Abstract
Background
Heterotopic ossification (HO) is a recognized complication following operative fixation of elbow trauma. While indomethacin prophylaxis is well established in hip and acetabular surgery, evidence for its role in elbow injuries remains limited. This randomized controlled trial evaluated whether oral indomethacin reduces the incidence and severity of HO after surgical treatment of acute elbow fractures and dislocations.
Methods
Eighty-six adults with acute elbow injuries requiring surgical fixation were randomized 1:1 to receive either oral indomethacin (25 mg 3 times daily for three weeks) or no prophylaxis. Eighty-four patients completed follow-up. HO was assessed radiographically at scheduled intervals through the final follow-up and graded using the Hastings and Graham system. The Mantel–Haenszel method was used to calculate a pooled odds ratio adjusted for injury severity.
Results
HO developed in 21 of 84 patients (25%), occurring in 14.3% of the indomethacin group vs. 35.7% of controls (P = .018), with indomethacin reducing both incidence (absolute risk reduction 21.4%; number needed to treat = 5) and severity (mean grade, 1.33 vs. 2.93, P = .013). All HO in the indomethacin group were low grade (G1-G2A), whereas high-grade HO (G2C-G3) occurred in 8 of 15 (53.3%) control patients. Stratified analysis demonstrated consistent trends across injury severity levels, with Mantel–Haenszel adjustment yielding similar estimates. Patients with HO in the indomethacin group had a greater range of motion compared to controls.
Discussion and Conclusion
Indomethacin prophylaxis was associated with reduced incidence and severity of HO following operative fixation of elbow injuries. These findings should be interpreted in the context of baseline differences in injury severity between groups. Indomethacin may be a useful adjunct in patients at risk of HO following elbow trauma given its low cost, ease of administration and favorable safety profile, although further multicenter studies are required.
Keywords: Elbow injuries, Heterotopic ossification, Indomethacin, Prophylaxis, Randomized controlled trials, Fracture fixation, Anti-inflammatory agents
Heterotopic ossification (HO) refers to the abnormal formation of bone in soft tissues, typically occurring where bone is not normally present. It most commonly develops following musculoskeletal trauma, orthopedic surgery, particularly hip and elbow procedures, and neurological injuries such as spinal cord or traumatic brain injury.12 The pathogenesis of HO involves inflammatory signaling at the site of injury, promoting the differentiation of mesenchymal stem cells into osteogenic precursors.10 Around the elbow, HO can severely restrict motion, impair functional recovery, and complicate rehabilitation following trauma surgery.
HO has been extensively studied in hip and acetabular surgery,16 where reported incidence rates range from 43% to 73% after total hip arthroplasty and other procedures.13 It has also been described following total knee arthroplasty, with a higher incidence in osteoarthritic than rheumatoid knees.6,15 In contrast, the true incidence of HO following elbow surgery remains less clearly defined due to inconsistent terminology and heterogeneity in radiographic interpretation.4,5,8 Estimates vary widely, from 3% to 89% following elbow trauma,4,5,14 although a recent publication reported an incidence of 28.7% after surgical repair of elbow fractures.7
Experimental studies have demonstrated that HO formation is associated with increased local prostaglandin activity, particularly prostaglandin-E2, in animal models.3 The involvement of prostaglandins is further supported by findings that cyclooxygenase inhibitors such as indomethacin, as well as corticosteroids, reduce HO formation.2 The effectiveness of indomethacin prophylaxis in reducing HO after total hip replacement has been confirmed in several clinical studies.9,17
Despite the recognized impact of HO on post-operative recovery, evidence guiding prophylactic strategies following elbow trauma remains limited. Therefore, this prospective randomized controlled trial (RCT) was designed to evaluate the effect of oral indomethacin on the incidence and severity of HO following surgical fixation of acute elbow injuries.
Materials and methods
This prospective, single-center, assessor-blinded, RCT was conducted at a tertiary-care referral center. Ethical clearance was obtained from the Institutional Ethics Committee (Approval No. INT/IEC/2021/SPL-1793, dated 27/12/2021), and the study was carried out in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. Patient recruitment occurred between January 2021 and September 2022, and all patients were followed prospectively until December 2024, ensuring a minimum of 24 months of follow-up. The trial was registered retrospectively with the Clinical Trials Registry–India (REF/2026/03/128,157) after completion of patient recruitment. The delay in registration was administrative and unintentional, and the study protocol remained unchanged throughout the study period.
Eligible participants were skeletally mature adults (≥18 years) presenting with acute elbow trauma requiring internal fixation within 14 days of injury. Fracture types included distal humerus, olecranon, radial head or neck, capitellum, coronoid, and fracture–dislocation configurations. Patients were excluded if they had sustained head injury, spinal cord injury, burns, prior elbow surgery, pre-existing HO, chronic neuromuscular disease or inflammatory arthropathy, renal impairment, active peptic ulcer disease, nonsteroidal anti-inflammatory drug (NSAID) allergy, pregnancy or breastfeeding, or were unable to comply with the post-operative protocol.
The sample size calculation was based on prior literature reporting HO incidence rates of approximately 30-60% following elbow trauma surgery,1,4,5 with NSAID prophylaxis shown to reduce HO rates to approximately 10-20% in orthopedic settings.9,11 Given the lack of prior randomized trials in elbow trauma, assumptions were extrapolated from available elbow observational studies and NSAID prophylaxis data in related orthopedic procedures. Accordingly, an expected reduction from 40% in the control group to 15% in the indomethacin group (absolute reduction 25%) was assumed for sample size estimation, with α = 0.05 and 80% power, yielding a minimum requirement of 39 patients per arm. To account for possible attrition, 86 patients were enrolled.
Randomization was performed using a computer-generated sequence with variable block sizes (1:1). Allocation concealment was ensured using sequentially numbered, opaque, sealed envelopes prepared by an independent coordinator. The intervention was administered by personnel not involved in outcome assessment. The radiologist interpreting the imaging and the statistician analyzing the data were blinded to group allocation. Range of motion (ROM) measurements were performed by a physiotherapist blinded to treatment allocation using a standardized goniometric technique. Treating surgeons were aware of the intervention but were not involved in outcome assessment. These measures were implemented to minimize selection and observer bias.
All procedures were performed by a senior fellowship-trained orthopedic surgeon using standard, widely accepted surgical approaches appropriate to the fracture pattern. These included posterior, lateral (Kocher or Kaplan), and combined approaches where indicated. The surgical goals were anatomical reduction, stable fixation, restoration of articular congruity, and preservation of soft tissues. Fixation strategies were selected based on fracture configuration and included plate osteosynthesis, screw fixation, and radial head arthroplasty where required. Efforts were made to minimize soft-tissue dissection and periosteal stripping in all cases. Operative details, including fracture classification, presence of dislocation, open or closed status, neurovascular findings, operative duration, and intraoperative complications, were recorded.
Patients in the intervention arm received oral indomethacin 25 mg 3 times daily for three weeks. Adherence to therapy was assessed through patient self-report and follow-up questioning at each visit. Patients were specifically asked about missed doses and treatment completion. All the patients completed the prescribed 3-week course, and no patient required premature discontinuation due to adverse effects (predefined as upper abdominal pain, dyspepsia, nausea, vomiting, gastrointestinal bleeding, and a rise in serum creatinine greater than 0.3 mg/dL or 50% above baseline, headache, dizziness, or hypersensitivity reactions). Renal function was monitored clinically and investigated biochemically only if indicated. All analyses were performed on an intention-to-treat basis. No crossover between groups occurred during the study period.
Post-operatively, all patients followed a uniform rehabilitation protocol. The elbow was immobilized in a posterior splint for 3 to five days to allow soft-tissue rest, after which supervised physiotherapy commenced. Early rehabilitation emphasized a gentle, pain-free passive and active-assisted ROM performed 3 to 5 times daily. Pronation–supination was restricted for two weeks in cases involving radial head or proximal ulna fixation and up to four weeks in fracture–dislocations or ligament repairs. Forced stretching and aggressive manipulation were avoided. Active ROM was advanced between 2 and six weeks to achieve functional motion by 8 to 10 weeks. Strengthening exercises were initiated after 10 to twelve weeks, once radiographic evidence of union permitted. Importantly, surgical technique, implant selection, and post-operative rehabilitation protocols were consistent across both groups, minimizing potential confounding related to operative management.
The primary outcome measure was the incidence of HO, identified on serial radiographs and graded using the Hastings and Graham classification. Secondary outcomes included HO severity, time to appearance, and elbow ROM, which were measured with a standard goniometer by a physiotherapist blinded to group allocation. Patients were evaluated clinically and radiographically at three weeks, six weeks, three months, six months, twelve months, twenty-four months, and final follow-up. Care was taken to differentiate HO from fracture fragments or callus formation based on serial radiographic progression, anatomical location, and maturation patterns. In equivocal cases (n = 4), computed tomography imaging was used selectively to confirm the diagnosis. The diagnosis and grading of HO were based on serial radiographic assessment in accordance with the Hastings and Graham classification, with clinical correlation of ROM limitation as measured by the physiotherapist. All imaging was interpreted by a single experienced musculoskeletal radiologist who was blinded to treatment allocation, reducing observer bias in HO detection.
Statistical analysis was performed using standard statistical software. The normality of the distribution was assessed using the Shapiro–Wilk test. Continuous variables were compared using independent t-tests or Mann–Whitney U tests, depending on the distribution of the data, and categorical variables using chi-square or Fisher exact tests. Relative risks and odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. All analyses were two-tailed, with P < .05 considered statistically significant. Both stratified (Mantel–Haenszel) analysis and a parsimonious multivariable logistic regression model were used to account for potential confounding by injury complexity. Given the limited number of outcome events, the regression model was restricted to clinically relevant variables to minimize overfitting.
Results
A total of 148 patients were screened during the study period, of whom 86 met the eligibility criteria and were randomized in equal numbers to 2 groups. One patient in each group was lost to follow-up, leaving 84 patients (42 per group) for the final analysis, as depicted in the Consolidated Standards of Reporting Trials diagram (Fig. 1). Baseline demographic variables were comparable between groups; however, baseline injury characteristics demonstrated some imbalance between groups, particularly in fracture distribution and open injury rates. High-complexity injuries were comparable between groups (33% vs. 38%), although certain fracture subtypes were unevenly distributed. Given the known association between injury severity and HO formation, these differences were considered in adjusted analyses (Table I).
Figure 1.
CONSORT flow diagram depicting patient enrollment, randomization, allocation, follow-up, and analysis. CONSORT, Consolidated Standards of Reporting Trials.
Table I.
Baseline demographic characteristics of the study population, comparing age, sex distribution, and dominant limb involvement and fracture characteristics between the group A and group B cohorts.
| Variable | Group A (n = 42) | Group B (n = 42) | P value |
|---|---|---|---|
| Demographics | |||
| Age (yr) | 39.61 ± 14.39 | 36.66 ± 14.48 | .352∗ |
| Male sex | 35 (83.3%) | 27 (64.3%) | .082† |
| Dominant arm involved | 26 (62%) | 24 (56%) | —‡ |
| Fracture distribution (AO/OTA classification) | |||
| Distal humerus (AO 13) | 24 (57%) | 13 (31%) | — |
| 13-A | 3 | 2 | |
| 13 B | 5 | 3 | |
| 13-C1 | 10 | 3 | |
| 13-C2/C3 | 6 | 5 | |
| Olecranon (AO 2U1) | 12 (29%) | 11 (26%) | — |
| 21-A | 6 | 6 | |
| 21 B | 3 | 1 | |
| 21-C | 3 | 4 | |
| Radial head/neck (AO 2R1) | 1 (2%) | 9 (21%) | — |
| Simple (A/B) | 0 | 3 | |
| Complex (C) | 1 | 6 | |
| Ligamentous injuries | 2 (4%) | 3 (7%) | — |
| Terrible triad injuries | 3 (6%) | 1 (2%) | — |
| Coronoid fractures | 0 | 5 (12%) | — |
| Injury severity and associated patterns | |||
| High-complexity injuries§ | 14 (33%) | 16 (38%) | .82 |
| Fracture–dislocations | 1 | 1 | — |
| Combined fractures | 1 | 1 | — |
| Bilateral fractures | 1 | 0 | — |
| Open fractures | 5 (12%) | 9 (21%) | .38 |
AO, arbeitsgemeinschaft fur osteosynthesefragen; OTA, orthopaedic trauma association.
Statistical comparisons for fracture subtypes were not performed due to small subgroup sizes.
Welch two-sample t-test.
Pearson chi-square test for independence.
Counts not available for statistical comparison; percentages appear comparable.
High-complexity injuries include AO type 13-C2/C3, AO 21-C fractures, and fracture–dislocation patterns.
Incidence and severity of heterotopic ossification
HO occurred in 6 patients (14.3%) in Group A and 15 patients (35.7%) in Group B (P = .018) (Table II). This corresponded to an absolute risk reduction of 21.4%, a relative risk of 0.40, a relative risk reduction of 60%, an OR of 0.30 (95% CI: 0.10-0.85), and a number needed to treat of 5.
Table II.
Summary of HO incidence and HO severity stratified by group, including calculated severity scores.
| Variable | Group A (n = 42) | Group B (n = 42) |
|---|---|---|
| HO incidence (any grade) | 6/42 (14.3%) | 15/42 (35.7%) |
| Mild HO (G1-G2A) | 5/6 (83.3%) | 6/15 (40%) |
| Moderate HO (G2B) | 1/6 (16.7%) | 1/15 (6.7%) |
| Severe HO (G2C-G3) | 0/6 (0%) | 8/15 (53.3%) |
| Mean HO severity score | 1.33 | 2.93 |
HO, heterotopic ossification.
The severity of HO differed between the groups (Fig. 2). Group A demonstrated predominantly low-grade lesions: 1 Grade 1, 4 Grade 2A, and 1 Grade 2C, with no Grade 3 HO. In contrast, Group B showed a broader and more advanced distribution: 2 Grade 1, 4 Grade 2A, 1 Grade 2B, 4 Grade 2C, and 4 Grade 3C. Higher-grade lesions (Grade 2C or above) occurred almost exclusively in the control group, affecting 53.3% (8 of 15) of HO-positive patients in Group B compared with 16.7% (1 of 6) in Group A. This difference was statistically significant (Fisher exact test, P = .041), with Group B showing 5.7-fold higher odds of developing advanced HO (OR: 5.71). The mean severity score was lower in Group A (1.33 vs. 2.93) (Mann–Whitney U test, P = .013).
Figure 2.
Distribution of HO grades (Hastings and Graham) between the indomethacin and control groups, demonstrating reduced severity in Group A. FU, follow-up; HO, heterotopic ossification.
Temporal progression of heterotopic ossification
The chronological pattern of HO appearance also differed between the groups. No early HO was detected at the 6-week follow-up in Group A, whereas 2 patients in Group B already demonstrated radiographic changes. By 3 months, 2 cases were observed in Group A and 4 in Group B. HO detection continued to rise through 12 months, after which it plateaued. The final cumulative incidence was 14.28% in Group A and 35.7% in Group B (Fig. 3).
Figure 3.
Temporal progression of HO, showing cumulative incidence and time of radiographic appearance in both groups from 6 weeks to final follow-up. HO, heterotopic ossification; FU, follow-up.
Influence of injury pattern on heterotopic ossification formation
Higher-grade HO was more frequently observed in patients with complex injuries, particularly arbeitsgemeinschaft fur osteosynthesefragen (AO) type C fractures and combined injury patterns. Distal humerus AO type C fractures, especially AO 13C2/C3, were commonly associated with higher HO grades (≥ 2C), as shown in Table III. Similarly, combined injuries, including fracture–dislocations and those involving both bony and ligamentous structures, demonstrated higher HO grades compared to isolated fractures.
Table III.
Fracture characteristics of patients who developed HO and corresponding HO grades according to the Hastings and Graham classification.
| Fracture pattern | Hastings and Graham grading of HO |
|---|---|
| Group A (n = 6) | |
| AO 13B1 | 2C |
| AO 13C1 | 2A |
| AO 13C2 | 1 |
| AO 2R1C | 2A |
| Terrible triad (n = 2) | 2A |
| Group B (n = 15) | |
| AO 13C1 | 2A |
| AO 13C2 | 3C |
| AO 13C3 (n = 3) | 2B, 2C, 2C |
| Combined fracture of distal humerus and olecranon | 2C |
| Terrible triad | 2A |
| AO 2R1C (n = 2) | 1, 2C |
| AO 2U1A | 1, 2C |
| AO 2U1C | 2C |
| Ligamentous injuries (n = 2) | 2A, 2A |
| Fracture–dislocation | 3C |
AO, arbeitsgemeinschaft fur osteosynthesefragen; HO, heterotopic ossification; OTA, orthopaedic trauma association.
Across groups, HO in the indomethacin group was predominantly low grade, whereas the control group showed a broader distribution of HO across multiple fracture patterns, including more frequent involvement of proximal radius and ulna injuries as well as combined injury patterns.
When stratified by injury severity, the adjusted OR obtained from multivariable logistic regression was identical to the Mantel–Haenszel pooled estimate, reflecting consistency between analytical approaches: 0.31 (95% CI: 0.10-0.90).
Range of motion outcomes
ROM recovery differed across subgroups (Table IV, Table V; Fig. 4). At final follow-up, the overall ROM was greater in the indomethacin group compared to controls. The mean flexion–extension arc was 138° ± 7.4 in Group A and 118° ± 23.3 in Group B (P < .001). Similarly, the mean pronation–supination arc was higher in Group A (84° vs. 70°, P < .001). Subgroup analyses showed that the non-HO subgroup had a mean flexion–extension arc of 140° in Group A and 135° in Group B. The mean pronation–supination arc was 85° and 80°, respectively. In patients with HO, the mean final flexion–extension arc was 125° ± 6.6 in Group A and 88° ± 6.2 in Group B. The mean pronation–supination arc was 80° ± 3.8 in Group A and 52° ± 4.2 in Group B.
Table IV.
Final follow-up flexion–extension arc demonstrating overall motion, subgroup means, and magnitude of motion loss attributable to heterotopic ossification in each group.
| Group | Whole group mean | Non-HO mean ± SD | HO only mean ± SD | Mean loss due to HO | P value |
|---|---|---|---|---|---|
| A (Indomethacin) | 138° | 140° ± 4.9 | 125° ± 6.6 | 15° | <.001 |
| B (Control) | 118° | 135° ± 4.2 | 88° ± 6.2 | 47° | <.001 |
HO, heterotopic ossification; SD, standard deviation.
Table V.
Final follow-up pronation–supination arc showing differences between non-HO and HO subgroups and the degree of rotational loss associated with HO in both groups.
| Group | Whole group mean | Non-HO mean ± SD | HO only mean ± SD | Mean loss due to HO | P value |
|---|---|---|---|---|---|
| A | 84° | 85° ± 4.1 | 80° ± 3.8 | 5° | .01 |
| B | 70° | 80° ± 4.3 | 52° ± 4.2 | 28° | <.001 |
HO, heterotopic ossification; SD, standard deviation.
Figure 4.
Temporal progression of HO formation and ROM recovery in the 4 study subgroups. Panel A and B demonstrate flexion–extension and pronation–supination recovery trajectories over sequential follow-up intervals (6 weeks, 3 months, 6 months, 12 months, 24 months, and final follow-up) in the indomethacin (Group A) and control (Group B) cohorts, stratified by HO status (HO vs. non-HO). HO, heterotopic ossification; ROM, range of motion; Flex-Ext, flexion–extension; Pro-Sup, pronation–supination.
Adverse events
Indomethacin was generally well tolerated. Mild gastrointestinal symptoms, including abdominal discomfort and dyspepsia, were reported in 5 patients (12%) in the indomethacin group. No cases of gastrointestinal bleeding, renal impairment, allergic reaction, or drug-related serious adverse events were recorded. No cases of delayed union or nonunion were identified in either group.
Discussion
This prospective RCT demonstrates that patients receiving indomethacin developed HO less frequently and, when present, predominantly exhibited lower-grade lesions. These differences were also reflected in improved ROM outcomes among patients who developed HO. These findings parallel extensive evidence from hip and acetabular surgery, in which NSAIDs have consistently demonstrated efficacy in reducing HO formation.11
The observed number needed to treat of 5 highlights the clinical relevance of indomethacin prophylaxis, indicating that a small number of patients need to be treated to prevent one case of HO. Indomethacin was well tolerated in this cohort, with no serious adverse events observed, supporting its feasibility as a prophylactic intervention in this setting. However, the analyses related to HO severity and ROM should be interpreted as supportive and exploratory, as the number of events for analyses was limited.
Incidence of heterotopic ossification
Abrams et al1 reported HO rates of up to 62% following elbow injuries, substantially higher than those observed in our cohort. Variations in injury pattern, surgical approach, and the exclusion of high-risk patients, such as those with burns or head injuries, likely contributed to these differences. Biologically, HO formation is driven by inflammatory signaling and prostaglandin-mediated osteogenic differentiation of mesenchymal stem cells. Cyclooxygenase inhibition by indomethacin reduces prostaglandin production, thereby limiting osteogenic differentiation of mesenchymal precursor cells and early ectopic bone formation. This mechanism is consistent with both experimental and clinical studies demonstrating reduced HO formation with NSAID use.
While several observational studies have suggested a protective effect, recent randomized data have reported conflicting findings. A recent randomized controlled study18 did not demonstrate a significant difference in HO incidence between indomethacin and placebo following elbow trauma. Differences in study design, patient selection, injury severity, and outcome assessment may account for these discrepancies. In contrast, the present study demonstrates a consistent reduction in both HO incidence and severity, particularly in clinically relevant higher-grade lesions.
Surgical technique and approach are recognized factors influencing HO following elbow trauma. In the present study, all procedures were performed by a single experienced surgeon using standardized principles of fixation and soft-tissue handling, thereby reducing variability related to operative technique. While individual fixation constructs varied based on fracture configuration, this reflects real-world clinical practice. Importantly, these variables were consistent across both study groups, suggesting that differences in HO incidence and severity are unlikely to be solely attributable to surgical factors.
Influence of injury severity on heterotopic ossification formation
A key consideration in interpreting these findings is the influence of injury severity on HO formation. In the present study, HO was more frequently associated with complex AO type C fractures and combined injury patterns, consistent with existing literature.4 As a tertiary referral center, our cohort included a higher proportion of high-energy and complex injuries, inherently associated with an increased baseline risk of HO. This provided a clinically relevant setting to evaluate the potential effect of indomethacin prophylaxis.
Although certain fracture subtypes were unevenly distributed between groups, the proportion of high-complexity injuries, an important determinant of HO risk, was comparable. This suggests that the observed reduction in HO incidence and severity with indomethacin is unlikely to be solely attributable to baseline differences in injury characteristics, although residual confounding cannot be entirely excluded. Furthermore, the consistency of findings across both stratified (Mantel–Haenszel) and multivariable regression analyses supports the robustness of the observed association and indicates that the treatment effect is not sensitive to the method of adjustment.
The temporal trend observed in this study suggests that indomethacin may influence both the timing and progression of HO formation. Earlier radiographic appearance and progression to higher-grade lesions were more frequently observed in the control group, whereas HO in the indomethacin group was delayed and remained predominantly low grade. This underscores the biologic plausibility of NSAID prophylaxis as a modulator of the inflammatory–osteogenic cascade that drives early HO maturation.
Functional impact of heterotopic ossification
Patients with HO in the control group demonstrated substantially greater loss of motion compared to those receiving indomethacin, particularly in the flexion–extension arc in the subgroup analyses. These differences were primarily attributable to the lower incidence and severity of HO in the indomethacin group, as reflected by the subgroup analysis. However, given that functional outcomes are influenced by multiple factors, including rehabilitation and injury characteristics, these findings should be interpreted cautiously. While the diagnosis and grading based on the interpretation of a single blinded radiologist and physiotherapist ensured consistency in grading, it may limit generalizability.
From a clinical perspective, the prevention of HO is particularly relevant in elbow trauma, where even small amounts of ectopic bone can significantly impair motion and function. Although concerns have been raised regarding the potential impact of NSAIDs on fracture healing, no cases of delayed union or nonunion were observed in this cohort. In addition, indomethacin was well tolerated, with only minor gastrointestinal symptoms reported. These findings support the short-term safety of indomethacin in this setting, although larger studies are required to fully evaluate rare adverse events. The findings of this study suggest that a short course of indomethacin may be a simple, low-cost intervention to reduce the risk of clinically significant HO, particularly in patients with high-risk injury patterns. Given its ease of administration and good tolerability observed in this study, indomethacin prophylaxis may be considered as an adjunct in the management of complex elbow injuries.
Limitations
This study has several limitations. First, it was conducted at a single center with a modest sample size and a limited number of outcome events, which may limit generalizability and the precision of effect estimates, particularly for subgroup and severity analyses. Second, baseline differences in injury characteristics were observed between groups, including variation in fracture subtypes. Although the proportion of high-complexity injuries was comparable and stratified analyses demonstrated consistent treatment effects across injury severity levels, residual confounding cannot be entirely excluded.
Third, the absence of a placebo control and the lack of patient blinding may have introduced performance and reporting bias, particularly for rehabilitation-dependent outcomes such as RO. However, outcome assessment was performed by blinded evaluators, which reduces the likelihood of measurement bias. In addition, while allocation concealment was ensured, treating surgeons were aware of group allocation, which may represent a potential source of bias.
Fourth, only a single NSAID regimen was evaluated, and the findings may not be generalizable to other agents, dosages, or durations of prophylaxis. Finally, the trial was registered retrospectively, which represents a methodological limitation. However, the study protocol, eligibility criteria, and outcome measures were predefined prior to patient enrollment and were not modified during the course of the study.
Despite these limitations, the present study provides supportive evidence supporting indomethacin as a safe and effective prophylactic strategy for reducing HO incidence, limiting radiographic progression, and improving functional outcomes following traumatic elbow fixation.
Conclusion
A 3-week course of oral indomethacin was associated with a reduction in both the incidence of HO following surgical fixation of elbow trauma, which was consistent across injury severity strata. Secondary analyses suggest a potential reduction in severity and improved functional outcomes; however, these findings should be interpreted cautiously given the modest sample size. Furthermore, multicenter studies are required to confirm these results.
Disclaimers:
Funding: No funding was disclosed by the authors.
Conflicts of interest: The authors, their immediate families, and any research foundations with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.
Footnotes
This study was approved by the Institutional Ethics Committee (IEC) of the Post Graduate Institute of Medical Education and Research (PGIMER), Chandigarh. The approval number is INT/IEC/2021/SPL-1793, dated 27/12/2021.
References
- 1.Abrams G.D., Bellino M.J., Cheung E.V. Risk factors for development of heterotopic ossification of the elbow after fracture fixation. J Shoulder Elbow Surg. 2012;21:1550–1554. doi: 10.1016/j.jse.2012.05.040. [DOI] [PubMed] [Google Scholar]
- 2.Ahrengart L., Lindgren U., Reinholt F.P. Comparative study of the effects of radiation, indomethacin, prednisolone, and ethane-1-hydroxy-1,1-diphosphonate (EHDP) in the prevention of ectopic bone formation. Clin Orthop. 1988;229:265–273. [PubMed] [Google Scholar]
- 3.Bartlett C.S., Rapuano B.E., Lorich D.G., Wu T., Anderson R.C., Tomin E., et al. Early changes in prostaglandins precede bone formation in a rabbit model of heterotopic ossification. Bone. 2006;38:322–332. doi: 10.1016/j.bone.2005.08.016. [DOI] [PubMed] [Google Scholar]
- 4.Foruria A.M., Augustin S., Morrey B.F., Sánchez-Sotelo J. Heterotopic ossification after surgery for fractures and fracture-dislocations involving the proximal aspect of the radius or ulna. J Bone Joint Surg Am. 2013;95 doi: 10.2106/JBJS.K.01533. [DOI] [PubMed] [Google Scholar]
- 5.Foruria A.M., Lawrence T.M., Augustin S., Morrey B.F., Sanchez-Sotelo J. Heterotopic ossification after surgery for distal humeral fractures. Bone Joint J. 2014;96-B:1681–1687. doi: 10.1302/0301-620X.96B12.34091. [DOI] [PubMed] [Google Scholar]
- 6.Furia J.P., Pellegrini V.D. Heterotopic ossification following primary total knee arthroplasty. J Arthroplasty. 1995;10:413–419. doi: 10.1016/s0883-5403(05)80139-1. [DOI] [PubMed] [Google Scholar]
- 7.Herman Z.J., Edelman D.G., Ilyas A.M. Heterotopic ossification after elbow fractures. Orthopedics. 2021;44:10–16. doi: 10.3928/01477447-20201119-03. [DOI] [PubMed] [Google Scholar]
- 8.Hong C.C., Nashi N., Hey H.W., Chee Y.H., Murphy D. Clinically relevant heterotopic ossification after elbow fracture surgery: a risk factors study. Orthop Traumatol Surg Res OTSR. 2015;101:209–213. doi: 10.1016/j.otsr.2014.10.021. [DOI] [PubMed] [Google Scholar]
- 9.Kölbl O., Knelles D., Barthel T., Kraus U., Flentje M., Eulert J. Randomized trial comparing early postoperative irradiation vs. the use of nonsteroidal antiinflammatory drugs for prevention of heterotopic ossification following prosthetic total hip replacement. Int J Radiat Oncol Biol Phys. 1997;39:961–966. doi: 10.1016/s0360-3016(97)00496-3. [DOI] [PubMed] [Google Scholar]
- 10.McCarthy E.F., Sundaram M. Heterotopic ossification: a review. Skeletal Radiol. 2005;34:609–619. doi: 10.1007/s00256-005-0958-z. [DOI] [PubMed] [Google Scholar]
- 11.McMahon J.S., Waddell J.P., Morton J. Effect of short-course indomethacin on heterotopic bone formation after uncemented total hip arthroplasty. J Arthroplasty. 1991;6:259–264. doi: 10.1016/s0883-5403(06)80173-7. [DOI] [PubMed] [Google Scholar]
- 12.Shehab D., Elgazzar A.H., Collier B.D. Heterotopic ossification. J Nucl Med. 2002;43:346–353. [PubMed] [Google Scholar]
- 13.Søballe K., Christensen F., Kristensen S.S. Ectopic bone formation after total hip arthroplasty. Clin Orthop. 1988;228:57–62. [PubMed] [Google Scholar]
- 14.Thompson H.C., Garcia A. Myositis ossificans: aftermath of elbow injuries. Clin Orthop. 1967;50:129–134. [PubMed] [Google Scholar]
- 15.Toyoda T., Matsumoto H., Tsuji T., Kinouchi J., Fujikawa K. Heterotopic ossification after total knee arthroplasty. J Arthroplasty. 2003;18:760–764. doi: 10.1016/s0883-5403(03)00194-3. [DOI] [PubMed] [Google Scholar]
- 16.Vanden Bossche L., Vanderstraeten G. Heterotopic ossification: a review. J Rehabil Med. 2005;37:129–136. doi: 10.1080/16501970510027628. [DOI] [PubMed] [Google Scholar]
- 17.Vastel L., Kerboull L., Dejean O., Courpied J.P., Kerboull M. Prevention of heterotopic ossification in hip arthroplasty. Int Orthop. 1999;23:107–110. doi: 10.1007/s002640050319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Atwan Y., Abdulla I., Grewal R., Faber K.J., King G.J.W., Athwal G.S. Indomethacin for heterotopic ossification prophylaxis following surgical treatment of elbow trauma: a randomized controlled trial. J Shoulder Elbow Surg. 2023;32:1242–1248. doi: 10.1016/j.jse.2023.02.119. [DOI] [PubMed] [Google Scholar]