- Email: [email protected]
External Beam Radiation Helps Prevent Heterotopic Bone Formation in Patients With a History of Heterotopic Ossification
The Journal of Arthroplasty Vol. 21 No. 5 2006
External Beam Radiation Helps Prevent Heterotopic Bone Formation in Patients With a History of Heterotopic Ossification Samuel T. Chao, MD,* Shih-Yuan Lee, MSPH,* Lester S. Borden, MD,y Michael J. Joyce, MD,y Viktor E. Krebs, MD,y and John H. Suh, MD*
Abstract: The aim of this study was to determine if radiation prevents heterotopic ossification (HO) in HO-forming patients after total hip arthroplasty (THA) or HO excision alone. Patients with HO in the ipsilateral hip (63 treated with THA revision and 25 treated with HO excision alone) and HO in the contralateral hip (36 treated with primary THA) were termed HO-forming patients. They underwent radiation to prevent HO. After excluding patients with inadequate follow-up, 84 patients were studied to determine if radiation prevents significant HO (Brooker Grade 3-4). For patients with ipsilateral hip HO, 12.3% developed significant HO. In patients with contralateral hip HO, 10.5% developed significant HO after THA. Sixty percent who received 6 Gy in 3 fractions after excision of ipsilateral HO developed significant HO, which was higher than for all dosefractionation schemes combined ( P = .01). In contrast, patients who received 7 Gy in 1 fraction developed significant HO 13.8% of the time, which was equivalent to all dose-fractionation schemes combined ( P = not significant). Radiation prevents HO in HO-forming patients. Key words: heterotopic ossification, radiation, hip arthroplasty, dose fractionation, prevention. n 2006 Elsevier Inc. All rights reserved.
patients [2]. Risk factors for heterotopic bone formation in the hip include a history of HO in the ipsilateral or contralateral hip, hypertrophic osteoarthritis, ankylosing spondylitis, diffuse idiopathic skeletal hyperostosis, Paget disease, posttraumatic arthritis, osteonecrosis, and rheumatoid arthritis [3]. Patients with a history of HO are considered at high risk for HO. Eighty percent of these patients form HO after surgery alone [4]. Radiation is effective in preventing HO. Typically, the radiation is delivered within 72 hours after surgery or 24 hours before surgery. However, studies showing the effectiveness of radiation in patients with a history of HO are limited. Therefore, the purpose of this study was to determine if external beam radiation can prevent significant HO in patients with a history of heterotopic bone formation, either in the same hip after HO resection
Heterotopic ossification (HO), bone formation in nonosseous tissues, usually occurs in the setting of trauma such as fractures and surgical procedures [1]. Heterotopic ossification of the hip is the most common complication of total hip arthroplasty (THA). It can occur in as many as 53% of patients who undergo THA and causes postsurgical disability from pain and limited range of motion in 7% of From the *Department of Radiation Oncology, The Cleveland Clinic Foundation, Cleveland, Ohio; and yDepartment of Orthopedics, The Cleveland Clinic Foundation, Cleveland, Ohio. Submitted January 25, 2005; accepted August 12, 2005. No benefits or funds were received in support of the study. Reprint requests: John H. Suh, MD, Department of Radiation Oncology, The Cleveland Clinic Foundation, 9500 Euclid Avenue, T-28, Cleveland, OH 44195. n 2006 Elsevier Inc. All rights reserved. 0883-5403/06/1906-0004$32.00/0 doi:10.1016/j.arth.2005.08.014
731
732 The Journal of Arthroplasty Vol. 21 No. 5 August 2006 or the contralateral hip. This study is a retrospective review of the experience at our institution.
Table 1. Dose Fractionation of Radiation Received by Patients Dose-fractionation scheme
Materials and Methods The records and radiographs of 195 patients who underwent hip surgery between August 31, 1982, and July 24, 2001, and postoperative external beam radiation therapy to prevent significant HO in the operated hip were reviewed. One hundred twenty-four (63.3%) of these patients had a history of HO, either in the ipsilateral hip or the contralateral hip, and served as the basis for this study. Eighty-eight of these patients had HO in the ipsilateral hip and were treated with revision of their THA (63 patients) or excision of their HO alone (25 patients). The remaining 36 of 124 patients had a history of HO in the hip opposite to that undergoing THA. Because these patients are also at high risk for HO, they were treated with postoperative radiation to prevent HO. Patients were excluded from analysis if they did not have both preoperative and follow-up Brooker Grades [5], and had not been followed for at least 6 months. The latter step was taken to ensure that the HO had enough time to mature after surgery. Of a possible 124 patients, 84 (68%) patients met our inclusion criteria: 65 (77%) had HO in the same hip, and 19 (23%) had HO in the contralateral hip; the remaining 40 patients not meeting our inclusion criteria had incomplete records or insufficient follow-up at the time of the analysis. The median age was 64 years (range, 22-83 years). These 84 patients were treated with one of the following radiation doses: 20 Gy in 10 fractions, 20 Gy in 5, 10 Gy in 5, 8 Gy in 1, 7 Gy in 1, 6 Gy in 3, 6 Gy in 2, or 6 Gy in 1. Table 1 lists the number of patients receiving each fractionation scheme from 1982 to 2001. The more protracted courses of radiation were used in the 1980s and early 1990s. Single fraction therapy was used starting in September 1990 after studies demonstrated its effectiveness. One patient received preoperative radiation. All but 8 (90%) patients received radiation within 3 days after surgery. The median follow-up time was 20.5 months (range, 6.2 to 147.0 months). The preoperative and follow-up Brooker Grades were compared to determine if radiation was effective in preventing significant HO, which was defined as Brooker Grade 3 or 4. In cases where a Brooker Grade was not reported by the radiologist or orthopedic surgeon, either preoperatively or in follow-up, radiographs were obtained by the investigators to determine the Brooker Grade.
No. of patients
7 Gy in 1 fraction 6 Gy in 1 fraction 10 Gy in 5 fractions 6 Gy in 3 fractions 20 Gy in 10 fractions 20 Gy in 5 fractions 8 Gy in 1 fraction 6 Gy in 2 fractions
38 20 11 8 4 1 1 1
Fisher exact tests were used to compare the development of significant HO between the various groups. A P value of .05 or less was considered statistically significant.
Results Ipsilateral vs Contralateral Hip HO The 65 patients with a history of HO on the same side had preoperative Brooker Grades ranging from 0 to 4, with 55 hips (85%) having preoperative Brooker Grades of 3 or 4. After surgery and radiation, 14 developed no HO, 21 developed Brooker Grade 1 HO, 22 developed grade 2 HO, 4 developed grade 3 HO, and 4 developed grade 4 HO after radiation. As such, 8 of 65 (12.3%) developed significant HO. All 19 patients with HO on the contralateral side had a preoperative Brooker Grade of 0. After surgery and radiation, 10 developed no HO, 3 developed Brooker Grade 1 HO, 4 developed grade 2 HO, and 2 developed grade 3 HO after radiation. Thus, only 2 (10.5%) of 19 developed significant HO. There was no difference in the development of HO in the operated hip if the same hip or the contralateral hip was initially affected with HO ( P = not significant [NS]). Table 2 summarizes these results. Dose Fractionation (6 Gy in 3 Fractions and Standard of Care, 7 Gy in 1 Fraction) Five patients received 6 Gy in 3 fractions after excision of their heterotopic bone in the same hip (1 with preoperative Brooker Grade 2 HO, 2 with grade 3 HO, and 2 with grade 4 HO). Two developed Brooker Grade 2 HO, and 3 developed grade 4 HO in follow-up. Of 5 patients, 3 (60%) developed significant HO. Thus, 6 Gy in 3 fractions was not effective in preventing HO when compared with the other dose-fractionation schemes as group ( P = .01).
Radiation Prevents Heterotopic Bone ! Chao et al 733 Table 2. Brooker Grade After Radiation
Ipsilateral hip HO (n = 65) Contralateral hip HO (n = 19) P
No HO
Grade 1
Grade 2
Grade 3
Grade 4
Rate of significant HO
14 (21.5) 10 (52.6)
21 (32.3) 3 (15.8)
22 (33.8) 4 (21.0)
4 (6.15) 2 (10.5)
4 (6.15) 0 (0)
12.3 10.5 NS
Values are presented as n (%) or percentages.
By contrast, 29 patients received 7 Gy in 1 fraction after excision of their heterotopic bone in the same hip. This dose-fractionation scheme was found to be effective by other studies and is the dose fractionation currently being used at our institution. Three of these patients developed Brooker Grade 3, and 1 developed grade 4 HO on follow-up. Thus, 4 (13.8%) of 29 treated with 7 Gy in 1 fraction developed significant HO. When compared with the other dose-fractionation schemes as a group, there was no significant difference in the rate of significant heterotopic bone formation ( P = NS). Table 3 summarizes these results. Complications Complications included postoperative infection (n = 3), all of which occurred in patients who underwent excision of HO in the ipsilateral hip. No patients developed symptomatic loosening of their prosthesis after removal of HO. One patient received an ingrowth stem and cup after HO removal, followed by radiation to 10 Gy in 5 fractions. During the radiation, the ingrowth components were blocked. He developed minor loosening documented in follow-up 2.5 years after x-ray but had good hip function with a Harris score of 88. Because of lack of symptoms, he never required reoperation. The second patient initially had right hip arthroplasty with a cementless acetabular component and cemented femoral component. He subsequently developed HO and had resection of his HO followed by radiation to 7 Gy in 1 fraction. Follow-up x-ray showed some evidence of loosen-
ing about the methacrylate-bone interface in the femur 7 months later and thigh pain. He had an aspiration arthrogram that was negative. Instillation of contrast from the arthrogram was negative for evidence of loosening. He subsequently developed other medical issues and was lost to further follow-up. One patient developed trochanteric bursitis in follow-up. This patient had HO in the ipsilateral hip and underwent revision with ingrowth stem and cup, followed by radiation to 6 Gy in 3 fractions. Other than trochanteric bursitis, the patient continued to have good range of motion and a Harris score of 78. In another patient, the acetabular component had migrated, but she had a history of rectal cancer treated with radiation. This patient initially had a THA followed by 2 other surgeries for dislocations. She was found to have Brooker Grade 3 HO in that hip and underwent repeat surgery for dislocation thought to be related to her HO. She had removal of HO and arthroplasty of the acetabular component, which was affixed with screws. Seven Gy in 1 fraction was delivered for prophylaxis. Follow-up approximately 1 year later showed migration of the acetabular component, with development of pain in the buttock. This migration and failure of fixation were attributed to her previous radiation for rectal cancer 14 years prior. The patient subsequently underwent revision with a Depuy acetabular cage, bone graft, and arthroplasty of the femoral component with subsequent improvement in her function. There were no known healing complications associated with radiation in this study.
Table 3. 6 Gy in 3 Fractions and 7 Gy in 1 Fraction Compared With Other Dose Fractionations as a Group
6 Gy in 3 fractions (n = 5) 7 Gy in 1 fraction (n = 29)
Grade 3
Grade 4
Rate of significant HO
P value when compared with all other fractionations combined
0 (0) 3 (10.4)
3 (60) 1 (3.4)
60 13.8
.01 NS
Values are presented as n (%) or percentages.
734 The Journal of Arthroplasty Vol. 21 No. 5 August 2006
Discussion Patients with a history of HO are at high risk for recurrence after resection. Given that pluripotential mesenchymal cells are responsible for the development of HO [6], it may be possible to stop the ossification by radiating these cells. Fractionated radiation (20 Gy in 10 fractions) was used and reported in 1981 [7]. In this landmark study, patients were classified into various groups; patients in groups 1, 3, 4, and 5 had a history of HO. Group 1 consisted of 12 hips where HO was removed. None of these hips developed severe or massive HO after excision and radiation. Group 3 consisted of 14 hips with a history of HO in that hip; a THA was performed as opposed to excision alone. Of these 14 hips, 3 (21%) had severe HO, but the postoperative function was better than that before surgery and radiation. Group 4 consisted of 7 patients who had a history of HO on the contralateral side. None of these patients developed severe or massive HO after surgery and radiation. Group 5 was composed of patients who had preexisting mild-to-moderate HO that was treated with radiation alone with no excision. Their HO progressed after radiation. Understandably, radiation alone has no effect on preexisting HO because the cells have already differentiated and have formed bone [8]. Since then, various fractionation schemes have been studied. A single fraction of radiation to 7 Gy was studied and found to be effective in preventing HO as well [9]. In this series of 25 patients (26 hips), 6 patients underwent hip revision and excision of HO, 4 underwent excision of preexisting HO, and 6 had a history of HO on the contralateral side. None of the patients had any functional impairment, and none of the patients developed Brooker Grade 3 or 4 HO. In another study, a single fraction of 5.5 Gy was compared with a single fraction of 7 Gy, retrospectively [10]. Sixty-three percent of the hips radiated with 5.5 Gy developed HO, with 2 of 19 being symptomatic. In comparison, only 10% of the patients in the 7 Gy group developed HO, and none were symptomatic. This study also categorized patients with a history of HO. In patients with ipsilateral HO, 2 of 5 treated with 5.5 Gy developed significant HO defined as Brooker Grade 3 and 4 vs 0 of 23 in the 7 Gy group. In patients with contralateral HO, 1 of 3 developed significant HO with 5.5 Gy vs 0 of 20 with 7 Gy. Therefore, 5.5 Gy was concluded to be ineffective in preventing HO, especially in patients with a history of HO. A dose of 6 Gy in a single fraction was also studied and found to be effective, with
none of the 17 patients in the series developing Brooker Grade 3 or 4 HO [11]. A randomized controlled trial was performed in Germany comparing preoperative radiation vs postoperative radiation for the prevention of HO [12]. Patients were randomized to receive 7 Gy of radiation within 4 hours before hip surgery or 17.5 Gy in 5 fractions less than 96 hours postsurgery. Five percent developed failures in the postoperative radiation arm vs 19% in the preoperative radiation arm ( P b .05). In subgroup analysis of patients without grade 3 or 4 HO before surgery, there was no difference between the preoperative and postoperative radiation groups. However, in patients who had radiation before the removal of their ipsilateral grade 3 or 4 HO, there was a failure rate of 39%, which was statistically significant ( P b .001) when compared with the postoperative arm, which failed only 9% of the time. As such, preoperative radiation was determined to be not as effective when compared with postoperative radiation for these high-risk patients. A recent study studied the optimal timing of preoperative radiation to prevent HO using a rabbit model [13]. Radiation was given at 4 different time points: 4 hours, 24 hours, 72 hours, and 3 weeks. Radiation given 24 hours before surgery was more effective than the other time points in preventing HO. It is speculated that cell cycling is necessary for the pluripotent mesenchymal cells to die. Twentyfour hours allows enough time for cell cycling with resulting cell death, whereas 4 hours is insufficient. This may explain why patients in the German study with a history of HO in the ipsilateral hip who received preoperative radiation do not do as well compared with postoperative radiation when the preoperative radiation was given only 4 hours before surgery. In our current retrospective study, the role of radiation in patients with a history of HO, either in the same hip or the contralateral hip, was assessed given that these patients represent the highest-risk group. In patients with a history of HO in the treated hip, the rate of HO after radiation was 12.3%. In patients with a history of HO in the hip contralateral to the treated hip, the rate of HO was 10.5%. There was no difference in the rate of HO when this was initially present in the same hip or the opposite hip ( P = NS). These results compare favorably with other series such as the German study mentioned previously despite these patients being at high risk for the development of HO. A variety of fractionation schemes have been used at our institution since 1982. This includes 20 Gy in 10 fractions, 20 Gy in 5, 10 Gy in 5, 8 Gy
Radiation Prevents Heterotopic Bone ! Chao et al 735
in 1, 7 Gy in 1, 6 Gy in 3, 6 Gy in 2, and 6 Gy in 1. Five patients in this study were treated using 6 Gy in 3 fractions. In these patients, 60% developed significant HO. When compared with the other fractionations, this fractionation scheme was not effective ( P = .01) in preventing HO. Given that this represents less dose biologically compared with 6 Gy in a single fraction, the lack of efficacy with this dose fractionation was not surprising. Fractionating therapy allows for repopulation of the pleuripotential mesenchymal cells between fractions. Thus, to achieve the same control rates, it is necessary to increase the total dose during fractionating therapy. Conversely, when decreasing the number of fractions, a larger dose per fraction is required. This is why 10 Gy in 5 fractions is equivalent to 7 Gy in 1 fraction but 6 Gy in 3 fractions is not equivalent to 6 Gy in 1 fraction. Seven Gy in 1 fraction has become a popular fractionation scheme because it requires only 1 day of treatment. Given that these patients are often in pain and immobile after surgery, the convenience of 1-day treatment is beneficial for both the patient and the staff. This dose fractionation is currently used in our department with 29 (45%) of 65 patients treated in this fashion in this series. Only 13.8% of the patients with HO in the same hip developed significant HO after radiation. When compared with the other dose fractionations, there was no difference in effectiveness. These results are comparable to published literature showing that a single fraction of 7 Gy is equally effective to fractionated doses of 10 Gy, which was shown to be as effective as fractionated doses of 20 Gy [9,14-17]. In this study, complications were rare. No patient had symptomatic loosening of their femoral components. One patient had shielding of the ingrowth femoral stem during radiation but, despite this, developed x-ray evidence of loosening. However, he continued to have good function in that hip and never required reoperation. Another patient only had excision of HO and no revision of his cemented THA. While on x-ray, he was felt to have some loosening; he was found not to have loosening based on the results of arthrogram. One patient had migration of the acetabular component, but this was mainly attributed to her history of pelvic radiation for rectal cancer because she was having problems with dislocations before receiving radiation for HO prophylaxis. She also developed the migration despite the use of fixation screws. Her acetabular component required eventual arthroplasty. In general, it is recommended that ingrowth components be shielded because radiation may
prevent bone ingrowth like it prevents HO bone growth. A study placed porous components into rabbit tibias and radiated some of these tibias to 10 Gy in 5 fractions [18]. The rabbits were killed. At 2 weeks, the force necessary to remove the component after radiation was significantly less than the force to remove the component from the untreated tibia. However, this difference was not statistically significant at 3 weeks and beyond. Despite this, the previously mentioned German study failed to show increased rates of loosening or instability in patients after radiation of their porous components without shielding [12]. Removal of HO by itself is not known to cause loosening. Currently, drug therapy with nonsteroidal antiinflammatory drugs (NSAIDs) such as indomethacin and cyclooxygenase-2 inhibitors is being used as an alternative to radiation for prophylaxis with good success [19-21]. Given these results, a recent meta-analysis was conducted using studies that compared radiation and NSAID therapy in the prevention of HO [22]. These studies included a variety of patients and were not limited to high-risk patients. This meta-analysis showed decreased risk for clinically significant HO (defined as Brooker Grade 3 or 4) with radiation vs NSAID. The study also suggests that higher doses beyond 6 Gy in 1 fraction are more effective, whereas a dose of 6 Gy is equally effective as NSAID therapy. Further investigation may be necessary in determining if NSAID therapy is effective in high-risk patients and if the decrease in risk with radiation also holds for high-risk patients when compared with NSAID therapy. If a patient is on NSAID therapy 72 hours after surgery and develops gastrointestinal complications or renal dysfunction, or require anticoagulation with heparin or Coumadin, preventing further use of NSAID, there is no longer any treatment that is effective. This would put the patient at high risk for redeveloping HO.
Conclusion The morbidity from HO may be severe with pain and limitation in the range of motion. In patients at risk for HO, prophylaxis with radiation or NSAID after hip surgery is often recommended to prevent HO. The highest-risk patients are those with a history of HO, either in the same hip or the contralateral hip. This study, as well as other studies, has confirmed that prophylaxis with radiation is effective in preventing HO formation in high-risk patients. Seven Gy in 1 fraction postoperatively has been shown to be as effective as fractionated
736 The Journal of Arthroplasty Vol. 21 No. 5 August 2006 therapy to 10 or 20 Gy and is the scheme being used at our institution. Preoperative radiation may be more convenient given the immobility and pain after surgery, which can make postoperative radiation difficult. Despite a study that showed that preoperative radiation is ineffective in patients with a history of HO, timing may play a role in the effectiveness of preoperative radiation. Clinical studies regarding the timing of preoperative radiation are indicated, especially in patients with a history of HO. Also, studies regarding NSAID therapy included a variety of patients and were not limited to patients with a history of HO. More studies are necessary to determine if NSAID therapy is as effective as radiation in this subgroup of high-risk patients.
11.
12.
13.
14.
References 15. 1. Kaplan FS, Glaser DL, Hebla N, et al. Heterotopic ossification. J Am Acad Orthop Surg 2004;12:116. 2. Thomas BJ, Amstutz HC. Prevention of heterotopic bone formation. Hip 1987;59. 3. Lo TCM. Radiation therapy for heterotopic ossification. Semin Radiat Oncol 1999;9:163. 4. Ayers DC, Pellegrini VD, Evarts CM. Prevention of heterotopic ossification in high-risk patients by radiation therapy. Clin Orthop 1991;263:87. 5. Brooker AF, Bowerman JW, Robinson RA, et al. Ectopic ossification following total hip replacement. J Bone Joint Surg 1973;55-A:1629. 6. Jowsey J, Coventry MB, Robins PR. Heterotopic ossification: theoretical consideration, possible etiologic factors, and a clinical review of total hip arthroplasty patients exhibiting this phenomenon, in the hip. In: The Hip Society, editor. In proceedings of the fifth open scientific meeting of the Hip Society. St Louis: CV Mosby; 1977. p. 210. 7. Coventry MB, Scanlon PW. The use of radiation to discourage ectopic bone: a nine-year study in surgery about the hip. J Bone Joint Surg Am 1981; 63:201. 8. Tonna EA, Cronkite EP. Auto radiographic studies of cell proliferation in the periosteum of intact and fractured femora of mice utilizing DNA labeling with ( 3H) thymidine. Proc Soc Exp Biol Med 1961;107:719. 9. Lo TC, Healy WL, Covall DJ, et al. Heterotopic bone formation after hip surgery: prevention with singledose postoperative hip irradiation. Radiology 1998;168:851. 10. Healy WL, Lo TC, DeSimone AA, et al. Single-dose irradiation for the prevention of heterotopic ossifica-
16.
17.
18.
19.
20.
21.
22.
tion after total hip arthroplasty. J Bone Joint Surg 1995;77-A:590. Hedley AK, Mead LP, Hendren DH. The prevention of heterotopic bone formation following total hip arthroplasty using 600 rad in a single dose. J Arthroplasty 1989;4:319. Seegenschmiedt MH, Keilholz L, Martus P, et al. Prevention of heterotopic ossification about the hip: final results of two randomized trials in 410 patients using either preoperative or postoperative radiation therapy. Int J Radiat Oncol Biol Phys 1997;39:161. Rumi MN, Deol GS, Bergandi JA, et al. Optimal timing of preoperative radiation for prophylaxis against heterotopic ossification: a rabbit hip model. J Bone Joint Surg Am 2005;87-A:366. Anthony P, Keys H, Evarts CM, et al. Prevention of heterotopic bone formation with early post operative irradiation in high risk patients undergoing total hip arthroplasty: comparison of 10.00 Gy vs 20.00 Gy schedules. Int J Radiat Oncol Biol Phys 1987;13:365. Sylvester JE, Greenberg P, Selch MT, et al. The use of postoperative irradiation for the prevention of heterotopic bone formation after total hip replacement. Int J Radiat Oncol Biol Phys 1988;14:471. Ayers DC, Evarts CM, Parkinson JR. The prevention of heterotopic ossification in high-risk patients after total hip arthroplasty. J Bone Joint Surg Am 1986; 68:1423. Pellegrini VD, Konski AA, Gastel JA, et al. Prevention of heterotopic ossification with irradiation after total hip arthroplasty. J Bone Joint Surg 1992;74-A:186. Konski A, Weiss C, Rosier R, et al. The use of postoperative irradiation for the prevention of heterotopic bone after total hip replacement with biologic fixation (porous coated) prosthesis: an animal model. Int J Radiat Oncol Biol Phys 1990; 18:861. Kienapfel H, Koller M, Wust A, et al. Prevention of heterotopic bone formation after total hip arthroplasty: a prospective randomized study comparing postoperative radiation therapy with indomethacin medication. Arch Orthop Trauma Surg 1999;119:296. Sell S, Willms R, Jany R, et al. The suppression of heterotopic ossification: radiation versus NSAID therapy — a prospective study. J Arthroplasty 1998; 13:854. Moore KD, Goss K, Anglen JO. Indomethacin versus radiation therapy for prophylaxis against heterotopic ossification in acetabular fractures—a randomized, prospective study. J Bone Joint Surg 1998;80-B:259. Pakos EE, Ioannidis JPA. Radiotherapy versus nonsteroidal anti-inflammatory drugs for the prevention of heterotopic ossification after major hip procedures: a meta-analysis of randomized trials. Int J Radiat Oncol Biol Phys 2004;60:888.
External Beam Radiation Helps Prevent Heterotopic Bone Formation in Patients With a History of Heterotopic Ossification Samuel T. Chao, MD,* Shih-Yuan Lee, MSPH,* Lester S. Borden, MD,y Michael J. Joyce, MD,y Viktor E. Krebs, MD,y and John H. Suh, MD*
Abstract: The aim of this study was to determine if radiation prevents heterotopic ossification (HO) in HO-forming patients after total hip arthroplasty (THA) or HO excision alone. Patients with HO in the ipsilateral hip (63 treated with THA revision and 25 treated with HO excision alone) and HO in the contralateral hip (36 treated with primary THA) were termed HO-forming patients. They underwent radiation to prevent HO. After excluding patients with inadequate follow-up, 84 patients were studied to determine if radiation prevents significant HO (Brooker Grade 3-4). For patients with ipsilateral hip HO, 12.3% developed significant HO. In patients with contralateral hip HO, 10.5% developed significant HO after THA. Sixty percent who received 6 Gy in 3 fractions after excision of ipsilateral HO developed significant HO, which was higher than for all dosefractionation schemes combined ( P = .01). In contrast, patients who received 7 Gy in 1 fraction developed significant HO 13.8% of the time, which was equivalent to all dose-fractionation schemes combined ( P = not significant). Radiation prevents HO in HO-forming patients. Key words: heterotopic ossification, radiation, hip arthroplasty, dose fractionation, prevention. n 2006 Elsevier Inc. All rights reserved.
patients [2]. Risk factors for heterotopic bone formation in the hip include a history of HO in the ipsilateral or contralateral hip, hypertrophic osteoarthritis, ankylosing spondylitis, diffuse idiopathic skeletal hyperostosis, Paget disease, posttraumatic arthritis, osteonecrosis, and rheumatoid arthritis [3]. Patients with a history of HO are considered at high risk for HO. Eighty percent of these patients form HO after surgery alone [4]. Radiation is effective in preventing HO. Typically, the radiation is delivered within 72 hours after surgery or 24 hours before surgery. However, studies showing the effectiveness of radiation in patients with a history of HO are limited. Therefore, the purpose of this study was to determine if external beam radiation can prevent significant HO in patients with a history of heterotopic bone formation, either in the same hip after HO resection
Heterotopic ossification (HO), bone formation in nonosseous tissues, usually occurs in the setting of trauma such as fractures and surgical procedures [1]. Heterotopic ossification of the hip is the most common complication of total hip arthroplasty (THA). It can occur in as many as 53% of patients who undergo THA and causes postsurgical disability from pain and limited range of motion in 7% of From the *Department of Radiation Oncology, The Cleveland Clinic Foundation, Cleveland, Ohio; and yDepartment of Orthopedics, The Cleveland Clinic Foundation, Cleveland, Ohio. Submitted January 25, 2005; accepted August 12, 2005. No benefits or funds were received in support of the study. Reprint requests: John H. Suh, MD, Department of Radiation Oncology, The Cleveland Clinic Foundation, 9500 Euclid Avenue, T-28, Cleveland, OH 44195. n 2006 Elsevier Inc. All rights reserved. 0883-5403/06/1906-0004$32.00/0 doi:10.1016/j.arth.2005.08.014
731
732 The Journal of Arthroplasty Vol. 21 No. 5 August 2006 or the contralateral hip. This study is a retrospective review of the experience at our institution.
Table 1. Dose Fractionation of Radiation Received by Patients Dose-fractionation scheme
Materials and Methods The records and radiographs of 195 patients who underwent hip surgery between August 31, 1982, and July 24, 2001, and postoperative external beam radiation therapy to prevent significant HO in the operated hip were reviewed. One hundred twenty-four (63.3%) of these patients had a history of HO, either in the ipsilateral hip or the contralateral hip, and served as the basis for this study. Eighty-eight of these patients had HO in the ipsilateral hip and were treated with revision of their THA (63 patients) or excision of their HO alone (25 patients). The remaining 36 of 124 patients had a history of HO in the hip opposite to that undergoing THA. Because these patients are also at high risk for HO, they were treated with postoperative radiation to prevent HO. Patients were excluded from analysis if they did not have both preoperative and follow-up Brooker Grades [5], and had not been followed for at least 6 months. The latter step was taken to ensure that the HO had enough time to mature after surgery. Of a possible 124 patients, 84 (68%) patients met our inclusion criteria: 65 (77%) had HO in the same hip, and 19 (23%) had HO in the contralateral hip; the remaining 40 patients not meeting our inclusion criteria had incomplete records or insufficient follow-up at the time of the analysis. The median age was 64 years (range, 22-83 years). These 84 patients were treated with one of the following radiation doses: 20 Gy in 10 fractions, 20 Gy in 5, 10 Gy in 5, 8 Gy in 1, 7 Gy in 1, 6 Gy in 3, 6 Gy in 2, or 6 Gy in 1. Table 1 lists the number of patients receiving each fractionation scheme from 1982 to 2001. The more protracted courses of radiation were used in the 1980s and early 1990s. Single fraction therapy was used starting in September 1990 after studies demonstrated its effectiveness. One patient received preoperative radiation. All but 8 (90%) patients received radiation within 3 days after surgery. The median follow-up time was 20.5 months (range, 6.2 to 147.0 months). The preoperative and follow-up Brooker Grades were compared to determine if radiation was effective in preventing significant HO, which was defined as Brooker Grade 3 or 4. In cases where a Brooker Grade was not reported by the radiologist or orthopedic surgeon, either preoperatively or in follow-up, radiographs were obtained by the investigators to determine the Brooker Grade.
No. of patients
7 Gy in 1 fraction 6 Gy in 1 fraction 10 Gy in 5 fractions 6 Gy in 3 fractions 20 Gy in 10 fractions 20 Gy in 5 fractions 8 Gy in 1 fraction 6 Gy in 2 fractions
38 20 11 8 4 1 1 1
Fisher exact tests were used to compare the development of significant HO between the various groups. A P value of .05 or less was considered statistically significant.
Results Ipsilateral vs Contralateral Hip HO The 65 patients with a history of HO on the same side had preoperative Brooker Grades ranging from 0 to 4, with 55 hips (85%) having preoperative Brooker Grades of 3 or 4. After surgery and radiation, 14 developed no HO, 21 developed Brooker Grade 1 HO, 22 developed grade 2 HO, 4 developed grade 3 HO, and 4 developed grade 4 HO after radiation. As such, 8 of 65 (12.3%) developed significant HO. All 19 patients with HO on the contralateral side had a preoperative Brooker Grade of 0. After surgery and radiation, 10 developed no HO, 3 developed Brooker Grade 1 HO, 4 developed grade 2 HO, and 2 developed grade 3 HO after radiation. Thus, only 2 (10.5%) of 19 developed significant HO. There was no difference in the development of HO in the operated hip if the same hip or the contralateral hip was initially affected with HO ( P = not significant [NS]). Table 2 summarizes these results. Dose Fractionation (6 Gy in 3 Fractions and Standard of Care, 7 Gy in 1 Fraction) Five patients received 6 Gy in 3 fractions after excision of their heterotopic bone in the same hip (1 with preoperative Brooker Grade 2 HO, 2 with grade 3 HO, and 2 with grade 4 HO). Two developed Brooker Grade 2 HO, and 3 developed grade 4 HO in follow-up. Of 5 patients, 3 (60%) developed significant HO. Thus, 6 Gy in 3 fractions was not effective in preventing HO when compared with the other dose-fractionation schemes as group ( P = .01).
Radiation Prevents Heterotopic Bone ! Chao et al 733 Table 2. Brooker Grade After Radiation
Ipsilateral hip HO (n = 65) Contralateral hip HO (n = 19) P
No HO
Grade 1
Grade 2
Grade 3
Grade 4
Rate of significant HO
14 (21.5) 10 (52.6)
21 (32.3) 3 (15.8)
22 (33.8) 4 (21.0)
4 (6.15) 2 (10.5)
4 (6.15) 0 (0)
12.3 10.5 NS
Values are presented as n (%) or percentages.
By contrast, 29 patients received 7 Gy in 1 fraction after excision of their heterotopic bone in the same hip. This dose-fractionation scheme was found to be effective by other studies and is the dose fractionation currently being used at our institution. Three of these patients developed Brooker Grade 3, and 1 developed grade 4 HO on follow-up. Thus, 4 (13.8%) of 29 treated with 7 Gy in 1 fraction developed significant HO. When compared with the other dose-fractionation schemes as a group, there was no significant difference in the rate of significant heterotopic bone formation ( P = NS). Table 3 summarizes these results. Complications Complications included postoperative infection (n = 3), all of which occurred in patients who underwent excision of HO in the ipsilateral hip. No patients developed symptomatic loosening of their prosthesis after removal of HO. One patient received an ingrowth stem and cup after HO removal, followed by radiation to 10 Gy in 5 fractions. During the radiation, the ingrowth components were blocked. He developed minor loosening documented in follow-up 2.5 years after x-ray but had good hip function with a Harris score of 88. Because of lack of symptoms, he never required reoperation. The second patient initially had right hip arthroplasty with a cementless acetabular component and cemented femoral component. He subsequently developed HO and had resection of his HO followed by radiation to 7 Gy in 1 fraction. Follow-up x-ray showed some evidence of loosen-
ing about the methacrylate-bone interface in the femur 7 months later and thigh pain. He had an aspiration arthrogram that was negative. Instillation of contrast from the arthrogram was negative for evidence of loosening. He subsequently developed other medical issues and was lost to further follow-up. One patient developed trochanteric bursitis in follow-up. This patient had HO in the ipsilateral hip and underwent revision with ingrowth stem and cup, followed by radiation to 6 Gy in 3 fractions. Other than trochanteric bursitis, the patient continued to have good range of motion and a Harris score of 78. In another patient, the acetabular component had migrated, but she had a history of rectal cancer treated with radiation. This patient initially had a THA followed by 2 other surgeries for dislocations. She was found to have Brooker Grade 3 HO in that hip and underwent repeat surgery for dislocation thought to be related to her HO. She had removal of HO and arthroplasty of the acetabular component, which was affixed with screws. Seven Gy in 1 fraction was delivered for prophylaxis. Follow-up approximately 1 year later showed migration of the acetabular component, with development of pain in the buttock. This migration and failure of fixation were attributed to her previous radiation for rectal cancer 14 years prior. The patient subsequently underwent revision with a Depuy acetabular cage, bone graft, and arthroplasty of the femoral component with subsequent improvement in her function. There were no known healing complications associated with radiation in this study.
Table 3. 6 Gy in 3 Fractions and 7 Gy in 1 Fraction Compared With Other Dose Fractionations as a Group
6 Gy in 3 fractions (n = 5) 7 Gy in 1 fraction (n = 29)
Grade 3
Grade 4
Rate of significant HO
P value when compared with all other fractionations combined
0 (0) 3 (10.4)
3 (60) 1 (3.4)
60 13.8
.01 NS
Values are presented as n (%) or percentages.
734 The Journal of Arthroplasty Vol. 21 No. 5 August 2006
Discussion Patients with a history of HO are at high risk for recurrence after resection. Given that pluripotential mesenchymal cells are responsible for the development of HO [6], it may be possible to stop the ossification by radiating these cells. Fractionated radiation (20 Gy in 10 fractions) was used and reported in 1981 [7]. In this landmark study, patients were classified into various groups; patients in groups 1, 3, 4, and 5 had a history of HO. Group 1 consisted of 12 hips where HO was removed. None of these hips developed severe or massive HO after excision and radiation. Group 3 consisted of 14 hips with a history of HO in that hip; a THA was performed as opposed to excision alone. Of these 14 hips, 3 (21%) had severe HO, but the postoperative function was better than that before surgery and radiation. Group 4 consisted of 7 patients who had a history of HO on the contralateral side. None of these patients developed severe or massive HO after surgery and radiation. Group 5 was composed of patients who had preexisting mild-to-moderate HO that was treated with radiation alone with no excision. Their HO progressed after radiation. Understandably, radiation alone has no effect on preexisting HO because the cells have already differentiated and have formed bone [8]. Since then, various fractionation schemes have been studied. A single fraction of radiation to 7 Gy was studied and found to be effective in preventing HO as well [9]. In this series of 25 patients (26 hips), 6 patients underwent hip revision and excision of HO, 4 underwent excision of preexisting HO, and 6 had a history of HO on the contralateral side. None of the patients had any functional impairment, and none of the patients developed Brooker Grade 3 or 4 HO. In another study, a single fraction of 5.5 Gy was compared with a single fraction of 7 Gy, retrospectively [10]. Sixty-three percent of the hips radiated with 5.5 Gy developed HO, with 2 of 19 being symptomatic. In comparison, only 10% of the patients in the 7 Gy group developed HO, and none were symptomatic. This study also categorized patients with a history of HO. In patients with ipsilateral HO, 2 of 5 treated with 5.5 Gy developed significant HO defined as Brooker Grade 3 and 4 vs 0 of 23 in the 7 Gy group. In patients with contralateral HO, 1 of 3 developed significant HO with 5.5 Gy vs 0 of 20 with 7 Gy. Therefore, 5.5 Gy was concluded to be ineffective in preventing HO, especially in patients with a history of HO. A dose of 6 Gy in a single fraction was also studied and found to be effective, with
none of the 17 patients in the series developing Brooker Grade 3 or 4 HO [11]. A randomized controlled trial was performed in Germany comparing preoperative radiation vs postoperative radiation for the prevention of HO [12]. Patients were randomized to receive 7 Gy of radiation within 4 hours before hip surgery or 17.5 Gy in 5 fractions less than 96 hours postsurgery. Five percent developed failures in the postoperative radiation arm vs 19% in the preoperative radiation arm ( P b .05). In subgroup analysis of patients without grade 3 or 4 HO before surgery, there was no difference between the preoperative and postoperative radiation groups. However, in patients who had radiation before the removal of their ipsilateral grade 3 or 4 HO, there was a failure rate of 39%, which was statistically significant ( P b .001) when compared with the postoperative arm, which failed only 9% of the time. As such, preoperative radiation was determined to be not as effective when compared with postoperative radiation for these high-risk patients. A recent study studied the optimal timing of preoperative radiation to prevent HO using a rabbit model [13]. Radiation was given at 4 different time points: 4 hours, 24 hours, 72 hours, and 3 weeks. Radiation given 24 hours before surgery was more effective than the other time points in preventing HO. It is speculated that cell cycling is necessary for the pluripotent mesenchymal cells to die. Twentyfour hours allows enough time for cell cycling with resulting cell death, whereas 4 hours is insufficient. This may explain why patients in the German study with a history of HO in the ipsilateral hip who received preoperative radiation do not do as well compared with postoperative radiation when the preoperative radiation was given only 4 hours before surgery. In our current retrospective study, the role of radiation in patients with a history of HO, either in the same hip or the contralateral hip, was assessed given that these patients represent the highest-risk group. In patients with a history of HO in the treated hip, the rate of HO after radiation was 12.3%. In patients with a history of HO in the hip contralateral to the treated hip, the rate of HO was 10.5%. There was no difference in the rate of HO when this was initially present in the same hip or the opposite hip ( P = NS). These results compare favorably with other series such as the German study mentioned previously despite these patients being at high risk for the development of HO. A variety of fractionation schemes have been used at our institution since 1982. This includes 20 Gy in 10 fractions, 20 Gy in 5, 10 Gy in 5, 8 Gy
Radiation Prevents Heterotopic Bone ! Chao et al 735
in 1, 7 Gy in 1, 6 Gy in 3, 6 Gy in 2, and 6 Gy in 1. Five patients in this study were treated using 6 Gy in 3 fractions. In these patients, 60% developed significant HO. When compared with the other fractionations, this fractionation scheme was not effective ( P = .01) in preventing HO. Given that this represents less dose biologically compared with 6 Gy in a single fraction, the lack of efficacy with this dose fractionation was not surprising. Fractionating therapy allows for repopulation of the pleuripotential mesenchymal cells between fractions. Thus, to achieve the same control rates, it is necessary to increase the total dose during fractionating therapy. Conversely, when decreasing the number of fractions, a larger dose per fraction is required. This is why 10 Gy in 5 fractions is equivalent to 7 Gy in 1 fraction but 6 Gy in 3 fractions is not equivalent to 6 Gy in 1 fraction. Seven Gy in 1 fraction has become a popular fractionation scheme because it requires only 1 day of treatment. Given that these patients are often in pain and immobile after surgery, the convenience of 1-day treatment is beneficial for both the patient and the staff. This dose fractionation is currently used in our department with 29 (45%) of 65 patients treated in this fashion in this series. Only 13.8% of the patients with HO in the same hip developed significant HO after radiation. When compared with the other dose fractionations, there was no difference in effectiveness. These results are comparable to published literature showing that a single fraction of 7 Gy is equally effective to fractionated doses of 10 Gy, which was shown to be as effective as fractionated doses of 20 Gy [9,14-17]. In this study, complications were rare. No patient had symptomatic loosening of their femoral components. One patient had shielding of the ingrowth femoral stem during radiation but, despite this, developed x-ray evidence of loosening. However, he continued to have good function in that hip and never required reoperation. Another patient only had excision of HO and no revision of his cemented THA. While on x-ray, he was felt to have some loosening; he was found not to have loosening based on the results of arthrogram. One patient had migration of the acetabular component, but this was mainly attributed to her history of pelvic radiation for rectal cancer because she was having problems with dislocations before receiving radiation for HO prophylaxis. She also developed the migration despite the use of fixation screws. Her acetabular component required eventual arthroplasty. In general, it is recommended that ingrowth components be shielded because radiation may
prevent bone ingrowth like it prevents HO bone growth. A study placed porous components into rabbit tibias and radiated some of these tibias to 10 Gy in 5 fractions [18]. The rabbits were killed. At 2 weeks, the force necessary to remove the component after radiation was significantly less than the force to remove the component from the untreated tibia. However, this difference was not statistically significant at 3 weeks and beyond. Despite this, the previously mentioned German study failed to show increased rates of loosening or instability in patients after radiation of their porous components without shielding [12]. Removal of HO by itself is not known to cause loosening. Currently, drug therapy with nonsteroidal antiinflammatory drugs (NSAIDs) such as indomethacin and cyclooxygenase-2 inhibitors is being used as an alternative to radiation for prophylaxis with good success [19-21]. Given these results, a recent meta-analysis was conducted using studies that compared radiation and NSAID therapy in the prevention of HO [22]. These studies included a variety of patients and were not limited to high-risk patients. This meta-analysis showed decreased risk for clinically significant HO (defined as Brooker Grade 3 or 4) with radiation vs NSAID. The study also suggests that higher doses beyond 6 Gy in 1 fraction are more effective, whereas a dose of 6 Gy is equally effective as NSAID therapy. Further investigation may be necessary in determining if NSAID therapy is effective in high-risk patients and if the decrease in risk with radiation also holds for high-risk patients when compared with NSAID therapy. If a patient is on NSAID therapy 72 hours after surgery and develops gastrointestinal complications or renal dysfunction, or require anticoagulation with heparin or Coumadin, preventing further use of NSAID, there is no longer any treatment that is effective. This would put the patient at high risk for redeveloping HO.
Conclusion The morbidity from HO may be severe with pain and limitation in the range of motion. In patients at risk for HO, prophylaxis with radiation or NSAID after hip surgery is often recommended to prevent HO. The highest-risk patients are those with a history of HO, either in the same hip or the contralateral hip. This study, as well as other studies, has confirmed that prophylaxis with radiation is effective in preventing HO formation in high-risk patients. Seven Gy in 1 fraction postoperatively has been shown to be as effective as fractionated
736 The Journal of Arthroplasty Vol. 21 No. 5 August 2006 therapy to 10 or 20 Gy and is the scheme being used at our institution. Preoperative radiation may be more convenient given the immobility and pain after surgery, which can make postoperative radiation difficult. Despite a study that showed that preoperative radiation is ineffective in patients with a history of HO, timing may play a role in the effectiveness of preoperative radiation. Clinical studies regarding the timing of preoperative radiation are indicated, especially in patients with a history of HO. Also, studies regarding NSAID therapy included a variety of patients and were not limited to patients with a history of HO. More studies are necessary to determine if NSAID therapy is as effective as radiation in this subgroup of high-risk patients.
11.
12.
13.
14.
References 15. 1. Kaplan FS, Glaser DL, Hebla N, et al. Heterotopic ossification. J Am Acad Orthop Surg 2004;12:116. 2. Thomas BJ, Amstutz HC. Prevention of heterotopic bone formation. Hip 1987;59. 3. Lo TCM. Radiation therapy for heterotopic ossification. Semin Radiat Oncol 1999;9:163. 4. Ayers DC, Pellegrini VD, Evarts CM. Prevention of heterotopic ossification in high-risk patients by radiation therapy. Clin Orthop 1991;263:87. 5. Brooker AF, Bowerman JW, Robinson RA, et al. Ectopic ossification following total hip replacement. J Bone Joint Surg 1973;55-A:1629. 6. Jowsey J, Coventry MB, Robins PR. Heterotopic ossification: theoretical consideration, possible etiologic factors, and a clinical review of total hip arthroplasty patients exhibiting this phenomenon, in the hip. In: The Hip Society, editor. In proceedings of the fifth open scientific meeting of the Hip Society. St Louis: CV Mosby; 1977. p. 210. 7. Coventry MB, Scanlon PW. The use of radiation to discourage ectopic bone: a nine-year study in surgery about the hip. J Bone Joint Surg Am 1981; 63:201. 8. Tonna EA, Cronkite EP. Auto radiographic studies of cell proliferation in the periosteum of intact and fractured femora of mice utilizing DNA labeling with ( 3H) thymidine. Proc Soc Exp Biol Med 1961;107:719. 9. Lo TC, Healy WL, Covall DJ, et al. Heterotopic bone formation after hip surgery: prevention with singledose postoperative hip irradiation. Radiology 1998;168:851. 10. Healy WL, Lo TC, DeSimone AA, et al. Single-dose irradiation for the prevention of heterotopic ossifica-
16.
17.
18.
19.
20.
21.
22.
tion after total hip arthroplasty. J Bone Joint Surg 1995;77-A:590. Hedley AK, Mead LP, Hendren DH. The prevention of heterotopic bone formation following total hip arthroplasty using 600 rad in a single dose. J Arthroplasty 1989;4:319. Seegenschmiedt MH, Keilholz L, Martus P, et al. Prevention of heterotopic ossification about the hip: final results of two randomized trials in 410 patients using either preoperative or postoperative radiation therapy. Int J Radiat Oncol Biol Phys 1997;39:161. Rumi MN, Deol GS, Bergandi JA, et al. Optimal timing of preoperative radiation for prophylaxis against heterotopic ossification: a rabbit hip model. J Bone Joint Surg Am 2005;87-A:366. Anthony P, Keys H, Evarts CM, et al. Prevention of heterotopic bone formation with early post operative irradiation in high risk patients undergoing total hip arthroplasty: comparison of 10.00 Gy vs 20.00 Gy schedules. Int J Radiat Oncol Biol Phys 1987;13:365. Sylvester JE, Greenberg P, Selch MT, et al. The use of postoperative irradiation for the prevention of heterotopic bone formation after total hip replacement. Int J Radiat Oncol Biol Phys 1988;14:471. Ayers DC, Evarts CM, Parkinson JR. The prevention of heterotopic ossification in high-risk patients after total hip arthroplasty. J Bone Joint Surg Am 1986; 68:1423. Pellegrini VD, Konski AA, Gastel JA, et al. Prevention of heterotopic ossification with irradiation after total hip arthroplasty. J Bone Joint Surg 1992;74-A:186. Konski A, Weiss C, Rosier R, et al. The use of postoperative irradiation for the prevention of heterotopic bone after total hip replacement with biologic fixation (porous coated) prosthesis: an animal model. Int J Radiat Oncol Biol Phys 1990; 18:861. Kienapfel H, Koller M, Wust A, et al. Prevention of heterotopic bone formation after total hip arthroplasty: a prospective randomized study comparing postoperative radiation therapy with indomethacin medication. Arch Orthop Trauma Surg 1999;119:296. Sell S, Willms R, Jany R, et al. The suppression of heterotopic ossification: radiation versus NSAID therapy — a prospective study. J Arthroplasty 1998; 13:854. Moore KD, Goss K, Anglen JO. Indomethacin versus radiation therapy for prophylaxis against heterotopic ossification in acetabular fractures—a randomized, prospective study. J Bone Joint Surg 1998;80-B:259. Pakos EE, Ioannidis JPA. Radiotherapy versus nonsteroidal anti-inflammatory drugs for the prevention of heterotopic ossification after major hip procedures: a meta-analysis of randomized trials. Int J Radiat Oncol Biol Phys 2004;60:888.