The present study examined 540 bowel resections and found that 54.6% of patients undergoing surgery for penetrating CD were likely to have a stoma formed. By utilizing the LASSO regression model, we were able to identify several significant risk factors associated with stoma formation. These factors include disease location, fistula resulted from previous anastomosis, absence of preoperative exclusive enteral nutrition support, albumin levels below 35 g/L, C-reactive protein levels exceeding 10 mg/L, hemoglobin levels below 100 g/L, perianal disease, internal fistula, and surgical approach. Based on these findings, we developed a predictive nomogram model that takes into account these nine factors to quantify the risk of stoma formation. The nomogram demonstrated a favorable overall performance in predicting stoma formation, as evidenced by its discriminatory ability, calibration, and net benefit.
In our nomogram model, we identified that the absence of preoperative exclusive enteral nutrition support as the strongest predictor for stoma formation, with an odds ratio (OR) of 4.343. Previous studies conducted by our team consistently demonstrate that preoperative total enteral nutrition is an independent protective factor against the development of postoperative intra-abdominal septic complications (IASCs)[11, 12]. These findings are in line with the guidelines set forth by the European Society for Clinical Nutrition and Metabolism (ESPEN), which recommend EEN as the preferred mode of nutritional support for patients. The aim is to facilitate intestinal mucosal healing, address nutritional imbalances, and reduce disease activity[13]. Consequently, it is reasonable to expect a higher likelihood of requiring a stoma when patients are unable to tolerate optimal preoperative total enteral nutrition.
Our study findings indicate that patients with fistula resulted from previous anastomosis in penetrating CD exhibit a preference for one-stage anastomosis during surgery. These fistulas commonly arise at the site of a previous surgical connection between two segments of the intestine, predominantly occurring in the small bowel or colon[14]. Surgeons who encounter anastomotic fistulas derive advantages from the presence of a pre-existing anastomosis, as it facilitates the planning and execution of the surgical repair with greater efficacy[15]. This advantage ultimately contributes to enhanced management of anastomotic fistulas in these patients.
Stoma formation is frequently chosen as the surgical approach for treating internal fistulas in CD. These fistulas can be highly intricate, involving multiple organs or structures, which presents a difficulty in achieving complete removal or repair without causing harm to the surrounding tissue[2]. The creation of a stoma provides a means to bypass the fistula, facilitating the healing process in the affected area[16, 17]. Moreover, the creation of a stoma can serve as a prudent strategy for managing the symptoms and complications associated with the disease[18]. It can effectively alleviate pain, mitigate inflammation, and prevent the recurrence of infections.
When patients with CD present with elevated levels of C-reactive protein, it indicates a heightened degree of disease activity and aggressiveness. Furthermore, there is typically more pronounced inflammation in the affected area, which can result in the thickening of intestinal walls, the development of strictures, and the emergence of fistulas. These complications can present difficulties in surgical treatment. Nevertheless, the utilization of an ostomy can effectively mitigate symptoms and facilitate healing in the affected region. Additionally, it decreases the likelihood of anastomotic leaks and postoperative infections[19-21].
Our research indicates that having serum albumin levels below 35 g/L and hemoglobin levels below 100 g/L increases the likelihood of requiring stoma formation. Patients with Crohn's disease may experience hypoalbuminemia due to factors such as intestinal inflammation, malabsorption of nutrients, and increased protein loss. Previous studies have demonstrated that preoperative low albumin levels are associated with a higher risk of IASC following anastomosis in patients with CD[22-24]. Additionally, preoperative anemia has been identified as a potential risk factor for postoperative complications in surgery for IBD[25-27]. Both albumin and hemoglobin play crucial roles in the incision healing process by maintaining adequate tissue perfusion, delivering nutrients and oxygen to the wound site, and promoting wound healing. Therefore, these unfavorable factors may have influenced the surgeon's decision to choose surgical stoma formation.
In patients with CD, the presence of perianal lesions increases the likelihood of requiring stoma formation. These perianal lesions are often indicative of a more severe form of the disease and have a high rate of recurrence even after surgical intervention[28, 29]. By diverting the fecal stream away from the perianal area, stoma formation can effectively manage these recurrent lesions, reducing the risk of additional damage and inflammation[17, 30]. This is why patients with CD and perianal lesions have a higher rate of stoma formation as part of their treatment.
We have observed a higher incidence of stoma requirement in patients undergoing open surgery, particularly in cases of Penetrating CD. This is attributed to the presence of bowel structure alterations and the development of adhesions, which pose challenges to laparoscopic surgery[31]. Additionally, Penetrating CD is a severe form of the disease that often presents with complications such as strictures, abscesses, or fistulas. In such complex cases, open surgery has demonstrated greater efficacy in addressing these issues compared to laparoscopic surgery. Recent studies have corroborated these findings, providing further support for this conclusion[32].
Our discovery of an increased likelihood of colonic CD stoma formation aligns with previous research[32]. Patients with colonic CD often exhibit a limited response to medical interventions, including immunosuppressants and biologic agents[33, 34]. Furthermore, Crohn's disease affecting the colon tends to present with more severe symptoms. Additionally, when colectomy is performed in CD patients, there is a higher incidence of IASCs following anastomosis[35, 36]. Consequently, surgeons may exercise greater caution when dealing with colonic lesions and aim to prevent the need for primary anastomoses.
This study has several limitations that should be acknowledged. Firstly, it is important to note that the study design is retrospective, and the datasets used for training the model and for validation are derived from the same center. This may introduce potential bias and limit the generalizability of the findings to other settings. Secondly, the inclusion period for the study was long, and during this time, there may have been changes in the selection and administration methods of drugs for penetrating CD. These changes could impact the accuracy and applicability of the model. Lastly, it is crucial to highlight that the nomogram developed in this study has not been validated by other central cohorts. Therefore, further research is needed to validate its generalizability and reliability in different populations and settings.
In summary, we have identified several factors that predict the likelihood of stoma creation in elective surgery for penetrating CD and have developed a new nomogram to accurately predict the probability of stoma formation. Our nomogram exhibits good accuracy, calibration, and clinical utility. This tool can assist surgeons in individualizing predictions regarding the likelihood of surgical stoma formation in patients with penetrating CD, thereby facilitating informed surgical decision-making.