PDF(1566 KB)
基于机器学习的结直肠癌R0切除术后延迟出院风险预测模型构建与验证
韩超, 张人超, 方圆, 项世骏, 裘正军, 黄陈
中国实用外科杂志 ›› 2026, Vol. 46 ›› Issue (9) : 1259-1266.
PDF(1566 KB)
PDF(1566 KB)
基于机器学习的结直肠癌R0切除术后延迟出院风险预测模型构建与验证
Construction and validation of a machine learning-based risk prediction model for delayed discharge after R0 resection of colorectal cancer
目的 基于术前参数，采用机器学习算法构建结直肠癌R0切除术后延迟出院风险预测模型并比较其性能。方法 回顾性分析2015年11月至2021年4月上海交通大学医学院附属第一人民医院接受R0切除术的2048例结直肠癌病人资料，按7∶3随机分为训练集（1433例）和验证集（615例）。采用逻辑回归、支持向量机（SVM）、随机森林和极限梯度提升（XGBoost），基于术前实验室参数及一般特征建模，以受试者工作特征曲线下面积（AUC）、真阳性率（TPR）、真阴性率（TNR）、准确率（ACC）及阳性预测值（PPV）评价性能。结果 术后感染是延迟出院的首要原因，占全部延迟出院原因的40.52%。训练集中XGBoost、随机森林、SVM和逻辑回归模型的AUC分别为0.762、0.772、0.747和0.750，验证集中分别为0.747、0.735、0.729和0.738。XGBoost模型综合性能最佳，其TNR高于SVM模型（P=0.046），TNR（P<0.001）和ACC（P=0.001）均高于逻辑回归模型，差异有统计学意义。预测贡献度居前10位的变量依次为体温、糖类抗原72-4（CA72-4）、激素治疗、白蛋白、尿酸（UA）、血钠（Na）、总胆固醇、白细胞计数（WBC）、癌胚抗原（CEA）及中性粒细胞百分比。结论 4种模型中，XGBoost模型预测效能最优；体温、CA72-4、激素治疗、白蛋白及UA为贡献度最高的前5位术前变量，可为早期识别高风险病人及制定干预措施提供参考。
Objective To construct a risk prediction model for delayed discharge after R0 resection of colorectal cancer using machine learning algorithms based on preoperative parameters and to compare the predictive performance of different models. Methods Clinical data from 2048 patients with colorectal cancer who underwent R0 resection at Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, from November 2015 to April 2021 were retrospectively analyzed. The data were randomly divided at a 7:3 ratio into a training set (1 433 cases) and a validation set (615 cases). Four algorithms—Logistic regression, support vector machine (SVM), random forest, and extreme gradient boosting (XGBoost)—were used to construct risk prediction models for delayed discharge based on preoperative laboratory parameters and general patient characteristics. Predictive performance was evaluated and compared using the area under the receiver operating characteristic curve (AUC), true-positive rate (TPR), true-negative rate (TNR), accuracy (ACC), and positive predictive value (PPV). Results Postoperative infection was the leading cause of delayed discharge, accounting for 40.52% of all causes of delayed discharge. In the training set, the AUCs of the XGBoost, random forest, SVM, and Logistic regression models were 0.762, 0.772, 0.747, and 0.750, respectively; the corresponding values in the validation set were 0.747, 0.735, 0.729, and 0.738. The XGBoost model had the best overall predictive performance. Its TNR was significantly higher than that of the SVM model (P=0.046), and its TNR (P<0.001) and ACC (P=0.001) were significantly better than those of the Logistic regression model. The top 10 variables contributing to XGBoost predictions, in descending order, were body temperature, carbohydrate antigen 72-4 (CA72-4), hormone therapy, albumin, uric acid (UA), serum sodium (Na), total cholesterol, white blood cell count (WBC), carcinoembryonic antigen (CEA), and neutrophil percentage. Conclusion Among the four models, XGBoost had the best predictive performance. Body temperature, CA72-4, hormone therapy, albumin, and UA were the top 5 preoperative variables contributing to model predictions. The model may facilitate early identification of high-risk patients and inform intervention planning.
结直肠癌 / 延迟出院 / 机器学习 / 预测模型 / R0切除 / XGBoost模型
colorectal cancer / delayed discharge / machine learning / prediction model / R0 resection / XGBoost model
| [1] |
|
| [2] |
Aim of the study was to report the experience of a single center in the surgical treatment of anastomotic leak after colorectal resection for cancer, focusing on its incidence, diagnosis and management, with particular attention to surgical options.Demograhic and clinical data from 1284 consecutive patients operated on for colorectal cancer during a period of 11 years, were prospectively collected and inserted into an electronic database. For the purpose of the study, only patients in whom an anastomosis was performed were considered. Therefore, 128 patients (9.9%) were excluded and 1156 represent the study population. All complications occurring after surgery were systematically recorded. The incidence of colorectal anastomotic leak (CAL), its management and outcome was analyzed.An ileo-colic anastomosis was performed in 426 cases (36.8%), ileo-rectal in 29 (2.5%), colo-colic in 409 (35.4), colo-rectal with partial mesorectal excision in 211 (18.3%) and lower colorectal or coloanal anastomosis with total mesorectal excision and temporary loop ileostomy in 81 (7%). Colorectal anastomotic leak occurred in 78/1156 patients (6.7%), accounting for 19% of overall complications. Thirty-six out of 78 patients (46.2%) were successfully treated conservatively, whereas 42 (53.8%) underwent re-operation. Nine out of 78 patients (11.5%) with CAL died owing to an uncontrolled sepsis. All had undergone previous re-intervention, with a post-operative mortality rate of 21.4% (9 out of 42).Mortality rate in patients undergoing re-operation for colorectal anastomotic leakeage is still high, and accounts for up to 40% of the deaths after colorectal resection for cancer. In the light of these data, strategic clinical decisions are mandatory to optimize the selection of patients who need an early and fast surgical approach. What does this paper add to the literature? Systematic and prospective data recording is an essential tool to assess the quality of healthcare and to plan quality improvement programs. Every effort should be done to reach an early diagnosis of CAL, possibly in a pre-clinical phase in which non clinical methods could be used to predict it.
|
| [3] |
肖毅. 近十余年结直肠外科领域的理想与现实[J]. 中国实用外科杂志, 2024, 44(4):398-402. DOI:10.19538/j.cjps.issn1005-2208.2024.04.05.
|
| [4] |
Background: The impact of developments in colorectal cancer surgery on length-of-stay (LOS) and re-admission have not been well described. In a population-based analysis, we investigated predictors of LOS and emergency readmission after the initial surgery episode.;Methods: Incident colorectal cancers (ICD-O2: C18-C20), diagnosed 2002-2008, were identified from the National Cancer Registry Ireland, and linked to hospital in-patient episodes. For those who underwent colorectal resection, the associated hospital episode was identified. Factors predicting longer LOS (upper-quartile, > 24 days) for elective and emergency admissions separately, and whether LOS predicted emergency readmission within 28 days of discharge, were investigated using logistic regression.;Results: 8197 patients underwent resection, 63% (n = 5133) elective and 37% (n = 3063) emergency admissions. Median LOS was 14 days (inter-quartile range (IQR) = 11-20) for elective and 21 (15-33) for emergency admissions. For both emergency and elective admissions, likelihood of longer LOS was significantly higher in patients who were older, had co-morbidities and were unmarried; it was reduced for private patients. For emergency patients only the likelihood of longer LOS was lower for patients admitted to higher-volume hospitals. Longer LOS was associated with increased risk of emergency readmission.;Conclusions: One quarter of patients stay in hospital for at least 25 days following colorectal resection. Over one third of resected patients are emergency admissions and these have a significantly longer median LOS. Patient- and health service-related factors were associated with prolonged LOS. Longer LOS was associated with increased risk of emergency readmission. The cost implications of these findings are significant.
|
| [5] |
Delay from surgery to adjuvant chemotherapy causes impaired survival among patients undergoing radical resection for stage III colon cancer, and the underlying mechanism for this is incompletely clarified. It is established that prolonged postoperative hospital length of stay (LOS) is associated with delayed initiation of the adjuvant treatment driving the assumption that prolonged LOS is prognostically unfavorable due to this fact and case mix factors. We hypothesize that prolonged LOS after surgery is a valuable marker for susceptibility to relapse that is not detected in established prognostic factors and, alone, associated with a shorter disease-free survival (DFS).A total of 690 consecutive patients undergoing elective radical resection for stage III colon cancer in 2000-2015 were identified in a prospective detailed facility database. Univariate and multivariate analyses were performed using Cox proportional hazards model in the evaluation of LOS as an independent prognostic factor.Short postoperative LOS, low comorbidity, and few complications were associated with longer DFS (p < 0.01). Fewer patients in the short and intermediate LOS groups had a relapse in their disease (28% and 33%, respectively), compared to the patients with longer LOS (40%, p < 0.05). LOS was a prognostic factor for DFS in the unadjusted univariate model (HR 1.04 per unit change) and remained statistically significant in the adjusted multivariate analysis, with a HR of 1.03 per hospital day (p < 0.01).Postoperative LOS independently correlates with the risk of recurrence and DFS, regardless of if adjuvant chemotherapy is given, along with the factors such as age, comorbidity, complications, and tumor features. We propose a further investigation into the causal mechanisms based on tumor and host biology linking LOS to DFS beyond established risk factors.© 2022. The Author(s).
|
| [6] |
|
| [7] |
|
| [8] |
To determine the cost-effectiveness of enhanced recovery pathways (ERPs) versus conventional care for patients undergoing elective colorectal surgery.ERPs for colorectal surgery are clinically effective, but their cost-effectiveness is unknown.A multi-institutional prospective cohort cost-effectiveness analysis was performed. Adult patients undergoing elective colorectal resection at 2 university-affiliated institutions from October 2012 to October 2013 were enrolled. One center used an ERP, whereas the other did not. Postoperative outcomes were recorded up to 60 days. Total costs were reported in 2013 Canadian dollars. Effectiveness was measured using the SF-6D, a health utility measure validated for postoperative recovery. Uncertainty was expressed using bootstrapped estimates (10,000 repetitions).A total of 180 patients were included (conventional care: n = 95; ERP: n = 95). There were no differences in patient characteristics except for a higher proportion of laparoscopy in the ERP group. Mean length of stay was shorter in the ERP group (6.5 vs 9.8 days; P = 0.017), but there were no differences in complications or readmissions. Patients in the ERP group returned to work quicker and had less caregiver burden. There was no difference in quality of life between the 2 groups. The cost of the ERP program was $153 per patient. Overall societal costs were lower in the ERP group (mean difference = -2985; 95% confidence interval, -5753 to -373). The ERP had a greater than 99% probability of cost-effectiveness. The results were insensitive to a range of assumptions and subgroups.Enhanced recovery is cost-effective compared with conventional perioperative management for elective colorectal resection.
|
| [9] |
Despite advances with Enhanced Recovery Pathways(ERP), some patients have unexpected prolonged lengths of stay(LOS). Our goal was to identify the patient and procedural variables associated with delayed discharge despite an established ERP.A divisional database was reviewed for minimally invasive colorectal resections with a multimodal ERP(8/1/13-7/31/15). Patients were stratified into ERP success or failure based on length of stay ≥5 days. Logistic regression modeling identified variables predictive of ERP failure.274 patients were included- 229 successes and 45 failures. Groups were similar in demographics. Failures had higher rates of preoperative anxiety(p = 0.0352), chronic pain(p = 0.0040), prior abdominal surgery(p = 0.0313), and chemoradiation(p = 0.0301). Intraoperatively, failures had higher conversion rates(13.3% vs. 1.7%, p = 0.0002), transfusions(p = 0.0032), and longer operative times(219.8 vs. 183.5min,p = 0.0099). Total costs for failures were higher than successes($22,127 vs. $13,030,p = 0.0182). Variables independently associated with failure were anxiety(OR 2.28, p = 0.0389), chronic pain(OR 10.03, p = 0.0045), and intraoperative conversion(OR 8.02, p = 0.0043).Identifiable factors are associated with delayed discharge in colorectal surgery. By prospectively preparing for patient factors and changing practice to address procedural factors and ERP adherence, postoperative outcomes could be improved.Copyright © 2017 Elsevier Inc. All rights reserved.
|
| [10] |
|
| [11] |
The risk factors for extended length of stay (LOS) have not been examined in a cohort of patients with complex social and medical barriers who undergo robotic assisted (RA) surgery for gynecologic malignancies. We sought to identify those patients with a LOS > 24 h after robotic surgery and the risk factors associated with delayed discharge. Then we aimed to develop a predictive model for clinical care and identify modifiable pre-operative risk factors.After IRB approval, data was abstracted from medical records of all patients with a gynecologic malignancy who underwent a RA laparoscopic surgery from 2010 to 2015. Univariable and multivariable logistic regression was performed to identify independent risk factors associated with delayed discharge defined as LOS > 24 h. A multi-variable logistic regression model was performed using a stepwise backward selection for the final prediction model. All testing was two-sided and a p-value < 0.05 was considered statistically significant.Of the 406 eligible and evaluable patients, 194 (48%) had a LOS > 24 h. Age ≥ 60 years, a higher usage of narcotic medication, a longer surgical time, and a larger estimated blood loss were all associated with LOS > 24 h (p < 0.05). Many of these women had a social work consultation and went home with home care services despite no surgical or post-operative complications. Our prediction model has the potential to correctly classified 75% of the patients discharged within 24 h.The development of a pre-hospitalization risk stratification and anticipating the possible need for home care services pre-operatively shows promise as a strategy to decrease LOS in patients classified as high-risk. These findings warrant prospective validation through the use of this prediction model in our institution.Published by Elsevier Inc.
|
| [12] |
|
| [13] |
|
| [14] |
Laparoscopic resection and a multimodal approach known as an enhanced recovery program (ERP) have been major changes in colorectal perioperative care that have improved clinical outcomes for colorectal cancer resection. EnROL (Enhanced Recovery Open Versus Laparoscopic) is a multicenter randomized controlled trial examining whether the benefits of laparoscopy still exist when open surgery is optimized within an ERP.Adults with colorectal cancer suitable for elective resection were randomly assigned at a ratio of 1:1 to laparoscopic or open surgery within an ERP, stratified by center, cancer site (colon v rectum), and age group (<66 v 66-75 v >75 years) using minimization. The primary outcome was physical fatigue at 1 month postsurgery. Secondary outcomes included hospital stay, complications, other patient-reported outcomes (PROs), and physical function. Patients and outcome assessors were blinded until 7 days postsurgery or discharge if earlier. Central independent and blinded pathologic assessment of surgical quality was undertaken.A total of 204 patients (laparoscopy, n=103; open surgery, n=101) were recruited from 12 UK centers from July 2008 to April 2012. One-month physical fatigue scores were similar in both groups (mean: laparoscopy, 12.28; 95% CI, 11.37 to 13.19 v open surgery, 12.05; 95% CI, 11.14 to 12.96; adjusted mean difference, -0.23; 95% CI, -1.52 to 1.07). Median total hospital stay was significantly shorter after laparoscopic surgery (median: laparoscopy, 5; interquartile range [IQR], 4 to 9 v open surgery, 7; IQR, 5 to 11 days; P=.033). There were no differences in other secondary outcomes or in specimen quality after central pathologic review.In patients treated by experienced surgeons within an ERP, physical fatigue and other PROs were similar in both groups, but laparoscopic surgery significantly reduced length of hospital stay.© 2014 by American Society of Clinical Oncology.
|
| [15] |
Prolonged hospitalisation not only increases cost, it is also associated with other complications. Length of stay (LOS) is one of the indicators that reflect total cost of care during hospitalisation. So, it is of paramount importance to find out why elderly patients overstay in acute care hospitals and address these issues proactively. Since no local data is available, a study was planned to ascertain reasons why the discharging of elderly patients from hospital is delayed and whether these reasons are avoidable.Long-stay patients were defined as those whose LOS had notably exceeded the average LOS for the Diagnosis Related Group (DRG), based on principal admitting diagnosis. A separate analysis showed that the specialty-specific long-stay marker for geriatric medicine was 28 days, so casenote review was done for all patients with LOS of more than 28 days who were discharged from geriatric medicine service of an acute care hospital during a 1-year study period. Information was collected on demographic profile, functional and cognitive status, past medical and social history, admitting medical diagnoses, discharge limiting and delaying factors.During the study period, 150 patients stayed over 28 days and 137 casenotes were available for review. The mean age of the patients was 84 years, 55.5% were female, 77.4% were Chinese, mean abbreviated mental test (AMT) score was 3 and mean modified Barthel's score was 11. The commonest primary diagnosis was sepsis followed by neurological problems, falls-related complication and cardiovascular diseases. The 2 most common discharge limiting factors (the final event which resulted in delay in discharge) were social issues (54, 39.4%) and sepsis (47, 34.3%). Of 47 patients with sepsis, 37 (78.7%) were nosocomial infection. Urinary tract infection and pneumonia were the 2 most common nosocomial infections. The 4 most common factors contributing to delayed discharge (various problems that surfaced throughout the hospital stay) were sepsis (94, 68.6%), deconditioning (65, 47.4%), social issues (52, 38.0%) and cardiovascular disorders (37, 27.0%).Elderly patients are more prone to hospitalisation-related complications like nosocomial infection and deconditioning leading to prolonged hospital stay. Early interventions can reduce these complications. Early identification of social issues and prompt discharge planning should be done to avoid delay in discharge.
|
| [16] |
We sought to identify independent modifiable risk factors for delayed discharge after total knee arthroplasty (TKA) that have been previously underrepresented in the literature, particularly postoperative opioid use, postoperative laboratory abnormalities, and the frequency of hypotensive events.Data from 1033 patients undergoing TKA for primary osteoarthritis of the knee between June 2012 and August 2014 at an academic orthopedic specialty hospital were reviewed. Patient demographics, comorbidities, inpatient opioid medication, postoperative hypotensive events, and abnormalities in laboratory values, all occurring on postoperative day 0 or 1, were collected. Multivariate logistic regression analysis was performed to identify independent risk factors for a prolonged length of stay (LOS) >3 days.The average age of patients undergoing primary TKA in our cohort was 65.9 (standard deviation, 9.1) years, and 61.7% were women. The mean LOS for all patients was 2.64 days (standard deviation, 1.14; range, 1-9). And 15.3% of patients had a LOS >3 days. On multivariate logistic regression analysis, nonmodifiable risk factors associated with a prolonged LOS included nonwhite race (odds ratio [OR], 2.01), single marital status (OR, 1.53), and increasing age (OR, 1.47). Modifiable risk factors included every 5 postoperative hypotensive events (OR, 1.31), 10-mg increases in oral morphine equivalent consumption (OR, 1.04), and postoperative laboratory abnormalities (hypocalcemia: OR, 2.15; low hemoglobin: OR, 2.63).This study identifies potentially modifiable factors that are associated with increased LOS after TKA. Doubling down on efforts to control the narcotic use and to use opioid alternatives when possible will likely have efficacy in reducing LOS. Attempts should be made to correct laboratory abnormalities and to be cognizant of patient opioid use, age, and race when considering potential avenues to reduce LOS.Copyright © 2019 Elsevier Inc. All rights reserved.
|
| [17] |
张宁, 南博, 拜云虎, 等. 胰腺创伤术后住院概率预测模型构建与验证[J]. 中国实用外科杂志, 2025, 45(11):1309-1316. DOI:10.19538/j.cjps.issn1005-2208.2025.11.13.
|
| [18] |
韩序, 刘亮, 楼文晖. 生成式人工智能大型语言模型在消化道癌症领域辅助科研创作的现状分析:基于2024年美国临床肿瘤学会中国学者数据[J]. 中国实用外科杂志, 2024, 44(8):894-899. DOI:10.19538/j.cjps.issn1005-2208.2024.08.14.
|
| [19] |
|
| [20] |
|
| [21] |
Artificial neural networks (ANNs) can be used to develop predictive tools to enable the clinical decision-making process. This study aimed to investigate the use of an ANN in predicting the outcomes from enhanced recovery after colorectal cancer surgery.Data were obtained from consecutive colorectal cancer patients undergoing laparoscopic surgery within the enhanced recovery after surgery (ERAS) program between 2002 and 2009 in a single center. The primary outcomes assessed were delayed discharge and readmission within a 30-day period. The data were analyzed using a multilayered perceptron neural network (MLPNN), and a prediction tools were created for each outcome. The results were compared with a conventional statistical method using logistic regression analysis.A total of 275 cancer patients were included in the study. The median length of stay was 6 days (range 2-49 days) with 67 patients (24.4 %) staying longer than 7 days. Thirty-four patients (12.5 %) were readmitted within 30 days. Important factors predicting delayed discharge were related to failure in compliance with ERAS, particularly with the postoperative elements in the first 48 h. The MLPNN for delayed discharge had an area under a receiver operator characteristic curve (AUROC) of 0.817, compared with an AUROC of 0.807 for the predictive tool developed from logistic regression analysis. Factors predicting 30-day readmission included overall compliance with the ERAS pathway and receiving neoadjuvant treatment for rectal cancer. The MLPNN for readmission had an AUROC of 0.68.These results may plausibly suggest that ANN can be used to develop reliable outcome predictive tools in multifactorial intervention such as ERAS. Compliance with ERAS can reliably predict both delayed discharge and 30-day readmission following laparoscopic colorectal cancer surgery.
|
| [22] |
Synchronous metastatic para-aortic lymph node (mPALN) dissectionin colorectal cancer has relatively good oncological outcomes, though many patients develop recurrence. Universal prognostic factor remain unclear and no definitive perioperative chemotherapy is available, making the treatment of mPALN controversial. In the present study, we aimed to establish a treatment strategy for synchronous mPALN.This retrospective study involved 20 patients with pathological mPALN below the renal vein who underwent R0 resection. Long-term outcomes, recurrence type, and prognostic factors for survival were investigated.The 5-year overall survival and recurrence-free survival rates were 39% and 25%, respectively. Seventeen patients (85%) developed recurrence, including 13 (76%) within 1 year after surgery, and ~ 70% of all recurrences were multiple recurrences. Four patients (20%) survived > 5 years. Pathological T stage (p= 0.011), time to recurrence (p = 0.007), and recurrence resection (p = 0.009) were identified as prognostic factors for long-term survival.R0 resection of synchronous mPALN in colorectal cancer resulted in acceptable oncological outcomes, though we found a high rate of early unresectable recurrence. If the recurrence occurs late or isolated, surgical resection should be considered for longer survival.
|
| [23] |
|
| [24] |
Postoperative length of stay is a key indicator in the management of medical resources and an indirect predictor of the incidence of surgical complications and the degree of recovery of the patient after cancer surgery. Recently, machine learning has been used to predict complex medical outcomes, such as prolonged length of hospital stay, using extensive medical information.
|
| [25] |
Objective. To explore the value of the heart rate, body temperature, and respiratory rate in the early prediction of anastomotic leakage after rectal cancer surgery. Methods. Clinical data from patients with rectal cancer who underwent anterior rectal resection in the Department of Gastroenterology, Renmin Hospital of Wuhan University, from January 2017 to December 2019 were collected and analyzed retrospectively. Based on the occurrence of anastomotic leakage after surgery, the patients were divided into two groups: those with and without anastomotic leakage. The quantitative values of the heart rate, body temperature, and respiration rate at day 7 postsurgery were compared between the two groups. The ROC curve was used to analyze their role in the early prediction of anastomotic leakage. Results. Among 441 patients with rectal cancer, 30 (6.81%) had clinical anastomotic leakage and were diagnosed at 7 ± 3 days postsurgery. Within 7 days postsurgery, the heart rate, body temperature, and respiratory rate in the anastomotic leakage group were higher than those in the nonanastomotic leakage group. The differences in heart rate (1‐5 d), body temperature (2‐7 d), and respiratory rate (1‐7 d) were statistically significant (P < 0.05). The three ROC curves were drawn, respectively. The predictive value of the heart rate is greatest at days 2‐3 postsurgery. The predictive value of the body temperature is greatest at days 4‐6 postsurgery. The predictive value of the respiratory rate is best at days 1‐4 postsurgery. Conclusion. The changes of vital signs (heart rate, body temperature, and respiratory rate) have a certain value in the early prediction of anastomotic leakage after rectal cancer surgery. Observation of postoperative vital signs at 7 days postsurgery is helpful for the early diagnosis of anastomotic leakage.
|
| [26] |
|
| [27] |
There had been no recognized serum tumor marker to predict the prognosis of colorectal cancer (CRC) patients with normal preoperative serum carcinoembryonic antigen (CEA) levels. The purpose of this study was to determine whether preoperative serum carbohydrate antigen 724 (CA724) was of predictive function for the prognosis of CRC patients with normal CEA levels.
|
| [28] |
A wide range of serum tumor biomarkers, including CA19-9, CA242, CA72-4, CA50, and CA125, has been studied in association with colorectal cancer (CRC). However, few previous studies have comprehensively considered the above tumor biomarkers to assess their clinical significance in predicting prognosis. Data from Fudan University Shanghai Cancer Center (FUSCC) between January 1, 2007 and December 30, 2012 was retrospectively analyzed. Univariate and multivariate analyses were performed to assess the association between prognostic factors and survival outcomes. Nomograms were established based on multivariate Cox regression model analysis for overall survival (OS) and disease free survival (DFS), and c-indexes were 0.772 (95% CI: 0.724-0.820) and 0.715 (95% CI: 0.678-0.752), respectively. Subgroup analyses according to CEA status (high/normal) suggested that CA724 was the only independent prognostic factor for OS (P = 0.001) and DFS (P < 0.001) in the CEA-high group, while, in the CEA-normal group, the only independent prognostic factor for OS (P = 0.031) and DFS (P = 0.043) was CA50. CA50 and CA724 could supplement CEA in monitoring recurrence and metastasis. Accordingly, nomograms based on CEA, CA50, CA724 and other clinical-pathological factors could improve prognosis prediction for colorectal cancer patients.
|
| [29] |
Anastomotic leakage is a potentially severe complication occurring after colorectal surgery and can lead to increased morbidity and mortality, permanent stoma formation, and cancer recurrence. Multiple risk factors for anastomotic leak have been identified, and these can allow for better prevention and an earlier diagnosis of this significant complication. There are nonmodifiable factors such as male gender, comorbidities and distance of tumor from anal verge, and modifiable risk factors, including smoking and alcohol consumption, obesity, preoperative radiotherapy and preoperative use of steroids or non-steroidal anti-inflammatory drugs. Perioperative blood transfusion was shown to be an important risk factor for anastomotic failure. Recent studies on the laparoscopic approach in colorectal surgery found no statistical difference in anastomotic leakage rate compared with open surgery. A diverting stoma at the time of primary surgery does not appear to reduce the leak rate but may reduce its clinical consequences and the need for additional surgery if anastomotic leakage does occur. It is still debatable if preoperative bowel preparation should be used, especially for left colon and rectal resections, but studies have shown similar incidence of postoperative leak rate.
|
| [30] |
|
| [31] |
This study examined risk factors for intra-abdominal sepsis after surgery in Crohn's disease.We reviewed 343 patients who underwent 1,008 intestinal anastomoses during 566 operations for primary or recurrent Crohn's disease between 1980 and 1997. Possible factors for intra-abdominal sepsis were analyzed by both univariate (chi-squared test) and multivariate (multiple regression) analyses.Intra-abdominal septic complications, defined as anastomotic leak, intra-abdominal abscess, or enterocutaneous fistula, developed after 76 operations (13 percent). Intra-abdominal septic complications were significantly associated with preoperative low albumin level (< 30 g/l; P = 0.04), preoperative steroids use (P = 0.03), abscess at the time of laparotomy (P = 0.03), and fistula at the time of laparotomy (P = 0.04). The intra-abdominal septic complication rate was 50 percent (8/16 operations) in patients with all of these four risk factors, 29 percent (10/35 operations) in patients with three risk factors, 14 percent (14/98 operations) in patients with two risk factors, 16 percent (33/209 operations) in patients with only one risk factor, and 5 percent (11/208 operations) in patients with none of these risk factors (P<0.0001). The following factors did not affect the incidence of septic complications; age, duration of symptoms, number of previous bowel resections, site of disease, type of operation (resection, strictureplasty, or bypass), covering stoma, and number, site, or method (sutured or stapled) of anastomoses.Preoperative low albumin level, steroid use, and the presence of abscess or fistula at the time of laparotomy significantly increased the risk of septic complications after surgery in Crohn's disease.
|
| [32] |
This study was designed to assess predictive factors of postoperative intra-abdominal septic complication in a homogenous group of patients undergoing ileocecal resection for Crohn's disease.From 1984 to 2004, 161 consecutive patients with Crohn's disease (81 males; mean age, 33 +/- 10 years) underwent, as a first operation, an elective ileocecal resection without temporary stoma. Postoperatively, 15 patients (9 percent) developed abdominal septic complications, including abscess and anastomotic leaks. Possible factors for postoperative intra-abdominal septic complication were analyzed by both univariate and multivariate analyses.There was no postoperative death. Multivariate analysis found only four independent factors associated with a higher risk of postoperative intra-abdominal septic complication: poor nutritional status (odds ratio, 6.23 (1.75-22.52)), intra-abdominal abscess discovered during surgery (odds ratio, 7.47 (1.5-37.69)), preoperative steroids use more than three months (odds ratio, 5.95 (1.04-34.1)), and recurrent clinical episode of Crohn's disease (odds ratio (per episode), 1.38 (1.03-1.9)).Recurrent clinical episode of Crohn's disease, preoperative steroids use, poor nutritional status, and the presence of abscess at the time of surgery significantly increased the risk of septic abdominal complications after first ileocecal resection for Crohn's disease. Knowledge of these risk factors could permit to propose a temporary stoma in very high-risk patients (i.e., with 3 or more risk factors).
|
| [33] |
|
| [34] |
Anastomotic leaks (AL) are among the most serious complications due to the substantial impact on the quality of life and mortality. Inspite of the advance in diagnostic tools such as laboratory tests and radiological adjuncts, only moderate improvement has been recorded in the rate of detected leaks. The purpose of the research was to assess the perioperative risk factors for AL.This study was achieved at MUH and MIH/Egypt within the period between January 2016 and January 2019 for the candidates who underwent bowel anastomosis for small intestinal (except duodenal one) and colorectal pathology. The collected data were analyzed using SPSS of V-26.This study included 315 cases, among them, 27 cases (8.57%) developed AL. The percentage of covering stoma was significantly higher in the non-leakage group vs leakage one (24.3% vs 11.1% respectively). lower albumin, operative timing, perforation, and covering stoma were shown as significant risk factors for leakage, but with multivariate analysis for these factors, the emergency operation, and serum albumin level was the only independent risk factors that revealed the significance consequently (p = 0.043, p = 0.015). The analysis of different predictors of AL on the third day showed that the cut-off point in RR was 29 with 83% sensitivity and 92% specificity in prediction of leakage, the cut-off point in RR was 118 with 74% sensitivity and 87% specificity in prediction of leakage and the cut-off point in CRP was 184.7 with 82% sensitivity and 88% specificity in prediction of AL and all had statistically significant value.The preoperative serum albumin level and the emergency operations are independent risk factors for anastomotic leakage. Moreover, leakage should be highly suspected in cases with rising respiratory rate, heart rate, and CRP levels.
|
| [35] |
the evidence of a potential beneficial role of antioxidants in preventing atherosclerotic disease is not entirely consistent.to assess the longitudinal association of serum total antioxidant capacity and serum antioxidants with the presence of subclinical carotid atherosclerosis.Prospective case-control study nested within an historical cohort. Cases were 150 individuals with elevated carotid intimal-medial thickness measured by B-mode ultrasound at the first two examinations of the Atherosclerosis Risk in Communities Study (1987-92). Controls were 150 age-gender-matched individuals with low carotid intimal-medial thickness. Serum antioxidant vitamins, uric acid, and serum total antioxidant capacity were measured in frozen serum samples collected from the same individuals in 1974 (13-15 years prior to the determination of case-control status).Compared to controls, atherosclerosis cases had significantly higher levels of serum total antioxidant capacity in 1974 than controls. This difference was almost entirely explained by increased serum concentration of uric acid in cases. In contrast with cross-sectional results, uric acid serum concentration in 1974, was significantly higher in cases than in controls, even after adjusting for the main cardiovascular risk factors. Cases had significantly lower levels of alpha-carotene in the 1974 sera than controls, but no other differences in serum antioxidant vitamin concentrations were observed.The higher serum uric acid concentration seemed associated with elevated total serum antioxidant capacity among individuals with atherosclerosis. This finding is consistent with experimental evidence suggesting that hyperuricemia may be a compensatory mechanism to counteract oxidative damage related to atherosclerosis and aging in humans.
|
| [36] |
|
| [37] |
|
| [38] |
To investigate the prognostic significance of uric acid level in colorectal cancer in addition to conventional factors in terms of survival.Observational Study.At the University of Health Sciences, Dr. Abdurrahman Yurtaslan Ankara Oncology Training and Research Hospital, Ankara, Turkey, between January 2012 and December 2019. Methodology: A total of selected 332 patients, who underwent surgery for colorectal cancer between 2012 and 2019, were retrospectively reviewed. Patients with a history of neoadjuvant therapy due to rectal cancer, GUT disease, renal failure, stage 4 disease, emergency surgery and severe preoperative infection, were excluded from the study. Peripheral blood samples were collected about a week before the operation. Serum uric acid (SUA) values were measured and recorded. Results: The patients comprised 198 males and 134 females with a mean age 62.2 ±11.7 years (14-91) years in total. Conventional surgery was performed in 228 patients, and laparoscopy in 104 patients. Uric acid level, number of pathological lymph nodes, number of pathological lymph nodes/total number of lymph nodes (LNO), perineural invasion, type of surgery and disease stage were found to be factors affecting the prognosis (p <0.05). Uric acid cut off value of 5.3 or higher was found to be statistically significant in terms of survival.Serum uric acid (SUA) value measured preoperatively was found to be a prognostic factor for colorectal cancer. Key Words: Serum uric acid (SUA), Colon cancer, Rectal cancer, Prognostic factor.
|
利益冲突 所有作者均声明不存在利益冲突
感谢上海森亿智慧信息科技股份有限公司李明在本研究数据库整合、机器学习算法开发与模型性能优化中给予的帮助
/
| 〈 |
|
〉 |
