OL Oncology Letters 1792-1074 1792-1082 D.A. Spandidos 10.3892/ol.2021.13037 OL-0-0-13037 Articles PET/MRI is useful for early detection of pelvic insufficiency fractures after radiotherapy for cervical cancer AzumiMaho 1 MatsumotoMasuyo 1 SuzukiKaho 1 SasakiRyohei 2 UenoYoshiko 3 NogamiMunenobu 3 TeraiYoshito 1 Department of Obstetrics and Gynecology, Kobe University Graduate School of Medicine, Kobe, Hyogo 650-0017, Japan Department of Radiation Oncology, Kobe University Graduate School of Medicine, Kobe, Hyogo 650-0017, Japan Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Hyogo 650-0017, Japan Correspondence to: Professor Yoshito Terai, Department of Obstetrics and Gynecology, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho Chuo-ku, Kobe, Hyogo 650-0017, Japan, E-mail: yterai@med.kobe-u.ac.jp 11 2021 10 09 2021 22 5 776 12062021 25082021 Copyright: © Azumi et al. 2021 This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

Radiotherapy (RT) is used to manage cervical cancer, and pelvic insufficiency fracture (PIF) is known as a late complication of RT. The present study identified risk factors for PIF after radiotherapy for cervical cancer, and investigated its incidence rate. It also considered the usefulness of positron emission tomography/magnetic resonance imaging (PET/MRI) in PIF diagnosis. A total of 149 patients with cervical cancer who received definitive or adjuvant RT with/without concurrent chemotherapy between January 2013 and December 2018 were investigated in the present study and followed up for more than one month after RT at Kobe University Hospital. The median follow-up period was 32 months (range, 1–87 months), and the median age of all patients was 66 years (age range, 34–90 years). Computed tomography (CT), MRI, PET/CT or PET/MRI were used for image examination. Among the 149 patients, 31 (20.8%) developed PIF. The median age of these patients was 69 years (age range, 44–87 years). Univariate analysis using the log-rank test demonstrated that age (≥60 years) was significantly associated with PIF. The median maximum standardized uptake value of PIF sites on PET/CT was 4.32 (range, 3.04–4.81), and that on PET/MRI was 3.97 (range, 1.21–5.96) (P=0.162). Notably, the detection time of PIF by PET/MRI was significantly earlier compared with PET/CT (P<0.05). The incidence of PIF after RT for cervical cancer was 20.8%, and age was significantly associated with risk factors for such fractures. Taken together, these results suggest that PET/MRI, which offers the advantage of decreased radiation exposure to the patient, is useful for diagnosing PIF and can detect it earlier than PET/CT imaging.

 uterine cervical neoplasms radiotherapy insufficiency fractures positron-emission tomography magnetic resonance imaging Introduction

Radiotherapy (RT) is used to manage cervical cancer, either postoperatively or as a primary treatment (1), and major improvements in definitive RT have improved survival rates over the past 15–20 years (2). Pelvic insufficiency fracture (PIF) is a late complication of the therapy (3). According to a meta-analysis by Sapienza et al (4), the overall incidence of PIF is between 10–18%, and the most common site for PIF is the sacrum. PIF can lead to intractable pain and limited mobility, which can seriously impact the quality of life of cervical cancer survivors (5). Risk factors associated with cervical cancer include age, postmenopausal period, low body weight and osteoporosis (5–14). Recently, intensity-modulated radiation therapy (IMRT) has been introduced to reduce acute and chronic gastrointestinal disorders (15). However, the most effective method in reducing PIF remains unclear.

Considering the increased number of cervical cancer survivors (16), it is important to clarify what factors increase the risk of PIF after RT for patients with cervical cancer, and detect PIF in the early phase before patients develop any symptoms. Most reports on PIF have used computed tomography (CT) and magnetic resonance imaging (MRI) as inspection methods (5–10,12–14,17–21). According to Lapina and Tiškevičius (22), the sensitivity and specificity in PIF diagnosis are 100 and 95.3% for MRI and 74.6 and 89.7% for CT, respectively. Recently, positron emission tomography (PET) is attracting great interest as an imaging test for the follow-up of cervical cancer recurrence (23). However, to the best of our knowledge, only a few reports have used PET (5,11,19,21), and PET/MRI is yet to be investigated (4). The present study aimed to identify risk factors for PIF and investigate its incidence rate. In addition, the usefulness of PET/MRI in PIF diagnosis was assessed.

 Materials and methods <sec> <title>Patient characteristics

The present study investigated 149 patients with cervical cancer who received definitive or adjuvant RT with/without concurrent chemotherapy between January 2013 and December 2018, and were followed up for more than one month after RT at Kobe University Hospital. Data collection ended in May 2020. The median follow-up period was 32 months (range, 1–87 months), and the median age of all patients was 66 years (age range, 34–90 years). The histopathological type was squamous cell carcinoma (SCC) in 121 cases (81.2%), adenocarcinoma in 26 cases (17.4%), adenosquamous in one case (0.7%) and clear cell carcinoma in one case (0.7%). The median body mass index (BMI) was 21 (range, 14–40). A total of 27 cases (18.1%) were in The International Federation of Gynecology and Obstetrics (FIGO) stage (24) was I, 62 cases in stage II (41.6%), 40 cases in stage III (26.8%) and 20 cases in stage IV (13.4%). A total of 123 patients (82.6%) received RT as a definitive treatment, 26 (17.4%) received RT as postoperative adjuvant therapy, and 110 (73.8%) received concurrent chemoradiotherapy. Table I presents patient characteristics, according to the presence or absence of PIF. Notably, no significant differences were observed between both groups.

The inclusion criteria were as follows: Patients whose medical records and prognosis could be obtained from medical records, and patients who had a definite diagnosis of cervical cancer and were followed up for at least one month after the end of radiotherapy. The exclusion criteria were as follows: Patients who were offered to not participate in this study based on publicly available information, and patients participating in or planning to participate in clinical studies involving drug therapy interventions.

 Treatment

For definitive RT, external beam RT (EBRT) was performed using the four-field box technique, and intracavitary brachytherapy (ICBT) was subsequently performed using a remote after-loading system, in combination with EBRT, using the central shield by anteroposterior/posteroanterior field technique at 10 MV photons. The upper limit of the standard irradiation field was the upper edge of the 5th lumbar vertebra, and the lower limit was at least 3 cm below the lower edge of the obturator foramen or the lower edge of the vaginal infiltration. The outer side of the anterior-posterior irradiation field was 1.5–2 cm outside the inner edge of the pelvis, and the anterior edge of the lateral irradiation field was ~0.5 cm in front of the anterior pubic symphysis. A total dose of EBRT (TrueBeam™; Varian Medical Systems) at 50.4 Gy (range, 40.0–50.4 Gy) and ICBT (Micro Selectron™; Nucletron BV) at 24 Gy (range, 18–30 Gy) was administered. For postoperative adjuvant RT, IMRT (TrueBeam™, Varian Medical Systems) was performed with a total dose of 50.4 Gy (range, 50.4–60.0 Gy). A total dose of 10.0 Gy (range, 3.6–16.2 Gy) lymph node boost (TrueBeam™; Varian Medical Systems) was given to 54 patients (36.2%) with positive lymph node metastasis. For concurrent chemoradiotherapy, cisplatin (Cisplatin Maruko™; Yakult Honsha Co., Ltd.) was administered once a week for 5–6 cycles at 40 mg/m2.

 Follow-up and diagnostic criteria of PIF

For cervical cancer, follow-up started 1 month after the end of radiation therapy, and in principle, it was continued for 5 years in the outpatient department. Follow-up was conducted every 1 to 3 months in the 1st and 2nd years, every 3 to 6 months in the 3rd year, and every 6 months in the 4 to 5th years. Patients underwent imaging tests twice a year for the first 2 years and then annually thereafter. CT, MRI, PET/CT or PET/MRI were used as follow-up imaging methods. At the discretion of the two radiologists and two gynecologists, a recurrent check or follow-up was performed, and a CT, PET/CT or PET/MRI was selected. In addition, an MRI was selected depending on the symptoms, such as buttock or lower back pains. During the median follow-up period, 121 patients (81.2%) underwent imaging with PET, while 66 patients (44.3%) underwent imaging with PET/MRI. Patients who received multiple types of tests for the recurrence check or follow-up were excluded from the analysis.

The diagnostic criteria for PIF was as follows: CT for a fracture line or osteosclerosis without osteolytic changes, MRI for T1 hypointensity and T2 hyperintensity without a soft tissue mass, and PET for mild and diffuse fluorodeoxyglucose (FDG) accumulation. All images were diagnosed by two radiologists at Kobe University Hospital.

 Statistical analysis

Statistical analysis was performed using R version 4.0.0 (https://cran.r-project.org/bin/windows/base/). For each factor in the fracture (n=31) and non-fracture groups (n=118), χ2 or Fisher's exact tests were used to assess the differences in frequency. The Mann-Whitney U test was used to compare differences between continuous variables. Each exposure or each predictor was assessed using the log-rank test as univariate analysis. P<0.05 was considered to indicate a statistically significant difference. This observational study complied with the STROBE guidelines (25).

 Results <sec> <title>Incidence

Among the 149 patients investigated in the present study, 31 (20.8%) developed PIF, and the median time for developing the condition was 17 months (range, 2–47 months). The median age of the fracture group was 69 years, of which 28 patients (90.3%) were postmenopausal. The median age of the non-fracture group was 66 years, of which 94 patients (79.7%) were postmenopausal. Notable, no significant differences in age and postmenopausal status were observed between the two groups. The median BMI was 20.9 in the fracture group and 21.0 in the non-fracture group, and the median number of deliveries was two in both the fracture and non-fracture groups, respectively. With regards to medical history, two patients (6.5%) had diabetes, while three (9.7%) had osteoporosis in the fracture group, whereas 15 patients (12.7%) had diabetes, four (3.4%) had rheumatic arthritis and 14 (11.9%) had osteoporosis in the non-fracture group. There were no significant differences in medical history between the two groups. No patients received hormone replacement therapy (HRT) in the fracture group, while 10 patients received HRT in the non-fracture group. The histopathological types were 27 cases (87.1%) of SCC, three (9.7%) of adenocarcinoma and one (3.2%) of adenosquamous in the fracture group, compared with 94 cases (79.7%) of SCC, 23 (19.5%) of adenocarcinoma and one (0.8%) of clear cell carcinoma in the non-fracture group. There were no significant differences in histopathological types between the two groups. A total of two cases (6.5%) were in FIGO stage I, 14 (45.2%) in stage II, 13 (41.9%) in stage III and two (6.5%) in stage IV in the fracture group. Conversely, 25 cases (21.2%) were in FIGO stage I, 48 (40.7%) in stage II, 27 (22.9%) in stage III and 18 (15.3%) in stage IV in the non-fracture group. These results suggest that the FIGO stage was slightly higher in the fracture group compared with the non-fracture group (P=0.0552). In the facture group, no patients received adjuvant concurrent chemoradiotherapy (CCRT), two (6.5%) received adjuvant RT, 22 (71.0%) received definitive CCRT and seven (22.6%) received definitive RT. Conversely, in the non-fracture group, 12 patients (10.2%) received adjuvant CCRT, 12 (10.2%) received adjuvant RT, 76 (64.4%) received definitive CCRT and 18 (15.3%) received definitive RT. A total of 28 patients (90.3%) received ICBT, 11 (35.5%) received a lymph node boost and two (6.5%) received IMRT in the fracture group, whereas 94 patients (79.7%) received ICBT, 43 (36.4%) received a lymph node boost and 23 (19.5%) received IMRT in the non-fracture group. Notably, no significant differences in the type of treatment were observed between the two groups. A total of 13 patients (41.9%) had PET/CT as the imaging test, while 15 (48.4%) had PET/MRI in the fracture group. Conversely, 68 patients (57.6%) had PET/CT as the imaging test, while 51 (43.2%) had PET/MRI in the non-fracture group. There were no significant differences in image examination between the two groups (Table I).

 Sites

Table II presents the result of 31 patients who were diagnosed with PIF, 17 of which (54.8%) were asymptomatic. A total of 28 patients has their fracture site in the sacrum, nine in the pubis and four in the lumber spinal vertebrae. A total of 11 patients (35.5%) had multiple fractures (Table II), while 17 (54.8%) were asymptomatic. Furthermore, 10 patients (32.3%) were diagnosed with PIF on CT, one (3.2%) on MRI, 10 (32.3%) on PET/CT and 10 (32.3%) on PET/MRI. In addition, 12/17 patients (70.6%) with asymptomatic PIF were diagnosed by PET-CT or PET-MRI.

 Analysis of risk factors

The associations between PIF, patient characteristics and treatment related factors were assessed (Table III). Univariate analysis using the log-rank test demonstrated that age (≥60 years) was significantly associated with PIF (P<0.05). However, BMI <21, postmenopausal status, parity ≥3, SCC, FIGO stage ≥III, ICBT, a lymph node boost, IMRT and definitive RT were not significant risk factors for developing PIF.

 Characteristics of PET/CT and PET/MRI in PIF diagnosis

The association between PIF diagnosis and the method of fracture detection, using either PET/CT or PET/MRI, was assessed. The median maximum standardized uptake value (SUVmax) of PIF sites on PET/CT was 4.32 (range, 3.04–4.81) and that on PET/MRI was 3.97 (range, 1.21–5.96), which was not statistically significant (P=0.162; Table II). Kaplan-Meier analysis was performed to compare patients who underwent image follow-up with PET/CT or PET/MRI during the median follow-up period, except for those who underwent both PET/CT and PET/MRI. The results demonstrated that the detection time of PIF by PET/MRI (n=40) was significantly earlier than that of PET/CT (n=55) (P=0.0367; Fig. 1).

 A diffuse linear FDG uptake

A diffuse linear FDG uptake was observed in parallel with the sacroiliac joint as a characteristic pattern of sacral fracture (Fig. 2). In addition, the features of FDG uptake were similar between PET/CT and PET/MRI in the present study (Fig. 2). The earliest MR sign, medullary edema, was seen and characterized as T1 hypointensity and T2 hyperintensity (Fig. 3).

 Discussion

The results of the present study demonstrated that PET/MRI was useful for early diagnosis of recurrent cervical cancer, and for early detection of PIF compared with PET/CT. Currently, a few studies have investigated the incidence and risk factors of PIF via PET/CT as the imaging method; however, the effects of PET/MRI have not yet been investigated (5,26–28). To the best of our knowledge, the present study was the first to report the incidence rate and risk factors of PIF via PET/MRI as the imaging method.

Radiation therapy of the pelvis can cause demineralization and decrease elastic resistance of the bone matrix (28). According to a recent meta-analysis, the overall incident rate of PIF was 14% (95% confidence interval, 10–18%, based on 21 studies) (4). The incidence rate of PIF in the present study was 20.8%, which was slightly higher than the meta-analysis. Age has also been reported as a risk factor in several studies (5–11), as well as the present study. However, the results of the present study failed to exhibit significant differences in other risk factors between the groups. This may have been due to the small sample size used in the present study. Previous studies have identified risk factors for PIF, including postmenopausal status (6–8,10,12), low body weight (5,9–11) and osteoporosis (13,14). In the present study, no significant differences were observed between the groups for some parameters such as menopausal status or BMI.

This study was a retrospective study and thus relies on medical records for information. However, data on the presence or absence of osteoporosis was insufficient and thus was not evaluated. Prospective studies are required to incorporate this data by screening patients prior to chemoradiotherapy. It has been reported that rheumatoid arthritis (RA) (12) and HRT (14) are risk factors for PIF. However, the present study was unable to assess the associations between PIF and RA or HRT as only a few patients had RA or had received HRT. Oh et al reported that a high radiation dose is a risk factor of PIF (11). At Kobe University Hospital, both EBRT and IMRT are performed during the treatment period in some cases and at the discretion of the radiation oncologist; therefore, it was not possible to calculate the total pelvic irradiation dose by simply adding the irradiation doses together in the present study. Concurrent chemoradiotherapy is considered a risk factor of PIF; however, no previous studies have identified chemotherapy as a risk factor (7,12,17), and neither did the present study. Ogino et al (29) reported that a history of three or more deliveries is a risk factor of PIF, but we did not find a significant difference in the present study. The sacrum is the predominant site of PIF and is reported to account for 73.6% of incidences (4). In the present study, fractures in the sacrum occurred in 27/31 patients (87.0%).

According to some reports, 43–77% of PIF cases are symptomatic, and patients experience pain that affects their quality of life (5,30,31). Maintaining the quality of life for cancer survivors has become an important issue in recent years. It is even more important in patients with cervical cancer as it has a relatively young age of onset and has a long survival time when completely cured (32). According to the results of the present study, 45.2% of patients with PIF were symptomatic. Treatment is generally conservative, with analgesic administration and rehabilitation (33). Early diagnosis and treatment may improve pain, immobilization and thus mortality (33).

Imaging tests are rarely performed to solely diagnose PIF, and PIF is often found when checking the recurrence of cervical cancer. In previous reports, PET/CT was demonstrated to be more sensitive in detecting recurrence or metastasis compared with CT or MRI (34,35) PET/MRI has been reported to exhibit comparable or better performance than PET/CT in detecting the recurrence of gynecologic cancer (36). Furthermore, PET/MRI has the advantage of lesion detection within the brain, breast, liver, kidneys, bone and pelvic lesion compared with PET/CT (37). In addition, PET/MRI offers the advantage of decreased radiation exposure compared with PET/CT (38). It is important for patients with cervical cancer to not only assess the detection of their recurrent disease but also to avoid the radiation-induced malignancies that can occur during a long follow-up period. Currently, several studies have reported on the usefulness of CT and MRI data in PIF diagnosis (20,22,39), and according to Lapina and Tiškevičius (22), the sensitivity and specificity in PIF diagnosis are 100 and 95.3% for MRI and 74.6 and 89.7% for CT, respectively. Although the characteristics of PET/CT in PIF diagnosis have been previously reported (5,27,28), their value remains unclear. In the present study, 54.8% of patients with PIF were asymptomatic, 70.6% of which were diagnosed by PET-CT or PET-MRI. Conversely, a previous report has indicated that only 39% of patients with asymptomatic PIF were diagnosed by CT or MRI (20).

It has been reported that low-grade FDG uptake is often observed at the fracture site (5). A diffuse linear FDG uptake is observed in parallel with the sacroiliac joint as a characteristic pattern of sacral fracture (Fig. 2). It is important to distinguish between benign and malignant fractures. The SUVmax value and the FDG uptake pattern differ between malignant and benign fractures (40,41). These reports are based on PET/CT and cannot be generally applied to PET/MRI; however, the results of the present study demonstrated no significant differences in the degree of SUVmax between PET/CT and PET/MRI. In addition, the features of FDG uptake were similar between PET/CT and PET/MRI in the present study (Fig. 2). Furthermore, the results demonstrated that PET/MRI may detect PIF earlier than PET/CT (Fig. 1). This suggests that the detection time of PIF is earlier in MRI compared with CT (22). According to Grangier et al (42), the earliest MR sign, medullary edema, which is characterized as T1 hypointensity and T2 hyperintensity, is seen as early as 18 days after the onset of symptoms and persist as long as 516 days (Fig. 3). These findings suggest that PET/MRI imaging may detect PIF at an earlier stage before patients have developed symptoms compared with PET/CT, CT or MRI alone.

The present study is not without limitations. Given that this study was retrospective, there may be data that has not yet been revealed. Furthermore, the present study failed to conduct a questionnaire to determine whether patients had any symptoms, and only depended on medical records. Thus, patients may have had symptoms, even in cases that were considered asymptomatic. Furthermore, image follow-up was performed mainly for the purpose of checking the recurrence of cervical cancer and not for PIF examination. Given that the specific timing of image follow-up was left to the attending physician, there may have been some variation in the imaging timing for each patient. In addition, the attending physician decided which imaging medium to follow up with, and since it was not randomized, there may have been selection bias. Patients who had many images taken due to recurrence may have had PIF found earlier. There is also the possibility that SUVmax may have had different values depending on the elapsed time after fracture. Furthermore, PIF may have been found secondarily during the recurrence check for cervical cancer, and it was not clear when it occurred. In addition, measured SUVs on PET/CT and PET/MRI could be significantly different even in the same patient due to the difference in detectors and reconstruction methods. The present study did not include dose volume histogram. In some cases, both EBRT and IMRT were performed during the treatment period, and it was not possible to calculate the total pelvic irradiation dose simply by adding the irradiation doses. Therefore, it is possible that the prescribed dose was clarified by using the dose volume histogram. It has been reported that high irradiation dose is a risk factor for PIF (30). In the present study, it may have been possible to clarify whether the prescribed dose is a risk factor of developing PIF or not by analyzing the dose volume histogram.

In conclusion, the incidence of PIF after RT for cervical cancer was 20.8% in the present study. Age was significantly associated with the development of PIF. Furthermore, PET/MRI, which offers the advantage of decreased radiation exposure to the patient, may detect PIF at an earlier phase before patients develop symptoms, thus allowing them to undergo early counseling and treatment for bone health and to maintain a better quality of life after cervical cancer treatment. However, further trials and reports are required to establish the usefulness of PET/MRI in PIF cases.

 Acknowledgements

The authors of the present study would like to thank Dr Sonoko Suda, Dr Keitaro Yamanaka, Dr Mamiko Sawada, Dr Masako Tomimoto, Dr Keiichi Washio, Dr Maho Shimizu, Dr Ryosuke Takahashi, Dr Satoshi Nagamata and Dr Yuka Murata for collecting the data. They would also like to thank Dr Satoshi Senoh for technical advice on radiation therapy. These persons were affiliated with Kobe University Hospital.

 Funding

No funding was received.

 Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

 Authors' contributions

MA, MM, KS participated in data collection and analyzed and interpreted patient data. RS, YU nad MN analyzed and interpreted the radiological data. YT and MM confirm the authenticity of all the raw data. YT contributed to the study conception and design. MA and YT drafted the manuscript. All authors have read and approved the final manuscript.

 Ethics approval and consent to participate

The present study was approved by Kobe University Hospital Clinical and Translational Research Center (IRB no. B200206, October 23rd, 2020). Given that this study was an observational study using existing data, patient consent was waived. All patient data were analyzed retrospectively, and no tissues were examined. However, all study details were disclosed to patients, and they were given the option to refuse registration via e-mail.

 Patient consent for publication

Not applicable.

 Competing interests

The authors declare that they have no competing interests.

 References KlaitongCMeannuchEKaewbunpermUKlaiphibulePSinthusekTEP-1512: Radiotherapy at pelvis region in menopausal cervix cancer induce osteopenia/osteoporosisRadiother Oncol127(Suppl 1)S819S820201810.1016/S0167-8140(18)31821-8 SmallWJrKachnicLPostradiotherapy pelvic fractures cause for concern or opportunity for future research?JAMA29426352637200510.1001/jama.294.20.263516304079 ViswanathanANLeeLJEswaraJRHorowitzNSKonstantinopoulosPAMirabeau-BealeKLRoseBSVon KeudellAGWoJYComplications of pelvic radiation in patients treated for gynecologic malignanciesCancer12038703883201410.1002/cncr.2884925056522 SapienzaLGSalcedoMPNingMSJhingranAKloppAHCalsavaraVFSchmelerKMLeite GomesMJde Freitas CarvalhoEBaiocchiGPelvic insufficiency fractures after external beam radiation therapy for gynecologic cancers: A meta-analysis and meta-regression of 3929 patientsInt J Radiat Oncol Biol Phys106475484202010.1016/j.ijrobp.2019.09.01231580930 ParkSKimJLeeJParkIKPelvic insufficiency fracture after radiotherapy in patients with cervical cancer in the era of PET/CTRadiat Oncol J29269276201110.3857/roj.2011.29.4.26922984680 RamlovAPedersenEMRøhlLWormEFokdalLLindegaardJCTanderupKRisk factors for pelvic insufficiency fractures in locally advanced cervical cancer following intensity modulated radiation therapyInt J Radiat Oncol Biol Phys9710321039201710.1016/j.ijrobp.2017.01.02628332986 BazireLXuHFoyJPAmessisMMalhaireCCaoKDe La RochefordiereAKirovaYMPelvic insufficiency fracture (PIF) incidence in patients treated with intensity-modulated radiation therapy (IMRT) for gynaecological or anal cancer: Single-institution experience and review of the literatureBr J Radiol9020160885201710.1259/bjr.2016088528291401 CaiZXLiYYangZHGongSZhangMDChenJYMRI and associated clinical characteristics of pelvic insufficiency fracture in cervical carcinoma patients after radiation therapyChinese J Med Imaging Technol31148314862015 TokumaruSToitaTOguchiMOhnoTKatoSNiibeYKazumotoTKodairaTKataokaMShikamaNInsufficiency fractures after pelvic radiation therapy for uterine cervical cancer: An analysis of subjects in a prospective multi-institutional trial, and cooperative study of the Japan radiation oncology group (JAROG) and Japanese radiation oncology Study Group (JROSG)Int J Radiat Oncol Biol Phys84e195e200201210.1016/j.ijrobp.2012.03.04222583605 SchmelerKMJhingranAIyerRBSunCCEifelPJSolimanPTRamirezPTFrumovitzMBodurkaDCSoodAKPelvic fractures after radiotherapy for cervical cancer: Implications for survivorsCancer116625630201010.1002/cncr.2481120052724 OhDHuhSJParkWJuSGNamHLeeJEClinical outcomes in cervical cancer patients treated by FDG-PET/CT-based 3-dimensional planning for the first brachytherapy sessionMedicine (Baltimore)95e3895201610.1097/MD.000000000000389527336876 YamamotoKNagaoSSuzukiKKogikuASendaTYanoHKitaiMShiozakiTMatsuokaKYamaguchiSGynecologic Oncology Pelvic fractures after definitive and postoperative radiotherapy for cervical cancer: A retrospective analysis of risk factorsGynecol Oncol147585588201710.1016/j.ygyno.2017.09.03529055558 WeidenbacherBBormKOechsnerMPelvic fractures after radiotherapy for cervical cancerStrahlenther Onkol1942018 ShihKKFolkertMRKollmeierMAAbu-RustumNRSonodaYLeitaoMMBarakatRRAlektiarKMPelvic insufficiency fractures in patients with cervical and endometrial cancer treated with postoperative pelvic radiationGynecol Oncol128540543201310.1016/j.ygyno.2012.12.02123262211 TaLJtGSeDCcHHjnAEffects of pelvic radiotherapy for primary pelvic cancers (Review)Cochrane Database Syst Rev2018 ElitLReadeCJRecommendations for follow-up care for gynecologic cancer survivorsObstet Gynecol12612071214201510.1097/AOG.000000000000112926551194 MehmoodQBeardwoodMSwindellRGreenhalghSWarehamTBarracloughLLivseyJDavidsonSEInsufficiency fractures in patients treated with pelvic radiotherapy and chemotherapy for uterine and cervical cancerEur J Cancer Care (Engl)234350201410.1111/ecc.1210523889218 KujiSHirashimaYKomedaSTanakaAAbeMTakahashiNTakekumaMAsakuraHHaradaHNishimuraTFeasibility of extended-field irradiation and intracavitary brachytherapy combined with weekly cisplatin chemosensitization for IB2-IIIB cervical cancer with positive paraaortic or high common iliac lymph nodes: A retrospective reviewInt J Clin Oncol19341347201410.1007/s10147-013-0551-823546545 JungJParkGKimYSDefinitive extended-field intensity-modulated radiotherapy with chemotherapy for cervical cancer with para-aortic nodal metastasisAnticancer Res3443614366201425075072 UezonoHTsujinoKOtaYNaganoFSoejimaTMorikiKSasakiRPelvic insufficiency fracture after definitive radiotherapy for uterine cervical cancer: Retrospective analysis of risk factorsJ Radiat Res5411021109201310.1093/jrr/rrt05523685668 KimJLeeKJParkKRHaBKimYJJungWLeeRKimSCMoonHSJuWTreatment outcomes after adjuvant radiotherapy following surgery for patients with stage I endometrial cancerRadiat Oncol J34265272201610.3857/roj.2016.0164827703126 LapinaOTiškevičiusSSacral insufficiency fracture after pelvic radiotherapy: A diagnostic challenge for a radiologistMed50249254201425458963 NCCN clinical practice guidelines in oncologyCervical Cancer Version 4Natl Compr Cancer Netw2019 PecorelliSZiglianiLOdicinoFRevised FIGO staging for carcinoma of the cervixInt J Gynecol Obstet105107108200910.1016/j.ijgo.2009.02.00919342051 Von ElmEAltmanDGEggerMPocockSJGøtzschePCVandenbrouckefJPSTROBE InitiativeThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for reporting observational studiesBull World Health Organ85867872200710.2471/BLT.07.04512018038077 HalaçMMutSSSönmezogluKYlmazMHOzerHUsluIAvoidance of misinterpretation of an FDG positive sacral insufficiency fracture using PET/CT scans in a patient with endometrial cancer: A case reportClin Nucl Med32779781200710.1097/RLU.0b013e318148b40817885357 SalavatiAShahVWangZJYehBMCostourosNGCoakleyFVF-18 FDG PET/CT findings in postradiation pelvic insufficiency fractureJ Clin Imaging35139142201110.1016/j.clinimag.2009.12.02621377053 ChungYKLeeYKYoonBHSuhDHKooKHPelvic insufficiency fractures in cervical cancer after radiation therapy: A meta-analysis and reviewIn Vivo (Brooklyn)3511091115202110.21873/invivo.1235633622908 OginoIOkamotoNOnoYKitamuraTNakayamaHPelvic insufficiency fractures in postmenopausal woman with advanced cervical cancer treated by radiotherapyRadiother Oncol686167200310.1016/S0167-8140(03)00128-212885453 KwonJWHuhSJYoonYCChoiSHJungJYOhDChoeBKPelvic bone complications after radiation therapy of uterine cervical cancer: Evaluation with MRIAJR Am J Roentgenol191987994200810.2214/AJR.07.363418806132 KidoAYoshidaSShimodaEIshidaYHasegawaMKobayashiHHonokiKHorikawaHTanakaYWalking disability in patients with pelvic insufficiency fracture after radiotherapy for uterine cervical cancerProg Rehabil Med120160009201610.2490/prm.2016000932789206 ThapaNMaharjanMXiongYJiangDNguyenTPPetriniMACaiHImpact of cervical cancer on quality of life of women in Hubei, ChinaSci Rep811993201810.1038/s41598-018-30506-630097622 van den BlinkQUGarcezKHensonCCDavidsonSEHighamCEPharmacological interventions for the prevention of insufficiency fractures and avascular necrosis associated with pelvic radiotherapy in adultsCochrane Database Syst Rev4CD010604201829683475 ChoiHJJuWMyungSKKimYDiagnostic performance of computer tomography, magnetic resonance imaging, and positron emission tomography or positron emission tomography/computer tomography for detection of metastatic lymph nodes in patients with cervical cancer: Meta-analysisCancer Sci10114711479201010.1111/j.1349-7006.2010.01532.x20298252 PantolaSKalaSKalaCSampathSShuklaMComparative study of positron emission tomography/computed tomography and computed tomography in the evaluation of post-treatment carcinoma cervix patientsIndian J Nucl Med33194201201829962714 BroskiSMGoenkaAHKempBJJohnsonGBClinical PET/MRI: 2018 updateAJR Am J Roentgenol211295313201810.2214/AJR.18.2000129949413 RosenkrantzABFriedmanKChandaranaHMelsaetherAMoyLDingYSJhaveriKBeltranLJainRCurrent status of hybrid PET/MRI in oncologic imagingAm J Roentgenol206162172201610.2214/AJR.15.1496826491894 SchwartzMGavaneSCBou-AyacheJKolevVZakashanskyKPrasad-HayesMTaouliBChuangLKostakogluLFeasibility and feasibility and diagnostic performance of hybrid pet/mri compared with pet/ct for gynecological malignancies: A prospective pilot studyAbdom Radiol4334623467201810.1007/s00261-018-1665-229948057 UgurluerGAkbasTArpaciTOzcanNSerinMBone complications after pelvic radiation therapy: Evaluation with MRIJ Med Imaging Radiat Oncol58334340201410.1111/1754-9485.1217624716673 HeXZhaoLGuoXZhaoLWuJHuangJSunLXieCChenHDifferential diagnostic value of18F-FDG PET/CT for benign and malignant vertebral compression fractures: Comparison with magnetic resonance imagingCancer Manag Res1021052115201810.2147/CMAR.S16837430050321 MauchJTCarrCMCloftHDiehnFEReview of the imaging features of benign osteoporotic and malignant vertebral compression fracturesAJNR Am J Neuroradiol3915841592201810.3174/ajnr.A552829348133 GrangierCGarciaJHowarthNRMayMRossierPRole of MRI in the diagnosis of insufficiency fractures of the sacrum and acetabular roofSkeletal Radiol26517524199710.1007/s0025600502789342810

Comparison of the incidence of PIF in PET/CT and PET/MRI during the median follow-up period. The detection time of PIF by PET/MRI was significantly earlier than that by PET/CT. PIF, pelvic insufficiency fracture; PET, positron emission tomography; CT, computed tomography; MRI, magnetic resonance imaging; CI, confidence interval; NA, not available.

A diffuse linear FDG uptake parallel with the sacrum. The features of FDG uptake were similar between PET/CT and PET/MRI. (A) Patient 11 diagnosed by PET/CT (SUVmax 4.11). (B) Patient 19 diagnosed by PET/MRI (SUVmax 3.92). FDG, fluorodeoxyglucose; PET, positron emission tomography; CT, computed tomography; MRI, magnetic resonance imaging; SUVmax, maximum standardized uptake value.

Earliest MR sign. The earliest MR sign, medullary edema, which is characterized as T1 hypointensity and T2 hyperintensity. (A) Patient 22 diagnosed by PET/MRI (SUVmax 2.5) (B) T1-weighted image: Hypointensity and (C) T2-weighted image: Hyperintensity on the left side of the sacrum. PET, positron emission tomography; MRI, magnetic resonance imaging; SUVmax, maximum standardized uptake value.

Patient characteristics.

Characteristic Fracture (+) (n=31) Fracture (−) (n=118) P-value
Median age (range), years 69 (44–87) 66 (34–90) 0.0727
Median age of menopause (range), years 50 (40–57) 50 (40–60) 0.3720
Menopausal status 0.2938
  Postmenopause 28 (90.3%) 94 (79.7%)
  Premenopause 2 (6.5%) 20 (16.9%)
  Unknown 1 (3.2%) 4 (3.4%)
Body mass index, kg/m2 20.9 (16.8–34.8) 21.0 (13.7–40.3) 0.5858
Parity 2 (0–5) 2 (0–6) 0.4134
Medical history
  Diabetes mellitus 2 (6.5%) 15 (12.7%) 0.5265
  Rheumatoid arthritis 0 (0.0%) 4 (3.4%) 0.5804
  Osteoporosis 3 (9.7%) 14 (11.9%) 1
Hormone replacement therapy 0 (0.0%) 10 (8.5%) 0.1224
Histopathological type 0.1571
  Squamous cell carcinoma 27 (87.1%) 94 (79.7%)
  Adenocarcinoma 3 (9.7%) 23 (19.5%)
  Adenosquamous 1 (3.2%) 0 (0.0%)
  Clear cell carcinoma 0 (0.0%) 1 (0.8%)
FIGO stage 0.0552
  I 2 (6.5%) 25 (21.2%)
  II 14 (45.2%) 48 (40.7%)
  III 13 (41.9%) 27 (22.9%)
  IV 2 (6.5%) 18 (15.3%)
Treatment 0.2129
  Adjuvant CCRT 0 (0.0%) 12 (10.2%)
  Adjuvant RT 2 (6.5%) 12 (10.2%)
  Definitive CCRT 22 (71.0%) 76 (64.4%)
  Definitive RT 7 (22.6%) 18 (15.3%)
ICBT 0.2009
  Yes 28 (90.3%) 94 (79.7%)
  No 3 (9.7%) 24 (20.3%)
Lymph node boost 1
  Yes 11 (35.5%) 43 (36.4%)
  No 20 (64.5%) 75 (63.6%)
IMRT 0.1069
  Yes 2 (6.5%) 23 (19.5%)
  No 29 (93.5%) 95 (80.5%)
PET/CT 0.1561
  Yes 13 (41.9%) 68 (57.6%)
  No 18 (58.1%) 50 (42.4%)
PET/MRI 0.6859
  Yes 15 (48.4%) 51 (43.2%)
  No 16 (51.6%) 67 (56.8%)

FIGO, International Federation of Gynecology and Obstetrics; CCRT, computer-controlled radiation therapy; RT, radiotherapy; ICBT, intracavitary brachytherapy; IMRT, intensity-modulated radiation therapy; PET, positron emission tomography; CT, computed tomography; MRI, magnetic resonance imaging.

Characteristics of PIF.

Patient Age, years Time, months Site Symptom Diagnostic medium PET/CT (SUVmax) PET/MRI (SUVmax)
1 72 32 Sacrum Asymptomatic CT NA NA
2 66 28 Sacrum Buttocks pain CT NA NA
3 80 12 Lumber spinal vertebrae Buttocks pain PET/CT 3.68 NA
4 82 38 Sacrum Buttocks pain CT NA NA
5 81 33 Sacrumpubis Buttocks pain CT NA NA
6 76 25 Sacrum Asymptomatic PET/CT 4.52 NA
7 87 12 Sacrum Buttocks pain MRI NA NA
8 61 15 Sacrum Buttocks pain PET/CT 4.81 NA
9 66 34 Sacrumpubis Asymptomatic PET/CT 4.30 NA
10 46 36 Sacrumpubis Buttocks pain PET/MRI NA 5.96
11 67 4 Sacrum Asymptomatic PET/CT 4.11 NA
12 67 47 Sacrumpubis Asymptomatic PET/CT 4.81 NA
13 51 22 Sacrum Asymptomatic CT NA NA
14 58 14 Sacrum Buttocks pain CT NA NA
15 69 7 Sacrum Buttocks pain PET/MRI NA 4.06
16 79 9 Sacrumlumber spinal vertebrae Asymptomatic PET/MRI NA 4.02
17 82 2 Lumber spinal vertebrae Buttocks pain CT NA NA
18 60 32 Sacrum Buttocks pain PET/MRI NA 1.21
19 66 29 Sacrum Asymptomatic PET/MRI NA 3.92
20 69 13 Sacrum Asymptomatic PET/MRI NA 5.23
21 70 29 Sacrumpubis Asymptomatic CT NA NA
22 75 15 Sacrum Asymptomatic PET/MRI NA 2.50
23 52 18 Sacrum Asymptomatic PET/CT 3.04 NA
24 74 7 Lumber spinal vertebrae Asymptomatic PET/CT 4.27 NA
25 73 23 Sacrumpubis Buttocks pain PET/CT 4.33 NA
26 44 17 Sacrum Asymptomatic PET/MRI NA 3.39
27 80 8 Sacrumpubis Asymptomatic CT NA NA
28 62 21 Sacrumpubis Buttocks pain PET/MRI NA 4.07
29 76 6 Sacrumpubis Buttocks pain PET/CT 4.34 NA
30 67 8 Sacrumthoracic spinal vertebrae Asymptomatic CT NA NA
31 56 8 Sacrum Asymptomatic PET/MRI NA 3.88

PIF, pelvic insufficiency fracture; PET, positron emission tomography; CT, computed tomography; MRI, magnetic resonance imaging; SUVmax, maximum standardized uptake value; NA, not available.

Univariate analysis of risk factors for PIF.

Variable 95% CI P-value
Age, years (≥60) 47-NA <0.05
BMI, kg/m2 (<21) NA-NA 0.505
Postmenopausal status NA-NA 0.199
Parity (≥3) 33-NA 0.289
SCC NA-NA 0.622
FIGO stage (≥III) 33-NA 0.0732
ICBT 36-NA 0.438
Lymph node boost 36-NA 0.188
IMRT NA-NA 0.125
Definitive radiotherapy NA-NA 0.0816

PIF, pelvic insufficiency fracture; CI, confidence interval; BMI, body mass index; SCC, squamous cell carcinoma; FIGO, International Federation of Gynecology and Obstetrics; ICBT, intracavitary brachytherapy; IMRT, intensity-modulated radiation therapy; NA, not available.