The effect of underlying bone on the beam angular correction in calculating the skin dose of the head in neuro-interventional imaging

Abstract

Skin dose depends on the surface shape, underlying tissue, beam energy, field size, and incident beam angle. These dependencies were determined in order to apply corrections in the skin-dose-tracking system (DTS) for accurate estimation of the risk of deterministic skin effects during fluoroscopically-guided neuro-interventional procedures. The primary-plus-scatter dose was calculated averaged over the skin thickness with underlying subcutaneous fat, and various thicknesses of skull bone on the surface of a cylindrical water phantom to simulate the head. The skin dose was calculated using EGSnrc Monte-Carlo (MC) software with 2×10¹⁰ incident photons and was normalized to the incident primary dose. Simulations were done for beam incident angles from 90 to 10 degrees with the skin surface, field sizes from 5 to 15 cm, bone thicknesses of 0, 1, 5, and 9 mm, and beam energies from 60 to 120 kVp. The results show the scatter-plus-primary to incident-primary dose ratio decreases with decreasing incident angle to the skin and with increasing thickness of underlying bone, while it increases with increasing field size and with increasing beam energy. The correction factor reduces the skin dose for angled rays and the reduction can be substantial for small angles of incidence, especially for angles below 50 degrees. For neuro-interventional procedures, the skin dose-area product (SDAP) with angular and bone correction is shown to be less than that without correction. The results of this study can be used to increase the accuracy of patient-skin-dose estimation for the head during fluoroscopic procedures.

1. INTRODUCTION

Fluoroscopically-guided interventional procedures can result in high dose to the skin on the entrance surface of the patient, which can cause threshold-dependent cutaneous injury. This high skin dose can cause deterministic cutaneous injury such as erythema, epilation and desquamation ^1–2. The skin injury can extend up to subcutaneous fat and muscle and, thus, it is necessary to calculate skin dose taking the tissue structure into consideration. The dose to the skin layer varies with the surface shape, underlying tissue, incident beam angle, and exposure parameters due to the change of primary attenuation and backscatter. Underlying bone, having a high density and high atomic number, will affect the amount of backscatter significantly. The purpose of this work is to evaluate the effect of underlying bone on the beam angular correction and determine the correction factors for clinical use. In this project, the dependence of skin dose on incident beam angle was calculated with underlying subcutaneous fat and for various thickness of bone to determine the correction factors for exposure of the head in the Dose Tracking System ^3–4. Skin dose is typically calculated with the beam at normal incidence to the flat surface of a water phantom, while correction is needed for various underlying tissues as a function of incident angles. Accurate determination of skin dose is important to predict the likelihood of deterministic effects.

2. METHODS

EGSnrc Monte-Carlo (MC) ⁵ software was used to calculate the skin dose as a function of incident beam angle to the surface of a 20 cm long cylindrical water phantom as shown in Fig. 1. This phantom is used to simulate the head and its diameter was set to 14.5 cm, which is the average width of the male head measured from the right tragion to the left ⁶. The normal scalp thickness in an adult person is about 5.8 mm ⁷ and the subcutaneous fat layer is 3.1 mm ⁸ so the dose in this project was averaged over a 2.7 mm “skin” thickness with an underlying layer of 3.1 mm subcutaneous fat and various thicknesses of bone, including 0, 1, 5, and 9 mm (ICRPBONE521ICRU). This total primary-plus-scatter skin dose was normalized to the incident primary dose, which was calculated as

$${\mathrm{P}}_0(\mathrm{E})=\int \mathrm{N}(\mathrm{E})\times \mathrm{E}\times \frac{{\mathrm{\mu }}_{\mathrm{en}}}{\mathrm{\rho }}(\mathrm{E})\mathrm{dE}$$ (1)

where N(E) is the photon fluence and $\frac{{\mathrm{\mu }}_{\mathrm{en}}}{\mathrm{\rho }}(\mathrm{E})$ is the mass energy absorption coefficient at energy E obtained from NIST tables ⁹.

Fig. 1.

Geometry of the exposure for determination of the angular-correction factor with underlying bone. The source to surface distance (SSD), 54 cm, is defined as the distance from the focal spot to the point of intersection of the central axis with the skin surface and the incident angle (θ) is the angle of the central ray with the skin surface. Field size is defined in a plane perpendicular to the central ray at its intersection with the skin.

The normalized skin dose was determined for incident beam angles from 90 to 10 degrees to the surface in the transverse direction of the phantom, for field sizes from 5 to 15 cm, and for beam energies from 60 to 120 kVp for the various thicknesses of underlying bone to determine angular-correction factors (ACF). All MC simulations used 3×10¹⁰ photons incident on the phantom. To determine the effect of including underlying bone in the skin dose calculation for a neuro-interventional procedure, the skin dose map was calculated using the parameter log files for a stent-assisted coiling procedure with and without the ACF and underlying bone. ACF’s with corresponding thickness of bone at each part of the skull were calculated by interpolation of the Monte Carlo results for different thicknesses and were applied in the DTS patient graphic phantom; the averaged thickness of skull bone used was 6 mm for frontal, 3.65 mm for temporal, 7.4 mm for occipital, and 5.7 mm for parietal bones ¹⁰. The skin dose values with angular and bone correction, with angular correction alone and without correction were compared.

3. RESULTS AND DISCUSSION

3.1. Primary plus scatter to incident primary ratio vs incident beam angle

The curves in Fig. 2 show that the primary plus scatter to incident primary ratio decreases as the incident angle decreases. This is mainly due to the longer path length of the primary x-rays through the 2.7 mm skin thickness and thus greater attenuation as the angle of incidence decreases.

Fig. 2.

Ratios of the primary plus scatter to the incident primary at the field center for a) 5 cm, b) 10 cm, and c) 15 cm field size at 80 kVp with 0, 1, 5, and 9 mm thick underlying bone, as a function of incident angle to the surface of a 20 cm long cylindrical water phantom with 14.5 cm diameter. The results show the dose ratio decreases with decreasing incident angle.

3.2. Primary plus scatter to incident primary ratio vs field size

The results of Fig. 3 show the scatter plus primary to incident primary ratio increases with increasing field size and the amount of increase is smaller for greater thickness of underlying bone due to more absorption of the transmitted primary beam and less scattering.

Fig. 3.

Ratios of the primary plus scatter to the incident primary ratio at the field center with 0, 1, 5, 9 mm underlying bone at 60 kVp at a) 30 degrees, b) 60 degrees, and c) 90 degrees incident angle as a function of incident field size to the surface of a 20 cm long cylindrical water phantom with 14.5 cm diameter.

3.3. Primary plus scatter to incident primary ratio vs thickness of underlying bone

The results shown in Fig. 4 illustrate that the primary plus scatter to incident primary ratio decreases with increasing thickness of underlying bone at 60, 80, 100, and 120 kVp. This is likely due to the increased absorption of backscatter by the bone with increased thickness.

Fig. 4.

Ratios of the primary plus scatter to the incident primary ratio at the field center at 60, 80, 100, 120 kVp at a) 30 degrees, b) 60 degrees, and c) 90 degrees incident angle as a function of underlying bone thickness for a 20 cm long cylindrical water phantom with 14.5 cm diameter. The result shows the dose ratio decreases with increasing thickness of underlying bone.

3.4. Percent dose reduction with bone vs thickness of underlying bone

The results in Fig. 5 shows that the percent dose reduction increases with increasing thickness of underlying bone at 60, 80, 100, and 120 kVp for all field sizes at 30, 60, and 90 degrees due to increased absorption of backscatter by the bone; the dose reduction decreases with increasing kVp due to more of the higher energy backscatter penetrating the bone.

Fig. 5.

The percent dose reduction* as a function of underlying bone thickness at the field center for all field sizes at 60, 80, 100, 120 kVp at a) 30 degrees, b) 60 degrees, and c) 90 degrees incident angle for a 20 cm long cylindrical water phantom with 14.5 cm diameter for a skin layer 2.7 mm thick with an underlying layer of 3.1 mm subcutaneous fat. [* Difference between the ratios of the primary plus scatter to the incident primary without bone and with bone divided by the ratio of the primary plus scatter to the incident primary without bone.]

3.5. The cumulative skin dose-area-histogram (DAH)

Figure 7 shows the cumulative dose-area-histogram (DAH) of the skin dose on the head for the selected stent coiling procedure whose DTS dose map is shown in Figure 6. The cumulative DAH is a plot of the area of skin with a dose exceeding the dose value on the abscissa. The repair of skin injury is dependent on the area of skin exposed to a dose exceeding a threshold and thus the DAH provides a good indication of the risk of serious cutaneous injury. The skin dose area products (SDAP) without any correction, with just angular correction, and with angular plus underlying bone corrections, are 0.0409 Gy·m², 0.0365 and 0.0341 Gy·m², respectively. The SDAP reductions are 1.12 with angular correction alone and 1.2 times with angular plus bone correction compared to no correction.

Fig. 7.

The cumulative DAH with angular plus bone correction and the DAH with angular correction only and the DAH without corrections.

Fig. 6.

The DTS display in playback mode for the stent-assisted coiling procedure selected for skin dose comparison in the head. Exposure outside the head occurred during catheter guidance.

4. CONCLUSIONS

Skin dose is dependent on the angle of beam incidence, decreasing with decreasing angle especially for angles less than 50 degrees with or without underlying bone due to the longer path length of the primary x-rays through the skin thickness and thus greater attenuation as the angle of incidence decreases. Due to increasing backscatter, skin dose increases with increasing field size and the amount of increase grows slower for thicker underlying bone due to more absorption of the primary x-rays and thus less scattering. The skin dose decreases and the percent dose difference reduction with bone increases as the thickness of underlying bone increases due to the absorption of backscatter by bone. The skin dose area products (SDAP) reductions and peak skin dose (PSD) reduction are 10.76% and 0.24% with angular correction alone and 16.6% and 2.67 % with angular plus bone correction compared to no correction on the head of a selected stent coiling procedure (Table 1). The results of this study shows the effect of underlying bone and can be used to increase the accuracy of the estimation of the skin dose for fluoroscopic procedures of the head using the Dose Tracking System.

Table 1.

Skin dose area products (SDAP), skin dose area products difference (SDAP Diff), peak skin dose (PSD), peak skin dose difference (PSD Diff), without any correction, with angular correction only, and with angular plus underlying bone corrections for the clinical neurointerventional procedure.

	Without correction	With angular correction only	With angular plus bone correction
SDAP (Gy · m2)	0.0409	0.0365	0.0341
% SDAP Diff	0	10.76	16.6
PSD (Gy)	0.823	0.821	0.801
% PSD Diff	0	0.24	2.67

% Diff = (without correction - with correction) / without correction *100