Development and validation of a machine learning-based risk model for prolonged length of stay after laparoscopic gastrointestinal surgery in women: A secondary analysis of the FDP-PONV trail

Development and validation of a machine learning-based risk model for prolonged length of stay after laparoscopic gastrointestinal surgery in women: A secondary analysis of the FDP-PONV trail | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Development and validation of a machine learning-based risk model for prolonged length of stay after laparoscopic gastrointestinal surgery in women: A secondary analysis of the FDP-PONV trail Jiankun Shi, Yabin Huang, Jiaxin Han, Shimin Zhang, Xingshan Cheng, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6659561/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Oct, 2025 Read the published version in BMC Gastroenterology → Version 1 posted 10 You are reading this latest preprint version Abstract Background Prolonged length of stay (PLOS) after surgery is associated with several clinical risks and increased medical costs. We aimed to develop a risk model for PLOS based on clinical features throughout pre-, intra-, and post-operative periods in women undergoing laparoscopic gastrointestinal surgery. Methods Women who underwent laparoscopic gastrointestinal surgery in the FDP-PONV randomized controlled trial were eligible for this secondary analysis. PLOS was defined as a duration longer than the median postoperative length of stay. All 96 clinical features prospectively collected in the FDP-PONV trial were used to generate the models. Six machine learning algorithms were employed: logistic regression, K-nearest neighbor, gradient boosting machine, random forest, support vector machine, and extreme gradient boosting. The model performance was assessed using numerous metrics and evaluated using bootstrapping with 1000 replicates. Results In total, 770 and 331 patients were assigned to the training and validation cohorts, respectively. The logistic regression model performed best in the validation cohort [area under the receiver operating characteristic curve (AUC) 0.751, 95% confidence interval (CI), 0.699–0.804] and the bootstrapping (AUC 0.760, 95% CI 0.722–0.796). A nomogram based on ten factors, including education level, preoperative hypoalbuminemia, preoperative insufficient sleep, duration of surgery, blood loss, postoperative hypotension, postoperative albumin infusion, postoperative serum phosphorus level, highest pain score during 73–120 h after surgery, and postoperative infection, was established. Conclusions A risk model was established using machine learning to characterize the key perioperative drivers and identify women at high risk of PLOS after laparoscopic gastrointestinal surgery. Trial registration: The FDP-PONV trial was registered at clinicaltrials.gov (NCT04853147) on 2021-04-27. length of hospital stay gastrointestinal surgery women Figures Figure 1 Figure 2 Figure 3 1. Background Laparoscopic gastrointestinal surgery is frequently performed to remove lesions and improve survival in patients with gastrointestinal cancer or benign disease. It is particularly relevant as colorectal cancer and stomach cancer were ranked among the top five most common cancers in the Global Cancer Statistics 2022 report [ 1 ]. Furthermore, the laparoscopic approach was associated with a decreased risk for all adverse outcomes [ 2 , 3 ]. The hospital length of stay (LOS) is an important quality metric for surgery, particularly in the era of enhanced recovery after surgery. Prolonged length of stay (PLOS) delays discharge, increases the risk of hospital-acquired infections, predicts a greater risk of short-term mortality, increases the use of medical resources (resulting in higher costs), and affects patient flow and access to healthcare [ 4 ]. Due to the clinical importance of LOS, studies have been undertaken globally to evaluate the factors associated with PLOS, and risk factors, such as age, sex, comorbidity, preoperative patient-reported outcomes, and laboratory test results (such as albumin and hemoglobin) have been reported [ 5 – 8 ]. However, most studies used a small number of subjects, did not evaluate a wide variety of clinical factors, or used information from clinical medical records, which might not be as accurate or complete as the information collected in prospective clinical trials. Additionally, the factors influencing LOS varied from surgery and population [ 9 ] and should be investigated in specific situations. Women live longer than men but spend more time experiencing ill health and disability [ 10 ]. It is essential to highlight the health of women as an important part of global health [ 11 ]. Previous studies have shown gender differences in treatment effect, and compared with men, women are at increased risk for postoperative complications, including PLOS and surgical site infection [ 12 – 15 ]. Therefore, the identification of specific factors associated with PLOS in women is important. However, to date, little research has been conducted on PLOS in women after laparoscopic gastrointestinal surgery. Currently, artificial intelligence can be used to analyze large clinical datasets to improve patient care and may help address challenges unique to the field of global health. Machine learning is one of the most common artificial intelligence technologies used for complex problem-solving, cost-benefit evaluations, and risk assessment [ 16 ]. Therefore, this study aimed to develop and validate a machine learning-based risk model for PLOS after laparoscopic gastrointestinal surgery in women, using comprehensive clinical features during the perioperative period, including the pre-, intra-, and postoperative periods. 2. Methods 2.1 Study population The study was a secondary analysis of the FDP-PONV trial, a randomized, controlled, double-blind study designed to evaluate the efficacy and safety of adding fosaprepitant to dexamethasone and palonosetron for preventing postoperative nausea and vomiting in patients undergoing laparoscopic gastrointestinal surgery [ 17 ]. The FDP-PONV trial was approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2021ZSLYEC-78) and registered at clinicaltrials.gov (NCT04853147). The trial was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients before their participation. The inclusion criteria of the FDP-PONV trial were age 18–75 years, scheduled for laparoscopic gastrointestinal surgery, and three or four Apfel risk factors [ 18 ]. Exclusion criteria included an American Society of Anesthesiologists (ASA) physical status of 4–6, severe hepatic dysfunction; contraindications to fosaprepitant, 5-hydroxytryptamine type-3 receptor antagonist, or dexamethasone; preoperative use of medications with known antiemetic properties; mental disorders or inability to communicate; and pregnant women or nursing mothers. Female patients who underwent laparoscopic gastrointestinal surgery in the FDP-PONV trial were included in this study. This study was also approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2024ZSLYEC-160) on 3 April 2024. 2.2 Outcome parameters The primary outcome was PLOS, defined as a LOS greater than the median in this study. The LOS was defined as the time from the date of operation completion to the date of hospital discharge. The criteria for hospital discharge included [ 19 – 21 ]: 1) tolerance of diet with no need for intravenous nutrition; 2) analgesic-free status, defined as a visual analog scale score ≤ 3 without intravenous analgesic drugs; 3) adequate mobility; and 4) afebrile status without major infectious complications. 2.3 Predictive variables associated with PLOS All 96 variables, including baseline characteristics, intraoperative information, postoperative outcomes, and adverse events prospectively collected in the FDP-PONV trial throughout the pre-, intra-, and postoperative periods were considered potential predictors (Additional file 1). 2.4 Sample size of machine learning model To ensure accurate estimates and reduce the risk of overfitting, we adopted a methodology outlined in the British Medical Journal for calculating the sample size required to develop a clinical prediction model [ 22 ]. This approach was grounded in the principle of events per variable (EPV), the aggregate number of participants, the outcome prevalence within the study population, and the anticipated predictive accuracy of the model. The minimum required total sample size for 10 candidate predictors was determined to be 990, with an events per variable (EPV) of 49.7 (events fraction, 0.5, criterion value rMAPE, 0.05), based on calculations available at https://mvansmeden.shinyapps.io/BeyondEPV/ . Consequently, a sample size of 1,101 patients proved adequate to sustain the development of this model with 10 predictors through the application of machine learning. 2.5 Statistical analysis A multiple interpolation method was used to fill the missing values. The data are presented as means ± standard deviations, medians [interquartile ranges], or numbers (percentages), as appropriate. Student's t-test for independent samples was employed to compare normally distributed continuous variables that exhibited equal variances across groups. The Wilcoxon rank-sum test was employed to analyze ranked variables, continuous variables with non-normal distributions, and variables exhibiting unequal variances across groups. Categorical data were analyzed using Pearson's chi-square test or, when appropriate, Fisher's exact test. Statistical significance was established at a two-sided p-value of less than 0.05. Prediction models were created by randomly splitting the dataset into a training set containing 70% of the patients and a testing set containing 30% of the patients. Least absolute shrinkage and selection operator (LASSO) was employed to identify variables featuring clinically significant attributes, alongside determining the optimal lambda value. Stepwise regression, guided by the Akaike Information Criterion (AIC), was employed to identify the final variables for inclusion in the model. Six machine learning algorithms—logistic regression (LR), K-nearest neighbor (KNN), gradient boosting machine (GBM), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost)—were utilized within R version 4.2.2 to construct risk models for PLOS. Hyperparameters for each model were optimized via grid search and 10-fold cross-validation. The performance of these models was evaluated in both training and validation cohorts, employing metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, Brier score, and calibration curve. Decision curve analysis (DCA) was applied to ascertain the net benefit of the models. Furthermore, an internal validation process involving 1000 bootstrap re-samples was conducted to ensure the predictive model's robustness. A nomogram derived from the logistic regression model was developed as a practical clinical tool. 3. Results 3.1 Patients and datasets In total, 1154 patients were randomly assigned to the FDP-PONV trial between April 2021 and April 2022. After excluding 40 male patients and 13 patients who did not undergo laparoscopic gastrointestinal surgery, 1101 female patients who underwent laparoscopic gastrointestinal surgery were included in the final dataset (Fig. 1 ). The median LOS of 1101 patients enrolled in this study was 7 days, and PLOS (LOS > 7 days) occurred in 535 patients (48.6%). The randomly split training (770 patients) and test (331 patients) sets are shown in Additional file 2. 3.2 Model development by machine learning method First, LASSO was used to select features after choosing an optimal lambda value of 0.01928917. Features with a coefficient of zero were considered redundant and removed, leaving 31 features for downstream stepwise regression. The details of feature selection using LASSO are shown in Additional file 3 and Additional file 4. Table 1 compares predictor variables selected by LASSO between patients with and without PLOS. Stepwise regression based on the AIC was used to further select the features. Finally, considering clinical significance and sample size, ten features were retained for inclusion in the models (Additional file 5). Table 1 Potential predictors selected by least absolute shrinkage and selection operator regression for PLOS Overall (N = 1101) Non-PLOS (N = 566) PLOS (N = 535) p value Use of fosaprepitant 0.07 No 550 (50.0%) 298 (52.7%) 252 (47.1%) Yes 551 (50.0%) 268 (47.3%) 283 (52.9%) Education level a < 0.01 Low 710 (64.5%) 330 (58.3%) 380 (71.0%) Medium 196 (17.8%) 107 (18.9%) 89 (16.6%) High 195 (17.7%) 129 (22.8%) 66 (12.3%) Yes 127 (11.5%) 64 (11.3%) 63 (11.8%) History of diabetes mellitus 0.05 No 999 (90.7%) 523 (92.4%) 476 (89.0%) Yes 102 (9.3%) 43 (7.6%) 59 (11.0%) Preoperative insufficient sleep b 0.01 No 787 (71.5%) 424 (74.9%) 363 (67.9%) Yes 314 (28.5%) 142 (25.1%) 172 (32.1%) Preoperative hypoalbuminemia < 0.01 No 903 (82.0%) 492 (86.9%) 411 (76.8%) Yes 198 (18.0%) 74 (13.1%) 124 (23.2%) Preoperative serum calcium (mmol/L) 2.25 (2.18, 2.33) 2.26 (2.19, 2.34) 2.25 (2.18, 2.33) 0.13 Preoperative hemoglobin (g/L) 117.00 (103.00, 127.00) 119.00 (105.00, 128.00) 115.00 (98.00, 126.00) < 0.01 Duration of surgery (min) 185.00 (146.00, 243.00) 166.00 (133.00, 208.75) 212.00 (164.50, 272.00) < 0.01 Dose of propofol (mg) 1,370.00 (1,070.00, 1,760.00) 1,220.00 (1,000.00, 1,509.25) 1,510.00 (1,180.00, 1,925.00) < 0.01 Use of neostigmine 0.40 No 408 (37.1%) 203 (35.9%) 205 (38.3%) Yes 693 (62.9%) 363 (64.1%) 330 (61.7%) Infusion of fluid in operating room (mL) 2,450.00 (2,250.00, 2,850.00) 2,350.00 (1,975.00, 2,750.00) 2,750.00 (2,250.00, 2,950.00) < 0.01 Intraoperative blood loss (mL) 50.00 (50.00, 100.00) 50.00 (30.00, 50.00) 50.00 (50.00, 100.00) < 0.01 Postoperative hypotension c 0.01 No 959 (87.1%) 511 (90.3%) 448 (83.7%) Yes 142 (12.9%) 55 (9.7%) 87 (16.3%) Postoperative opioids consumption (morphine equivalents, mg) 38.28 (30.80, 46.67) 38.00 (30.13, 46.67) 38.56 (31.36, 46.67) 0.11 Postoperative albumin infusion < 0.01 No 169 (15.3%) 123 (21.7%) 46 (8.6%) Yes 932 (84.7%) 443 (78.3%) 489 (91.4%) Postoperative hypoalbuminemia < 0.01 No 310 (28.2%) 190 (33.6%) 120 (22.4%) Yes 791 (71.8%) 376 (66.4%) 415 (77.6%) Postoperative fasting blood glucose (mmol/L) 8.25 (6.95, 10.30) 7.99 (6.85, 9.86) 8.60 (7.10, 10.55) < 0.01 Postoperative serum calcium (mmol/L) 2.10 ± 0.12 2.12 ± 0.11 2.09 ± 0.13 < 0.01 Postoperative serum phosphorus (mmol/L) 1.20 (1.05, 1.33) 1.20 (1.05, 1.32) 1.20 (1.05, 1.34) 0.79 Postoperative hemoglobin (g/L) 106.00 (92.00, 116.00) 108.00 (96.00, 117.00) 103.00 (89.00, 114.00) < 0.01 Postoperative monocyte count (*10 9 /L) 0.40 (0.23, 0.61) 0.34 (0.19, 0.56) 0.45 (0.29, 0.64) < 0.01 Postoperative nausea and vomiting (0–24 h) 0.19 No 647 (58.8%) 322 (56.9%) 325 (60.7%) Yes 454 (41.2%) 244 (43.1%) 210 (39.3%) Postoperative nausea (73-120h) < 0.01 No 863 (78.4%) 475 (83.9%) 388 (72.5%) Yes 238 (21.6%) 91 (16.1%) 147 (27.5%) Postoperative vomiting (0-120 h) 0.92 No 729 (66.2%) 374 (66.1%) 355 (66.4%) Yes 372 (33.8%) 192 (33.9%) 180 (33.6%) Postoperative pain score d (0–24 h) 4.00 (3.00, 5.00) 4.00 (3.00, 5.00) 4.00 (3.00, 5.00) 0.68 Postoperative pain score d (73–120 h) 2.00 (2.00, 3.00) 2.00 (2.00, 3.00) 3.00 (2.00, 4.00) < 0.01 Time to first flatus after surgery (h) 48.00 (30.00, 68.00) 44.00 (25.00, 62.00) 54.00 (38.00, 75.00) < 0.01 Time to first defaecation after surgery (h) 70.00 (46.00, 99.00) 66.50 (43.00, 91.00) 76.00 (50.00, 112.50) < 0.01 Postoperative bleeding 0.01 No 1,081 (98.2%) 563 (99.5%) 518 (96.8%) Yes 20 (1.8%) 3 (0.5%) 17 (3.2%) Postoperative infection e < 0.01 No 1,033 (93.8%) 559 (98.8%) 474 (88.6%) Yes 68 (6.2%) 7 (1.2%) 61 (11.4%) Anastomotic leak < 0.01 No 1,070 (97.2%) 565 (99.8%) 505 (94.4%) Yes 31 (2.8%) 1 (0.2%) 30 (5.6%) PLOS, prolonged length of stay after surgery. a Education level was categorized as low (no education or primary school only), medium (only secondary school), and high level (professional education or university). b Preoperative insufficient sleep was defined as sleep duration < 4 h on the night before surgery. c Postoperative hypotension was defined as mean arterial pressure < 65 mmHg. d Postoperative pain score meant the highest pain score (numeric rating scale) at mobility. e Postoperative infection consists of surgical site infection, anastomotic leak, abdominal abscess, pelvic abscess, hepatic abscess, central venous catheter-related infection, urinary tract infection or pneumonia. The AUCs for the six models ranged from 0.689 to 0.751, and the receiver operating characteristic (ROC) curves are shown in Fig. 2 . Table 2 summarizes the performance of each model. The logistic regression model performed best, achieving an AUC of 0.751 (95% CI, 0.699–0.804), with 0.713 (95% CI, 0.712–0.714) accuracy, 0.829 sensitivity, 0.607 specificity, 0.658 positive predictive value, 0.795 negative predictive value and 0.734 F1 score on the validation cohort. The Brier score of the logistic model for predicting PLOS was 0.205 in the validation cohort. The bootstrapping method by 1000 times sampling on the whole data set (Table 3 ) also showed that the logistic regression model performed best, achieving an AUC of 0.760 (95% CI, 0.722–0.796). These results indicated that the proposed model was reliable. The DCA of the logistic regression model in the validation cohort showed that if the threshold probability of patients was 25–70%, using this model to predict the risk of PLOS could add more benefit than either the treat-all tactics or the treat-none tactics (Fig. 2 ). Calibration curve analyses were performed for further model evaluation (Additional file 6). Table 2 Model performance in the training and validation cohorts Model AUC (95% CI) Accuracy (95% CI) Sensitivity Specificity PPV NPV F1 score Brier Training cohort LR 0.776 (0.743–0.808) 0.713 (0.712–0.714) 0.700 0.725 0.710 0.716 0.673 0.191 KNN 0.806 (0.777–0.836) 0.725 (0.724–0.725) 0.653 0.794 0.752 0.704 0.699 0.178 GBM 0.835 (0.807–0.863) 0.781 (0.780–0.781) 0.716 0.842 0.813 0.756 0.761 0.164 RF 0.987 (0.981–0.992) 0.945 (0.945–0.946) 0.944 0.947 0.944 0.947 0.944 0.042 SVM 0.814 (0.783–0.844) 0.753 (0.753–0.754) 0.743 0.763 0.751 0.756 0.747 0.172 XGBoost 0.821 (0.792–0.850) 0.758 (0.758–0.759) 0.743 0.774 0.759 0.758 0.751 0.172 Validation cohort LR 0.751 (0.699–0.804) 0.713 (0.712–0.714) 0.829 0.607 0.658 0.795 0.734 0.205 KNN 0.733 (0.680–0.787) 0.683 (0.681–0.684) 0.627 0.734 0.683 0.683 0.654 0.208 GBM 0.744 (0.692–0.797) 0.695 (0.694–0.696) 0.696 0.694 0.675 0.714 0.685 0.205 RF 0.689 (0.633–0.745) 0.656 (0.656–0.657) 0.677 0.636 0.629 0.683 0.652 0.222 SVM 0.733 (0.679–0.788) 0.680 (0.678–0.681) 0.778 0.590 0.634 0.745 0.699 0.208 XGBoost 0.732 (0.678–0.786) 0.686 (0.685–0.687) 0.677 0.694 0.669 0.702 0.673 0.209 LR, logistic regression; KNN, K-nearest neighbor; GBM, gradient boosting machine; RF, random forest; SVM, support vector machine; XGBoost, extreme gradient boosting; AUC, area under the receiver operating characteristic curve; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value. Table 3 Model effect evaluated by the bootstrap method Model AUC Lower Upper p -value LR 0.760 0.722 0.796 0.02 KNN 0.637 0.592 0.679 0.02 GBM 0.742 0.705 0.781 0.02 RF 0.661 0.619 0.702 0.02 SVM 0.732 0.692 0.770 0.02 XGBoost 0.742 0.705 0.780 0.02 LR, logistic regression; KNN, K-nearest neighbor; GBM, gradient boosting machine; RF, random forest; SVM, support vector machine; XGBoost, extreme gradient boosting; AUC, area under the receiver operating characteristic curve. 3.3 Nomogram for PLOS after laparoscopic gastrointestinal surgery As the logistic regression model performed well, it was transformed into a nomogram for easier use in different clinical situations. As shown in Fig. 3 , ten factors—including education level, preoperative hypoalbuminemia, preoperative insufficient sleep (sleep duration < 4 h), duration of surgery, intraoperative blood loss, postoperative hypotension (mean arterial pressure < 65 mmHg), postoperative albumin infusion, postoperative serum phosphorus, highest pain score at mobility during 73–120 h after surgery, and postoperative infection—were screened to form the final nomogram. 4. Discussion In this study, we used perioperative data from the FDP-PONV trial of 1101 female patients undergoing laparoscopic gastrointestinal surgery to develop and validate six machine learning-based risk models for PLOS. Three models (LR, GBM, and XGBoost) achieved AUC values > 0.741 in internal validation by bootstrapping 1000 times, with the potential to help in the clinical prevention and management of PLOS. The LR model achieved the best discrimination, with an AUC of 0.760, and was converted into a nomogram to enhance its potential clinical applicability. Machine learning algorithms allow computers to learn from numerous data to provide evidence to clinicians and play a considerable role in clinical decision-making. Risk factors associated with PLOS are complex and may exist in every link during hospitalization. This study used 96 variables prospectively collected in the FDP-PONV trial throughout the pre-, intra-, and postoperative periods to expand the coverage as widely as possible, different from previous studies focusing on preoperative factors or searching clinical medical records retrospectively [ 23 – 25 ]. Risk models were established based on three preoperative, two intraoperative, and five postoperative factors. Among these, five classic factors, including preoperative hypoalbuminemia, duration of surgery, intraoperative blood loss, postoperative pain score, and postoperative infection, were reported to be associated with PLOS in many previous studies [ 26 – 30 ]. Additionally, five relatively new and interesting factors were identified: education level, postoperative sleep duration, postoperative serum phosphorus, postoperative hypotension, and postoperative albumin infusion. Compared to the high level of education (professional education or university), low (no education or primary school only), and medium (only secondary school) levels of education [ 31 ] were related to the occurrence of PLOS in the study. With the promotion of ERAS, the level of education is increasingly important in recovery since it affects the patients’ comprehension of the strategies for rehabilitation and their compliance [ 32 ]. Sleep is vital for maintaining healthy mental, physical, and metabolic states [ 33 ]. Severe insomnia (sleep duration < 3.5 h) in women was associated with an increase in mortality [ 34 ], insufficient sleep duration (< 4 h) could elevate the risk of cognitive disorders [ 35 ], and higher preoperative sleep quality induced a shorter stay after total joint arthroplasty [ 36 ]. Based on previous studies and the collected data, this study offered a cut-off value of sleep duration on the night before surgery and found that a sleep duration of < 4 h was a risk factor for PLOS in women. As usual, the serum phosphorus level after surgery did not attract as much attention as potassium or sodium levels. A few previous studies have shown that hypophosphatemia following hepatectomy is associated with decreased morbidity and mortality [ 37 , 38 ], but hypophosphatemia following pancreatectomy may be associated with postoperative pancreatic fistula [ 39 ], suggesting that the clinical significance of serum phosphorus varies with the surgical type. This study used data on several electrolytes, including serum potassium, sodium, calcium, and phosphorus, both before and after surgery for analysis, and only the level of serum phosphorus after surgery was entered into the models as a risk factor for PLOS. Most studies have focused on the relationship between intraoperative hypotension and adverse events [ 40 , 41 ]. This study fully investigated blood pressure events, including intraoperative hypotension and hypertension, as well as postoperative hypotension and hypertension; only postoperative hypotension was selected. This finding was consistent with a cohort study showing that only postoperative hypotension, and not intraoperative hypotension, was associated with myocardial injury in 1710 patients undergoing non-cardiac surgery [ 42 ], suggesting that postoperative hypotension should also be paid more attention to. Although albumin infusion is usually considered to reduce tissue edema [ 43 ], this study found that postoperative albumin infusion was a risk factor for PLOS. A retrospective cohort study of 1051441 patients undergoing elective total hip and knee arthroplasties also found that perioperative fluid resuscitation with albumin 5% was associated with an increased risk of acute renal failure and most other complications [ 44 ]. These results suggest that after surgery, albumin infusion should be administered as minimally as possible. Our study had some limitations. First, although data from the FDP-PONV trial were collected carefully and prospectively, a retrospective analysis of PLOS was performed. Second, although a bootstrap testing approach was used to validate our model, further external validation in a wider population is also essential. Third, although 96 variables were considered throughout the perioperative period, some potential risk factors with large missing values, such as C-reactive protein and procalcitonin levels, were not included in this study. In conclusion, we successfully developed a simple and reliable tool to predict the occurrence of PLOS after laparoscopic gastrointestinal surgery in women. Such individualized risk analyses allow clinicians to monitor and treat patients in a targeted manner to shorten LOS. Abbreviations LOS Length of Stay PLOS Prolonged Length of Stay ASA American Society of Anesthesiologists EPV Events Per Variable LASSO Least Absolute Shrinkage and Selection Operator AIC Akaike Information Criterion LR Logistic Regression KNN K-Nearest Neighbor GBM Gradient Boosting Machine RF Random Forest SVM Support Vector Machine XGBoost Extreme Gradient Boosting AUC Area Under the Receiver Operating Characteristic Curve DCA Decision Curve Analysis ROC Receiver Operating Characteristic Declarations Ethics approval and consent to participate The FDP-PONV trial was approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2021ZSLYEC-78). Written informed consent was obtained from all patients before their participation. The FDP-PONV trial was conducted in accordance with the Declaration of Helsinki. The secondary analysis of the FDP-PONV trail was also approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2024ZSLYEC-160). Consent for publication Not applicable. Availability of data and materials All data of this study are available upon reasonable request to the principal investigator. Competing interests The authors declare that they have no competing interests. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Authors' contributions Study design: TL; Data collection: JS, YH, JH, SZ, XC; Statistical analysis: TL, JS, YH, JH; Data interpretation: SZ, XC; Writing of paper: JS, YH, JH; Manuscript revision: TL. All authors read and approved the final manuscript. Acknowledgements We would like to thank Editage (www.editage.cn) for English language editing. References Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin . 2024;74:229-63. van den Bosch T, Warps AK, de Nerée Tot Babberich MPM, Stamm C, Geerts BF, Vermeulen L, et al. Predictors of 30-day mortality among Dutch patients undergoing colorectal cancer surgery, 2011-2016. JAMA Netw Open . 2021;4:e217737. Cai Z, Song H, Huang Z, Fingerhut A, Xu X, Zhong H, et al. Safety and feasibility of laparoscopic surgery for colorectal and gastric cancer under the Chinese multi-site practice policy: admittance standards of competence are needed. Gastroenterol Rep (Oxf) . 2022;10:goac046. Rosman M, Rachminov O, Segal O, Segal G. Prolonged patients' In-Hospital Waiting Period after discharge eligibility is associated with increased risk of infection, morbidity and mortality: a retrospective cohort analysis. BMC Health Serv Res . 2015;15:246. Gohil R, Rishi M, Tan BHL. Pre-operative serum albumin and neutrophil-lymphocyte ratio are associated with prolonged hospital stay following colorectal cancer surgery. Br J Med Med Res . 2014;4:481-7. Nishimuta H, Kusachi S, Watanabe M, Asai K, Kiribayashi T, Niitsuma T, et al. Impact of postoperative remote infection on length of stay and medical costs in hospitals in Japan. Surg Today . 2021;51:212-8. Sutherland J, Liu G, Crump T, Bair M, Karimuddin A. Relationship between preoperative patient-reported outcomes and hospital length of stay: a prospective cohort study of general surgery patients in Vancouver, Canada. J Health Serv Res Policy . 2019;24:29-36. Zeidan M, Goz V, Lakomkin N, Spina N, Brodke DS, Spiker WR. Predictors of readmission and prolonged length of stay after cervical disc arthroplasty. Spine . 2021;46:487-91. Gokhale S, Taylor D, Gill J, Hu Y, Zeps N, Lequertier V, et al. Hospital length of stay prediction for general surgery and total knee arthroplasty admissions: systematic review and meta-analysis of published prediction models. Digit Health . 2023;9:589821209:20552076231177497. Connolly A, Regan DL. Women's health matters. Br J Gen Pract . 2022;72:556-7. Koplan JP, Bond TC, Merson MH, Reddy KS, Rodriguez MH, Sewankambo NK, et al. Towards a common definition of global health. Lancet . 2009;373:1993-5. Toren P, Wilkins A, Patel K, Burley A, Gris T, Kockelbergh R, et al. The sex gap in bladder cancer survival - a missing link in bladder cancer care? Nat Rev Urol . 2024;21:181-92. Seeman MV. Men and women respond differently to antipsychotic drugs. Neuropharmacology . 2020;163:107631. Wilson LAB, Zajitschek SRK, Lagisz M, Mason J, Haselimashhadi H, Nakagawa S. Sex differences in allometry for phenotypic traits in mice indicate that females are not scaled males. Nat Commun . 2022;13:7502. Heyer JH, Cao NA, Amdur RL, Rao RR. Postoperative complications following orthopedic spine surgery: is there a difference between men and women? Int J Spine Surg . 2019;13:125-31. Schwalbe N, Wahl B. Artificial intelligence and the future of global health. Lancet . 2020;395:1579-86. Huang Q, Wang F, Liang C, Huang Y, Zhao Y, Liu C, et al. Fosaprepitant for postoperative nausea and vomiting in patients undergoing laparoscopic gastrointestinal surgery: a randomised trial. Br J Anaesth . 2023;131:673-81. Apfel CC, Läärä E, Koivuranta M, Greim CA, Roewer N. A simplified risk score for predicting postoperative nausea and vomiting: conclusions from cross-validations between two centers. Anesthesiology . 1999;91:693-700. Fiore JF, Browning L, Bialocerkowski A, Gruen RL, Faragher IG, Denehy L. Hospital discharge criteria following colorectal surgery: a systematic review. Colorectal Dis . 2012;14:270-81. Lee TG, Kang SB, Kim DW, Hong S, Heo SC, Park KJ. Comparison of early mobilization and diet rehabilitation program with conventional care after laparoscopic colon surgery: a prospective randomized controlled trial. Dis Colon Rectum . 2011;54:21-8. Guner A, Kim KY, Park SH, Cho M, Kim YM, Hyung WJ, et al. Safe discharge criteria after curative gastrectomy for gastric cancer. J Gastric Cancer . 2022;22:395-407. Riley RD, Ensor J, Snell KIE, Harrell FE, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. BMJ . 2020;368:m441. Jo YY, Han J, Park HW, Jung H, Lee JD, Jung J, et al. Prediction of prolonged length of hospital stay after cancer surgery using machine learning on electronic health records: retrospective cross-sectional study. JMIR Med Inform . 2021;9:e23147. Triana AJ, Vyas R, Shah AS, Tiwari V. Predicting length of stay of coronary artery bypass grafting patients using machine learning. J Surg Res . 2021;264:68-75. Daghistani TA, Elshawi R, Sakr S, Ahmed AM, Al-Thwayee A, Al-Mallah MH. Predictors of in-hospital length of stay among cardiac patients: A machine learning approach. Int J Cardiol . 2019;288:140-7. Shankar H, Sureshkumar S, Gurushankari B, Samanna Sreenath G, Kate V. Factors predicting prolonged hospitalization after abdominal wall hernia repair - a prospective observational study. Turk J Surg . 2021;37:96-102. Li Z, Li H, Lv P, Peng X, Wu C, Ren J, et al. Prospective multicenter study on the incidence of surgical site infection after emergency abdominal surgery in China. Sci Rep . 2021;11:7794. Charalambides M, Mavrou A, Jennings T, Powar MP, Wheeler J, Davies RJ, et al. A systematic review of the literature assessing operative blood loss and postoperative outcomes after colorectal surgery. Int J Colorectal Dis . 2022;37:47-69. Mallard SR, Clifford KA, Park R, Trainee Intern Research Group, Cousins K, Patton A, et al. Role for colorectal teams to support non-colorectal teams to improve clinical outcomes and adherence to ERAS guidelines for segmental colectomy: a cohort study. BMC Surg . 2021;21:132. Wang X, Naito Y, Nakatani H, Ida M, Kawaguchi M. Prevalence of undernutrition in surgical patients and the effect on length of hospital stay. J Anesth . 2022;36:89-95. den Bakker CM, Schaafsma FG, Consten ECJ, Schraffordt Koops SE, van der Meij E, van de Ven PM, et al. Personalised electronic health programme for recovery after major abdominal surgery: a multicentre, single-blind, randomised, placebo-controlled trial. Lancet Digit Health . 2023;5:e485-94. Zhou YY, Zhang BK, Ran TF, Ke S, Ma TY, Qin YY, et al. Education level has an effect on the recovery of total knee arthroplasty: a retrospective study. BMC Musculoskelet Disord . 2022;23:1072. Consensus Conference Panel, Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, et al. Joint consensus statement of the American Academy of Sleep Medicine and sleep research society on the recommended amount of sleep for a healthy adult: methodology and discussion. Sleep . 2015;38:1161-83. Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Arch Gen Psychiatry . 2002;59:131-6. Xu W, Tan CC, Zou JJ, Cao XP, Tan L. Sleep problems and risk of all-cause cognitive decline or dementia: an updated systematic review and meta-analysis. J Neurol Neurosurg Psychiatry . 2020;91:236-44. Ding Z, Li J, Xu B, Cao J, Li H, Zhou Z. Preoperative high sleep quality predicts further decrease in length of stay after total joint arthroplasty under enhanced recovery short-stay program: experience in 604 patients from a single team. Orthop Surg . 2022;14:1989-97. Squires MH, Dann GC, Lad NL, Fisher SB, Martin BM, Kooby DA, et al. Hypophosphataemia after major hepatectomy and the risk of post-operative hepatic insufficiency and mortality: an analysis of 719 patients. HPB (Oxford) . 2014;16:884-91. Hallet J, Karanicolas PJ, Zih FSW, Cheng E, Wong J, Hanna S, et al. Hypophosphatemia and recovery of post-hepatectomy liver insufficiency. Hepatobiliary Surg Nutr . 2016;5:217-24. Mueller JL, Chang DC, Fernandez-Del Castillo CC, Ferrone CR, Warshaw AL, Lillemoe KD, et al. Lower phosphate levels following pancreatectomy is associated with postoperative pancreatic fistula formation. HPB (Oxford) . 2019;21:834-40. Wesselink EM, Kappen TH, Torn HM, Slooter AJC, van Klei WA. Intraoperative hypotension and the risk of postoperative adverse outcomes: a systematic review. Br J Anaesth . 2018;121:706-21. Wesselink EM, Wagemakers SH, van Waes JAR, Wanderer JP, van Klei WA, Kappen TH. Associations between intraoperative hypotension, duration of surgery and postoperative myocardial injury after noncardiac surgery: a retrospective single-centre cohort study. Br J Anaesth . 2022;129:487-96. Liem VGB, Hoeks SE, Mol KHJM, Potters JW, Grüne F, Stolker RJ, et al. Postoperative hypotension after noncardiac surgery and the association with myocardial injury. Anesthesiology . 2020;133:510-22. Hasselgren E, Zdolsek M, Zdolsek JH, Björne H, Krizhanovskii C, Ntika S, et al. Long intravascular persistence of 20% albumin in postoperative patients. Anesth Analg . 2019;129:1232-9. Opperer M, Poeran J, Rasul R, Mazumdar M, Memtsoudis SG. Use of perioperative hydroxyethyl starch 6% and albumin 5% in elective joint arthroplasty and association with adverse outcomes: a retrospective population based analysis. BMJ . 2015;350:h1567. Additional Declarations No competing interests reported. Supplementary Files Additionalfile1.doc Additionalfile2.doc Additionalfile3.doc Additionalfile4.doc Additionalfile5.doc Additionalfile6.doc Cite Share Download PDF Status: Published Journal Publication published 08 Oct, 2025 Read the published version in BMC Gastroenterology → Version 1 posted Editorial decision: Revision requested 20 Jun, 2025 Reviews received at journal 19 Jun, 2025 Reviewers agreed at journal 19 Jun, 2025 Reviews received at journal 18 Jun, 2025 Reviewers agreed at journal 15 Jun, 2025 Reviewers invited by journal 11 Jun, 2025 Editor assigned by journal 11 Jun, 2025 Editor invited by journal 19 May, 2025 Submission checks completed at journal 18 May, 2025 First submitted to journal 18 May, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6659561","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":471587857,"identity":"e4be53b7-ac0a-46e2-a575-b8bf055a7a50","order_by":0,"name":"Jiankun Shi","email":"","orcid":"","institution":"Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Jiankun","middleName":"","lastName":"Shi","suffix":""},{"id":471587860,"identity":"56771ef6-a9b4-48c9-8e8f-9617d5192e04","order_by":1,"name":"Yabin Huang","email":"","orcid":"","institution":"Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Yabin","middleName":"","lastName":"Huang","suffix":""},{"id":471587861,"identity":"bcdec8fc-47e4-4116-8e77-664a715ba912","order_by":2,"name":"Jiaxin Han","email":"","orcid":"","institution":"Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Jiaxin","middleName":"","lastName":"Han","suffix":""},{"id":471587865,"identity":"47e4f3fb-7d50-4b6e-9766-a730723c1004","order_by":3,"name":"Shimin Zhang","email":"","orcid":"","institution":"Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Shimin","middleName":"","lastName":"Zhang","suffix":""},{"id":471587866,"identity":"29238f63-6a09-4343-9737-9143a43c6418","order_by":4,"name":"Xingshan Cheng","email":"","orcid":"","institution":"Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Xingshan","middleName":"","lastName":"Cheng","suffix":""},{"id":471587868,"identity":"02bdcba3-90e1-42b3-961e-6a226250a883","order_by":5,"name":"Tongfeng Luo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA80lEQVRIiWNgGAWjYBACPhBR8cOGgYGZuQEmaIBXCxsDA2PD2Z40oBZGUrScYTvMAKKJ1CKRfPzBAZ7z0fztjA2MP9vqEhvYm7dJMNTcwaMlLbHhgMXt3BmHGRuYedsOJzbwHCuTYDj2DI+WHMPmDzy3cxtAWhjbDiQ2SOSYSTA2HMarpeEA27nc+YdhDpN/Q5SWA7kbgFoYeNuYgbbwENDC8yxxxsGe5NyNQC2Hec4dNm7jSSu2SDiGWws/e/KBDwd+2OXOO3/44MMfZXWy/eyHN974UINbCwo4wMgGjikGhgTiNIDAH+KVjoJRMApGwcgBAIDmV/dJLXtdAAAAAElFTkSuQmCC","orcid":"","institution":"Sixth Affiliated Hospital of Sun Yat-sen University","correspondingAuthor":true,"prefix":"","firstName":"Tongfeng","middleName":"","lastName":"Luo","suffix":""}],"badges":[],"createdAt":"2025-05-14 02:38:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6659561/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6659561/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12876-025-04330-y","type":"published","date":"2025-10-08T15:57:11+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":84819158,"identity":"383c5054-4e40-4298-8bd3-e43cfe92cfd3","added_by":"auto","created_at":"2025-06-17 15:57:37","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":248493,"visible":true,"origin":"","legend":"\u003cp\u003eA flowchart visualization of the study design.\u003c/p\u003e","description":"","filename":"fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/48fcf1975dc89ba5b123ec60.jpg"},{"id":84817466,"identity":"303c67b3-4d5f-481e-b742-7bca111e416a","added_by":"auto","created_at":"2025-06-17 15:49:37","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":296505,"visible":true,"origin":"","legend":"\u003cp\u003eROC and DCA curves of the six machine learning models in the training and validation cohorts. A, B, the ROC curves of different models in the training (A) and validation (B) cohorts. C, D, the DCA curves of different models in the training (C) and validation (D) cohorts. ROC curve, receiver operating characteristic curve; DCA, decision curve analysis; Logistic, logistic regression; KNN, K-nearest neighbor; GBM, gradient boosting machine; RF, random forest; SVM, support vector machine; XGBoost, extreme gradient boosting; AUC, area under the receiver operating characteristic curve.\u003c/p\u003e","description":"","filename":"fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/14fa0854e3ff3426c3f9752c.jpg"},{"id":84819160,"identity":"bc0cac31-3020-46b6-86ca-f1181398b764","added_by":"auto","created_at":"2025-06-17 15:57:37","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":280814,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram for estimating prolonged length of stay after laparoscopic gastrointestinal surgery in women. \u003csup\u003ea\u003c/sup\u003eEducation level was categorized as low (no education or primary school only), medium (only secondary school), and high level (professional education or university). \u003csup\u003eb\u003c/sup\u003ePreoperative insufficient sleep was defined as sleep duration \u0026lt;4 h on the night before surgery. \u003csup\u003ec\u003c/sup\u003ePostoperative hypotension was defined as mean arterial pressure \u0026lt;65 mmHg. \u003csup\u003ed\u003c/sup\u003ePostoperative pain score (73–120 h) meant the highest pain score (numeric rating scale) at mobility during 73–120 h after surgery. \u003csup\u003ee\u003c/sup\u003ePostoperative infection consisted of surgical site infection, anastomotic leak, abdominal abscess, pelvic abscess, hepatic abscess, central venous catheter-related infection, urinary tract infection, or pneumonia.\u003c/p\u003e","description":"","filename":"fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/fbdc5dadae7320415e13e08a.jpg"},{"id":93419709,"identity":"fe392c90-2937-4fa4-b0c4-a3182a7927d3","added_by":"auto","created_at":"2025-10-13 16:06:19","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1955727,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/320e2298-4da8-4c25-8cd8-53578298b0ec.pdf"},{"id":84817469,"identity":"2e783c78-0145-4bf4-a1df-5a73df75ca9c","added_by":"auto","created_at":"2025-06-17 15:49:37","extension":"doc","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":309129,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile1.doc","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/76c8a58b04306742cdebc665.doc"},{"id":84819157,"identity":"821e2331-74a7-40d8-878d-a1b78386ccfe","added_by":"auto","created_at":"2025-06-17 15:57:37","extension":"doc","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":115200,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile2.doc","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/eecca69948b42ce4d0d40c51.doc"},{"id":84819159,"identity":"8f58d45c-e0d4-437b-91f8-c27a89a99161","added_by":"auto","created_at":"2025-06-17 15:57:37","extension":"doc","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":113152,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile3.doc","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/b5468a9d392f7b2aecd86884.doc"},{"id":84817476,"identity":"8420b7ab-c222-42fb-93e9-4275641c027c","added_by":"auto","created_at":"2025-06-17 15:49:37","extension":"doc","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":40448,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile4.doc","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/26bff9d387dd03464687b82b.doc"},{"id":84817480,"identity":"7561a5b4-171c-46c5-9b21-b6ec7c41141e","added_by":"auto","created_at":"2025-06-17 15:49:37","extension":"doc","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":50176,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile5.doc","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/9be227cbe3ceeeab50f5c47e.doc"},{"id":84819978,"identity":"08b55ac5-d758-4489-87e1-bbb7295ef5d2","added_by":"auto","created_at":"2025-06-17 16:05:37","extension":"doc","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":104448,"visible":true,"origin":"","legend":"","description":"","filename":"Additionalfile6.doc","url":"https://assets-eu.researchsquare.com/files/rs-6659561/v1/73f8365278f92f4dc50098bd.doc"}],"financialInterests":"No competing interests reported.","formattedTitle":"Development and validation of a machine learning-based risk model for prolonged length of stay after laparoscopic gastrointestinal surgery in women: A secondary analysis of the FDP-PONV trail","fulltext":[{"header":"1. Background","content":"\u003cp\u003eLaparoscopic gastrointestinal surgery is frequently performed to remove lesions and improve survival in patients with gastrointestinal cancer or benign disease. It is particularly relevant as colorectal cancer and stomach cancer were ranked among the top five most common cancers in the Global Cancer Statistics 2022 report [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Furthermore, the laparoscopic approach was associated with a decreased risk for all adverse outcomes [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The hospital length of stay (LOS) is an important quality metric for surgery, particularly in the era of enhanced recovery after surgery. Prolonged length of stay (PLOS) delays discharge, increases the risk of hospital-acquired infections, predicts a greater risk of short-term mortality, increases the use of medical resources (resulting in higher costs), and affects patient flow and access to healthcare [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDue to the clinical importance of LOS, studies have been undertaken globally to evaluate the factors associated with PLOS, and risk factors, such as age, sex, comorbidity, preoperative patient-reported outcomes, and laboratory test results (such as albumin and hemoglobin) have been reported [\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. However, most studies used a small number of subjects, did not evaluate a wide variety of clinical factors, or used information from clinical medical records, which might not be as accurate or complete as the information collected in prospective clinical trials. Additionally, the factors influencing LOS varied from surgery and population [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e] and should be investigated in specific situations. Women live longer than men but spend more time experiencing ill health and disability [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. It is essential to highlight the health of women as an important part of global health [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Previous studies have shown gender differences in treatment effect, and compared with men, women are at increased risk for postoperative complications, including PLOS and surgical site infection [\u003cspan additionalcitationids=\"CR13 CR14\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Therefore, the identification of specific factors associated with PLOS in women is important. However, to date, little research has been conducted on PLOS in women after laparoscopic gastrointestinal surgery.\u003c/p\u003e \u003cp\u003eCurrently, artificial intelligence can be used to analyze large clinical datasets to improve patient care and may help address challenges unique to the field of global health. Machine learning is one of the most common artificial intelligence technologies used for complex problem-solving, cost-benefit evaluations, and risk assessment [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Therefore, this study aimed to develop and validate a machine learning-based risk model for PLOS after laparoscopic gastrointestinal surgery in women, using comprehensive clinical features during the perioperative period, including the pre-, intra-, and postoperative periods.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study population\u003c/h2\u003e \u003cp\u003eThe study was a secondary analysis of the FDP-PONV trial, a randomized, controlled, double-blind study designed to evaluate the efficacy and safety of adding fosaprepitant to dexamethasone and palonosetron for preventing postoperative nausea and vomiting in patients undergoing laparoscopic gastrointestinal surgery [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The FDP-PONV trial was approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2021ZSLYEC-78) and registered at clinicaltrials.gov\u003c/p\u003e \u003cp\u003e(NCT04853147). The trial was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients before their participation. The inclusion criteria of the FDP-PONV trial were age 18\u0026ndash;75 years, scheduled for laparoscopic gastrointestinal surgery, and three or four Apfel risk factors [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Exclusion criteria included an American Society of Anesthesiologists (ASA) physical status of 4\u0026ndash;6, severe hepatic dysfunction; contraindications to fosaprepitant, 5-hydroxytryptamine type-3 receptor antagonist, or dexamethasone; preoperative use of medications with known antiemetic properties; mental disorders or inability to communicate; and pregnant women or nursing mothers. Female patients who underwent laparoscopic gastrointestinal surgery in the FDP-PONV trial were included in this study. This study was also approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2024ZSLYEC-160) on 3 April 2024.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Outcome parameters\u003c/h2\u003e \u003cp\u003eThe primary outcome was PLOS, defined as a LOS greater than the median in this study. The LOS was defined as the time from the date of operation completion to the date of hospital discharge. The criteria for hospital discharge included [\u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]: 1) tolerance of diet with no need for intravenous nutrition; 2) analgesic-free status, defined as a visual analog scale score\u0026thinsp;\u0026le;\u0026thinsp;3 without intravenous analgesic drugs; 3) adequate mobility; and 4) afebrile status without major infectious complications.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Predictive variables associated with PLOS\u003c/h2\u003e \u003cp\u003eAll 96 variables, including baseline characteristics, intraoperative information, postoperative outcomes, and adverse events prospectively collected in the FDP-PONV trial throughout the pre-, intra-, and postoperative periods were considered potential predictors (Additional file 1).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Sample size of machine learning model\u003c/h2\u003e \u003cp\u003eTo ensure accurate estimates and reduce the risk of overfitting, we adopted a methodology outlined in the British Medical Journal for calculating the sample size required to develop a clinical prediction model [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. This approach was grounded in the principle of events per variable (EPV), the aggregate number of participants, the outcome prevalence within the study population, and the anticipated predictive accuracy of the model. The minimum required total sample size for 10 candidate predictors was determined to be 990, with an events per variable (EPV) of 49.7 (events fraction, 0.5, criterion value rMAPE, 0.05), based on calculations available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mvansmeden.shinyapps.io/BeyondEPV/\u003c/span\u003e\u003cspan address=\"https://mvansmeden.shinyapps.io/BeyondEPV/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Consequently, a sample size of 1,101 patients proved adequate to sustain the development of this model with 10 predictors through the application of machine learning.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Statistical analysis\u003c/h2\u003e \u003cp\u003eA multiple interpolation method was used to fill the missing values. The data are presented as means\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviations, medians [interquartile ranges], or numbers (percentages), as appropriate. Student's t-test for independent samples was employed to compare normally distributed continuous variables that exhibited equal variances across groups. The Wilcoxon rank-sum test was employed to analyze ranked variables, continuous variables with non-normal distributions, and variables exhibiting unequal variances across groups. Categorical data were analyzed using Pearson's chi-square test or, when appropriate, Fisher's exact test. Statistical significance was established at a two-sided p-value of less than 0.05. Prediction models were created by randomly splitting the dataset into a training set containing 70% of the patients and a testing set containing 30% of the patients. Least absolute shrinkage and selection operator (LASSO) was employed to identify variables featuring clinically significant attributes, alongside determining the optimal lambda value. Stepwise regression, guided by the Akaike Information Criterion (AIC), was employed to identify the final variables for inclusion in the model. Six machine learning algorithms\u0026mdash;logistic regression (LR), K-nearest neighbor (KNN), gradient boosting machine (GBM), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost)\u0026mdash;were utilized within R version 4.2.2 to construct risk models for PLOS. Hyperparameters for each model were optimized via grid search and 10-fold cross-validation. The performance of these models was evaluated in both training and validation cohorts, employing metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, Brier score, and calibration curve. Decision curve analysis (DCA) was applied to ascertain the net benefit of the models. Furthermore, an internal validation process involving 1000 bootstrap re-samples was conducted to ensure the predictive model's robustness. A nomogram derived from the logistic regression model was developed as a practical clinical tool.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Patients and datasets\u003c/h2\u003e \u003cp\u003eIn total, 1154 patients were randomly assigned to the FDP-PONV trial between April 2021 and April 2022. After excluding 40 male patients and 13 patients who did not undergo laparoscopic gastrointestinal surgery, 1101 female patients who underwent laparoscopic gastrointestinal surgery were included in the final dataset (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The median LOS of 1101 patients enrolled in this study was 7 days, and PLOS (LOS\u0026thinsp;\u0026gt;\u0026thinsp;7 days) occurred in 535 patients (48.6%). The randomly split training (770 patients) and test (331 patients) sets are shown in Additional file 2.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Model development by machine learning method\u003c/h2\u003e \u003cp\u003eFirst, LASSO was used to select features after choosing an optimal lambda value of 0.01928917. Features with a coefficient of zero were considered redundant and removed, leaving 31 features for downstream stepwise regression. The details of feature selection using LASSO are shown in Additional file 3 and Additional file 4. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e compares predictor variables selected by LASSO between patients with and without PLOS. Stepwise regression based on the AIC was used to further select the features. Finally, considering clinical significance and sample size, ten features were retained for inclusion in the models (Additional file 5).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePotential predictors selected by least absolute shrinkage and selection operator regression for PLOS\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOverall (N\u0026thinsp;=\u0026thinsp;1101)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon-PLOS (N\u0026thinsp;=\u0026thinsp;566)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003ePLOS (N\u0026thinsp;=\u0026thinsp;535)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUse of fosaprepitant\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.07\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e550 (50.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e298 (52.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e252 (47.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e551 (50.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e268 (47.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e283 (52.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducation level\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLow\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e710 (64.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e330 (58.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e380 (71.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMedium\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e196 (17.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e107 (18.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e89 (16.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e195 (17.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e129 (22.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e66 (12.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e127 (11.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e64 (11.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e63 (11.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHistory of diabetes mellitus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.05\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e999 (90.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e523 (92.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e476 (89.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e102 (9.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e43 (7.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e59 (11.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePreoperative insufficient sleep\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e787 (71.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e424 (74.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e363 (67.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e314 (28.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e142 (25.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e172 (32.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePreoperative hypoalbuminemia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e903 (82.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e492 (86.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e411 (76.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e198 (18.0%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e74 (13.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e124 (23.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePreoperative serum calcium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.25 (2.18, 2.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.26 (2.19, 2.34)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.25 (2.18, 2.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePreoperative hemoglobin (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e117.00 (103.00, 127.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e119.00 (105.00, 128.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e115.00 (98.00, 126.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDuration of surgery (min)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e185.00 (146.00, 243.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e166.00 (133.00, 208.75)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e212.00 (164.50, 272.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDose of propofol (mg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,370.00 (1,070.00, 1,760.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1,220.00 (1,000.00, 1,509.25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1,510.00 (1,180.00, 1,925.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUse of neostigmine\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.40\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e408 (37.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e203 (35.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e205 (38.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e693 (62.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e363 (64.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e330 (61.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInfusion of fluid in operating room (mL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2,450.00 (2,250.00, 2,850.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2,350.00 (1,975.00, 2,750.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2,750.00 (2,250.00, 2,950.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIntraoperative blood loss (mL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e50.00 (50.00, 100.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e50.00 (30.00, 50.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e50.00 (50.00, 100.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative hypotension\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e959 (87.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e511 (90.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e448 (83.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e142 (12.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e55 (9.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e87 (16.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative opioids consumption (morphine equivalents, mg)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e38.28 (30.80, 46.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e38.00 (30.13, 46.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e38.56 (31.36, 46.67)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative albumin infusion\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e169 (15.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e123 (21.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e46 (8.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e932 (84.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e443 (78.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e489 (91.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative hypoalbuminemia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e310 (28.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e190 (33.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e120 (22.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e791 (71.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e376 (66.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e415 (77.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative fasting blood glucose (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e8.25 (6.95, 10.30)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7.99 (6.85, 9.86)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e8.60 (7.10, 10.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative serum calcium (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.10\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.12\u0026thinsp;\u0026plusmn;\u0026thinsp;0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2.09\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative serum phosphorus (mmol/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.20 (1.05, 1.33)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.20 (1.05, 1.32)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.20 (1.05, 1.34)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.79\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative hemoglobin (g/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e106.00 (92.00, 116.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e108.00 (96.00, 117.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e103.00 (89.00, 114.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative monocyte count (*10\u003csup\u003e9\u003c/sup\u003e/L)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.40 (0.23, 0.61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.34 (0.19, 0.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.45 (0.29, 0.64)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative nausea and vomiting (0\u0026ndash;24 h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.19\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e647 (58.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e322 (56.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e325 (60.7%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e454 (41.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e244 (43.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e210 (39.3%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative nausea (73-120h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e863 (78.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e475 (83.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e388 (72.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e238 (21.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e91 (16.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e147 (27.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative vomiting (0-120 h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.92\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e729 (66.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e374 (66.1%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e355 (66.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e372 (33.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e192 (33.9%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e180 (33.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative pain score\u003csup\u003ed\u003c/sup\u003e (0\u0026ndash;24 h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e4.00 (3.00, 5.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e4.00 (3.00, 5.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e4.00 (3.00, 5.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative pain score\u003csup\u003ed\u003c/sup\u003e (73\u0026ndash;120 h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2.00 (2.00, 3.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e2.00 (2.00, 3.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e3.00 (2.00, 4.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime to first flatus after surgery (h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e48.00 (30.00, 68.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e44.00 (25.00, 62.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e54.00 (38.00, 75.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime to first defaecation after surgery (h)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e70.00 (46.00, 99.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e66.50 (43.00, 91.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e76.00 (50.00, 112.50)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative bleeding\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,081 (98.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e563 (99.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e518 (96.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e20 (1.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3 (0.5%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e17 (3.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePostoperative infection\u003csup\u003ee\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,033 (93.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e559 (98.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e474 (88.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e68 (6.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e7 (1.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e61 (11.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAnastomotic leak\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1,070 (97.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e565 (99.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e505 (94.4%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYes\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e31 (2.8%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1 (0.2%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e30 (5.6%)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003ePLOS, prolonged length of stay after surgery.\u003c/td\u003e\u003c/tr\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003csup\u003ea\u003c/sup\u003eEducation level was categorized as low (no education or primary school only), medium (only secondary school), and high level (professional education or university). \u003csup\u003eb\u003c/sup\u003ePreoperative insufficient sleep was defined as sleep duration\u0026thinsp;\u0026lt;\u0026thinsp;4 h on the night before surgery. \u003csup\u003ec\u003c/sup\u003ePostoperative hypotension was defined as mean arterial pressure\u0026thinsp;\u0026lt;\u0026thinsp;65 mmHg. \u003csup\u003ed\u003c/sup\u003ePostoperative pain score meant the highest pain score (numeric rating scale) at mobility. \u003csup\u003ee\u003c/sup\u003ePostoperative infection consists of surgical site infection, anastomotic leak, abdominal abscess, pelvic abscess, hepatic abscess, central venous catheter-related infection, urinary tract infection or pneumonia.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe AUCs for the six models ranged from 0.689 to 0.751, and the receiver operating characteristic (ROC) curves are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes the performance of each model. The logistic regression model performed best, achieving an AUC of 0.751 (95% CI, 0.699\u0026ndash;0.804), with 0.713 (95% CI, 0.712\u0026ndash;0.714) accuracy, 0.829 sensitivity, 0.607 specificity, 0.658 positive predictive value, 0.795 negative predictive value and 0.734 F1 score on the validation cohort. The Brier score of the logistic model for predicting PLOS was 0.205 in the validation cohort. The bootstrapping method by 1000 times sampling on the whole data set (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e) also showed that the logistic regression model performed best, achieving an AUC of 0.760 (95% CI, 0.722\u0026ndash;0.796). These results indicated that the proposed model was reliable. The DCA of the logistic regression model in the validation cohort showed that if the threshold probability of patients was 25\u0026ndash;70%, using this model to predict the risk of PLOS could add more benefit than either the treat-all tactics or the treat-none tactics (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Calibration curve analyses were performed for further model evaluation (Additional file 6).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eModel performance in the training and validation cohorts\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAccuracy (95% CI)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSensitivity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSpecificity\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003ePPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eNPV\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eF1 score\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eBrier\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTraining cohort\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.776 (0.743\u0026ndash;0.808)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.713 (0.712\u0026ndash;0.714)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.700\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.725\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.710\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.673\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.191\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.806 (0.777\u0026ndash;0.836)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.725 (0.724\u0026ndash;0.725)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.653\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.794\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.752\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.704\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.178\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGBM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.835 (0.807\u0026ndash;0.863)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.781 (0.780\u0026ndash;0.781)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.716\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.842\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.813\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.756\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.761\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.164\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.987 (0.981\u0026ndash;0.992)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.945 (0.945\u0026ndash;0.946)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.947\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.947\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.042\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.814 (0.783\u0026ndash;0.844)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.753 (0.753\u0026ndash;0.754)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.743\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.763\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.751\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.756\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.747\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.172\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.821 (0.792\u0026ndash;0.850)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.758 (0.758\u0026ndash;0.759)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.743\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.774\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.759\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.758\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.751\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.172\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eValidation cohort\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.751 (0.699\u0026ndash;0.804)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.713 (0.712\u0026ndash;0.714)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.829\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.607\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.658\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.795\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.734\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.205\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.733 (0.680\u0026ndash;0.787)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.683 (0.681\u0026ndash;0.684)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.627\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.734\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.683\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.683\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.654\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.208\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGBM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.744 (0.692\u0026ndash;0.797)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.695 (0.694\u0026ndash;0.696)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.696\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.694\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.675\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.714\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.685\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.205\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.689 (0.633\u0026ndash;0.745)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.656 (0.656\u0026ndash;0.657)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.677\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.636\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.629\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.683\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.652\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.222\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.733 (0.679\u0026ndash;0.788)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.680 (0.678\u0026ndash;0.681)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.778\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.590\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.634\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.745\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.699\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.208\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.732 (0.678\u0026ndash;0.786)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.686 (0.685\u0026ndash;0.687)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.677\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.694\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0.669\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e \u003cp\u003e0.673\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e \u003cp\u003e0.209\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"9\"\u003eLR, logistic regression; KNN, K-nearest neighbor; GBM, gradient boosting machine; RF, random forest; SVM, support vector machine; XGBoost, extreme gradient boosting; AUC, area under the receiver operating characteristic curve; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eModel effect evaluated by the bootstrap method\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAUC\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLower\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eUpper\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.760\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.722\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.796\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eKNN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.637\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.592\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.679\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGBM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.781\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.661\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.619\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.702\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSVM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.732\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.692\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.770\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXGBoost\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.742\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.705\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.780\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.02\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eLR, logistic regression; KNN, K-nearest neighbor; GBM, gradient boosting machine; RF, random forest; SVM, support vector machine; XGBoost, extreme gradient boosting; AUC, area under the receiver operating characteristic curve.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Nomogram for PLOS after laparoscopic gastrointestinal surgery\u003c/h2\u003e \u003cp\u003eAs the logistic regression model performed well, it was transformed into a nomogram for easier use in different clinical situations. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, ten factors\u0026mdash;including education level, preoperative hypoalbuminemia, preoperative insufficient sleep (sleep duration\u0026thinsp;\u0026lt;\u0026thinsp;4 h), duration of surgery, intraoperative blood loss, postoperative hypotension (mean arterial pressure\u0026thinsp;\u0026lt;\u0026thinsp;65 mmHg), postoperative albumin infusion, postoperative serum phosphorus, highest pain score at mobility during 73\u0026ndash;120 h after surgery, and postoperative infection\u0026mdash;were screened to form the final nomogram.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eIn this study, we used perioperative data from the FDP-PONV trial of 1101 female patients undergoing laparoscopic gastrointestinal surgery to develop and validate six machine learning-based risk models for PLOS. Three models (LR, GBM, and XGBoost) achieved AUC values\u0026thinsp;\u0026gt;\u0026thinsp;0.741 in internal validation by bootstrapping 1000 times, with the potential to help in the clinical prevention and management of PLOS. The LR model achieved the best discrimination, with an AUC of 0.760, and was converted into a nomogram to enhance its potential clinical applicability.\u003c/p\u003e \u003cp\u003eMachine learning algorithms allow computers to learn from numerous data to provide evidence to clinicians and play a considerable role in clinical decision-making. Risk factors associated with PLOS are complex and may exist in every link during hospitalization. This study used 96 variables prospectively collected in the FDP-PONV trial throughout the pre-, intra-, and postoperative periods to expand the coverage as widely as possible, different from previous studies focusing on preoperative factors or searching clinical medical records retrospectively [\u003cspan additionalcitationids=\"CR24\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Risk models were established based on three preoperative, two intraoperative, and five postoperative factors. Among these, five classic factors, including preoperative hypoalbuminemia, duration of surgery, intraoperative blood loss, postoperative pain score, and postoperative infection, were reported to be associated with PLOS in many previous studies [\u003cspan additionalcitationids=\"CR27 CR28 CR29\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Additionally, five relatively new and interesting factors were identified: education level, postoperative sleep duration, postoperative serum phosphorus, postoperative hypotension, and postoperative albumin infusion.\u003c/p\u003e \u003cp\u003eCompared to the high level of education (professional education or university), low (no education or primary school only), and medium (only secondary school) levels of education [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] were related to the occurrence of PLOS in the study. With the promotion of ERAS, the level of education is increasingly important in recovery since it affects the patients\u0026rsquo; comprehension of the strategies for rehabilitation and their compliance [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Sleep is vital for maintaining healthy mental, physical, and metabolic states [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Severe insomnia (sleep duration\u0026thinsp;\u0026lt;\u0026thinsp;3.5 h) in women was associated with an increase in mortality [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], insufficient sleep duration (\u0026lt;\u0026thinsp;4 h) could elevate the risk of cognitive disorders [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e], and higher preoperative sleep quality induced a shorter stay after total joint arthroplasty [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Based on previous studies and the collected data, this study offered a cut-off value of sleep duration on the night before surgery and found that a sleep duration of \u0026lt;\u0026thinsp;4 h was a risk factor for PLOS in women. As usual, the serum phosphorus level after surgery did not attract as much attention as potassium or sodium levels. A few previous studies have shown that hypophosphatemia following hepatectomy is associated with decreased morbidity and mortality [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], but hypophosphatemia following pancreatectomy may be associated with postoperative pancreatic fistula [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e], suggesting that the clinical significance of serum phosphorus varies with the surgical type. This study used data on several electrolytes, including serum potassium, sodium, calcium, and phosphorus, both before and after surgery for analysis, and only the level of serum phosphorus after surgery was entered into the models as a risk factor for PLOS. Most studies have focused on the relationship between intraoperative hypotension and adverse events [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. This study fully investigated blood pressure events, including intraoperative hypotension and hypertension, as well as postoperative hypotension and hypertension; only postoperative hypotension was selected. This finding was consistent with a cohort study showing that only postoperative hypotension, and not intraoperative hypotension, was associated with myocardial injury in 1710 patients undergoing non-cardiac surgery [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e], suggesting that postoperative hypotension should also be paid more attention to. Although albumin infusion is usually considered to reduce tissue edema [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e], this study found that postoperative albumin infusion was a risk factor for PLOS. A retrospective cohort study of 1051441 patients undergoing elective total hip and knee arthroplasties also found that perioperative fluid resuscitation with albumin 5% was associated with an increased risk of acute renal failure and most other complications [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. These results suggest that after surgery, albumin infusion should be administered as minimally as possible.\u003c/p\u003e \u003cp\u003eOur study had some limitations. First, although data from the FDP-PONV trial were collected carefully and prospectively, a retrospective analysis of PLOS was performed. Second, although a bootstrap testing approach was used to validate our model, further external validation in a wider population is also essential. Third, although 96 variables were considered throughout the perioperative period, some potential risk factors with large missing values, such as C-reactive protein and procalcitonin levels, were not included in this study.\u003c/p\u003e \u003cp\u003eIn conclusion, we successfully developed a simple and reliable tool to predict the occurrence of PLOS after laparoscopic gastrointestinal surgery in women. Such individualized risk analyses allow clinicians to monitor and treat patients in a targeted manner to shorten LOS.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLOS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLength of Stay\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePLOS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eProlonged Length of Stay\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eASA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAmerican Society of Anesthesiologists\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eEPV\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eEvents Per Variable\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLASSO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLeast Absolute Shrinkage and Selection Operator\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAIC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAkaike Information Criterion\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLR\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eLogistic Regression\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eKNN\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eK-Nearest Neighbor\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGBM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGradient Boosting Machine\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eRF\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eRandom Forest\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSVM\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSupport Vector Machine\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eXGBoost\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eExtreme Gradient Boosting\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eArea Under the Receiver Operating Characteristic Curve\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eDCA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eDecision Curve Analysis\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eROC\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eReceiver Operating Characteristic\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe FDP-PONV trial was approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2021ZSLYEC-78). Written informed consent was obtained from all patients before their participation. The FDP-PONV trial was conducted in accordance with the Declaration of Helsinki. The secondary analysis of the FDP-PONV trail was also approved by the Ethics Committee of the Sixth Affiliated Hospital of Sun Yat-sen University (2024ZSLYEC-160).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data of this study are available upon reasonable request to the principal investigator.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStudy design: TL; Data collection: JS, YH, JH, SZ, XC; Statistical analysis: TL, JS, YH, JH; Data interpretation: SZ, XC; Writing of paper: JS, YH, JH; Manuscript revision: TL. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe would like to thank Editage (www.editage.cn) for English language editing.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin\u003cem\u003e. \u003c/em\u003e2024;74:229-63.\u003c/li\u003e\n\u003cli\u003evan den Bosch T, Warps AK, de Ner\u0026eacute;e Tot Babberich MPM, Stamm C, Geerts BF, Vermeulen L, et al. Predictors of 30-day mortality among Dutch patients undergoing colorectal cancer surgery, 2011-2016. JAMA Netw Open\u003cem\u003e. \u003c/em\u003e2021;4:e217737.\u003c/li\u003e\n\u003cli\u003eCai Z, Song H, Huang Z, Fingerhut A, Xu X, Zhong H, et al. Safety and feasibility of laparoscopic surgery for colorectal and gastric cancer under the Chinese multi-site practice policy: admittance standards of competence are needed. Gastroenterol Rep (Oxf)\u003cem\u003e. \u003c/em\u003e2022;10:goac046.\u003c/li\u003e\n\u003cli\u003eRosman M, Rachminov O, Segal O, Segal G. Prolonged patients\u0026apos; In-Hospital Waiting Period after discharge eligibility is associated with increased risk of infection, morbidity and mortality: a retrospective cohort analysis. BMC Health Serv Res\u003cem\u003e. \u003c/em\u003e2015;15:246.\u003c/li\u003e\n\u003cli\u003eGohil R, Rishi M, Tan BHL. Pre-operative serum albumin and neutrophil-lymphocyte ratio are associated with prolonged hospital stay following colorectal cancer surgery. Br J Med Med Res\u003cem\u003e. \u003c/em\u003e2014;4:481-7.\u003c/li\u003e\n\u003cli\u003eNishimuta H, Kusachi S, Watanabe M, Asai K, Kiribayashi T, Niitsuma T, et al. Impact of postoperative remote infection on length of stay and medical costs in hospitals in Japan. Surg Today\u003cem\u003e. \u003c/em\u003e2021;51:212-8.\u003c/li\u003e\n\u003cli\u003eSutherland J, Liu G, Crump T, Bair M, Karimuddin A. Relationship between preoperative patient-reported outcomes and hospital length of stay: a prospective cohort study of general surgery patients in Vancouver, Canada. J Health Serv Res Policy\u003cem\u003e. \u003c/em\u003e2019;24:29-36.\u003c/li\u003e\n\u003cli\u003eZeidan M, Goz V, Lakomkin N, Spina N, Brodke DS, Spiker WR. Predictors of readmission and prolonged length of stay after cervical disc arthroplasty. Spine\u003cem\u003e. \u003c/em\u003e2021;46:487-91.\u003c/li\u003e\n\u003cli\u003eGokhale S, Taylor D, Gill J, Hu Y, Zeps N, Lequertier V, et al. Hospital length of stay prediction for general surgery and total knee arthroplasty admissions: systematic review and meta-analysis of published prediction models. Digit Health\u003cem\u003e. \u003c/em\u003e2023;9:589821209:20552076231177497.\u003c/li\u003e\n\u003cli\u003eConnolly A, Regan DL. Women\u0026apos;s health matters. Br J Gen Pract\u003cem\u003e. \u003c/em\u003e2022;72:556-7.\u003c/li\u003e\n\u003cli\u003eKoplan JP, Bond TC, Merson MH, Reddy KS, Rodriguez MH, Sewankambo NK, et al. Towards a common definition of global health. Lancet\u003cem\u003e. \u003c/em\u003e2009;373:1993-5.\u003c/li\u003e\n\u003cli\u003eToren P, Wilkins A, Patel K, Burley A, Gris T, Kockelbergh R, et al. The sex gap in bladder cancer survival - a missing link in bladder cancer care? Nat Rev Urol\u003cem\u003e. \u003c/em\u003e2024;21:181-92.\u003c/li\u003e\n\u003cli\u003eSeeman MV. Men and women respond differently to antipsychotic drugs. Neuropharmacology\u003cem\u003e. \u003c/em\u003e2020;163:107631.\u003c/li\u003e\n\u003cli\u003eWilson LAB, Zajitschek SRK, Lagisz M, Mason J, Haselimashhadi H, Nakagawa S. Sex differences in allometry for phenotypic traits in mice indicate that females are not scaled males. Nat Commun\u003cem\u003e. \u003c/em\u003e2022;13:7502.\u003c/li\u003e\n\u003cli\u003eHeyer JH, Cao NA, Amdur RL, Rao RR. Postoperative complications following orthopedic spine surgery: is there a difference between men and women? Int J Spine Surg\u003cem\u003e. \u003c/em\u003e2019;13:125-31.\u003c/li\u003e\n\u003cli\u003eSchwalbe N, Wahl B. Artificial intelligence and the future of global health. Lancet\u003cem\u003e. \u003c/em\u003e2020;395:1579-86.\u003c/li\u003e\n\u003cli\u003eHuang Q, Wang F, Liang C, Huang Y, Zhao Y, Liu C, et al. Fosaprepitant for postoperative nausea and vomiting in patients undergoing laparoscopic gastrointestinal surgery: a randomised trial. Br J Anaesth\u003cem\u003e. \u003c/em\u003e2023;131:673-81.\u003c/li\u003e\n\u003cli\u003eApfel CC, L\u0026auml;\u0026auml;r\u0026auml; E, Koivuranta M, Greim CA, Roewer N. A simplified risk score for predicting postoperative nausea and vomiting: conclusions from cross-validations between two centers. Anesthesiology\u003cem\u003e. \u003c/em\u003e1999;91:693-700.\u003c/li\u003e\n\u003cli\u003eFiore JF, Browning L, Bialocerkowski A, Gruen RL, Faragher IG, Denehy L. Hospital discharge criteria following colorectal surgery: a systematic review. Colorectal Dis\u003cem\u003e. \u003c/em\u003e2012;14:270-81.\u003c/li\u003e\n\u003cli\u003eLee TG, Kang SB, Kim DW, Hong S, Heo SC, Park KJ. Comparison of early mobilization and diet rehabilitation program with conventional care after laparoscopic colon surgery: a prospective randomized controlled trial. Dis Colon Rectum\u003cem\u003e. \u003c/em\u003e2011;54:21-8.\u003c/li\u003e\n\u003cli\u003eGuner A, Kim KY, Park SH, Cho M, Kim YM, Hyung WJ, et al. Safe discharge criteria after curative gastrectomy for gastric cancer. J Gastric Cancer\u003cem\u003e. \u003c/em\u003e2022;22:395-407.\u003c/li\u003e\n\u003cli\u003eRiley RD, Ensor J, Snell KIE, Harrell FE, Martin GP, Reitsma JB, et al. Calculating the sample size required for developing a clinical prediction model. BMJ\u003cem\u003e. \u003c/em\u003e2020;368:m441.\u003c/li\u003e\n\u003cli\u003eJo YY, Han J, Park HW, Jung H, Lee JD, Jung J, et al. Prediction of prolonged length of hospital stay after cancer surgery using machine learning on electronic health records: retrospective cross-sectional study. JMIR Med Inform\u003cem\u003e. \u003c/em\u003e2021;9:e23147.\u003c/li\u003e\n\u003cli\u003eTriana AJ, Vyas R, Shah AS, Tiwari V. Predicting length of stay of coronary artery bypass grafting patients using machine learning. J Surg Res\u003cem\u003e. \u003c/em\u003e2021;264:68-75.\u003c/li\u003e\n\u003cli\u003eDaghistani TA, Elshawi R, Sakr S, Ahmed AM, Al-Thwayee A, Al-Mallah MH. Predictors of in-hospital length of stay among cardiac patients: A machine learning approach. Int J Cardiol\u003cem\u003e. \u003c/em\u003e2019;288:140-7.\u003c/li\u003e\n\u003cli\u003eShankar H, Sureshkumar S, Gurushankari B, Samanna Sreenath G, Kate V. Factors predicting prolonged hospitalization after abdominal wall hernia repair - a prospective observational study. Turk J Surg\u003cem\u003e. \u003c/em\u003e2021;37:96-102.\u003c/li\u003e\n\u003cli\u003eLi Z, Li H, Lv P, Peng X, Wu C, Ren J, et al. Prospective multicenter study on the incidence of surgical site infection after emergency abdominal surgery in China. Sci Rep\u003cem\u003e. \u003c/em\u003e2021;11:7794.\u003c/li\u003e\n\u003cli\u003eCharalambides M, Mavrou A, Jennings T, Powar MP, Wheeler J, Davies RJ, et al. A systematic review of the literature assessing operative blood loss and postoperative outcomes after colorectal surgery. Int J Colorectal Dis\u003cem\u003e. \u003c/em\u003e2022;37:47-69.\u003c/li\u003e\n\u003cli\u003eMallard SR, Clifford KA, Park R, Trainee Intern Research Group, Cousins K, Patton A, et al. Role for colorectal teams to support non-colorectal teams to improve clinical outcomes and adherence to ERAS guidelines for segmental colectomy: a cohort study. BMC Surg\u003cem\u003e. \u003c/em\u003e2021;21:132.\u003c/li\u003e\n\u003cli\u003eWang X, Naito Y, Nakatani H, Ida M, Kawaguchi M. Prevalence of undernutrition in surgical patients and the effect on length of hospital stay. J Anesth\u003cem\u003e. \u003c/em\u003e2022;36:89-95.\u003c/li\u003e\n\u003cli\u003eden Bakker CM, Schaafsma FG, Consten ECJ, Schraffordt Koops SE, van der Meij E, van de Ven PM, et al. Personalised electronic health programme for recovery after major abdominal surgery: a multicentre, single-blind, randomised, placebo-controlled trial. Lancet Digit Health\u003cem\u003e. \u003c/em\u003e2023;5:e485-94.\u003c/li\u003e\n\u003cli\u003eZhou YY, Zhang BK, Ran TF, Ke S, Ma TY, Qin YY, et al. Education level has an effect on the recovery of total knee arthroplasty: a retrospective study. BMC Musculoskelet Disord\u003cem\u003e. \u003c/em\u003e2022;23:1072.\u003c/li\u003e\n\u003cli\u003eConsensus Conference Panel, Watson NF, Badr MS, Belenky G, Bliwise DL, Buxton OM, et al. Joint consensus statement of the American Academy of Sleep Medicine and sleep research society on the recommended amount of sleep for a healthy adult: methodology and discussion. Sleep\u003cem\u003e. \u003c/em\u003e2015;38:1161-83.\u003c/li\u003e\n\u003cli\u003eKripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality associated with sleep duration and insomnia. Arch Gen Psychiatry\u003cem\u003e. \u003c/em\u003e2002;59:131-6.\u003c/li\u003e\n\u003cli\u003eXu W, Tan CC, Zou JJ, Cao XP, Tan L. Sleep problems and risk of all-cause cognitive decline or dementia: an updated systematic review and meta-analysis. J Neurol Neurosurg Psychiatry\u003cem\u003e. \u003c/em\u003e2020;91:236-44.\u003c/li\u003e\n\u003cli\u003eDing Z, Li J, Xu B, Cao J, Li H, Zhou Z. Preoperative high sleep quality predicts further decrease in length of stay after total joint arthroplasty under enhanced recovery short-stay program: experience in 604 patients from a single team. Orthop Surg\u003cem\u003e. \u003c/em\u003e2022;14:1989-97.\u003c/li\u003e\n\u003cli\u003eSquires MH, Dann GC, Lad NL, Fisher SB, Martin BM, Kooby DA, et al. Hypophosphataemia after major hepatectomy and the risk of post-operative hepatic insufficiency and mortality: an analysis of 719 patients. HPB (Oxford)\u003cem\u003e. \u003c/em\u003e2014;16:884-91.\u003c/li\u003e\n\u003cli\u003eHallet J, Karanicolas PJ, Zih FSW, Cheng E, Wong J, Hanna S, et al. Hypophosphatemia and recovery of post-hepatectomy liver insufficiency. Hepatobiliary Surg Nutr\u003cem\u003e. \u003c/em\u003e2016;5:217-24.\u003c/li\u003e\n\u003cli\u003eMueller JL, Chang DC, Fernandez-Del Castillo CC, Ferrone CR, Warshaw AL, Lillemoe KD, et al. Lower phosphate levels following pancreatectomy is associated with postoperative pancreatic fistula formation. HPB (Oxford)\u003cem\u003e. \u003c/em\u003e2019;21:834-40.\u003c/li\u003e\n\u003cli\u003eWesselink EM, Kappen TH, Torn HM, Slooter AJC, van Klei WA. Intraoperative hypotension and the risk of postoperative adverse outcomes: a systematic review. Br J Anaesth\u003cem\u003e. \u003c/em\u003e2018;121:706-21.\u003c/li\u003e\n\u003cli\u003eWesselink EM, Wagemakers SH, van Waes JAR, Wanderer JP, van Klei WA, Kappen TH. Associations between intraoperative hypotension, duration of surgery and postoperative myocardial injury after noncardiac surgery: a retrospective single-centre cohort study. Br J Anaesth\u003cem\u003e. \u003c/em\u003e2022;129:487-96.\u003c/li\u003e\n\u003cli\u003eLiem VGB, Hoeks SE, Mol KHJM, Potters JW, Gr\u0026uuml;ne F, Stolker RJ, et al. Postoperative hypotension after noncardiac surgery and the association with myocardial injury. Anesthesiology\u003cem\u003e. \u003c/em\u003e2020;133:510-22.\u003c/li\u003e\n\u003cli\u003eHasselgren E, Zdolsek M, Zdolsek JH, Bj\u0026ouml;rne H, Krizhanovskii C, Ntika S, et al. Long intravascular persistence of 20% albumin in postoperative patients. Anesth Analg\u003cem\u003e. \u003c/em\u003e2019;129:1232-9.\u003c/li\u003e\n\u003cli\u003eOpperer M, Poeran J, Rasul R, Mazumdar M, Memtsoudis SG. Use of perioperative hydroxyethyl starch 6% and albumin 5% in elective joint arthroplasty and association with adverse outcomes: a retrospective population based analysis. BMJ\u003cem\u003e. \u003c/em\u003e2015;350:h1567.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-gastroenterology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmge","sideBox":"Learn more about [BMC Gastroenterology](http://bmcgastroenterol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmge/default.aspx","title":"BMC Gastroenterology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"length of hospital stay, gastrointestinal surgery, women","lastPublishedDoi":"10.21203/rs.3.rs-6659561/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6659561/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eProlonged length of stay (PLOS) after surgery is associated with several clinical risks and increased medical costs. We aimed to develop a risk model for PLOS based on clinical features throughout pre-, intra-, and post-operative periods in women undergoing laparoscopic gastrointestinal surgery.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWomen who underwent laparoscopic gastrointestinal surgery in the FDP-PONV randomized controlled trial were eligible for this secondary analysis. PLOS was defined as a duration longer than the median postoperative length of stay. All 96 clinical features prospectively collected in the FDP-PONV trial were used to generate the models. Six machine learning algorithms were employed: logistic regression, K-nearest neighbor, gradient boosting machine, random forest, support vector machine, and extreme gradient boosting. The model performance was assessed using numerous metrics and evaluated using bootstrapping with 1000 replicates.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eIn total, 770 and 331 patients were assigned to the training and validation cohorts, respectively. The logistic regression model performed best in the validation cohort [area under the receiver operating characteristic curve (AUC) 0.751, 95% confidence interval (CI), 0.699\u0026ndash;0.804] and the bootstrapping (AUC 0.760, 95% CI 0.722\u0026ndash;0.796). A nomogram based on ten factors, including education level, preoperative hypoalbuminemia, preoperative insufficient sleep, duration of surgery, blood loss, postoperative hypotension, postoperative albumin infusion, postoperative serum phosphorus level, highest pain score during 73\u0026ndash;120 h after surgery, and postoperative infection, was established.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eA risk model was established using machine learning to characterize the key perioperative drivers and identify women at high risk of PLOS after laparoscopic gastrointestinal surgery.\u003c/p\u003e\u003ch2\u003eTrial registration:\u003c/h2\u003e \u003cp\u003eThe FDP-PONV trial was registered at clinicaltrials.gov (NCT04853147) on 2021-04-27.\u003c/p\u003e","manuscriptTitle":"Development and validation of a machine learning-based risk model for prolonged length of stay after laparoscopic gastrointestinal surgery in women: A secondary analysis of the FDP-PONV trail","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-17 15:49:32","doi":"10.21203/rs.3.rs-6659561/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-06-20T05:33:48+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-19T21:54:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"250985104023604695757099369073354544502","date":"2025-06-19T21:53:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-18T21:42:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"78897670715236449363569897070810338244","date":"2025-06-15T22:10:37+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-11T22:05:28+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-11T10:46:52+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-05-19T05:51:02+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-18T11:41:12+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Gastroenterology","date":"2025-05-18T11:40:05+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-gastroenterology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmge","sideBox":"Learn more about [BMC Gastroenterology](http://bmcgastroenterol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bmge/default.aspx","title":"BMC Gastroenterology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"df98df3a-055a-4a70-8ee2-29c28994715a","owner":[],"postedDate":"June 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-10-13T16:01:16+00:00","versionOfRecord":{"articleIdentity":"rs-6659561","link":"https://doi.org/10.1186/s12876-025-04330-y","journal":{"identity":"bmc-gastroenterology","isVorOnly":false,"title":"BMC Gastroenterology"},"publishedOn":"2025-10-08 15:57:11","publishedOnDateReadable":"October 8th, 2025"},"versionCreatedAt":"2025-06-17 15:49:32","video":"","vorDoi":"10.1186/s12876-025-04330-y","vorDoiUrl":"https://doi.org/10.1186/s12876-025-04330-y","workflowStages":[]},"version":"v1","identity":"rs-6659561","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6659561","identity":"rs-6659561","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}