Chin-Yee Nicolas J, Yan Andrew T, Kumachev Alexander, Ko Dennis, Earle Craig, Tomlinson George, Trudeau Maureen E, Krahn Murray, Krzyzanowska Monika, Pal Raveen, Brezden-Masley Christine, Gavura Scott, Lien Kelly, Chan Kelvin
University of Toronto (Chin-Yee, Yan, Kumachev, Ko, Tomlinson, Trudeau, Krahn, Brezden-Masley, Chan); St. Michael's Hospital (Yan, Brezden-Masley); Institute for Clinical and Evaluative Sciences (Ko, Earle); Sunnybrook Health Sciences Centre (Ko, Earle, Trudeau, Chan); University Health Network (Tomlinson, Krzyzanowska), Toronto, Ont.; Kingston General Hospital (Pal), Kingston; Cancer Care Ontario (Gavura); Canadian Centre for Applied Research in Cancer Control (Chan), Toronto, Ont.
CMAJ Open. 2016 Feb 18;4(1):E66-72. doi: 10.9778/cmajo.20150033. eCollection 2016 Jan-Mar.
Adjuvant trastuzumab is the standard of care for patients with HER2 overexpressing breast cancer, but use of trastuzumab may lead to cardiotoxicity. Our goal was to evaluate the relationship between hospital and physician case volume and cardiac outcomes in this population.
In this retrospective cohort study, we identified all female patients in Ontario with a breast cancer diagnosis in 2003-2009 who underwent treatment with trastuzumab through a provincial drug-funding program and linked these patients to administrative databases to ascertain patient demographics, treating hospital and physician characteristics, admissions to hospital, cardiac risk factors, cardiac imaging and comorbidities. Insufficient cardiac monitoring was defined as per the Canadian Trastuzumab Working Group guideline. Cardiotoxicity was defined as receiving fewer than 16 of 18 doses of trastuzumab because of heart failure admission, heart failure diagnosis or discontinuation of the drug after cardiac imaging. We constructed hierarchical multivariable logistic regression models to evaluate the effect of annual hospital volume, cumulative physician volume and treatment period on cardiac monitoring and cardiotoxicity.
Of 3777 women treated by 214 oncologists at 68 hospitals, 918 (24.3%) had insufficient cardiac monitoring and cardiotoxicity developed in 640 (16.9%). Cardiotoxicity occurred in 389 (42.4%) and 251 (8.8%) patients in the insufficient- and sufficient-monitoring groups, respectively. Higher annual hospital and cumulative physician volumes, and more recent calendar period, were all independent predictors for decreased cardiotoxicity. Adjustment for rates of cardiac monitoring annulled the relationships between case volume and cardiotoxicity.
Greater hospital and physician case volumes are associated with reduced rates of trastuzumab-related cardiotoxicity, most likely because of better cardiac monitoring at higher volume centres.
辅助性曲妥珠单抗是HER2过表达乳腺癌患者的标准治疗方法，但使用曲妥珠单抗可能会导致心脏毒性。我们的目标是评估该人群中医院和医生病例数量与心脏结局之间的关系。
在这项回顾性队列研究中，我们确定了2003年至2009年安大略省所有通过省级药物资助计划接受曲妥珠单抗治疗的乳腺癌女性患者，并将这些患者与行政数据库相链接，以确定患者的人口统计学特征、治疗医院和医生特征、住院情况、心脏危险因素、心脏成像和合并症。根据加拿大曲妥珠单抗工作组指南定义心脏监测不足。心脏毒性定义为因心力衰竭住院、心力衰竭诊断或心脏成像后停药而接受的曲妥珠单抗剂量少于18剂中的16剂。我们构建了分层多变量逻辑回归模型，以评估年度医院病例数量、累积医生病例数量和治疗时间对心脏监测和心脏毒性的影响。
在68家医院的214名肿瘤学家治疗的3777名女性中，918名（24.3%）心脏监测不足，640名（16.9%）发生了心脏毒性。心脏监测不足组和充足组分别有389名（42.4%）和251名（8.8%）患者发生心脏毒性。年度医院病例数量和累积医生病例数量增加以及更近的日历时间都是心脏毒性降低的独立预测因素。对心脏监测率进行调整后，病例数量与心脏毒性之间的关系消失。
医院和医生的病例数量增加与曲妥珠单抗相关心脏毒性发生率降低相关，最可能的原因是病例数量较多的中心心脏监测更好。
