Primary operator radiation dose in the cardiac catheter laboratory

Abstract

Objectives:

Radiation from cardiac angiography procedures is harmful to patients and the staff performing them. This study sought to investigate operator radiation dose for a range of procedures and different operators in order to investigate trends and optimise dose.

Methods:

Real-time dosemeters (RTDs) were worn by operators for angiography procedures for 3 years. Dose–area product (DAP) and RTD were collected. RTD was normalised to DAP (RTD/DAP) to compare radiation dose and radiation protection measures. Comparisons were made across procedure categories and individual operators.

Results:

In 7626 procedures, median and 75th percentile levels were established for operator dose for 8 procedure categories. There was a significant difference in all operator dose measures and DAP across procedure categories (p<0.001). DAP, RTD, and RTD/DAP were significantly different across 22 individual operators (p<0.001).

Conclusion:

DAP was significantly different across procedure categories and a higher RTD was seen with higher DAP. RTD/DAP can demonstrate radiation protection effectiveness and identified differences between procedures and individual operators with this measure. Procedures and individuals were identified where further optimisation of radiation protection measures may be beneficial. A reference level for operator dose can be created and audited against on a regular basis.

Advances in knowledge:

This study demonstrates that operator dose can be easily and routinely measured on a case by case basis to investigate dose trends for different procedures. Normalising the operator dose to DAP demonstrates radiation protection effectiveness for the individual operator which can then be optimised as part of an ongoing audit program.

Introduction

Radiation used during cardiac angiography and interventional procedures is known to be harmful to both the patients undergoing the procedure and the staff performing them.^1–3 Determining how much radiation is used for a particular procedure is important in estimating the risk to the patient from the radiation and reference levels are useful in ensuring that the dose is optimised. Most commonly, the dose–area product (DAP) is the comparative measure for reference levels (termed diagnostic reference levels (DRL)) and its use is advocated by the international committee of radiation protection (ICRP), who have outlined the establishment and use of DRLs in ICRP 135.⁴ The DRL is used for local dose audit activity in order to optimise medical imaging equipment, imaging protocols and procedures, ultimately optimising patient dose. Reducing DAP (patient dose) should in turn result in the reduction of the dose to the staff performing the procedure.⁵ However, the relationship between staff and patient dose may not be simply a linear one. Optimising equipment and imaging protocols does not necessarily mean a similar level of optimisation of staff dose, as operator performance factors will affect this. For example, the optimal use of the radiation protection principles of time (fluoroscopy time and number of digital acquisitions), distance (how far the operator stands from the patient during exposure) and shielding use (effective use of ceiling suspended and table side shields) have significant effects on operator dose, regardless of the level of equipment and protocol optimisation.This is outlined in some detail by the ICRP in publication 139.⁶ The dose to the primary catheter operator will be the highest of all the staff present, since they are the closest to the X-ray source. Typically, their dose is measured using thermoluminescent dosemeters that are worn continuously by each staff member and read once a month. Though these dosemeters are important in continuously auditing the dose to the individual, they have limitations. Firstly, the results are not available immediately. Secondly, they only measure dose accumulated over a period of time and hence do not display results for each procedure. Determining how much dose is received per procedure or when especially high doses were received is not possible. This can only be easily achieved by measuring dose to the operator for each individual procedure, using real time digital dosemeters (RTDs). Whilst the DAP level and the resulting operator RTD dose can be measured, an indication of the effective use of the available shielding of the operator can be also achieved by dividing the RTD dose by the DAP, and this can then be used to optimise an individual operator’s radiation protection practices.

The aim of this study was to measure dose to operators performing a range of cardiac angiography procedures and investigate differences between procedure types and the individual operators. In addition, it was envisaged that a reference dose for the operator could be established for these procedures. This would allow individual operators to compare their exposure with others, giving them a simple benchmark that if they routinely exceed should trigger a review of their radiation safety practices.

Methodology

The study was a retrospective analysis of prospectively collected radiation dose data collected in a single tertiary centre from March 2015 to September 2018 and the hospital human research ethics committee approved the study.

Four different X-ray systems were used during the study period [Siemens Axiom Artis dBc, (Siemens Healthcare, Erlangen, Germany) Philips Allura Clarity FD10/10 (Philips Healthcare, Netherlands), and two Siemens Artis Q biplane]. All X-ray systems had a DAP meter housed within the X-ray c-arm assembly and had their DAP meters calibrated on an annual basis as part on ongoing compliance testing.

DAP was collected as the primary measure of patient dose. Other measures that were collected were:procedure type, fluoroscopy time, air kerma at the reference point, patient age, patient gender, patient BMI, catheter access route, operator dose and operator.

Operator dose was measured with RTDs, (Raysafe^TMi2 (Unfors,Billdal, Sweden)) which were worn high on the left side of the lead collar, on the outside of the lead apron. In line with previous studies that used similar dosemeters,⁷ the dosemeter measures the Hp₁₀ dose (an estimate of dose 10 mm below the skin surface). However, since the RTD was worn above the lead apron, the effective dose calculation is made by dividing the Hp₁₀ dose by 21. This can give a whole body effective dose measurement – E⁸

The dosemeter was worn by the primary and secondary operator (when present) for each procedure and handed to the radiographer who placed the dosemeter in a dedicated reader which was connected to a computer and read by the Raysafe^TM software. The dosemeter was then cleared prior to the next procedure.

Standard radiation protection measures at this facility consisted of table side drapes of 1 mm lead equivalence and a ceiling suspended lead acrylic shield of 0.5 mm lead equivalence. Operators wore lead aprons and skirts with a lead equivalence of 0.5 mm at the front, thyroid collar and lead goggles. These measures are mandated at this centre for all procedures. Some operators also wore lead caps.

Individual procedures were grouped into 29 procedure categories, however, in line with other studies creating reference levels^9,10 and guidelines⁴ only procedure categories with >50 individual procedures were included in the final analysis and only procedures where the RTD was worn by the primary operator were included in the analysis. Median, 25th and 75th percentile values were generated for DAP and RTD.

As the dose to the RTD (and the calculated E dose) is a function of the DAP, RTD was normalised to DAP by dividing RTD by DAP (RTD/DAP). RTD/DAP was the key measure in demonstrating the effective use of radiation protection measures. RTD/DAP was compared across procedure categories to investigate radiation protection differences between procedure categories and between operators.

Each operator was anonymised and allocated a number for comparison between operators.

As with the procedure categories, only operators who had performed >50 procedures each were included in this part of the analysis. Again, Median, 25th and 75th percentile values were generated for DAP and RTD and RTD/DAP and were compared across operators.

Previous studies had noted a significant difference in operator dose between femoral and radial access.¹¹ Therefore, procedures were also split by catheter access route to investigate differences in RTD/DAP for radial and femoral access.

Analysis

Continuous data were tested for normal distribution and median (interquartile ranges) or means ( ± standard deviation) were used to describe data. Mann–Whitney U tests or Kruskal–Wallis tests were used to compare median values across different categories or groups. χ² tests were used to compare categorical variables. SPSS v. 23 (International Business Machines (IBM) Armonk, NY, USA) was used for statistical analysis.

Results

Of the 10,317 procedures performed in study period, 7849 were performed with an RTD worn by the primary operator. Of the 29 procedure categories, there were 8 procedure categories with >50 procedures, leaving 7626 procedures meeting inclusion for analysis.

Median (interquartile range) patient age was 67 (58–75) years and 67% were male. Radial access was most common and used in 71.2% of cases. Overall median (interquartile range) DAP was 24.6 (13.1–44.8) Gycm², median RTD was 7(2-18) µSv, and median RTD/DAP was 0.3 (0.1–0.7) µSv/Gycm².

Procedure categories

The 8 procedure categories with >50 procedures and their various measures are given in Table 1. The elements of the procedures that led to the grouping into the eight classifications are outlined below:

Table 1.

Baseline patient and radiation measures for the different procedure categories

	Aortic valvuloplasty	Cardiac biopsy	Complex PCI	Diagnostic	CTO	Grafts	PCI	RHC
N=	111	63	228	5205	70	413	1406	130
Age (years)	84 (78.0–88.0)	60 (41–64)	69 (61–77)	67 (58–75)	67 (57–72)	73 (65–79)	65 (57–74)	56 (46–67)
BMI (kg/m2)	25.9 (22.7–30.1)	27.1 (23.9–30.5)	29.4 (26.0–33.3)	29.3 (25.5–33.8)	29.9 (27.2–34.8)	29.4 (26.2–33.4)	29.1 (25.6–33.0)	28.5 (24.3–33.0)
Male gender	61 (55%)	53 (84%)	174 (76%)	3264 (63%)	59 (84%)	356 (86%)	1054 (75%)	71 (55%)
FT (min)	10.2 (7.5–15.5)	3.2 (2.3–5.5)	29.0 (20.4–40.4)	4.7 (3−8.1)	56.6 (41.7–92.0)	10.3 (6.9–15.2)	16.2 (11.5–23.7)	4.7 (3.0–8.7)
AK (mGy)	170 (85–424)	30 (14–50)	1702 (1053–2658)	326 (188–537)	2397 (1817–3788)	550 (345–840)	1042 (637–1700)	42 (15–103)
DAP (Gycm2)	15.9 (9.6–37.2)	2.7 (1.3–5.2)	80.8 (49.5–115.8)	19.3 (11.3–32.1)	124.9 (78.2–188.0)	35.8 (22.3–53.1)	49.3 (32.2–81.5)	4.6 (1.5–10.3)
RTD (µSv)	5 (2–18)	2 (1–7)	21 (8–51)	6 (2–15)	32 (14–75)	5 (2–12)	13 (4–21)	1 (0–4)
E (µSv)	0.2 (0.1–0.9)	0.1 (0.1–0.3)	1.0 (0.4–2.4)	0.3 (0.1–0.7)	1.5 (0.7–3.6)	0.2 (0.1–0.6)	0.6 (0.19–1.5)	0.1 (0.0–0.2)
RTD/DAP (µSv/Gycm2)	0.2 (0.1–0.8)	1.4 (0.1–3.1)	0.3 (0.1–0.6)	0.3 (0.1–0.7)	0.3 (0.1–0.7)	0.2 (0.1–0.3)	0.3 (0.1–0.6)	0.2 (0.0–1.1)
ODRL - RTD/DAP (µSv/Gycm2)	0.3	2.4	0.4	0.5	0.3	0.3	0.3	0.4

AK, air kerma; BMI, body mass index; CTO, chronic total occlusion; DAP, dose area product; E, calculated effective dose; FT, fluoroscopy time; ODRL, operator dose reference level; PCI, percutaneous coronary intervention; RHC, right heart catheterisation; RTD, real time dosimeter reading.

This table demonstrates the median values (IQR) for the various measures collected for each procedure category. All measures demonstrated a significant difference across procedure categories to the p<0.001 level.

Diagnostic procedures include: coronary angiography ± left ventriculography ± aortography (AO), ± fractional flow reserve (FFR), ± intravascular ultrasound (IVUS), ± optical coherence tomography (OCT), ± iliac/femoral angiography.

Grafts procedures includes coronary angiography and angiography of saphenous vein grafts and arterial grafts.

Percutaneous coronary intervention (PCI) includes the criteria for diagnostic procedures in addition to single vessel angioplasty ± stenting. Procedures with prior diagnostic procedures and staged PCI were also included in this category.

Complex PCI includes the elements for PCI category + rotational atherectomy or multivessel PCI or IVUS/OCT-guided PCI. Procedures with prior diagnostic procedures and staged complex PCI were also included in this category.

The Chronic total occlusion (CTO) category is for stable lesions of chronic totally occluded vessels in an elective presentation ± complex PCI elements.

Aortic valvuloplasty includes balloon valvuloplasty of the aortic valve ± any diagnostic elements.

Right heart catheterisation (RHC) includes only venous catheterisation and insertion of a catheter for invasive pressure wave analysis ± venous oxygen saturation measurement.

Cardiac biopsy category includes procedures where there were myocardial biopsy samples taken ± RHC elements.

There was a significant difference in DAP (p < 0.001), RTD (p < 0.001), and E(p < 0.001) across procedure categories with CTO intervention having the highest of all these values and cardiac biopsy having the lowest DAP and RTD. However, right heart catheterisation studies had the lowest RTD/DAP (p < 0.001) and cardiac biopsy the highest (p < 0.001) (Table 1).

There was a significant difference in overall DAP between the four X-ray units used, with the Siemens Artis dBc (SAdBc) having the highest DAP and the second Siemens Artis Q (SAQB2) system having the lowest. RTD/DAP was compared across the four X-ray units and the first Siemens Artis Q biplane system (SAQB1) showed the highest RTD/DAP values and the SAdBc the lowest RTD/DAP (p < 0.001) (Table 2).There was a significant difference between procedures with regard to catheter access. The differences between radial and femoral access are demonstrated in Table 3.

Table 2.

Comparison of X-ray equipment

Equipment	DAP (Gycm2)	RTD/DAP (µSv/Gycm2)
SAQB1	22.3 (11.5–42.8)	0.3 (0.1–0.7)
SAdBc	37.10 (22.5–61.5)	0.2 (0.1–0.6)
PAC	24.97 (16.3–42.0)	0.2 (0.1–0.6)
SAQB2	18.1(10.1–36.0)	0.3 (0.1–0.7)
p=	<0.001	<0.001

DAP, dose area product; PAC, Philips Allura Clarity; RTD/DAP, dosimeter dose normalised to DAP; SAQB1, Siemens Artis Q Biplane 1; SAQB2, Siemens Artis Q Biplane 2; SAdBc, Siemens Artis dBc.

Table 3.

Dose differences related to access

	Radial	Femoral	Other	p-value
N=	5426	2020	180
Age (years)	66 (57–73)	72 (63–80)	58 (46–65)	<0.001
BMI(kg/m2)	29.4 (25.7–33.7)	28.7 (24.9–32.9)	28.1 (24.2–31.6)	<0.001
Male Gender	3603 (66%)	1370 (68%)	121 (67%)	0.509
DAP (Gycm2)	23.4 (13.0–42.1)	30.6 (15.7–54.4)	3.7 (1.5–7.3)	<0.001
RTD (µSv)	8 (3–21)	5 (2–13.5)	2 (0–6)	<0.001
E (µSv)	0.4 (0.1–1.0)	0.3 (0.1–0.6)	0.1 (0.0–0.3)	<0.001
RTD/DAP(µSv/Gycm2)	0.3 (0.1–0.8)	0.2 (0.1–0.4)	0.4 (0–2.5)	<0.001

BMI, body mass index; DAP, dose area product; E, calculated effective dose; RTD, real time digital dosimeter reading.

This table demonstrates the differences in radiation dose measures in relation to catheter access route and demonstrates that operator dose is generally higher for radial access but the effective dose normalised to DAP (RTD/DAP) ratio is highest with ‘other’ access routes

Interoperator comparisons

There were 22 different operators, of varying experience who had performed >50 procedures in the study. Median DAP and operator dose (RTD) were significantly different across operators (p < 0.001) (Figures 1 and 2). RTD/DAP was also significantly different across operators (Figure 3) (p < 0.001). Trainee operators used less DAP than specialist (consultant) operators (21.0 (12.0–37.7) vs 27.3 (14.0–49.7) p < 0.001) and had slightly lower RTD/DAP than specialist operators (0.2 (0.1–0.5) vs 0.3 (0.1–0.7) p < 0.001).

Figure 1.

DAP comparisons between operators. This figures graphically quantifies the difference in median (interquartile range) DAP across individual operators. Error bars demonstrate 95% CI, °=Outlier, *=Extreme outlier. CI, confidence interval; DAP, dose–area product

Figure 2.

Radiation dose comparisons across operators. This figure graphically demonstrates the difference in median (interquartile range) RTD across individual operators. Error bars demonstrate 95% CI, °=Outlier, *=Extreme outlier. CI, confidence interval; RTD, real time dosemeter.

Figure 3.

Operator dose normalised to DAP. This figuregraphically quantify the difference in median (interquartile range) RTD normalised to DAP across individual operators to demonstrate the effectiveness of radiation protection measures between the operators. Error bars demonstrate 95% CI, °=Outlier, *=Extreme outlier. CI, confidence interval; DAP, dose area product; RTD, real time dosemeter.

Discussion

Patient dose levels (DAP) for the procedure categories in this study can be compared to previous studies and the results compare favourably, with a median DAP of 49.3 Gycm² for PCI procedures compared to 87.4 Gycm² in a local DRL study.⁹ The median DAP for the complex PCI category in the present study is 80.8 Gycm² which is again lower than the 2014 study.⁹ The lower dose is possibly a result of using newer equipment and possibly changes in imaging protocols. The highest DAP in the present study is for CTO procedures, with a median value of 124.9 Gycm². This compares to a mean value of 119.8 Gycm² in a recent study investigating CTO procedures performed on different imaging systems.¹²

Some studies have published operator dose data, in addition to the patient data as part of randomised trials and observational studies.^11,13,14 One study found that operator effective doses per procedure ranged from 0.02 to 38 µSv for diagnostic procedures and 0.17 to 31 µSv for PCI procedures.¹⁵ This compares to a median of 0.3 µSv for diagnostic procedures and 0.6 µSv for PCI procedures in this study.

A particular aim of this study was the comparison of operator dose across procedure categories and between individual operators in the same facility. When comparing operator dose, one should use caution at comparing solely the RTD or E value, as some operators will perform only one type of procedure, perhaps ones with a relatively high or low dose. Comparing operator RTD dose for the same procedure type is warranted but the RTD/DAP ratio measure has significant value in this setting. The RTD and resulting E value are a direct function of the amount of patient exposure (DAP) used for the procedure. Therefore, if the radiation protection measures are used equally effectively for all procedure types, the amount of radiation to the operator per unit DAP should be similar. The RTD/DAP normalises the DAP and gives insight into the effective protection of operators and should be relatively consistent across procedure categories and individual operators given the similar position of the operator to the patient for all procedures. By measuring the dose to operators and comparing the RTD/DAP, one may get a better insight of an individual’s radiation protection practice.

Figures 1–3 demonstrate how dose differs between operators. Some operators receive high doses, and some receive low doses, though once normalised to the DAP, the RTD/DAP value is more consistent between operators, but it does demonstrate that some operators could be further optimised (Figure 3).

If significantly different RTD/DAP values exist for the same operators performing the same procedures, in different rooms, the difference may be due to the protection measures of a particular room and this should be investigated and further optimised. Likewise, an RTD/DAP difference between operators in the same room, with the same protection measures in place could indicate poor use of the protection apparel or could be an indication of differences in practice: i.e. the operator stands closer/further from the patient.Though not investigated here, a difference in RTD/DAP could also be due to the table height difference between operators and operator height, which has been shown previously to impact on operator dose.^16,17

Table 2 demonstrates that there is a significant difference in DAP between the four X-ray systems used in this study. The RTD/DAP between the systems also significantly varies and this demonstrates that there could possibly be more optimisation performed in the radiation protection/shielding measures across the different systems/rooms. The X-ray system with the highest RTD/DAP (SAQB1) is the system where the cardiac biopsy procedures were performed, which may have led to the RTD/DAP differences.

Radial access has a higher RTD and RTD/DAP when compared to femoral access in this study. Higher operator doses and higher dose normalised to DAP for radial access has been noted previously by Sciahbasi et al in their study of acute patients undergoing PCI. Their effective doses of 2.3 µSv for radial access and 1.2 µSv for femoral access appear higher than the calculated effective dose for PCI in this study and the dose normalised to DAP of 0.8 and 0.5 µSv /Gycm² also appear to be higher than the median 0.3 µSv /Gycm² calculated for PCI in this study. The difference in dosemeter type and conversion methodology could explain this difference and, in addition, the dosemeter was placed in the outside pocket of the lead apron, as opposed to the thyroid collar, as in this study.¹¹ The higher radiation doses associated with radial access have been noted in other studies^18,19 and is likely due to the operator standing closer to the patient, thus receiving more backscatter. In addition to this, the ceiling mounted shield may be more difficult to use or less effective with radial access. The “other“ access category in this study has the highest RTD/DAP which is likely due to the jugular access route for cardiac biopsy procedures. The operators performing these procedures also had the highest RTD/DAP levels. The high RTD/DAP is possibly due to a difficulty in effectively utilising the under table and ceiling mounted lead shielding for this access type. With this information, these operators’ doses may now be further optimised.

Another finding was that trainee operators used less DAP and this is likely due to consultant operators taking on the more complex and therefore higher dose procedures. However, trainee operators had a lower overall RTD/DAP, all be it a small difference. The reasons for this are unknown but it indicates that trainee operators used their radiation protection measures more effectively than consultants and this warrants further investigation.

Reference levels for audit

The ICRP advise in ICRP 135 to use the 75th percentile value of the distribution of the medians for the DRL in studies investigating patient dose. The median values for a procedure, usually for a room or centre, should fall below the established DRL.⁴ If any facility consistently obtains a median result higher than the 75th percentile, then that facility should investigate why this may be the case.^4,5,20 The same methodology could be adapted for operator dose, where in addition to measuring operator dose for different procedure categories, median and 75th percentile values for RTD and RTD/DAP can be calculated for each operator. When setting an operator dose reference level, this could be established using the 75th percentile of the distribution of the medians for a particular procedure (Table 1). Those with a median dose above this level should review their doses and discuss further optimisation strategies with a medical physicist. These operator dose reference levels can be calculated on an annual basis for common procedures for ongoing auditing, with a view to optimisation. Other centres could also adopt this methodology and compare their doses. In addition, this methodology is not only relevant to cardiac angiography procedures but could be implemented for any fluoroscopically guided procedure.

Limitations

The main limitation of this study is that it is a single centre study. However, the methodology has the most utility in the single centre where many individual operators may be using the same equipment. Differences in dose can then be explored and radiation protection measures/practices can be optimised. However, it should be noted that where DRL studies assess dose for a population of patients, the ODRL compares doses to individuals, which is quite different. Individuals have attributes that may impact on dose (e.g. height) which could affect their dose and subsequent success of optimisation.

Conclusion

This study demonstrates how readily available digital dosimeters (RTDs) can be used on a case by case basis, as a core part of operational workflow in the CCL. They can be used to measure and compare operator dose between operators and across different procedure types with the aim of dose optimisation. By normalising the operator dose to the DAP (RTD/DAP), the effectiveness of radiation protection strategies can also be compared. This methodology is an important auditing tool for managing the occupational radiation exposure of staff working in the CCL. A reference level can be created and audited against, with the goal of further optimisation.