Is etoricoxib effective in preventing heterotopic ossification after primary total hip arthroplasty?

Abstract

Purpose

Heterotopic ossification is a common complication after total hip arthroplasty. Non-steroidal anti-inflammatory drugs (NSAIDs) are known to prevent heterotopic ossifications effectively, however gastrointestinal complaints are reported frequently. In this study, we investigated whether etoricoxib, a selective cyclo-oxygenase-2 (COX-2) inhibitor that produces fewer gastrointestinal side effects, is an effective alternative for the prevention of heterotopic ossification.

Methods

We investigated the effectiveness of oral etoricoxib 90 mg for seven days in a prospective two-stage study design for phase-2 clinical trials in a small sample of patients (n = 42). A cemented primary total hip arthroplasty was implanted for osteoarthritis. Six months after surgery, heterotopic ossification was determined on anteroposterior pelvic radiographs using the Brooker classification.

Results

No heterotopic ossification was found in 62 % of the patients that took etoricoxib; 31 % of the patients had Brooker grade 1 and 7 % Brooker grade 2 ossification.

Conclusions

Etoricoxib seems effective in preventing heterotopic ossification after total hip arthroplasty. This finding further supports the use of COX-2 inhibitors for the prevention of heterotopic ossification following total hip arthroplasty.

Introduction

A common complication following total hip arthroplasty (THA) is the development of heterotopic bone ossification (HO). The incidence of HO after total hip and acetabular fracture repair varies from 43 % to 51 % [1]. The development of HO hampers the rehabilitation process and gives rise to pain and major functional impairment of the hip joint.

To prevent HO, low dose radiation therapy [2] and non-steroidal anti-inflammatory drugs (NSAID), including indomethacin, ibuprofen, tenoxicam, naproxen, flurbiprofen, ketorolac, and diclofenac are proven to be effective [3, 4]. However, both treatment options have disadvantages. For radiation therapy, the potential risk of cancer, infertility, transportation of patients to the radiation department, and its associated higher costs have been suggested [5, 6]. When using NSAIDs, prolonged bleeding time, gastrointestinal side effects, and an increase in non-union of associated fractures have been observed [7]. In addition, in some studies up to 37 % of the patients that used NSAIDs had to cease these medications because of serious side effects [8]. Therefore, the best practice to prevent HO is still under debate [6].

Since there is still a need for potential effective medication to prevent HO (which coincides with less adverse events), we investigated whether etoricoxib (Arcoxia®, MSD), a selective cyclo-oxygenase-2 (COX-2) inhibitor, is effective in preventing HO after THA. This selective COX-2 inhibitor is associated with significantly fewer gastrointestinal side effects [9] and is therefore recommended [10]. In a prospective two-stage study design for phase 2 clinical trials, we investigated the efficacy of etoricoxib 90 mg once daily oral dose for seven days in a small sample of patients. The heterotopic bone formation was assessed on antero-posterior radiographs using the Brooker classification [11]. We postulated that etoricoxib 90 mg once daily is equally effective as the non-selective COX-2 inhibitor indomethacin in preventing heterotopic ossifications in patients undergoing total hip surgery.

Material and methods

After approval from the Dutch authorities and the Institutional Ethical Review Board, 42 patients (aged 19–83 years) with osteoarthritis undergoing elective primary single hip arthroplasty were consented. All patients were recruited at the Department of Orthopaedics, Radboud University Nijmegen Medical Centre, the Netherlands. The clinical trial is registered at EudraCT (#2009-013161-26) and at ClinicalTrials.gov (#NCT01022190). The study was conducted according to the Declaration of Helsinki on biomedical research involving human subjects.

Two-stage study design

We used a two-stage study design for phase two clinical trials [12] in this study. In this design, a small group of patients was exposed to the experimental drug and only if the effect in this small group was effective, was the study group expanded. This type of study design was previously used successfully by van der Heide et al. [13]. By exposing only a small group of patients, they studied the efficacy of rofecoxib (a COX-2 in inhibitor) to prevent HO. These positive results were confirmed later in a randomised controlled trial using a large patient cohort [14]. For calculation of the number of subjects needed in this two-stage study design, the data of two historical patient groups with total hip arthroplasty from our department were used [15]. In the past, one group did not receive any prophylaxis and the second group received seven days of indomethacin (Table 1). In the group without prophylaxis, 29 % developed grade 3 or 4 HO and 71 % did not. In the indomethacin group, only 2 % developed grade 3 or 4 HO and 98 % did not. Based on these data, a treatment of seven days with a COX-2 inhibitor Arcoxia® (etoricoxib, MSD) was considered sufficient if at least 80 % of the cases had a Brooker classification 0, 1, or 2. In addition, a drug which is very effective and shows less than five percent grade 3 or 4 HO needs to be detected with a high probability given the results of the historical treatment group. In the two-stage Simon design [12], with an actual effectiveness of less than 80 %, the medication (etoricoxib) is not to be recommended (accept the corresponding hypothesis with probability [1 – α] = 0.95). An actual effectiveness of at least 95 % needs to be detected with a high probability (≥ 90 %). When entering these data in the table in the article by Simon [12], or using his software, the number of patients needed was calculated. In the first stage, 19 cases were necessary. Six months after surgery, antero-posterior radiographs were scored using the Brooker classification (Table 2). When two patients with a Brooker classification 3 or 4 were detected in this first cohort, the hypothesis was seen to be wrong and the study did not proceed to the second stage. However, if the criteria of the first stage were fulfilled, the second stage was started and completed with another 23 patients. Treatment was declared promising if at least 38 patients with a Brooker classification 0, 1, or 2 were found in 42 patients.

Table 1.

Heterotopic ossification after primary cemented total hip arthroplasty in two historical patient groups (no prophylaxis and indomethacin) used for calculation of the two-phase study design and the incidence in the group which received etoricoxib

Brooker grade	No prophylaxis (n = 170)	Indomethacin 7 days (n = 99)	Arcoxia 7 days (n = 42)
0	40 (24 %)	72 (73 %)	26 (62 %)
1	33 (19 %)	24 (24 %)	13 (31 %)
2	48 (28 %)	1 (1 %)	3 (7 %)
3	29 (17 %)	1 (1 %)	–
4	20 (12 %)	1 (1 %)	–

Table 2.

Classification of heterotopic ossification by Brooker [11]

Brooker grade	Bone formation
I	Isolated bone islands
II	Bone spurs from the pelvis or proximal femur; space between opposing surface ≥ 1 cm
III	Bone spurs from the pelvis or proximal femur; space between opposing surface < 1 cm
IV	Apparent bony ankylosis

Patients

Patients were included in the study if they were listed for a cemented primary total hip arthroplasty at our department. Exclusion criteria were rheumatoid arthritis, ankylosing spondylitis, femoral neck fractures, previous allergic reaction on anti-inflammatory drugs, a history of gastrointestinal ulcers or perforations, inflammatory bowel disease, hepatic dysfunction, renal dysfunction with a clearance below 30 ml/min and cardiac insufficiency. All patients that entered the study gave their written consent and the study was approved by the local ethical committee of the University hospital.

Pain medication

The administration of NSAID was ceased ten days before surgery; if necessary, acetaminophen (paracetamol) and/or tramadol were allowed to relieve pain. The morning before surgery, patients received their first dose of etoricoxib two hours before surgery [16]. Patients were administered 90 mg etoricoxib orally, and continued until the sixth postoperative day; thus, the trial period was seven days in total. Patients were not allowed to use any other NSAIDS, except the study medication, during the first six weeks after surgery. If pain had to be relieved, patients were allowed to take acetaminophen or tramadol. If patients used medication because of gastrointestinal complaints prior to surgery, they were allowed to continue these medications during the trial period. Patients received pantoprazol 40 mg a day if severe gastrointestinal complaints occurred during the first seven days after operation.

Operation

In all patients, a posterolateral approach without trochanteric osteotomy was used and a cemented total hip prosthesis (Exeter total hip prosthesis, Stryker Howmedica, Newbury, UK) was inserted. In 30 patients (71 %), the primary total hip arthroplasty was combined with bone impaction grafting. The patients received antibiotic prophylaxis at the start of the operation (cefazoline, 2 g intravenously). Antithrombotic prophylaxis was established by low molecular-heparin (nadroparin 5,500 IE a day, subcutaneous) during the first six weeks after the operation, starting on the evening before operation. Suction wound drainage was used for two days. Two days after surgery, full-weight bearing was allowed, using two elbow crutches.

Clinical examination

Clinical examinations were performed the day before surgery, at six weeks and six months postoperatively. Roentgenographic examinations were routinely performed immediately after operation, at six weeks and six months postoperatively. Conventional antero-posterior pelvic radiographs were assessed; the classification of heterotopic ossification by Brooker [11] was used (Table 1). The radiographs were evaluated by an experienced radiologist. To prevent a possible favourable influence of the radiologist (since the radiologist might know that all radiographs were part of the study), all radiographs of the patients of the study group were mixed with radiographs of a group of control patients. This group of control patients were treated in the same year as the patients, using the same operation technique and implants. A random sample of these controls was collected using the statistical program SPSS 18.0 (SPSS Inc., Chicago, IL, USA). All radiographs were coded and the names and birth dates of the patients were blinded.

Results

Nineteen patients were included (14 women, five men) for the first stage of the study. The indication for the hip prosthesis was primary or secondary osteoarthritis of the hip joint. During the second stage of the study, 23 patients were included (11 women, 12 men); no patients were lost to follow-up. The average age of the whole group of patients at time of surgery was 54 years (range 19–83).

All 19 patients who were included for the first stage of the two-stage design study completed the seven days protocol for etoricoxib and were available for follow-up. After six months, the radiographs were analysed for heterotopic ossification: 13 patients had a Brooker grade 0; four patients a Brooker grade 1; and two patients had a Brooker grade 2. The criteria to enter the second stage were fulfilled (no more than two patients with Brooker 3 or 4) and the second phase was started.

Another 23 patients were included in the study and completed the seven-day protocol. After six months, the radiographs of the patients of the second stage were analysed for heterotopic ossification: 13 patients had a Brooker grade 0; nine patients a Brooker grade 1; and one patient a Brooker grade 2. At the end of the study, the results of the patients of the first and second stage were combined. No heterotopic ossification was found in 26 of the 42 patients (62 %). Brooker grade 1 ossification was found in 13 (31 %), and 3 (7 %) displayed a Brooker grade 2 ossification. No widespread ossification (Brooker grade 3 and 4) was seen in patients that used Arcoxia. The data of the etoricoxib group was entered in the two-stage study design. The probability of an inhibitory effect of etoricoxib on severe heterotopic bone formation was calculated to be 100 % (95 % confidence interval 92–100 %) (Table 1).

Complications

Three patients of the etoricoxib group had a postoperative complication. One patient had a urinary tract infection, which was treated with antibiotics. A second patient had a persistent wound leakage and a pulmonary embolus, which was treated with anti-coagulation. Another patient had a peri-operative perforation of the femur shaft, which was corrected during the surgery and had no effect on the after treatment protocol. No re-operations were necessary. These three complications were not related to the use of etoricoxib.

Discussion

The main finding of our study is that etoricoxib, a COX-2-selective inhibitor, in a dose of 90 mg once daily for seven days, was effective in preventing heterotopic ossification after cemented primary total hip arthroplasty. Since etoricoxib is claimed to be associated with fewer gastrointestinal site effects than non-selective NSAIDs [9], etoricoxib is a promising treatment option to prevent heterotopic ossification. Our results further support the use of COX-2 inhibitors in the prevention of heterotopic ossifications after total hip arthroplasty.

Meta-analyses on randomised clinical trials on the efficacy of selective COX-2 inhibitors and non-selective NSAIDS found both medications equally effective in the prevention of HO after THA [10]. Considering the side effects of non-selective NSAIDs, selective COX-2 inhibitor was recommended as the preferable medication for the prevention of HO after THA [9, 10, 17]. The results of our study supports the use of the COX-2 inhibitor etoricoxib.

In a previous study that was performed by our research group, we found Rofecoxib, a selective COX-2 inhibitor, equally effective as indomethacin to prevent HO after THA [13, 14]. However, serious concerns were raised over the safety of this selective COX-2 inhibitor for the cardiovascular system. Long-term use (over three years) was associated with an increased cardiovascular risk [18]. Therefore, rofecoxib (marketed under the brand names Vioxx®, Ceoxx®, and Ceeoxx®) was withdrawn voluntarily from the market by Merck on September 2004. However, meta-analyses indicates that all NSAIDs (selective or not) have a comparable cardiovascular safety risk (except for naproxen) [19].

A possible limitation of our study is the small sample size. However, due to ethical considerations, a two phase model is necessary to indicate whether etoricoxib is effective in prevention of HO. Future research is, therefore, warranted to further investigate the effectivity of etoricoxib in a larger cohort of patients. This research should preferably be done in a randomised controlled setting in which etoricoxib is compared to a non-selective NSAID, for example, indomethacin.

An advantage of COX-2 inhibitors compared with non-selective NSAIDS is that COX-2 inhibitors can reduce perioperative blood loss. In a study conducted by Weber et al. [20], it was found that COX-2 inhibitors reduce the perioperative blood loss. They found that total perioperative blood loss after meloxicam was significantly less than observed after indomethacin in patients that underwent total hip arthroplasty. These results indicate that the COX-2 class drugs, such as etoricoxib, which do not interfere with platelet function, may be effective in reducing blood loss during orthopaedic surgery.

In conclusion, this is the first study indicating that a seven-day treatment with etoricoxib 90 mg once daily (a COX-2 inhibitor) is a promising intervention to prevent the formation of heterotopic bone after cemented primary total hip arthroplasty. These promising results should be studied in a larger trial, to further elucidate the beneficial effects of etoricoxib in the prevention of heterotopic bone ossification after total hip arthroplasty.