Abstract
Background/Aim: Despite advances in diagnosis and pharmacotherapy, surgery remains crucial for Crohn’s disease (CD). Postoperative intra-abdominal septic complications (IASC) occur in 1.2-16.7% of cases. We investigated the frequency of postoperative IASC in elective surgeries for CD and the risk factors and potential biomarkers for postoperative IASC.
Patients and Methods: We conducted a retrospective single-center cohort study of patients who underwent abdominal surgery for CD at Fukuoka University Chikushi Hospital between January 2015 and December 2023. The primary focus was the incidence of IASC within 60 days postoperatively. Patient-related variables were examined using univariate and multivariable analyses.
Results: The analysis included 206 of 249 surgeries. Postoperative IASC occurred in 26 patients (12.6%). Univariate analysis identified history of steroid use requiring steroid coverage (p=0.002), penetrating type (p=0.020), WBC count (p=0.037), neutrophil count (0.009), C-reactive protein (CRP) (p=0.035), CRP-albumin ratio (CAR) (p=0.034), neutrophil-to-lymphocyte ratio (NLR) (p=0.002), and operation duration (p=0.010) as significant factors. Multivariable analysis identified history of steroid use requiring steroid coverage (OR=6.23, 95%CI=1.61-24.1, p=0.008), high NLR (OR=3.43, 95%CI=1.30-9.04, p=0.013), and long duration of operation (OR=2.63, 95%CI=1.01-6.88, p=0.049) as independent predictors. The optimal cutoffs for predicting IASC were an NLR of 3.98 (sensitivity, 61.5%; specificity, 77.8%) and an operation time of 173 min (sensitivity, 65.4%; specificity, 65.0%), respectively.
Conclusion: History of steroid use requiring steroid coverage, preoperative NLR ≧3.98, and duration of operation ≧173 min are independent risk factors for postoperative IASC in elective surgeries for CD. Recognition of high-risk patients would contribute to the decision-making process for perioperative management.
- Crohn’s disease
- intra-abdominal septic complication
- neutrophil-to-lymphocyte ratio
- postoperative complication
- steroid use
- biomarker
Introduction
Despite significant reductions in lifetime surgical risk owing to advances in early diagnostic techniques and pharmacotherapy, surgery remains a key component in the treatment of Crohn’s disease (CD). The cumulative risk of initial surgery in patients diagnosed with CD in the 21st century is reported to be 12.3% at 1 year, 18.0% at 5 years, and 26.2% at 10 years, whereas the risk of reoperation is reported to be 14.8% at 5 years and 25.5% at 10 years (1).
Surgery for patients with CD is associated with higher complication rates than surgeries for other benign diseases owing to factors such as immunosuppression, malnutrition, and active bowel inflammation. Among these complications, postoperative intra-abdominal septic complications (IASC) are common, occurring in 1.2% to 16.7% of cases (2). Postoperative IASC not only prolong hospitalization for aggressive treatment, thus lowering the patient’s quality of life, but also significantly increase the rate of endoscopic and surgical postoperative recurrence (3-6). Multiple surgeries are also reported as a risk factor for postoperative IASC (7, 8). Therefore, IASC should be evaluated separately from the other complications. It is essential to predict and stratify high-risk patients with IASC, as this is crucial for decision-making and for obtaining informed consent prior to surgery. Although factors such as steroid use, preoperative low albumin levels, preoperative abscess formation, and immunosuppressive agents have been reported as risk factors for postoperative IASC (9-11), there is no consensus regarding these factors, except for relatively consistent reports on steroid use.
Recently, the usefulness of simple indicators like the C-reactive protein (CRP)-albumin ratio (CAR), the neutrophil-to-lymphocyte ratio (NLR), and the platelet-to-lymphocyte ratio (PLR), which can be easily calculated from routine clinical data, has attracted attention in several conditions including various malignancies (12, 13), autoimmune disease (14), cardiovascular disease (15), and immunotherapy (16). However, few studies have reported the role of these indicators in CD, and no consensus has been reached regarding their utility (17-20).
This study aimed to identify the frequency of postoperative IASC in elective surgeries for patients with CD at our institution and to explore risk factors and potential biomarkers for the occurrence of postoperative IASC, including several simple indicators derived from preoperative blood tests.
Patients and Methods
Patients. This was a retrospective single-center cohort study that included patients who underwent abdominal surgery for CD at Fukuoka University Chikushi Hospital between January 2015 and December 2023. All patients were followed-up for at least 60 days postoperatively. Cases involving emergency surgery, laparoscopic surgery, preoperative or postoperative pathological diagnosis of cancer, or missing data were excluded. The study protocol was approved by the Institutional Review Board of Fukuoka University (C21-01-013) and was disclosed on the hospital website, providing an opportunity to refuse through the opt-out approach.
Surgical procedure. Patients scheduled for surgery were preconditioned by fasting and intensive total parenteral nutrition for 4-6 weeks to achieve bowel rest and decompression. If intra-abdominal abscess formation was observed preoperatively, antibiotic therapy and, if possible, percutaneous drainage was used. Elective surgery was performed once the inflammatory response normalized and bowel decompression was achieved. Anastomosis was performed using the Gambee technique, considering maximal preservation of the bowel and ease of postoperative surveillance.
Dependent and independent variables. The primary outcome measure was the incidence of IASC within the 60-day postoperative period. Postoperative IASC included anastomotic leakage and intra-abdominal abscesses. IASC was diagnosed based on the presence of any of the following: intestinal fluid or fecal discharge from the incisional wound or drain tract, intra-abdominal abscess or peritonitis, and an anastomotic defect verified by imaging. An intraperitoneal abscess near the anastomotic site without an obvious fistula was also diagnosed as a clinical leakage.
Independent clinical variables included patient demographics (sex, age, body mass index, diabetes), disease characteristics (disease location, age at diagnosis, preoperative treatment, clinical forms, indication for surgery, abscess, previous laparotomy for CD), preoperative laboratory values [white blood cell (WBC) count, hemoglobin (Hb), platelets, CRP, albumin], and surgical characteristics (surgical procedure, duration of surgery, blood transfusion). The patients were classified into two clinical forms, penetrating and non-penetrating, according to the classification by Greenstein et al. (21). The following indices, which consider inflammation and immunonutrition, were calculated from the preoperative blood tests: CAR, NLR, PLR, and prognostic-nutrition index (PNI). PNI was calculated as follows: (10×serum albumin (g/dl))+ (0.005× total lymphocyte value) (22). Preoperative laboratory data were evaluated based on blood test data obtained closest to the day of surgery. Patients without available blood test data within three weeks prior to surgery were excluded from the analysis as “missing data”. Preoperative treatment was classified as systemic steroids [budesonide (BD) or prednisolone (PSL)], immunomodulators [azathioprine (AZA), 6-mercaptopurine (6-MP), or methotrexate (MTX)], and biologics [infliximab (IFX), adalimumab (ADA), ustekinumab (UST), or vedolizumab (VDZ)]. In this study, patients who had been administered steroids at a prednisone dose of 5 mg/day or more within six months prior to surgery received perioperative steroid coverage as stress-dose steroids.
Statistical analysis. The normality of the distribution of the numerical data was evaluated using the Kolmogorov–Smirnov test. Continuous variables were presented as medians with ranges according to data distribution, while nominal variables were expressed as absolute values with percentages. For the numerical data, a two-sided Mann–Whitney U-test was used to compare differences between the no IASC and IASC groups. For categorical variables, the two-sided Fisher’s exact test was used to compare the differences between the no IASC and IASC groups. After univariate analysis, variables with a p-value ≤0.05 were selected for multivariable analysis using a logistic regression analysis. Receiver operating characteristic (ROC) curve analysis was used to identify the optimal cut-off values for CAR, NLR, and duration of operation. For the multivariable analysis, logistic regression analysis was used to estimate the odds ratio for each independent variable. Statistical significance was set at p<0.05. All analyses were performed using JMP pro 17 software (SAS Institute, Cary, NC, USA).
Results
Of the 249 abdominal surgeries performed on patients with CD during the study period, 206 were included in the analysis after excluding 12 laparoscopic surgery cases, nine cancer cases, eight emergency surgery cases, and 14 missing data cases. The median age was 42 years (range=15-73 years), and 153 (74.2%) were men. Of the 206 patients, 22 (10.7%) received systemic steroids, 40 (19.4%) received immunomodulators, and 113 (54.9%) received biologic therapy, preoperatively. The most common indication for surgery was structuring in 111 patients (53.9%), followed by penetration in 85 patients (41.3%). Three patients were classified as refractory to therapy, a category that included those with abdominal pain or hemorrhage.
Preoperative abscess formation was noted in 22 (10.7%) patients, who were subsequently treated with antibiotics with or without percutaneous drainage. Intraoperative drainage of pus was observed in four patients in the non-IASC group and two patients in the IASC group. The two patients in the IASC group did not have a preoperative diagnosis of intra-abdominal abscess.
A total of 126 (61.2%) patients underwent prior abdominal surgery for CD. Postoperative IASC occurred in 26 (12.6%) patients. There were no deaths among the patients of interest. The median postoperative period for diagnosis of IASC was 7.5 days (range=3-21 days). None of the patients required re-operation. Twelve patients required radiological intervention, of whom six required adjustments of the drains placed during surgery, and the remaining six required new percutaneous drainage. The remaining 14 were successfully treated with antibiotics alone. The IASC group included two procedures that were performed on the same patient with a history of heavy particle therapy of the abdomen.
Univariate analysis. None of the independent variables related to patient demographics were associated with abscess formation (Table I). Among the independent variables related to disease characteristics, steroid use (p= 0.041), clinical forms (0.020), and indication for surgery (p= 0.046) were significantly associated with the development of IASC (Table II). Specifically, history of steroid use requiring steroid coverage significantly correlated with the development of IASC (p=0.002). Among the independent variables related to preoperative laboratory values, the WBC count (p=0.037), neutrophil count (p=0.009), CRP (p=0.035), lymphocyte count (p=0.002), CAR (p=0.034), and NLR (p= 0.002) were significantly associated with IASC development (Table III). Among the independent variables related to surgical characteristics, duration of surgery (p= 0.010) was significantly associated with a higher incidence of IASC (Table IV).
Patient demographics and intra-abdominal septic complications (IASC) development.
Disease characteristics and intra-abdominal septic complications (IASC) development.
Preoperative laboratory values and intra-abdominal septic complications (IASC) development.
Surgical characteristics and intra-abdominal septic complications (IASC) development.
Multivariable analysis. Table V summarizes the results of the multivariable analysis. The best cutoff of NLR for predicting postoperative IASC was 3.98 with a sensitivity of 61.5% and with a specificity of 77.8%. Positive and negative predictive values were 28.6% and 93.3%, respectively. The ROC curve analysis is shown in Figure 1. The best cutoff for duration of operation for prediction of postoperative IASC was 173 min, with a sensitivity of 65.4% and a specificity of 65.0%. Positive and negative predictive values were 21.3% and 92.9%, respectively. The ROC curve analysis is shown in Figure 2. History of steroid use requiring steroid coverage (OR=6.23, 95%CI=1.61-24.1, p=0.008), high NLR (OR=3.43, 95%CI=1.30-9.04, p=0.013), and long duration of operation (OR=2.63, 95%CI=1.01-6.88, p=0.049) remained as significant independent predictors of postoperative IASC development.
Multivariable analysis.
Graph showing the receiver operating characteristic (ROC) curve for efficacy of the neutrophil-to-lymphocyte ratio (NLR) as a predictor of postoperative IASC development.
Graph showing the receiver operating characteristic (ROC) curve for efficacy of the duration of operation as a predictor of postoperative IASC development.
Discussion
In this study, we identified history of steroid use requiring steroid coverage, high preoperative NLR (≧3.98), and long duration of operation (≧173 min) as risk factors for postoperative IASC. Although we could not specify the duration or total amount of steroid administration, steroid use is a crucial determinant of postoperative IASC formation, as mentioned in the guidelines (23). In addition, the duration of surgery (24, 25) has been reported to be a risk factor for postoperative IASC in CD surgery, which is consistent with our results. In other studies, CAR has been reported to be effective in evaluating the activity of CD (26, 27). Moreover, the penetrating type (5, 28) has been reported to be a risk factor for postoperative IASC in CD surgery. Additionally, perioperative blood transfusion has been reported as a risk factor for postoperative infections, including IASC (29). However, our results did not identify these associations in multivariable analysis.
Several studies have reported the utility of the NLR in CD. As a screening tool, NLR has been reported to be useful in distinguishing patients with CD from healthy individuals (20, 30). Additionally, NLR is reportedly effective in differentiating between the active and inactive phases of CD (31-33). In the context of surgical treatment, preoperative NLR values have been reported to be correlated with postoperative complications and recurrence (18, 19, 34). In our study, neutrophil count was significantly correlated with postoperative IASC, while lymphocyte count tended to be lower in the IASC group. Additionally, the ratio of these two counts, the NLR, showed an even more sensitive correlation. This finding aligns with that of Tsunoda et al. (35); however, the correlation between NLR and other complications, such as wound infection, ileus, and catheter-related bloodstream infection, was not evident in this study (data not shown). Generally, neutrophils reflect ongoing inflammation, whereas lymphocytes are markers of immune regulatory responses (36). In highly inflammatory states, neutrophil apoptosis is delayed (37, 38), and lymphocyte apoptosis is promoted (39). Lymphocytes in IBD patients are known to exhibit functional abnormalities at both peripheral blood and mucosal levels (40). Therefore, the NLR may be a useful biomarker reflecting disease activity in CD. Indeed, some studies have suggested that the NLR reflects disease activity in CD, indicating that intestinal surgeries performed in states of high NLR (active inflammation) pose a risk for postoperative IASC. Interestingly, in pediatric patients with IBD, the NLR at diagnosis correlated with CRP but did not show a correlation during remission (32). Additionally, in our study, the NLR was not correlated with the occurrence of wound infection. Collectively, these findings suggest that the preoperative NLR might reflect intestinal-specific inflammatory activity that does not manifest as changes in CRP or WBC levels.
The guidelines recommend preoperative optimization of nutrition via enteral or parenteral routes (23). It has been reported that preoperative nutritional therapy lowers CRP and that changes in preoperative CRP correlate with postoperative IASC incidence (33-35). In seven patients who underwent their first surgery for CD laparoscopically, although they were excluded from the current analysis, preoperative nutritional optimization significantly reduced the preoperative NLR compared with that at admission (Figure 3). These findings suggest that the NLR could be a modifiable biomarker during preoperative nutritional conditioning to improve patient outcomes.
Analysis of neutrophil-to-lymphocyte ratio (NLR) values in seven patients who underwent laparoscopic surgery. Boxplot representing the median, first, and third quartiles and minimum and maximum NLR values. The Mann–Whitney U-test was conducted to compare the differences between the NLR values at the time of admission and before surgery. *Statistically significant difference (p<0.05).
We found that history of steroid use requiring steroid coverage, a high preoperative NLR, and long surgery duration were risk factors for postoperative IASC. Preventing postoperative IASC is crucial for the surgical treatment of CD. For patients with these risk factors, laparoscopic or two-stage surgery may be necessary, as mentioned in the guidelines (23), and the risk of postoperative IASC should be explained. Additionally, NLR may be particularly advantageous, given that it is minimally invasive, economical, and easily calculated from blood count data in general clinical practice.
Study limitations. The limitations of this study include its retrospective nature and heterogeneity of patient backgrounds and surgical procedures. Additionally, we could not evaluate smoking history, which is a known risk factor for anastomotic leakage. Nevertheless, we believe that we have added new evidence regarding risk factors and potential biomarkers for postoperative IASC development in patients with CD. Validation of our findings is highly desirable.
Conclusion
History of steroid use requiring steroid coverage, high preoperative NLR (≧3.98), and long duration of operation (≧173 min) are independent risk factors for postoperative IASC in elective surgeries for CD. Surgeons should recognize high-risk patients in order to contribute to the decision-making process for perioperative management. Additionally, the NLR could be a potential biomarker during preoperative conditioning.
Acknowledgements
The Authors thank all members of the staff of the Department of Surgery, Gastroenterology, and Pathology at Fukuoka University Chikushi Hospital for the medical treatment provided to the patients. The Authors also express their sincere gratitude to Natsumi Eguchi for her invaluable support throughout this study.
Footnotes
Authors’ Contributions
Conceptualization: MK. Data curation: MK, KK, NT. Formal analysis and investigation: MK, DH, YM, HA. Writing – original draft: MK. Writing – review and editing: DH, YM, SN, TH. Methodology: MK, HA. Funding acquisition: MK. Supervision: MW.
Conflicts of Interest
The Authors declare that they have no competing interests in relation to this study.
Funding
This work was supported by the JSPS KAKENHI grant number JP21K20817 (to MK) and research grants from the Nakatomi Foundation (to MK).
- Received February 10, 2025.
- Revision received February 20, 2025.
- Accepted February 21, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).









