Risk factors for prolonged postoperative ileus in elderly patients undergoing laparoscopic radical resection for colorectal cancer

Abstract

BACKGROUND

As the elderly population of patients with colorectal cancer (CRC) has increased, the risk of postoperative complications, particularly prolonged postoperative ileus (PPOI), has similarly increased in frail patients; this finding has hindered the postoperative recovery process. However, the risk factors for PPOI in elderly patients following laparoscopic radical resection for CRC remain unclear.

AIM

To investigate and analyze the incidence and potential risk factors for PPOI in elderly patients with CRC following laparoscopic radical resection.

METHODS

A single-center, retrospective study was conducted involving 167 patients who underwent laparoscopic radical CRC resection from November 2018 to June 2025. Clinical data on baseline characteristics, preoperative laboratory tests, intraoperative and postoperative parameters, and pathological outcomes were collected and analyzed. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for PPOI, and a P value of < 0.05 was considered to indicate statistical significance. Receiver operating characteristic curves were plotted, bootstrap validation was performed, and calibration curves and decision curve analysis were constructed.

RESULTS

In this study, 49 patients developed PPOI (29.30%), whereas 71 patients experienced frailty (42.50%). The multivariate analysis revealed that modified 5-item Frailty Index (mFI-5) ≥ 2 (P = 0.028), postoperative opioid use (P = 0.021), and serum potassium levels on postoperative day 1 (POD 1) < 3.61 mmol/L (P = 0.003) were associated with PPOI. The area under the curve (AUC) analysis demonstrated a predictive performance of 0.730 [95% confidence interval (CI): 0.649-0.811] for PPOI in elderly patients, with a bootstrap-validated AUC of 0.730 (95%CI: 0.644-0.800).

CONCLUSION

PPOI is a common postoperative complication in elderly patients with CRC following laparoscopic radical resection, while an mFI-5 ≥ 2, postoperative opioid use, and low serum potassium levels on POD 1 are independent risk factors. The predictive model constructed on the basis of these indicators demonstrated moderate discrimination. The optimization of perioperative evaluation and management protocols is crucial for reducing the incidence of PPOI.

Key Words: Frailty; Elderly patients; Colorectal cancer; Prolonged postoperative ileus; Risk factors

Core Tip: This single-center retrospective study investigated and analyzed the incidence and risk factors for prolonged postoperative ileus (PPOI) in geriatric patients with colorectal cancer following laparoscopic radical resection. A more comprehensive and accurate risk stratification model for PPOI was developed by jointly assessing the patients' global functional status and modifiable clinical indicators, which provided empirical evidence to support the development of individualized and precise perioperative management in the future. For elderly patients, optimizing perioperative evaluation and management protocols is crucial to reduce the incidence of PPOI and promote postoperative recovery.

INTRODUCTION

As a reflection of socioeconomic development, colorectal cancer (CRC) ranks third in terms of incidence but second in terms of mortality and has been steadily increasing in countries undergoing major transitions[1]. Despite the widespread use of minimally invasive techniques and enhanced recovery after surgery (ERAS) protocols in the treatment of CRC, prolonged postoperative ileus (PPOI), a major complication of gastrointestinal functional recovery, remains an important contributor to increased socioeconomic burden, with an incidence ranging from 10% to 30%[2-4]. The occurrence of PPOI is attributable to the interaction of multiple factors via distinct pathways. Its main pathogenesis can be categorized into three interrelated and synergistic mechanisms-neural, inflammatory and pharmacological-that collectively impede intestinal function recovery, as elaborated below.

As the initiating trigger of PPOI, the neural mechanism is activated immediately upon surgical incision. Sympathetic activation induced by the incision elicits an early intestinal inflammatory response, which is transient and typically resolves within hours post-operatively[5]. The inflammatory and intestinal barrier mechanism is the core driver of persistent PPOI symptoms, which are initiated 3-4 hours post-operatively and last for several days[5]. Surgical manipulation activates local macrophages to release chemokines, recruiting circulating leukocytes to amplify inflammation and spread it from the surgical site to the entire gastrointestinal tract[6,7]. Additionally, inflammatory mediators from activated immune cells not only directly exacerbate gastrointestinal dysfunction but also inhibit myenteric neuron activity by enhancing intestinal sympathetic output, thereby suppressing intestinal function recovery[8]. Pharmacological factors are important iatrogenic contributors to PPOI, with postoperative opioid analgesics having the most significant effect. These drugs bind to intestinal μ-receptors, ultimately inhibiting intestinal peristalsis and hindering intestinal function recovery[5]. In summary, the neural mechanism initiates PPOI, the inflammatory mechanism sustains its symptoms, and the pharmacological mechanism serves as an important auxiliary factor. Their interaction forms the complex pathogenic network of PPOI. Currently, the diagnosis of PPOI is generally based on the standardized definition proposed by Vather et al[9] in 2013, which was also adopted in the present study.

With the accelerated progression of population aging[10], the proportion of elderly patients undergoing intestinal surgery continues to increase[11]. These patients often present with complex clinical features, including comorbidities, limited mobility, and polypharmacy[12], which pose increasing challenges for managing colorectal surgery in high-risk patients. Traditionally, patient age and the American Society of Anesthesiologists (ASA) score have been regarded as important factors associated with the prognosis of CRC[5,13-15], particularly with respect to postoperative gastrointestinal recovery, which remains a significant clinical challenge.

Frailty is particularly prevalent in elderly patients, with an incidence ranging from 28% to 38.4% according to previous studies[16-18]. Notably, a meta-analysis suggested that compared with age and ASA score, frailty better reflects the decline in the body's resistance and reserve capacity in response to stressors[19]. Frailty, as a clinical syndrome, is characterized by the decline of multiple physiological systems with advancing age, leading to a significant reduction in an individual’s ability to maintain homeostasis and cope with stress[20]. Therefore, assessing frailty may offer a more comprehensive understanding of a patient’s physiological reserve and ability to tolerate stress, thereby enabling more accurate preoperative risk stratification[16]. However, findings to date remain inconsistent, which is attributable to variations in study populations and the diversity of frailty assessment tools used[21-23]; thus, no consensus has yet been reached regarding the optimal assessment method[16]. One of the earliest and most comprehensive risk assessment tools is the 70-item frailty index based on the Canadian Study of Health and Aging, which has demonstrated significant prognostic value in predicting adverse surgical outcomes. On the basis of the American College of Surgeons National Surgical Quality Improvement Program database, researchers developed a comparable 11-item frailty index. This was subsequently refined into a 5-factor model known as the modified 5-item Frailty Index (mFI-5)[24]. The mFI-5 is distinguished by its ease of clinical application; moreover, its predictive efficacy and clinical practicality have been verified in the Chinese population[12]. Multiple studies have confirmed that frailty status assessed via the mFI-5 is significantly associated with PPOI occurrence in patients undergoing abdominal surgery[16,17]. This scale facilitates efficient screening in clinical settings, promoting early detection and intervention to increase a patient’s physiological reserve[14]. However, whether it can serve as a risk factor for PPOI in elderly patients with CRC remains insufficiently characterized.

Owing to significant variations in clinical complexity among different age groups and the strict limitations on participants' physiological reserves imposed by most prospective randomized controlled trials, evidence-based medical research targeting elderly patients with CRC is relatively rare. On the basis of current research, this real-world study was conducted to clarify the risk factors for PPOI in elderly patients with CRC, to promote precision and individualized management, and to reduce the incidence of PPOI.

MATERIALS AND METHODS

Study design and patient selection

This was a single-center, retrospective study conducted at Chaoyang Central Hospital of China Medical University. A total of 723 patients who underwent laparoscopic radical resection for CRC between November 2018 and June 2025 were initially enrolled. After the inclusion and exclusion criteria were applied, 167 patients were ultimately included in the final analysis (Figure 1). To safeguard the internal validity of the study, we reviewed the institutional clinical practice guidelines during the study period. The core surgical procedures of laparoscopic radical resection for CRC and perioperative management protocols were conducted by the same surgical team at our center and remained consistent and stable throughout the study period, adhering to standardized clinical guidelines with no substantial modifications.

Figure 1 Flowchart of participants throughout the study. CRC: Colorectal cancer; PPOI: Prolonged postoperative ileus; AJCC: American Joint Committee on Cancer; ICU: Intensive care unit.

Data collection and definitions

Demographic and clinical data collected through the electronic medical records system included sex, age, body mass index, smoking history, American Joint Committee on Cancer (AJCC) stage (8^th edition), tumor marker levels, ASA score, complete peripheral blood cell count, operative time, postoperative opioid use, and other relevant indicators on the first postoperative day (POD). The reference ranges of the relevant indicators in this study were as follows: Serum potassium concentration, 3.5-5.3 mmol/L (flame photometric method); carcinoembryonic antigen, 0-3.40 ng/mL (enzyme-linked immunosorbent assay); serum albumin (ALB) concentration, 35-55 g/L (colorimetry); and serum calcium concentration, 2.11-2.52 mmol/L (flame photometric method). In accordance with the study protocol, additional clinical recovery indicators, such as the presence of nausea or vomiting, oral intake tolerance, passage of flatus, degree of abdominal distension, and relevant imaging findings, were recorded up to 72 to 96 hours post-operatively. The above information, along with functional dependency scores assessed on the day of admission-covering daily activities such as eating, bathing, dressing, toileting, and ambulation-was extracted from admission records and subsequent medical documentation. All surgeries were performed at our hospital by senior surgeons with experience performing laparoscopic radical resection for CRC. Data collection was performed by two independent researchers to ensure quality control. Evaluation and perioperative management were conducted in strict accordance with unified clinical management protocols and documentation standards.

The inclusion criteria: Patients with pathologically confirmed colorectal malignancy, patients aged ≥ 65 years, patients who underwent laparoscopic radical resection for CRC, and patients with AJCC stage I-III disease.

The exclusion criteria: Incomplete medical records, patients who underwent reoperation during the same hospitalization, patients with known inflammatory or hematologic conditions affecting the complete blood count, patients receiving neoadjuvant therapy or using immunosuppressants, patients undergoing emergency surgery for ileus and related conditions, patients with concurrent cancer or CRC recurrence, patients who underwent Miles surgery or preventive stoma, and patients who were transferred to the ICU postoperatively.

PPOI diagnostic criteria: PPOI was defined according to the criteria established by Vather et al[9] in 2013. A diagnosis of PPOI was established when a patient presented with at least two of the following five criteria at 96 hours postoperative or later: (1) Nausea or vomiting; (2) Inability to tolerate oral feedings; (3) Failure to resume flatus; (4) Abdominal distension; and (5) Imaging confirmation.

mFI-5 diagnostic criteria: The mFI-5 was calculated on the basis of the 2018 instrument[12], which comprises five items: Hypertension, diabetes, chronic obstructive pulmonary disease, congestive heart failure, and dependent physical status. This study utilized patient medical history and the assessment of daily living activities conducted on the day of admission. Dependent physical status was determined from the nursing records and specifically assessed five activities of daily living: Eating, bathing, dressing, toileting, and walking. Each mFI-5 item was recorded as “yes” or “no” and scored as 1 or 0 points, respectively. A total score of ≥ 2 indicates frailty[12,16].

Inflammation indicators

This study included assessments of various previously reported inflammatory indicators: The neutrophil-to-lymphocyte ratio (NLR) was calculated as the neutrophil count/Lymphocyte count; the systemic immune inflammation index (SII) was calculated as (platelet count × neutrophil count)/Lymphocyte count; and the pan-immune-inflammation value (PIV) was calculated as (neutrophil count × monocyte count × platelet count)/Lymphocyte count.

Nutritional indicators and calculation formulas

The prognostic nutritional index (PNI) was calculated as 10 × ALB (g/dL) + 0.005 × total lymphocyte count (per mm³).

Statistical analysis

Statistical analysis was performed using STATA 18.0. Missing data for key clinical variables (such as the mFI-5) were handled by complete case analysis to ensure robustness. Categorical variables were reported as n (%), compared using the χ² test. Continuous variables were tested for normality via the Shapiro-Wilk test. Normally distributed data were presented as the mean ± SD and compared using independent samples t tests; nonnormally distributed data were reported as the median (interquartile range) and analyzed using the Mann-Whitney U test. Standardized mean differences were calculated for continuous variables. Univariate logistic regression was performed, and optimal cutoff values for significant indicators were determined using the Youden index. Variables with P < 0.1 in the univariate analysis were entered into the multivariable logistic regression. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated, with P < 0.05 considered to indicate statistical significance. Multicollinearity was assessed using the variance inflation factor (VIF), with a VIF > 5.0 defined as severe collinearity. The events-per-variable principle was applied to limit the number of variables in the final model. Model performance was evaluated using receiver operating characteristic curves, with bootstrap resampling to assess overfitting and stability. Calibration curves and clinical decision curve analysis (DCA) were used to validate the predictive accuracy and clinical utility.

RESULTS

A total of 167 patients who met the study criteria were included, comprising 128 patients with rectal cancer and 39 with colon cancer. The clinicopathological characteristics of the patients are shown in Table 1. PPOI occurred in 49 patients (29.30%), and 71 patients (42.50%) had an mFI-5 ≥ 2.

Table 1 Clinical baseline characteristics and pathological data of patients, n (%).

Baseline characteristics and pathological data	n = 167
Sex
Male	72 (43.1)
Female	95 (56.9)
Age (years)	71.4 ± 6.6
BMI (kg/m2)	23.9 ± 3.3
Smoking history
Yes	110 (65.9)
No	57 (34.1)
Preoperative CEA (ng/mL)	3.78 (2.13-8.88)
Preoperative albumin (g/L)	38.3 (35.9-41.2)
History of abdominal surgery	27 (16.2)
mFI-5 ≥ 2	71.0 (42.5)
Coronary heart disease	26 (15.6)
Operation time (minute)	173.6 ± 53.3
AJCC staging
I	26 (15.6)
II	72 (43.1)
III	69 (41.3)
Tumor location
Rectal cancer	128 (76.7)
Colon cancer	39 (23.4)
Lymphovascular/perineural invasion	48 (28.7)
PPOI	49 (29.3)

BMI: Body mass index; CEA: Carcinoembryonic antigen; mFI-5: Modified 5-item Frailty Index; AJCC: American Joint Committee on Cancer; PPOI: Prolonged postoperative ileus.

No significant differences in patient baseline characteristics were found between the PPOI and non-PPOI groups (Table 2). The univariate analysis in which patients were divided into PPOI and non-PPOI groups revealed statistically significant associations between an mFI-5 ≥ 2 (OR = 3.36; 95%CI: 1.68-6.73; P = 0.001), postoperative opioid use (OR = 2.49; 95%CI: 1.26-4.91; P = 0.009), and a serum potassium concentration on POD 1 < 3.61 mmol/L (OR = 0.14; 95%CI: 0.05-0.41; P < 0.001) and PPOI.

Table 2 Comparison between non-prolonged postoperative ileus and prolonged postoperative ileus groups, n (%).

Variable	Non-PPOI (n = 118)	PPOI (n = 49)	SMD	P value
Sex			-	0.520
Male	49 (41.5)	23 (46.9)
Female	69 (58.5)	26 (53.1)
Age (years)	71.7 ± 5.4	70.9 ± 8.8	0.113	0.504
BMI (kg/m2)	23.8 ± 3.3	24.2 ± 3.2	0.138	0.415
Smoking history	40 (33.9)	17 (34.7)	-	0.921
History of abdominal surgery	18 (15.3)	9 (18.4)	-	0.619
Coronary heart disease	19 (16.1)	7 (14.3)	-	0.768
mFI-5 ≥ 2	40 (33.9)	31 (63.3)	-	< 0.001
Preop CEA (ng/mL)	3.74 (2.09-10.3)	4.33 (2.52-7.42)	0.104	0.699
Preoperative albumin (g/L)	38.35 (36.1-41.2)	38 (35.3-41.1)	0.133	0.684
AJCC staging			-	0.251
I	15 (12.7)	11 (22.5)
II	54 (45.8)	18 (36.7)
III	49 (41.5)	20 (40.8)
ASA score			-	0.080
I	2 (1.7)	1 (2)
II	87 (73.7)	28 (57.1)
III	29 (24.6)	20 (40.8)
Auxiliary incision length (cm)	5 (5-5)	5 (5-5)	0.120	0.401
Operation time (minute)	176 ± 50	168 ± 60.9	0.151	0.374
Tumor location			-	0.167
Rectal cancer	87 (73.7)	41 (83.7)
Colon cancer	31 (26.3)	8 (16.3)
Lymphovascular/perineural invasion	36 (30.5)	12 (24.5)	-	0.434
Postoperative opioid use			-	0.008
Yes	39 (33.1)	27 (55.1)
No	79 (66.9)	22 (44.9)
Postoperative albumin (g/L)	32.7 (31-35.08)	32.06 (30.3-33.96)	0.093	0.129
POD 1 fluid (mL)	3611 (2900-4100)	3660 (2766-4130)	0.026	0.934
POD 2 fluid (mL)	3494 (2880-3930)	3300 (2844-3890)	0.175	0.340
POD 3 fluid (mL)	3465 (2780-4000)	3320 (2800-3936)	0.124	0.566
POD 1 K+ (mmol/L)	4.265 (4-4.5)	4.1 (3.63-4.4)	0.499	0.012
POD 1 Ca2+ (mmol/L)	2.03 (1.96-2.12)	2.01 (1.94-2.11)	0.085	0.402
POD 1 NLR	9.035 (7.6-12.6)	8.4 (6.91-11.38)	0.179	0.141
POD 1 PNI	37.54 (35.1-40.52)	37.55 (35.11-39.9)	0.025	0.965
POD 1 SII	1866.215 (1321.88-2471.95)	1683.44 (1156.09-2238.31)	0.274	0.174
POD 1 PIV	1238.155 (749.32-2038.9)	1071.18 (704.18-1790.47)	0.054	0.551

SMD: Standardized mean differences; BMI: Body mass index; Pre-op CEA: Preoperative carcinoembryonic antigen; mFI-5: Modified 5-item Frailty Index; PPOI: Prolonged postoperative ileus; ASA: American Society of Anesthesiologists; POD 1: Postoperative day 1; NLR: Neutrophil-to-lymphocyte ratio; PNI: Prognostic nutritional index; SII: Systemic immune-inflammation index; PIV: Pan-immune-inflammation value.

The multivariate analysis revealed that an mFI-5 ≥ 2 (OR = 2.31; 95%CI: 1.10-4.86; P = 0.028), postoperative opioid use (OR = 2.39; 95%CI: 1.14-5.01; P = 0.021), and a serum potassium concentration on POD 1 < 3.61 mmol/L (OR = 0.17; 95%CI: 0.05-0.55; P = 0.003) were associated with PPOI. The other indicators were not significant according to the univariate or multivariate analyses, as shown in Table 3. The area under the curve (AUC) analysis demonstrated a predictive performance of 0.730 (95%CI: 0.649-0.811) for PPOI in elderly patients (Figure 2A); the model sensitivity was 0.674, the specificity was 0.636, and the optimal cutoff value was 0.268. The bootstrap-validated AUC was 0.730 (95%CI: 0.644-0.800). This model outperformed single indicators in predicting PPOI in elderly patients, such as the mFI-5 ≥ 2 (AUC = 0.647, 95%CI: 0.554-0.739; Figure 2B), postoperative opioid use (AUC = 0.610, 95%CI: 0.515-0.705; Figure 2C), and serum potassium concentration on POD 1 < 3.61 mmol/L (AUC = 0.601, 95%CI: 0.501-0.701, Figure 2D). Calibration was assessed using the Hosmer-Lemeshow test (P = 0.775) (Figure 3). The DCA was shown in Figure 4.

Figure 2 Receiver operating characteristic curves and area under the curve for predicting prolonged postoperative ileus in elderly patients with colorectal cancer. A: Receiver operating characteristic (ROC) curves and area under the curve (AUC) values on the basis of an modified 5-item Frailty Index (mFI-5) ≥ 2, postoperative opioid use, and low serum potassium on postoperative day 1 (POD 1); B: ROC curves and AUC values of mFI-5 ≥ 2; C: ROC curves and AUC values of postoperative opioid use; D: ROC curves and AUC values of serum potassium concentration on POD 1 < 3.61 mmol/L. AUC: Area under the curve; CI: Confidence interval.

Figure 3 Hosmer-Lemeshow test for model calibration. AUC: Area under the curve.

Figure 4  Decision curve analysis was performed to evaluate the clinical utility of the model.

Table 3 Univariate and multivariate logistic regression analysis of risk factors for prolonged postoperative ileus, n (%).

Variable	Univariable analysis		Multivariable analysis
	OR (95%CI)	P value	OR (95%CI)	P value
Sex
Male	Reference	-
Female	0.80 (0.41-1.57)	0.520
Age (years)	0.98 (0.94-1.03)	0.506
BMI (kg/m2)	1.04 (0.94-1.16)	0.413
Smoking history	1.04 (0.51-2.09)	0.921
History of abdominal surgery	1.25 (0.52-3.01)	0.619
Coronary heart disease	0.87 (0.34-2.22)	0.768
mFI-5 ≥ 2	3.36 (1.68-6.73)	0.001	2.31 (1.10-4.86)	0.028
Preop CEA (ng/mL)	0.99 (0.98-1.01)	0.546
Preoperative albumin (g/L)	1.02 (0.96-1.09)	0.443
AJCC staging
I	Reference	-
II	0.45 (0.18-1.17)	0.101
III	0.56 (0.22-1.42)	0.220
ASA score
I	Reference	-
II	0.64 (0.06-7.37)	0.723
III	1.38 (0.12-16.26)	0.798
Auxiliary incision length (cm)	0.82 (0.48-1.42)	0.484
Operation time (minute)	1.00 (0.99-1.00)	0.373
Tumor location
Rectal cancer	Reference	-
Colon cancer	0.55 (0.23-1.3)	0.171
Lymphovascular/perineural invasion	0.74 (0.35-1.58)	0.435
Postoperative opioid use
Yes	2.49 (1.26-4.91)	0.009	2.39 (1.14-5.01)	0.021
No	Reference	-
Postoperative albumin (g/L)	0.98 (0.9-1.06)	0.584
POD 1 fluid (per mL)	1.00 (0.97-1.04)	0.878
POD 2 fluid (per mL)	0.98 (0.93-1.02)	0.303
POD 3 fluid (per mL)	0.99 (0.96-1.02)	0.501
POD 1 K+ (mmol/L)	0.34 (0.16-0.72)	0.005
≥ 3.61	0.14 (0.05-0.41)	< 0.001	0.17 (0.05-0.55)	0.003
< 3.61	Reference	-
POD 1 Ca2+ (mmol/L)	0.55 (0.05-5.71)	0.614
POD 1 NLR	0.96 (0.90-1.03)	0.293
POD 1 PNI	1.00 (0.93-1.06)	0.880
POD 1 SII (per 100)	0.97 (0.94-1.01)	0.114
POD 1 PIV (per 100)	0.99 (0.96-1.03)	0.749

BMI: Body mass index; Pre-op CEA: Preoperative carcinoembryonic antigen; mFI-5: Modified 5-item Frailty Index; PPOI: Prolonged postoperative ileus; ASA: American Society of Anesthesiologists; POD 1: Postoperative day 1; NLR: Neutrophil-to-lymphocyte ratio; PNI: Prognostic nutritional index; SII: Systemic immune-inflammation index; PIV: Pan-immune-inflammation value; OR: Odds ratio; CI: Confidence interval.

DISCUSSION

Population aging, an inevitable aspect of demographic structural transformation, presents significant challenges to the modern health care system. This demographic shift suggests that the proportion of elderly patients undergoing CRC surgery will continue to increase, potentially leading to a corresponding increase in the incidence of PPOI[3]. PPOI is a major complication that adversely affects patient recovery, and its pathophysiology is multifactorial[25,26].

Numerous studies have confirmed that the ERAS protocol can significantly shorten hospital stay, reduce complication rates, and promote the recovery of intestinal function[27]. However, there is still room for improvement in terms of optimizing intestinal function recovery in elderly and high-risk patients[28]. The PPOI risk prediction model established in this study combines preoperative functional status (such as the mFI-5) in elderly patients with modifiable clinical indicators (postoperative opioid use and hypokalemia), which is consistent with the “individualized management” concept of ERAS. Although ERAS already includes modifiable measures such as preoperative optimization[27], its clinical implementation remains insufficient[29]. Integrating the risk stratification system of this study into the ERAS workflow may improve the pertinence and effectiveness of ERAS implementation and reduce the incidence of PPOI in elderly patients.

ERAS promotes intestinal function recovery through multimodal interventions[27], with a compliance rate of more than 90% in the elderly population and clear benefits[27,29]. In recent years, the association between preoperative frailty and PPOI has attracted increasing attention[16]. Traditional decision-making is mostly based on age, which tends to ignore physiological differences among patients of the same age[3,5]. However, postoperative risk in elderly patients is not determined by age alone[18], and geriatric frailty significantly increases the risk of postoperative intestinal dysfunction. On the basis of real-world data from our center, after adjusting for confounding factors, an mFI-5 ≥ 2 was significantly associated with PPOI risk following laparoscopic radical resection in elderly patients with CRC, whereas age was not independently correlated, which is consistent with previous findings[16,18]. These findings suggest that risk stratification based on preoperative frailty status has greater clinical value.

Preoperative rehabilitation and technical integration are important directions for the future development of ERAS[29]. The functional status assessment in this study can provide a basis for preoperative rehabilitation, allowing the early identification of high-risk patients for inclusion in rehabilitation programs. In summary, under the standardized ERAS framework, this study proposes a precise risk prediction strategy based on functional status and modifiable indicators, providing new insights for optimizing postoperative intestinal function management. Further studies are needed to explore the dynamic integration of the above indicators into the ERAS workflow, promoting the improvement of perioperative management from “standardization” to “precision”.

In this study, a potential association was observed between serum potassium concentrations on POD 1 < 3.61 mmol/L and PPOI. Although electrolyte disturbance is not the core initiating mechanism of PPOI, physiological evidence has demonstrated that potassium ion homeostasis is essential for maintaining the function of gastrointestinal interstitial cells of Cajal and the contractile rhythm of smooth muscles[30], serum potassium concentrations on POD 1 < 3.61 mmol/L may inhibit intestinal myoelectric activity and peristalsis by altering the membrane potential. Existing mechanistic and clinical evidence supports the incorporation of postoperative potassium assessment and optimization into the multimodal management strategy of ERAS[27].

Notably, although several previous studies have confirmed that peripheral blood inflammatory markers (such as the NLR and SII) are associated with the occurrence of PPOI[3,5,31], none of the POD 1 peripheral blood inflammatory markers (NLR, PNI, SII, and PIV) were significantly associated with PPOI in the present study. Specifically, the OR of the POD 1 NLR was 0.96 (95%CI: 0.90-1.03; P = 0.293), and the OR of the SII (per 100 units) was 0.97 (95%CI: 0.94-1.01; P = 0.114). PNI and PIV also showed OR values close to the null value with relatively wide CI. These negative findings may be attributed to the limited sample size, single-time point detection, and insufficient differentiation of early postoperative inflammation. This suggests that given the statistical power of the present study, the above markers are not sufficient to serve as independent predictors of PPOI; moreover, this study’s findings reflect significant heterogeneity in the predictive performance of inflammatory markers across different populations, surgical types, and detection time points.

The multivariate analysis in the present study demonstrated a significant association between postoperative opioid use and an increased risk of PPOI (P = 0.021), which is consistent with known pharmacological mechanisms and findings from previous studies[27]. However, these results should be interpreted with caution: Severe pain itself impairs gastrointestinal function, and patients receiving opioid therapy often experience more intense pain. Therefore, the observed association may partially reflect pain severity rather than a direct causal effect of opioid use. Al-Jasim et al[32] investigated the relationship between postoperative analgesia and PPOI and highlighted this confounding effect. Their study linked opioid use to an increased risk of obstruction, whereas improved pain control was associated with a lower obstruction rate, suggesting that pain may contribute to excessive opioid use and a heightened risk of obstruction.

In our study, preoperative frailty and serum potassium levels were adjusted for in the multivariate analysis, and the association between postoperative opioid analgesic use and PPOI remained significant. Nevertheless, since postoperative pain was not included in the analysis, the potential confounding effect of pain cannot be completely ruled out. Future studies should incorporate postoperative pain scores to clarify the independent effects of pain and opioid use and to explore optimized analgesic regimens with improved efficacy and lower PPOI risk.

In this study, a predictive model for PPOI was established by integrating three indicators: An mFI-5 ≥ 2, postoperative opioid use, and a serum potassium concentration on POD 1 < 3.61 mmol/L. The model achieved an AUC of 0.730, which was superior to that of any single indicator (Figure 2), and showed a marked improvement in predictive performance compared with that of previous reports[16,18]. Internal validation using bootstrapping yielded an AUC of 0.730 (95%CI: 0.644-0.800), confirming favorable reproducibility.

These findings suggest that the combination of multidimensional clinical features can significantly improve the identification of PPOI risk and enhance the utility of the mFI-5 in perioperative assessments. Compared with conventional approaches, the joint evaluation of global functional status (such as the mFI-5, postoperative serum potassium concentration and postoperative opioid use) allows the construction of a more comprehensive and accurate risk stratification system for PPOI, thereby providing a basis for individualized and precise perioperative management. Given that the relevant indicators are readily available at the bedside, we believe that this model holds promise as a bedside scoring system. For elderly patients, timely monitoring and correction of preoperative frailty and postoperative electrolyte imbalance, as well as optimized analgesic regimens, may effectively promote postoperative gastrointestinal functional recovery and reduce the risk of PPOI.

Certainly, the present study has several limitations. As a single-center retrospective study conducted in a single-ethnicity population, it has inherent confounding bias, a limited sample size and lacks external validation, which limits the generalizability of the predictive model. Further prospective and multicenter studies must be conducted in the future to control confounding effects by using methods such as propensity score matching to improve the applicability and stability of the model.

CONCLUSION

This retrospective study confirmed that there is a high incidence of PPOI in elderly patients following laparoscopic radical resection for CRC; furthermore, an mFI-5 ≥ 2, postoperative opioid use, and low serum potassium concentrations on POD 1 are independent risk factors. The predictive model constructed based on these indicators demonstrated moderate discrimination. Comprehensive perioperative evaluation and management protocols are necessary to reduce the incidence of PPOI and improve postoperative recovery.

ACKNOWLEDGEMENTS

We express our sincere gratitude to the Guo Shi-Gang Model Worker Innovation Studio for their invaluable guidance and assistance.