Objectives

The aim of this study was to evaluate institutional variability in high radiation doses during percutaneous coronary intervention (PCI).

Background

It is unknown whether radiation safety practices are optimally applied across institutions performing PCI.

Methods

Using data from a large statewide registry, PCI discharges between July 1, 2016, and March 31, 2018, with a procedural air kerma (AK) recorded were analyzed. PCI procedures were grouped by the performing hospital, and institutional frequency of procedural AK ≥5 Gy was calculated. Fitted hierarchical Bayesian modeling was performed to identify variables independently associated with an AK ≥5 Gy. The performing hospital was included as a random effect in the hierarchical model.

Results

Among 36,201 PCI procedures at 28 hospitals, procedural AK was ≥5 Gy in 1,477 cases (4.1%), ≥10 Gy in 185 (0.5%), and ≥15 Gy in 105 (0.3%). The institutional frequency of procedural AK ≥5 Gy ranged from 0.0% to 10.9%. Bayesian modeling identified body mass index, dyslipidemia, diabetes, prior coronary bypass surgery, use of mechanical circulatory support, and the performing hospital as independent predictors of an AK ≥5 Gy. The median odds ratio for the performing hospital, representing an estimate of the contribution of interhospital variability in determining the odds of having a procedural AK ≥5 Gy, was 3.08 (95% confidence interval: 3.01 to 3.16).

Conclusions

Wide variability exists in the institutional frequency of procedural AK ≥5 Gy during PCI. After accounting for patient characteristics and procedural variables, the performing hospital appears to be a major factor in determining patient radiation dose in contemporary PCI.

中文翻译：

---

经皮冠状动脉介入治疗期间患者放射剂量≥5Gy的机构变异性

目标

这项研究的目的是评估经皮冠状动脉介入治疗（PCI）期间高剂量放射的机构变异性。

背景

未知辐射安全实践是否在执行PCI的机构中得到最佳应用。

方法

使用来自全州大型注册表的数据，分析了2016年7月1日至2018年3月31日之间的PCI放电情况，并记录了程序性空气比释动能（AK）。PCI手术按执行医院分组，并计算手术AK≥5Gy的机构频率。进行拟合的分层贝叶斯建模以识别独立于与AK≥5Gy相关的变量。绩效医院作为随机效应包括在层次模型中。

结果

在28家医院的36201例PCI手术中，程序性AK在1477例病例中≥5Gy（4.1％），在185例中≥10Gy（0.5％），在105例中≥15Gy（0.3％）。程序性AK≥5Gy的机构频率为0.0％至10.9％。贝叶斯模型确定了体重指数，血脂异常，糖尿病，冠状动脉搭桥手术之前的经验，机械循环支持的使用以及表现良好的医院为AK≥5 Gy的独立预测因子。表现医院的中位优势比为3.08（95％的置信区间：3.01至3.16），代表医院间变异性对确定程序性AK≥5Gy的优势的估计。

结论

PCI期间手术AK≥5 Gy的机构频率存在很大差异。在考虑了患者特征和程序变量之后，执行医院似乎是确定当代PCI中患者放射剂量的主要因素。