Published online Jul 15, 2026. doi: 10.4251/wjgo.120582
Revised: March 10, 2026
Accepted: March 31, 2026
Published online: July 15, 2026
Processing time: 133 Days and 3.1 Hours
The study of prolonged postoperative ileus (PPOI) after laparoscopic colorectal cancer (CRC) surgery is a clinically significant concern, but there is little research on predicting gastrointestinal function of CRC patients through the characteristics of bowel sounds.
To analyze differences in bowel sound characteristics in patients with PPOI after surgery, and aims to establish a predictive model to provide clinicians with a new method for evaluating postoperative gastrointestinal function.
A retrospective analysis was conducted on 133 patients diagnosed with CRC who underwent surgical treatment in the Department of General Surgery II of Shaanxi Provincial People’s Hospital from January 2022 to January 2024. This study analyzes the characteristics of bowel sounds in PPOI patient pre-operation 1 day, on operation day, and post-operation 3 days, clarifying their differences and trends. The Mann-Whitney U test, Kolmogorov-Smirnov test, and receiver operating characteristic (ROC) curve analysis were used to examine the relationship between clinical indicators and bowel sound characteristics with postoperative PPOI. Univariate and multifactorial analyses were performed to clarify the differences between the PPOI and no-PPOI groups. Subsequently, significant variables were selected and incorporated into the model for further modeling.
The analysis found that patients with PPOI had significant differences in number of bowel sounds (NBS) and recovery time of bowel sounds (RTBS) on the post-operation 1 day. The characteristics of bowel sounds predicted the occurrence of postoperative PPOI with certain predictive value according to the ROC curve. The NBS cutoff value was 1.201 counts per minute, with a sensitivity of 56.67% and specificity of 80.58%. The RTBS cutoff value was 16.9 hours, with a sensitivity of 90.00% and specificity of 43.75%. Univariate analysis revealed significant differences in operation time, preoperative hypoproteinemia, RTBS, and NBS between the PPOI group and the no-PPOI group. The LASSO regression and the Boruta algorithm were used in conjunction with univariate and multivariate logistic regression to screen for relevant variables, ultimately including four variables in the model: Operation time, preoperative hypoproteinemia, RTBS, and NBS. Decision curve analysis indicated that the risk nomogram for PPOI after CRC surgery provides a good clinical net benefit.
The characteristics of bowel sounds have certain predictive value for PPOI after laparoscopic CRC surgery. The intelligent auscultation system collects bowel sounds, which helps establish a predictive model for the occurrence of PPOI. It provides objective reference indicators for its early detection and warrants further investigation.
Core Tip: This study constitutes the first application of an intelligent auscultation system for extended monitoring of gastrointestinal function in colorectal cancer patients following surgery, enabling prediction of gastrointestinal functional outcomes through acoustic feature analysis.
- Citation: Shi S, Wang C, Zhao CS, Chen Z, Yan L, Liang Y, Yue X, Duan XL, Wang ZZ. Clinical study predicting prolonged postoperative ileus after laparoscopic colorectal cancer surgery with intelligent bowel sound auscultation system. World J Gastrointest Oncol 2026; 18(7): 120582
- URL: https://www.wjgnet.com/1948-5204/full/v18/i7/120582.htm
- DOI: https://dx.doi.org/10.4251/wjgo.120582
Colorectal cancer (CRC) is a highly prevalent gastrointestinal cancer worldwide[1]. Despite significant advancements in modern medicine regarding its diagnosis and treatment, surgery remains the primary treatment method[2,3]. With the development of enhanced recovery, postoperative recovery has increasingly attracted the attention of clinicians[4]. Prolonged postoperative ileus (PPOI) is a common complication after surgery, primarily manifested as slow or stagnant recovery of gastrointestinal function in patients postoperatively, presenting as severe abdominal distension, constipation, and inability to eat[5,6]. This not only affects postoperative recovery but also significantly prolongs hospital stay, increases the incidence of postoperative complications, and leads to rising medical costs[7]. Compared to postoperative ileus (POI) that can resolve on its own in a short period, PPOI often lasts longer and is more difficult to recover from, becoming a challenge in postoperative management[8,9].
The etiology of PPOI is extremely complex, and is not limited to a single factor. Various factors may trigger PPOI, including the stress response triggered by surgical trauma, excessive activation of the sympathetic nervous system, systemic inflammatory response, improper management of the surgical site, and electrolyte imbalance[10]. Additionally, patient-specific constitutional factors, surgical methods, use of anesthetic drugs, and improper postoperative manage
Currently, clinical practice mainly relies on gas passage and defecation as diagnostic criteria for PPOI, but this method often lags behind the actual occurrence of gastrointestinal dysfunction, making it difficult to promptly reflect the postoperative recovery status[12,13]. Based on this, monitoring bowel sounds as a non-invasive method has gradually entered the field of gastrointestinal function assessment[12]. Bowel sounds are the noises produced during intestinal peristalsis, which can reflect the activity level of the intestines[14]. By continuously recording the changes in bowel sounds of patients after surgery over an extended period and studying their association with PPOI, new ideas and tools for early prediction may be provided. Therefore, this study aims to continuously record the bowel sounds of CRC patients before and after surgery. The purpose of this study is to continuously record bowel sounds in CRC patients before and after surgery using an intelligent auscultation system, to investigate the differences in bowel sound characteristics between the PPOI group and the no-PPOI group, and to explore the application value of bowel sound indicators in predicting PPOI.
This study retrospectively collected information on 133 patients diagnosed with CRC who underwent surgical treatment in the Department of General Surgery II of Shaanxi Provincial People’s Hospital from January 2022 to January 2024. Inclusion criteria were as follows: (1) Preoperative diagnostic tests confirmed CRC; (2) No obvious contraindications to surgery and indications for laparoscopic CRC surgery; (3) Patients who underwent bowel sound monitoring both preoperatively and postoperatively with complete monitoring data; and (4) Patients with complete clinical data and postoperative follow-up information. Exclusion criteria included: (1) Diagnostic tests indicating non-CRC or the presence of other tumors; (2) Patients in poor general condition who could not undergo surgical treatment or who did not receive surgical treatment; (3) Patients who did not undergo bowel sound monitoring or whose monitoring data were incom
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Medical Ethics Committee of Shaanxi Provincial People’s Hospital.
This study uses a continuous auscultation recorder (model: YM-TYJL-01) jointly developed by Tsinghua University and Beijing Yimai Medical Technology Co., Ltd., designed for continuous monitoring of bowel sounds. The intelligent auscultation system collects the patient's bowel sounds through a highly sensitive sensing device attached firmly to the patient’s abdomen. It then performs efficient noise reduction processing via a collection patch processor. The audio data is then transmitted to the receiving processor via wireless transmission. The receiving processor sends the data to the server in the doctor’s office, where dedicated software installed on the computer processes the raw data. The bowel sound collector is placed on the patient’s right lower abdomen and automatically records bowel sounds for 5 minutes every 30 minutes, continuously over 24 hours each day.
Bowel sound recognition is achieved through mel-frequency cepstral coefficients (MFCC) feature extraction. This is followed by building a CRNN neural network composed of 5 convolutional layers, a bidirectional GRU, and a fully connected layer. A large amount of clinically collected bowel sound data was obtained, and then the audio data was manually annotated by clinicians, segmenting the sounds into standardized clips for training. Afterward, a new convolutional neural network design method was employed, adapting popular CNN modules from the field of image recognition to bowel sound segmentation. Subsequently, MFCC features were extracted from these sound recordings, and the neural network was trained to achieve autonomous recognition of bowel sounds[15] (Figure 2).
Bowel sound-related indicators include the following[15]: (1) Number of bowel sounds (NBS): The NBS per minute, measured in counts per minute (cpm); (2) Bowel sound vibration amplitude (BSVA): The intensity of the bowel sound in the time or frequency domain, measured in decibels; (3) Frequency of bowel sounds (FBS): The significant frequency component of the bowel sound, representing the frequency band where the bowel sound energy is concentrated, measured in Hz; and (4) Recovery time of bowel sounds (RTBS): Defined as the time interval between the end of surgery and the first postoperative occurrence of bowel sounds, which is identified as the first instance when the NBS exceeds 1 cpm (Figure 3).
Judgment is made by two associate chief physicians based on the following criteria: 96 hours post-operation or later, with at least two of the following conditions met[12,13]: (1) Nausea and vomiting have occurred in the past 12 hours. Nausea is assessed using a 10-point scale, where 1-3 points indicate mild, 4-7 points moderate, and 8-10 points severe nausea; a score greater than 4 points meets this criterion; (2) The patient has been unable to tolerate solid food during the last two meals. A self-reported food intake of less than 25% is considered meeting this criterion; (3) No passage of gas or stool has occurred in the past 24 hours. For patients with an ileostomy or colostomy, no gas or stool in the ostomy bag meets this criterion; (4) Moderate to severe abdominal distension is present, clinically assessed by the physician through percussion; such distension meets this criterion; and (5) Imaging examination (abdominal X-ray or computed tomography) performed within the past 24 hours confirms ileus, defined by the presence of at least two of the following findings: Gastric dis
All patients undergo relevant preoperative examinations and receive nutritional support before surgery. They are instructed to stop smoking and to perform respiratory function exercises prior to the procedure. Fasting is required before surgery, and bowel preparation is carried out according to standard protocols. General anesthesia is selected, and the patient’s vital signs are monitored during the operation, with goal-directed fluid management. A gastric tube, abdominal drainage tube, and urinary catheter are usually placed postoperatively. Patients are encouraged to get up and move around on the first day after surgery to prevent complications such as deep vein thrombosis. The gastric tube is removed once the patient passes gas. They are then encouraged to try drinking water and gradually progress to a liquid diet, provided there is no abdominal discomfort, followed by slowly increasing solid food intake.
We collected patients’ general clinical data, including gender, age, body mass index, diabetes, hypertension, heart disease, smoking and drinking habits, histories of abdominal surgeries, tumor location, cTNM stage, and preoperative obs
According to clinical physicians, patients are divided into the PPOI group and the no-PPOI group based on diagnostic indicators. Clinical data and bowel sounds characteristics on the pre-operation 1 day, operation day, and post-operation 3 days are then analyzed. Additionally, the overall trend of bowel sounds characteristics monitored over an extended period (e.g., several days) is analyzed. A total of 97 patients without intestinal obstruction on the pre-operation 1 day are included in the control group and used to analyze the differences in bowel sounds indicators among the PPOI group, no-PPOI group, and the control group on the post-operation 1 day. Using univariate and multivariate analysis, the diffe
Statistical analysis was conducted using R version 4.3.2, and data analysis was performed using GraphPad Prism software. Categorical data were presented as frequencies or percentages, and between-group comparisons were made using χ2 tests or Fisher’s exact tests, as appropriate. Normally distributed continuous data were reported as mean ± SD and compared between groups using independent samples t-tests, while skewed continuous data were presented as median (interquartile range). To identify relevant predictors, we applied LASSO regression and the Boruta algorithm, followed by univariate and multivariate logistic regression analyses to determine independent risk factors for PPOI in CRC. Subsequently, we developed a predictive model using logistic regression. The model’s performance was assessed using receiver operating characteristic (ROC) curves and calibration curves, and its clinical utility was evaluated through decision curve analysis (DCA). Finally, the model was visualized as a nomogram. A significance level of P < 0.05 was considered statistically significant.
This study included 133 patients with CRC who underwent surgical treatment, including 61 cases of colon cancer and 72 cases of rectal cancer. Of these, 74 patients (55.64%) were male and 59 patients (44.36%) were female. Preoperatively, 36 patients (27.07%) had ileus. Among these, 30 (22.5%) developed PPOI (Table 1).
| Variables | Total (n = 133) | 0 (n = 103) | 1 (n = 30) | Statistic | P value |
| Age (year) | 64.00 (56.00, 71.00) | 64.00 (56.00, 70.00) | 65.50 (61.25, 73.75) | Z = -0.81 | 0.419 |
| Gender | χ2 = 0.02 | 0.898 | |||
| Male | 74 (55.64) | 57 (55.34) | 17 (56.67) | ||
| Female | 59 (44.36) | 46 (44.66) | 13 (43.33) | ||
| BMI (kg/m2) | 23.18 (20.81, 24.91) | 23.18 (20.90, 25.04) | 22.75 (19.81, 24.38) | Z = -1.27 | 0.204 |
| Diabetes mellitus | χ2 = 0.74 | 0.391 | |||
| No | 111 (83.46) | 88 (85.44) | 23 (76.67) | ||
| Yes | 22 (16.54) | 15 (14.56) | 7 (23.33) | ||
| Hypertension | χ2 = 0.41 | 0.521 | |||
| No | 82 (61.65) | 62 (60.19) | 20 (66.67) | ||
| Yes | 51 (38.35) | 41 (39.81) | 10 (33.33) | ||
| Coronary disease | χ2 = 1.60 | 0.206 | |||
| No | 124 (93.23) | 94 (91.26) | 30 (100.00) | ||
| Yes | 9 (6.77) | 9 (8.74) | 0 (0.00) | ||
| Smoking | χ2 = 0.98 | 0.321 | |||
| No | 98 (73.68) | 78 (75.73) | 20 (66.67) | ||
| Yes | 35 (26.32) | 25 (24.27) | 10 (33.33) | ||
| Drinking | χ2 = 1.87 | 0.172 | |||
| No | 105 (78.95) | 84 (81.55) | 21 (70.00) | ||
| Yes | 28 (21.05) | 19 (18.45) | 9 (30.00) | ||
| History of abdominal operation | χ2 = 0.97 | 0.325 | |||
| No | 102 (76.69) | 81 (78.64) | 21 (70.00) | ||
| Yes | 31 (23.31) | 22 (21.36) | 9 (30.00) | ||
| Accompanied by incomplete intestinal obstruction | χ2 = 0.00 | 0.955 | |||
| No | 97 (72.93) | 75 (72.82) | 22 (73.33) | ||
| Yes | 36 (27.07) | 28 (27.18) | 8 (26.67) | ||
| Stage cTNM | - | 0.953 | |||
| 0 | 2 (1.50) | 2 (1.94) | 0 (0.00) | ||
| I | 20 (15.04) | 15 (14.56) | 5 (16.67) | ||
| II | 52 (39.10) | 41 (39.81) | 11 (36.67) | ||
| III | 49 (36.84) | 38 (36.89) | 11 (36.67) | ||
| IV | 10 (7.52) | 7 (6.80) | 3 (10.00) | ||
| Combined resection | χ2 = 0.01 | 0.942 | |||
| No | 126 (94.74) | 97 (94.17) | 29 (96.67) | ||
| Yes | 7 (5.26) | 6 (5.83) | 1 (3.33) | ||
| Surgical scope | χ2 = 3.28 | 0.194 | |||
| Right hemicolon | 38 (28.57) | 29 (28.16) | 9 (30.00) | ||
| Left hemicolon | 34 (25.56) | 30 (29.13) | 4 (13.33) | ||
| Rectum | 61 (45.86) | 44 (42.72) | 17 (56.67) | ||
| Reinforced suture of the anastomosis | χ2 = 1.74 | 0.188 | |||
| No | 57 (42.86) | 41 (39.81) | 16 (53.33) | ||
| Yes | 76 (57.14) | 62 (60.19) | 14 (46.67) | ||
| Enterostomy | χ2 = 1.51 | 0.219 | |||
| No | 100 (75.19) | 80 (77.67) | 20 (66.67) | ||
| Yes | 33 (24.81) | 23 (22.33) | 10 (33.33) | ||
| Anemia | χ2 = 0.16 | 0.690 | |||
| No | 77 (64.71) | 58 (63.74) | 19 (67.86) | ||
| Yes | 42 (35.29) | 33 (36.26) | 9 (32.14) | ||
| Hypoproteinemia | χ2 = 6.68 | 0.010 | |||
| No | 92 (69.17) | 77 (74.76) | 15 (50.00) | ||
| Yes | 41 (30.83) | 26 (25.24) | 15 (50.00) | ||
| CEA | χ2 = 0.85 | 0.357 | |||
| Normal | 79 (59.40) | 59 (57.28) | 20 (66.67) | ||
| Abnormal | 54 (40.60) | 44 (42.72) | 10 (33.33) | ||
| CA19-9 | χ2 = 1.38 | 0.240 | |||
| Abnormal | 100 (75.19) | 75 (72.82) | 25 (83.33) | ||
| Normal | 33 (24.81) | 28 (27.18) | 5 (16.67) | ||
| Potassium in blood | - | 0.493 | |||
| Hypokalemia | 5 (4.20) | 5 (5.49) | 0 (0.00) | ||
| Normokalemia | 113 (94.96) | 85 (93.41) | 28 (100.00) | ||
| Hyperkalemia | 1 (0.84) | 1 (1.10) | 0 (0.00) | ||
| Operation time (minute) | 235.00 (185.00, 300.00) | 230.00 (165.00, 295.00) | 272.50 (231.25, 365.00) | Z = -3.36 | < 0.001 |
| Anesthesia time (minute) | 295.00 (225.00, 350.00) | 295.00 (222.50, 350.00) | 297.50 (232.50, 348.75) | Z = -0.76 | 0.446 |
| NBS (cpm) | χ2 = 13.54 | < 0.001 | |||
| ≤ 1.201 | 36 (27.07) | 20 (19.42) | 16 (53.33) | ||
| > 1.201 | 97 (72.93) | 83 (80.58) | 14 (46.67) | ||
| RTBS (hour) | χ2 = 10.88 | < 0.001 | |||
| ≤ 16.90 | 47 (35.34) | 44 (42.72) | 3 (10.00) | ||
| > 16.90 | 86 (64.66) | 59 (57.28) | 27 (90.00) |
This study found that by examining bowel sound characteristics before surgery, on the operation day, and three days after surgery, there were no significant differences between PPOI patients and no-PPOI patients before surgery and pre-operation one day. However, on the first post-operation day, RTBS and NBS in PPOI patients were significantly different compared to those in no-PPOI patients (P < 0.05). Moreover, on the post-operation 2 day, the BSVA difference between the two groups was statistically significant. Analysis of PPOI patients, no-PPOI patients, and the control group showed significant differences in the NBS in PPOI patients compared to both no-PPOI patients and controls; however, there were no significant differences in FBS among the three groups. Further analysis of bowel sound characteristics before surgery, on the operation day, and three days postoperatively showed that the PPOI group’s NBS, which was higher than that of the no-PPOI group before surgery, gradually returned to a level comparable to the no-PPOI group three days after surgery. Although no significant differences were found in BSVA between the two groups before and after surgery, both groups exhibited low BSVA on the first postoperative day. Notably, the PPOI group showed a decrease in FBS compared to the no-PPOI group on the post-operation 1 day, indicating impaired bowel motility (Table 2 and Figures 4 and 5).
| Variables | Total (n = 133) | No-PPOI (n = 103) | PPOI (n = 30) | Z value1 | P value |
| Pre-operation 1 day NBS | 7.05 (4.29, 9.59) | 6.86 (4.28, 9.45) | 7.62 (5.14, 9.86) | -0.82 | 0.415 |
| Pre-operation 1 day BSVA | 49.92 (47.66, 52.33) | 50.08 (47.87, 52.37) | 49.20 (47.09, 51.82) | -0.75 | 0.453 |
| Pre-operation 1 day FBS | 407.56 (351.78, 452.70) | 415.95 (352.99, 452.76) | 375.37 (321.35, 439.87) | -1.43 | 0.151 |
| Operation day NBS | 2.41 (1.30, 3.26) | 2.38 (1.28, 3.23) | 2.48 (1.58, 3.35) | -0.40 | 0.688 |
| Operation day BSVA | 43.44 (41.95, 45.71) | 43.68 (42.12, 46.21) | 42.77 (41.76, 44.60) | -1.46 | 0.145 |
| Operation day FBS | 399.59 (351.03, 461.91) | 399.59 (350.22, 462.89) | 399.53 (356.64, 451.15) | -0.37 | 0.712 |
| RTBS | 19.70 (15.00, 25.30) | 18.10 (14.55, 24.95) | 23.70 (18.23, 27.10) | -2.33 | 0.020 |
| Post-operation 1 day NBS | 2.46 (1.18, 3.61) | 2.85 (1.42, 3.85) | 1.19 (0.69, 2.53) | -3.81 | <0.001 |
| Post-operation 1 day BSVA | 5.93 (3.42, 8.24) | 6.16 (3.91, 8.39) | 5.18 (3.05, 6.94) | -1.27 | 0.205 |
| Post-operation 1 day FBS | 406.38 (364.14, 451.16) | 406.38 (363.79, 449.21) | 403.31 (368.13, 460.78) | -0.57 | 0.566 |
| Post-operation 2 day NBS | 3.73 (2.26, 5.14) | 3.79 (2.71, 5.15) | 3.24 (1.38, 4.83) | -1.73 | 0.083 |
| Post-operation 2 day BSVA | 49.63 (47.85, 51.37) | 49.93 (48.29, 51.53) | 48.71 (46.50, 50.51) | -2.44 | 0.015 |
| Post-operation 2 day FBS | 411.17 (371.22, 456.93) | 409.40 (363.93, 456.14) | 413.56 (389.37, 462.64) | -1.40 | 0.162 |
| Post-operation 3 days NBS | 5.87 (3.60, 7.96) | 5.81 (3.51, 8.02) | 6.05 (4.43, 7.65) | -0.38 | 0.700 |
| Post-operation 3 days BSVA | 44.75 (42.23, 46.46) | 44.75 (42.12, 46.47) | 44.68 (42.53, 46.26) | -0.19 | 0.853 |
| Post-operation 3 days FBS | 381.25 (342.42, 448.98) | 391.13 (345.84, 456.54) | 361.53 (328.83, 424.37) | -1.56 | 0.118 |
ROC analysis found that NBS can predict postoperative PPOI, with a cutoff value of 1.201 cpm, while RTBS can also predict postoperative PPOI, with a cutoff value of 16.90 hours. Additionally, surgery duration can predict postoperative PPOI, with a cutoff value of 205 minutes. Univariate analysis revealed that NBS, RTBS, operation time, and hypoproteinemia are factors influencing the occurrence of PPOI after surgery. Multivariate analysis confirmed that these factors are independent risk factors for developing PPOI (Figure 6 and Table 3).
| Variables | β | SE | Z value | P value | OR (95%CI) |
| NBS (> 1.201 cpm/≤ 1.201 cpm) | -1.03 | 0.50 | -2.06 | 0.039 | 0.36 (0.13-0.95) |
| RTBS (> 16.90 hours/≤ 16.90 hours) | 1.85 | 0.69 | 2.68 | 0.007 | 6.39 (1.65-24.73) |
| Operation time (> 205 minutes/≤ 205 minutes) | 1.65 | 0.73 | 2.26 | 0.024 | 5.21 (1.25-21.74) |
| Hypoproteinemia (no/yes) | 1.37 | 0.53 | 2.60 | 0.009 | 3.92 (1.40-10.98) |
Based on the combined results of LASSO and Boruta methods, and informed by clinical experience, a prediction model for postoperative PPOI in CRC was constructed using four indicators: NBS, RTBS, surgical time, and hypoproteinemia. The ROC curve for the model’s performance shows an area under the curve (AUC) value of 0.816 (95%CI: 0.738-0.894), indicating good discriminative ability in predicting postoperative PPOI in CRC patients. Internal validation was conducted through 500 random samplings using the bootstrap method. The AUC was used to assess the predictive efficacy of the model, with a value of 0.816 (95%CI: 0.733-0.891), indicating good predictive performance and clinical practicality. In addition, the calibration curve shows that the predicted probabilities of the model agree well with the actual incidence rates, and model calibration is further improved after bias correction. In the DCA, the model shows a substantial net benefit when the high-risk threshold is between 0.05 and 0.3, indicating that the model helps optimize decision-making in clinical practice and provides effective support for postoperative patient management. Overall, the model has good discriminative efficacy and calibration performance and can be used for clinical prediction of postoperative PPOI risk (Figures 7 and 8).
This study used an intelligent auscultation system to perform real-time monitoring of bowel sounds in the human body. By collecting a large amount of clinical bowel sound data and extracting MFCC features, the study trained a neural network system for recognition. The test results showed that under the same recording environment, the sensitivity of this method reached 90.92%, with an overall accuracy of 92.56%. Even in different environments with significant noise interference, its specificity was maintained at 96.2%, with an overall accuracy of 94.2%[15]. Therefore, it is being applied in clinical practice for further research.
Borborygmi is the sound produced when the intestines contract to push liquids and gases through different parts of the gastrointestinal tract during the digestive process[16,17]. The occurrence of borborygmi is significantly related to the migrating motor complex of the gastrointestinal tract[18]. Previous studies have reported a significant association between characteristics of bowel sounds and various gastrointestinal diseases, including irritable bowel syndrome, POI, and delayed gastric emptying, and have demonstrated high sensitivity and specificity for certain diseases[19,20]. This study predicts postoperative PPOI occurrence in patients using ROC curve analysis, demonstrating that both RTBS and NBS have predictive value.
Yoshino et al[21] collected and analyzed bowel sounds in 21 patients with mechanical intestinal obstruction and found that the frequency characteristics of bowel sounds (high-frequency sounds above 900 Hz) were significantly correlated with the severity of the patients, and patients with high-frequency sounds had a higher likelihood of surgery. This indicates that analyzing bowel sound frequencies can assist clinicians in assessing the severity of intestinal obstruction. Dalle et al[22] recorded and analyzed bowel sounds in 67 subjects and found that the FBS in mechanical intestinal obstruction differed from that in acute diffuse peritonitis, which could help differentiate between the two diseases. The aforementioned research on bowel sounds further demonstrates the differences in bowel sound characteristics among different patients and the feasibility of their clinical application. This study did not find a significant difference in FBS on the post-operation 1 day between the PPOI group and the no-PPOI group. A possible explanation is that previous studies mainly focused on patients with mechanical intestinal obstruction, whose sound characteristics are pronounced, making their results more significant. In contrast, patients in this study all received general anesthesia, and bowel sounds on the post-operation 1 day were weak, resulting in no significant difference in FBS. Kaneshiro et al[23] monitored 28 patients undergoing abdominal surgery for 60 minutes before surgery and continuously after surgery, finding that the bowel rate on the first and second postoperative days in PPOI patients was significantly lower than that in patients without PPOI. This study constructed a gastrointestinal function algorithm for predicting PPOI, with sensitivity, specificity, and negative predictive values of 63%, 72%, and 81%, respectively. Spiegel et al[24] found in a study on postoperative gast
Postoperative PPOI following colorectal surgery is a common gastrointestinal complication after abdominal surgery[25]. Previous studies have found that surgical stimulation affects the release of local and systemic inflammatory mediators in the intestine, which leads to impaired function of intestinal nerves and smooth muscles, thereby inhibiting intestinal motility[26,27]. In gastrointestinal surgery, the removal of lesions and intestinal reconstruction during the operation can significantly affect the recovery of gastrointestinal function due to damage to the enteric nervous system. Additionally, anesthetics and analgesics (such as opioids) used during and after surgery primarily inhibit intestinal activity via central and peripheral mechanisms[28-30]. Previous studies have found that the occurrence of postoperative PPOI in CRC is associated with factors such as patient age, operation time, preoperative hypoproteinemia, opioid analgesic use, and hypokalemia, which further corroborate some results of this study[11,31,32]. The reason for not including postoperative related indicators in this study is that it is considered that various indicators and the use of related drugs interact in various ways, ultimately affecting gastrointestinal motility, while bowel sounds are directly related to gastrointestinal functional status. Therefore, only bowel sounds were included in the study to avoid interference from postoperative variables and to further clarify the reliability of the prediction of PPOI based on the single characteristic of bowel sounds.
This study incorporates significant results into the model through multifactorial analysis and conducts an evaluation of the model’s efficacy. The ROC curve analysis indicates that the AUC value of the model is 0.816. After internal validation using self-sampling, the results indicate that the model has a certain predictive efficacy. The calibration curve shows that the predicted probabilities of the model closely correspond to the actual incidence rates. After bias correction, the calibration effect of the model is further improved. In the DCA curve analysis, the model shows a large net benefit when the high-risk threshold is between 0.05 and 0.3, indicating that the model is helpful in optimizing decision-making in actual clinical applications and can provide certain support for postoperative management of patients.
The advantage of this study lies in the use of an intelligent auscultation system for real-time monitoring of bowel sounds in patients. This system incorporates bowel sound characteristics into the prediction model for PPOI in CRC and investigates the predictive value of these characteristics along with clinical indicators for postoperative gastrointestinal function. The study found that on the post-operation 1 day, NBS = 1.201 cpm and RTBS = 16.9 hours represent important thresholds. Clinicians should closely monitor patients whose NBS and RTBS fall below these critical values on the first postoperative day and consider implementing appropriate measures to promote gastrointestinal function recovery. The selection of bowel sound characteristics on the post-operation 1 day is partly based on this study’s analysis indicating significant differences on that day. These early postoperative results offer valuable guidance for future clinicians in early intervention for patients with gastrointestinal function abnormalities. In future research, first, we will conduct multi-center prospective observational studies and perform external validation to enhance the completeness of the evidence chain, thereby improving the model’s generalizability. Second, we will implement early intervention measures for PPOI patients and monitor bowel sounds to assess gastrointestinal function recovery after intervention. Third, from a technological perspective, we will collaborate further with acoustic experts to identify clinical bowel sound characteristics to include, incorporating additional sound indicators into the model to improve its accuracy and discriminative ability.
However, this study has certain limitations. First, as a retrospective study, the long recall period of clinical data may lead to some inaccuracies. In addition, the inclusion of clinical research indicators may not be comprehensive enough. Second, the sample size of this study is relatively small, and the number of outcome variables is limited. These factors may cause overfitting of the models and reduce the generalizability of the results. Third, the bowel sounds collector is not widely used in clinical applications, and its reliability across various environments requires further validation and enhancement. Fourth, this study included relatively few sound features, which limits the predictive accuracy of the model. Fifth, this model is only a preliminary attempt at establishing the system. The layout of the intelligent auscultation system requires significant expenditure, which limits the ability to conduct multicenter research in a short timeframe. This constraint prevents external validation and subsequently affects the assessment of model stability and generalizability.
In summary, bowel sound characteristics have predictive value for PPOI following laparoscopic CRC surgery. Data collected by the intelligent auscultation system helps establish a predictive model for the occurrence of postoperative PPOI, providing objective reference indicators for its early detection, which warrants further investigation.
This study was completed in the Department of General Surgery II, Shaanxi Provincial People’s Hospital. We are grateful to all staff members for their assistance in the experiment.
| 1. | Luo T, Li J, Pu K, Yang G. Association between periodontitis and gastrointestinal cancer risk and prognosis: evidence from a nested case-control study in Southwest China. Eur J Med Res. 2025;30:225. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 2. | Kalogera E, Dowdy SC. Enhanced Recovery After Surgery and Acute Postoperative Pain Management. Clin Obstet Gynecol. 2019;62:656-665. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 5] [Cited by in RCA: 8] [Article Influence: 1.1] [Reference Citation Analysis (0)] |
| 3. | Patterson JM, Freeman M, Hickman LC. Postoperative Enhanced Recovery After Surgery Pathways in Gynecologic Surgery. Clin Obstet Gynecol. 2025;68:501-507. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3] [Cited by in RCA: 3] [Article Influence: 3.0] [Reference Citation Analysis (0)] |
| 4. | Huang Y, Wang X, Li J, He Q, Wang R. Knowledge, attitude, and practice towards enhanced recovery after surgery among patients underwent thoracoscopy surgery. PLoS One. 2024;19:e0309821. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 4] [Reference Citation Analysis (0)] |
| 5. | Wacha H. [Postoperative mechanical ileus]. Langenbecks Arch Chir Suppl Kongressbd. 1991;154-160. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 1] [Reference Citation Analysis (0)] |
| 6. | Wolthuis AM, Bislenghi G, Lambrecht M, Fieuws S, de Buck van Overstraeten A, Boeckxstaens G, D'Hoore A. Preoperative risk factors for prolonged postoperative ileus after colorectal resection. Int J Colorectal Dis. 2017;32:883-890. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 21] [Cited by in RCA: 33] [Article Influence: 3.7] [Reference Citation Analysis (0)] |
| 7. | Zhao L, Guo Y, Zhou X, Mao W, Zhu H, Chen L, Liu X, Zhang L, Xie Y, Li L. The research progress of perioperative non-pharmacological interventions on postoperative cognitive dysfunction: a narrative review. Front Neurol. 2024;15:1369821. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 8] [Reference Citation Analysis (0)] |
| 8. | Weledji EP. Perspectives on paralytic ileus. Acute Med Surg. 2020;7:e573. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 41] [Cited by in RCA: 33] [Article Influence: 5.5] [Reference Citation Analysis (0)] |
| 9. | Lin Z, Li Y, Wu J, Zheng H, Yang C. Nomogram for prediction of prolonged postoperative ileus after colorectal resection. BMC Cancer. 2022;22:1273. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 23] [Reference Citation Analysis (0)] |
| 10. | de Gaay Fortman DPE, Kroon HM, Bedrikovetski S, Fitzsimmons TR, Dudi-Venkata NN, Sammour T. A snapshot of intraoperative conditions to predict prolonged postoperative ileus after colorectal surgery. ANZ J Surg. 2022;92:2199-2206. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 3] [Reference Citation Analysis (0)] |
| 11. | Guo F, Sun Z, Wang Z, Gao J, Pan J, Zhang Q, Ren S. Nomogram for predicting prolonged postoperative ileus after laparoscopic low anterior resection for rectal cancer. World J Surg Oncol. 2023;21:380. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 9] [Reference Citation Analysis (0)] |
| 12. | Liang WQ, Zhang KC, Cui JX, Xi HQ, Cai AZ, Li JY, Liu YH, Liu J, Zhang W, Wang PP, Wei B, Chen L. Nomogram to predict prolonged postoperative ileus after gastrectomy in gastric cancer. World J Gastroenterol. 2019;25:5838-5849. [PubMed] [DOI] [Full Text] |
| 13. | Pei L, Wang G, Yang S, Zhou S, Xu T, Zhou J, Zhang W, Lu K, Hu L, Wang Y, Wang K, You D, Wu Y, Li L, Guo J, Sun J. Electroacupuncture Reduces Duration of Postoperative Ileus After Laparoscopic Gastrectomy for Gastric Cancer: A Multicenter Randomized Trial. Gastroenterology. 2025;169:73-84. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 35] [Cited by in RCA: 32] [Article Influence: 32.0] [Reference Citation Analysis (3)] |
| 14. | Saito SN, Otsuka S, Zenbutsu S, Hori S, Honda M, Nakagawa S. Generation mechanisms of bowel sounds by simultaneous measurements of X-ray fluoroscopy and bowel sounds. Annu Int Conf IEEE Eng Med Biol Soc. 2021;2021:1593-1596. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1] [Cited by in RCA: 4] [Article Influence: 0.8] [Reference Citation Analysis (0)] |
| 15. | Zhao K, Jiang H, Wang Z, Chen P, Zhu B, Duan X. Long-Term Bowel Sound Monitoring and Segmentation by Wearable Devices and Convolutional Neural Networks. IEEE Trans Biomed Circuits Syst. 2020;14:985-996. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 41] [Cited by in RCA: 22] [Article Influence: 3.7] [Reference Citation Analysis (0)] |
| 16. | Hebbelstrup Jensen B, Olsen KE, Struve C, Krogfelt KA, Petersen AM. Epidemiology and clinical manifestations of enteroaggregative Escherichia coli. Clin Microbiol Rev. 2014;27:614-630. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 129] [Cited by in RCA: 147] [Article Influence: 13.4] [Reference Citation Analysis (0)] |
| 17. | Wiseman LR, Faulds D. Cisapride. An updated review of its pharmacology and therapeutic efficacy as a prokinetic agent in gastrointestinal motility disorders. Drugs. 1994;47:116-152. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 171] [Cited by in RCA: 152] [Article Influence: 4.8] [Reference Citation Analysis (0)] |
| 18. | Tomomasa T, Morikawa A, Sandler RH, Mansy HA, Koneko H, Masahiko T, Hyman PE, Itoh Z. Gastrointestinal sounds and migrating motor complex in fasted humans. Am J Gastroenterol. 1999;94:374-381. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 67] [Cited by in RCA: 53] [Article Influence: 2.0] [Reference Citation Analysis (0)] |
| 19. | Inderjeeth AJ, Webberley KM, Muir J, Marshall BJ. The potential of computerised analysis of bowel sounds for diagnosis of gastrointestinal conditions: a systematic review. Syst Rev. 2018;7:124. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 38] [Cited by in RCA: 28] [Article Influence: 3.5] [Reference Citation Analysis (0)] |
| 20. | Gibbons JC, Sullivan JF. Chronic idiopathic pseudo-obstructive bowel disease. Am J Gastroenterol. 1978;70:306-313. [PubMed] |
| 21. | Yoshino H, Abe Y, Yoshino T, Ohsato K. Clinical application of spectral analysis of bowel sounds in intestinal obstruction. Dis Colon Rectum. 1990;33:753-757. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 61] [Cited by in RCA: 43] [Article Influence: 1.2] [Reference Citation Analysis (0)] |
| 22. | Dalle D, Devroede G, Thibault R, Perrault J. Computer analysis of bowel sounds. Comput Biol Med. 1975;4:247-256. [PubMed] [DOI] [Full Text] |
| 23. | Kaneshiro M, Kaiser W, Pourmorady J, Fleshner P, Russell M, Zaghiyan K, Lin A, Martinez B, Patel A, Nguyen A, Singh D, Zegarski V, Reid M, Dailey F, Xu J, Robbins K, Spiegel B. Postoperative Gastrointestinal Telemetry with an Acoustic Biosensor Predicts Ileus vs. Uneventful GI Recovery. J Gastrointest Surg. 2016;20:132-9; discussion 139. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 36] [Cited by in RCA: 27] [Article Influence: 2.7] [Reference Citation Analysis (0)] |
| 24. | Spiegel BM, Kaneshiro M, Russell MM, Lin A, Patel A, Tashjian VC, Zegarski V, Singh D, Cohen SE, Reid MW, Whitman CB, Talley J, Martinez BM, Kaiser W. Validation of an acoustic gastrointestinal surveillance biosensor for postoperative ileus. J Gastrointest Surg. 2014;18:1795-1803. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 40] [Cited by in RCA: 29] [Article Influence: 2.4] [Reference Citation Analysis (3)] |
| 25. | Kronberg U, Kiran RP, Soliman MS, Hammel JP, Galway U, Coffey JC, Fazio VW. A characterization of factors determining postoperative ileus after laparoscopic colectomy enables the generation of a novel predictive score. Ann Surg. 2011;253:78-81. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 161] [Cited by in RCA: 142] [Article Influence: 9.5] [Reference Citation Analysis (3)] |
| 26. | The F, Cailotto C, van der Vliet J, de Jonge WJ, Bennink RJ, Buijs RM, Boeckxstaens GE. Central activation of the cholinergic anti-inflammatory pathway reduces surgical inflammation in experimental post-operative ileus. Br J Pharmacol. 2011;163:1007-1016. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 63] [Cited by in RCA: 69] [Article Influence: 4.6] [Reference Citation Analysis (0)] |
| 27. | Schmidt J, Stoffels B, Moore BA, Chanthaphavong RS, Mazie AR, Buchholz BM, Bauer AJ. Proinflammatory role of leukocyte-derived Egr-1 in the development of murine postoperative ileus. Gastroenterology. 2008;135:926-936, 936.e1. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 45] [Cited by in RCA: 53] [Article Influence: 2.9] [Reference Citation Analysis (0)] |
| 28. | Jauregui-Amezaga A, Rimola J, Ordás I, Rodríguez S, Ramírez-Morros A, Gallego M, Masamunt MC, Llach J, González-Suárez B, Ricart E, Panés J. Value of endoscopy and MRI for predicting intestinal surgery in patients with Crohn's disease in the era of biologics. Gut. 2015;64:1397-1402. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 63] [Cited by in RCA: 78] [Article Influence: 7.1] [Reference Citation Analysis (3)] |
| 29. | Dai F, Zhang R, Deng R, Wang G, Guo H, Guo C. Regular use of low-dose of opioids after gastrointestinal surgery may lead to postoperative gastrointestinal tract dysfunction in children: a Chinese national regional health center experience sharing. BMC Gastroenterol. 2023;23:369. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 4] [Reference Citation Analysis (0)] |
| 30. | Varga A, Matrai AA, Bedocs-Barath B, Fazekas LA, Brasil FS, Mehta A, Vanyolos E, Deak A, Lesznyak T, Peto K, Nemeth N. Local and Systemic Micro-Rheological Changes during Intestinal Anastomosis Operation: A Metabolic Dependence in an Experimental Model. Metabolites. 2024;14:458. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 1] [Reference Citation Analysis (0)] |
| 31. | Fan ZQ, Chen Y, Fu XA, Yin HT, Li JS, Wang WS, Yuan JQ, Guo SG. Nomogram for predicting prolonged postoperative ileus in colorectal cancer based on age and inflammatory markers. Biomark Med. 2023;17:921-933. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1] [Cited by in RCA: 10] [Article Influence: 3.3] [Reference Citation Analysis (0)] |
| 32. | Ju H, Shen K, Li J, Feng Y. Total postoperative opioid dose is an independent risk factor for prolonged postoperative ileus after laparoscopic colorectal surgery: a case-control study. Korean J Anesthesiol. 2024;77:133-138. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 2] [Cited by in RCA: 6] [Article Influence: 3.0] [Reference Citation Analysis (0)] |