Introduction

Heterotopic ossification (HO) is the formation of mature lamellar bone in aberrant body sites. It is a common complication after surgical trauma, especially after major hip procedures such as internal fixation of acetabular fractures and total hip arthroplasty (THA). The reported incidence of HO after THA varies between 2 and 90% [10, 20]. The pathogenesis is unclear, although surgical trauma to soft tissues or bone appears to induce the process. The most common proposed risk factors for the development of HO is ankylosing spondylitis, hypertrophic osteoarthritis, diffuse idiopathic skeletal hyperostosis, biochemical markers, male sex, age over 60, and previous formation of heterotopic bone [3, 9, 25]. Most of the patients with HO after THA remain asymptomatic and only those with advanced disease suffer from symptoms such as pain and decreased range of motion that can lead to functional problems.

Radiotherapy (RT) and non-steroidal anti-inflammatory drugs (NSAIDs) have been widely used for the prevention of HO after total hip arthroplasty. RT can be given either postoperatively within 3 days surgery or preoperatively within a few hours before surgery with various results in the reported series [1, 8, 14]. Moreover, NSAIDs, especially indomethacin, have been extensively used as well for the prevention of HO in various daily doses and durations of therapy [7, 18, 19]. However, despite the widespread use of both therapies, the type of therapy, the dose, and the duration of treatment have not yet been well established. The combination of both preventative treatment entities (radiotherapy and NSAIDs) has not been reported. This study represents the experience of a University Hospital with a combined therapeutic protocol consisting of postoperative radiotherapy and non-steroidal anti-inflammatory drugs for the prevention of HO after THA.

Materials and methods

Sixty patients with hip arthritis who underwent THA were included in the study. All eligible patients were at a high risk of the development of HO with hypertrophic arthritis, previous formation of HO, or previous surgery for removal of HO (Figs. 1, 2). All procedures were performed in the Department of Orthopaedic Surgery in the Ioannina University Hospital by the same team of surgeons and all were primary arthroplasties for osteoarthrosis of the hip or secondary arthritis due to congenital hip disease. A standard posterolateral approach was used in all patients with 1 of the 3 types of prosthetic components: cemented, hybrid or cementless. A standard closed-suction drainage was used. One drainage tube was brought out through the skin under the fascia and was removed on the second postoperative day. Postoperative pain was treated with paracetamol and pethidine. Perioperatively and postoperatively for 5 weeks all patients received prophylaxis against venous thromboembolic disease with low molecular weight heparins in doses according to the body mass index. The same antibiotic therapy that began the evening before surgery and continued till the second postoperative day was used in all patients. For the first 15 postoperative days all patients received 75 mg indomethacin for the prevention of HO. Two divided doses of 300 mg of ranitidine were administered concurrently to alleviate gastrointestinal side effects.

Fig. 1
figure 1

Radiograph of a 64-year-old female patient with hypertrophic osteoarthrosis

Full size image
Fig. 2
figure 2

Radiograph of a 66-year-old female patient with hypertrophic secondary arthritis due to congenital hip disease

Full size image

Within the first 3 postoperative days patients were admitted to the Radiation Therapy Department. Prior to RT all patients underwent simulation. Depending on body size, an individual portal of 12–14×12–14 cm was chosen to encompass all periarticular soft tissue. RT was delivered by either a linear accelerator (6 MV), or a Cobalt-60 unit with two parallel-opposed fields (anteroposterior and posteroanterior). Irradiation was given in a single fraction of 7.0 Gy to all patients.

All patients were followed up on an outpatient basis after THA. The efficacy of treatment was assessed both clinically and radiographically (for the presence of HO) by the same group of doctors in the Department of Orthopaedic Surgery 1 year after surgery and annually thereafter. The radiographic assessment of HO was based on comparison of roentgenograms performed at 1 year with those performed preoperatively and immediately postoperatively and was classified using the Brooker’s grading system [2]. The clinical evaluation was performed 1-year after the operation using the Merle d’Aubigne score [15]. Patients were also evaluated for treatment-related adverse effects and complications during the follow-up period.

The follow-up time was calculated from the day of operation. The outcomes were assessed for the whole group and according to the different types of arthritis. The group differences were tested using exact inference (Fisher’s test) or the Mann–Whitney test, as appropriate. A t test for independent samples was used to determine whether or not there was a significant difference in the mean values of the Merle d’Aubigné score. p values <0.05 were considered formally statistically significant; all p values were two-tailed. All statistical analyses were performed using the Statistical Package for Social Sciences (SPSS 11.0, Chicago, IL, USA).

Results

Patients

A total of 60 patients with arthritis of the hip treated with THA and supplementary combined postoperative RT and indomethacin for the prevention of heterotopic ossification were included. Six patients were excluded due to violation of the protocol (higher RT doses or expanded use of indomethacin). The remaining 54 patients were analyzed for clinical and radiographical evidence of HO. Out of the 54 eligible patients 30 had osteoarthrosis of the hip and 24 had secondary arthritis due to congenital hip disease. The mean age of patients was 66 years (SD: 8.13, range: 43–86 years) and the male/female ratio was 0.35 (14 males and 40 females). All the hip arthroplasties were primary operations. Twenty-six THA were right-sided, whereas 28 were left-sided. A cemented prosthesis was used in 4 patients (7%), a hybrid in 29 (54%), and a cementless in 21 patients (39%). Within the first 3 postoperative days (mean time: 1.37 days) all patients received radiotherapy. Fifty-one patients received RT with a linear accelerator and 3 with a Cobalt-60 unit. The main patient characteristics are presented in Table 1.

Table 1 Characteristics of 54 patients included in the study
Full size table

All patients had an X-ray at 1 year’s follow-up, while most of them had later radiographic evaluations as well. All cases of HO were clearly manifested at 1 year and no differences were observed between the X-rays at 1  year and those at the end of follow-up. No acute side effects due to RT or indomethacin were seen. One female patient with secondary arthritis due to congenital hip disease had pulmonary embolism in the early postoperative period. One male patient with osteoarthrosis had a peri-prosthetic fracture 2 years after surgery that was managed with internal fixation. None of the patients had aseptic loosening of the hip prosthesis and no revision arthroplasty was performed during the follow-up period. No prosthesis infection was seen and none of the patients died.

Radiographic evidence HO

Eleven patients had radiographic evidence of HO at 1 year (Table 1). The overall incidence of heterotopic ossification of any Brooker grade was 20.4% (95% CI 10.6–33.5%). Ten patients with radiographic evidence of HO were of Brooker I–II grade (18.5% [95% CI 9.3–31.4%]). One patient with severe HO (Brooker III) was seen (1.9% [95% CI 0.04–9.9%]). The radiographic evidence of HO of any grade was higher in patients with secondary arthritis due to congenital hip disease (33.3% [95% CI 15.6–55.3%]) compared with those with osteoarthrosis (10% [95% CI 2.1–26.5%]). The difference was statistically significant (p=0.046). All patients with radiographic evidence of HO were females. The difference in the incidence of HO between the female and male patients was statistically significant (p=0.048). Despite the fact that the overall incidence of HO was low there was no evidence of a statistically significant difference between right- and left-sided prostheses (p=1.00) or among the types of prosthesis.

Clinical evaluation of patients

Forty-one patients had available clinical assessment with the Merle d’Aubigné score (Table 1). The clinical score at 1 year was almost excellent for the vast majority of patients with an overall mean (±SD) score of 17.59 (±0.71). The mean (±SD) score for patients with osteoarthrosis (n=22) was 17.91 (±0.43), while the respective value for patients with secondary arthritis due to congenital hip disease (n=19) was 17.21 (±0.79). The difference between the two groups was statistically significant (p=0.001). The improvement of Merle d’Aubigné could not be presented since data on preoperative scores were not available. Patients with radiographic documentation of HO had a lower mean Merle d’Aubigné score (17.28±0.75) compared with those with no evidence of HO (17.65±0.69), but the difference was not significant (p=0.22).

Discussion

This study presents our Department’s experience of combined postoperative RT and NSAID for the prevention of HO after THA. The overall incidence of any grade HO was very low both clinically and radiologically. Only one case of clinically significant HO (Brooker grade III) was observed. No early or later side effects were observed with the combined treatment. Subgroup analyses showed that the incidence of HO was statistically significantly higher in patients with secondary arthritis due to congenital hip disease and female patients. No evidence of differences according to side of THA and type of prosthesis was observed. The clinical evaluation showed that almost all patients had an excellent Merle d’Aubigné score. Patients with radiographic evidence of HO had lower scores, although not statistically significantly.

The effectiveness of RT for the prevention of HO after major hip procedures such as total hip arthroplasty and acetabular fractures has been extensively studied with various results [6, 8, 12, 14]. However, the time of RT and the dose–response effect of RT on the incidence of HO is still a matter for investigation [5, 8, 11, 13, 21–23]. On the other hand, NSAIDs have been extensively used perioperatively for the prevention of HO after THA. Several types of NSAID in different daily doses and durations of treatment have been employed with a variety of efficacies in the reported outcomes [4, 7, 19, 24]. Several randomized trials have compared the efficacy of RT and NSAIDs [12, 16]. A large meta-analysis of randomized trials comparing RT with NSAIDs reported that RT is on average more effective than NSAIDs (apart from aspirin) and found a statistically significant dose–response relationship of RT with HO [17]. The effectiveness of postoperative RT is increased with doses exceeding 6 Gy, although RT doses higher than 8 Gy seem to be unnecessary due to the low incidence of treatment failure.

Despite the fact that the literature is dominated by a large amount of studies evaluating the efficacy of therapeutic methods for the prevention of HO after THA, our literature search did not identify any other studies that evaluated the benefit from the combined therapy of the two treatment modalities. Moreover, to our knowledge this is the first study to report the outcomes of combined therapy in secondary arthritis due to congenital hip disease and the first study to report different efficacy compared with osteoarthrosis. We identified only one prospective study with a combination of RT at 7.0 Gy and diclofenac as analgesic therapy. However, the irradiation was delivered preoperatively and the duration of diclofenac was less than 1 week at different doses individualized according to the reporting of postoperative pain [11].

The observed differences in the efficacy of the combined therapy in osteoarthritic hips and hips with secondary arthritis due to congenital disease that were seen in our trial may represent differences in the incidence of HO between the two entities. On the other hand, we should acknowledge that the present study represents our experience and therefore the results should be interpreted with caution, due to the small sample size and the lack of control groups receiving radiotherapy alone, NSAIDs alone, or no therapy at all. However, the limited number of patients with THA at a high risk of developing heterotopic ossification admitted to our hospital, along with the fact that the combined protocol is the only one used in our Department, precluded designing and establishment of a more valid and evidence-based study. In future, larger randomized trials should examine the efficacy of combined RT and NSAID therapy for the prevention of HO after THA and target the issue of the different efficacy of this therapy according to the type of arthritis.