Exposure to ionizing radiation (IR) is a consequence of the improvement in minimally invasive techniques used in interventional cardiology. While this contributes to improved knowledge regarding coronary heart disease (CAD), the introduction of more sophisticated catheters and stents may lead to more complex and time-consuming interventions.1,2

Percutaneous coronary intervention (PCI) is the first line diagnostic and therapeutic method to treat acute coronary syndromes (ACS) and is also an important therapeutic strategy in symptomatic chronic coronary syndromes (CCS). The downside of such procedure is the potentially harmful exposure to IR. The two main biological effects of IR are deterministic (direct effects in tissues like skin burns and cataracts) and stochastic effects (long term carcinogenic alterations), and the risk of radiation-induced cancer increases in direct proportion to effective dose.3–6

Thus, interest in radiation doses in clinical practice has garnered greater interest recently, in particular, the exposure of patients and catheterization laboratory (Cath-Lab) personnel to radiation. Nevertheless, very few Cath-Labs communicate their patients’ IR exposure data.7–10

Diagnostic reference levels (DRLs) are an essential tool to promote optimization in IC procedures, highlighted by the European Commission Basic Safety Standard.11,12 Air kerma area product (PKA) and cumulative air kerma at patient entrance reference points (Ka,r) are the dose descriptors used to calculate DRLs, which are key to defining which levels may lead to harmful consequences.11,13

Diagnostic reference levels should be set and regularly updated at local (hospital) and national levels by the competent national health authority. Fourteen years ago, the SENTINEL project and more recently, the European Commission for Radiation Protection (EUCLID) suggested DRLs in interventional cardiology. There have also been some recent DRLs studies focusing on coronary angiography (CA) and PCI, however, for complex procedures such as the treatment of coronary chronic total occlusions (CTO), left main coronary artery (LMCA) or structural interventions, information on DRLs remain scarce.2,11,14,15

The goal of this study was to establish, for the first time, local (hospital) DRLs for complex PCI procedures, encouraging other Cath-Labs to establish their own, thus contributing to setting national DRLs.

The study aimed to access the IR dose exposure during complex percutaneous coronary intervention (PCI), and to set the respective DRLs (primary outcome) and future achievable doses (ADs) (secondary outcome).

This was a single center retrospective cohort study, taking place between 2019 and 2020, involving patients who underwent complex PCI at a local hospital in Portugal.

The technological evolution in IC allowed for a more generalist approach to CAD, not only for diagnostic purposes, but also in therapeutic procedures for more complex coronary disease. Therefore, procedures have become more time-consuming and thus increasingly subjected to IR. For this reason, professionals are becoming more aware of the adverse effects of radiation, and safer techniques and equipment are needed to minimize radiation exposure. The focus on complex disease and PCI should not, therefore, compromise the safety methods used to reduce exposure. Cath-Lab IR dose exposure data during clinical practice are scarce, and evidence from large radiation dose surveys is missing in the literature.2,4,7

As previously defined, complex PCI encompassed patients undergoing CTO or LMCA. Male gender was associated with significantly higher radiation exposure measured by PKA and Ka,r. This finding corroborates that of the PROTECTION VIII study, but reasons for this remain unclear. One plausible explanation in the present study was the higher body surface area and an expected IR dose exposure.7,17,18

The characteristics of CTO lesions and the complexity of the technical procedures makes this intervention more time-consuming and requiring greater image optimization (thus, higher rates of radiation time and contrast doses used). In the present study, CTO and LMCA PCI and fluoroscopy time were not statistically different and CTO had similar procedure times compared to other European and international CTO registries.19–21 Interestingly, a more time-consuming procedure in CTO group was not related to significantly higher IR dose exposure measured in Ka,r and PKA. The CTO group had also more ad-hoc procedures, which are associated with greater PKA radiation exposure in European registries.2 In our analysis, ad-hoc regimes showed statistically significant lower rates of IR exposure, probably associated with less severe CAD, although without statistical significance (57.1% vs. 65.4%, p=0.245). IR exposure was not associated with the complexity of CAD, probably due to the complex anatomy in the entire sample.

Diagnostic radiation levels help identifying those facilities that are using unusual high IR exposure doses, and are a relevant tool for optimization in IC.11,16,22 As mentioned above, the DRLs were calculated based on the third quartile of the median for the defined IR doses (Ka,r and PKA).13 There is very little international data on DRLs for complex PCI. European registries reflect DRLs for CA, PCI and both procedures. The IR dose exposure and DRLs presented in international registries (Table 5) were significantly lower than those presented in the current study, which can be explained by scarce complex procedures data.7,11,15,23,24 The European register of Siiskonen et al. showed IR dose and respective DRLs according to the type of interventional cardiology procedure. The data reported for complex PCI were specific for CTO procedures, and albeit scarce, were closer to the DRLs presented in the current study. In international CTO registries, the IR dose exposure was also similar to those presented.2,20,21 Miller et al.,25 in a North American register, presented DRLs superior to European data and similar to those in this study, and argue that complex procedures were probably the main reason for the differences reported (Table 4).11

To emphasize the “as low as reasonably achievable” (ALARA) principle, optimization of DRLs is encouraged by using the ADs. ADs are set at approximately the 50th percentile (median) of the study dose distribution. In the present study, ADs were set and should supplant DRLs in future assessments.16

Our multivariate analysis showed that independent predictors of IR dose exposure with positive correlation was demonstrated between Ka,r and PKA, Ka,r and fluoroscopy time and Ka,r and PCI time, which corroborates the results from international registries.2,7,11,23

Cath-Labs IR safety performance should be critically reviewed, and safety methods to minimize IR exposure should be implemented. These results are an opportunity for improvement in radiation safety quality practice.

The participation of institutions in IR dose exposure surveys are important, because it will enable the setting of national DRLs to improve radiation “best practice” in Cath-Labs and consequently ensure safer procedures for patients and laboratory personnel. Cath-Lab certification in IR should also become a reality, helping doctors and patients to select safer centers with low IR dose exposure.

This study reports the results of IR in complex PCI. DRLs were set for IR dose exposure measured in Ka,r (3012 mGy) and PKA (162 Gy cm2). DRLs in the study were relatively higher than the DRLs reported in international registries, but data from complex PCI in these registries are scarce. ADs were set in Ka,r (1917 mGy) and PKA (101 Gy cm2) and DRLs should be updated taking these results into account. National surveys should be a reality, and probably will allow the implementation of new safety methods to minimize IR exposure.

Some limitations must be acknowledged. This was an observational study, therefore collected data are limited to medical records. Our sample included only successful CTO or LMCA PCI. Unsuccessful CTO or LMCA PCI was not address and may limit the validity of the results. Also, defining complex PCI as CTO or LMCA was simplistic and an important limitation. Finally, the data in our study are from only one PCI center (composed of 1 Cath-Lab room) with a small sample, so our conclusions may not be generalizable to other cardiac centers.

The authors have no conflicts of interest to declare.