Heterotopic Ossification Following Total Hip Arthroplasty: A Contemporary Approach-Based Review for Surgeons Utilizing the Direct Anterior Approach

Abstract

Background

The direct anterior approach (DAA) has become the most frequently utilized surgical approach for total hip arthroplasty (THA) in the United States. There is a growing need to clarify the differential risk of complications associated with the DAA compared to other surgical approaches. Heterotopic ossification (HO) is a well-recognized complication following THA that may result in pain and stiffness. Despite its clinical relevance, the risk of HO in DAA patients remains poorly described. In this review, we provide a comprehensive overview of HO following THA, with a particular focus on the DAA.

Methods

A comprehensive literature search of the PubMed/Medline database was performed using search terms “total hip arthroplasty,” “heterotopic ossification,” and “approach.” Article titles and abstracts were reviewed to determine relevance. Randomized controlled trials, systematic reviews, meta-analyses, comparative studies, and case series were included. Articles not published in the English language were excluded.

Results

HO results from trauma-induced differentiation of osteoprogenitor cells in soft tissue surrounding the hip joint. Surgical approach is a determinant of HO risk after THA, with the DAA producing the lowest incidence of HO. Other risk factors for the development of HO have been identified, including male gender and advanced age. Prophylaxis against HO includes perioperative nonsteroidal anti-inflammatory drugs, tranexamic acid, and local radiation therapy, while management involves surgical resection of ectopic bone in symptomatic patients.

Conclusions

This review synthesizes data from recent systematic reviews and case series to provide arthroplasty surgeons with an understanding of postoperative HO risk, stratified by surgical approach.

Introduction

Total hip arthroplasty (THA) is a safe and effective treatment for end-stage osteoarthritis, offering pain relief and improved joint mobility. A variety of surgical approaches are used in THA. The direct anterior approach (DAA) has gained popularity in recent years and is now the most commonly used approach in the United States, surpassing the posterior approach (PA) in 2024 [1]. The increase in utilization of DAA may be attributable to the potential for improved recovery in the early postoperative period. Compared to PA, DAA may lead to shorter hospital stays and earlier return to function [2]. As the utilization of DAA continues to rise, there is a growing need to clarify the differential risk of postoperative complications associated with each surgical approach. Heterotopic ossification (HO) is a significant complication of THA, with clinically relevant disease affecting roughly 1.3-3.1% of THA patients. It is associated with postoperative pain and limited range of motion [3]. To date, systematic reviews of postoperative HO do not include all available evidence describing rates among DAA patients. Additionally, existing literature provides limited information regarding clinical considerations for arthroplasty surgeons. Therefore, this review aims to summarize the current literature on HO following THA, including its pathogenesis, clinical presentation, and incidence across different surgical approaches, with a particular emphasis on the DAA. This study will also review strategies for prevention and management, providing surgeons with practical, evidence-based recommendations for addressing HO in DAA patients.

Materials and methods

A comprehensive literature search of the PubMed/Medline database was performed using search terms “total hip arthroplasty,” “heterotopic ossification,” and “approach.” Article titles and abstracts were reviewed to determine relevance. Randomized controlled trials, systematic reviews, meta-analyses, comparative studies, and case series were included. Articles not published in the English language were excluded.

Results and discussion

Pathogenesis

The formation of HO requires a progenitor cell population, an inducing agent, and a microenvironment conducive to bone formation [4]. The process is thought to be initiated by the differentiation of susceptible cell populations into osteogenic progenitors [5]. The stem cell population capable of this transformation has not been definitively described, but several populations have been identified as potential candidates [[6], [7], [8], [9], [10], [11], [12], [13], [14]]. A number of studies using transgenic animal models of HO have found that mesenchymal stem cells are the predominant cell type that undergoes osteogenic differentiation [6,7]. More specifically, muscle-derived mesenchymal stem cells and vascular endothelial precursors have been identified as cells capable of commitment to the osteoblastic lineage [[8], [9], [10]]. There is also evidence that endoneurial cells, perivascular cells, and vascular smooth muscle cells may contribute to the formation of ectopic bone [[10], [11], [12], [13]]. Femoral canal–derived cells are another cell population that has been reported to contribute to the HO formation in an animal model. Bone particulate debris created during surgery may distribute osteogenic cells from the femoral canal into soft tissue surrounding the hip, seeding HO [14].

Activation of the BMP signaling pathway is known to be important for osteogenic differentiation. Proteins in the BMP family activate serine-threonine kinase receptors, which in turn phosphorylate R-Smad proteins. Phosphorylated R-Smad proteins form a complex with their cofactor, co-Smad, which then enters the nucleus and regulates transcription of genes related to osteogenic differentiation [15]. Animal models have shown that BMP-2, BMP-4, and BMP-9 induce HO when injected into soft tissue [10,16,17]. Additionally, BMPs have been found to be elevated in trauma-induced ectopic bone lesions in humans [18].

BMP signaling is activated by inflammation. Tumor necrosis factor-α (TNFα), an inflammatory cytokine, upregulates the expression of BMP-2 in endothelial cells through activation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) [19]. However, the key inflammatory cytokine involved in the formation of HO is prostaglandin E2 (PGE2). PGE2 activates osteogenic differentiation in mesenchymal cells and has led to ectopic bone formation in animal models [20,21]. In THA, soft tissue trauma adjacent to joint implants may induce inflammation that results in osteogenic differentiation and, ultimately, HO.

Clinical presentation

In the clinical setting, symptoms of HO typically present between 3 and 12 weeks following surgery. Patients may experience pain, swelling, stiffness, and decreased range of motion in the affected joint [22,23]. A definitive diagnosis of HO is made based on imaging using either plain film radiographs or computed tomography (CT) scans. These imaging modalities can detect calcifications consistent with the formation of HO as early as 6 weeks after an operation [22]. HO is graded according to the Brooker system, which uses anteroposterior radiographs to categorize HO into 4 classes depending on the degree and severity of ectopic bone formation (Fig. 1). Functional deficits in hip flexion and abduction may occur in some patients with class 2 HO, but only classes 3 and 4 are considered clinically relevant.

Figure 1.

Brooker classification of heterotopic bone in the hip joint, shown in anteroposterior (AP) and lateral (L) radiographs. Class 1 is defined as isolated islands of bone in the soft tissue surrounding the hip (a); class 2 is defined as bony projections from the pelvis or the femur with at least 1 cm of space between projections from opposite surfaces (b); class 3 is defined as bony projections from the pelvis or the femur with ˂1 cm of space between projections from opposite surfaces (c); and class 4 is defined as complete ankylosis of the hip joint (d) [24].

The location of HO relative to the hip joint may present diagnostic challenges. With a standard anteroposterior radiograph, ossification lateral and posterior to the hip is easily detectable, but ossification anterior to the hip is not as apparent. One study suggested that a lateral radiograph is needed to identify HO anterior to the hip in some cases. Dorn et al. [25] assessed 209 patients who received THA via the direct lateral approach (DLA) and found that 13% developed HO in the anterior intertrochanteric region that was missed by standard anteroposterior imaging but detectable with a lateral radiograph.Although modified HO diagnostic protocols that include lateral radiographs have been proposed, they have not been widely adopted [[25], [26], [27]]. However, detection of HO anterior to the hip is of clinical importance, as anterior ossification can impede flexion and become symptomatic if impingement occurs.

The localization of HO could have implications for postoperative screening practices based on surgical approach. To our knowledge, no comparative studies have examined the influence of surgical approach on the location of HO [28,29]. Nevertheless, the various approaches employed in THA produce soft tissue trauma in different locations relative to the hip and therefore may lead to approach-dependent localization of HO. The PA and DLA cause soft tissue trauma posterior and lateral to the hip, while DAA produces soft tissue trauma anterior to the hip. Accordingly, standard anteroposterior radiographs are likely sufficient to screen for HO in most PA and DLA patients, but lateral radiographs may be indicated in DAA patients. There is, however, some evidence from noncomparative studies that DLA can lead to ossification anterior to hip [25,30].

Risk factors

Many factors have been shown to be associated with increased risk for the development of HO following THA. Among the most commonly reported risk factors are male gender and advanced age. A 2015 meta-analysis performed by Zhu et al. reported that the odds ratio for development of postoperative HO in males compared to females was 2.11. The study also reported that bilateral operations, history of ankylosing spondylitis, and previous hip ankylosis were risk factors for HO development. [31] A retrospective case series conducted by Aprato et al. [32] corroborated the increased risk of HO among males and also reported an elevated risk in patients whose operations lasted more than 90 min and those with prolonged use of postoperative drains. Concerning patient age, it is generally accepted that advanced age is associated with an increase in HO rates. One retrospective case series demonstrated an association between HO (Brooker 1-4) incidence and older patient age, while another found that patients over 65 have an increased incidence of clinically relevant HO (Brooker 3,4) [32,33]. However, some studies have found no significant association between advanced age and HO incidence [31,34]. Future studies are needed to confirm the effects of age on the development of postoperative HO.

Several other risk factors for HO development have been identified, though they are less widely reported in the literature. The type of hip prosthetic used has been found to impact rates of HO development, with cemented implants associated with a higher incidence than uncemented implants [31,35]. A recent retrospective review of 2541 primary THA patients showed that black patients and patients with osteoporosis, spine disease, or low estrogen levels had an increased risk of developing postoperative HO [36]. A link between elevated body mass index (BMI) and higher incidence of postoperative HO has been reported, but the association remains controversial [33,37]. It has also been reported that hypertrophic osteophytes are a risk factor for HO, both independently and in males with substantial osteophytic osteoarthritis, but Zhu et al. found no association between the presence of osteophytes and the incidence of HO [31,[38], [39], [40]]. Patients with osteonecrosis may have an elevated risk of HO development due to increased intraoperative bleeding and hematoma formation, which have been shown to be associated with HO [[41], [42], [43], [44]]. Recently, surgical approach has been identified as a determinant of ectopic bone formation following THA. Table 1 summarizes reported risk factors for development of HO.

Table 1.

Summary of associations between patient factors and the development of HO.

Reference	N	Risk factors	Protective factors	No association
Zhu et al., 2015 [31]	6468	Male gender, cemented implants, bilateral operations, ankylosing spondylitis, previous hip ankylosis	Rheumatoid arthritis	Age, use of NSAIDs, femoral neck fracture, previous hip fracture, hypertrophic osteoarthritis, and presence of osteophytes
Aprato et al., 2023 [32]	1225	Male gender, >65 years of age, operative time >90 min, prolonged use of postoperative drains	-	-
Alijanipour et al., 2017 [33]	1482	Male gender, increased age, elevated BMI, increased length of hospital stay, increased operative time	-	-
Eggli et al., 2001 [34]	1318	Male gender, heterotrophic osteoarthritis, previous hip surgery, previous HO after contralateral THA, trochanteric osteotomy, subtrochanteric femoral osteotomy	Rheumatoid arthritis	Age, weight, height, diabetes, obesity, type of anesthesia, hematoma, type of femoral component fixation, type of acetabular component fixation, preoperative pain, preoperative flexion, preoperative walking capacity, and preoperative intermalleolar spreading distance
Pavlou et al., 2012 [35]	920	Male sex, cemented implants, previous HO following THA	-	-
Singh et al., 2022 [36]	2541	Black/African American race, osteoporosis, spine disease, low estrogen state	-	-
Ritter et al., 1977 [38]	507	Males with bilateral osteophytic osteoarthritis, previous ectopic ossification	-	-
Shaffer, 1989 [39]	-	Males with osteophytic osteoarthritis, ankylosing spondylitis, diffuse idiopathic spinal hyperostosis, previous hip surgery, previous HO	-	Prosthesis type, use of trochanteric osteotomy, cemented implants
Handel et al., 2004 [40]	589	Male, high BMI, low preoperative ROM, long operative time, presence of osteophytes	-	Age, cemented implants

ROM, range of motion.

Incidence of HO by surgical approach

The incidence of HO following THA varies depending on the selected surgical approach. Systematic reviews demonstrate that the incidence of HO following DAA may be decreased compared to either PA or DLA. In a systematic review of 6512 patients, Herzberg et al. reported HO rates of 15.2% for DAA patients, 21% for PA patients, and 57.2% for DLA patients. Incidence of clinically relevant HO followed a similar trend, with 2.3% of DAA patients, 4.4% of PA patients, and 5.4% of DLA patients developing HO with Brooker class 3 or higher [3]. Unfortunately, this study did not provide statistical analyses. A 2021 systematic review and meta-analysis examining 24,853 hip replacements also found that DAA resulted in a lower incidence of HO than DLA, but the difference was not statistically significant (12.16% vs 26.47%; P = .07). Clinically relevant HO was slightly lower in DAA patients (3.0%) than in DLA patients (3.9%), but this difference was also insignificant [45].

Findings from comparative case series generally support the pattern of HO incidence observed in systematic reviews. In a retrospective, single-surgeon study performed by Newman et al., 24.3% of DAA patients developed HO and 27.5% of PA patients developed HO. Clinically relevant HO was significantly lower in the DAA (3.0%) group than the PA group (7.5%) [46]. Wilke et al. demonstrated a similar trend in a multi-surgeon retrospective case series, finding that PA patients had significantly higher odds of developing HO (OR = 2.99, 95% CI, 2.40-3.72, P < .001) and clinically relevant HO (OR = 2.31, 95% CI, 1.45-3.68, P < .001) than DAA patients [29]. It has also been reported that DAA results in a lower incidence of HO compared to DLA [33]. In a retrospective, multisurgeon case series, Alijanipour et al. found a statistically significant decrease in HO incidence in DAA patients compared to DLA patients, with 19.4% of DAA patients and 36.1% of DLA patients developing HO (P < .001) [33]. However, Hürlimann et al. [47] reported that DAA patients showed an increase in HO compared to DLA patients but provided no statistical comparison between the 2 approaches.

The reported incidence of HO ranges from 9.6% to 33% following DAA, 10.4% to 71.1% following PA, and 14.3% to 84.8% following DLA. The incidence of clinically relevant HO ranged from 1.3% to 6.7% in DAA patients, 1.6% to 23.1% in PA patients, and 3.1% to 15.8% in DLA patients [33,[45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60]]. Table 2 summarizes comparative and single-approach case series examining HO incidence across different surgical approaches.

Table 2.

Incidence of HO across various surgical approaches.

Reference	N	Approach	Incidence of HO	Incidence of clinically relevant HO (Brooker ≥3)
Newman et al., 2016 [46]	235	DAA	24.3%	3.0%
	120	PA	27.5%	7.5%
Wilke et al., 2023 [29]	1295	DAA	9.6%	–
	1565	PA	24.1%	OR = 2.31 (compared to DAA)
Alijanipour et al., 2017 [33]	746	DAA	19.4%	3.0%
	736	DLA	36.1%	3.9%
Hürlimann et al., 2017 [47]	39	DAA	23.1%	5.1%
	28	DLA	14.3%	3.6%
Hayashi et al., 2019 [48]	114	PA	71.1%	8.8%
	33	DLA	84.8%	15.2%
Rüdiger et al., 2020 [49]	401	DAA	29.9%	3.3%
Knapp et al., 2020 [50]	45	DAA	26.7%	6.7%
Leunig et al., 2018 [51]	964	DAA	12.6%	1.7%
Hartford & Bellino, 2017 [52]	442	DAA	17.7%	3.8%
Tippets et al., 2014 [53]	233	DAA	41.5%	9.4%
Chémaly et al., 2013 [54]	39	PA	38.5%	23.1%
Edwards et al., 2015 [55]	510	PA	10.4%	1.6%
Malhotra et al., 2019 [56]	18	PA	11.1%	5.6%
Rashed et al., 2020 [57]	111	PA	12.2%	4.4%
Ateschrang et al., 2014 [58]	250	DLA	64.1%	11.7%
Busch et al., 2020 [59]	67	DLA	41.3%	3.1%
Syed et al., 2015 [60]	38	DLA	63.2%	15.8%

Data were extracted from observational retrospective case series with a minimum follow-up time of 3 months. Additional information can be found in Appendix 1.

The lower observed incidence of HO following DAA could be due to a decrease in trauma than other approaches. DAA may be associated with less soft tissue trauma given the internervous plane utilized. [3,4,61] Several studies have found evidence for a decrease in serum markers of muscle inflammation in DAA compared to both DLA and PA [62,63]. It has been suggested that the gluteus minimus specifically is spared with the DAA. Two retrospective studies utilizing postoperative MRIs found a significant decrease in damage to the gluteus minimus in DAA compared to DLA. Neither study found a significant difference in soft tissue damage between DAA and PA [64,65]. However, a cadaveric study found that DAA produced less soft tissue damage to the gluteus minimus compared to PA [66]. The soft tissue sparing nature of DAA may result in less inflammation, contributing to the lower associated incidence of HO. Therefore, meticulous intraoperative soft tissue handling is recommended to minimize the risk of HO in THA patients, regardless of surgical approach.

The incidence of HO after DAA may also be influenced by the use of an orthopaedic table during surgery. Several comparative studies have reported rates of HO with and without orthopaedic tables [53,57]. In a retrospective case series, Tippets et al. found that the HO rate in DAA patients was 47.9% higher when an orthopaedic table was not used. Additionally, they showed that the Brooker class of patients who received THA without an orthopaedic table was significantly higher than those who received THA with an orthopaedic table. The authors suggested a mechanism for this observation, proposing that THA performed without an orthopaedic table produces more soft tissue trauma, specifically during femoral elevation [53]. It has also been proposed that operations performed with a standard table require greater soft tissue release and more extensive use of retractors [67,68]. These factors may explain the higher rates of HO formation observed following standard table DAA surgery.

The benefit of DAA over other surgical approaches is reduced when an orthopaedic table is not used. A retrospective case series reported that the HO rate following DAA without an orthopaedic table was higher than the HO rate following PA (47.7% vs 27.6%, P < .01) [67]. The use of an orthopaedic table may be a surrogate measure for soft tissue damage in DAA.

Prophylaxis

Perioperative prophylaxis has been recommended for high-risk patients to reduce the risk of HO following THA. The most commonly used prophylactic method is oral NSAIDs. Through inhibition of cyclooxygenase (COX), NSAIDs downregulate the synthesis of prostaglandins, preventing proliferation of osteogenic cells and reducing ectopic bone formation [69]. The efficacy of nonselective NSAIDs as a prophylactic measure against HO is well established [70]. Indomethacin has been shown to significantly reduce the incidence of HO compared to controls in a large number of studies [37,[71], [72], [73], [74], [75], [76], [77]]. The standard dosing protocol for indomethacin is 100 mg per day for 6 weeks, but a variety of regimens have been shown to be effective [[73], [74], [75], [76],78]. A randomized trial of 154 patients found that 50 mg of indomethacin taken twice a day for 6 weeks significantly decreased HO incidence compared to controls [73]. Another randomized trial of 41 patients showed that 25 mg of indomethacin taken 3 times a day for 2 weeks also led to a significant reduction in HO incidence [74]. Two separate studies examining the duration of indomethacin use found no significant difference in Brooker scores between 7- and 14-day regimens of 50 mg given twice a day [75,76]. A prospective, nonrandomized study showed that a 3-day regimen 50 mg of indomethacin 3 times a day following an initial 100 mg dose was insufficient to protect against HO, finding no significant difference in HO incidence between the indomethacin and control groups [77]. Though historically considered the standard of care for HO prophylaxis, indomethacin is no longer routinely prescribed [79]. In contemporary practice, postoperative THA regimens typically include alternative medications with prophylactic efficacy against HO that carry side-effect profiles more favorable to that of indomethacin. Other nonselective and COX-II selective NSAIDs, such as meloxicam and celecoxib, respectively, present fewer tolerability issues than indomethacin, demonstrating a lower risk of gastrointestinal side effects [[80], [81], [82]].

Aspirin, which is often prescribed for VTE prophylaxis following THA, may provide a secondary benefit as an inhibitor of HO development. A recent systematic review found that aspirin significantly reduced both all-class and clinically relevant HO when 325 mg were given twice a day for 4 to 6 weeks postoperatively. There was an increase in HO incidence in patients who received low doses of aspirin (81 mg twice a day) compared to those who received 325 mg twice a day, but the difference was not significant [83]. Ketorolac, which is used primarily for postoperative pain management in THA patients, may be protective against HO as well. A randomized control trial found that ketorolac significantly reduced HO incidence compared to controls. Additionally, none of the 152 patients treated with ketorolac developed clinically relevant HO, while 7.3% of control patients developed HO with Brooker ≥3. In this study, 60 mg of ketorolac were administered intraoperatively, and 30 mg were given every 8 hours for a total of five doses [84].

Other nonselective NSAIDs have been studied for their prophylactic efficacy but are less frequently prescribed. Ibuprofen has been shown to reduce the postoperative incidence of HO with efficacy equivalent to indomethacin. A recent comparative study examining HO rates after ibuprofen and indomethacin use found that ibuprofen was as effective as indomethacin at preventing both all-class HO and clinically relevant HO [85]. In a systematic review, diclofenac was found to significantly reduce the incidence of HO compared to controls [86]. Another systematic review found that naproxen reduced overall incidence of HO and incidence of Brooker class 1 and 2 HO but did not reduce rates of clinically relevant HO (Brooker ≥3) [87].

COX-II selective NSAIDs have been studied as an alternative to traditional NSAIDs due to their more favorable side-effect profile. Several recent meta-analyses have shown that HO rates in patients treated with COX-II selective NSAIDs were not significantly different from those in patients who received nonselective NSAIDs [70,88,89]. Celecoxib is a frequently prescribed COX-II selective prophylactic agent in THA patients. A retrospective case control found that patients given 100 mg of celecoxib twice a day for 10 days had significantly reduced HO incidence and significantly increased improvement in their Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) stiffness scores compared to controls. The incidence of clinically relevant HO in the celecoxib group was not significantly different from the control group (2.9% vs 3.8%, P = .64) [90]. The protective effects of celecoxib extend to patients who receive THA with DAA. A retrospective case series of 688 DAA patients found that celecoxib significantly reduced all-class HO (OR = 0.22) and clinically relevant HO compared to controls [91].

Meloxicam is a preferentially selective COX inhibitor, displaying inhibition of COX-II at doses ˂15 mg daily and inhibition of both COX-I and COX-II at doses ˃15 mg daily. Two separate prospective cohort studies have demonstrated the protective effects of meloxicam. The first found a significant decrease in HO incidence compared to controls in patients who received 15 mg of meloxicam preoperatively followed by 15 mg daily for 7 days. Additionally, this study showed no significant difference in prophylactic efficacy between meloxicam and indomethacin, with 100 mg of indomethacin given 1 hour before surgery followed by 150 mg daily for 1 week [92]. The second study also found meloxicam to be as effective as indomethacin at preventing overall HO, with meloxicam administered at 15 mg daily and indomethacin at 100 mg daily, both for 12 days. The incidence of clinically relevant HO was the same in both treatment groups [93]. Additionally, a recent systematic review comparing indomethacin to celecoxib, rofecoxib, meloxicam, and ibuprofen found that no NSAID was superior in the prevention of HO [94]. One prospective randomized trial contradicted these results, finding that meloxicam led to a significantly higher incidence of overall HO than indomethacin, with meloxicam given at 15 mg daily and indomethacin given at 100 mg daily, both for 14 days. However, in this study, meloxicam was associated with a slightly lower incidence of clinically relevant HO, though statistical significance was not reported [95]. As such, meloxicam appears to be an effective prophylactic agent against HO, but its efficacy relative to indomethacin has not yet been determined.

Non-NSAID drugs also play an important role in HO prophylaxis. Tranexamic acid (TXA), a synthetic lysine analog, inhibits the activation of plasminogen to plasmin and is used to control bleeding in surgical patients. TXA may serve a prophylactic role among THA patients as it may reduce excessive bleeding and, consequently, prevent the formation of ectopic bone [[96], [97], [98], [99]]. There is also evidence that TXA may reduce HO by decreasing inflammation [[97], [98], [99]]. A retrospective review of 357 patients found that patients who received perioperative TXA had a significantly lower incidence of HO than patients who did not (OR = 0.58). In this study, 2 doses of TXA were administered at 10 mg/kg, with one dose given preoperatively and one given 2 hours later [100]. Another retrospective study of 401 primary THAs supported these results, also finding that TXA significantly reduced the incidence of clinically relevant HO [101]. Taken together, the current evidence suggests TXA may serve a secondary prophylactic role in HO prevention in THA patients. Studies examining the efficacy of different HO prophylactic agents are summarized in Table 3.

Table 3.

Dosing, prophylactic outcomes, and adverse effect considerations for HO prophylaxis drugs.

Reference	Medication	Dose	Duration	Outcomes	Contraindications
Wang et al., 2023 [83]	Aspirin	325 mg twice a day	-	Significantly reduced HO Brooker ≥3	History of GI bleeding, peptic ulcers, or gastritis, hepatic or renal dysfunction, bleeding disorders
Pritchett et al., 1995 [84]	Ketorolac	60 mg intraoperatively, followed by 30 mg for 5 doses	N/A	Significantly reduced HO Brooker ≥3	History of GI bleeding, peptic ulcers, or gastritis, renal dysfunction, bleeding disorders
Kjaersgaard-Andersen et al., 1993 [74]	Indomethacin	25 mg 3 times a day	2 weeks	Significantly reduced HO incidence	History of GI bleeding, peptic ulcers, or gastritis, renal dysfunction, warfarin
Schneider et al., 2023 [85]	Ibuprofen	600 mg BID	-	Significantly reduced HO Brooker ≥3	History of GI bleeding, peptic ulcers, or gastritis, IBD, hepatic or renal dysfunction
Haffer et al., 2022 [86]	Diclofenac	50 mg 3 times a day	6 weeks	Significantly reduced clinically relevant HO (Hierton classification)	History of GI bleeding, peptic ulcers, or gastritis, IBD, hepatic or renal dysfunction
Zhang et al., 2019 [87]	Naproxen	Variable	Variable	Significantly reduced HO incidence; HO Brooker ≥3, not significantly reduced	Active GI bleeding or ulcers, IBD, renal or hepatic dysfunction
Badi et al., 2023 [90]	Celecoxib	100 mg twice a day	10 days	Significantly reduced HO incidence; insignificantly reduced HO Brooker ≥3	Active GI bleeding, peptic ulcers, or gastritis, IBD, hepatic or renal dysfunction
Naylor et al., 2021 [91]	Celecoxib	200 mg twice a day	3 weeks	Significantly reduced HO, HO Brooker ≥3	-
van der Heide et al., 2004 [92]	Meloxicam	15 mg perioperatively, 15 mg everyday	7 days	Significantly reduced HO Brooker ≥3	Active GI bleeding or peptic ulcers, history or peptic ulcers, IBD, hepatic or renal dysfunction
Legenstein et al., 2003 [93]	Meloxicam	15 mg everyday	12 days	Reduced HO Brooker ≥3 equivalent to indomethacin	-
Barthel et al., 2002 [95]	Meloxicam	15 mg everyday	14 days	Significantly increased overall HO, decreased HO Brooker ≥3 compared to indomethacin	-
Johnson et al., 2025 [100]	Tranexamic Acid	One 10 mg/kg dose preoperatively, second dose 2 h later	N/A	Significantly reduced HO incidence	Bleeding disorder, renal dysfunction
Debre et al., 2021 [101]	Tranexamic Acid	-	-	Significantly reduced HO Brooker ≥3	-

GI, gastrointestinal; IBD, irritable bowel disease.

Radiation therapy has been shown to be an effective prophylactic agent against HO in arthroplasty patients. Local perioperative radiation results in a significantly reduced incidence of HO following THA [102]. However, radiation therapy is not routinely indicated for standard-risk patients and should not be viewed as mandatory, even in higher-risk individuals, including those with a history of HO in the contralateral hip, ankylosing spondylitis, or hip ankylosis [31,34]. Rather, its use may be considered on a case-by-case basis after weighing patient-specific risk factors, surgeon preference, institutional resources, and alternative prophylactic measures such as NSAID-based regimens. Logistical challenges related to the perioperative timing of dosage may limit radiation usage among THA patients. Radiation therapy is typically administered either preoperatively within 24 h of surgery or postoperatively within 72 h of surgery [103]. There does not appear to be a difference in prophylactic efficacy between preoperative and postoperative dosing. A randomized trial measured HO formation in THA patients who received single dose, 700-800 cGy radiation therapy either preoperatively or postoperatively and found no significant difference in HO incidence between the 2 groups [104]. In terms of dosage quantity, many perioperative regimens exist, but the optimal dosage remains unknown. A retrospective study found that a single preoperative dose of 700-800 cGy was an effective prophylactic dose, preventing HO in 85% of high-risk patients [105]. Several other studies have reported that there is a minimum required dose to mitigate HO formation. A randomized trial comparing a single 400 cGy dose to a single 700 cGy dose found that the incidence of HO was significantly greater in the lower dose group [106]. A similar increase in HO incidence has been reported in patients who received 550 cGy compared to those who received 700 cGy [107]. However, 1 randomized trial found that there was no significant difference in postoperative HO incidence between patients who received 500 cGY and those who received 1000 cGy [108]. Accordingly, the current literature does not establish a single preferred regimen. Further study is needed to better define the role of prophylactic radiation in modern THA practice. Studies reporting outcomes of various prophylactic radiation regimens are summarized in Table 4.

Table 4.

Dosing regimens and prophylactic outcomes of radiation therapy.

Study	N	Dose (cGy)	Timing	Doses	Incidence of HO
Gregoritch et al., 1994 [104]	55	700-800	Within 4 h preoperatively	1	26% (2% Brooker ≥3)
	43	700-800	Within 2 days postoperatively	1	28% (P = 1.0) (5% Brooker ≥3; [P = .58])
Hanna et al., 2022 [105]	39	700-800	2-4 h preoperatively	1	15.4%
Liu et al., 2016 [106]	71	400	1-2 days postoperatively	1	57.8% of patients showed no increase in Brooker score
	76	700	1-2 days postoperatively	1	75.0% of patients showed no increase in Brooker score (P = .027)
Healy et al., 1995 [107]	19	550	0-5 days postoperatively	1	63% (21.1% Brooker ≥3)
	88	700	0-5 days postoperatively	1	10% (2.3% Brooker ≥3)
Padgett et al., 2003 [108]	29	500	2-4 days postoperative	2 (250 cGy each)	69.0%
	30	1000	2-4 days postoperative	5 (200 cGy each)	43.3%; (P = .086)

In summary, HO prophylaxis regimens should be individualized based on patient-specific risk. For standard-risk patients, chemoprophylaxis alone is typically sufficient and should include perioperative NSAIDs and TXA. As most NSAIDs demonstrate similar prophylactic efficacy, selection can be guided by side-effect profile and secondary function in the management of THA patients. COX-II selective NSAIDs should be considered for patients with elevated risk for gastrointestinal complications. Aspirin, commonly used for VTE prophylaxis, has efficacy in HO prevention, and ketorolac may be considered in the early postoperative period. Administration of TXA preoperatively and intraoperatively is recommended to reduce intraoperative bleeding and mitigate HO formation. For high-risk patients, including those with previous HO in the contralateral hip, ankylosing spondylitis, or hip ankylosis, local radiation therapy can be considered as adjunct prophylaxis. However, radiation may be logistically challenging and not required for patients without risk factors for HO.

Management

Once HO has developed, surgical excision is the standard of care for patients with severely restricted motion or activity-limited pain. (Fig. 2) Resection of HO is indicated for patients experiencing symptoms, which typically occur only in high-grade disease (Brooker ≥3). There is conflicting evidence regarding the success of surgical excision in alleviating symptoms of HO. A study evaluating outcomes of HO excision following THA found that excision produced a significant increase in hip range of motion but did not improve patient-reported pain [109]. A more recent systematic review corroborated the finding that surgical excision leads to an increased range of motion, but reported that pain relief after excision was inconsistent [110]. Surgical excision of HO should be performed only once the ectopic bone has matured. Following surgery, HO maturation is typically complete 12 to 18 months postoperatively [111]. However, longer maturation times have been observed in some THA patients, with increases in Brooker class reported as late as 3 years after THA [23]. Decreasing bone scan activity and the return to baseline of serum alkaline phosphatase levels, which are elevated during deposition of ectopic bone, are both useful indicators in marking the maturation of HO [111].

Figure 2.

a. Anterior–posterior radiograph of Brooker class 4 HO in a patient who received THA. b. Anterior–posterior intraoperative fluoroscopic image of the patient’s hip following surgical resection of HO. c. Macroscopic view of heterotrophic bone excised from the patient’s hip joint.

Perioperative prophylaxis is recommended for patients undergoing surgical excision of HO to prevent recurrence. A recent systematic review examined HO recurrence rates in patients who underwent HO excision and received prophylaxis with indomethacin, radiation, or both. Of the 41 patients included in the study, only 2 patients developed recurrent HO, including one who received radiation alone and one who received indomethacin alone [110]. Another study of 14 patients who underwent HO excision found that no patients developed recurrent HO after treatment with a single dose of 700 cGy on postoperative day 1 followed by 50 mg of indomethacin daily for 2 weeks. The average modified Harris Hip Score and the average nonarthritic Hip Score of patients both increased after excision [112].

There is evidence that suggests delayed radiation therapy may be effective in limiting the progression of existing HO. A prospective case series of 17 patients measured the effect of delayed radiation on HO progression in patients who developed HO within 6 weeks postoperatively. Patients who received 700 cGy following their 6-week follow-up visit showed no progression of HO after 6 months and demonstrated a significant decrease in the amount of ectopic bone formed compared to nonirradiated patients (32% vs 86%, P < .05) [113]. Delayed radiation should be considered for patients with immature HO.

Conclusions

HO is a significant complication following THA, with clinically relevant disease occurring in 1.3-3.1% of patients postoperatively. HO results from the inflammation-induced differentiation of osteoprogenitors and is influenced by a number of risk factors, including male gender and advanced age. DAA, the most commonly utilized surgical approach in contemporary THA, may offer a lower incidence of HO than alternative approaches. Systematic reviews and large cohort case series demonstrate a lower rate of HO in DAA patients than PA and DLA patients. However, the incidence of HO varies greatly between approaches, and further investigation is needed to determine the effects of surgical approach on HO development. A decrease in soft tissue trauma in DAA patients may be responsible for the apparent differences in HO incidence between alternative surgical approaches, enforcing the importance of meticulous soft tissue handling in THA. Prophylaxis against HO includes perioperative NSAIDs, ASA, and TXA in standard-risk patients. Clinical management involves surgical resection of high-grade ossification in symptomatic patients.

CRediT authorship contribution statement

Connor Buchanan: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis, Conceptualization. McKenzie W. Culler: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Conceptualization. Mohammad Poursalehian: Writing – review & editing, Writing – original draft, Methodology, Investigation, Formal analysis. Lucas Anderson: Writing – review & editing, Supervision, Formal analysis, Conceptualization. Andrew G. Yun: Writing – review & editing, Supervision, Formal analysis, Conceptualization. Nathanael D. Heckmann: Writing – review & editing, Supervision, Methodology, Formal analysis, Conceptualization.

Conflicts of interest

Andrew Yun receives royalties from Smith & Nephew, is a speaker for Avanos, and is a paid consultant for Avanos and Mizuho. Nathanael Heckmann receives royalties from Corin USA, is a paid consultant for Intellijoint Surgical, MicroPort Orthopaedics, Corin USA, and Zimmer, holds stock in Intellijoint Surgical, and serves as a board member of AAOS, AJRR, AAHKS, the Knee Society, and the Hip Society. Lucas Anderson receives royalties from OrthoGrid, is a speaker for Medacta, is an unpaid consultant or OrthoGrid, hold stock in OrthoGrid, receives, receives other financial or material support from Smith & Nephew. The other authors declare that there are no conflicts of interest.

For full disclosure statements, refer to https://doi.org/10.1016/j.artd.2026.102110.