Postradiation Damage to the Pelvic Girdle in Cervical Cancer Patients: Is Intensity-Modulated Radiation Therapy Safer Than Conventional Radiation?

Ioffe, Yevgeniya Jane Mikhailovna MD*; Hillen, Travis J. MD; Zhou, Gongfu PhD; Schwarz, Julie K. MD, PhD§; Massad, L. Stewart MD*; Powell, Matthew A. MD*; Hagemann, Andrea R. MD*; Mutch, David G. MD*; Thaker, Premal H. MD, MSCI*

International Journal of Gynecological Cancer:
doi: 10.1097/IGC.0000000000000117
Quality of Life

Objectives: Intensity-modulated radiation therapy (IMRT) is frequently utilized in the treatment of cervical cancer. Our study compared instances of pelvic fractures, osteonecrosis, and osteomyelitis posttreatment with conventional radiation therapy (RT) versus IMRT in patients with cervical carcinomas.

Methods: Eighty-three patients primarily treated with IMRT were case matched with 83 historical control subjects treated with conventional RT. Pretreatment and posttreatment computed tomography scans were reviewed. Logistic regression analysis was utilized to examine the effects of treatment type (conventional RT vs IMRT) on the occurrence of posttreatment pelvic bony structure complications while adjusting for confounders.

Results: In the IMRT group, 3 (4%) of 83 patients developed posttreatment sacral fractures (median follow-up, 51 months). In the conventional RT group, there were 14 pelvic girdle complications (17%): 9 fractures, 2 cases of osteonecrosis, and 3 cases of osteomyelitis (median follow-up, 43.5 months; odds ratio, 4.49 for conventional vs IMRT groups, P = 0.01; 95% confidence interval, 1.4–14.1). In addition, there were 4 cases of posttreatment osteoporosis in the conventional RT group. All patients with complications in the IMRT group and 11 of 13 in the conventional RT group were symptomatic.

Conclusions: Intensity-modulated radiation therapy is associated with a lower risk for pelvic girdle complications than conventional RT.

Author Information

*Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, and Departments of †Radiology, ‡Statistics, and §Radiation Oncology, Washington University in St Louis School of Medicine, St Louis, MO.

Address correspondence and reprint requests to Yevgeniya Jane Mikhailovna Ioffe, MD, Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Washington University in St Louis School of Medicine, 660 S Euclid Ave, 4th Floor Maternity Bldg, St Louis, MO 63110. E-mail:

The authors declare no conflicts of interest.

Received November 17, 2013

Accepted January 28, 2014

Article Outline

Cervical cancer is the fourth most common cause of death in women worldwide.1 Mortality rates for cervical cancer have decreased over the last several decades, likely due to successful prevention and early detection efforts2 and the addition of cisplatin chemotherapy to radiation regimens.3,4 Radiation therapy (RT) remains the core of definitive treatment for women with cervical carcinomas with both early- and late-stage and recurrent disease.1 As disease survival has improved because of addition of cisplatin chemotherapy and advances in RT, complications of treatment become increasingly relevant. However, there is a paucity of data regarding long-term complications of RT, particularly the effects of intensity-modulated RT (IMRT) on the pelvic girdle of women with cervical cancer.

In 1995, Grigsby and colleagues5 published the results of a single-institution study of 207 cases of advanced carcinoma of the cervix, endometrium, vulva, and vagina that had been treated with pelvic and groin irradiation between 1954 and 1992. Women received treatment via anterior posterior–posterior anterior (AP/PA) fields encompassing the pelvis and the groins. In this cohort, the femoral neck fracture rate was 4.8%. Cigarette smoking was associated with risk of femoral fracture, but not age, weight, or radiation dose.

Several other retrospective studies have described the incidence and site distribution of pelvic insufficiency fractures in patients with cervical carcinomas treated with conventional RT during the 1990s and early 2000s. A recent study by Schmeler et al6 from MD Anderson Cancer Center reported an overall pelvic fracture rate of 9.7%, with most fractures diagnosed within 2 years after treatment. Other retrospective studies examining the insufficiency fracture rate in women with cervical cancers who had received RT found the fracture rate to range between 8.2% and 20%.7–12 Only 1 small study prospectively followed a cohort of women treated with RT in the setting of advanced cervical cancer with imaging. The incidence of pelvic fractures as detected by magnetic resonance imaging in this study was as high as 89%.13 In all the previously mentioned studies, most fractures occurred within 2 years after completion of the treatment.

Since 2005, the Division of Radiation Oncology at Washington University School of Medicine in St Louis has been utilizing IMRT as the modality of choice for the treatment of cervical carcinomas requiring RT.14 To our knowledge, there are no reports describing posttreatment pathology of the pelvic girdle after IMRT treatment, utilizing split-field technique, in the setting of curative primary treatment of locoregionally advanced cervical carcinoma. In this study, we compared the prevalence of pelvic fractures, as well as other pelvic girdle pathology (osteonecrosis, osteomyelitis), after completion of intensity-modulated pelvic radiation versus conventional RT based. The diagnosis of pelvic fractures was established by pelvic computed tomography (CT) scans, a highly sensitive imaging modality for the diagnosis of pelvic fractures.15

Back to Top | Article Outline


After obtaining approval from the Washington University Medical Center Human Research Protection Office, a cohort of 500 consecutive cases of patients with stages IB1 to IVB cervical cancer were identified by International Classification of Diseases, Ninth Revision billing codes and treatment modalities from the Division of Radiation Oncology’s institutional database. Patients received intensity-modulated pelvic RT or conventional pelvic RT for the treatment of primary cervical carcinoma between the years of 1998 and 2009 at the Division of Radiation Oncology, Washington University School of Medicine.

To be included in the study, at least 1 pretreatment CT scan and at least 1 posttreatment CT scan had to be available for imaging review. Centralized imaging review was conducted by a board-certified musculoskeletal radiologist (T.J.H.). Preradiation and postradiation positron emission tomography (PET)/CT scans were obtained as part of the surveillance regimen after chemoradiation treatment. Women were asked to present for follow-up PET/CT scans at 3, 6, 12, 24, 36, 48, and 60 months after treatment. However, the timing of available posttreatment CT scans was variable in both patient groups.

Sacral fracture was defined as a vertically oriented linear lucency or sclerosis with or without cortical discontinuity just medial to the sacroiliac joints in the typical locations of a sacral insufficiency fracture. Pubic ramus fracture was defined as a linear lucency with callous formation consistent with subacute fracture. Osteonecrosis was defined as a geographic area of sclerosis in the sacrum that did not meet criteria for fracture. The patients with osteomyelitis had either a large ulcer extending to the ischium with associated productive bone formation and/or erosion typical of acute on chronic osteomyelitis. Medical records of the patients found to have posttreatment pelvic fractures, osteonecrosis, or osteomyelitis were reviewed for presence of symptoms around the time of the complication.

Women younger than 18 years or with a diagnosis of concurrent pregnancy were excluded. Patients treated in the adjuvant setting after hysterectomy were excluded. Patients who received pelvic IMRT for the treatment of malignancies other than cervical carcinoma had concurrent prolonged use of oral steroid therapy (≥3 months) at the time of RT or discontinuation of prolonged oral steroid therapy 6 months or less before initiation of RT, as well as patients with active parathyroid disease or with bone metastases from any type of cancer, were also excluded.

Once patients matching the inclusion criteria were identified, their charts curated for demographics, tumor histology and stage, smoking history, weight/body mass index (BMI), age, prolonged (≥3 months) use of oral steroid therapy, parathyroid disease, presence of bone metastases, use of bisphosphonates, calcitonin, parathyroid hormone, and duration of follow-up were abstracted. We also abstracted the data on the type of RT received, as well as use of concurrent chemotherapy.

Conventional RT was delivered according to the midline step-wedge central blocking technique developed at Mallinckrodt Institute of Radiology and used for more than 30 years. External irradiation target doses of 45 to 50 Gy are prescribed to the “whole pelvis,” that is, cervix, lymph nodes, bladder, and rectum. Midline blocking was used after 20 Gy. The goal of the midline block technique is to spare the bladder and rectum, while permitting increased doses to the cervix from the intracavitary brachytherapy. Patients receive intracavitary brachytherapy treatments to deliver a total dose of more than 85 Gy to point A. According to this protocol, the majority of the prescribed total dose equivalent dose of 85 Gy to point A is delivered with brachytherapy.

Intensity-modulated radiation therapy technique has been developed over the last decade to replicate our step-wedge technique and is referred to as pseudo–step-wedge intensity modulation. The goals of this technique are to limit bladder, rectal, and bone marrow doses and to maximize point A and pelvic lymph node doses. The dose to the pelvic bones was kept under 45 Gy.14,16

Preradiation and postradiation PET/CT scans were obtained as part of usual care accompanying chemoradiation administration. Available scans were re-reviewed for presence of posttreatment pelvic girdle bony pathology. The goal of the surveillance protocol was to obtain follow-up PET/CT scans at set intervals mentioned previously. However, despite the intent to have imaging available for review at regular intervals, the CT scans available for review were highly variable in posttreatment timing in both patient groups.

Back to Top | Article Outline
Statistical Analysis

After data were collected, we performed a propensity score–matching procedure: case-control pairs were matched up (1 IMRT patient and 1 conventional RT patient) based on age, stage, race, tumor histology, BMI, and smoking history. The propensity score–matched data were analyzed by a logistic regression analysis examining the effects of treatment type received (conventional RT vs IMRT) on the occurrence of posttreatment pelvic bony structure complications, while adjusting for the previously mentioned factors. SAS software, version 9.2 (SAS Institute, Cary, NC), was utilized to carry out the analyses. A Kaplan-Meier estimate of the survivor function was utilized to calculate the proportion of patients per group who experienced pelvic fractures as a function of time in both the IMRT and conventional RT cohorts. In addition, the demographic data (mean age, race, smoking, average BMI, stage and histology distribution, and treatment received) were tested for similarity with the appropriate statistical test (unpaired t, χ2, or Fisher exact test) utilizing GraphPad Prism software (San Diego, CA).

Back to Top | Article Outline


From the initial cohort, 183 eligible patients were identified. Only 166 patients, or 83 pairs, met the propensity score–matching procedure cutoff.

Demographic characteristics of the 2 groups are shown in Table 1 and are similar between the 2 groups. The mean age at presentation was approximately 50 years for both groups. The majority of patients were of white descent in both groups: 71% in the IMRT group and 61% in the conventional RT group. Approximately 50% of patients in both groups were former or current smokers; average BMIs of the 2 cohorts were on the border of obesity: 29.4 kg/m2 in the IMRT group and 30.4 kg/m2 in the conventional RT group (Table 1).

Most patients had locally advanced disease in both groups, and overall stage distributions were similar (Table 2). Only 4 (5%) of 83 in the IMRT group and 3 (4%) of 83 patients in the conventional RT group had stage IV disease. In terms of treatment modalities, 82 patients (99%) received chemotherapy concurrently with radiation in the IMRT group; similarly, 81 patients (96%) received chemotherapy along with radiation in the conventional RT group. Median follow-up for the 2 groups was as follows: 27 months for the IMRT group and 31 months for the conventional RT group (Table 2).

There were 3 pelvic girdle complications (4% of subjects) were identified in the IMRT group, and all 3 were bilateral sacral fractures. All 3 of those patients were symptomatic with back or hip pain (Table 3A). In the conventional RT group, there were 14 complications (17% subjects). The overall race distribution was statistically different between the 2 groups, as analyzed by the χ2 test (Table 1). Subsequently, the difference in race distribution was controlled for in the logistic regression analysis.

All patients in the IMRT group with bilateral sacral fractures experienced low-back or hip pain necessitating narcotic analgesics. Eleven of 13 patients in the conventional RT group (with available follow-up clinic notes) experienced low-back/abdominal/leg pain (Table 3B). Specifically, there were 9 cases of pelvic fractures, 2 cases of osteonecrosis, and 3 cases of osteomyelitis. There were 3 patients with right sacral fractures, 1 patient with insufficiency fractures of the right sacrum and right pubic ramus, 3 left sacral fractures, and 2 bilateral sacral fractures. The odds ratio for developing a post-RT complication was 4.49 when comparing the conventional and intensity-modulated therapy treatment groups, with P = 0.01 (95% confidence interval [CI], 1.4–14.1). The fraction of patients remaining fracture-free at a given length of follow-up after pretreatment CT scan (in months) is illustrated in Figure 1.

Median follow-up for the subjects with complications was 51 months in the IMRT group, and 46 months in the conventional RT group. Median time to fracture detection by CT scans was 19 months in the IMRT group, and 29 months in the conventional RT group (Table 3C). Of note, there were also 4 new cases of osteoporosis in the conventional RT group that developed after chemoradiation and were not present upon pretreatment CT examination; these were not included as pelvic girdle complications in the current analysis as complications.

Back to Top | Article Outline


In this study, we noted a significant decrease in posttreatment pelvic girdle complications in patients treated with IMRT, compared with the group treated with conventional RT in the setting of advanced cervical cancer. Our study evaluated women receiving definitive chemoradiation treatment, without preceding (radical) hysterectomy; as such, this report is one of the first to characterize pelvic girdle complications in a group of patients undergoing definitive IMRT predominately for treatment of locoregionally advanced cervical cancer.

A number of investigations in the last several years, both prospective and retrospective, have examined the outcomes of IMRT administered postoperatively. Shih et al17,18 and Folkert et al19 recently published a series of reports describing the outcomes of postoperative IMRT in patients with cervical and endometrial cancers treated at Memorial Sloan-Kettering Cancer Center. Of 222 patients receiving postoperative pelvic RT, 11 (5%) developed pelvic insufficiency fractures at a median time of 11.5 months from RT completion. The rate of pelvic insufficiency fractures was similar whether patients received IMRT or conventional RT.17 While the results are seemingly discordant with our findings, there are a number of differences between our study and that conducted by Shih and colleagues precluding direct comparison. In the study population at Memorial Sloan-Kettering Cancer Center, all patients were treated postoperatively (none in the definitive setting); only 35% of patients had the diagnosis of cervical cancer, and only 36.5% of all patients were treated with IMRT. There was no matching procedure for the conventional and IMRT cases. Furthermore, not all patients, (64% and 79% in the conventional and IMRT groups, respectively) were evaluated with posttreatment CT scans. Therefore, outcomes described by Shih et al cannot be meaningfully compared with our data, as different patient populations are described.

The pelvic fracture rate detected after conventional RT treatment in our report is similar to those described in the earlier literature.6,8,9,12 One of the largest studies6 published in 2010 reported on a cohort of 300 patients receiving RT. The vast majority of patients included in the study of Schmeler et al6 received conventional RT (AP/PA fields), with only 3% of patients receiving IMRT; 86% of patients received radiation in the primary setting, and approximately 12% received postoperative radiation. The pelvic fracture rate reported by Schmeler et al was 9% versus 11% observed in the current study (conventional RT group). The fracture sites predilection observed in the current study was also similar to those described in the literature. All patients in our cohort with pelvic fractures had a sacral component (100%), whereas in Schmeler and colleagues’ study 93% of patients had a sacral fracture component.

In this investigation, the inclusion criteria were limited to the group of patients receiving IMRT along with concurrent chemotherapy, as primary treatment for locoregionally advanced and metastatic cervical cancer in order to further eliminate confounding factors, such as preoperative versus postoperative treatment setting, IMRT and conventional RT treatment planning techniques, and post-RT imaging modalities.

Some of the disadvantages of this retrospective study include small sample size, single institutional setting, and the inconsistency of the timing and number of follow-up CT studies (some of the patients had as few as 2 follow-up scans available for review, and some of the patients had 6 scans available for review). This may have impaired the accuracy of determining the timing of fracture occurrence. Ascertainment bias might have impacted evidence of asymptomatic disease, although use of CT scan for disease monitoring has increased with time and should have resulted in finding more fractures in IMRT patients. Therefore, detailed reporting on the timing of pelvic fractures/osteomyelitis/osteonecrosis after RT treatment is beyond the capacity of this study.

The general applicability of our conclusions may be limited because of variations in treatment planning strategies implemented across the medical centers. At Washington University, cervical cancer patients were treated conventionally with AP/PA fields, as opposed to a 4-field box technique. Currently, treatment planning is based on split-field technique, with the AP/PA treatment concept adapted to IMRT, and central pelvic dose is largely driven by brachytherapy dosing.14 Further investigations is necessary to ascertain whether our findings are applicable to the patients with cervical cancer definitively treated with 4-field-box conventional RT technique versus 4-field box–derived IMRT and reduced-dose brachytherapy.

Our investigation identified another potential advantage of a now widely utilized RT modality. Although beyond the scope of this study, it has previously been shown by Kidd and colleagues20 from Washington University in St Louis that patients’ posttreatment PET/CT findings were not significantly different between the IMRT and non-IMRT patients. Furthermore, in the cohort described by Kidd et al, the IMRT and conventional RT groups had similar overall and cause-specific survivals, with a trend toward improved survival in the IMRT group.20

As extended cancer survivorship becomes a reality, improving quality of life (QOL) of patients in lasting remission comes to the forefront of long-term treatment goals. Several studies conducted in the last 13 years have measured the QOL of long-term cervical cancer survivors.21–25 A recent study by Le Borgne et al23 demonstrated that cervical cancer survivors endorsed globally similar good QOL, as compared with control subjects. The study also suggested that compared with cervical cancer survivors treated by surgery alone, QOL of those who received radiotherapy was significantly more affected in terms of cervical cancer–specific problems, such as sexual dysfunction, voiding and abdominal symptoms, and lymphedema.23 Our study describes another possible advantage of IMRT that may improve long-term functioning of cervical cancer survivors—decrease in pelvic girdle complications, which are likely to decrease pain levels and chronic morbidity and thus contribute to a greater psychosocial and sexual function; the vast majority of the patients in our study who experienced pelvic girdle complications post-RT were symptomatic with pain that became chronic or required the use of narcotic medications. Earlier intervention by instituting better symptom screening, bone scans, or pharmacologic prophylaxis can perhaps contribute to QOL and fracture prevention. Also, more widespread use of IMRT may lead to decrease in the overall number complications.

Future studies are needed to identify the patients at risk and to examine the patterns of post-RT pelvic girdle damage (conventional RT vs IMRT). Table 3A illustrates the difference between fracture site predilection between the conventional RT and IMRT groups. However, at this point, it is unclear whether the fracture site difference occurred by chance or is due to a true pattern. The fracture/complication pattern may also vary by institutions, as different RT treatment protocols are utilized. Identifying bony sites that are at highest risk, the postcomplication symptom pattern, and educating the clinicians to such may also aid with early symptom recognition and treatment. In our review, we noted that the vast majority of patients with complications were likely to suffer from a form of chronic bony pelvic, abdominal, or leg pain. As treatment modalities improve, and cancer survivorship lengthens, a larger body of investigations and reports are needed to draw conclusions and ideas for improving the QOL of the long-term survivors.

Back to Top | Article Outline


1. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin. 2011; 61: 69–90.
2. 2. American Cancer Society. Cancer Facts & Figures 2013. 2013.
3. Peters WA, Liu PY, Barrett RJ, et al. Concurrent chemotherapy and pelvic radiation therapy compared with pelvic radiation therapy alone as adjuvant therapy after radical surgery in high-risk early-stage cancer of the cervix. J Clin Oncol. 2000; 18: 1606–1613.
4. Reducing uncertainties about the effects of chemoradiotherapy for cervical cancer: individual patient data meta-analysis. Cochrane Database Syst Rev. 2010: CD008285.
5. Grigsby PW, Roberts HL, Perez CA. Femoral neck fracture following groin irradiation. Int J Radiat Oncol Biol Phys. 1995; 32: 63–67.
6. Schmeler KM, Jhingran A, Iyer RB, et al. Pelvic fractures after radiotherapy for cervical cancer: implications for survivors. Cancer. 2010; 116: 625–630.
7. Baxter NN, Habermann EB, Tepper JE, et al. Risk of pelvic fractures in older women following pelvic irradiation. JAMA. 2005; 294: 2587–2593.
8. Huh SJ, Kim B, Kang MK, et al. Pelvic insufficiency fracture after pelvic irradiation in uterine cervix cancer. Gynecol Oncol. 2002; 86: 264–268.
9. Ikushima H, Osaki K, Furutani S, et al. Pelvic bone complications following radiation therapy of gynecologic malignancies: clinical evaluation of radiation-induced pelvic insufficiency fractures. Gynecol Oncol. 2006; 103: 1100–1104.
10. Kwon JW, Huh SJ, Yoon YC, et al. Pelvic bone complications after radiation therapy of uterine cervical cancer: evaluation with MRI. AJR Am J Roentgenol. 2008; 191: 987–994.
11. Ogino I, Okamoto N, Ono Y, et al. Pelvic insufficiency fractures in postmenopausal woman with advanced cervical cancer treated by radiotherapy. Radiother Oncol. 2003; 68: 61–67.
12. Oh D, Huh SJ, Nam H, et al. Pelvic insufficiency fracture after pelvic radiotherapy for cervical cancer: analysis of risk factors. Int J Radiat Oncol Biol Phys. 2008; 70: 1183–1188.
13. Blomlie V, Rofstad EK, Talle K, et al. Incidence of radiation-induced insufficiency fractures of the female pelvis: evaluation with MR imaging. AJR Am J Roentgenol. 1996; 167: 1205–1210.
14. Macdonald DM, Lin LL, Biehl K, et al. Combined intensity-modulated radiation therapy and brachytherapy in the treatment of cervical cancer. Int J Radiat Oncol Biol Phys. 2008; 71: 618–624.
15. Fu CY, Wu SC, Chen RJ, et al. Evaluation of pelvic fracture stability and the need for angioembolization: pelvic instabilities on plain film have an increased probability of requiring angioembolization. Am J Emerg Med. 2009; 27: 792–796.
16. Chang AJ, Richardson S, Grigsby PW, et al. Split-field helical tomotherapy with or without chemotherapy for definitive treatment of cervical cancer. Int J Radiat Oncol Biol Phys. 2012; 82: 263–269.
17. Shih KK, Folkert MR, Kollmeier MA, et al. Pelvic insufficiency fractures in patients with cervical and endometrial cancer treated with postoperative pelvic radiation. Gynecol Oncol. 2013; 128: 540–543.
18. Shih KK, Milgrom SA, Abu-Rustum NR, et al. Postoperative pelvic intensity-modulated radiotherapy in high risk endometrial cancer. Gynecol Oncol. 2013; 128: 535–539.
19. Folkert MR, Shih KK, Abu-Rustum NR, et al. Postoperative pelvic intensity-modulated radiotherapy and concurrent chemotherapy in intermediate- and high-risk cervical cancer. Gynecol Oncol. 2013; 128: 288–293.
20. Kidd EA, Siegel BA, Dehdashti F, et al. Clinical outcomes of definitive intensity-modulated radiation therapy with fluorodeoxyglucose–positron emission tomography simulation in patients with locally advanced cervical cancer. Int J Radiat Oncol Biol Phys. 2010; 77: 1085–1091.
21. Kim SH, Kang S, Kim YM, et al. Prevalence and predictors of anxiety and depression among cervical cancer survivors in Korea. Int J Gynecol Cancer. 2010; 20: 1017–1024.
22. Korfage IJ, Essink-Bot ML, Mols F, et al. Health-related quality of life in cervical cancer survivors: a population-based survey. Int J Radiat Oncol Biol Phys. 2009; 73: 1501–1509.
23. Le Borgne G, Mercier M, Woronoff AS, et al. Quality of life in long-term cervical cancer survivors: a population-based study. Gynecol Oncol. 2013; 129: 222–228.
24. Vistad I, Fossa SD, Dahl AA. A critical review of patient-rated quality of life studies of long-term survivors of cervical cancer. Gynecol Oncol. 2006; 102: 563–572.
25. Wenzel L, DeAlba I, Habbal R, et al. Quality of life in long-term cervical cancer survivors. Gynecol Oncol. 2005; 97: 310–317.

Cervical cancer; Intensity-modulated radiation therapy; Survivorship

© 2014 by the International Gynecologic Cancer Society and the European Society of Gynaecological Oncology.