Radiation Reduction in the Pediatric Catheterization Laboratory Using a Novel Imaging System

Abstract: Objectives. Radiation dose was compared between two modern imaging systems with different x-ray tube technology (Megalix vs Gigalix) and detector type (amorphous vs crystalline silicon) at the same institution. Background. Further reduction in radiation dose than currently reported may be achievable with advances in x-ray tube and detector technology. Methods. Radiation dose (air kerma, dose-area product [DAP]) was retrospectively compared for post-transplant pediatric patients undergoing right heart catheterization/biopsy (fluoroscopy only) or “annual” catheterization with coronary angiography in one of two imaging systems between January 2014 and December 2016. Comparisons were also made with published radiation doses. Results. A total of 122 right heart catheterizations with biopsy were performed in the Megalix/amorphous silicon (Si) lab and 168 in the Gigalix/crystalline Si lab. Age and weight were not statistically different for the two groups. There was a 50% decrease in median air kerma (2.2 mGy vs 1.1 mGy; P<.001) and 66% decrease in median DAP (52.2 µGy•m² vs 18.0 µGy•m²; P<.001) for the Gigalix/crystalline Si lab. A total of 24 “annual” catheterizations were performed in the Megalix/amorphous Si lab and 22 were performed in the Gigalix/crystalline Si lab. There was a 57% reduction in median air kerma (458.6 mGy vs 198.6 mGy; P<.001) and a 46% reduction in median DAP (2548.0 µGy•m² vs 1367.1 µGy•m²; P<.01) for the Gigalix/crystalline Si lab. Similar reductions were found on comparison with published doses. Conclusion. The Gigalix tube and crystalline Si detector decrease radiation dose by 50%-60% for fluoroscopy and cine acquisition in pediatric patients. 

J INVASIVE CARDIOL 2018;30(1):28-33. Epub 2017 October 15.

Key words: pediatric cardiology, radiation dosage, dose-area product

Routine surveillance endomyocardial biopsy performed to assess for graft rejection is an integral component of managing patients after orthotopic heart transplantation.¹ Coronary angiography is also performed every 1-2 years to evaluate for post-transplant coronary vasculopathy.² Because x-ray is utilized in these procedures, patients are frequently exposed to radiation as part of their medical care. Children appear to be more sensitive to the stochastic effects of ionizing radiation, such as malignancy.³ This stems from the frequent cell division necessary for growth as well as the relatively longer life expectancy, allowing for a larger window of opportunity for the expression of damage from ionizing radiation.

Recognizing these long-term consequences of radiation exposure, adherence to the As Low As Reasonably Achievable (ALARA) concept by the operator, and incorporation of advances in x-ray technology to reduce radiation exposure are of paramount importance in the pediatric catheterization laboratory.⁴ Significant developments in x-ray technology include the emergence of isocentric biplane imaging systems, digital systems allowing instant replay, pulse fluoroscopy with last-image hold, optimal copper filtration, and replacement of image-intensifier systems with high-efficiency flat-panel x-ray detector (FD) systems.⁵ The recently developed Gigalix x-ray tube is more powerful than the Megalix tube, maximizing contrast, spatial, and temporal resolution at lower voltage and higher copper filtration than its predecessor. A significant further advancement has been the replacement of amorphous silicon (amorphous Si) with crystalline silicon (crystalline Si) as the photodetector in the FD systems. Crystalline Si FD systems decrease electronic noise at the detector level, thereby reducing the radiation necessary to obtain the same quality image.

In January 2014, our institution upgraded its equipment in one of two dedicated biplane pediatric catheterization laboratories to a system with the more powerful Gigalix x-ray tube and crystalline Si detector (Artis Q.zen; Siemens Healthcare). The other laboratory (Megalix tube/amorphous Si detector), in operation since October 2011, was supplied by the same vendor (Artis zee; Siemens Healthcare), and the settings in both laboratories are optimized for pediatric patients. The goal of this study is to compare radiation dose during fluoroscopy and cine acquisition in the two laboratories when used for frequent similar procedures, specifically, routine-surveillance endomyocardial biopsy and annual coronary angiography in a pediatric cardiac transplant population. 

Methods

This study was approved by the Cleveland Clinic Foundation Institutional Review Board. We retrospectively reviewed radiation exposure data of all cardiac transplant patients undergoing right heart catheterization with endomyocardial biopsy and surveillance for post-transplant coronary arteriopathy at our institution between January 1, 2014 and December 31, 2016. To minimize operator influence, only cases performed by one practitioner (GJB) were included in this study. Cases had been assigned to either of the two laboratories based on room and personnel availability. Each imaging system has the capability of reporting the radiation dose administered during use. Radiation data collected and analyzed in this study included total fluoroscopy time, cumulative air kerma, dose-area product (DAP), and the number of cine acquisitions obtained with the “A” and “B” planes. Patient age and weight were also collected.

Radiation protocols. At the time of installation and for several days thereafter, system settings on both imaging platforms were optimized with the assistance of medical physicists from Siemens Healthcare utilizing the ALARA concept as guiding principle. Instant feedback was given by the operators to optimize image quality. The final settings are summarized in Table 1. The Artis Zee system (Megalix tube) has an amorphous Si FD (A plane, 30 x 40 cm; B plane, 20 x 20 cm), while the Q.zen system (Gigalix tube) has a crystalline Si FD (A and B planes, both 26 x 30 cm). Both systems have a “biopsy” protocol setting with three fluoroscopy options with radiation dose of 23 nGy/pulse, 36 nGy/pulse, and 45 nGy/pulse for the Artis Zee system, and 10 nGy/pulse, 23 nGy/pulse, and 45 nGy/pulse for the Q.zen system (levels I, II, and III, respectively). The biopsies are universally performed at the lowest dose (level I) with a pulse rate of 4 frames/sec in both laboratories. Protocols for cine acquisition are based on weight (<6 kg, <20 kg, <40 kg, >40 kg = adult), and for each weight category there is an “extra-low” and a “low” option. The vast majority of post-transplant patients are in one of two weight categories: <40 kg or >40 kg = adult. The extra-low protocol is universally chosen for coronary angiography acquisition in post-transplant patients, and it is performed at 15 frames/sec. The system allows selection of 7.5 frames/sec, and that is typically chosen for left ventricular angiography. Settings corresponding to these two weight categories for each of the two laboratories are reported in Table 1.

Best practices to minimize radiation exposure in accordance with the ALARA concept were followed, including minimizing the source to detector distance, collimating to limit exposure only to the thoracic areas needed to be visualized, and using magnification only as needed. Although the air-gap technique (source to image distance maximized to achieve desired magnification, anti-scatter grid removed) is used in our laboratory for patients <20 kg, the vast majority of the patients in this study were >20 kg, and therefore this technique was rarely used. 

Catheterization protocol. All procedures were performed with the patients either moderately sedated or under general anesthesia depending on the patient’s age. Right heart pressure and pulmonary artery saturation were obtained in the standard fashion. Direct measurement of cardiac output was also obtained via the thermodilution technique. A minimum of 4 (but not more than 5) endomyocardial biopsy samples were obtained from the trabecular portion of the right ventricle. For cases involving right and left heart catheterization, a left ventricular angiogram was obtained by power injection of contrast. Thereafter, multiple orthogonal views of the coronary arteries were acquired via hand injections of contrast in the standard fashion. Intravascular ultrasound was performed in one vessel, almost always the left anterior descending coronary artery, if the diameter of the artery was ≥2 mm.

Statistical methods. All data collected were analyzed in aggregate. Univariable analysis was performed to compare characteristics between the Megalix/amorphous Si lab and the Gigalix/crystalline Si lab, by two-sample t-test on age and weight, and Wilcoxon rank-sum test on outcomes and fluoroscopy time, based on distribution of continuous variables, and Pearson’s Chi-square test or Fisher’s exact test for categorical variables. 

Outcomes, air kerma, and DAP values were logarithmically transformed before analysis to satisfy normal distribution. Linear regression with adjustment of age, weight, and fluoroscopy time was used to assess the difference in radiation between the two labs. Considering the potential correlation within repeated measurements in the same patient, linear mixed model with compound symmetry covariance structure was also performed to compare radiation between labs with adjustment for age, weight, and fluoroscopy time as fixed effects. All P-values are two-sided, with .05 as the level of statistical significance. Statistical analysis was performed using SAS software v. 9.4 (SAS Institute, Inc.). 

Results

During the study period, a total of 336 right heart catheterizations with endomyocardial biopsy and annual (or every 2 years) coronary angiography procedures were performed, of which 146 were in the Megalix/amorphous Si lab and 190 were in the Gigalix/crystalline Si lab (Table 2). There were 290 right heart catheterization with endomyocardial biopsy procedures performed on 68 patients and 46 coronary angiography cases performed on 37 patients. Tables 3 and 4 summarize the demographics and descriptive statistics of the patient population at the time of each procedure. There was no statistical difference in age or weight for the patients who underwent either type of catheterization in either of the two laboratories. There was also no statistical difference in fluoroscopy time for “annual” cases with coronary angiography between the two laboratories. However, fluoroscopy time for right heart catheterization with endomyocardial biopsy procedures was longer in the Megalix/amorphous Si lab (median, 5.3 min vs 3.9 min; P<.001) (Table 3).

Fluoroscopy-only procedures. Despite similar patient ages and weights, there was a very significant reduction in radiation exposure for fluoroscopy-only cases (right heart catheterization with endomyocardial biopsy) performed in the Gigalix/crystalline Si lab (Table 3). The median air kermas for the Megalix/amorphous Si lab and the Gigalix/crystalline Si lab were 2.2 mGy and 1.1 mGy, respectively (P<.001), corresponding to a 50% decrease in radiation exposure. Similarly, there was a 66% reduction in the DAP between the two lab (52.2 µGy•m² for the Megalix/amorphous Si lab and 18.0 µGy•m² for the Gigalix/crystalline Si lab; P<.001). The fluoroscopy time for the Megalix/amorphous Si lab was longer than for the Gigalix/crystalline Si lab (5.3 min vs 3.9 min, respectively; P<.01). This is likely explained by the higher percentage of trainees participating in cases in the Megalix/amorphous Si lab (80% vs 60%; P<.001). Using patient age, weight, and fluoroscopy time as covariates, a linear model analysis reiterated statistically significant radiation dose reduction between the two groups (Table 5). 

Fluoroscopy and cine-acquisition procedures. For cases consisting of right and left heart catheterization with endomyocardial biopsy and coronary angiography, a procedure requiring cine acquisition, there continued to be a significant reduction in radiation exposure for the Gigalix/crystalline Si lab (Table 4). The median number of angiograms per case was similar (19 in the Megalix/amorphous Si lab and 18 in the Gigalix/crystalline Si lab). Median air kerma was 458.6 mGy for the Megalix/amorphous Si lab and 198.6 mGy for the Gigalix/crystalline Si lab (P<.001), representing a 57% reduction for the Gigalix/crystalline Si system. A similar 46% reduction in the median DAP was observed (2548.0 µGy•m² for the Megalix/amorphous Si lab vs 1367.1 µGy•m² for the Gigalix/crystalline Si lab; P<.01). Using patient weight, fluoroscopy time, and number of angiograms as covariates, linear modeling confirmed statistically significant radiation dose reduction between the two groups (Table 6). 

Comparison to published radiation dose. A comparison of our results to published radiation dose for the same procedure, reported as “transplant with coronary angiography” by Glatz et al,⁶ using the same imaging system as our Megalix/amorphous Si lab (Artis Zee) is shown in Table 7. This publication reports radiation dose stratified by weight range and procedure type. The two weight ranges shown in Table 7 best approximate the weight of our patient groups. Noting the mean weight of our patients to be greater than their lower weight range, our radiation dose compares favorably to the published doses even for the same imaging system despite the slightly longer fluoroscopy time for our cohort. This is likely explained by the difference in radiation settings between the two institutions, with their fluoroscopy pulse rate typically at 10-15 pulses/sec delivering 23–29 nGy/pulse, and digital acquisition at 30 frames/sec for intracardiac structures (assuming coronary angiography was performed at 30 frames/sec). The slightly longer fluoroscopy time may be related to intravascular ultrasound performed at our institution, which is not routinely done in most other programs. A very significant reduction of approximately 50% for both air kerma and DAP is observed when comparing their radiation doses to our Gigalix/crystalline Si lab, despite the mean weight of our patients being at the upper end of their lower weight range and a slightly longer fluoroscopy time.

Published radiation doses for right heart catheterization with biopsy from two other centers are shown in Table 8. In comparison to the study by Verghesi et al,⁷ there is a huge 70-150 fold decrease in air kerma and 30-80 fold decrease in DAP between this report and our results, which is more pronounced for the Gigalix/crystalline Si lab but equally significant for the Megalix/amorphous Si lab. Although the fluoroscopy time is longer in their study, it is only so by a few minutes and is clearly not the explanation. Notably, the study period for that report goes back as many as 10 years before our study. Many technical advances and operator protocols aimed at decreasing radiation exposure over the last decade explain this decrease, not the least of which are the features associated with the more powerful Gigalix tube and crystalline Si detector. A more recent study by Sutton et al⁸ specifically sought to optimize ALARA techniques for this group of patients with the same system as our Megalix/amorphous Si lab. They used a pulse rate of 3 frames/sec for most cases and 10-18 nGy/frame depending on weight. The mean weight of their patients was 20.4 ± 16.6 kg vs 43.5 ± 24.5 kg for our patients in the Gigalix/crystalline Si lab, making valid comparison of radiation dose not possible. Yet despite the much lower weight of their patients, the air kerma for our patients in the Gigalix/crystalline Si lab was slightly lower, and the DAP was higher by only 2.2 µGy•m². 

Discussion

Because of the potential short-term and long-term effects of ionizing radiation, all efforts should be made to reduce radiation exposure in the catheterization laboratory, particularly in pediatric patients requiring repeat catheterization procedures. Although definite proof of an increase in cancer risk as a result of radiation exposure in the catheterization laboratory in childhood is sparse,^9,10 the theoretical concern is valid, as evidenced by the finding of chromosome damage following cardiac catheterization in children with congenital heart disease.¹¹ Furthermore, catheterization procedures are becoming increasingly complex, with longer fluoroscopy times than in prior decades. In addition, our patients are living longer and are requiring more procedures over the course of their lifetimes. Not only do patients benefit from a reduction in radiation exposure, medical personnel exposed to the long-term risks of occupational radiation exposure also stand to benefit. 

In this study, we document a >50% radiation dose reduction using a more powerful x-ray tube and crystalline Si FD compared with a slightly older system with an amorphous Si FD for cases consisting of right heart catheterization with biopsy, a procedure utilizing only fluoroscopy. Similarly, there was an approximately 50% radiation dose reduction for cases consisting of right and left heart catheterization with coronary angiography, requiring both fluoroscopy and cine acquisition. Lamers et al¹² recently reported very significant radiation dose reduction for one specific interventional procedure (patent ductus arteriosus closure) when utilizing the same system as our Gigalix/crystalline Si lab in a dedicated pediatric catheterization laboratory, as compared with an older-generation system in an adult catheterization laboratory.Furthermore, they compared their dose to recently published benchmarks for the same procedure, and reported a 50%-75% reduction in air kerma and 30%-60% reduction in DAP with the Gigalix/crystalline Si system.^6,13

Similarly, comparison of our results with published data demonstrated at least a 50% reduction in both air kerma and DAP for the same procedure (post-transplant coronary angiography) when performed in the Gigalix/crystalline Si lab.⁶ Only rough comparisons can be made with reports using different imaging systems, different procedure types, and different patient sizes. Other imaging systems have been recently developed with real-time noise reduction algorithms not based on crystalline Si. The mean DAP for 117 patients >40 kg reported for one of these systems was 3840 µGy•m². Our mean patient weight for the Gigalix/crystalline Si lab was 43.5 kg, and therefore a number of patients were <40 kg. Nonetheless, our median DAP of 1367.1 µGy•m² appears to compare favorably with this alternate technology.¹⁴ 

Our results show lower radiation dose than published data even for the Megalix/amorphous Si lab, particularly when compared with reports dating back 5-10 years. The importance of radiation dose reduction has finally taken center stage over the past few years. Both vendors and operators have made great strides toward minimizing dose while maintaining the image quality necessary for the procedure being performed. The flexibility afforded by the ability to choose between different radiation setting protocols is well borne out in this study. A post-transplant right heart catheterization with biopsy requires much less image definition than a complex intervention, and therefore can be performed with minimal radiation dose, as we and others have shown.⁸ In addition to the well-known practices of decreasing the source to image distance, appropriate collimation, using magnification only when necessary, using shallow angles whenever possible, and minimizing fluoroscopy time as much as possible, we have strived to minimize radiation exposure to both patient and operator by optimizing radiation settings with the help of medical physicists in both our laboratories. Currently, we default to level I fluoroscopy dose with a pulse rate of 4 pulses/sec for all cases in both laboratories, and it is uncommon to have to increase to levels II or III. When we do increase, it is typically for only short portions of the procedure. We have not used more than 15 frames/sec for cine acquisition for any size patient since 2012, and often use 7.5 frames/sec in larger patients with slower heart rates. Stored fluoroscopy instead of acquisition is commonly used for balloon inflations or guiding hand injections via delivery sheaths prior to device placement. Both laboratories have four weight categories (<6 kg; <20 kg; <40 kg; and adult). Radiation parameters for each of the weight categories were set in conjunction with medical physicists from the vendor while optimizing image quality for several days after installing each of the two laboratories. Further adjustments were made to the Artis zee system 1 year after installation.

Our study compared two modern imaging systems in two dedicated pediatric catheterization laboratories in the same institution with the same operator. Our data show the robust reduction in radiation exposure achieved by the more powerful Gigalix x-ray tube technology and crystalline Si detector. Although the procedures compared in our study did not involve an intervention, image quality was important in the evaluation of post-transplant coronary artery disease. The equal number of angiograms in the two laboratories supports the contention that image quality was comparable, which is also the experience of the operators. The magnitude of the reduction in radiation exposure with this new technology makes it one of the most important recent advances in keeping with the ALARA concept in the pediatric cardiac catheterization laboratory. 

Study limitations. The retrospective, non-randomized nature of the study and the small sample size analyzed for the cine-acquisition group are limitations of this study. Furthermore, while we aimed to limit operator influence by selecting cases performed by one staff operator, trainees with little catheterization experience assisted the operator during many of the cases. This may have accounted for the slightly longer fluoroscopy times documented in this study as compared to data published by others.⁸

Conclusion

This study compared the radiation dose associated with two specific procedures performed in the same institution in one of two imaging systems, both in dedicated pediatric catheterization laboratories. The more powerful Gigalix tube maximizes image quality while minimizing dose. The crystalline Si detector technology is a recent development that decreases electronic noise at the detector level, thereby reducing the amount of radiation needed to obtain the same image quality. There was a >50% reduction in both air kerma and DAP for procedures consisting of right heart catheterization with biopsy (fluoroscopy only) and an approximately 50% reduction in both air kerma and DAP for procedures consisting of right and left heart catheterization with coronary angiography, a procedure requiring cine acquisition. Comparison to published data also demonstrated significant radiation dose reduction for the Gigalix tube and crystalline Si imaging system. This technology should become the new gold standard in the pediatric catheterization laboratory. 

Acknowledgments. We would like to thank Martin von Roden for his input regarding the technical differences between the two x-ray tubes studied. 

References

1.    Chin C, Akhtar MJ, Rosenthal DN, Bernstein D. Safety and utility of the routine surveillance biopsy in pediatric patients 2 years after heart transplantation. J Pediatr. 2000;136:238-242.

2.    Pahl E, Naftel DC, Kuhn MA, et al. The impact and outcome of transplant coronary artery disease in a pediatric population: a 9-year multi-institutional study. J Heart Lung Transplant. 2005;24:645-651. 

3.    Kleinerman RA. Cancer risks following diagnostic and therapeutic radiation exposure in children. Pediatr Radiol. 2006;36(Suppl 2):121-125.

4.    Justino H. The ALARA concept in pediatric cardiac catheterization: techniques and tactics for managing radiation dose. Pediatr Radiol. 2006;36(Suppl 2):146-153.

5.    Holmes DR Jr, Laskey WK, Wondrow MA, Cusma JT. Flat-panel detectors in the cardiac catheterization laboratory: revolution or evolution-what are the issues? Catheter Cardiovasc Interv. 2004;63:324-330.

6.    Glatz AC, Patel A, Zhu X, et al. Patient radiation exposure in a modern, large-volume, pediatric cardiac catheterization laboratory. Pediatr Cardiol. 2014;35:870-878.

7.    Verghese GR, McElhinney DB, Strauss KJ, Bergersen L. Characterization of radiation exposure and effect of a radiation monitoring policy in a large volume pediatric cardiac catheterization lab. Catheter Cardiovasc Interv. 2012;79:294-301. 

8.    Sutton NJ, Lamour J, Gellis LA, Pass RH. Pediatric patient radiation dosage during endomyocardial biopsies and right heart catheterization using a standard “ALARA” radiation reduction protocol in the modern fluoroscopic era. Catheter Cardiovasc Interv. 2014;83:80-83.

9.    Modan B, Keinan L, Blumstein T, Sadetzki S. Cancer following cardiac catheterization in childhood. Int J Epidemiol. 2000;29:424-428.

10.    McLaughlin JR, Kreiger N, Sloan MP, et al. An historical cohort study of cardiac catheterization during childhood and the risk of cancer. Int J Epidemiol. 1993;22:584-591.

11.    Andreassi MG, Ait-Ali L, Botto N, et al. Cardiac catheterization and long-term chromosomal damage in children with congenital heart disease. Eur Heart J. 2006;27:2703-2708.

12.    Lamers LJ, Moran M, Torgeson JN, Hokanson JS. Radiation reduction capabilities of a next-generation pediatric imaging platform. Pediatr Cardiol. 2016;37:24-29.

13.    Ghelani SJ, Glatz AC, David S, et al. Radiation dose benchmarks during cardiac catheterization for congenital heart disease in the United States. JACC Cardiovasc Interv. 2014;7:1060-1069.

14.    Haas NA, Happel CM, Mauti M, et al. Substantial radiation reduction in pediatric and adult congenital heart disease interventions with a novel x-ray imaging technology. Int J Cardiol Heart Vasc. 2015;6:101-109.

From the ¹Department of Pediatric Cardiology, Cleveland Clinic Children’s Hospital, Cleveland, Ohio; and the ²Department of Quantitative Health Sciences, Cleveland Clinic Foundation, Cleveland, Ohio.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted April 27, 2017, provisional acceptance given May 5, 2017, final version accepted May 22, 2017.

Address for correspondence: Lourdes R. Prieto, MD, Department of Pediatric Cardiology, M 41 9500 Euclid Avenue, Cleveland, OH 44195. Email: prietol@ccf.org