Different Classes of Antibiotics, by Provoking Distinct Patterns of Dysbiosis, May Affect the Occurrence of Inflammatory Bowel Disease

1. Introduction

Inflammatory bowel disease (IBD) refers to diseases that cause chronic inflammation in the gastrointestinal (GI) tract. Symptoms include abdominal pain, diarrhea (often with blood), fatigue, and unintended weight loss, though they can vary depending on the disease's severity and location. Symptoms include abdominal pain, diarrhea, and weight loss, which can come and go in periods of flare-up and remission. Crohn's disease can develop anywhere along the digestive tract from the mouth to the anus, whereas ulcerative colitis is restricted to the large intestine. These diseases are non-communicable and incurable, affecting millions of people worldwide. Usually, IBD is considered a Western disease. By the 21st century, emerging industrialized nations in Asia and Latin America had experienced a rapid increase in cases, making inflammatory bowel disease a global issue (1, 2). Patients with IBD might produce a variety of symptoms that are different in UC and CD; however, frequent symptoms include persistent abdominal pain, diarrhea (with or without blood), weakness, and weight loss. IBD’s clinical manifestations are characterized by recurrent episodes of gastrointestinal tract inflammation caused by an abnormal immune response to gut microflora. Ulcerative colitis (UC) generally presents itself as bloody diarrhea with or without mucus. Patients frequently complain of tenesmus, a sensation of incomplete evacuation, and abdominal pain. On physical exam, the patients complain of feeling predominantly left lower or left upper quadrant abdominal pain. In more severe cases, signs of an acute abdomen may develop, including guarding, rebound tenderness, or percussion tenderness, warranting investigation for toxic megacolon. The clinical manifestations of Crohn's disease (CD) vary considerably depending on the anatomical location of gastrointestinal involvement. Manifestations differ depending on the underlying inflammation, fistula formation, or the development of strictures. The symptom complex of right lower quadrant pain, weight loss, and non-bloody diarrhea is suggestive of Crohn's disease flare-up. Fistula formation can appear as fecaluria, pneumaturia, and rectovaginal fistula development. Masses in the right lower quadrant are generally related to abscess formation. Children with CD may present with growth retardation and delayed sexual maturation (3-5). The incidence of IBD has risen in the past decades, and now it is a global issue and not only a “Western” disease. Approximately 1.3 million people in Europe suffer from inflammatory bowel disease (IBD), which is about 0.2% of the population, but its rate varies by geographical regions. IBD incidence ranges from 10.5 to 46.14 per 100,000 in Europe. Crohn’s disease [CD] incidence ranged from 4.1 to 22.78 per 100,000 in Europe, and ulcerative colitis [UC] estimates ranged from 3.0 to 23.36 per 100,000 in Europe. Northern European countries and the UK show the highest rates of UC, while southern and eastern European countries historically have lower rates, though these are increasing. The UK has the highest prevalence rate with 570/100000 inhabitants, followed by Denmark (530/100000) and Norway (503/100000). The lowest prevalence rates were reported from Portugal (1,6/100000) and Romania (2,42/100000). The prevalence of Crohn's disease in Europe ranges from approximately 61.6 to 178 per 100,000 people (6, 7). Genetic analyses have identified more than 200 genes associated with an increased risk of developing IBD, but most of the associations are relatively weak. Twins studies indicate a concordance rate of around 4% for dizygotic pairs. In contrast, concordance rates for monozygotic twins have been reported as 50% with higher rates among pairs with Crohn's disease than pairs with ulcerative colitis. Among pairs with Crohn's disease, the majority have disease in the exact location of the bowel (8). The single greatest known risk factor for developing IBD is having a close family member with IBD, which might raise the possibility, apart from genetics, that the “transfer” of the IBD-related microbiome also occurs (9). It is observed that about 5% and 23% of people with IBD have a first-degree relative with IBD. Among families with multiple affected family members, there appears to be a high degree of clinical similarity in disease location, disease behavior, age at diagnosis, and the presence and character of extra-intestinal manifestations (10). Reports on the environmental risk factors for IBD indicated that maternal smoking, early life antibiotic exposure, otitis media, living in urban area (for CD), adult smoking (for CD), longer NSTAID drugs taking, combined oral contraceptives, hormonal medical treatment, adult antibiotic exposure, carnivorous diet, and consuming ultra-processed foods (UBD) augment the possibility of developing CD, but not UC. Breastfeeding up to 12 months, higher consumption of vegetables and fish, exposure to farm animals, bed sharing, fruit and vegetables (UC), fiber intake (CD), appendectomy (UC), tobacco smoking (UC), presence of Helicobacter pylori, and Mediterranean diet are considered protective factors. The opposite effect of smoking for CD (aggravating the symptoms) and UC (ameliorating the disease process) is well known, but no appropriate explanation exists. Reports also indicate the role of psychological factors in the development of IBD (6, 11). Inflammatory bowel diseases arise from a convergence of genetic risk, environmental factors, and gut microbiota, where each is necessary but not sufficient alone to cause disease. However, environmental factors (diet, antibiotics, etc.) can influence the composition of gut flora by promoting or inhibiting the proliferation of different microbial taxa. The pro-colitogenic effects of intestinal dysbiosis and the amelioration of the inflammatory burden through fecal microbiota transplantation are well established.

2. Concept/working hypothesis:

Considerable scientific knowledge accumulated over the past decades suggests that the principal driving force in the development of inflammatory bowel diseases (UC, CD) is pro-inflammatory dysbiosis, which arises from various external factors in (probably) genetically susceptible individuals. Different diets either promote (ultra-processed food) or protect (Mediterranean diet) the development of IBD. Antibiotic exposure, particularly at early ages, is considered a risk factor for the development of IBD. Still, different antibiotic classes, depending on the susceptibility of the intestinal microbial taxa, may either promote or inhibit the growth of IBD-related dysbiosis. Antibiotic consumption reports (ECDC) from the 30 European countries included in the study indicate manifold differences in the average yearly consumption of different antibiotic classes, expressed in Defined Daily Dose/1000 inhabitants/Day (DID). The consumption of narrow-spectrum, beta-lactamase-sensitive penicillin (J01CE), calculated as an average yearly consumption, is highest in Denmark (27.733 DID) and lowest in Italy (0.009 DID). The highest average annual cephalosporin consumption was reported in Greece (23.814 DID), and the lowest in the UK (2.067). It is hypothesized that if the consumption of any antibiotic group shows a significant positive statistical association with the incidence of IBD, it might indicate that higher use of that class is linked to an increased likelihood of belonging to a higher IBD-incidence category. In contrast, a significant negative association may suggest that greater consumption of a given antibiotic class is associated with a higher probability of belonging to a lower IBD-incidence category, potentially reflecting differing impacts on gut flora. Countries with higher consumption of antibiotic classes that show positive associations might report higher IBD incidence. In contrast, increased use of classes that show negative associations may correspond to lower IBD incidence.

3. Objective

Identification of antibiotic classes that show positive or negative statistical association with the IBD incidence (ASIR 2021), which might promote or inhibit the IBD-related dysbiosis.

4. Materials and methods

An average yearly antibiotic consumption database was developed on antibiotic consumption in 30 European countries from the ECDC annual reports of 2004-2023 (twenty years) accessed in August 2025 (https://www.ecdc.europa.eu/en/publications-data/antimicrobial-consumption-eueea-esac-net-annual-epidemiological-report-2023 ). Total antibiotic consumption (public, hospital) was recorded for the countries included in the study for each year (2004-2023, twenty years) to calculate the yearly average consumption. The ECDC reports included 15 antibiotic categories, expressed as Defined Daily Dose/1000 Inhabitants/Day (DID), coded according to the ATC classification at level II. and III. https://qap.ecdc.europa.eu/public/extensions/AMC2_Dashboard/AMC2_Dashboard.html#eu-consumption-tab

Antibiotic categories (codes): Total systemic antibiotics: J01, tetracyclines: J01A, amphenicols: J01B, penicillin group (all): J01C, broad spectrum, beta-lactamase susceptible penicillin: J01CA, narrow spectrum, beta-lactamase susceptible penicillin: J01CE, narrow spectrum, beta-lactamase resistant, narrow spectrum penicillin: J0CF, broad spectrum, beta-lactamase resistant combination penicillin: J01CR, cephalosporin: J01D, sulfonamides: J01E, macrolides and lincomycin: J01F, aminoglycosides: J01G, quinolones: J01M, antibiotic combinations: J01R, other antibiotics: J01X, which includes glycopeptides, polymyxines, steroid antibacterials, imidazole derivates, nitrofurantoin derivates, fusidic acid etc. The average yearly total antibiotic consumption (J01, in DID) was calculated as 587. 226 DID for the 30 countries. The highest consumption was reported from the UK (41.961 DID), and the lowest from the Netherlands (9.917). Several differences in antibiotic consumption were observed across European countries. The Nordic countries preferred different narrow-spectrum penicillins, whereas in the Mediterranean countries, broad-spectrum antibiotics were predominantly used (Table 1). To develop the antibiotic database for comparison, we calculated the relative share of antibiotic consumption across different classes from the total amount (J01, DID) and expressed it as a percentage (%). Antibiotic groups of very low consumption (below 0.5 DID/year: amphenicols /J01B/ and combinations /J01R/) were left out of the calculation. The remaining amount covered the 99,87% of the average total yearly consumption. The database on the incidence rate of IBD in 2021 was extracted from the Global Burden of Disease (GBD) study 2021, and the age-standardized incidence rate/100,000 inhabitants (ASIR) for 2021, as reported by Chen L et al. The database included data on young people aged 15-39 years, for whom IBD occurs more frequently (12, 13). The results were featured in Table 1. The highest incidence of IBD (age-stratified incidence rate: ASIR 2021) among young people was reported in the Netherlands (28.5/100000 inhabitants), and the lowest in Romania (2.5/100000 inhabitants). The results were categorized into three ordinal groups (low, medium, and high incidence) based on tertiles of the overall distribution: Group 1 = IBD incidence < 9.867, Group 2 = IBD incidence >= 9.867 and < 17.067, and Group 3 = IBD incidence >= 17.067. Statistics: Pearson correlation coefficients were first calculated to assess the bivariate relationships between each antibiotic class and IBD incidence. Ordinal logistic regression was performed to model the association between antibiotic class proportions and IBD incidence category. Each antibiotic class was entered separately into the model as a continuous independent variable. This approach was chosen to avoid multicollinearity among antibiotic courses, as these variables represent mutually dependent components of total antibiotic consumption. A positive association indicated higher odds of belonging to a higher IBD-incidence category with increasing antibiotic consumption, whereas a negative association indicated higher odds of belonging to a lower IBD-incidence category. The regression models estimated odds ratios (OR) with corresponding 95% confidence intervals (95% CI) and p-values. Non-parametric tests were used due to violations of normality and homogeneity of variance, as assessed by the Shapiro-Wilk and Levene’s tests. Therefore, differences in antibiotic consumption patterns across IBD incidence categories were examined using the Kruskal-Wallis test, followed by Dunn’s test with the Dwass-Steel-Critchlow-Fligner (DSCF) correction for pairwise post-hoc comparisons. Model performance was evaluated using likelihood-ratio tests, and the proportional odds assumption was checked where applicable. Statistical significance was set at p < 0.05. Pearson correlation coefficients were reported alongside regression outputs and the results of the non-parametric tests for comparative interpretation. All analyses were performed using Jamovi (version 2.3.28.0) statistical software.

5. Results

In the 30 European countries analyzed, average annual antibiotic consumption varied considerably by class. Pearson correlation analyses identified positive associations between the proportional use of tetracyclines (J01A, r = 0.519, p = 0.003), narrow spectrum, beta-lactamase–sensitive penicillin (J01CE, r = 0.465, p = 0.010), narrow spectrum, beta-lactamase–resistant penicillin (J01CF, r = 0.503, p = 0.005), and sulfonamides (J01E, r = 0.424, p = 0.020) with IBD incidence. In contrast, cephalosporins (J01D, r = –0.478, p = 0.008), macrolides (J01F, r = –0.337, p = 0.068), aminoglycosides (J01G, r = –0.524, p = 0.003), and quinolones (J01M, r = –0.452, p = 0.012) demonstrated inverse correlations. Other antibiotic classes did not show significant associations (Table 1).

The results of univariate ordinal logistic regression analyses paralleled these patterns. Higher consumptions of tetracycline: J01A (OR = 1.215, 95% CI [1.074-1.412]), narrow-spectrum, beta-lactamase-sensitive penicillin: J01CE (OR = 1.223, 95% CI [1.065-1.513]), narrow-spectrum, beta-lactamase resistant penicillin: J01CF (OR = 1.715, 95% CI [1.234-2.779]) and sulfonamides: J01E (OR = 1.555, 95% CI [1.062-2.399]) were associated with increased odds of belonging to higher IBD-incidence categories, respectively. On the other hand, higher consumptions of cephalosporin: J01D (OR = 0.843, 95% CI [0.742-0.943]), macrolides: J01F (OR = 0.795, 95% CI [0.652-0.936]), aminoglycosides: J01G (OR = 0.041, 95% CI [0.002-0.314]), and quinolones: J01M (OR = 0.772, 95% CI [0.609-0.944]) were associated with lower odds of higher IBD incidence. (Table 1.) These trends were visualized in a forest plot summarizing odds ratios and 95% confidence intervals across all antibiotic classes (Figure 1) and in diagrams (Figure 2 and Figure 3).

Non-parametric Kruskal–Wallis tests confirmed significant differences in antibiotic consumption across IBD incidence categories for multiple classes, including J01A, J01C, J01CE, J01CF, J01CR, J01D, J01E, and J01 G. Post-hoc pairwise comparisons using the DSCF method revealed that the most significant differences were between low- vs high-incidence groups (G1 vs G3) and medium- vs high-incidence groups (G2 vs G3) (Table 1). For example, J01A consumption differed significantly between G1 and G3 (p = 0.018) and G2 vs G3 (p = 0.006), while J01CF showed differences between G1 vs G2 (p = 0.034) and G2 vs G3 (p = 0.003).

Overall, results demonstrate a consistent pattern: certain antibiotic classes, particularly those targeting Gram-positive bacteria, were positively associated with higher IBD incidence, whereas classes with broader Gram-negative activity were inversely associated, with some overlap.

6. Discussion

The possible role of the gut microbiome in the development of inflammatory bowel diseases (IBD) was first raised over 20 years ago through the process of dysbiosis, an imbalance of gut bacteria that leads to reduced diversity and an increase in pro-inflammatory microbes (14). Early efforts to correct microbiome composition through fecal microbiota transplantation and achieve clinical remission had also been reported (15). The role of the gut microbiota in the establishment and maintenance of health, and in the pathogenesis of different diseases, including IBD, has also been identified over the past two decades. The gut microbiota interacts with the host through metabolites, small molecules produced as intermediates or end products of microbial metabolism. These metabolites can arise from bacterial metabolism of dietary substrates, modification of host molecules, such as bile acids, or directly from bacteria (16). In healthy adults, Firmicutes and Bacteroidetes are the predominant phyla in the intestines, with Proteobacteria and Actinobacteria accounting for the majority of the remaining bacteria (17). The possible roles of viruses and plant metabolites have also been considered (18). The most crucial change observed in IBD-related dysbiosis is the significant reduction of the diversity of microbiome, with the decrease in taxa with anti-inflammatory or protective properties like Lachnospiraceae, Bifidobacterium, Faecalibacterium prausnitzii, Roseburia species, Bacteroides, and the proliferation of the pro-inflammatory Proteobacteria, Fusobacterium species, and Ruminococcus gnavus species has also been observed. Biopsy specimens from the ileum and the rectum showed similar dysbiosis, but the biopsy and fecal isolates were not identical (19). Firmicutes and Bacteroides, together, constitute approximately 90% of gut microbiota and are key producers of short-chain fatty acids (SCFAs). Bacterial ‘dysbiosis’ in IBD has been typically characterized by an increased ratio of potentially pathogenic to beneficial bacteria and lower overall diversity. Recently, the focus of microbiome research has shifted from microbial composition to the functional properties of specific bacterial species and strains. Comprehensive multi-omics investigations have elucidated the composition of bacterial species and their metabolic profiles in patients with IBD, enabling more accurate detection of alterations in bacterial ecology that coincide with disease initiation and progression. An extensive, detailed review summarizes the roles of different microbial metabolites and their interactions with the immune system (20). A multi-omics study of UC patients indicated that B. vulgatus proteases are associated with UC disease activity, and this observation has also been confirmed in animal experiments (21). The mechanisms by which the microbiota damages the intestinal epithelial barrier are complex and remain a subject of study (22). Several risk factors are considered for the development of IBD. However, the decisive role in the development of IBD is the modification of the microbiome, because external factors can induce dysbiosis, which may act as a "risk factor" for IBD or have a "protective" effect against its development (23). The crucial role of the microbiome in the development of IBD or in the amelioration of symptoms is further supported by fecal microbiota transfer (FMT) experiments (24). Fecal microbiota transplantation from healthy donors induced clinical remission and endoscopic improvement in active ulcerative colitis (UC) and is associated with distinct microbial changes that relate to outcome (25). The clinical symptoms of Crohn's disease had also been ameliorated after using FMT (26). By restoring microbial diversity and correcting dysbiosis, FMT offers a novel, microbiota-targeted alternative to conventional therapies. While data support its efficacy in improving disease remission, variability in outcomes underscores the need for standardized protocols and additional large-scale, controlled studies (27). The role of antibiotics in the development of IBD and their use as therapeutic agents are controversial. A combination of amoxicillin, metronidazole, and tetracycline was reported to be beneficial in a limited number of steroid-resistant UC patients without a control group (28, 29). An extensive study from Denmark reported that antibiotic use is a risk factor, with practically all antibiotic classes identified as risk factors, except nitrofurantoin. A meta-analysis of publications reporting on the association of antibiotic use and the development of IBD confirmed that antibiotic exposure could be considered a risk factor for IBD. This study was based on prescription analyses and questionnaires. They concluded that almost all antibiotic classes were associated with increased IBD risk, and the extent of the risk varied across classes (30). Recent reports indicate a significant shift in healthy gut flora even before IBD diagnosis (31). These changes include a general decrease in microbial diversity, a reduction in beneficial bacteria like Faecalibacterium and Ruminococcaceae, and an increase in potentially harmful bacteria such as Enterobacteriaceae and Fusobacterium (32).