Introduction

Heterotopic ossification is a condition of abnormal osteogenesis in the non-skeletal tissue. Although it is not a malignant condition, the sequelae associated with HO can range from being asymptomatic to severe ankylosis of the joint with loss of range of motion [1]. Other conditions such as infection including osteomyelitis, thrombophlebitis, cellulitis, deep vein thrombosis, and benign and malignant tumors, can mimic heterotopic bone formation clinically and on imaging studies and should be properly differentiated [2,3,4]. On imaging studies, clear localization of abnormalities is crucial for proper detection of HO.

The diagnosis of heterotopic ossification is based on radiographic findings; however, X-rays can be normal for up to 4 weeks or longer even after the onset of the condition. Bone scintigraphy is considered the method of choice for the earliest detection of heterotopic ossification and for assessing its maturity [5].

Accurate localization of the uptake on bone scintigraphy is the key to differentiating the condition from other pathologies. This is considered a dilemma on planar bone scintigraphy. The addition of SPECT imaging of the area of concern can help to localize the uptake; however, registering the acquired data from SPECT with low-dose CT images obtained from the new hybrid systems SPECT-CT should provide superior activity localization. No studies on the value of CT component on localization and detection of HO have been performed, although studies on the use of SPECT/CT have evaluated the stage of maturation and the metabolism of neurogenic HO in a limited number of patients [6,7,8]. The purpose of this study was to find the added value of SPECT-CT in the scintigraphic detection of heterotopic ossification.

Materials and methods

All reports on the PACS reporting system of patients that underwent bone scintigraphy at the Department of Nuclear Medicine in Mubarak Alkabeer Hospital from the year 2010 to 2017 were filtered for the term “heterotopic ossification or heterotopic bone formation” and “SPECT-CT” either mentioned in the study request or the report of the patients’ studies. In the pool of 1780 cases, only 34 patients that had both three-phase bone scintigraphy and SPECT-CT in the same occasion fulfilled this criterion and were included in the study. None of the 34 patients had follow-up bone scintigraphy. For each case, two sets of data were read on two separate sessions approximately 7 days apart. The first set included “planer images with SPECT only” and the second set included “planer images with SPECT-CT”. The two sets were reviewed separately by two experienced nuclear physicians independently. Both were blinded to the original report. Any disagreement in interpretation was resolved by consensus.

Planer images included blood flow, blood pool, and delayed images including multiple spot views of the area of concern. SPECT images were displayed in the transaxial, coronal, and sagittal planes with the maximum intensity projection MIP images. The CT component was displayed in the tomographic planes as fused and non-fused sets with SPECT images. The certainty of the presence of extra-skeletal soft tissue ossification on the planar alone and planar with SPECT were graded as definitely present, maybe present, or not present, in which case no uptake was seen, or there is uptake which could not be viewed as HO due to lack of separation from bony structures. On another session, findings of planar and SPECT were correlated with SPECT-CT images, which was used to confirm the presence of HO as positive or negative.

For bone scintigraphy, a dual-head gamma camera equipped with a large field of view, low-energy, high-resolution, parallel hole collimator was used with energy window set at 20% centered at 140 keV. All patients received 740–925 MBq (20–25 mCi) of Tc-99 m-methylene diphosphonate (MDP) intravenously. The IV bolus was administered while the patient is under the camera and dynamic imaging of the area of interest is acquired. Flow images were obtained at 3 s/20 frames for 60 s using a 64 × 64 matrix while the patient under the camera is receiving IV bolus of Tc-99 m MDP. Immediate regional blood pool images were acquired for 500 k counts using a 128 × 128 matrix of the area of interest. Delayed whole-body imaging was then performed 3 h post-injection using 256 × 1024 matrix with a speed of 8 min/m where the patient was positioned supine, arms at side, with a knee pillow and foot band in place. This is followed by static views of the area of interest using a 256 × 256 matrix. SPECT/CT was then performed of the area of interest. For SPECT, a matrix size of 64 × 64 with non-circular acquisition mode was used. Sixty projections were acquired at a speed of 20 s/projection with a 6° angle and 180° arch. A Butterworth filter was used with a cut-off of 0.43, order 7. For CT, helical acquisition with the voltage set at 140 (kV), current at 80 (mA), pitch of 0.5, rotation speed of 1 s/rotation, and slice thickness of 2.5 mm.

Statistical analysis

IBM’s Statistical Package for Social Sciences version 23 (SPSS-Inc., Chicago, IL, USA) was used to perform all statistical analysis. The independent samples t test was conducted to compare the numbers of cases with definite heterotypic ossification by planar images with SPECT/CT vs. planar images with SPECT alone in order to determine statistical significance.

Results

Thirty-four patients who fulfilled the selection criteria (mean age, 54.2 years; median age 58 years; age range, 21–81 years) were included. All 34 patients had prior radiographs at least 4 weeks prior to the bone scintigraphy. The results in all were either inconclusive or showed no evidence of HO.

The time interval between the inciting event and the bone scintigraphy date ranged from 3 months to 6 years. Nine patients had a prior history of arthroplasty, 18 had traumatic injury resulting from fall, sport, or road traffic injury, and one patient had spinal cord injury. The location of the lesions, history of inciting events, and final diagnosis are all shown in Tables 1 and 2.

Table 1 Details of patients with positive results for HO on planer images and SPECT/CT
Full size table
Table 2 Details of patients with negative results for HO on planer images and SPECT/CT
Full size table

There were 19 (54%) males and 15 (44%) females. The results of final consensus showing the degree of certainty for detecting heterotopic ossification is displayed in Table 3 comparing planar images with SPECT only to planer images with SPECT/CT.

Table 3 Comparison of planar images with SPECT only to planer images with SPECT/CT for the detection of heterotopic ossification
Full size table

Based on the SPECT/CT findings, 23 (67.6%) patients were found to have heterotopic ossification, and 11 (32.4%) did not have findings to confirm heterotopic ossification. Of the 23 positive patients on planer with SPECT/CT, only eight (34.8%) showed definitely present HO on planer with SPECT alone in which CT did not add to the certainty of the diagnosis. Twelve (52.2%) patients were categorized under maybe present on planer with SPECT only. Three (13%) patients were negative (not viewed as HO) on the planer with SPECT alone and showed evidence of heterotopic ossification when the CT component was added. In two of these cases, the lesions were at the vicinity of the right greater trochanteric region and in the third case the lesion was at the lateral aspect of the left femoral condyle. All of these lesions were at least 12 mm in the shortest dimension, which excludes the partial volume effect as a cause of non-visualization on the planer and SPECT images.

Of the 11 patients that did not demonstrate evidence of heterotopic ossification on planer with SPECT/CT, 7/11 (63.6%) were negative on planer with SPECT alone and 4/11 (36.4%) were under the unclear category “maybe present” on planer with SPECT alone. These four cases showed osteoarthritic changes on the CT component.

The t test conducted to compare the means of HO in the two groups showed a P value of 0.014, indicating a statistical significance in determining the definite presence of HO with planar with SPECT/CT vs. planar with SPECT alone.

Discussion

Heterotopic ossification, in etiology, can be either hereditary or acquired. The rare hereditary form of HO can be of two types; fibrodysplasia ossificans progressiva or myositis ossificans progressiva and progressive osseous heteroplasia, which are both due to gene mutations [9].

In the common acquired form, it is mainly following trauma including surgery along the long bones leading to the entity of myositis ossificans; or following nervous system insult, most commonly spinal cord injury, leading to the neurogenic heterotopic ossification. It may occur in less common forms as in tumoral calcinosis or calcification within end-stage valvular disease. The pathophysiology of HO is believed to be secondary to the transformation of pluripotent mesenchymal cells present in the fibrous and connective tissue septa within muscle into osteogenic cells line that takes place under permissive environment and induction agent [10]. This explains the formation of the heterotopic bone at some distance from normal bone, moving towards it later.

The onset of HO can occur as early as 3 weeks or as late as 12 weeks after injury, yet most commonly occurring at 2 months post-injury [4].

The acquired heterotopic bone formation will have a relatively benign course in more than 80% of cases. However, significant loss of motion and ankylosis can occur in up to 10% of the remaining cases, which may result in loss of function in the affected part [11,12,13]. Therefore, early intervention and treatment are essential as soon as the diagnosis of early HO has been confirmed. Treatment choices can be in the form of passive range-of-motion exercises, diphosphonates, and nonsteroidal anti-inflammatory drugs, radiation therapy and surgical resection of mature HO. The treatment choice depends on clinical, laboratory, radiographic, and scintigraphic criteria. However, multiphase bone scintigraphy is the most useful investigation in detecting early HO at the onset of clinical symptoms and disease monitoring as in maturity and response to therapy [14]. Assessment of the maturity of the HO is essential because of the fact that resection before maturity almost always leads to a recurrence of HO. In addition, if excision is performed before the heterotopic ossification has matured, incomplete and inadequate excision may result.

CT scan has low specificity in the early stage of HO, and its primary role focuses on staging the extent of bone formation and helping guide radiotherapy or surgery [15]. The superiority of the multi-phase bone scintigraphy lies in its early changes particularly on flow and blood pool images that can show evidence of HO, even in the absence of bone uptake on delayed images and if an abnormal alkaline phosphatase is the only evidence of disease activity [16].

The diagnosis of HO can be difficult since clinically it may present similar to other conditions particularly infections and tumors [3, 4, 17]. Scintigraphic findings can mimic several conditions and may lead to incorrect diagnosis [3]. The identification of the abnormality as extraskeletal is crucial. This can be difficult since abnormalities are commonly adjacent to bony structures (Fig. 1). This study involves cases with heterotopic ossification in patients who had SPECT/CT to assess value of the CT component in determining precisely the location of osteoblastic activity in relation to bony structures as a key to the diagnosis of the condition (Fig. 2).

Fig. 1
figure 1figure 1

a A 67-year-old female with left shoulder pain and limitation of movement for 2 months. The delayed whole-body images show intense uptake within the left shoulder joint region corresponding to the area of moderate hyperemia on the blood pool images. b Fused SPECT-CT images localize increased tracer uptake within the left shoulder joint and the surrounding soft tissue calcification, indicating heterotopic ossification. These findings were in consensus reading

Full size image
Fig. 2
figure 2figure 2

a A 44-year-old male 6 months after revision of total right knee replacement complaining of pain in the right knee and limitation of movement. Immediate whole-body images demonstrate marked hyperemia surrounding the right knee joint mainly superiorly with delayed images showing intense uptake within the anterior aspect of the distal right femur. b Fused SPECT-CT images demonstrate increased tracer uptake within the right distal posterior femoral cortex as well as within the soft tissue calcification anteriorly indicating heterotopic ossification. These findings were in concordance with consensus reading

Full size image

A study on 12 patients with spinal cord injury by Lima et al. used SPECT/CT to determine osteoblastic activity in patients to determinate which patients have a greater risk of relapse after surgical resection, however the study did not look at the role of CT in detecting HO but rather in evaluating activity to determine maturity [6]. A case was reported by Lin in a patient with paraplegia 6 months after brain injury about the value of SPECT/CT in determining activity and help directing management to conservative rather than surgery [7]. Another case of a patient with Guillain–Barré syndrome was reported showing the value of SPECT/CT in localizing the abnormal activity to the soft tissues surrounding the hips [8]. However, no study was found to assess the value of SPECT/CT in differentiating HO from other mimicking conditions in a large number of patients.

Our study showed the value of adding CT component in the detection of HO in 34 patients. In our study, a significant number of cases were not diagnosed with certainty on planar or SPECT and confirmed by adding the CT component due to more accurate localization of activity. In this study, 65.2% (15 of the total 23 positive cases) of cases were missed using SPECT only for the confirmation of the presence of heterotopic ossification attributed mainly to poor anatomical localization of bone scintigraphy of the extra-skeletal uptake in the absence of hybrid imaging. Additionally, in our study adding CT excluded the presence of heterotopic ossification in 36.4% of cases in which the findings were unclear on SPECT alone.

In the three cases where the SPECT alone did not show extraskeletal uptake, CT findings were confirmatory for the presence of heterotopic ossification. This is attributed to the full maturity of the heterotopic ossification in these cases in which the scintigraphic uptake normalized and is therefore not seen on SPECT images.

Conclusions

Although the study population is small, our findings suggest that SPECT-CT bone scintigraphy is a useful tool that helps in accurately detecting heterotopic ossification and improving the level of confidence in reporting this condition. This allows physicians to differentiate heterotopic bone from other mimicking diseases.