Dynamic changes in the aggregate index of systemic inflammation (ΔAISI) predict postoperative complications following ileostomy closure in rectal cancer patients

Abstract

Background

Ileostomy closure following rectal cancer surgery is generally considered a low-risk procedure; however, postoperative complications remain clinically relevant. Reliable and easily accessible biomarkers for predicting these complications are lacking. This study aimed to evaluate the predictive value of dynamic changes in the aggregate index of systemic inflammation (ΔAISI) for postoperative complications following ileostomy closure.

Methods

This retrospective cohort study included 137 patients who underwent curative rectal cancer surgery with diverting ileostomy followed by ileostomy closure between January 2015 and March 2026. Inflammatory and nutritional markers were recorded prior to primary surgery and before ileostomy closure. ΔAISI was calculated as the difference between these two time points. Comparisons were performed using the Mann–Whitney U test and chi-square test. Receiver operating characteristic (ROC) curve analysis and multivariate logistic regression were conducted to evaluate predictive performance.

Results

Postoperative complications occurred in 23 patients (16.8%). No significant differences were observed in inflammatory markers measured before primary surgery. In contrast, several inflammatory markers measured prior to ileostomy closure were significantly associated with postoperative complications. Among the evaluated indices, ΔAISI demonstrated good discriminative ability (AUC: 0.755, 95% CI: 0.628–0.882, p < 0.001). In multivariate analysis, high ΔAISI remained an independent predictor of postoperative complications (OR: 4.13, 95% CI: 1.41–12.03, p = 0.009).

Conclusion

ΔAISI is a promising and independent predictor of postoperative complications following ileostomy closure in rectal cancer patients. Both pre-closure AISI and ΔAISI demonstrated good predictive performance, whereas AISI measured before primary surgery showed poor discrimination. These findings suggest that inflammatory status immediately before ileostomy closure may serve as a clinically useful biomarker for perioperative risk stratification.

Introductıon

Colorectal cancer continues to represent a significant global health burden, accounting for substantial morbidity and mortality worldwide [1]. Based on an analysis of previous years’ cancer statistics, an estimated 158,850 new cases of colorectal cancer are expected in the United States in 2026. Nearly half (45%) of these new diagnoses are projected to be among individuals currently under 65; this figure was 27% in 1995. Furthermore, an estimated 55,230 people will die from colorectal cancer in 2026, with approximately one-third of them being under 65 [1]. Surgical resection, particularly total mesorectal excision combined with neoadjuvant therapy in selected patients, represents the cornerstone of curative treatment for rectal cancer [2]. Deep pelvic colorectal anastomosis after resection is associated with a significant risk of anastomotic leakage, and this remains one of the most feared complications due to its association with increased morbidity, mortality, and impaired oncological outcomes [3].

To mitigate clinical outcomes in complications such as sepsis that may develop secondary to anastomotic leakage, the use of loop ileostomy for guidance in rectal cancer surgery has become a widely accepted strategy. Protective stomas have been shown to reduce the severity of leakage and decrease the need for reoperation, thus improving short-term outcomes [4].

Although ileostomy closure is generally considered a relatively low-risk procedure, postoperative morbidity rates are not negligible. Complications such as postoperative ileus, surgical site infection, anastomotic complications, and incisional hernia have been frequently reported, negatively impacting patient recovery and quality of life [5]. Postoperative morbidity following ileostomy closure has been reported with considerable variability across studies, with rates ranging between approximately 10% and 48%. Frequently encountered complications include small bowel obstruction, surgical site infection, stoma-related hernia, anastomotic complications, and enterocutaneous fistula formation. These early and late postoperative morbidities lead to prolonged hospital stays, additional costs, and decreased quality of life [6]. Furthermore, factors such as timing of stoma closure, perioperative treatment strategies, and patient-related characteristics have been investigated, yet no clear consensus has been established regarding optimal risk stratification [7].

Recent evidence highlights the impact of systemic inflammation on surgical outcomes. Inflammation plays a key role in tumor progression, immune response, and tissue healing, and is closely associated with postoperative complications [8]. Easily accessible inflammation-based biomarkers derived from routine blood tests—including neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), and composite indices such as the systemic immune-inflammation index (SII) and aggregate index of systemic inflammation (AISI)—have emerged as valuable prognostic tools in oncology [9].

AISI represents a composite index that captures the interaction between inflammatory activity and immune response. Previous studies have demonstrated its prognostic significance in various clinical settings, including cancer survival, cardiovascular outcomes, and chronic inflammatory diseases [8, 10–12]. Similarly, other inflammation-based indices have been shown to predict metastasis, survival, and treatment response in colorectal cancer patients [3]. In addition, nutritional and combined inflammatory-nutritional indices have also been associated with clinical outcomes, highlighting the importance of systemic physiological status in surgical recovery.

While the prognostic significance of inflammatory markers has been extensively investigated, data specifically evaluating their role in predicting complications after ileostomy closure remain limited. Moreover, most existing studies focus on single preoperative measurements, whereas the potential impact of dynamic changes in inflammatory status between primary rectal surgery and stoma closure has not been adequately explored.

We aimed to determine whether preoperative inflammatory and nutritional markers are associated with postoperative complications following ileostomy closure in patients with rectal cancer. Furthermore, we assessed whether dynamic changes in these markers between primary surgery and ileostomy closure improve the prediction of postoperative outcomes. Identification of reliable, cost-effective, and easily obtainable biomarkers may contribute to improved risk stratification and optimization of perioperative management in this patient population.

Materıals and methods

Study design and patient population

This retrospective cohort study included patients who underwent curative rectal cancer surgery with diverting ileostomy followed by ileostomy closure at the General Surgery Clinic of Gazi University Faculty of Medicine between January 2015 and March 2026. A total of 137 patients were included in the final analysis.

Inclusion criteria were: (1) patients aged 18 years and older (2), histopathologically confirmed rectal adenocarcinoma (3), elective and curative rectal resection and ileostomy, and (4) ileostomy closure performed at the same institution.

Exclusion criteria were: (1) emergency surgery due to obstruction or bleeding (2), presence of active infection at ileostomy closure (3), presence of hematological or chronic inflammatory disease that may affect systemic inflammatory markers (4), chronic immunosuppressive therapy, and (5) incomplete clinical or laboratory data.

Ethical approval for this study was obtained from the Ethics Committee of Gazi University Rectorate (Approval No: 2026-765, 14.04.2026). All procedures were performed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. Due to the retrospective design of the study, the requirement for informed consent was waived by the ethics committee.

Demographic, clinical, and pathological data were collected from hospital electronic medical records.

Data collection

The collected variables included:

• Demographic characteristics: age, gender

• Tumor-related parameters: tumor location (upper, middle, lower rectum), tumor size, TNM stage, overall stage (AJCC 8th Edition Cancer Staging Manual)

• Histopathological characteristics: lymphovascular invasion (LVI), perineural invasion (PNI), and modified Ryan tumor regression score in patients receiving neoadjuvant therapy

• Treatment-related variables: neoadjuvant therapy and adjuvant therapy status

• Time interval between primary surgery and ileostomy closure

Inflammatory and nutritional markers

Preoperative laboratory parameters were obtained at two time points:

1. Before primary rectal cancer surgery

2. Before ileostomy closure

Hematological and biochemical parameters included neutrophil, lymphocyte, monocyte, platelet counts, hemoglobin, and serum albumin levels.

The following inflammation-based indices were calculated:

• Neutrophil-to-lymphocyte ratio (NLR)

• Platelet-to-lymphocyte ratio (PLR)

• Lymphocyte-to-monocyte ratio (LMR)

• Systemic immune-inflammation index (SII = platelet × neutrophil / lymphocyte)

• Aggregate index of systemic inflammation (AISI = neutrophil × platelet × monocyte / lymphocyte)

• Hemoglobin, albumin, lymphocyte, platelet (HALP) score

Additionally, the change (Δ) in inflammatory markers between primary surgery and ileostomy closure was calculated for each parameter.

Outcome measures

The primary outcome of the study was the occurrence of postoperative complications following ileostomy closure.

Complications were classified according to the Clavien–Dindo classification system. Patients were divided into two groups:

• No complication group

• Complication group

Secondary outcomes included major complications, defined as Clavien–Dindo grade ≥ III.

Statistical analysis

All statistical analyses were performed using IBM SPSS Statistics version 31.0 (IBM Corp., Armonk, NY, USA).

Continuous variables were expressed as mean ± standard deviation or median (interquartile range), depending on data distribution, which was assessed using the Kolmogorov–Smirnov test. Categorical variables were presented as frequencies and percentages.

Comparisons between groups were performed using:

• Independent samples t-test or Mann–Whitney U test for continuous variables

• Chi-square test or Fisher’s exact test for categorical variables

Receiver operating characteristic (ROC) curve analysis was used to determine the predictive performance of inflammatory markers and to identify optimal cut-off values using the Youden index.

Variables with a p-value < 0.10 in univariate analysis were included in multivariate logistic regression analysis to identify independent predictors of postoperative complications. Results were presented as odds ratios (OR) with 95% confidence intervals (CI).

A p-value < 0.05 was considered statistically significant.

Results

A total of 159 patients were initially assessed for eligibility. Of these, 17 patients were excluded because primary rectal surgery had been performed at another institution, 1 patient due to the presence of synchronous malignancy in another organ, 2 patients because diverting ileostomy was not created during primary surgery but was performed later as part of salvage surgery following anastomotic leakage, and 2 patients due to a history of hematological disease.

Ultimately, 137 patients were included in the final analysis, of whom 23 (16.8%) developed postoperative complications following ileostomy closure. The patient selection process is summarized in Fig. 1. The types of complications included surgical site infection in 9 patients, prolonged ileus in 8 patients, intestinal obstruction in 3 patients, anastomotic leakage in 1 patient, postoperative bleeding in 1 patient, and acute renal failure in 1 patient.

Fig. 1.

Flow diagram of patient selection and postoperative outcomes following ileostomy closure

According to the Clavien–Dindo classification, 10 patients had grade I complications, 7 had grade II, 5 had grade III, and 1 had grade IV complications. Overall, 6 patients (26.1% of those with complications) experienced major complications (Clavien–Dindo ≥ III).

No early postoperative mortality was observed. The patient who developed acute renal failure required a single session of hemodialysis, after which renal function recovered. Three patients with intestinal obstruction underwent reoperation, and intraoperative findings revealed adhesions involving terminal ileal loops as the underlying cause.

One patient with postoperative bleeding required reoperation due to melena and a decrease in hemoglobin levels; mucosal bleeding at the anastomotic site was identified and controlled. Among patients with prolonged ileus, only one required prokinetic therapy, while the others were managed conservatively.

Of the patients who developed surgical site infection, three had superficial infections that resolved without antibiotic therapy, whereas six patients developed cellulitis requiring antibiotic treatment.

Baseline demographic and clinicopathological characteristics are summarized in Table 1. There were no significant differences between patients with and without complications in terms of age (p = 0.484), sex (p = 0.352), neoadjuvant therapy (p = 0.249), tumor stage (p = 0.760), lymphovascular invasion (p = 0.893), perineural invasion (p = 0.113), tumor differentiation (p = 0.338), tumor regression score (p = 0.086), surgical approach (p = 0.728), adjuvant therapy (p = 0.764), or interval to stoma closure (p = 0.758).

Table 1.

Baseline characteristics of patients according to postoperative complication status

Variable	No complication (n = 114)	Complication (n = 23)	p-value
Age (years)	63 (53–71)	64 (55–77)	0.484
Sex
Male	38 (%33.3)	10 (%43.5)	0.352
Female	76 (%66.7)	13 (%56.5)
Neoadjuvant therapy
No	36 (%31.6)	5 (%21.7)	0.249
Yes	78 (%68.4)	18 (%78.3)
Tumor location
Upper	35 (%30.7)	5 (%21.8)	0.026
Middle	62 (%54.4)	9 (%39.1)
Lower	17 (%14.9)	9 (%39.1)
Stage
0 (pCR)	15 (%13.1)	5 (%21.7)	0.760
1	27 (%23.7)	5 (%21.7)
2	41 (%36)	7 (%30.4)
3	31 (%27.2)	6 (%26.2)
Lymphovascular invasion
No	73 (%73.7)	13 (%72.2)	0.893
Yes	26 (%26.3)	5 (%27.8)
Perineural invasion
No	86 (%86.9)	13 (%72.2)	0.113
Yes	13 (%13.1)	5 (%27.8)
Differentiation
Well	34 (%34.3)	8 (%44.4)	0.338
Moderate	56 (%56.6)	7 (%38.9)
Poorly	9 (%9.1)	3 (%16.7)
Tumor regression score
0	15 (%19.2)	5 (%25)	0.086
1	5 (%6.4)	4 (%44.4)
2	37 (%47.5)	4 (%9.8)
3	21 (%26.9)	5 (%19.8)
Hospital stay	5 (4–6)	7 (6–10)	< 0.001
Surgical procedure
Open	73 (%64)	13 (%56.5)	0.728
Laparoscopic	26 (%22.8)	7 (%30.4)
Robotic	15 (%13.2)	3 (%13)
Adjuvant therapy
No	36 (%31.6)	8 (%34.8)	0.764
Yes	78 (%68.4)	15 (%65.2)
Interval to closure (months)	5 (3–8)	5 (2–10)	0.758

Data are presented as median (interquartile range) or n (%). Continuous variables were compared using the Mann–Whitney U test, and categorical variables using the chi-square test. Bold values indicate statistically significant differences (p < 0.05). Tumor regression score was evaluated in patients receiving neoadjuvant therapy (n = 96). Lymphovascular and perineural invasion were not assessed in patients with pathological complete response (n = 20)

Abbreviations: pCR pathological complete response

However, tumor location was significantly associated with postoperative complications (p = 0.026), with a higher proportion of lower rectal tumors observed in patients who developed complications. In addition, patients with complications had a significantly longer hospital stay compared to those without complications (7 vs. 5 days, p < 0.001).

There were no significant differences in inflammatory markers measured before primary rectal cancer surgery between the two groups (all p > 0.05), as detailed in Table 2.

Table 2.

Comparison of inflammatory markers before primary surgery and before ileostomy closure

Markers	No complication (n = 114)	Complication (n = 23)	p-value
AISI (primary)	391.19 (271.63–685.16)	501.125 (219.48–763.76)	0.712
AISI (closure)	315.16 (203.81–472.79)	692.05 (381.11–1065.55)	< 0.001
NLR (primary)	3.47 (2.36–4.78)	3.33 (2.19–4.8)	0.954
NLR (closure)	3.13 (2.00–4.17)	3.72 (3.11–6)	0.011
PLR (primary)	195.95 (136.25–328.5)	196 (146–300)	0.904
PLR (closure)	174.55 (128.48–240.15)	211.81 (147.39–315.44)	0.031
SII (primary)	758.95 (480.19–1220.00)	679.36 (462.22–1523.14)	0.922
SII (closure)	600.54 (415.00–834.53)	823.14 (741.36–1514.53)	< 0.001
LMR (primary)	1.82 (1.37–3.02)	1.87 (1.45–2.75)	0.947
LMR (closure)	2.20 (1.55–3.00)	1.61 (1.14–2.05)	0.011
HALP (primary)	2.57 (1.54–4.14)	2.89 (2.07–3.21)	0.867
HALP (closure)	3.05 (2.18–4.12)	2.31 (1.33–3.09)	0.019
Neutrophil 109/L (primary)	3.80 (3.00–4.75)	3.62 (2.90–4.80)	0.998
Neutrophil 109/L (closure)	3.50 (2.70–4.52)	4.37 (3.50–5.79)	< 0.001
Lymphocyte 109/L (primary)	0.99 (0.72–1.70)	1.00 (0.80–1.30)	0.820
Lymphocyte 109/L (closure)	1.10 (0.80–1.60)	1.10 (0.70–1.60)	0.682
Monocyte 109/L (primary)	0.54 (0.40–0.76)	0.59 (0.40–0.70)	0.842
Monocyte 109/L (closure)	0.50 (0.40–0.66)	0.60 (0.58–0.70)	0.012
Platelet 109/L (primary)	228.00 (176.50–281.25)	219.00 (159.00–269.00)	0.584
Platelet 109/L (closure)	192.50 (152.75–254.25)	244.00 (198.00–327.00)	0.023
Hemoglobin g/dL (primary)	13.00 (11.80–14.02)	12.00 (11.10–14.3)	0.381
Hemoglobin g/dL (closure)	12.60 (11.50–14.00)	12.60 (11.50–14.00)	0.263
Albumin g/L (primary)	42.00 (39.00–44.00)	43.00 (40.00–44.00)	0.571
Albumin g/L (closure)	42.50 (40.00–45.00)	42.00 (39.00–44.00)	0.326

Data are presented as median (interquartile range). Comparisons were performed using the Mann–Whitney U test. Bold values indicate statistically significant differences (p < 0.05)

In contrast, inflammatory markers measured prior to ileostomy closure demonstrated significant associations with postoperative complications. Patients who developed complications had significantly higher values of AISI (p < 0.001), NLR (p = 0.011), PLR (p = 0.031), SII (p < 0.001), neutrophil count (p < 0.001), monocyte count (p = 0.012), and platelet count (p = 0.023) before stoma closure.

Conversely, LMR (p = 0.011) and HALP (p = 0.019) values were significantly lower in patients with complications. No significant differences were observed in lymphocyte count, hemoglobin, or albumin levels prior to closure (all p > 0.05), as summarized in Table 2.

Changes in inflammatory markers between primary surgery and ileostomy closure were significantly associated with postoperative complications, as shown in Table 3.

Table 3.

Comparison of changes in inflammatory markers (Δ) between patients with and without postoperative complications

Markers	No complication (n = 114)	Complication (n = 23)	p-value
ΔAISI	-66.58 (-288.04–43.57)	207.74 (-18.49–389.63)	< 0.001
ΔNLR	-0.21 (-1.48–0.65)	0.46 (-1.12–1.7)	0.128
ΔPLR	-31.08 (-106.13–25.16)	38.61 (-61.59–85.47)	0.026
ΔSII	-106.86 (-405.93–148.83)	258.09 (-325.76–540.93)	0.005
ΔLMR	0.26 (-0.27–0.87)	-0.25 (-0.97–0.37)	0.008
ΔHALP	0.4 (-0.38–1.36)	-0.47 (-1.11–1.26)	0.077

Data are presented as median (interquartile range). Comparisons were performed using the Mann–Whitney U test. Bold values indicate statistically significant differences (p < 0.05)

ΔAISI demonstrated the most significant difference between groups and was significantly higher in patients who developed postoperative complications (p < 0.001). In addition, ΔSII (p = 0.005), ΔLMR (p = 0.008), and ΔPLR (p = 0.026) were also significantly associated with complication development.

Specifically, patients with complications showed an increase in AISI, SII, and PLR values, along with a decrease in LMR values over time. In contrast, no significant differences were observed for ΔNLR (p = 0.128) and ΔHALP (p = 0.077), as summarized in Table 3.

These findings highlight the potential value of dynamic inflammatory indices as predictive markers for postoperative complications.

Additional ROC analyses were performed to compare the predictive performance of AISI measured before primary rectal cancer surgery, AISI measured immediately before ileostomy closure, and ΔAISI, as shown in Table 4. AISI measured before primary surgery demonstrated poor discriminative ability for postoperative complications (AUC = 0.524, 95% CI: 0.387–0.662, p = 0.728). In contrast, AISI measured before ileostomy closure demonstrated the highest predictive performance (AUC = 0.784, 95% CI: 0.677–0.891, p < 0.001), followed by ΔAISI (AUC = 0.755, 95% CI: 0.628–0.882, p < 0.001).

Table 4.

ROC analysis comparing AISI measurements for predicting postoperative complications

Markers	AUC	95% CI	p-value
AISI before primary surgery	0.524	0.387–0.662	0.728
AISI before ileostomy closure	0.784	0.677–0.891	< 0.001
ΔAISI	0.755	0.628–0.882	< 0.001

ROC Receiver operating characteristic, AUC Area under the curve, CI Confidence interval, AISI Aggregate index of systemic inflammation, ΔAISI Change in AISI between primary surgery and ileostomy closure

Receiver operating characteristic (ROC) curve analysis demonstrated that ΔAISI had good discriminative ability in predicting postoperative complications, with an area under the curve (AUC) of 0.755 (95% CI: 0.628–0.882, p < 0.001). The optimal cut-off value for ΔAISI was identified as − 50.23, yielding a sensitivity of 78.3% and a specificity of 55.3% (Fig. 2).

Fig. 2.

Receiver operating characteristic (ROC) curve of ΔAISI for predicting postoperative complications. The area under the curve (AUC) was 0.755 (95% CI: 0.628–0.882, p < 0.001). The optimal cut-off value was − 50.23, with a sensitivity of 78.3% and a specificity of 55.3%

Additionally, a clinically relevant cut-off value of + 50 was evaluated, representing a clear increase in inflammatory burden and reflecting true postoperative inflammatory progression; however, this threshold was not used in the primary model.

Multivariate logistic regression analysis was performed to identify independent predictors of postoperative complications, as presented in Table 5. ΔAISI was dichotomized according to the ROC-derived cut-off value (− 50.23) into low and high ΔAISI groups. Patients in the high ΔAISI group had a significantly increased risk of complications (odds ratio [OR]: 4.13, 95% CI: 1.41–12.03, p = 0.009).

Table 5.

Multivariate logistic regression analysis for predictors of postoperative complications

Variable	OR	95% CI	p-value
High ΔAISI	4.13	1.41–12.03	0.009
Age	1.016	0.978–1.056	0.406
Stage	0.893	0.564–1.412	0.627
Neoadjuvant therapy	1.339	0.434–4.134	0.612

Data are presented as odds ratios (OR) with 95% confidence intervals (CI). ΔAISI was dichotomized according to the ROC-derived cut-off value (− 50.23) into low and high groups. Variables with p < 0.05 were considered statistically significant. The model demonstrated an acceptable goodness-of-fit according to the Hosmer–Lemeshow test (p = 0.104)

In contrast, age (p = 0.406), tumor stage (p = 0.627), and neoadjuvant therapy (p = 0.612) were not significantly associated with postoperative complications.

The model demonstrated an acceptable goodness-of-fit according to the Hosmer–Lemeshow test (p = 0.104).

When major complications (Clavien–Dindo ≥ III) were analyzed separately, ΔAISI was not significantly associated with major complications (p = 0.110).

Although ΔAISI was associated with overall postoperative complications, it did not show a statistically significant association with major complications, likely due to the limited number of major complication events. Therefore, the findings of the present study should primarily be interpreted in the context of overall postoperative morbidity rather than severe complications alone.

Discussion

The present study demonstrated that dynamic changes in systemic inflammatory status, particularly ΔAISI, are significantly associated with postoperative complications following ileostomy closure in patients undergoing rectal cancer surgery. Among the evaluated dynamic inflammatory markers, ΔAISI showed significant predictive ability and remained an independent predictor of postoperative complications in multivariate analysis.

Although ileostomy closure is generally considered a low-risk procedure, postoperative morbidity remains clinically relevant. Reported complication rates vary widely in the literature, ranging from 10% to nearly 50%, depending on patient selection and definitions used [6, 7, 13–15]. In line with previous studies, the overall complication rate in our cohort was 16.8%, confirming that ileostomy closure is not a trivial procedure and that improved risk stratification is warranted.

Systemic inflammation has increasingly been recognized as a key determinant of surgical outcomes. Various inflammation-based markers, including NLR, PLR, LMR, and SII, have been extensively studied in colorectal cancer and shown to be associated with tumor progression, metastasis, and clinical outcomes [3, 9, 16, 17]. However, most previous studies have focused on static preoperative measurements obtained prior to primary tumor resection. In contrast, our study highlights the importance of temporal changes in inflammatory status, demonstrating that markers measured prior to ileostomy closure, as well as their dynamic changes, appear to provide more clinically relevant information than baseline values obtained at the time of primary surgery. The wide interquartile range observed for ΔSII likely reflects substantial interindividual variability in inflammatory response between primary surgery and ileostomy closure.

Notably, no significant differences were observed in inflammatory markers measured before primary rectal cancer surgery, whereas multiple markers measured prior to ileostomy closure showed significant associations with postoperative complications. An important finding of the present study was that AISI measured immediately before ileostomy closure demonstrated slightly higher discriminative ability than ΔAISI. This suggests that a single preoperative measurement may capture much of the clinically relevant inflammatory burden while offering greater practical applicability in routine clinical practice. Nevertheless, the significant association observed for ΔAISI supports the concept that dynamic inflammatory alterations between primary surgery and stoma closure may also reflect postoperative vulnerability.

Several mechanisms may explain elevated inflammatory markers before ileostomy closure. Persistent low-grade inflammation following rectal cancer surgery, stoma-related irritation, subclinical infectious processes, nutritional impairment, delayed recovery after neoadjuvant treatment, and alterations in gut microbiota may all contribute to a sustained pro-inflammatory state. These factors may impair tissue healing, compromise immune regulation, and increase susceptibility to postoperative complications following stoma closure.

Additional ROC analyses demonstrated that AISI measured immediately before ileostomy closure had the highest discriminative performance (AUC = 0.784), followed by ΔAISI (AUC = 0.755), whereas AISI measured before primary surgery showed poor predictive ability (AUC = 0.524). These findings suggest that inflammatory status at the time of stoma closure is more relevant than baseline inflammatory status measured before primary rectal cancer surgery.

From a pathophysiological perspective, AISI integrates multiple components of the inflammatory and immune response, including neutrophils, monocytes, platelets, and lymphocytes. Previous studies have demonstrated that AISI reflects the balance between inflammatory activation and immune competence and is associated with adverse outcomes in various clinical conditions [8, 10–12, 18, 19]. An increase in ΔAISI may therefore indicate a shift toward a pro-inflammatory and immunosuppressive state, which could impair tissue healing, increase susceptibility to infection, and predispose patients to postoperative complications.

Importantly, dynamic changes in inflammatory markers may provide complementary information beyond static measurements by reflecting longitudinal alterations in systemic inflammatory status [20, 21]. In this context, ΔAISI may capture changes occurring between primary surgery and ileostomy closure, potentially reflecting postoperative vulnerability and perioperative risk.

In addition to the statistically derived cut-off value, we also explored a clinically intuitive threshold of + 50, representing a clear increase in inflammatory burden. Although this threshold was not used in the primary model, it may provide a more practical interpretation in clinical settings, allowing surgeons to easily identify patients with worsening inflammatory status.

When major complications were analyzed separately, ΔAISI did not show a statistically significant association. This finding is likely attributable to the limited number of major complication events in our cohort, which may have reduced the statistical power of the analysis. Larger studies are needed to further evaluate the relationship between ΔAISI and severe postoperative outcomes.

The findings of this study have important clinical implications. ΔAISI is a simple, cost-effective, and readily available biomarker that can be calculated from routine laboratory parameters without additional burden. Its use may facilitate preoperative risk stratification prior to ileostomy closure and help identify patients who may benefit from closer monitoring, delayed closure, or optimization of inflammatory status before surgery. Such optimization strategies may include nutritional assessment, treatment of occult infections, correction of reversible inflammatory conditions, and reassessment of the timing of stoma closure. This approach is consistent with enhanced recovery strategies and personalized perioperative care models [15, 22].

This study has several limitations. First, its retrospective design may introduce selection bias. Second, the sample size is relatively limited, particularly for the analysis of major complications. Third, the study was conducted at a single center, which may limit generalizability. In addition, potentially relevant parameters such as C-reactive protein levels, body mass index, nutritional assessment scores, and sarcopenia-related measurements were not routinely available and therefore could not be evaluated in the present study. Furthermore, the present study focused exclusively on diverting ileostomies after rectal cancer surgery, and whether these findings can be extrapolated to colostomy closure remains uncertain. Finally, external validation in independent cohorts is required before ΔAISI can be widely adopted in clinical practice.

In conclusion, ΔAISI is a promising and independent predictor of postoperative complications following ileostomy closure in patients with rectal cancer. Both pre-closure AISI and ΔAISI demonstrated good predictive performance, whereas AISI measured before primary surgery showed poor discrimination. These findings suggest that inflammatory status immediately before ileostomy closure may represent a clinically useful biomarker for perioperative risk stratification. Future prospective, multicenter studies are warranted to validate these findings and to further explore the role of ΔAISI in clinical decision-making.

Abbreviations

AISI

Aggregate Index of Systemic Inflammation

ΔAISI

Change in Aggregate Index of Systemic Inflammation between primary surgery and ileostomy closure

NLR

Neutrophil-to-Lymphocyte Ratio

PLR

Platelet-to-Lymphocyte Ratio

LMR

Lymphocyte-to-Monocyte Ratio

SII

Systemic Immune-Inflammation Index

HALP

Hemoglobin, Albumin, Lymphocyte, Platelet Score

ROC

Receiver Operating Characteristic

AUC

Area Under the Curve

OR

Odds Ratio

CI

Confidence Interval

Authors’ contributions

MY conceived the study, collected data, performed statistical analyses, interpreted the results, prepared the tables and figures and drafted the manuscript.SK contributed to data collection and manuscript revision.RK contributed to study design, interpretation of findings, and critical revision of the manuscript.MAT contributed to data interpretation and manuscript revision.MS supervised the study and critically revised the manuscript.All authors read and approved the final manuscript.

Funding

The authors received no financial support for this study.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval for this study was obtained from the Ethics Committee of Gazi University Rectorate (Approval No: 2026 − 765, 14.04.2026). All procedures were performed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. Due to the retrospective design of the study, the requirement for informed consent was waived by the Ethics Committee of Gazi University Rectorate.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.