Pushing the limits: establishing an ultra-low framerate and antiscatter grid-less radiation protocol for left atrial ablations

Abstract

Aims

Despite the use of 3D mapping systems and new developments of non-fluoroscopic options, most centres still rely at least in part on fluoroscopy for catheter visualization during catheter ablations. The purpose of this study was to assess the feasibility of using an ultra-low frame rate and antiscatter grid-less radiation protocol during complex left atrial ablations to minimize radiation exposure for the patient and staff.

Methods and results

A total of 150 consecutive patients undergoing left atrial ablations in our hospital were included in the analysis. The procedures were performed between January 2015 and November 2016. Of the included patients 75 (50%) underwent ablation before and 75 (50%) after the ultra-low frame rate (reduced from 4 to 2 FPS) and antiscatter grid-less radiation protocol was established. Procedures performed after the dose reduction protocol was established showed a 64% reduction of the dose area product (630.28 ± 550.96 vs. 226.44 ± 277.44 µGym², P < 0.001), while fluoroscopy duration (14.22 ± 4.47 vs. 13.62 ± 7.11 min, P = 0.066) and procedural duration (1:48 ± 0:28 vs. 1:53 ± 0:34 min, P = 0.525) were not prolonged. Acute procedural success was achieved in all procedures. Two complications occurred before and one complication after the protocol was established. During four procedures, operators decided to re-introduce the antiscatter grid. This was due to impaired visibility in morbidly obese patients (n = 2) or technically difficult transseptal puncture (n = 2).

Conclusion

The use of an ultra low framerate and antiscatter grid-less radiation protocol effectively reduced radiation dose for complex left atrial ablation procedures and lead to very low average patient doses. Reduced image quality did not impair procedural and fluoroscopy duration or acute procedural success.

Radiation dose , Antiscatter grid , Framerate reduction

What’s new?

• Evaluation of the on top effect on radiation dose reduction during complex left atrial ablations using a combined ultra low framerate and antiscatter grid-less approach.

• Reported dose area products for these procedures are among the lowest ever published in the literature.

• Reduced image quality did not impair procedural and fluoroscopy duration or acute procedural success.

Introduction

Catheter ablations of atrial fibrillation and left atrial tachycardias are amongst the most complex electrophysiological (EP) procedures with long procedure and fluoroscopy durations. Constant improvements of non-fluoroscopic options in the EP lab have helped to substantially reduce fluoroscopy times but patient doses remain high, especially if repeated procedures are necessary. Several effective approaches for radiation dose reduction during cardiologic interventions have been proposed so far. These measurements include fluoroscopy framerate reduction,¹ asymmetric collimation,² avoidance of unnecessary cine loops, implementation of three-dimensional mapping systems and image integration.³ Additionally, ‘low dose’ programs with optimized image processing and exposure system settings allow further dose reduction.^4,5

Antiscatter grids were introduced to improve contrast by removing scattered X-rays from the X-ray beam. They are inserted at the input of the image detector on an X-ray imaging system and can usually be removed easily by pushing a single button. Re-introduction is also easy and fast. Antiscatter grids not only remove scattered but also some straight X-ray beams before they reach the image receptor. Removing the antiscatter grid will, therefore, effectively reduce patient dose, as less radiation is needed to reach the detector entrance dose.⁶ This happens at the cost of reduced image contrast. As even during the most complex ablation procedures, image quality demand is usually modest, an antiscatter grid-less radiation approach has been proved to be effective for various EP procedures.⁷

The purpose of this study was to assess the feasibility of a combined approach using an ultra-low frame rate, low-dose, and scatter grid-less radiation protocol to minimize radiation exposure during complex left atrial ablations.

Methods

Study design

This presented study is based on a single-centre analysis of 150 left atrial ablation procedures in our hospital before and after an ultra low-frame-rate and antiscatter grid-less protocol was established.

All procedures were performed with the same X-ray system, to avoid potential machine-related differences. The procedures were performed between January 2015 and November 2016. Operators were the same before and after implementation of the dose reduction protocol.

Left atrial ablations

Left atrial ablations (pulmonary vein isolation (PVI), or PVI plus left atrial tachycardia ablation or ablation of the cavo-tricuspid isthmus (CTI)) were performed under conscious sedation with midazolam and propofol after exclusion of intracardiac thrombi by transesophageal echocardiography.

A 10-pole diagnostic electrophysiology catheter (Inquiry, St Jude Medical, Saint Paul, MN, USA) was placed in the coronary sinus. Double transseptal puncture was performed in all cases. A circular 10-pole pulmonary vein mapping catheter with an electrode size of 1mm and an interelectrode spacing of 7–7–7 mm (Inquiry Optima, St Jude Medical, Saint Paul, MN, USA) was used. As ablation and mapping catheter, an open-irrigated 4-mm tip CoolFlex or Tacticath (IBI/St. St Jude Medical, Saint Paul, MN, USA) was used. Both were introduced via two long left atrial sheats (Agilis and SLO 8.5 Fr sheath, St- Jude Medical, Saint Paul, MN, USA).

After electroanatmical reconstruction of the left atrium, catheter ablation was performed with irrigated radiofrequency energy with power of 20 W at the posterior left atrial wall and 35 W at all other sites.

Measures to reduce radiation dose

A biplane Siemens Axiom Artis dBC (Siemens, Erlangen, Germany) X-ray system was used. Before the framerate reduction, images were acquired at 4 frames per second (FPS). A ‘low dose’ program, established in cooperation with the X-ray system manufacturer, was already in use. It included optimized image processing and exposure system settings to enable dose reduction. Among those optimizations, was the introduction of copper filtration (0.1–0.2-mm thickness), a reduced peak kilovoltage from (102 kV) a low detector dose (100 nGy/Pulse for cine acquisition and 15 nGy/Pulse for fluoroscopy), a short pulsewidth (4 ms) and the selection of a smaller focus.

Two simple changes were introduced to further reduce radiation dose: setting the framerate to 2 FPS and removing the antiscatter grids. Whenever image quality was insufficient, antiscatter grids were re-introduced.

For all procedures, included standard ALARA principles⁸ were applied. This included the use of collimation and minimal magnification, using anterior–posterior view whenever possible and avoiding cine loop acquisition.

To evaluate the effect on dose reduction, the reduction of the dose area product (DAP) was measured, as it is used for objective comparisons between X-ray systems⁹ and strongly correlates with skin dose.¹⁰

Statistical analysis

For comparison of categorical variables among groups, χ² test was used. Independent sample t-test (normally distributed data) or Mann–Whitney U test (skewed data) was used to compare continuous variables. All analyses were performed using SPSS software version 22 (SPSS Inc., Chicago, IL, USA). Data are presented as absolute numbers and percentages for categorical variables or mean ± standard deviation (SD) for continuous variables. A P value of < 0.05 was considered statistically significant.

Results

Study cohort and ablation procedures

Of the 150 consecutive procedures included in this analysis, 75 procedures were performed before and 75 (50%) after the antiscatter grids were routinely removed and framerate was reduced to 2 FPS. Patient characteristics are summarized in Table 1.

Table 1

Patient characteristics

Parameter	Total (n = 150)	4 F/S (n = 75)	2 F/S (n = 75)	Significance
Age, mean ± SD (years)	63.56 ± 11.00	62.31 ± 10.67	64.81 ± 11.26	P=0.103
Male gender, n (%)	90/150 (60%)	47/75 (63%)	43/75 (57%)	P=0.505
Body mass index, mean ± SD (kg/m2)	27.85 ± 5.36	27.83 ± 4.90	27.86 ± 5.81	P=0.595
LVEF ± SD	52.29 ± 6.11	52.45 ± 5.21	52.13 ± 6.94	P=0.463
Paroxysmal atrial fibrillation, n (%)	76/150 (51%)	32/75 (43%)	44/75 (59%)	P=0.097
Arterial hypertension, n (%)	106/150 (71%)	53/75 (71%)	53/75 (71%)	P=1.0
Diabetes mellitus, n (%)	21/150 (14%)	11/75 (15%)	10/75 (13%)	P=0.814
COLD, n (%)	9/150 (6%)	2/75 (3%)	7/75 (9%)	P=0.086
CAD, n (%)	33/150 (22%)	16/75 (21%)	17/75 (23%)	P=0.844
Procedure type
PVI + AT/CTI-ablation (%)	30/150 (20%)	11/75 (15%)	19/75 (25%)	P=0.102

SD, standard deviation; LVEF, left ventricular ejection fraction; COLD, chronic obstructive lung disease; CAD, coronary artery disease; PVI, pulmonary vein isolation; AT, atrial tachycardia; CTI, cavo-tricuspid isthmus.

Of the included patients, 120 (80%) underwent PVI and 30 (20%) PVI plus ablation of one or more atrial tachycardias or ablation of the CTI.

Dose reduction/procedural outcome

By removing the antiscatter grid and minimizing framerate, the average dose area product was reduced by 64% (630.28 ± 550.96 vs. 226.44 ± 277.44 µGym², P < 0.001). Fluoroscopy time (14.22 ± 4.47 vs. 13.62 ± 7.11 min, P = 0.066) and procedural duration (1:48 ± 0:28 vs. 1:53 ± 0:34 min, P = 0.525) were not prolonged after these changes were made (see Table 2). Acute procedural success was achieved in all procedures. One major complication (pericardial effusion) occurred before the protocol was established. There was one minor complication before and after the dose reduction protocol was set up.

Table 2

Dose reduction/procedural outcome

Parameter	Total (n = 150)	4 F/S (n = 75)	2 F/S (n = 75)	Significance
Procedural duration, mean ± SD (h:min)	1:50 ± 0:31	1:48 ± 0:28	1:53 ± 0:34	P=0.525
Fluoroscopy duration, mean ± SD (min)	13.92 ± 5.93	14.22 ± 4.47	13.62 ± 7.11	P=0.066
Dose area product, mean ± SD (µGym2)	428.36 ± 479.62	630.28 ± 550.96	226.44 ± 277.44	P<0.001
Dose area product per minute, mean ± SD (μGym2/min)	30.33 ± 38.28	44.55 ± 48.73	16.12 ± 12.88	P<0.001
Acute procedural success, n (%)	150 (100%)	75 (100%)	75 (100%)	P=1.0
Minor complications, n (%)	2/150 (1%)	1/75 (1%)	1/75 (1%)	P=1.0
Major complications, n (%)	1/150 (1%)	1/75 (1%)	0/75 (0%)	P=0.316

SD, standard deviation.

Re-introduction of antiscatter grids

During four procedures, the antiscatter grids were re-introduced for various reasons. For the statistical analysis, they remained the antiscatter grid-less group.

In one patient with a BMI of 40 kg/m², the antiscatter grid had to be re-introduced for the whole procedure because of insufficient wire and catheter visibility. In three patients, the antiscatter grid was removed only after the transseptal puncture (TSP). In one patient with a BMI of 37 kg/m² due to insufficient visibility of the wires and in the two other normal weight patients, it was due to technically difficult TSP.

Discussion

Wherever fluoroscopy is needed for medical procedures, patient and operator dose must be kept as low as possible to minimize the risk of potentially fatal complications caused by radiation. Complex left atrial ablations were initially associated with long fluoroscopy times and high-radiation doses with average dose area products of several thousand µGym².^11,12 With the use of 3D mapping systems and additional novel non-fluoroscopic catheter tracking systems, average X-ray dose levels of around 700–800 µGym² were achieved.^13,14 As these additional catheter tracking systems are still not widely spread, establishing low-dose setups is the most effective option for radiation dose reduction in most centres. It is always advisable to contact the technical support of the X-ray systems’ manufacturer to establish a setup that minimizes radiation dose and is tailored to the operators’ needs. With modern X-ray systems equipped with flat detector technology and programmed to very low detector entrance doses, this may lead to extremely low-radiation doses. In a recently published trial, a dose area product of only 199 ± 159.6 µGym² during complex left atrial procedures (atrial fibrillation- and atypical flutter-ablation) was reported.⁴ These settings did not lead to prolonged procedural times, fluoroscopy times, or to reduced procedural success. Such low doses should become the goal for every centre that performs these kinds of procedures.

With the current low-dose settings in our hospital, the average dose area product for complex left atrial procedures was around 600 µGym², so further measures to reach lower doses were evaluated. As it was shown that removing the antiscatter grid lead to significant dose reduction during various EP procedures,⁷ it was routinely removed before all left atrial ablations and only re-introduced if image quality was insufficient. Additionally, the framerate was reduced to the lowest level the operators felt comfortable with. This was at 2 FPS. Both changes had the advantage that can be easily implemented by the operator and became effective immediately.

By applying this approach, the dose area product could be reduced by 64%. This similarly leads to very low doses of around 200 µGym². The accompanying reduction of image quality was not at the cost of longer procedure or fluoroscopy times and did not influence acute procedural success.

In 4 of the 75 cases where the antiscatter grid had initially be removed, it was re-introduced during the procedure. Two of these procedures were performed in patients with a very high BMI (37 and 40 kg/m²). We would, therefore, not recommend to remove the grids in morbidly obese patients. The other two procedures were characterized by a technically difficult TSP and the operators chose to improve visibility. In contrast, the reduction to 2 FPS was tolerated by all operators and did not have to be upregulated in any single case.

Limitations

This study represents a single centre study and the interventions were performed by a small number of (experienced) operators. They may accept lower image qualities compared to operators in lower volume centres. Additionally, every measure to reduce radiation dose leads to increased radiation awareness of the operators. Improvements in collimation or angulation may be in part responsible for the observed effects.

Conclusion

To achieve the lowest possible radiation dose during complex left atrial ablations, different paths may lead to the same goal. In this analysis, the average dose area product was reduced by 64% by removing the antiscatter grid and reducing framerate to 2 FPS. As antiscatter grids can usually be easily removed and re-introduced, it could become a standard to start left atrial ablation procedures without the antiscatter grid and only reintroduce them when catheter visibility is insufficient.

Conflict of interest: none declared.