Heterotopic Ossification

Key Points

  • While the etiology of heterotopic ossification (HO) is diverse, it most commonly occurs through a process of endochondral ossification after operative procedures about the hip.

  • While HO is typically only a radiographic finding without clinical consequences, clinically meaningful stiffness may result from extensive ectopic bone formation in as many as 10% of patients after total hip arthroplasty (THA).

  • Patients at greatest risk for HO are those who have made heterotopic bone after prior hip surgery, have hypertrophic osteoarthritis, ankylosing spondylitis, diffuse idiopathic skeletal hyperostosis, Parkinson disease, or are of male gender.

  • Effective prophylaxis of high-risk patients with nonsteroidal antiinflammatory agents or external beam radiation must be instituted within 5 days of operation to avoid HO-related compromise of function after THA.

  • In patients who cannot receive radiation prophylaxis, indomethacin is the drug of choice for preventing the development of HO. Cyclooxygenase-2 (COX-2) inhibitors may be considered as second-line agents if there are contraindications to indomethacin.

  • When surgical excision of HO is necessary to restore function, recurrence is most likely unless prophylaxis is administered in the early postoperative period.


Heterotopic ossification (HO) is the process by which aberrant bone is formed in the soft connective tissues, joint capsule, or skeletal muscle outside of the normal confines of the skeleton. There are several etiologies for HO, which are generally grouped as neurologic, genetic, and traumatic, with orthopedic procedures included in the latter. This pathologic process most often occurs about the hip and is typically incited by a variety of surgical procedures, such as total hip arthroplasty (THA), repair of acetabular fractures, pelvic osteotomy, intramedullary nailing of femoral shaft fractures, and even hip arthroscopy with femoral or acetabuloplasty. HO also occurs about other joints, such as the shoulder, elbow, and knee. It can also appear in conjunction with other nonsurgical events, such as head injury, cerebrovascular accident, burns, and genetic disorders such as fibrodysplasia ossificans progressive (FOP). Most recently, related to armed conflicts and the use of improvised explosive devices (IEDs), HO has appeared in the residual limbs of up to two-thirds of combat-related amputees after blast injury. Nonetheless, it is clinically most often manifest as ectopic bone formation in the abductor musculature after THA. Factors intrinsic to the patient, injury to the abductor musculature, surgical approach, hemorrhage, and bone debris remaining in the soft tissues all contribute to the overall risk of HO in any individual patient.

As a radiographic phenomenon without accompanying symptoms, HO has been reported to occur in as many as 90% of patients who have undergone THA. In this setting, it is manifest as small islands of bone in the soft tissues about the hip or as bony outgrowths from the margin of the acetabulum or the proximal femur, measuring less than 1 cm in length. However, in up to 10% of patients, ectopic bone formation THA occurs to such an extent or in such a location that it results in clinical manifestations of pain soon after operation and late restriction of motion that are of consequence to the patient. In the most extreme cases, the resulting symptoms of pain and stiffness replace those that prompted the patient to seek surgery in the first place and may actually negate the beneficial effect of the arthroplasty.

Epidemiology and Risk Factors

Patients judged to be at increased risk for making heterotopic bone should receive some form of prophylaxis for HO; this constitutes less than 20% of those undergoing THA in the author's practice. Classically, these high-risk patients include those with a diagnosis of hypertrophic osteoarthritis and prominent marginal osteophytes ( Fig. 112.1 ), diffuse idiopathic skeletal hyperostosis, ankylosing spondylitis, and those having formed heterotopic bone after a previous procedure or injury about the hip. Nearly all patients who have formed heterotopic bone after undergoing surgery about the hip will develop heterotopic bone again, usually to a greater degree, after reoperation on the involved hip or primary operation on the contralateral hip. Men are twice as likely as women to form heterotopic bone, but women in these high-risk categories develop ectopic bone at rates comparable with those of men. It is increasingly recognized that central nervous system disorders—specifically Parkinson disease ( Fig. 112.2 ), spinal cord injury, and perioperative cerebrovascular accident—also substantially predispose patients to HO about the hip.

Fig. 112.1, (A) Radiograph of a 66-year-old male patient with primary osteoarthritis and diffuse hypertrophic skeletal hyperostosis. The “whiskering” of periosteal new bone about the ischial tuberosity and iliac crest is evident. Stigmata of enthesopathy are seen as calcification in the tendinous insertion of the iliopsoas on the lesser trochanter. (B) Postoperative radiograph after bilateral total hip arthroplasty, staged 1 year apart. Grade III heterotopic ossification (HO) is evident on the right after the initial hip replacement. No HO developed on the left following postoperative radiation prophylaxis administered after total hip arthroplasty.

Fig. 112.2, Radiograph of a 72-year-old male patient with a history of Parkinson disease following total hip replacement without radiation prophylaxis. Grade IV heterotopic ossification is apparent and the patient's hip was clinically ankylosed. He ultimately came to surgical resection and received adjunctive postoperative radiation to an expanded field with no recurrence of ossification and a good functional result.

Risk factors for developing HO following THA are as follows:

  • Prior heterotopic bone formation after hip surgery or trauma

  • Hypertrophic osteoarthritis with prominent osteophytosis

  • Diffuse idiopathic skeletal hyperostosis (DISH)

  • Ankylosing spondylitis

  • Male gender

  • Central nervous system disorders: Parkinson disease, perioperative stroke, traumatic brain injury


In 1975, Chalmers et al. postulated that 3 requisite conditions must be met to allow heterotopic bone formation: (1) the presence of an inducing agent, (2) an osteogenic precursor cell, and (3) an environment conducive to osteogenesis. The etiologies of heterotopic bone formation are generally classified as neurologic, genetic, and traumatic. Over the past decade, much research has been done to identify the osteogenic precursor cell as well as the molecular signals that stimulate pluripotent mesenchymal stem cell differentiation down osteogenic cell lines ( Fig. 112.3 ). While the specific osteogenic precursor remains unknown, much progress has been made in understanding the inductive role of bone morphogenetic protein (BMP) signaling.

Fig. 112.3, Basic schematic of the pathophysiology of heterotopic ossification and the points of prophylactic intervention along the pathway. BMP, bone morphogenetic protein; NSAIDs, nonsteroidal anti-inflammatory drugs.

Several bone morphogenic proteins exist, and “bone dust” (particulate bone fragments resulting from reaming or machining of bone with power instruments) recovered from patients who formed heterotopic bone has been shown to stimulate proliferation of isolated bone progenitor cells in culture 6-fold more than similar material extracted from patients who did not form bone. It is suggested that the induction of pro-osteogenic pathways occurs through a dysregulation of BMP signaling in the presence of inflammatory triggers. In the local environment, BMPs are overexpressed and, in combination with an underexpression of antagonists, a powerful morphogenetic gradient is created. Inflammatory factors such as interleukin 1β and prostaglandin E1 and E2 have been shown to enhance the expression of BMPs and ultimately the appearance of heterotopic bone. The role of these inflammatory signals is further supported by the known inhibitory effects of nonsteroidal antiinflammatory drugs (NSAIDs) on heterotopic ossification. Attesting to the metabolic hyperactivity of this tissue, histomorphometric and biochemical data have shown that heterotopic bone contains more than twice the number of active osteoclasts and has a rate of appositional new bone formation nearly 3 times that of normal age-matched bone.

Central to the issue of HO is the question of the anatomic origin of the pluripotential mesenchymal cells participating in the process of bone formation within the soft tissues. Transformation of local cells into bone-producing elements has been proposed by many investigators. Moreover, the demonstrated efficacy of preoperative radiation in preventing HO provides strong circumstantial evidence in support of a local source of osteoprogenitor cells resulting in ectopic ossification. Trauma to muscle leads to hemorrhage, muscle degeneration, and proliferation of perivascular connective tissue, culminating in the production of heterotopic bone. Studies in an animal model using hydrogen-3-labeled thymidine and uridine have shown that local soft tissue fibroblasts adjacent to an implanted demineralized bone fragment were induced to transform into pluripotential mesenchymal cells that differentiated into osteoblasts. Pluripotential mesenchymal stem cells are ubiquitous in the soft tissues about the hip; it is postulated that these cell lines may be induced to undergo atypical differentiation into osteogenic stem cells capable of participating in the process of HO. Maturation of these cells down either osteoblastic or chondroblastic stem cell lines could then lead to the formation of ectopic bone. Increasing evidence suggests that the pathologic process of HO proceeds through a pathway of secondary bone formation, first passing through an intermediate phase of cartilage model bone, which then undergoes ossification to mature lamellar bone ( Fig. 112.4 ).

Fig. 112.4, Histologic section of heterotopic bone from a rabbit animal model 12 weeks after hip surgery. Periosteal new bone is appearing from the cortical surface. Chondroid tissue is giving rise to islands of lamellar bone through a process of calcification of hypertrophic cartilage, reminiscent of the growth plate.

Several groups have identified stem cells with osteogenic capacities. One candidate is a muscle-derived stem cell isolated from skeletal muscle that has been shown to commit to an osteogenic lineage in response to BMP2 and BMP4. Another potential source is a vascular endothelial precursor cell known to respond to an inflammatory trigger that forms heterotopic bone in response to overactive BMP signaling. Consistent with this observation, inhibiting the migration of endothelial-derived mesenchymal stem cells has been shown to decrease the formation of HO in response to BMP exposure. Yet another group has recently identified mesenchymal progenitor cells (MPCs) obtained from the traumatized muscle of combatants with extensive soft tissue extremity wounds. These cells function as osteoprogenitor cells in the appropriate biochemical environment. It is conceivable that several cell populations have the potential to form bone following the proper signals.

Alternatively, it is possible that distant migratory hematopoietic stem cells may be essential to inducing local connective tissue elements to form heterotopic bone. Local pluripotential mesenchymal cells and osteocytes are liberated from the marrow space of the ilium and femoral canal during hip arthroplasty surgery and are likely present in the local hematoma. They may contribute to the process of HO either directly by bone formation or indirectly via stimulation of local cells to express osteogenic phenotypes. Distant hematopoietic stem cells, known as circulating osteoprogenitor cells, transported to the wound by virtue of the normal response to surgical injury also possess the capability to mature along osteogenic cell lines under the influence of mitogenic stimuli in the wound environment. These circulating osteogenic precursors may play an important role in ectopic bone formation in locations remote from any direct injury, such as seen after closed head injury or burns.

Regardless of the site of origin of these pluripotential mesenchymal stem cells, heterotopic ossification seems to be clearly dependent on cellular differentiation down osteoprogenitor cell lines. Ionizing radiation is known to exert its greatest influence on rapidly dividing cells by interfering with the normal production of nuclear deoxyribonucleic acid. One study has shown that ionizing radiation in doses from 0 to 20 Gy reduces the formation of the BMP2/BMP receptor complex in a dose-dependent fashion. The authors of this study postulate that ionizing radiation acts to downregulate the BMP receptor. Tonna and Cronkite demonstrated in mice that differentiation of pluripotential mesenchymal cells into osteoblasts began 16 hours after fracture of the femur and peaked at approximately 32 hours. In their model, the critical events of cellular differentiation occurred during the immediate postoperative period. A similar chronology may be extrapolated to the sequence of HO even though the actual ectopic bone is not detectable radiographically for several weeks after surgery. Therefore, to be most effective, it seems essential that irradiation or other prophylactic measures be administered early during the postoperative period to prevent osteoblastic differentiation of pluripotential mesenchymal stem cells, effectively arresting osteoid and subsequent heterotopic bone formation during the initial phases of cellular reorganization. Experience with postoperative radiation therapy for prevention of HO has empirically defined a window of 4 to 5 days for the effective institution of external beam irradiation after operation about the hip.

Tissue environment has been shown to influence the development of HO in vivo. One study has demonstrated that BMP-9, a strong inducer of osteogenesis in vitro, is insufficient to induce heterotopic ossification in the absence of muscle injury. Furthermore, a study comparing the effects of muscle, bone, and combination injury has shown the greatest increase in osteogenic potential in animals subjected to both muscle and bone injury. This is consistent with clinical observations that increased rates of HO are observed in the setting of more extensive bone and soft tissue trauma.

Fibrodysplasia Ossificans Progressiva Genetics

Fibrodysplasia ossificans progressiva (FOP) is a heritable disorder of connective tissue characterized by congenital malformation of the great toes and postnatal formation of HO. Over the past decade, much progress has been made in understanding the pathophysiology of FOP. Heterotopic bone in FOP undergoes an endochondral process of ossification similar to HO following THA and, although the etiology of HO in FOP is different than that following THA, study of this rare genetic disorder may provide some insight into the condition following surgery about the hip. It was first discovered that BMP4 messenger ribonucleic acid and protein are specifically overexpressed in FOP. Further investigation revealed that FOP cells have a defect in the BMP-antagonist response. Whereas a normal response to BMP4 activation onto its receptor causes upregulation of BMP antagonists such as Noggin, FOP cells showed a dramatically attenuated response to the BMP antagonist and consequently an accumulation of much higher levels of BMP4. These cells also have significantly higher levels of BMPR1A cell surface receptors, marked reduction of internalization and degradation of receptors, and a dysregulation of downstream BMP signaling and hyperresponsiveness to BMP. Most recently, a mutation in the activation domain of the activin A type 1 receptor (ACVR1), a BMP type 1 receptor, has been shown to enhance receptor signaling.

Radiographic Staging

A radiographic classification system popularized by Brooker et al. is most commonly used to describe the pattern and extent of ossification on the anteroposterior pelvic radiograph: stage I, islands of bone appearing in the soft tissues; stage II, bone spurs arising from the pelvis or proximal femur with greater than 1 cm between adjacent bone surfaces; stage III, bone spurs arising from the pelvis or proximal femur with less than 1 cm between adjacent bone surfaces; and stage IV, confluent bone bridging the pelvis and proximal femur and apparent bony ankylosis of the hip. It has been shown, however, that some patients with apparent stage IV heterotopic ossification may have near normal clinical range of motion (ROM). Furthermore, while the Brooker grading system has been shown to correlate with hip ROM, it does not correlate with outcome measurements such as the Harris hip score. This has led to consideration of combined use of both anteroposterior and lateral radiographs to allow for more accurate grading and better correlation with clinical findings.

Further description of the radiographic extent of ossification correlates with the degree of functional impairment attributed to the ectopic bone formation. One grading system describes the proportion of the area involved in the triangle defined by the base of the greater trochanter, the anterior iliac spine, and the inferior aspect of the ischium. Grade A bone involves less than or equal to 33% of this area, grade B involves 34% to 66%, and grade C involves 67% to 100%. Greater extent of involvement correlated with more clinically significant limitation of motion about the hip.

Clinical Presentation

Vague discomfort or frank pain in the hip region is thought to be the direct result of the inflammatory process that is involved in the process of HO. It is characteristically present at rest, is unaffected by activity, and often interferes with sleep. This typically occurs during the first 6 months after surgery and spontaneously resolves as the biologic activity of the process subsides and the radiographic appearance of the bone matures. The extent of radiographic ossification is variable at this point in time and difficult to assess because the symptoms often occur before the bone is well mineralized. Ultimately, grade III or grade IV bone formation is apparent in those patients with clinically bothersome restriction of motion. NSAIDs or nonnarcotic analgesics are the cornerstone of treatment until the inflammatory process runs its self-limited course.

Greater trochanteric bursitis may occur in association with small bone spurs (stage II) originating from the base or lateral surface of the greater trochanter. Although not extensive, the strategic location of these spurs at the prominence of the greater trochanter beneath the iliotibial band may provoke troublesome irritation of the bursa located in this area. These symptoms are aggravated by postoperative gluteal abductor weakness secondary to disuse during a prolonged period of arthritic disease before surgery. Effective treatment typically consists of NSAIDs and a gluteal abductor strengthening program. Occasionally, a steroid injection into the region of the bursa and the offending bone spur is necessary to break the cycle of symptoms for the patient. Symptoms usually resolve over a period of several months, but the bursitis associated with these spurs of ectopic bone can be particularly troublesome and refractory to the usual treatment measures. A series of several steroid injections may be indicated and, rarely, surgical removal of the spurs might be considered in conjunction with postoperative measures to prevent recurrence.

Loss of ROM about the hip is the most problematic long-term consequence of HO occurring after THA and is typically present only when grade III or grade IV bone is evident on the radiograph. Most commonly, the operating surgeon finds it remarkable that rather prominent radiographic ossification is accompanied by a modest decrease in range of motion with comparatively little, if any, functional restriction in most patients. The degree of loss of ROM necessary to compromise function varies from patient to patient and is dependent on the individual patient's needs for mobility during daily activities in addition to the degree of arthritic disease of the adjacent joints and low back. Considering that the lumbar spine is most often called upon to compensate for a stiff hip, high-grade HO is most problematic in patients with concurrent lumbar spondylosis and spinal stenosis. In such patients, a hip flexion contracture resulting from HO results in a compensatory exaggeration of the lumbar lordosis which, in turn, aggravates the symptoms of lumbar spondylosis and claudication associated with spinal stenosis.

Typical limitations occur secondary to restricted rotation and flexion ranges of the hip, producing compromised sitting ability, donning of shoes and socks, and foot hygiene. Neither medication nor injection therapy is effective in restoring ROM once the radiographs show evidence of ectopic bone or clinical restriction of motion is present. Although relatively few patients seek surgery to restore lost mobility secondary to HO after THA, surgical excision of the offending bone is the only effective intervention once restricted motion is clinically evident. Surgery is rarely indicated for less than radiographic stage III or stage IV ossification. Moreover, adjunctive prophylactic measures, such as radiation or NSAIDs, are essential in conjunction with surgical excision to prevent postoperative recurrence of bone formation to the same or even greater degree. Because of this fact, the development of effective methods of prophylaxis has attracted even more attention than the surgical management of this problem.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here