Author + information
- Received August 22, 2017
- Revision received January 8, 2018
- Accepted January 11, 2018
- Published online March 12, 2018.
- James A. Crowhurst, BSc (Hons)a,b,∗ ( )(, )
- Gregory M. Scalia, MBBS, MMedScia,b,
- Mark Whitby, MSc, PhDc,
- Dale Murdoch, MBBSa,b,
- Brendan J. Robinson, BappScia,
- Arianwen Turner, BappScid,
- Liesie Johnston, BappScid,
- Swaroop Margale, MBBSe,
- Sarvesh Natani, MBBS, MDe,
- Andrew Clarke, MBBSf,
- Darryl J. Burstow, MBBSa,b,
- Owen C. Raffel, MBa,b and
- Darren L. Walters, MBBS, MPhila,b
- aCardiology Department, The Prince Charles Hospital, Chermside, Queensland, Australia
- bUniversity of Queensland, St. Lucia, Queensland, Australia
- cBiomedical Technical Services, The Prince Charles Hospital, Chermside, Queensland, Australia
- dMedical Imaging Department, The Prince Charles Hospital, Chermside, Queensland, Australia
- eDepartment of Anaesthesia, The Prince Charles Hospital, Chermside, Queensland, Australia
- fDepartment of Cardio-thoracic Surgery, The Prince Charles Hospital, Chermside, Queensland, Australia
- ↵∗Address for correspondence:
Mr. James Crowhurst, Cardiac Investigations Unit, The Prince Charles Hospital, Rode Road, Chermside, Queensland, Australia.
Background Transesophageal echocardiography operators (TEEOP) provide critical imaging support for percutaneous structural cardiac intervention procedures. They stand close to the patient and the associated scattered radiation.
Objectives This study sought to investigate TEEOP radiation dose during percutaneous structural cardiac intervention.
Methods Key personnel (TEEOP, anesthetist, primary operator [OP1], and secondary operator) wore instantly downloadable personal dosimeters during procedures requiring TEE support. TEEOP effective dose (E) and E per unit Kerma area product (E/KAP) were calculated. E/KAP was compared with C-arm projections. Additional shielding for TEEOP was implemented, and doses were measured for a further 50 procedures. Multivariate linear regression was performed to investigate independent predictors of radiation dose reduction.
Results In the initial 98 procedures, median TEEOP E was 2.62 μSv (interquartile range [IQR]: 0.95 to 4.76 μSv), similar to OP1 E: 1.91 μSv (IQR: 0.48 to 3.81 μSv) (p = 0.101), but significantly higher than secondary operator E: 0.48 μSv (IQR: 0.00 to 1.91 μSv) (p < 0.001) and anesthetist E: 0.48 μSv (IQR: 0.00 to 1.43 μSv) (p < 0.001). Procedures using predominantly right anterior oblique (RAO) and steep RAO projections were associated with high TEEOP E/KAP (p = 0.041). In a further 50 procedures, with additional TEEOP shielding, TEEOP E was reduced by 82% (2.62 μSv [IQR: 0.95 to 4.76] to 0.48 μSv [IQR: 0.00 to 1.43 μSv] [p < 0.001]). Multivariate regression demonstrated shielding, procedure type, and KAP as independent predictors of TEEOP dose.
Conclusion TEE operators are exposed to a radiation dose that is at least as high as that of OP1 during percutaneous cardiac intervention. Doses were higher with procedures using predominantly RAO projections. Radiation doses can be significantly reduced with the use of an additional ceiling-suspended lead shield.
Percutaneous interventional procedures, guided by fluoroscopy for structural pathology of the heart, are now commonplace, and procedures such as transcatheter aortic valve replacement (TAVR) offer an alternative treatment option to open-heart surgery (1). These procedures and many others, including left atrial appendage occlusion and transcatheter mitral valve repair/implantations, require guidance with transesophageal echocardiography (TEE) in addition to fluoroscopy (2,3). This exposes the echocardiographer and/or echocardiologist who operate the TEE probe and console to the harmful effects of scattered ionizing radiation. Although recent publications have highlighted the risks of radiation to the staff performing fluoroscopically guided cardiac procedures (4–6), none of these studies were inclusive of radiation dose to TEE operators in this environment.
Primary operators, usually the cardiologist performing the procedure, have tableside and ceiling-suspended lead shields in place to protect them from the harmful effects of radiation. These are usually only in place on the right side of the procedure table, where the primary operator and their assistant stand. There is often no specific additional protection installed at the head end or left side of the procedure table where the TEE operator (TEEOP) would stand. Recent guidelines have highlighted the risks of radiation to TEEOP and the lack of evidence surrounding radiation dose to TEEOP (7). All staff working in this environment are monitored with optically stimulated luminescent dosimeters that are interrogated on a monthly basis. Although all staff wear these dosimeters, they do not show how much radiation dose each staff member may receive on a case-by-case basis.
With the lack of evidence surrounding this issue, this study sought to measure the radiation dose to TEEOP during percutaneous interventional procedures that require TEE guidance and compare their dose to other members of the multidisciplinary heart team. Secondly, we sought to investigate procedural factors that would have an impact on total radiation dose and the TEEOP dose. Additional lead shielding was also implemented for the TEEOP, and the effectiveness of the shielding solution was measured.
This single, tertiary center, observational study, conducted between September 2014 and November 2015, included all x-ray–guided procedures requiring TEE guidance in a single hybrid operating theatre, equipped with a contemporary cardiovascular x-ray system (Siemens Artis Zee ceiling, Siemens Healthcare, Erlangen, Germany).
Four key roles (primary catheter operator [OP1], secondary catheter operator [OP2], anesthetist [ANA], and the TEEOP) who were in attendance wore an instantly downloadable dosimeter (IDD) (Instadose, Mirion Technologies, San Ramon, California) for the duration of each procedure. The IDD was worn on the outside of the lead thyroid protection collar by all key staff, except for the TEEOP, who attached the IDD to the posterior aspect of their left shoulder, so that the dosimeter faced the radiation source.
The IDD is a direct ion storage dosimeter, the basic principles of which are well described in the published reports (8–10). The design of the dosimeter is based upon the storage of charge in a nonvolatile analogue (MOSFET) memory cell, surrounded by a small volume of air, essentially acting as an ion chamber. When scattered radiation emanating from the patient is incident on the detector, ionization occurs within the ion chamber, altering the stored charge. This change in charge provides a measure of the air Kerma (mGy) incident on the detector. The exposure can then be downloaded by connecting the dosimeter to a computer. The personal dose equivalent Hp(10) (mSv) is then calculated by the application of an appropriate algorithm, by the computer, which takes into consideration the typical photon energy incident on the dosimeter. The IDD has a broad dose and energy response of 0.01 mSv to 5 Sv and 5 KeV to 6 MeV, respectively (NVLAP lab code 100555-0 for ANSI N13.11-2009 categories IA, IIA, and IIC).
The personal dose equivalent Hp(d) is defined as the dose equivalent in soft tissue below a specified point on the body at an appropriate depth. For penetrating x-rays, this depth is 10 mm and therefore denoted as Hp(10). Hp(10) is normally considered to provide a conservative or close approximation to effective dose (E) (11). However, because the IDD was located outside the protective apron, Hp(10) in this case would significantly overestimate effective dose. The results in Hp(10) were therefore converted to effective dose, by dividing the value by 21, in line with the methodology outlined in the NCRP 168 document (12). This conversion allows for the dose across key staff to be compared along and with other studies.
All personnel working in the hybrid theatre wore a lead apron or apron and skirt with a minimum of 0.25 mm of lead-equivalent properties at the back and 0.5 mm at the front. A collar for thyroid protection was also worn. In addition, there was the option for staff to wear a lead-lined theatre hat and lead goggles with lead-equivalent properties of 0.5 mm. OP1 and OP2 stood on the patient’s right side for all procedures, with the exception of surgical access TAVR procedures, where OP1 stood either on the patients’ left side (transapical access TAVR) or at the head of the procedure table (transaortic TAVR procedures). OP1 and OP2 were protected by a table-mounted lower body radiation lead shield (skirt), with 1-mm lead-equivalent properties, and when standing on the patient’s right side, a ceiling-mounted “drop down” lead acrylic shield with 0.5-mm lead-equivalent properties. The ANA was positioned at the head of the patient and was provided with a lead acrylic shield on wheels with 1-mm lead-equivalent properties. The TEEOP was positioned at the head of the patient, toward the patient’s left side and stood obliquely, predominantly with their back to the x-ray source. The TEEOP had access to use the acrylic shield on wheels but had no other specific additional protection. A plan view of approximate staff and shielding positions is demonstrated in Figure 1.
Procedural radiation measures collected were:
• Procedure type.
• Fluoroscopy time.
• Total radiation dose (Kerma area product [KAP]).
• Patient body mass index (BMI).
• Patient sex.
• C-arm projection most used during each procedure.
To aid analysis, procedures were grouped into the following categories:
• Transfemoral TAVR or intervention.
• TAVR or intervention using open surgical access, either transaortic or transapical (TAVIS).
• Mitral valve intervention.
• Left atrial device implantation (LADI).
• Ventricular or atrial septum intervention/implants.
Additional protection measures for TEEOP
After reviewing data from the initial procedures, steps were taken to reduce TEEOP dose, and an additional ceiling-suspended lead acrylic shield with 0.5-mm lead-equivalent properties (Mavig, Munich, Germany), identical to that available to OP1, was mounted to the ceiling of the hybrid theatre, such that the TEEOP could use it for additional protection (Figure 2). With this solution in place, dosimeters were worn for a further 50 procedures, and the results from the IDD, worn by the 4 key roles, were compared for procedures before and after the additional protection was installed. Facility human research ethics committee approval was granted for this study.
In line with other studies investigating radiation dose levels (13), and international guidelines (14), the KAP was used as the primary measure for overall radiation dose and patient dose. Effective doses were compared across procedure type and TEEOP individuals. The C-arm projection most used during each procedure were divided into 5 categories, and each key operator effective dose per unit of KAP (E/KAP) was analyzed across these categories. C-arm projection categories were defined as: steep right anterior oblique (RAO) (>−20°), RAO (−6° to −20° RAO), posteroanterior (PA) (−5° to +5°), left anterior oblique (LAO) (6° to 20° LAO), and steep LAO (>20°). A Pearson correlation coefficient was used for correlation of TEEOP E against the different radiation measures. Categorical data were compared across groups with a chi-square or Fisher exact test. Continuous variables were tested for normality, and groups were compared using a Mann-Whitney U test or Student's t-test as appropriate. Statistical significance across multiple groups was performed using a Kruskal-Wallis or analysis of variance test as appropriate. Multivariate linear regression was performed to investigate independent predictors for TEEOP dose from the variables collected. A small constant was added to all TEEOP dose values, and natural logarithmic transformations were applied. Scatter plots of continuous predictors by the log-transformed outcomes were examined, and KAP was also log-transformed for use in regression modeling. Variables with p values <0.20 in univariable linear regression models were entered into a multivariable linear regression model. Variables with p values <0.05 remained in the final model. Regression coefficients were exponentiated to obtain the relative effects of predictors on the outcome variable measured in the original units. SPSS version 22 (IBM, Armonk, New York) was used for analysis.
There were 98 procedures performed before the installation of the additional ceiling-mounted lead acrylic shield and a further 50 procedures after installation. Overall, patients were 54.7% male with a median BMI of 28.70 kg/m2 (interquartile range [IQR]: 24.09 to 32.64 kg/m2). There was no significant difference in sex for procedures before and after the additional lead protection was installed (58.2% male vs. 48.0% male; p = 0.240), though there was a difference in patient BMI (29.73 kg/m2 [IQR: 24.39 to 33.96 kg/m2] before, 26.26 kg/m2 [IQR: 23.66 to 30.84 kg/m2] after; p = 0.012).
Procedures without additional TEEOP protection
Before the additional protection was present, a significant difference in radiation dose across the key roles was demonstrated (p < 0.001). Median TEEOP E was the highest of the key roles, though not statistically higher than OP1 E: 2.62 μSv (IQR: 0.95 to 4.76 μSv) versus 1.91 μSv (IQR: 0.48 to 3.81 μSv) (p = 0.101), but was significantly higher than OP2 E: 0.48 μSv (IQR: 0.00 to 1.91 μSv) (p < 0.001) and ANA E: 0.48 μSv (IQR: 0.00 to 1.43 μSv) (p < 0.001) (Central Illustration, panel A).
Many of the patient and radiation measures were significantly different across the procedural categories. Median TEEOP E was significantly different across procedure groups, with LADI procedures demonstrating the highest E at 4.76 μSv (IQR: 3.81 to 11.91 μSv) and TAVIS the lowest E at 0.95 μSv (IQR: 0.00 to 1.91 μSv) (p < 0.001). A predominantly steep RAO projection was more likely in LADI procedures (100%) and not used in other procedures, including TAVIS (0%) and transfemoral TAVR or intervention (0%) (p < 0.001) (Table 1). TEEOP E had the strongest correlation with KAP (r = 0.547; p < 0.001).
The TEEOP effective dose/KAP (E/KAP) composite value, when grouped by predominant C-arm angle demonstrated that doses were higher in procedures where more RAO and steep RAO projections were used (p = 0.041). OP1 E/KAP was also significantly different across C-arm projections, with the PA projection demonstrating the highest value (p = 0.045). E/KAP for the other team members did not differ significantly across C-arm projection categories: OP2 (p = 0.856), ANA (p = 0.366) (Figure 3).
There was a large difference in the number of procedures that individual TEEOP performed, ranging from 1 procedure for 1 operator to 39 procedures for another. However, there was no statistically significant difference in any patient-, procedural-, or radiation-related variable across the individual operators. There was no significant difference in radiation dose to the other key personnel across TEEOP categories.
Impact of additional TEEOP protection
After the additional protection was installed, median TEEOP E was reduced by 81.7%: 2.62 μSv (IQR: 0.95 to 4.76 μSv) versus 0.48 μSv (IQR: 0.00 to 1.43 μSv) (p < 0.001) (Central Illustration, panel B). Radiation dose to the other key personnel did not change significantly with the installation of the additional protection. All other variables were not significantly different, other than patient BMI and KAP. The procedure numbers across procedure categories were not significantly different after the installation of the additional shielding (p = 0.173). The TEEOP E/KAP value was significantly lower with the additional protection in place; 28.42 μSv (IQR: 13.47 to 52.64 μSv) versus 8.40 μSv (IQR: 0.00 to 18.53 μSv) (p < 0.001) (Table 2).
Univariate and multivariate linear regression demonstrated that the use of the shield was significantly associated with TEEOP radiation dose. A 76% (95% confidence interval: 59% to 86%) (p < 0.001) reduction in TEEOP dose is demonstrated in procedures where the shield was in place, compared with those without the shield after correcting for the other variables in the multivariate model. The procedure type was also demonstrated as a significant predictor for TEEOP dose, with LADI procedures demonstrating an 8.7 times greater dose than the reference group: TAVIS (p < 0.001). KAP was also demonstrated as a significant predictor (p < 0.001).
These results demonstrate that before the additional ceiling-suspended protection was in place, there was a significant radiation dose to the TEE operator for structural intervention procedures requiring TEE support (Central Illustration). The dose is at least as high as the dose to the primary operator, with the associated risks from radiation. Radiation dose to primary operators performing percutaneous cardiac procedures has been associated with cataracts (4) and higher rates of orthopedic illness and cancer (6). Other studies have also found a higher incidence in left-sided brain tumors in physicians performing interventional procedures (5). The American Society of Echocardiography issued guidelines in 2014 in an effort to highlight the radiation risks associated with percutaneous structural intervention procedures and ways to reduce that risk, including additional shielding (7). It too highlighted the paucity of data on this important topic (7). There are good data for radiation dose to operators during coronary angiography and intervention (15,16), and additional protection measures have proved to be effective in reducing radiation exposure (17). However, to date, there are no data with regard to radiation dose to TEE operators assisting with these procedures.
Drews et al. (18) investigated radiation dose to key personnel during surgical TAVR procedures with a mean KAP of 86.61 Gy ∙ cm2, similar to the 88.26 Gy ∙ cm2 median overall KAP in this study. They reported a mean E of 17.5 μSv for the ANA/TEEOP, compared with a median E of 2.62 μSv for the TEEOP in the present study. However, in that study, the ANA performed the echocardiography, which is a different model to the dedicated TEEOP seen in the present study. In addition, their methodology in calculating effective dose also differed (18).
The dose to the primary operator can also be compared. The median OP1 E of 1.9 μSv in the present study appears similar to the 2.3 μSv and 1.2 μSv of the radial and femoral access arms, respectively, of another study investigating operator radiation dose during acute myocardial infarction intervention (15). Again, however, a slightly different method to calculate E was used, as E cannot be directly measured.
At this center, the TEEOP stand with their back to the radiation source, thereby potentially lowering the effectiveness of the lead garment. As such, TEE operators should take care to ensure that there is adequate lead-equivalent protection in the rear of the lead apron, and “backless” style lead aprons should be avoided completely. TEEOP dose was not significantly different between operators, indicating that the high doses seen were not attributed to the working practices of 1 or 2 individuals but more likely related to the close proximity of the TEEOP to the radiation source. Increasing their distance from the source would reduce the dose, but this is difficult to achieve while operating the TEE probe, and the use of the lead shield on wheels is ergonomically challenging.
Procedures with increasing steepness of RAO C-arm projections are shown in this study to deliver higher doses to the TEEOP. In comparison, the lowest E/KAP to the OP1 is seen in procedures with steep RAO projections. This is in agreement with a previous study that demonstrated the lowest operator dose with RAO projections and considerably higher doses from backscatter when LAO projections were used (19). In a similar manner, highest TEEOP E/KAP is seen in the present study with increasing steepness of RAO projections, because they stand on the left side of the patient and are more exposed to the higher backscattered radiation. TEEOP should be mindful of this when performing left atrial and mitral valve procedures, where RAO projections are predominantly used. If it is possible to stand on the patient’s right side for these procedures, then this should be considered. Radiation doses to the TEEOP were significantly lower after the implementation of the ceiling-suspended shield; however, doses to OP1 did not significantly decrease and were higher than the TEEOP when both had ceiling-suspended shields present. The higher dose to OP1 is likely due to the shorter distance between the patient and OP1 during procedures and the fact that the shield may have been difficult to use for certain procedures, such as TAVIS, where the PA projection is commonly used, and in this study, higher doses are demonstrated.
Normalizing the operator’s dose to KAP is advantageous as it then accounts for confounding factors that have an impact on operator dose. KAP is the primary measure for the amount of radiation used for the procedure. It is a measure of the dose output from the x-ray system and the area exposed. KAP was lower for procedures after the installation of the additional protection and is possibly due to the lower patient BMI in that group; however, E/KAP was demonstrated to be significantly lower after the installation of the additional protection, indicating that the lower KAP was not the reason for the lower TEEOP doses.
This is a single center observational study, with all the inherent weaknesses of such. It is possible that a different work flow and/or room layout would produce different results.
This study highlights a comparatively significant scattered radiation dose to TEE operators during percutaneous structural intervention, with the associated risks involved. The radiation dose is at least as high as that to OP1, and doses are higher for procedures with predominantly RAO C-arm projections. With the additional ceiling-mounted lead protection in place, radiation dose to TEEOP was reduced dramatically. On the basis of these results, the authors advocate that similar shielding devices should be implemented in cardiac catheterization theatres and hybrid operating theatres where TEE operators are used to facilitate these procedures on a regular basis.
COMPETENCY IN SYSTEMS-BASED PRACTICE: The exposure of operators performing transesophageal echocardiography to ionizing radiation during percutaneous interventions on patients with structural heart disease is higher than that of other staff.
TRANSLATIONAL OUTLOOK: Further efforts are needed to develop optimum ceiling-mounted lead shielding and other safety measures in catheterization laboratories and hybrid operating facilities to reduce the exposure of team members providing imaging support.
The authors thank Dr. Karen Hay for her assistance with statistical analysis and to the staff that assisted with the study by wearing the IDDs.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- body mass index
- effective dose
- instantly downloadable dosimeter
- interquartile range
- Kerma area product
- left atrial device implantation
- left anterior oblique
- primary operator
- secondary operator
- right anterior oblique
- transcatheter aortic valve replacement or intervention using open surgical access
- transcatheter aortic valve replacement
- transesophageal echocardiography
- transesophageal echocardiography operator(s)
- Received August 22, 2017.
- Revision received January 8, 2018.
- Accepted January 11, 2018.
- Reddy V.Y.,
- Doshi S.K.,
- Sievert H.,
- et al.
- Feldman T.,
- Kar S.,
- Rinaldi M.,
- et al.
- Andreassi M.G.,
- Piccaluga E.,
- Guagliumi G.,
- Del Greco M.,
- Gaita F.,
- Picano E.
- Wernli C.,
- Fiechtner A.,
- Kahlilainen J.
- National Council on Radiation Protection and Measurements
- Crowhurst J.A.,
- Whitby M.,
- Thiele D.,
- et al.
- Sciahbasi A.,
- Frigoli E.,
- Sarandrea A.,
- et al.
- Kuon E.,
- Dahm J.B.,
- Empen K.,
- Robinson D.M.,
- Reuter G.,
- Wucherer M.