Effect of gastrointestinal gas on the temperature distribution in pancreatic cancer hyperthermia treatment planning

Figures & data

Table 1. Relevant gastrointestinal gas (GG) volume (i.e. GG located within the radiotherapy 20% isodose surface) for CT and CBCTs [40].

Patient	Relevant GG volume [cm^3]
		CBCT
	CT	Mean	Range
P1	154	51***	6–171
P2	73	56	13–143
P3	119	61**	5–132
P4	111	53***	20–99
P5	69	63	18–155
P6	60	65	15–150
P7	394	212*	77–669
P8	273	161**	21–503
P9	101	55***	21–97
Average	150	86

Significance of difference in relevant GG volume between CBCTs and CT (two-sided Wilcoxon rank-sum test): *p < 0.05; **p < 0.01; ***p < 0.001.

Bold values show the patients used in this study.

Figure 1. Schematic of data analysis: (a) CT, (b) segmented CT (sCT), (c) sCT with gas replaced by muscle (sCT₀), (d) cone-beam CT (CBCT), (e) segmented gas from CBCT (GG-CBCT), (f) sCT with GG-CBCT inserted (sCT_CBCT), (g) calculated temperature distributions, (h) cumulative volume-temperature histograms. Structures sCT, sCT₀ and sCT_CBCT were used for temperature calculations. Note: for calculations, segmented external air was replaced by water bolus.

Table 2. Values of the dielectric and thermal properties for gastrointestinal gas and different tissue types at 70 MHz, used in the simulations.

Tissue	σ [S/m]	εr [–]	ρ [kg/m^3]	c [J/kg/°C]	k [W/m/°C]	Wb [kg/m^3/s]
Bone	0.05	10	1595	1420	0.65	0.12
Fat	0.06	10	888	2387	0.217	1.1
Gastrointestinal gas	0.0	1.0	1.29	10000	0.024	0
Lung	0.2	20	750	3000	0.25	3.6
Muscle	0.75	75	1050	3639	0.56	3.6
CTV (tumor)	0.74	65	1050	3639	0.56	1.8
CTV (fat)	0.06	10	888	2387	0.217	1.1

Note: To allow for larger time steps in the thermal computations, the value of c for gastrointestinal gas was taken ten times too high; this has a negligible effect on the steady-state temperature [37].

σ: conductivity; ε_r: relative permittivity; ρ: density; c: specific heat capacity; k: thermal conductivity; W_b: perfusion.

Figure 2. Example of imaging data and calculation results for patient P7, with (a) CT scan, (b) CBCT of 5^th fraction, (c) segmented CT (sCT), (d) sCT_CBCT, i.e. sCT with GG-CT replaced by muscle and GG-CBCT inserted, (e) temperature distribution for sCT, (f) temperature distribution for sCT_CBCT. GG volume was 699 cm³ for sCT (c) and 260 cm³ for the sCT_CBCT (d). Note: for calculations, segmented external air was replaced by water bolus.

Table 3. Results of this study.

Patient	GG volume [cm^3]			T50[°C]		ΔT50[°C]
	sCT	sCTCBCT mean	sCT + sCTCBCT range	sCT	sCT + sCTCBCT	sCT	sCT + sCTCBCT range		Max–min
P7	699	389	138–991	40.5	40.0–40.6	3.5	3.0–3.6		0.6
P8	407	243	30–661	41.1	40.6–41.2	4.1	3.6–4.2		0.6
P1	265	81	9–265	40.2	40.0–40.4	3.2	3.0–3.4		0.4
P6	107	99	32–190	41.6	41.3–41.8	4.6	4.3–4.8		0.5

For the four patients, gastrointestinal gas (GG) volume on sCT, sCT_CBCT (mean over the 8 fractions) and sCT + sCT_CBCT combined (range). Also, T₅₀ (temperature in 50% of the CTV volume) and ΔT₅₀ (increase in T₅₀ relative to 37 °C) are given, for sCT as well as for sCT + sCT_CBCT combined (range). In addition, the difference between the maximum and minimum ΔT₅₀ is given.

Figure 3. (Left column) For four patients, cumulative temperature-volume histograms of clinical target volume (CTV), for sCT, sCT₀ and sCT_CBCT data for eight fractions. (Right column) Relation between ΔT₅₀ and volume of gastrointestinal gas (GG) for sCT (dark gray double circle), sCT₀ (light gray double circle) and sCT_CBCT (colored dots); R² is the coefficient of determination for the linear fit (dashed line) to the ten data points.

Figure 4. sCT_CBCT with colorwash of corresponding temperature distribution, for patient P1, 11th fraction (a; GG volume = 158 cm³) and 13th fraction (b; 58 cm³), and patient P7, 1st fraction (c; 991 cm³) and 9th fraction (d; 275 cm³). Regions of high temperature in normal tissue were generally located near GG. For patient P1, who had more fat tissue, high temperature regions were also found at the anterior fat-muscle interface (a and b). Note: these 2D images show a limited view of the 3D temperature distribution; colorwash only shown for high-temperature regions (ΔT ≈≥ 3.5 °C).

Figure 5. Volume of normal tissue reaching T_ref +1 °C and T_ref +2 °C, for patients (a) P7, (b) P8, (c) P1 and (d) P6, with T_ref = ΔT₅₀ + 37 °C for the CTV for sCT of that patient. Data from temperature distributions for sCT (black), sCT₀ (dark gray) and sCT_CBCT (colors corresponding to colors in Figure 3). Linear fit and corresponding R² are given for each.

Figure 6. Maximum temperature in normal tissue, T_max, as well as the maximum temperature in normal tissue when ignoring 1 ml (64 voxels) of normal tissue with highest temperature, T_max–1 mL; both relative to T_ref, with T_ref = ΔT₅₀ + 37 °C for the CTV for sCT of that patient. Data from temperature distributions for sCT (black), sCT₀ (dark gray) and sCT_CBCT (colors corresponding to colors in Figure 3). Linear fit and corresponding R² are given for each.

van der Horst A, Houweling AC, van Tienhoven G, et al. Dosimetric effects of anatomical changes during fractionated photon radiation therapy in pancreatic cancer patients. J Appl Clin Med Phys. 2017;18(6):142–151.

 PubMed Web of Science ®Google Scholar

Kok HP, Korshuize-van Straten L, Bakker A, et al. Online adaptive hyperthermia treatment planning during locoregional heating to suppress treatment-limiting hot spots. Int J Radiat Oncol Biol Phys. 2017;99(4):1039–1047.

 PubMed Web of Science ®Google Scholar