A 78-year-old gentleman presented 2.2 years after a left total hip arthroplasty performed for osteoarthritis through a posterior approach. At the time of surgery, the physician noted an acetabular defect and placed bone graft and BMP into the acetabulum. A deep venous thrombosis mandating 6 weeks of low-molecular-weight heparin complicated his postoperative course. The patient noted a progressive decrease in motion, as well as an increase in groin pain since surgery. On exam, the left hip was fixed at 25° of flexion with a 10° adduction contracture. It was difficult to examine hip muscular strength, but the abductors and flexors did fire with attempted range of motion. There were two large bony protrusions on the lateral aspect of his hip next to the greater trochanter. He was neurovascularly intact in the lower extremities. On radiographic examination, he had Brooker type IV heterotopic ossification of the left hip (Fig. 21.4a, b). His uncemented femoral and acetabular components were well fixed.

Brooker’s [1] original classification of heterotopic ossification (HO) found that 20% of 100 consecutive patients treated with THA developed some level of HO. Eighty-four percent of those THAs were performed through a lateral approach, and the rest through a trochanteric osteotomy. The HO was classified (Table 21.1) as Brooker I in 7%, II in 5%, III in 7%, and IV in 2% [1]. Recent reviews have found that the overall incidence of HO following primary THA is from 10 to 60% depending on the risk factors studied [2–13]. Differences in technique and approach have been identified in the epidemiology of HO with a rate of HO in direct anterior THA ranging from 19 to 41.5%, posterior THA from 10 to 27.5%, and anterolateral THA of 34% [10–13]. It is important to note that the severity of the HO by Brooker classification varies widely by article since HO that forms after THA is often minor and not clinically significant [14]. Brooker III or IV is more clinically relevant than Brooker I [15]. Brooker III HO has been reported to occur in 1.4–19% and Brooker IV in 0–5% [7, 10, 11, 13, 16].

Heterotopic ossification after total hip arthroplasty can have a devastating impact on clinical function, and is often unpredictable. It is important, however, to inventory the risk factors for HO and to better understand the epidemiology of HO as reviewed above to better prevent its occurrence. Heterotopic ossification occurs when mesenchymal cells present in bone marrow, periosteum, muscle, and fascia differentiate into osteoprogenitor cells. This transformation occurs within 18 h after the index surgical procedure. An osteoid matrix is then calcified. HO consists of mature lamellar bone with trabeculae formation in soft tissue [17].

A previously ankylosed hip has been identified as perhaps the highest risk for HO, with an odds ratio of 9.85 in one study [18]. In general, patients who have post-traumatic arthritis, spinal cord injuries, diffuse idiopathic skeletal hyperostosis, ankylosing spondylitis, or had multiple operations on the hip are at higher risk [19–22]. Male gender has been identified as twice the risk for HO than female gender, but female patients over the age of 65 with osteoarthritis approach the same risk as men [23]. Rheumatoid arthritis may be protective for the development of HO [18]. There is a higher incidence of HO with a trochanteric osteotomy, lateral or anterolateral approach, previous hip surgery, sub-trochanteric femoral osteotomy, and male gender or combination of any of these factors [2, 7, 10, 22, 24, 25]. Resurfacing hip arthroplasty may have a greater risk of grades III and IV HO than patients with THA [16]. Revision THA or those with excessive bleeding may also be at higher risk [26]. Additionally, patients with secondary arthritis because of congenital hip disease had a statistically significantly higher incidence of HO compared with those with osteoarthritis [27, 28]. Approach for THA and effect on incidence of HO have been widely studied and results are extremely variable form study to study. Unequivocally, more disruption of muscle tissue or release of progenitor cells into the musculature places patients at higher risk of HO.

The complete pathogenesis of HO is unknown, but surgical trauma to soft tissue or bone appears to induce the process. Current prophylactic measures generally adhere to one or more of the following three principles: disrupting the relevant inductive signaling pathways, altering the relevant osteoprogenitor cells in the target tissue, or modifying the environment conducive to heterotopic osteogenesis [29].

There are a multitude of surgery-related factors that may modify the environment regarding heterotopic osteogenesis. Incision length, approach [6, 9, 11], localized tissue trauma and muscle damage [30] and ischemia [31], blood loss, anesthetic type, and length of surgery may all contribute to the local inflammatory response. Pulsed lavage may also spread osteoblast precursors, thereby creating an osteoconductive environment [11]. Smaller incisions may reduce the “zone of injury” to the skin and underlying soft tissues and subsequent risk of HO formation. However, if the smaller incision leads to a more difficult procedure, and more muscle damage, HO may be increased. Common intraoperative principles to avoid HO include the suctioning of marrow contents during femoral broaching, irrigating any bone dust on muscle tissue, and avoiding any muscular dissection that is not necessary. These principles again are based on limiting the spread of osteoblast precursors.

Several studies suggest that nonsteroidal anti-inflammatory drugs (NSAIDS) are efficacious in preventing HO after THA [32–39]. Indomethacin is one of the most commonly used agents and inhibits PGE2 via COX-1 downregulation and osteoprogenitor cell differentiations to osteoblasts. A typical treatment dose of 25 mg three times a day or 75 mg once a day has been shown to be efficacious if given directly after surgery [40]. More recently, naproxen 500 mg twice daily has been shown in a prospective double-blind placebo-controlled trial to lower rates of HO after hip arthroscopy from 46 to 4% [41].

A number of randomized trials have compared nonselective NSAIDS to selective NSAIDS. In a study by Saudan et al. [42], patients were allocated to receive either ibuprofen 400 mg three times daily or celecoxib 200 mg twice daily. There was an incidence of Brooker II and III HO in 13% in the ibuprofen group, and 5.1% in the celecoxib group [34]. However, in systematic reviews [37, 43], there have been no differences identified in the selective versus nonselective NSAIDS in the prevention of HO. Yet selective NSAIDs have enhanced compliance due to gastrointestinal side effects compared with the nonselective NSAIDs [37]. The beneficial action of NSAIDS for prophylaxis against heterotopic ossification is attributed to the inhibition of cyclooxygenase 2 (COX-2) enzyme, an inducible enzyme in the osteoblasts. COX-2 is the enzyme that catalyzes the first reaction of arachidonic acid toward prostaglandin formation. The increased concentration of prostaglandins, especially PGE2, results in new bone matrix production and thus in heterotopic ossification formation [44].

Utilizing NSAIDS to prevent HO has been presented as an adjunctive medication for the sole purpose of prevention of HO. However, patients who received aspirin for DVT prophylaxis compared to warfarin experienced a decreased incidence and severity of HO [34, 39, 45]. Cohn et al. demonstrated a rate of 0% Grade III or IV HO in a cohort of 35 hips after THA treated with aspirin for DVT prophylaxis. The dosage that has been studied in multiple studies is aspirin 325 mg twice daily for 6 weeks [39].

Prevention with perioperative radiation treatment (RT) for prophylaxis of HO has been widely studied [20, 46–49]. RT is attractive as compliance is 100%, there is a low side effect profile, and it avoids the possible gastrointestinal complications of NSAIDs. Since the postoperative treatment may be uncomfortable for patients, our center prefers immediate preoperative treatment. Historically, treatment included 2000 cGy over 10 days. However, recent studies have demonstrated efficacy at a single dose of 500 cGy [48]. Ionizing radiation appears to interfere with the processing of nuclear DNA during cell division and inhibits transformation of pluripotential mesenchymal cells [48]. RT also influences the cellular responsiveness to BMP-2-signaled osteoblast differentiation [50]. Because treatment is directed at a local area, adverse reactions are limited to this region, mitigating systemic reactions [51, 52]. Cost is much higher than NSAID treatment, provided that there are no complications due to NSAID use. In an analysis of patients who were diagnosed with a malignancy after hip replacement, there was a 4% rate of malignancy in a cohort of 238 patients who had RT, compared with a 7% rate of malignancy in a control group of 476 patients who did not have RT. No patients in the radiation group had a malignancy in the field of radiation, demonstrating reasonable safety of RT for HO prophylaxis [52].

The diagnosis of HO is made at postoperative visits both by physical exam and radiographic evaluation. Range of motion may be significantly reduced or eliminated altogether. Patients often complain of achy pain that has been progressive with concurrent decrease in motion. Decreased range of motion correlates with higher grades of HO [15]. Occasionally, contractures in flexion and adduction cause additional difficulties for activities of daily living. Primary classification is made by the system of Brooker [1]. Once the initial diagnosis of HO has been made, a CT scan may be obtained to provide the clinician a more detailed view of the locations and extent of the HO, and its relationship to anatomical structures including muscle and the sciatic nerve—especially if operative intervention is planned. If there is concern that the HO may still be in an immature state, a bone scan may help the clinician by showing little activity in the bone at a safe time to remove.

Range-of-motion limitations due to HO itself will not respond to physical therapy. However, gentle stretching and strengthening may prevent further stiffness, and encourage improved gait, and optimal positioning and function of the limb. NSAIDS may be helpful on a more chronic basis to reduce inflammation associated with impingement due to the excessive bone. Generally, after 1 year [53] the bone is considered to be matured, and at that time if the decreased range of motion and pain are bad enough, excision may be contemplated. Pain as the only reason for excision may be a contraindication for operative management. Heterotopic bone usually does not cause pain after maturation is complete [53]. However, the perception may continue and lead to a chronic pain syndrome that removal cannot fully resolve. In one study, patients who had HO excised for pain only, none had complete resolution of symptoms [54].

A careful preoperative plan is needed. Prior operative reports will help the surgeon know the prior surgical approach and any complications or issues noted in the initial procedure. The surgeon may have to perform an approach that they are not familiar to due to the prior approach, or location of HO. Implant design and manufacturer should be noted as in many cases implant exchange is necessary to gain access, or to enhance stability. Constraint or dual-mobility constructs may be considered as HO oftentimes includes much of the abductor muscle mass with a consequent high risk of postoperative instability.

Radiographs or CT scan should be reviewed to again determine the location of the bone relative to anatomical structures. Specifically, if the HO is near the sciatic nerve, preparation to isolate the sciatic nerve and perhaps perform a neurolysis should be made. Prior to revision for HO excision, perioperative radiation therapy, and postoperative NSAID treatment should be considered to give the best chance for non-recurrence of the HO.

Intraoperatively, painstaking care must be taken to identify the HO, and carefully remove it from native tissues. A good technical tip is to find the border of normal tissue and progress to the joint along the abnormal tissue, removing HO off of normal bone until the joint is reached. Often an osteotome or a Cobb elevator can aid in pushing the tissue off of the HO. Typically the HO has to be removed in piecemeal fashion to continue to gain access. Bleeding may be encountered if the HO invades normal tissue, and preparations should be made for increased blood loss. Autogenous blood recycling may be considered if available. As mentioned above, care should be taken around the neurovascular structures. The surgeon must be aware that these structures may be adjacent to, or perhaps in the middle of, the HO. Again, careful review of preoperative imaging will assist. Neurosurgical consultation could be made if the surgeon is not comfortable with neurolysis. If appropriate preparations are made, and careful and efficient surgical technique is employed, safe and effective removal of HO will be achieved.

In cases of significant HO formation after THA, where pain and limitation of function are significant, excision may be entertained as described above. Delaying excision for at least 1 year until the bone has matured and formed a stable fibrous capsule will improve results and allow the bone to be removed [53]. A bone scan revealing decreased activity may also aid in the timing of surgery [54]. Alkaline phosphatase levels are often elevated with immature HO, and return to normal that may indicate maturing of the HO [53]. Attempting removal before this time has been shown to have inferior results, and more difficult removal. However, there is significant debate that the bone may be removed earlier with perioperative radiation as soon as the HO is clinically important [55]. After excision of HO, patients may expect an average of an increase in flexion range of motion (ROM) of 30°–40°, abduction–adduction of 20°–30°, and internal–external rotation of 20°–30° [54, 56]. Although not well studied, it is important to note that pain is generally improved for most patients, but in modest amounts for many patients [54].

Complications of HO excision follow many of the normal complications of revision total hip arthroplasty. Extensive dissection may lead to sciatic nerve direct or stretch injury leading to long-standing neuropraxia. When removing HO from the undersurface of the abductor musculature there is a risk of damaging the superior gluteal neurovascular bundle which may lead to further defects in the abductors. In cases of significant ankyloses, osteotomy of the femur may be required to lift up the abductors and gain access to the joint.

Often large amounts of bone and tissue, up to 1 L, may be removed during excision of HO. This may lead to cases of instability due to a large dead space, or dislocation due to impingement if bone is selectively removed in one range of motion. Component position and range of motion before impingement must be inspected closely. A low threshold for dual-mobility or constrained constructs may be entertained in those patients that do exhibit instability intraoperatively.

The results of radiation after HO excision are encouraging. RT should be given either right before or the day after excision, but may be effective out to 3 days after excision [57, 58]. Using 500–2000 cGy in one dose has been shown to be safe and effective in prevention of recurrence of HO after excision in the setting of THA [48, 54, 58–60]. Recurrence rates of HO after RT are 5–15% clinically significant HO [58, 59]. Concerns remain for fixation of newly implanted uncemented implants, and shielding of these implants may be indicated during RT to prevent lack of bony ingrowth [58].