 hours and found that hydrogen production was higher in this early period than in the late stages (7 days). Gut microbiota alterations have been reported to occur within 72 hours of AP diagnosis[35]. Studies have shown reduced bacterial diversity and the increased presence of potentially harmful bacteria, such as Bacteroidetes and Proteobacteria, compared with healthy individuals[21,35-37].

The pancreas influences the microbiota directly by secreting antimicrobial peptides and indirectly by playing a role in digestion, thereby shaping the composition of the gut microbiome. In patients with AP, alterations in the gut-pancreas axis lead to bowel stasis and an increase in colonic-type bacteria in the small intestine, resulting in SIBO[38,39]. In their research, Wang et al[4] observed that the colonic transit time (CTT) was increased in patients with AP compared with HCs. In addition, this study found that CTT was significantly higher in patients with severe AP than in those with moderately severe AP. CTT was not measured in the present study; however, the observed association of SIBO with ileus, obstipation, and bloating supports the role of impaired gastrointestinal motility in its development during AP. In AP, ileus is a common complication resulting from splanchnic hypoperfusion, inflammatory mediators, electrolyte imbalances, and opioid use[40].

In our study, the SIBO burden exhibited a strong graded association with disease severity, being present in all patients with severe AP and in three-quarters of those with moderately severe AP. This finding is consistent with the observations of Zhang et al[19], who reported that SIBO was present in 25.82% of patients with severe AP and 8% of those with mild AP (P < 0.05). The increased incidence of SIBO in severe AP can be attributed to various pathophysiological factors, such as pancreatic exocrine insufficiency, reduced intestinal motility, altered intestinal barrier integrity, and disrupted microbial homeostasis, which result in intestinal dysbiosis and bacterial overgrowth. The findings from this study highlight the bidirectional relationship between intestinal dysbiosis and AP severity. Severe disease is linked to significant alterations in gut motility, permeability, and immune balance, which promote SIBO. These changes can intensify systemic inflammation, contribute to pancreatic necrosis, and worsen clinical outcomes, emphasizing the early detection of gut dysfunction in AP.

Data on SIBO in AP are limited, but several studies have documented a high prevalence of SIBO in CP. According to Sanjeevi et al[41], SIBO was present in one-third of patients with CP based on jejunal aspirate cultures, and hypoalbuminemia is the only significant predictor. Capurso et al[16] performed a meta-analysis, which confirmed a high pooled prevalence of SIBO in CP (36%). Furthermore, Chonchubhair et al[42] observed that SIBO was more common in patients with complications related to CP, such as diabetes mellitus, pancreatic exocrine insufficiency, and proton pump inhibitor use, especially in alcohol-induced CP.

The only study evaluating the prevalence of SIBO in pancreatic malignancy was performed by Ma et al[18], who reported it to be 63.3%. Alcohol use is a major cause of AP and is independently linked to a heightened risk of SIBO. Studies by Mutlu et al[43] and Schnabl and Brenner[44] established that alcohol causes intestinal dysbiosis and bacterial overgrowth, promoting the translocation of bacterial endotoxins. Clinically, Gabbard et al[45] noted a high prevalence of SIBO (58%) in patients with alcoholism without cirrhosis. Moreover, SIBO often occurs in patients with liver cirrhosis, particularly in those with alcoholic cirrhosis. Vaithiyam et al[24] conducted a study on SIBO in severe alcoholic hepatitis, which revealed that the condition was more common than in controls (25% vs 3.33%; P = 0.017).

This study is the first to identify the predictors of SIBO in AP, as the other two available studies failed to assess it[19,33]. Univariate analysis indicated that clinical features such as abdominal bloating, obstipation, ileus, and severe AP; markers of systemic inflammation (SIRS, elevated NLR, and hsCRP); renal dysfunction (elevated creatinine); and radiological severity indexes (ANC and higher CTSI scores) were significantly linked to SIBO. These findings suggest that SIBO in AP is closely linked to systemic inflammatory burden and local pancreatic complications. Univariate associations highlight the key pathophysiological links among intestinal dysmotility, inflammation, and bacterial overgrowth in AP.

This study has several strengths. This investigation is the first to identify the predictors of SIBO in patients with AP. This work employed standardized Atlanta diagnostic criteria, comprehensive clinical and radiological evaluations, and measurements of both hydrogen and methane production, which enabled the identification of methanogenic SIBO. Nonetheless, this study also has certain limitations. The relatively modest sample size and high exclusion rate might have introduced selection bias. However, post-hoc power analysis demonstrated a study power of 99.4% for the observed difference between cases and controls. Moreover, only patients aged 18-60 years were included in this study. The exclusion of adolescents and elderly patients might limit the generalizability of the findings to these age groups. Future studies involving age-stratified analyses are warranted. Moreover, although the GHBT is noninvasive and widely used, it may underestimate SIBO and does not offer direct microbiological characterization. In addition, this study did not assess the long-term outcomes of SIBO or methanogenic SIBO in patients with AP or the temporal effects of SIBO treatment. Kalia et al[46] examined the effectiveness of oral rifaximin in preventing infected pancreatic necrosis in patients with predicted severe AP. The administration of this drug reduced the duration of hospital stay but did not affect mortality or the development of infected pancreatic necrosis. This study did not evaluate SIBO; targeted treatment after its identification could have altered the findings. Several potential confounding factors, such as paralytic ileus and analgesic use, were not excluded in this study. Opioid and nonopioid analgesic use, which is known to impair gastrointestinal motility, might have influenced the development of SIBO. Furthermore, alcohol-related pancreatitis accounted for 33.3% of cases, and alcohol itself has been implicated in the pathogenesis of SIBO via multiple mechanisms, including intestinal mucosal damage, altered gut motility, and increased intestinal permeability. Paralytic ileus, noted in 36.7% of cases in our study and identified as a predictor of SIBO in univariate analysis, could have independently exacerbated the risk of SIBO owing to its association with intestinal stasis.

Despite these limitations, this study provides valuable insights into the burden and clinical significance of SIBO in patients with AP. The high prevalence of SIBO, particularly in severe cases, indicates that gut-focused approaches, such as early enteral nutrition, reduced opioid use, prokinetic agents, and selective application of nonabsorbable antibiotics or microbiota-modulating therapies, must be explored further. Larger multicenter prospective studies that include microbiome analysis and outcome-based interventions are essential to determine whether targeted treatment of SIBO can positively affect the clinical course of AP.

CONCLUSION

SIBO is significantly more prevalent in patients with AP than in HCs and is strongly linked to disease severity and systemic inflammation. These findings support the bidirectional gut–pancreas axis, in which intestinal dysbiosis both reflects and aggravates disease severity. Larger prospective studies must be conducted in the future to determine the causal role of SIBO in AP complications and to assess whether targeted gut-focused therapies can improve clinical outcomes.