Heterotopic ossification: Pathophysiology, clinical features, and the role of radiotherapy for prophylaxis

The evolution of radiotherapy for HO prophylaxis

Coventry et al. established that RT could successfully be used in the prevention of HO. The next step was to pursue reduced doses for prophylaxis, especially in light of the concern regarding radiation-induced malignancies. Sylvester  reported on a retrospective comparison of patients receiving 20 Gy in 10 fractions as compared with 10 Gy in 5 fractions. A total of 27 hips were irradiated after THA. Two hips receiving 20 Gy and one hip receiving 10 Gy developed clinically significant HO. Additionally, all three of the hips developing clinically significant HO had radiation delivered >4 days after surgery. The authors concluded that, though the sample size was small, the two radiation regimens appear to be similarly effective. They also noted that RT should be delivered postoperatively within 4 days of surgery.

The impetus for a decrease in dose from 20 Gy to 10 Gy according to Sylvester et al., was the need to decrease hospital stays. Further supporting this trend to decrease dose and fractions is that the decrease in dose theoretically may diminish radiation-induced cancer risk, and the decrease in fractionation reduces cost and inconvenience to patients. Hence, the next step in the evolution of RT for HO prevention was the exploration of single-fraction treatment. Lo, in a retrospective series, reviewed the use of single-fraction RT in patients considered to be at high risk for HO after THA. A total of 24 hips in 23 patients were treated with a single dose of 7 Gy. Only 1 patient was treated beyond 72 h and anteroposterior/posteroanterior fields were used. None of the patients developed Brooker Grade 3 or 4 HO. The authors concluded that single-dose RT with 7 Gy is effective in preventing HO. Pellegrini  further explored the efficacy of single-fraction therapy for HO prophylaxis after THA in a prospective, randomized trial of single fraction vs. fractionated therapy. A total of 62 hips in 55 patients undergoing hip surgery and considered to be at high risk for HO were randomized to 10 Gy in five fractions vs. 8 Gy in one fraction postoperatively. RT was initiated within 4 days of surgery in all but 1 patient. HO (of any grade) developed in 21% of the hips receiving 10 Gy and in 21% of those receiving 8 Gy. The authors concluded that single-fraction therapy is similar in efficacy to fractionated therapy.

Having established the effectiveness of single-dose RT of 7–8 Gy, an attempt was made to consider lower doses of single fraction treatment. Healy  examined single-dose irradiation with 7 Gy in comparison to 5.5 Gy. This was a retrospective analysis of 107 hips deemed to be at high risk for the development of HO after THA. All except 1 patient were treated within 3 days of the procedure and anteroposterior/posteroanterior fields were used. HO developed in 10% of those receiving 7 Gy and in 63% of those receiving 5.5 Gy (p = 0.03). Clinically significant HO (Grade 3 or 4) was noted in 2 patients receiving 7 Gy and 4 patients receiving 5 Gy. The authors concluded that 5.5 Gy is not a sufficient dose for HO prophylaxis. Reduced-dose RT was more recently tested by Padgett  in a prospective, randomized trial. Fifty-nine patients considered to be at high risk for HO after THA were randomized to receive either 5 Gy in two fractions or 10 Gy in five fractions. All patients were treated within 4 days of surgery. There was a trend toward increased HO of any grade in the 5-Gy group (69% vs. 43%, p = 0.09). Most HO was clinically insignificant, however. Clinically significant HO was noted in 2 of 19 patients receiving 5 Gy vs. 1 of 30 patients receiving 10 Gy (no p value indicated). The authors concluded that there was no significant difference in HO development between those receiving 5 Gy in two fractions vs. 10 Gy in five fractions, but added that a true difference may not have been detected because of a small sample size. Though 5 Gy may be inferior to 10 Gy for HO prophylaxis, the difference may be marginal when comparing the incidence of clinically significant HO. Studies with larger numbers are needed to determine whether reduced radiation doses are indeed sufficient for HO prophylaxis.

With evidence to support the effectiveness of preoperative RT in HO prevention, Seegenschmiedt  performed a prospective, randomized trial of preoperative vs. postoperative radiotherapy. One hundred and sixty-one patients considered to be at high risk for HO development were randomized to receive RT either preoperatively (<4 h before surgery) or postoperatively (<72 h after surgery). Patients receiving preoperative treatment received 7 Gy in one fraction. Patients receiving postoperative treatment received 17.5 Gy in five fractions. Within the preoperative therapy group (n = 80), there were 19 treatment failures, and in the postoperative therapy group (n = 81), there were 4 treatment failures. The difference between the groups was statistically significant (p < 0.05). However, the authors did note that this difference was not apparent when comparing patients who preoperatively had a low “preload of ectopic bone”—Brooker grades 0–2. Though it is unclear if this should impact efficacy of HO prevention, there is a notable difference in the biologic equivalent dose between the preoperative and postoperative RT regimens (based on an α/β of 10, biologic equivalent dose = 11.9 and 23.6, respectively). A randomized, controlled trial of preoperative vs. postoperative therapy was performed by Gregoritch et al., in this case using the same RT dose and fractionation . A total of 122 patients (124 hips) receiving THA and at high risk of HO were randomized to receive 7–8 Gy in one fraction either preoperatively (<4 h before surgery) or postoperatively (<72 hours postoperatively). The authors reported no significant difference between the treatment groups with an HO overall incidence of 26% in the preoperative RT group vs. 28% in the postoperative RT group. Clinically significant HO was noted in 2% of the preoperative group vs. 5% in the postoperative group. The authors concluded that the preoperative and postoperative regimens are similar in efficacy. However, the sample size was insufficient to determine true equivalence.

Shielding of the prosthesis in radiation therapy for HO prevention

The majority of prostheses used in THA are cementless prostheses with porous elements permitting bony ingrowth. Concern has arisen that RT may inhibit bony ingrowth into the prosthesis and cause prosthesis failure at either the bony interface of the prosthetic acetabulum or at the proximal femoral region where the shaft of the prosthesis abuts the native femur. Konski  studied the impact of RT on bony growth into a porous coated rod in rabbits to investigate the validity of this concern. The rabbits underwent a procedure placing porous coated rods into the bilateral tibias. Each animal had one tibia irradiated 1 day postoperatively to a total of 10 Gy in five fractions. The animals were then sacrificed at weekly intervals beginning at 2 weeks and up to 6 weeks after surgery, and the amount of force necessary to pull the rod out of the medullary cavity of the treated and untreated tibias was compared. The authors noted that at 2 weeks, there was a statistically significant difference between the amount of force necessary to remove the rod, with less force required for the treated tibia as compared with the untreated tibia. After 3 weeks, there was no difference in force required to remove the rod. The authors concluded that there is decreased bony ingrowth in radiated bone, but that this only results in transient instability of the implanted prosthesis. Given these findings, the authors advocated shielding of the prosthesis. Despite these findings, however, clinical data thus far have not substantiated an increased risk of prosthesis failure in the setting of RT. For example, in the aforementioned study of preoperative vs. postoperative RT by Seegenschmiedt et al. (n = 188 patients with uncemented implants), there was no evidence of prosthesis failure despite the fact that the prostheses were not shielded.

Shielding of the prosthesis has raised concern for potential reduced efficacy of prophylactic RT for HO. One study by Jasty  evaluated the impact of shielding of the acetabular and femoral prosthetic components in a small population of patients undergoing THA. This was a retrospective review of 16 patients (18 hips) who were considered to be at high risk for HO development. Patients received prophylactic RT to a total of 15 Gy in five fractions initiated within 48 h of surgery. The femoral and acetabular components of the prosthesis were shielded with Cerrobend blocks. Only 2 of 18 hips developed HO and both were Brooker Grade 1 lesions. Although the number of patients was small, the authors concluded that RT with precision shielding of the prosthetic components remains an effective means of preventing heterotopic bone formation.

Radiation side effects

The most concerning potential side effect of RT is carcinogenesis, although there have been no documented cases of radiation-induced tumors after RT for HO prevention. This may reflect the relatively low dose used for treatment. In a review of their 50-year experience of radiation-induced sarcomas, Kim et al. reported no cases of bone or soft-tissue sarcomas in patients exposed to doses lower than 30 Gy  In addition to the low doses used, another factor that may contribute to the lack of observed cases of second malignancies in patients receiving RT for HO prophylaxis is that these patients tend to be older. In the meta-analysis of Pakos and Ioannidies, for example, the average age of 1143 patients receiving HO prophylaxis was 61 With the latency for radiation-induced tumors typically 10 years or longer, it is possible that the lack of documented second malignancies is partially attributable to the relatively small number of patients exposed to radiation who live long enough to experience a second tumor. It therefore remains possible that as more patients are followed for a longer interval after exposure to radiation for HO prophylaxis, second tumors may be observed. This concern seems particularly worthy of consideration when young patients at risk for HO development are referred for RT prophylaxis.

Trochanteric nonunion is also a potential side effect of RT. Trochanteric osteotomy is occasionally necessary to facilitate removal of a hip prosthesis at the time of revision. Studies not using shielding of the osteotomy site have found trochanteric nonunion rates of approximately 12–30% after RT. For comparison, rates of nonunion after trochanteric osteotomy range from approximately 2–15% . The use of shielding to lower the risk of nonunion does not thus far appear to diminish the efficacy of radiation as prophylaxis for HO, though there are few data to answer this question. Of note, the osteotomy technique in these studies—the Charnley trochanteric osteotomy—has largely been abandoned given the high rate of nonunion in unirradiated patients. An extended trochanteric osteotomy—in which the excision of the greater trochanter extends into the diaphysis of the femur—is now typically performed to access and remove the prosthesis as this permits a greater surface area for bony union to occur. The authors are unaware of any studies examining the effect of RT on rates of union after this procedure.

Last, radiation dose to the testis is also of concern given the potential for reduction in sperm counts and the theoretical risk of radiation-produced hereditary effects. Doses as low as 20 to 70 cGy have been noted to result in reversible oligospermia and doses of 120 cGy or higher are considered to confer a risk of permanent azoospermia. Furthermore, according to the Committee on the Biologic Effects of Ionizing Radiation and the United Nations Scientific Committee on the Effects of Atomic Radiation, the doubling dose for hereditary genetic effects is 100 cGy. These data are subject to controversy, however, given that they are largely based on animal data. In an abstract by Patel , 800 cGy in one fraction for HO prophylaxis was found to result in a mean testicular dose of 25.1 cGy (range, 13–50 cGy). A testicular shield was found to reduce this dose by approximately 54%, yielding an average dose of 11.3 cGy (range, 3–26 cGy). Given these findings, the authors advocate the use of a testicular shield in men. They also recommend informing patients in whom shielding is not used of a potential reduction in sperm count and possible genetic abnormalities within the sperm for 6–12 months after RT.