肿瘤化疗患者输液港相关性血栓的简易风险评分构建及验证
作者: |
1陈洁雅,
1张国龙,
1李平东,
1石海霞,
1宫玉翠
1 广州医科大学附属第一医院呼吸健康研究院,广州 510120 |
通讯: |
宫玉翠
Email: gongyucuifang@sohu.com |
DOI: | 10.3978/j.issn.2095-6959.2020.03.024 |
摘要
目的:了解肿瘤化疗患者输液港相关性血栓的临床发生情况和影响因素,并构建和验证简易风险评分系统,为肿瘤化疗患者的临床护理提供参考。方法:应用自行设计的患者资料调查表回顾性收集2015年1月至2018年12月在广州医科大学附属第一医院呼吸健康研究院留置输液港的肿瘤化疗患者资料,数据应用SPSS 22.0软件进行统计分析。结果:620例患者的输液港相关性血栓发生率为6.5%(40/620),训练样本和验证样本的发生率依次为6.7%(25/372)和6.1%(15/248);通过logistic回归分析可构建包含化疗次数、癌症分期、血栓病史、D-二聚体、同侧中心静脉置管史5个因子的简易评分系统,模型的评分在0~28,可以将其划分为低危(<9分)、中危(9~19分)和高危(>19分)3个风险等级,并且训练样本和验证样本在不同风险等级的血栓发生率差异无统计学意义(P>0.05)。ROC结果示:训练样本和验证样本的曲线下面积分别为0.748(0.707~0.873)和0.837(0.749~0.925),差异均具有统计学意义(P<0.05)。结论:肿瘤化疗患者输液港相关性血栓的临床发生率相对较高,新建的简易评分系统具有一定的预测能力,适用于肿瘤化疗患者输液港相关性血栓的筛查和预防。
关键词:
肿瘤;化学药物治疗;血栓;输液港;影响因素;预测模型
Construction and validation of a simple risk score for venous thromboembolisms related to implanted ports in cancer patients undergoing chemotherapy
CorrespondingAuthor: GONG Yucui Email: gongyucuifang@sohu.com
DOI: 10.3978/j.issn.2095-6959.2020.03.024
Abstract
Objective: To understand the clinical occurrence and influencing factors of venous thromboembolisms related to implanted ports in cancer patients undergoing chemotherapy, and to construct and validate a simple risk scoring system, so as to provide reference for clinical nursing of cancer patients undergoing chemotherapy. Methods: The data of cancer patients undergoing chemotherapy in our hospital from January 2015 to December 2018 were retrospectively collected with self-designed patient information questionnaire. The data were analyzed by SPSS 22.0 software. Results: The incidence of venous thromboembolisms related to implanted ports in 620 patients was 6.5% (40/620), the incidence in training samples and validation samples were 6.7% (25/372) and 6.1% (15/248), respectively. After logistic regression analysis, the frequency of chemotherapy, cancer stage, thrombosis history, d-dimers and ipsilateral central venous catheterization history were included in the simple scoring system. The score of the model ranged from 0 to 28, which could be divided into three risk levels: low risk (<9), medium risk (9–19) and high risk (>19). There was no significant difference in the incidence of thrombosis between training samples and validation samples at different risk levels (P>0.05). ROC results showed that the area under the curve of training samples and validation samples was 0.748 (0.707–0.873) and 0.837 (0.749–0.925), respectively, and there was no statistical significance (P<0.05). Conclusion: The clinical incidence of venous thromboembolisms related to implanted ports in cancer patients undergoing chemotherapy is relatively high. The new simple scoring system has certain predictive ability and is suitable for the screening and prevention of port-related thrombosis in cancer patients undergoing chemotherapy.
Keywords:
cancer; chemotherapy; thrombosis; implanted ports; influencing factors; prediction model