临床药师基于颅内肿瘤患者切除术后感染风险列线图模型的药学服务模式探讨

Discussion on Pharmaceutical Care Model of Clinical Pharmacists Based on Nomogram Model of Infection Risk After Resection of Intracranial Tumor Patients

关键词:  颅内肿瘤  术后感染  Logistic回归  列线图  药学服务

Key words:  intracranial tumor  postoperative infection  Logistic regression  nomogram  pharmaceutical care

摘要:

目的 探讨神经外科颅内肿瘤患者术后感染危险因素及风险列线图模型的建立,基于该模型筛选重点人群实施药学服务。方法 回顾性分析2020年1月-2021年12月盐城市第一人民医院神经外科行颅内肿瘤切除术患者的病历资料,通过单因素分析和多因素Logistic回归分析得到术后感染的危险因素,建立相关列线图预测模型。结果 共收集288例有效病历,其中发生术后感染91例(31.60%)。多因素Logistic分析提示住院总天数(≥ 20 d)、留置导尿管(≥ 21 d)及机械通气是术后发生感染的独立危险因素(P<0.05)。基于以上独立危险因素构建的风险列线图模型,ROC曲线下面积为0.986(95%CI:0.972~1.000)。结论 颅内肿瘤切除术后感染风险的列线图模型预测效能良好,可筛选出术后感染的高危患者,基于该模式实施重点人群药学服务是切实可行的。

Abstract:

OBJECTIVE To explore the risk factors of postoperative infection in neurosurgical patients with intracranial tumors and construct the nomogram model, and to screen key populations based on this model to implement pharmaceutical care. METHODS The medical records of patients who underwent intracranial tumor resection in the Department of Neurosurgery of Yancheng First People's Hospital from January 2020 to December 2021 were retrospectively analyzed, the risk factors of postoperative infection were obtained by univariate analysis and multivariate Logistic regression analysis, and a relevant nomogram prediction model was established. RESULTS A total of 288 valid medical records were collected, including 91 cases(31.60%) of postoperative infection. Multivariate Logistic analysis suggested that the total length of hospital stay(≥ 20 d), indwelling catheter(≥ 21 d), and mechanical ventilation were independent risk factors for postoperative infection(P<0.05). The risk nomogram model constructed based on the above independent risk factors had an area under the ROC curve of 0.986 (95% CI:0.972-1.000). CONCLUSION The nomogram model for predicting the risk of postoperative infection after intracranial tumor resection has good predictive power and can screen out patients with postoperative infection. It is feasible to implement pharmaceutical care for key populations based on this model.