PDF(1562 KB)
经脐单孔腹腔镜与传统腹腔镜全子宫切除手术并发症的影响因素分析及预测模型构建
任小玉, 冯剑敏, 杨涵琳, 朱焱, 龚芫, 田维杰, 訾聃
中国实用妇科与产科杂志 ›› 2025, Vol. 41 ›› Issue (7) : 755-759.
PDF(1562 KB)
PDF(1562 KB)
经脐单孔腹腔镜与传统腹腔镜全子宫切除手术并发症的影响因素分析及预测模型构建
Analysis of influencing factors and construction of a prediction model for surgical complications in transumbilical single-Port Laparoscopic and conventional laparoscopic total hysterectomy
目的 探讨经脐单孔腹腔镜和传统腹腔镜全子宫切除术患者手术并发症的共同影响因素及预测模型构建。 方法 选取2021年9月至2024年12月于贵州省人民医院因良性妇科病变行经脐单孔或传统腹腔镜全子宫切除术的585例患者为研究对象，收集患者临床资料，按照7∶3比例划分为训练集和验证集。在训练集中分别用LASSO-logistic回归、全变量logistic回归及逐步logistic回归模型筛选并发症影响因素，整合曲线下面积（AUC）、Akaike信息准则（AIC）/贝叶斯信息准则（BIC）、净重分类指数（NRI）/综合判别改善指数（IDI）多维度评估模型效能并开发网页列线图工具，最终经验证集验证其区分度、校准度与临床适用性。结果 术前中重度贫血、剖宫产史、高血压及子宫>12孕周为独立危险因素，LASSO-logistic回归模型具有较好预测性能[AUC=0.792（95%CI 0.736~0.849），AIC =341.059 ，BIC = 374.115，NRI>0，IDI>0]。验证集AUC=0.786（95%CI 0.725~0.845）。校准曲线显示模型预测与实测风险吻合良好。决策曲线表明当风险阈值>0.04时，模型可提供临床净收益（训练集阈值范围0.07~0.98，验证集0.04~0.97）。结论 术前贫血、剖宫产史、高血压和子宫大小是经脐单孔腹腔镜或传统腹腔镜全子宫切除手术并发症的共同影响因素。LASSO-logistic回归模型具有一定临床预测价值。
Objective To investigate the influencing factors of surgical complications in patients undergoing transumbilical single-port laparoscopic total hysterectomy and traditional laparoscopic total hysterectomy and to construct a prediction model. Methods The clinical data of patients undergoing transumbilical single-port or traditional laparoscopic total hysterectomy for benign diseases at Guizhou Provincial People's Hospital from September 2021 to December 2024 were retrospectively collected. The data were divided into a training set (n=409) and a validation set (n=219). LASSO-logistic regression，full-variable logistic regression，and stepwise logistic regression models were applied to screen for influencing factors in the training set. Model performance was evaluated in multiple dimensions using the area under the receiver operating characteristic curve (AUC)，Akaike information criterion (AIC)/Bayesian information criterion (BIC)，net reclassification index (NRI)/integrated discrimination improvement index (IDI)，and a nomogram was developed. The validation set was used to assess its discrimination，calibration，and clinical applicability. Results Moderate to severe preoperative anemia，history of cesarean delivery，hypertension，and uterine size >12 gestational weeks were identified as independent risk factors. The LASSO-logistic regression model demonstrated optimal predictive performance [AUC=0.792 (95%CI 0.736-0.849)；AIC=341.059；BIC=374.115；NRI > 0；IDI > 0]. Validation yielded an AUC of 0.785 (95%CI 0.700-0.870). Calibration curves showed good consistency between predicted and observed risks. Decision curve analysis confirmed clinical net benefit at risk thresholds >0.06 (training set：0.07-0.98；validation set：0.06-0.86). Conclusions Preoperative anemia，history of cesarean section，hypertension，and uterine size are identified as influencing factors for complications in laparoscopic total hysterectomy. The LASSO-logistic regression model demonstrates certain clinical prediction value.
子宫切除术 / 经脐单孔腹腔镜 / 手术并发症 / 预测模型
hysterectomy / transumbilical single-port laparoscopy / surgical complications / prediction model
| [1] |
中国医师协会妇产科医师分会妇科肿瘤专业委员会(学组). 良性子宫疾病子宫切除术手术路径的中国专家共识(2021年版)[J]. 中国实用妇科与产科杂志, 2021, 37(8):821-825. DOI: 10.19538/j.fk2021080109.
子宫切除是妇科临床最常实施的手术操作之一，子宫（子宫颈）良性疾病患者多因指征需要而选择子宫切除术［1］。经典子宫切除术的适应证主要包括子宫肌瘤、子宫内膜异位症/子宫腺肌病、异常子宫出血、子宫脱垂等，广义上子宫切除指征还包括其他盆腔疾病及不明原因的盆腔疼痛等。传统意义上的开腹子宫切除术是经典的手术方式，然而随着医疗技术的不断进步和认识的提高，子宫切除的路径选择已形成经腹、阴式、腹腔镜、腹腔镜辅助阴式和机器人辅助等多种路径并存的局面，并且各种手术路径和手术方式的子宫切除术均已在我国得到了蓬勃发展并广泛应用于临床。浏览更多请关注本刊微信公众号及当期杂志。
|
| [2] |
To estimate the incidence and risk factors for bowel injury in women undergoing hysterectomy for benign indications.A retrospective cohort study was conducted among women undergoing hysterectomy for benign indications from 2012 to 2016 at institutes participating in the American College of Surgeons National Surgical Quality Improvement Program, including both inpatient and outpatient settings. Bowel injury was identified using Current Procedural Terminology codes as patients who underwent bowel repair at the time of hysterectomy or postoperatively within 30 days. Multivariate logistic regression models were used to control for patient clinical factors and perioperative factors.Bowel injury occurred in 610 of 155,557 (0.39%) included women. After bivariate analysis, factors associated with bowel injury included age, race, body mass index, American Society of Anesthesiologists classification, increased operative time, surgical approach, type of hysterectomy, lysis of adhesions, and operative indication. After adjusting for potential confounders, bowel injury was found associated with older age, surgical indication of endometriosis, and abdominal surgical approach. Compared with the surgical indication of endometriosis (n=63/10,625), the surgical indications of menstrual disorder (odds ratio [OR] 0.33, 95% CI 0.23-0.47; adjusted odds ratio [aOR] 0.33, 95% CI 0.23-0.48; n=67/34,168), uterine leiomyomas (OR 0.80, 95% CI 0.61-1.05; aOR 0.44, 95% CI 0.33-0.59; n=243/51,232), and genital prolapse (OR 0.30, 95% CI 0.20-0.45; aOR 0.41, 95% CI 0.25-0.67; n=36/20,384) were each associated with lower odds of bowel injury. Compared with the vaginal approach to hysterectomy (n=27/27,434), the abdominal approach was found to have significantly increased odds of bowel injury (OR 10.80, 95% CI 7.31-15.95; aOR 10.49 95% CI 6.42-17.12; n=401/38,106); the laparoscopic approach had smaller but significantly increased odds (OR 2.06, 95% CI 1.37-3.08; aOR 2.03 95% CI 1.24-3.34; n=182/90,017) as well.Increased risk of bowel injury is associated with endometriosis and the abdominal surgical approach to hysterectomy. These findings have implications for the surgical care of women with benign uterine disease.
|
| [3] |
| [4] |
|
| [5] |
| [6] |
吴云肖, 唐瑜芬, 郑莉, 等. 重度阻塞性睡眠呼吸暂停患儿临床预测模型研究[J]. 中国实用儿科杂志, 2022, 37(9):701-707. DOI:10.19538/j.ek2022090611.
|
| [7] |
罗玉婷, 陈纯玲, 郭予雄, 等. 先天性心脏病术后脓毒症患儿列线图预后模型的构建与验证[J]. 中国实用儿科杂志, 2024, 39(2):119-124.DOI:10.19538/j.ek2024020609.
|
| [8] |
张沥今, 魏夏琰, 陆嘉琦, 等. Lasso回归:从解释到预测[J]. 心理科学进展, 2020, 28(10):1777-1791. DOI: 10.3724/SP.J.1042.2020.01777.
传统的最小二乘回归法关注于对当前数据集的准确估计, 容易导致模型的过拟合, 影响模型结论的可重复性。随着方法学领域的发展, 涌现出的新兴统计工具可以弥补传统方法的局限, 从过度关注回归系数值的解释转向提升研究结果的预测能力也愈加成为心理学领域重要的发展趋势。Lasso方法通过在模型估计中引入惩罚项的方式, 可以获得更高的预测准确度和模型概化能力, 同时也可以有效地处理过拟合和多重共线性问题, 有助于心理学理论的构建和完善。
|
| [9] |
| [10] |
|
| [11] |
In recent decades, the global rates of cesarean delivery have rapidly increased. Nonetheless, the influence of cesarean deliveries on surgical complications later in life has been understudied.To investigate whether previous cesarean delivery increases the risk of reoperation, perioperative and postoperative complications, and blood transfusion when undergoing a hysterectomy later in life.This registry-based cohort study used data from Danish nationwide registers on all women who gave birth for the first time between January 1, 1993, and December 31, 2012, and underwent a benign, nongravid hysterectomy between January 1, 1996, and December 31, 2012. The dates of this analysis were February 1 to June 30, 2016.Cesarean delivery.Reoperation, perioperative and postoperative complications, and blood transfusion within 30 days of a hysterectomy.Of the 7685 women (mean [SD] age, 40.0 [5.3] years) who met the inclusion criteria, 5267 (68.5%) had no previous cesarean delivery, 1694 (22.0%) had 1 cesarean delivery, and 724 (9.4%) had 2 or more cesarean deliveries. Among the 7685 included women, 3714 (48.3%) had an abdominal hysterectomy, 2513 (32.7%) had a vaginal hysterectomy, and 1458 (19.0%) had a laparoscopic hysterectomy. In total, 388 women (5.0%) had a reoperation within 30 days after a hysterectomy. Compared with women having vaginal deliveries, fully adjusted multivariable analysis showed that the adjusted odds ratio of reoperation for women having 1 previous cesarean delivery was 1.31 (95% CI, 1.03-1.68), and the adjusted odds ratio was 1.35 (95% CI, 0.96-1.91) for women having 2 or more cesarean deliveries. Perioperative and postoperative complications were reported in 934 women (12.2%) and were more frequent in women with previous cesarean deliveries, with adjusted odds ratios of 1.16 (95% CI, 0.98-1.37) for 1 cesarean delivery and 1.30 (95% CI, 1.02-1.65) for 2 or more cesarean deliveries. Blood transfusion was administered to 195 women (2.5%). Women having 2 or more cesarean deliveries had an adjusted odds ratio for receiving blood transfusion of 1.93 (95% CI, 1.21-3.07) compared with women having no previous cesarean delivery.Women with at least 1 previous cesarean delivery face an increased risk of complications when undergoing a hysterectomy later in life. The results support policies and clinical efforts to prevent cesarean deliveries that are not medically indicated.
|
| [12] |
|
| [13] |
|
| [14] |
|
| [15] |
Deep learning neural network models such as multilayer perceptron (MLP) and convolutional neural network (CNN) are novel and attractive artificial intelligence computing tools. However, evaluation of the performance of these methods is not readily available for practitioners yet. We provide a tutorial for evaluating classification accuracy for various state-of-the-art learning approaches, including familiar shallow and deep learning methods. For qualitative response variables with more than two categories, many traditional accuracy measures such as sensitivity, specificity, and area under the receiver operating characteristic curve are not applicable and we have to consider their extensions properly. In this paper, a few important statistical concepts for multicategory classification accuracy are reviewed and their utilities for various learning algorithms are demonstrated with real medical examples. We offer problem-based R code to illustrate how to perform these statistical computations step by step. We expect that such analysis tools will become more familiar to practitioners and receive broader applications in biostatistics.© 2019 John Wiley & Sons, Ltd.
|
/
| 〈 |
|
〉 |
