**Background:** Rumen cannulation is widely used in animal nutritional research to study digestive physiology, and gastric fistulation is also practiced in human medicine (e.g., percutaneous endoscopic gastrostomy). While previous work indicates that cannulation does not seriously affect feed intake or growth, there is evidence in humans that PEG can alter gastrointestinal immune responses. Separately, the plane of milk nutrition in pre-weaning calves is known to influence intestinal development and immune competence. This study hypothesized that rumen cannulation performed in early life would modulate the intestinal and systemic immune system of cattle in later life, and that a restrictive milk replacer (MR) feeding regime would exacerbate these effects. The study was conducted in 24 female Holstein calves at the Research Institute for Farm Animal Biology (FBN), Germany, with ethical approval (registration number 7221.3-1.1-009/16).
**Methods:** Calves were assigned to two feeding groups: 10MR (10% of body weight MR per day) and 20MR (20% of body weight MR per day), with n=12 per group. At week 3 of life (18.4 ± 0.8 days), seven animals per feeding group received a permanent rumen cannula (RC) under local anesthesia, analgesics, sedatives, and antibiotics, with analgesics and antibiotics continued for 5 ± 1 days post-surgery. The remaining five animals per group served as non-cannulated controls (NRC). All calves were gradually weaned between weeks 11-12 and received a total mixed ration thereafter. Animals were slaughtered at week 34 of life. Tissue samples were collected from jejunum (epithelium, lamina propria, Peyer's patches), ileum (epithelium, lamina propria), mesenteric lymph nodes (MSL), spleen, thymus, and blood. Leucocytes were isolated and T cell subsets (CD2+, CD4+, CD8+) and B cell subsets (CD21+) were quantified by flow cytometry. Gene expression in jejunal mucosa and spleen was measured by RT-qPCR for barrier function, cytokine, and toll-like receptor (TLR) genes. Intestinal content was analyzed for pH, dry matter, lactate, and volatile fatty acids. Plasma and tissue IgA/sIgA were measured by ELISA. Statistical analysis used a mixed model with fixed factors of MR feeding intensity (MRF), rumen cannula (RC), and their interaction, plus dehorning as a covariate. Differences were significant at P < 0.05 and trends at P < 0.1.
**Key Results:** At slaughter, 20MR heifers were heavier (312 ± 7 kg) than 10MR heifers (280 ± 7 kg, P < 0.05), with no body weight difference between RC and NRC animals. In MSL, 20MR NRC heifers showed significantly higher CD8+ T cell subsets compared to all other groups (P < 0.05). In jejunal intraepithelial lymphocytes (IELs), pooled CD4+ T cell subsets were higher in NRC vs RC heifers (P = 0.012), and this was most pronounced between 10MR NRC and 10MR RC animals. In ileal IELs, pooled CD4+ T cells were lower in NRC vs RC heifers (P < 0.05), while CD21+ B cells were higher in NRC vs RC heifers (P = 0.029). In the spleen, CD8+ T cell subsets tended to be lower in 20MR NRC heifers compared to all other groups (P < 0.1), and CD21+ B cells were significantly higher in 20MR NRC vs RC heifers (P = 0.033). Splenic TLR6 gene expression was higher in RC vs NRC heifers (P = 0.036), and IL4 expression tended to be higher in RC heifers (P < 0.1). Splenic TLR2 expression was significantly lower in 10MR vs 20MR heifers (P = 0.022), with TLR3 and TLR10 showing similar trends (P < 0.1). In jejunal mucosa, prostaglandin endoperoxide synthase 2 (PTGS2/COX2) expression was higher in RC vs NRC heifers (P = 0.044), and MUC2 expression tended to be lower in 10MR vs 20MR heifers (P = 0.098). A significant positive correlation was found between jejunal PTGS2 and IL17A expression (r = 0.719, P < 0.001). Intestinal content pH was higher in NRC vs RC heifers (P = 0.044), and dry matter tended to be lower in NRC animals (P = 0.091). No significant effects of cannulation or MRF were observed on IgA concentrations in plasma, spleen, or jejunal content, nor on T and B cell subsets in blood.
**Clinical Implications (Veterinary/Research):** This study demonstrates that rumen cannulation performed in early life has lasting modulatory effects on intestinal and systemic T and B cell subsets in cattle persisting until at least 34 weeks of age, several months after surgery. The findings indicate that surgical fistulation should be considered a potential confounding factor in nutritional and immunological research using cannulated animals. The restrictive (10%) milk feeding regime produced immune alterations in the spleen, MSL, and thymus that paralleled those induced by cannulation, suggesting that early nutritional stress can durably shape immune development. The increase in jejunal PTGS2 (COX2) expression in cannulated animals may indicate low-grade inflammatory or wound-healing processes. The absence of changes in blood lymphocyte subsets despite changes in lymphoid organs highlights the importance of tissue-level immune analysis. These results should inform the interpretation of data from fistulated research animals and may have implications for managing the health and immune development of surgically or nutritionally stressed neonatal calves.