Prophylactic Radiotherapy of Hip Heterotopic Ossification: A Narrative Mini Review

Abstract

Evidence on prophylactic radiotherapy (RT) in hip heterotopic ossification (HO) is sparse and conflicting. The aim of this literature review was to collect and summarize the available data on RT efficacy in preventing hip HO. The results of this review show that RT is effective in the prevention of hip HO, albeit with large variability across series. Effective prophylactic RT requires optimal treatment fields and time intervals with surgery. On the contrary, there is no clear evidence on the optimal timing (post-operative versus pre-operative RT). Comparisons between prophylactic RT and use of non-steroidal anti-inflammatory drugs showed conflicting results, although most were in favor of RT. In conclusion, RT is an established prophylactic treatment for hip HO. However, optimal dose, technique and timing remain unclear, as does the usefulness of combining RT with drugs.

Heterotopic ossification (HO) is defined as the formation of new bone in soft tissue outside the skeletal system (1). HO can be differentiated into three main groups: traumatic HO (mainly following fractures), non-traumatic HO (usually occurring after burns), and neurological HO (2).

Several prophylactic treatments for HO have been proposed, such as non-steroidal anti-inflammatory drugs (NSAIDs), Noggin (an extracellular peptide that binds and antagonizes bone morphogenetic proteins), pulsed electromagnetic fields, and free radical scavengers (3-11).

HO is a particularly frequent complication after total hip arthroplasty, with reported rates ranging from 15% to 90%. In patients with a significant amount of ossification, hip mobility can be impaired (12). Main risk factors for HO after total hip arthroplasty are male gender, hip ankylosis, and previous history of HO (13,14). The only effective treatment of symptomatic, established HO is surgical resection (15).

From the early 1980s (16), radiotherapy (RT) has been extensively studied and used in this setting. Most evidence on RT efficacy in preventing hip HO comes from non-randomized studies (17-45), although some randomized trials (46-61) and systematic-reviews and meta-analyses (62-68) have been performed.

However, evidence on prophylactic RT is sparse and conflicting, no international guidelines are available, and several questions remain unanswered. Therefore, the aim of this literature review was to collect and summarize the main available evidence on RT efficacy in preventing hip HO.

Review Method

A literature search was conducted on PubMed on 30 May 2021. The search strategy was as follows: “heterotopic ossification”[All Fields] AND “hip”[All Fields] AND (“radiotherapy”[All Fields] OR “radiation”[All Fields]). In this review, we included clinical studies and meta-analyses published in English after 1990 reporting on patients treated with RT to prevent hip HO development, including results in terms of efficacy or toxicity, with data on RT dose and fractionation, with prospective or retrospective design, and enrolling more than 50 patients. Studies were excluded in the case of their being a letter, commentary, editorial, case report, conference proceedings, reports on study protocols, preclinical studies, studies on animal models, imaging or planning studies, surveys, guidelines, or recommendations, or due to inclusion of prophylactic RT for sites other than the hip, or duplication of data.

A summary of the main findings of selected non-randomized studies, randomized trials, and meta-analyses are summarized in Table I, Table II, and Table III, respectively. Figure 1 shows computed tomographic scans of an extensive HO of the right hip of a 55-year-old male patient with paralysis of the lower extremity.

Table I. Number of patients and main findings of nonrandomized studies.

HHS: Harris Hip Score, an outcome measure after hip surgery; HO: heterotopic ossification; RT: radiotherapy; THA: total hip arthroplasty. *Based on pure pain, mobility of the leg and ability to walk, grading each from 1 to 6 points.

Table II. Number of patients and main findings of randomized trials.

HO: Heterotopic ossification; LD-/HD-RT: low-dose/high-dose radiotherapy; NSAID: non-steroidal anti-inflammatory drug.

Table III. Number of patients, comparison, and main findings of meta-analyses.

BED: Biologically effective dose; CI: confidence interval; COX-II: cyclo-oxygenase 2; HO: heterotopic ossification; NSAID: non-steroidal antiinflammatory drug; OR: odds ratio; RR: risk ratio; RT: radiotherapy.

Figure 1. Extensive heterotopic ossification of the right hip in a 55-yearold male patient with paralysis of the lower extremity. Computerized tomography: A: coronal reconstruction; B: sagittal reconstruction.

Discussion

All studies on the efficacy of RT in the prevention of hip HO reported a significant advantage over surgery alone (19,22,23,41,56,57). However, the incidence of HO after prophylactic RT was highly variable, with overall rates ranging between 6% and 28% (17,19,27,29,30,46,49), and grade III-IV HO rates ranging from 0% to 5% (19,24,27,29,30,46,49,57). Nevertheless, no grade III-IV hip HO cases were recorded in most series of patients treated with RT (19,24,27,46,57).

In terms of dose, many studies analyzed the impact of RT regimens on HO rates (18,20,26,32,38,46,48,51,52,59,61,65-67). In some studies, single 7- to 8-Gy fractions were found to be equivalent to regimens of 3×4 Gy (18) and 5×2 Gy (26,38,46). Furthermore, other studies have shown the superiority of 1×7 Gy over regimens of 1×4 Gy (61), 1×5 Gy (51,52), 1×5.5 Gy (20), and 3×2 Gy (32). Moreover, some analyses reported the equivalence of 5×2 Gy versus 1×7 Gy (38), 1×8 Gy (46), 10×2 Gy (48), 5×3.5 Gy (48), and 2×2.5 Gy (59). In addition, one study showed the greater efficacy of 4×3 Gy compared to 1×5 Gy (51). Finally, two meta-analyses did not record a significant impact of RT dose on postoperative HO incidence (64,65), while another showed an advantage of multi-fractionated RT over single-fraction RT (67).

In terms of the timing between surgery and prophylactic RT, some studies reported superior results in patients with an interval of less than 6 days (17), and 3 days (19,28). Furthermore, another study confirmed that delays in postoperative RT can reduce its efficacy in terms of HO prophylaxis (47). Finally, a study showed that the effectiveness of preoperative RT is greater in the case of an RT-surgery interval ≤8 hours (28).

Comparing pre- versus post-operative prophylactic RT, three studies did not show significant differences between the efficacy of the two strategies (28,49,50). This equivalence was confirmed by three meta-analyses (64,65,67).

Comparisons with other HO prophylaxis methods was conducted exclusively between RT and NSAIDs (37,39,51,52,54,55,58,60,63,65,66,68). Three studies reported the superiority of RT over indomethacin (37,39,60) or acetylsalicylic acid (51). Two other studies confirmed the superiority of prophylactic RT over NSAIDs in general (52,55). In contrast, one study showed the superiority of NSAIDs over RT in terms of preventing all HOs but equivalence between the two treatments in terms of grade III-IV HO (54). Finally, two meta-analyses showed the superiority of RT over NSAIDs (62,66), two meta-analyses showed the two treatments to be equivalent (58,63), and one meta-analysis showed the superiority of NSAIDs (68).

Only two studies provided information on different outcomes of prophylactic RT in different HO subgroups. Ebinger et al. compared the outcome after surgery and prophylactic RT in patients with hip HO that developed after brain injury, local hip trauma, or the combination of both (25). The authors recorded similar recurrence rates among groups, but better clinical outcome in patients with hip trauma. Cipriano et al. reported a lack of protective effect by prophylactic RT (1×7 Gy) in patients with resected neurogenic HO (36).

In terms of surgical outcome and side-effects, one study showed that prophylactic RT combined with indomethacin did not have an impact on the stability of cementless cups (35). Moreover, another article reported prophylactic RT to be associated with delayed wound-healing rates similar to those after surgery alone (36). Furthermore, one study showed the same rate of implant loosening after prophylactic RT or indomethacin (45). In contrast, another analysis showed a higher incidence of postoperative sepsis in patients undergoing prophylactic RT (44). Finally, one study showed the absence of radiation-induced toxicity in patients undergoing prophylactic RT (36) and another reported the absence of RT-induced tumors at the treated site after 10 years of follow-up (45).

In terms of RT technique, one study showed that incorrect positioning of shielding blocks is associated with a higher incidence of HO (17). Another study showed that shielding was associated with higher rates of HO (34). Finally, a planning study on patients treated with prophylactic RT reported lower mean and maximum doses to testicles in patients whose therapy was planned with a split-beam technique (42).

In summary, the results of this literature review show that RT is effective in the prevention of hip HO, albeit with wide ranges of efficacy across series. Low RT doses (4-5.5 Gy) seem to be less effective compared to intermediate doses (7-8 Gy), while higher doses do not provide further advantages. However, we can note that German guidelines recommend a 5×3.5 Gy regimen in patients with a high risk of developing HO (2). For effective prophylactic RT, it is important to respect the correct intervals between preoperative RT and surgery or between surgery and postoperative RT, and a careful definition of treatment fields is needed. On the contrary, clear evidence of the superiority of post-operative versus pre-operative RT is lacking. However, some authors suggested the use of pre-operative RT in order to reduce logistical problems (31) and discomfort and possible complications of post-operative RT (50).

Comparisons between prophylactic RT and administration of NSAIDs have shown conflicting results, although most evidence is in favor of RT. Furthermore, there is very little evidence on the efficacy of RT in the prevention of HO recurrence in patients undergoing HO removal and on the efficacy of RT in the different HO subgroups based on etiology. Moreover, RT is not correlated with clinically detectable side-effects, peri- or post-operative complications, or radiation-induced tumor rates.

This analysis has several limitations. Most of the evidence comes from retrospective studies. This type of study design is obviously associated with the risk of selection bias. Indeed, some authors explicitly admitted that patients at higher risk of HO were preferentially referred to RT over observation or drug treatment alone (26,36). It is clear that this bias may have limited the detection of benefits in patients undergoing RT. In addition, many studies evaluated the incidence of HO based on the Brooker classification (69), a widely used quantitative and qualitative assessment tool. However, this classification presents some ambiguities that may limit its generalizability between different centers and specialists (70). Furthermore, no study included patient-reported outcome measures among the main objectives of the analysis. Therefore, there is a lack of data on the real impact of RT on quality of life. Finally, in most cases, the evaluation of prophylactic RT efficacy was performed considering all HO grades. However, only higher-grade HOs are known to affect physical functions (30).

In conclusion, after 40 years of experience, RT is an established prophylactic treatment for hip HO. However, optimal doses, techniques, and timing remain undefined, as does the usefulness of combining RT with drug treatments, at least for some categories of patients.

Therefore, further studies are needed, in particular to i) evaluate the efficacy of RT in secondary HO prevention in combination with the resection of already developed HOs; ii) evaluate the effectiveness of RT in different HO subgroups based on etiology; iii) define optimal RT timing, technique, combinations with drugs, and dose to achieve the best therapeutic results, according to the risk categories.

Conflicts of Interest

None declared.

Authors’ Contributions

AGM and MDP had the idea for the article; EG, CG, MB, and SC performed the literature search and data collection; EG, CG, MB, and AGM drafted the article; all Authors critically revised the work.