Analysis of magnetic resonance signal intensity changes in the sacrococcygeal region of patients with uterine fibroids treated with high intensity focused ultrasound ablation

Figures & data

Figure 1. Sagittal view of MR images obtained from a 41-year-old patient with uterine fibroids before and 1 day after HIFU treatment. (A) Pre-HIFU T2WI image showed normal signal intensity in lumbar vertebrae (arrow); (B) Post-HIFU T2WI image showed a sheet-like hypointense area in the vertebrae of lumbar 5 (arrow); (C) Pre-HIFU contrast-enhanced image showed normal enhancement in the vertebrae of lumbar 5 (arrow); (D) Post-HIFU contrast-enhanced image showed a partial perfusion area corresponds to the hypointense area in the vertebrae of lumbar 5 on post-HIFU T2WI (arrow).

Figure 2. Transverse view of T2WI and contrast enhanced MR images obtained from a 47-year-old patient with uterine fibroids before and 1 day after HIFU. (A) Pre-HIFU T2WI image showed no abnormal signal intensity in the soft tissue adjacent to the fibroid (arrow); (B) Post-HIFU T2WI image showed a hyperintense area in the right piriformis (arrow); (C) Pre-HIFU contrast-enhanced image showed normal enhancement in the in the right piriformis (arrow); (D) Post-HIFU contrast-enhanced image showed a decreased perfusion in the area corresponds to the hyperintense area in the right piriformis on post-HIFU T2WI compared with pre-HIFU contrast-enhanced image (arrow).

Figure 3. Transverse section through 2nd sacral vertebrae of MR images obtained from a 43-year-old patient with uterine fibroids before and 1 day after HIFU. (A) Pre-HIFU T2WI showed normal signal intensity in the sacrum and the soft tissue adjacent to the sacrum; (B) Post-HIFU T2WI showed a sheet-like hyperintense area in the sacrum and the soft tissue anterior to the sacrum (arrows); (C) Pre-HIFU contrast-enhanced image showed normal perfusion in the sacrum and the soft tissue adjacent to the sacrum; (D) Post-HIFU contrast-enhanced image showed no perfusion in the area corresponds to the signal changed area in the sacrum and the soft tissue adjacent to the sacrum on post-HIFU T2WI (arrows).

Table 1. Baseline characteristics of patients with or without MR signal changes in the sacrum and/or soft tissue adjacent to the sacrum.

Variable	Without signal changes (N = 145)	With signal changes (N = 122)	p
Age (years)	41 (36–44)	41 (35–44)	.589
BMI (kg/cm^2)	22.2 (20.5–23.8)	22.0 (20.3–23.1)	.315
Uterine volume (cm^3)	219.9 (159.1–387.1)	259.3 (179.0–371.2)	.183
Fibroid volume (cm^3)	79.1 (41.2–14.9)	93.2 (53.0–163.2)	.132
Maximum diameter of fibroids (mm)	57 (48–73)	62 (52–74)	.125
Fibroid ventral side to abdominal wall (mm)	40 (28–51)	45 (33–66)	.007
Fibroid dorsal side to sacrum(mm)	30 (19–46)	16 (11–23)	.000
Position of uterus			.000
Anteverted	123 (61%)	78 (39%)
Retroverted	2 (33%)	44 (67%)
Type of fibroids			.003
Submucosa	17 (59%)	12 (41%)
Subserosal	47 (71%)	19 (29%)
Intramural	81 (47%)	91 (53%)
Location of fibroids			.001
Adjacent	91 (65%)	49 (35%)
Posterior wall	21 (34%)	41 (66%)
Lateral wall	23 (49%)	24 (51%)
Fundus	10 (56%)	8 (44%)
Signal intensity on T2WI^#			.102
Hypo	44 (63%)	26 (37%)
Iso	71 (54%)	56 (44%)
Hetero-hyper	22 (46%)	26 (54%)
Slighty-hyper	6 (32%)	13 (68%)
Marked-hyper	2 (67%)	1 (33%)
Enhancement type on T1WI			.786
Slight enhancement	45 (56%)	36 (44%)
Moderate enhancement	35 (51%)	34 (49%)
Significant enhancement	65 (56%)	52 (44%)

# Hypo: hypointense; Iso: isointense; Hetero-hyper: heterogeneoushyperintense; Marked-hyper: markedly homogeneous hyperintense; Slight-hyper: slightly homogeneous hyperintense.

Table 2. Comparison of HIFU treatment parameters between patients with and without MR signal changes in the sacrum and soft tissue adjacent to the sacrum.

Variable	Without signal changes (N = 145)*	With signal changes (N = 122)*	p
Treatment intensity(s/h)	524.2 ± 145.6	598.9 ± 180.3	.000
Therapeutic dose (J)	215,000 (132,150–354,000)	341,000 (219,863–561,100)	.000
Treatment efficiency(mm^3/s)	121.3 (58.0–243.6)	74.7 (34.1–157.2)	.003
EEF*(J/mm^3)^#	2.3 (1.1–4.8)	3.7 (1.7–7.9)	.003
NPV ratio*	80% (70–100%)	80% (60–100%)	.052

^#EEF: energy efficiency factor.

*NPV ratio: non-perfused volume ratio.

Table 3. Binary logistic regression analysis to evaluate related factors to the MR signal intensity changes in the sacrum and the soft tissue adjacent to the sacrum.

Predictor variables	B	S.E.	p	Ors [Exp(B)]	95% CI
					Lower	Upper
Position of the uterus
Anteverted	0^a		.053	1.000
Retroverted	0.985	0.418	.018	2.678	1.181	6.073
Type of fibroids
Submucosa	0^a		.121	1.000
Subserosal	−0.834	0.576	.148	0.434	0.140	1.345
Intramural	−0.047	0.498	.925	0.954	0.360	2.530
Location of the fibroids
Adjacent	0^a		.566	1.000
Posterior wall	0.320	0.458	.485	1.377	0.561	3.381
Lateral wall	0.238	0.437	.586	1.269	0.539	2.985
Fundus	−0.628	0.660	.341	0.533	0.146	1.945
Fibroid ventral side to abdominal wall (mm)	−0.004	0.010	.669	0.996	0.976	1.015
Fibroid dorsal side to the sacrum (mm)	−0.072	0.015	.000	0.931	0.905	0.958
Treatment intensity (s/h)	0.002	0.001	.142	1.002	0.999	1.004
Therapeutic dosage (J)	0.000	0.000	.515	1.000	1.000	1.000
Treatment efficiency(mm^3/s)	−0.003	0.001	.008	0.997	0.994	0.999
EEF^#(J/mm^3)	0.025	0.018	.168	1.026	0.989	1.063
Constant	1.023	1.086	.347	2.780

^#EEF: energy efficiency factor.

^a indicates the contrast in binary logistic regression analysis.

Table 4. Comparison of the fibroids and baseline characteristics of patients with MR signal intensity changes in the sacrum and/or the soft tissue adjacent to the sacrum after HIFU.

Variable	Signal intensity changes in sacrum (N = 21)	Signal intensity changes in sacrum and soft tissue adjacent to the sacrum (N = 66)	Signal intensity changes in the soft tissue adjacent to the sacrum (N = 35)	p
Age	42 (37–45)	41 (35–44)	40 (33–45)	.290
BMI	22.6 (20.9–24.8)	21.9 (20.3–23.0)	21.6 (20.1–22.9)	.166
Uterine volume (cm^3)	430 (227–720)	257 (182–342)	219 (154–342)	.015
Fibroid volume (cm^3)	131 (51–322)	89 (56–153)	74 (43–142)	.146
Maximum diameter of fibroids (mm)	74 ± 29	63 ± 17	61 ± 57	.188
Fibroid ventral side to abdominal wall (mm)	37 (27–62)	45 (32–66)	52 (37–70)	.203
Fibroid dorsal side to the sacrum(mm)	14 (12–20)	16 (10–21)	19 (13–29)	.156
Position of uterus				.216
Anteverted	15 (71%)	41 (63%)	22 (63%)
Retroverted	6 (29%)	25 (38%)	13 (37%)
Type of fibroids				.729
Submucosa	3 (14%)	7 (11%)	2 (6%)
Subserosal	2 (10%)	12 (18%)	5 (14%)
Intramural	16 (76%)	47 (71%)	28 (80%)
Location of fibroids				.793
Anterior	9 (43%)	26 (39%)	14 (40%)
Posterior wall	6 (29%)	24 (36%)	11 (31%)
Lateral wall	5 (24%)	10 (15%)	9 (26%)
Fundus	1 (5%)	6 (9%)	1 (3%)
Signal intensity on T2WI				.270
Hypo	4 (19%)	12 (18%)	10 (29%)
Iso	12 (57%)	31 (47%)	13 (37%)
Hetero-hyper	2 (10%)	14 (21%)	10 (29%)
Slight-hyper	3 (14%)	9 (14%)	1 (3%)
Marked-hyper	0 (0)	0 (0)	1 (3%)
Enhancement type on T1WI				.017
Slight	11 (52%)	12 (18%)	13 (37%)
Moderate	6 (29%)	21 (32%)	7 (20%)
Significant	4 (19%)	33 (50%)	15 (43%)

Table 5. Comparison of HIFU treatment results of patients with MR signal changes in the sacrum and/or the soft tissue adjacent to the sacrum.

Variable	Signal intensity changes in sacrum (N = 21)	Signal intensity changes in sacrum and soft tissue adjacent to the sacrum (N = 66)	Signal intensity changes in the soft tissue adjacent to the sacrum (N = 35)	p
Treatment intensity(s/h)	600 (560–745)	650 (491–750)	533 (41–-636)	.041
Therapeutic dose (J)	340,000 (184,000–672,775)	379,375 (238,000–531,100)	304,400 (167,268–414,725)	.321
Treatment efficiency(mm^3/s)	125.8 (54.0–320.2)	72.0 (34.0–142.4)	70.3 (31.7–162.5)	.081
EEF (J/mm^3)	2.2 (0.9–5.2)	3.9 (2.0–7.9)	4.0 (1.6–8.8)	.087
NPV ratio	90% (70–100%)	80 (60–90%)	80 (60–100%)	.215

Table 6. Multivariate logistic regression analysis results of the factors affecting MR signal intensity changes in the sacrum and/or the soft tissue adjacent to the sacrum.

Signal intensity change region^a	Predictor variables	B	S.E.	p	ORs[Exp(B)]	95% CI
						Lower	Upper
Sacrum and soft tissue adjacent to the sacrum	Uterine volume^# (cm^3)
	>719.9	−2.064	0.919	.025	0.127	0.021	0.768
	430.0-719.9	0.313	1.062	.768	1.368	0.171	10.975
	227.3–430.0	0.050	0.907	.956	1.051	0.178	6.214
	<227.3	0^b			1.000
	TI*(s/h)
	>729.2	−0.174	0.940	.853	0.840	0.133	5.307
	600–729.2	0.170	0.933	.856	1.185	0.190	7.381
	480–600	−1.339	0.919	.145	0.262	0.043	1.588
	<480	0^b			1.000
	Enhancement type on T1WI
	Slight	−1.993	0.736	.007	0.136	0.032	0.576
	Moderate	−0.426	0.766	.577	0.653	0.146	2.927
	Significant	0^b			1.000
	Constant	2.939	0.952	.002
Soft tissue adjacent to the sacrum	Uterine volume^# (cm^3)
	>719.9	−1.771	0.945	.061	0.170	0.027	1.084
	430.0–719.9	0.294	1.088	.787	1.342	1.159	11.313
	227.3–430.0	−0.432	0.933	.643	0.649	0.104	4.038
	<227.3	0^b			1.000
	TI*(s/h)	−0.002	0.002	.404	0.998	0.995	1.002
	>729.2	−1.382	1.033	.181	0.251	0.033	1.903
	600–729.2	−0.279	0.960	.277	0.370	0.062	2.221
	480–600	−0.994	0.914	.277	0.370	0.062	2.221
	<480	0^b			1.000
	Enhancement type on T1WI
	Slight	−1.102	0.753	.143	0.332	0.076	1.454
	Moderate	−0.709	0.848	.403	0.492	0.093	2.593
	Significant	0^b			1.000
	Constant	2.505	0.962	.009

*TI: treatment intensity.

^adependent variable reference category: abnormal sacrum signal group.

^bindicates the degree of reinforcement reference category.

^#Uterus volume and treatment intensity are assigned by the interquartile range: 25, 50, 75% as the cut-off value.

Table 7. Comparison of the signal intensity change volume between patients with sacral pain and those without sacral pain.

Variable	Patients without sacral pain (N = 97)	Patients with sacral pain (N = 25)	p
Signal intensity change volume in the sacrum on T2WI (cm^3)	4.3 (2.4–6.6)	4.6 (2.5–11)	.476
Signal intensity change volume in soft tissue on T2WI (cm^3)	14.6 (9.6–24.6)	30.3 (15.1–45.5)	.003
Non-perfused volume (cm^3)	2.7 (1.7–5.2)	3.0 (1.8–8.6)	.447

Stewart EA, Laughlin-Tommaso SK, Catherino WH, et al. Uterine fibroids. Nat Rev Dis Primers. 2016;2(1):16043.

 PubMed Web of Science ®Google Scholar

Zhang L, Zhang W, Orsi F, North American Hyperthermia Group. et al. Ultrasound-guided high intensity focused ultrasound for the treatment of gynaecological diseases: a review of safety and efficacy. Int J hyperth. 2015;31(3):280–284.

 PubMed Web of Science ®Google Scholar