Non-contrast enhanced MRI for efficiency evaluation of high-intensity focused ultrasound in adenomyosis ablation

Objective: To investigate the value of T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) in evaluating the therapeutic effect of high-intensity focused ultrasound (HIFU) in adenomyosis ablation.

One hundred eighty-nine patients with adenomyosis were treated with HIFU. The ablation areas on T2WI and DWI sequences were classified into different types: type I, relatively ill- defined rim or unrecognizable; subtype IIa, well-defined rim with hyperintensity; subtype IIb, well- defined rim with hypointensity. The volume of ablation areas on T2WI (V T2WI ) and DWI (V DWI ) was measured and compared with the non-perfused volume (NPV), and linear regression was conducted to analyze their correlation with NPV.

The V T2WI of type I and type II (subtype IIa and subtype IIb) were statistically different from the corresponding NPV (p ¼ 0.004 and 0.024, respectively), while no significant difference was found between the V DWI of type I and type II with NPV (p ¼ 0.478 and 0.561, respectively). In the linear regression analysis, both V T2WI and V DWI were positively correlated with NPV, with R 2 reaching 0.96 and 0.97, respectively.

Both T2WI and DWI have the potential for efficient evaluation of HIFU treatment in adenomyosis, and DWI can be a replacement for CE-T1WI to some extent. Abbreviations: HIFU: high-intensity focused ultrasound ; NPV: non-perfused volume ; NPVR: non-per - fused volume ratio ; V T2WI : ablation volume on T2WI ; V DWI : ablation volume on DWI ; ICC: intraclass cor - relation coefficient ; MRI: magnetic resonance image ; T2WI: T2-weighted imaging ; DWI: diffusion- weighted imaging ; CE-T1WI: contrast-enhanced T1WI ; BMI: body mass index ; IT: interval time between HIFU procedure and MRI scanning post-treatment ; EEF: energy efficiency factor ARTICLE HISTORY Received 13 June 2023 Revised 11 December 2023 Accepted 12 December 2023

High-intensity focused ultrasound; magnetic resonance imaging; adenomyosis; non-perfused volume; efficiency evaluation

Adenomyosis is a common and benign uterine disease that affects women during their childbearing age. Its typical pathological findings are ectopic endometrial glands, stromal tissue in the myometrium, and hyperplasia of the peripheral smooth muscle [ 1 ]. Adenomyosis can seriously impair the quality of life. About 2/3 of patients have symptoms of dys - menorrhea or menorrhea and some patients may experience infertility [ 2 , 3 ]. In addition to medication and surgery, high- intensity focused ultrasound (HIFU) ablation is a novel

for adenomyosis treatment. By focusing ultrasonic energy on the target tissue, HIFU treatment can cause coagulative necrosis and reduce the volume of the lesion [ 4 ]. HIFU treatment contains the advantage of being highly effi - cient, noninvasive, and inexpensive [ 5 , 6 ]. To estimate the prognosis of the patients, the evaluation of treatment efficacy is necessary. MRI is one of the most important methods for efficacy evaluation post-treatment. The ablation rate, represented by the non-perfused volume ratio (NPVR), is the most widely used index for evaluating the short-term efficacy of HIFU. NPVR is defined as the ratio of the non-perfused volume (NPV) measured on post-treat - ment contrast-enhanced T1-weighted imaging (CE-T1WI) to the adenomyosis volume measured on pretreatment T2- weighted imaging (T2WI). It has been confirmed that NPVR has a significant positive correlation with the prognosis of the patients. Therefore, the NPVR has been the gold standard in the efficacy evaluation of HIFU treatment [ 7–9 ]. However, the use of gadolinium-based contrast agents in CE-T1WI may increase money consumption and time costs. Furthermore, it CONTACT Yang Liu [email protected] State Key Laboratory of Ultrasound Engineering in Medicine and Engineering, College of Biomedical Engineering, Chongqing Medical University, Yixueyuan Road, Yuzhong District, Chongqing, 400016, China; Chongqing Key Laboratory of Biomedical Engineering, Chongqing Medical University, Chongqing, China; Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China Si Ma and Fajin Lv are both first authors. � 2024 The Author(s). Published with license by Taylor & Francis Group, LLC This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. INTERNATIONAL JOURNAL OF HYPERTHERMIA 2024, VOL. 41, NO. 1, 2295813 https://doi.org/10.1080/02656736.2023.2295813 may also pose potential risks to patient health, such as neph - rogenic systemic fibrosis (NSF), contrast-induced nephrop - athy, and anaphylaxis [ 10–12 ]. An MRI method that can quickly and effectively evaluate the efficacy of HIFU in adenomyosis treatment without the use of contrast agents may significantly improve its economy and safety. T2-weighted imaging (T2WI) and diffusion-weighted imag - ing (DWI) are MR sequences with fast imaging and without contrast agent use. Several studies have reported their value in the efficiency evaluation of HIFU treatment in leiomyomas [ 13–15 ]. The signal change of micro hemorrhage on T2WI can make the boundary between ablation areas and fibroid tissue. Furthermore, Liao DF’s study found that a high-signal ring on DWI after HIFU ablation could help to measure the area of necrotic tissue and evaluate the success of the pro - cedure. However, their value in efficiency evaluation for HIFU treatment in adenomyosis remains unknown.

Patients Patients with symptomatic adenomyosis who underwent HIFU treatment (between June 2013 and September 2019) were enrolled in this study ( Figure 1 ). The inclusion criteria were as follows: (1) Adult women with symptomatic adenomyosis; (2) the patients should be completely autonomous and cooperate to complete HIFU treatment; and (3) the patients should undergo pelvic MRI scanning before and after HIFU treatment. The exclusion criteria were as follows: (1) age < 18 years, (2) inability to com - plete the treatment, (3) lack of clinical or imaging data, (4) other pelvic malignancies, and (5) acute pelvic inflamma - tion and pregnancy. HIFU treatment The patient ingested liquid food three days before treatment and a single dose of intestinal preparation solution (2000 ml of composite polyethylene glycol electrolyte solution) in the afternoon before treatment. An enema was performed on the morning of the day of treatment. The treated area was shaved, degreased, and degassed in advance. The patient was prone on the operating table, and the anterior abdom - inal wall was in contact with degassed water. A catheter was inserted to control bladder volume by injecting saline, and a degassed water balloon was used to push away the intestine in the acoustic path. The HIFU procedure was performed using a Focused Ultrasound Tumor Therapeutic System (Model-JC or Model-JC 200, Chongqing Haifu Medical Technology Co., Ltd.). The ultrasound parameters used in this study were as follows: a working frequency of 0.8 MHz, an acoustic power range of 300–400 W, and a focal area of 1.5 mm � 1.5 mm � 10 mm. The ultrasound device provided real-time monitoring during the HIFU procedure. The patient was treated under intraven - ous conscious sedation with fentanyl and midazolam hydro - chloride. The ultrasonic energy was adjusted based on both patient’s feedback and changes in gray scale on ultrasono - graphic imaging during the treatment. The physician discon - tinued the treatment until the grayscale covered most of the lesion or the patient could not endure the pain of the pro - cedure. When the treatment ended, the patient was asked to observe for 2 h before they returned to the ward. MRI evaluation All patients were scanned with 3.0 T MRI equipment (single HD excite, GE Healthcare, USA) before and after HIFU treat - ment. The eight-channel phased-array abdominal coil was Figure 1. The recruitment process of patients in this study. 2 M. SI ET AL. fixed in a supine position. Patients were asked to breathe calmly and avoid body movements during the examination. The scanning parameters are presented in Table 1 . The types of ablation areas, non-perfused volume (NPV), the ablation volume on T2WI (V T2WI ), the ablation volume on DWI (V DWI ), and the volume of adenomyosis were evaluated. The types of ablation areas were classified independently by radiologists 1 and 2 (both with eight years of experience in MRI diagnosis). In cases of disagreement in classification between the two radiologists, the final decisions were made by radiologist 3 (10 years of experience in MRI diagnosis). All three radiologists were blinded to the patient’s information and post-treatment CE-T1WI images. The ablation areas on T2WI and DWI were respectively classified according to the following criteria ( Figure 2 ): (1) type I, the ablation area had a relatively ill-defined rim which was either hyperintensity or hypointensity or unrecognizable; (2) subtype IIa, the ablation area had a well-defined rim which was hyperintensity; (3) and subtype IIb, the ablation area had a well-defined periph - eral rim which was hypointensity. When evaluating the abla - tion areas, the central slice that could manifest the largest part of the ablation area was used and pretreatment images were selected as a reference. The consistency of the classifi - cation of the ablation areas between the two radiologists was evaluated using the kappa consistency test. The ablation volume on T2WI (V T2WI ), the ablation volume on DWI (V DWI ), and NPV were measured on post-treatment T2WI, DWI (b ¼ 800 m 2 /s) and CE-T1WI. Adenomyosis volume was measured on pretreatment T2WI. The volumes were measured by delineating all the slices of the target areas using the ITK-SNAP 3.4 software (Cognitica, Philadelphia, PA, USA) and the volumes were calculated automatically ( Figure 3 ). When measuring the type I ablation volumes, the regions with suspicious signal change at the rim should all be cov - ered. For patients whose ablation areas were unrecognized, we defined the V T2WI or V DWI as 0 cm 3 . V T2WI and V DWI were compared to the corresponding NPV. Linear regression ana - lysis was performed to assess their correlations. Based on their ablation types on T2WI, the patients were divided into the T2WI-type I group and the T2WI- type II group, the T2WI-type II group was then divided into T2WI-subtype IIa group and T2WI-subtype IIb group. The patients were also classified into DWI-type I group, DWI-type II group, DWI-subtype IIa group, and DWI-subtype IIb group according to their ablation types on DWI. The clinical and treatment parameters were compared between the groups. Statistical analysis All analyses were performed using the SPSS software (SPSS 24.0 IBM Company, Armonk, NY, USA). Normally distributed data were reported as mean ± standard deviation, whereas non-normally distributed data were reported as medians and interquartile ranges. The t-test or rank sum test was used to compare differences in measurement data, and the chi- square test was used to compare differences in counting data. The interpretation of the Kappa value is based on the following criteria: a Kappa value < 0.2 indicates poor consist - ency; a Kappa value between 0.21 and 0.40 indicates fairly poor consistency; a Kappa value between 0.41 and 0.60 indi - cates moderate consistency; a Kappa value between 0.61 and 0.80 indicated fairly good consistency; and a Kappa value > 0.80 indicates good consistency. An intraclass correl - ation coefficient above 0.90 was classified as having good reliability. Statistical significance was set at p < 0.05.

General characteristic In total, 189 patients were enrolled in this study. The mean age was 41 ± 5 years, and the mean BMI was 32.1 ± 2.9. The median adenomyosis volume was 94.4 (53.5–196.2) cm 3 . The median NPV was 43.2 (23.7–78.8) cm 3 and the median NPVR was 44.1 (30.9–62.3) %. The interval time between the HIFU procedure and post-treatment MRI scanning (IT) was 1 (1–3) days and 141 patients underwent MRI examination within 2 days. The classification of ablation areas on T2WI and DWI Of the 189 adenomyosis on T2WI, 31 of them presented type I ablation areas, 119 of them presented subtype IIa ablation areas and 39 of them presented subtype IIb ablation areas. In all 189 ablation areas on DWI, 15 of them were type I ablation areas, 121 of them were subtype IIa ablation areas, and 53 of them were subtype IIb ablation areas. The kappa value of the classifi - cation for the ablation area on T2WI was 0.627 (p < 0.001) between the two radiologists, and that on DWI was 0.651 (p < 0.001). Both reached fairly good consistency ( Table 2 ). The measurement of ablation volume on T2WI and DWI The median V T2WI of type I and type II ablation area were 15.0 (5.3–33.5) cm 3 and 50.4 (23.2–114.8) cm 3 , both of which were significantly different (p ¼ 0.004 and p ¼ 0.024, Table 1. MRI parameters. TR (ms) TE (ms) NEX FOV (cm � cm) Matrix size Slice thickness (mm) Slice gap (mm) Imaging planes T2WI (FSE) 4080 105 2 38 � 47.2 512 � 512 6 2 Transverse Sagittal T1WI (SE) 600 10 1 38 � 47.2 512 � 512 6 2 Transverse DWI (SE-EPI) (b ¼ 0, 800 m 2 /s) 5700 65.7 2 38 � 38 256 � 256 5 0 Transverse CE-MRI (LAVA) 4.2 1.9 0.72 38 � 38 512 � 512 4 0 Transverse Sagittal T2WI: T2-weighted imaging; T1WI: T1-weighted imaging; DWI: diffusion-weighted imaging; CE-MRI: contrast-enhanced MRI; TR: repetition time; TE: echo time; FOV: field of view; NEX: number of excitations. INTERNATIONAL JOURNAL OF HYPERTHERMIA 3 respectively) with their corresponding NPV, which were 21.2 (8.7–42.8) cm 3 and 49.1 (27.9–94.9) cm 3 , respectively. And the V DWI of type I and type II ablation area were 9.4 (3.7– 24.6) cm 3 and 57.1 (33.5–122.3) cm 3 , both of which with no significant difference (p ¼ 0.078 and p ¼ 0.561) with their cor - responding NPV, which were 13.9 (6.3–26.6) cm 3 and 58.8 (30.1–118.4) cm 3 . Seven patients got invisible ablation areas on T2WI, their largest NPV and NPVR were 18.5 cm 3 and 17.7% and their smallest NPV and NPVR were 2.2 cm 3 and 2.0%. Five patients had completely invisible ablation areas on DWI, their largest NPV and NPVR were 6.3 cm 3 and 13.8% and their smallest NPV and NPVR were 2.2 cm 3 and 2.0%. The ICC of the measurement for V T2WI was 0.951 in the type I group and that was 0.990 in type II group. The ICC of the measurement for V DWI was 0.947 in type I group and that was 0.975 in type II group. All achieved good consistency ( Table 3 ). In the linear regression analysis, both V T2WI and V DWI showed significant correlations with the NPV. The linear equation between V T2WI and NPV was NPV ¼ 0.98 V T2WI þ 1.32 (R 2 ¼ 0.96, p < 0.001). The linear equation between V DWI and NPV was NPV ¼ 0.96 V T2WI þ 1.07 (R 2 ¼ 0.97, p < 0.001) ( Figure 4 ). Clinical and treatment parameters of the different groups The age, BMI, adenomyosis volume, adenomyosis type, ute - rus position, NPV, sonication power, treatment time, sonication time, sonication dose, energy efficiency factor (EEF, defined as the ultrasound energy delivered for ablating 1 mm 3 of the adenomyotic lesion tissue), NPVR, and IT of the different groups were compared. The NPV in the T2WI-type I group was 21.2 (8.7–42.8) cm 3 , while the NPV was 49.1 (27.9–94.9) cm 3 in the T2WI-type II group (p < 0.01). The NPVR in the T2WI-type I group was 27.2 (17.4–44.7)%, while that was 54.6 (42.0–73.9)% in the T2WI-type II group (p < 0.01). The NPV in the DWI-type I group was 13.9 (6.3– 26.6) cm 3 , and that was 58.8 (30.1–118.4) cm 3 in the DWI- type II group (p < 0.01). The same difference was found between the NPVR in the DWI-type I and DWI-type II group (32.5 (13.4–48.5)% vs. 43.0 (30.0–68.7)%, p < 0.001). The EEF in the DWI-type I group was 4.1 (2.5–6.8) kJ/cm 3 , and the EEF in the DWI-type II group was 2.0 (1.4–4.3) kJ/cm 3 (p ¼ 0.044). And the volume of adenomyosis in the DWI-type I group was smaller than that in the DWI-type II group (68.7 (33.6–93.9) cm 3 vs. 148.2 (76.3–266.1) cm 3 , p ¼ 0.001). The

are shown in Tables 4 and 5 . A remarkable statistical significance was found between the ITs in the T2WI-subtype IIa and T2WI-subtype IIb groups (1 (1–2) days vs. 5 (2–7) days, p < 0.001), and the same differ - ence was also found between the DWI-subtype IIa and IIb groups (1 (1–3) days vs. 6 (3-6) days, p < 0.001). The EEF and sonication time [2.4 (1.5–7.8) kJ/cm 3 and 597 (397–1213) s] in the T2WI-subtype IIa group were smaller than those [2.9 (1.9–4.6) kJ/cm 3 and 912 (489–1200) s] in the T2WI-subtype IIb group. The treatment time in the T2WI-subtype IIa group was 52 (35–93) min, which was smaller than the treatment Figure 2. T2WI (A1, B1, C1), DWI (A2, B2, C2), and CE-T2WI (A3, B3, C3) images post-treatment. As we can see, the subtype IIa and IIb ablation areas on T2WI or DWI had similar morphology with corresponding non-perfused area on CE-T1WI (B3 and C3). 4 M. SI ET AL. time of 88 (61–102) min in the T2WI-subtype IIb group. And the NPV and NPVR in the DWI-subtype IIb group were larger than those in the DWI-subtype IIa group [54.6 (19.3–83.2) cm 3 vs. 52.6 (21–113.5) cm 3 , p ¼ 0.026 and 67.7 (35.2–79.5)% vs 44.1 (35.8–68.5)%, p ¼ 0.013). Furthermore, the sonication power was slightly higher in the DWI-subtype IIa group (400 (373-400) W vs. 398 (371–400) W, p ¼ 0.013). The results are shown in Tables 6 and 7 .

The ultrasound energy in HIFU ablation leads to an interrup - tion of blood flow in the ablation tissue. As a result, the treated tissue appears as the non-perfused area on CE-T1WI [ 16 ]. At present, the NPVR has become the gold standard for efficient evaluation of HIFU treatment since it has a positive correlation with the patient’s prognosis [ 7–9 ]. However, the use of gadolinium-based contrast agents is time- and money-consuming and may pose potential risks. Therefore, it is more economical and safer to use the non-contrast MRI technique for efficacy evaluation. In leiomyomas, the value of T2WI and DWI in the efficacy evaluation of HIFU treatment has been revealed [ 13–15 ]. However, their potential for eval - uating HIFU treatment in adenomyosis remains unknown. According to the image features of the ablation area on T2WI or DWI, they were classified into type I (with an ill- defined rim) and type II (with a well-defined rim). The main differences in treatment parameters between the type I group and the type II group were their NPV and NPVR. Adenomyosis patients with type I ablation areas on T2WI or Figure 3. The measurement of V T2WI , V DWI, NPV and volume of adenomyosis. A. Measurement of volume of adenomyosis on pretreatment T2WI; B. Measurement of V T2WI on post-treatment T2WI; C. Measurement of V DWI on post-treatment DWI; D. Measurement of NPV on post-treatment CE-T1WI. Table 2. Inter-reader agreement in classifying ablation types on T2WI and DWI. T2WI types Radiologist2 Kappa pRadiologist1 Type I Subtype IIa Subtype IIb Type I 21 1 3 Subtype IIa 6 94 22 Subtype IIb 3 5 34 0.627 0.000 DWI types Radiologist2 Kappa PRadiologist1 Type I Subtype IIa Subtype IIb Type I 9 1 3 Subtype IIa 5 102 16 Subtype IIb 0 9 44 0.651 0.000 Table 3. Inter-reader agreement in measuring V T2WI and V DWI . Radiologist 1 Radiologist 2 ICC P V T2WI of type I (cm 3 ) 15.0 (5.3–33.5) 13.9 (4.6–38.0) 0.951 0.000 V T2WI of type II (cm 3 ) 50.4 (23.2–114.8) 53.2 (26.1–124.4) 0.958 0.000 V DWI of type I (cm 3 ) 9.4 (3.7–24.6) 7.1 (3.0–24.2) 0.947 0.000 V DWI of type II (cm 3 ) 57.1 (33.5–122.3) 55.9 (34.2–123.8) 0.975 0.000 INTERNATIONAL JOURNAL OF HYPERTHERMIA 5 DWI were associated with lower NPV and NPVR, indicating that a type I ablation area on T2WI or DWI may suggest rela - tively poor treatment efficiency. Previous studies have also found that lesion volume and EEF are related to NPVR [ 17 , 18 ], which may explain why EEF and lesion volume were also different between type I and type II. Figure 4. Linear correspondence between V T2WI and NPV and V DWI and NPV. A. The scatter diagram of correspondence betweenV T2WI and NPV; B. The scatter dia - gram of correspondence between V DWI and NPV. Table 4. Comparison of clinical and treatment parameters between T2WI-type I and T2WI- type II group. Ablation types on T2WI pType I Type II Age 39 ± 5 41 ± 5 0.081 BMI 23.1 ± 2.4 23.1 ± 2.9 0.987 Adenomyosis volume (cm 3 ) 80.7 (45.6–112.7) 109.9 (51.2–225.7) 0.084 Types of adenomyosis (diffuse/focus) 22/9 126/32 0.340 Position involved in uterus Anterior (yes/no) 18/13 76/82 0.310 Posterior (yes/no) 21/10 105/53 0.890 Fundus (yes/no) 8/23 56/102 0.310 NPV (cm 3 ) 21.2 (8.7–42.8) 49.1 (27.9–94.9) 0.000 � Sonication power (W) 400 (366–400) 400 (373–400) 0.400 treatment time (min) 80.7 ± 29.5 78.8 ± 51.9 (40.8–94.3) 0.535 Sonication time (s) 840.2 ± 373.3 882.4 ± 581.0 0.754 Sonication dose (kJ) 309.9 ± 159.3 343.4 ± 229.3 0.446 EEF (kJ/cm 3 ) 4.0 (2.5-5.5) 2.8 (1.4–4.7) 0.061 NPVR (%) 27.2 (17.4–44.7) 54.6 (42.0–73.9) 0.000 � IT (d) 2.0 (1.0–3.0) 2.5 (1.0–6.0) 0.238 T2WI: T2-weighted imaging; T1WI: T1-weighted imaging; DWI: diffusion-weighted imaging; CE-MRI: contrast-enhanced MRI; TR: repetition time; TE: echo time; FOV: field of view; NEX: number of excitations. Table 5. Comparison of clinical and treatment parameters between DWI-type I and DWI-type II group. Ablation types on DWI PType I Type II Age 40 ± 6 41 ± 6 0.628 BMI 23.7 ± 2.5 23.1 ± 2.9 0.444 Adenomyosis volume (cm 3 ) 68.7 (33.6–93.9) 148.2 (76.3-266.1) 0.001 � Types of adenomyosis (diffuse/focus) 10/5 138/36 0.254 Position involved in uterus Anterior (yes/no) 7/8 87/87 0.804 Posterior (yes/no) 11/4 115/59 0.568 Fundus (yes/no) 1/14 63/111 0.020 NPV (cm 3 ) 13.9 (6.3–26.6) 58.8 (30.1–118.4) 0.000 � Sonication power (W) 400 (359–400) 400 (390–400) 0.458 treatment time (min) 70.2 ± 32.1 79.8 ± 49.9 0.481 Sonication time (s) 785.3 ± 349.2 882.7 ± 564.9 0.526 Sonication dose (kJ) 272.3 ± 151.4 343.3 ± 23.6 0.245 EEF (kJ/cm 3 ) 4.1 (2.5–6.8) 2.0 (1.4–4.3) 0.044 � NPVR (%) 32.5 (13.4–48.5) 43.0 (30.0–68.7) 0.006 � IT (d) 1.0 (1.0–1.3) 2.5 (1.0–5.3) 0.105 BMI: body mass index; EEF: energy efficiency factor; IT: interval time between HIFU procedure and MRI scanning post-treatment. 6 M. SI ET AL. The type II ablation areas on T2WI and DWI were further classified into subtypes IIa and IIb on the basis of the signal intensity of their rims. Sixty-three percent (119/189) of the ablation areas on T2WI belonged to subtype IIa, while 64.0% (121/189) of the ablation areas on DWI belonged to subtype IIa. Each of them had a core of heterogeneous signal inten - sity and a rim of hyperintensity that delineated the edge of the ablation area. This distinctive appearance may be due to the difference in energy distribution in the ablation area [ 19 , 20 ]. The central part absorbed more ultrasonic energy and resulted in coagulative necrosis, while the peripheral part absorbed less ultrasonic energy. Its pathological change was cytotoxic edema, which presented high intensity on T2WI and DWI. This phenomenon was also observed in previous studies [ 21 , 22 ]. The remaining ablation areas on T2WI and DWI were clas - sified as subtype IIb, and they had apparent rims of hypoin - tensity. It is noteworthy that the median IT of patients with subtype IIb ablation areas on T2WI and DWI was longer. The IT of some patients in our study exceeded 2 days. Generally, MRI scans within 2 days after the HIFU treatment would allow physicians to better understand the success of the pro - cedure. We still enrolled these patients in our study because we found the IT may somehow impact the image features of the ablation area on T2WI and DWI. When the IT was long enough, the local micro-hemorrhage would gradually trans - form into paramagnetic material, such as deoxyhemoglobin or methemoglobin, which could appear as low signal inten - sity on T2WI and DWI [ 23 , 24 ]. In the meantime, the local cytotoxic edema also faded over time [ 21 ]. The MR findings of subtype IIb on T2WI and DWI were not observed in previ - ous reports [ 13 , 25 , 26 ]. This difference might be caused by the different objects and IT in various studies. The EEF, treat - ment time, and sonication time between the T2WI-subtype IIa group and the T2WI-subtype IIb group showed statistical differences. While, the NPVR, NPV, and sonication power between the DWI-subtype IIa group and the DWI-subtype IIb group also showed statistical differences. The results above showed that the HIFU parameters and treatment difficulty of adenomyosis may affect the appearance of lesions on T2WI and DWI to some extent. However, the specific mechanism remained unclear. Both V T2WI and V DWI presented high correlations with NPV, which may imply that T2WI and DWI have the potential to be optional methods for evaluating the efficiency of HIFU treatment for adenomyosis. The V T2WI of types I and II were Table 6. Comparison of clinical and treatment parameters between T2WI-subtype IIa and T2WI-subtype IIb group. Ablation subtypes on T2WI pSubtype IIa Subtype IIb Age 41 ± 6 40 ± 5 0.118 BMI 23.1 ± 2.5 23.4 ± 3.4 0.290 Adenomyosis volume (cm 3 ) 83.3 (41.7–164.9) 96.4 (56.7–203.7) 0.600 Types of adenomyosis (diffuse/focus) 98/20 28/12 0.076 Position involved in uterus Anterior (yes/no) 59/59 17/23 0.412 Posterior (yes/no) 77/41 28/12 0.451 Fundus (yes/no) 43/75 13/27 0.653 NPV (cm 3 ) 43.5 (19.7–80.6) 49.9 (28.9–104.4) 0.300 Sonication power (W) 400 (385–400) 400 (369–400) 0.064 treatment time (min) 78.3 ± 55.5 80.8 ± 40.0 0.006 � Sonication time (s) 841.2 ± 544.4 1004.1 ± 670.6 0.019 � Sonication dose (kJ) 331.3 ± 216.4 386.5 ± 260.3 0.055 EEF (kJ/cm 3 ) 2.4 (1.5–7.8) 2.9 (1.9–4.6) 0.015 � NPVR (%) 50.7 (40.1–67.1) 58.0 (38.0–75.2) 0.145 IT (d) 1 (1–2) 5 (2–7) 0.000 � BMI: body mass index; EEF: energy efficiency factor; IT: interval time between HIFU procedure and MRI scanning post-treatment. Table 7. Comparison of clinical and treatment parameters between DWI-subtype IIa and DWI-subtype IIb group. Ablation subtypes on DWI pSubtype IIa Subtype IIb Age 43 ± 4 38 ± 7 0.604 BMI 23.1 ± 2.1 22.5 ± 2.7 0.277 Adenomyosis volume (cm 3 ) 88 (57.5–260.9) 98.6 (51.7–175.3) 0.064 Types of adenomyosis (diffuse/focus) 99/21 39/15 0.122 Position involved in uterus Anterior (yes/no) 62/58 25/29 0.512 Posterior (yes/no) 77/43 38/16 0.424 Fundus (yes/no) 45/75 18/36 0.597 NPV (cm 3 ) 52.6 (21–113.5) 54.6 (19.3–83.2) 0.013 � Sonication power (W) 400 (373–400) 398 (371–400) 0.013 � treatment time (min) 78.0 ± 54.6 83.3 ± 37.1 0.326 Sonication time (s) 870 ± 477 784 ± 483 0.172 Sonication dose (kJ) 353.7 ± 181 309 ± 187.5 0.241 EEF (kJ/cm 3 ) 2.6 (1.2–7.9) 3.1 (1.6–4.3) 0.881 NPVR (%) 44.1 (35.8–68.5) 67.7 (35.2–79.5) 0.026 � IT (d) 1 (1–3) 6 (3–6) 0.000 � BMI: body mass index; EEF: energy efficiency factor; IT: interval time between HIFU procedure and MRI scanning post-treatment. INTERNATIONAL JOURNAL OF HYPERTHERMIA 7 statistically different from their corresponding NPV. Neither V DWI of type I nor V DWI type II V DWI differed significantly from their corresponding NPV. Even though most type I ablation areas on DWI were relatively ill-defined, their outlines were still visible, and their V DWI was comparable with NPV. For patients whose ablation areas were unrecognized, we defined their V DWI as 0 cm 3 . However, no difference was found between their V DWI and NPV. Because they had rather low NPV (less than 10 cm 3 ). For these reasons, DWI has high accuracy in the volume measurement of ablation areas of all types. However, T2WI is not suitable for evaluating the effi - ciency of HIFU treatment for adenomyosis independently, considering the limited accuracy of volume measurement for ablation areas. The classification of ablation area types on T2WI and DWI between two radiologists reached “fairly good” standards (j ¼ 0.627 and 0.651) in Kappa consistency tests. This meant that a “gray zone” existed in the classifica - tion of area types on DWI. However, it may have little effect on the value of DWI in efficiency evaluation, because DWI achieved high accuracy in the measurement of ablation areas of both type I and type II. Based on the above analysis, we believe that DWI may replace CE-T1WI in NPVR measure - ments and calculations to some extent. There are still some limitations in this study: first, it is a retrospective study, and bias is inevitable; furthermore, the interval time of MRI reexamination post-treatment for some patients exceeded 2 days, which may have some impact on the results; and finally, patients were treated by physicians with different seniorities.

Both T2WI and DWI have the potential to evaluate the effi - ciency of high-intensity focused ultrasound in adenomyosis ablation, and DWI may be an alternative to CE-T1WI. Author contributions Si Ma: manuscript writing, data management, and analysis. Mingmei Tang, Xueke Qiu, and Yang Liu (third author): Classification and ROI delineation. Chunmei Gong: Conceptualization. Yan Hu: chart drawing. Fajin Lv and Yang Liu: project development and administration. Ethical approval The protocol of this retrospective study was approved by the Ethics Committee, and the requirement for informed consent was waived (eth - ics approval number K2023-115). All patient data were anonymized for reporting purposes. Disclosure statement No potential conflict of interest was reported by the author(s). Funding The authors reported there is no funding associated with the work fea - tured in this article. Data availability statement The authors confirm that data supporting the findings of this study are available upon request from the corresponding author. The data are not publicly available because they contain information that can comprom - ise the privacy of the research participants.

0[1] Zhai J, Vannuccini S, Petraglia F, et al. Adenomyosis: mechanisms and pathogenesis. Semin Reprod Med. 2020;38(2-03):129–143. doi: 10.1055/s-0040-1716687. 0[2] Chen Q, Li YW, Wang S, et al. Clinical manifestations of adeno - myosis patients with or without pain symptoms. J Pain Res. 2019; 12:3127–3133. doi: 10.2147/JPR.S212117. 0[3] Harada T, Khine YM, Kaponis A, et al. The impact of adenomyosis on women’s fertility. Obstet Gynecol Surv. 2016;71(9):557–568. doi: 10.1097/OGX.0000000000000346. 0[4] Bachu VS, Kedda J, Suk I, et al. High-intensity focused ultrasound: a review of mechanisms and clinical applications. Ann Biomed Eng. 2021;49(9):1975–1991. doi: 10.1007/s10439-021-02833-9. 0[5] Marques ALS, Andres MP, Kho RM, et al. Is high-intensity focused ultrasound effective for the treatment of adenomyosis? A system - atic review and meta-analysis. J Minim Invasive Gynecol. 2020; 27(2):332–343. doi: 10.1016/j.jmig.2019.07.029. 0[6] Ou KY, Jeng CJ, Long CY, et al. A comparison of the cost-utility of ultrasound-guided high-intensity focused ultrasound and hys - terectomy for adenomyosis: a retrospective study: is the cost- effectiveness of HIFU for adenomyosis and fibroids feasible? BJOG. 2018;125(6):763–764. doi: 10.1111/1471-0528.15115. 0[7] Fan TY, Zhang L, Chen W, et al. Feasibility of MRI-guided high intensity focused ultrasound treatment for adenomyosis. Eur J Radiol. 2012;81(11):3624–3630. doi: 10.1016/j.ejrad.2011.05.036. 0[8] Mindjuk I, Trumm CG, Herzog P, et al. MRI predictors of clinical success in MR-guided focused ultrasound (MRgFUS) treatments of uterine fibroids: results from a single Centre. Eur Radiol. 2015; 25(5):1317–1328. doi: 10.1007/s00330-014-3538-6. 0[9] Keserci B, Duc NM. Magnetic resonance imaging features influ - encing high-intensity focused ultrasound ablation of adenomyo - sis with a nonperfused volume ratio of �90% as a measure of clinical treatment success: retrospective multivariate analysis. Int J Hyperthermia. 2018;35(1):626–636. doi: 10.1080/02656736.2018. 1516301. [10] Prince MR, Zhang H, Zou Z, et al. Incidence of immediate gado - linium contrast media reactions. AJR Am J Roentgenol. 2011; 196(2):W138–W143. doi: 10.2214/AJR.10.4885. [11] Rogosnitzky M, Branch S. Gadolinium-based contrast agent tox - icity: a review of known and proposed mechanisms. Biometals. 2016;29(3):365–376. doi: 10.1007/s10534-016-9931-7. [12] Laurent S, Elst LV, Copoix F, et al. Stability of MRI paramagnetic contrast media: a proton relaxometric protocol for transmetalla - tion assessment. Invest Radiol. 2001;36(2):115–122. doi: 10.1097/ 00004424-200102000-00008. [13] Liao D, Xiao Z, Lv F, et al. Non-contrast enhanced MRI for assess - ment of uterine fibroids’ early response to ultrasound-guided high-intensity focused ultrasound thermal ablation. Eur J Radiol. 2020;122:108670. doi: 10.1016/j.ejrad.2019.108670. [14] Ikink ME, Voogt MJ, van den Bosch MA, et al. Diffusion-weighted magnetic resonance imaging using different b-value combina - tions for the evaluation of treatment results after volumetric MR-guided high-intensity focused ultrasound ablation of uterine fibroids. Eur Radiol. 2014;24(9):2118–2127. doi: 10.1007/s00330- 014-3274-y. [15] Liu J, Wei J, Keserci B, et al. T2 � -weighted imaging in the assess - ment of the non-perfused volume of uterine fibroids following magnetic resonance-guided high-intensity focused ultrasound ablation. Int J Gynaecol Obstet. 2016;132(1):100–102. doi: 10. 1016/j.ijgo.2015.07.005. 8 M. SI ET AL. [16] Wu F, Chen WZ, Bai J, et al. Tumor vessel destruction resulting from high-intensity focused ultrasound in patients with solid malignancies. Ultrasound Med Biol. 2002;28(4):535–542. doi: 10. 1016/s0301-5629(01)00515-4. [17] Gong C, Lin Z, Lv F, et al. Magnetic resonance imaging parame - ters in predicting the ablative efficiency of high-intensity focused ultrasound for uterine fibroids. Int J Hyperthermia. 2021;38(1): 523–531. doi: 10.1080/02656736.2021.1904152. [18] Gong C, Yang B, Shi Y, et al. Factors influencing the ablative effi - ciency of high intensity focused ultrasound (HIFU) treatment for adenomyosis: a retrospective study. Int J Hyperthermia. 2016; 32(5):496–503. doi: 10.3109/02656736.2016.1149232. [19] Chu KF, Dupuy DE. Thermal ablation of tumours: biological mech - anisms and advances in therapy. Nat Rev Cancer. 2014;14(3):199– 208. doi: 10.1038/nrc3672. [20] Willis WT, Jackman MR, Bizeau ME, et al. Hyperthermia impairs liver mitochondrial function in vitro. Am J Physiol Regul Integr Comp Physiol. 2000;278(5):R1240–R1246. doi: 10.1152/ajpregu. 2000.278.5.R1240. [21] Harary M, Essayed WI, Valdes PA, et al. Volumetric analysis of magnetic resonance-guided focused ultrasound thalamotomy lesions. Neurosurg Focus. 2018;44(2):E6. doi: 10.3171/2017.11. FOCUS17587. [22] Walker MR, Zhong J, Waspe AC, et al. Acute MR-guided high- intensity focused ultrasound lesion assessment using diffusion- weighted imaging and histological analysis. Front Neurol. 2019; 10:1069. doi: 10.3389/fneur.2019.01069. [23] Gomori JM, Grossman RI, Goldberg HI, et al. Intracranial hemato - mas: imaging by high-field MR. Radiology. 1985;157(1):87–93. doi: 10.1148/radiology.157.1.4034983. [24] Gomori JM, Grossman RI, Hackney DB, et al. Variable appearances of subacute intracranial hematomas on high-field spin-echo MR. Am J Roentgenol. 1988;150(1):171–178. doi: 10.2214/ajr.150.1.171. [25] Jacobs MA, Gultekin DH, Kim HS. Comparison between diffusion- weighted imaging, T2-weighted, and postcontrast T1-weighted imaging after MR-guided, high intensity, focused ultrasound treatment of uterine leiomyomata: preliminary results. Med Phys. 2010;37(9):4768–4776. doi: 10.1118/1.3475940. [26] Pilatou MC, Stewart EA, Maier SE, et al. MRI-based thermal dosim - etry and diffusion-weighted imaging of MRI-guided focused ultra - sound thermal ablation of uterine fibroids. J Magn Reson Imaging. 2009;29(2):404–411. doi: 10.1002/jmri.21688. INTERNATIONAL JOURNAL OF HYPERTHERMIA 9