Translate this page into:
Imaging in osteomyelitis: Unveiling the secrets of bone infections
-
Received: ,
Accepted: ,
How to cite this article: Gupta NA, Tyagi V, Saxena S, Katyal A, Shah AB. Imaging in osteomyelitis: Unveiling the secrets of bone infections. Indian J Musculoskelet Radiol. 2026;8:85-96. doi: 10.25259/IJMSR_55_2025
Abstract
“Osteomyelitis” or “infection of the bone marrow” is a commonly encountered clinical scenario in which imaging plays a crucial role in diagnosis, localization, and follow-up. The mode of spread as well as pathophysiology of osteomyelitis (OM) differ in various situations, and also depend on the age of the patient. Direct spread of OM is commonly found in adults with certain risk factors, whereas indirect/hematogenous spread of OM is more frequently seen in the pediatric age group. The imaging findings differ based on the mode of spread, the stage of disease, and the age of the patient. Choosing the appropriate imaging modality for the particular clinical scenario is of utmost importance for guiding patient management. We discuss the role of various imaging modalities in these diverse scenarios and common imaging findings.
Keywords
Abscess
Bone infection
Infection
Osteomyelitis
INTRODUCTION
Osteomyelitis (OM) is an inflammatory process involving the bone marrow secondary to infection. It is an important cause of morbidity in children as well as adults. OM may occur due to direct spread following a penetrating injury or prior surgery, or due to indirect (hematogenous) spread, which is more commonly found in immunocompromised patients.[1]
The diagnosis of OM is based on a combination of clinical signs, symptoms, laboratory tests, microbiological evaluation of blood or joint fluid, and imaging modalities. Laboratory tests may reveal elevated C-reactive protein, elevated erythrocyte sedimentation rate, raised white cell counts, and positive blood cultures. Clinically, patients present with fever, pain, reduction in mobility of the involved limb or joint, joint swelling, erythema, and/or a discharging wound/sinus tract. However, these classical symptoms and signs may be absent in infants, who may present only with poor feeding or irritability.[2] To add to the diagnostic dilemma, serum inflammatory markers may be normal in neonates,[3] as they may be in cases of chronic OM. In such cases, imaging plays a crucial role in establishing the diagnosis of OM.
The role of imaging in OM includes confirmation of the diagnosis of OM and ruling out other differential diagnoses, determining the extent of infection, guiding aspiration or biopsy, as well as in follow-up and evaluating complications.[2] Imaging may also provide important information about the involvement of important adjacent structures, such as joints and growth plates. In cases of penetrating injury, a retained foreign body may also be demonstrated on imaging.
In all cases of suspected OM, a bone biopsy with bacterial culture should be considered for definitive diagnosis.
DISCUSSION
Pathophysiology of OM
OM can occur at any age, from childhood to the elderly. The occurrence of OM in different age groups depends on the type of spread of OM.
Routes of spread and age-wise differences
Indirect (hematogenous) spread of OM is more commonly seen in the pediatric age group, particularly between 2 and 12 years of age.[4] Hematogenous spread is less common in adults,[2] and, if it occurs, it involves the axial skeleton or vertebral column, as opposed to more predominant involvement of the appendicular skeleton or extremities in children.
Direct OM may involve any site and age group. Direct OM may occur due to direct inoculation of microorganisms following open compound fractures or penetrating injuries or surgery with insertion of metallic implants, or due to contiguous spread from adjacent soft-tissue infection. Contiguous spread of direct OM is most commonly seen in patients with diabetes mellitus with peripheral vascular disease or neuropathy, presenting with foot ulcers or wounds. It also commonly affects the lower extremities and the pelvis in people with paralysis, due to predisposition to repeated microtrauma in the presence of peripheral neuropathy.
Pathogens
Staphylococcus aureus is the most common pathogen responsible for OM in children, most of which are caused by the community-acquired methicillin-resistant strains.[2,5,6] Other common organisms include Staphylococcus epidermidis and Enterobacter.[2] Pseudomonas and Salmonella species predominate in certain clinical scenarios, such as intravenous drug users and sickle-cell anemia, respectively.[2] Atypical OM, on the other hand, is caused by organisms including Mycobacterium species, fungi (Histoplasma, Cryptococcus), Coxiella and Bartonella, and is usually endemic to certain areas or affects immunosuppressed patients.
Acute OM
Acute OM begins with inoculation of bacteria within the bone marrow, which incites an inflammatory response. The metaphyses of long bones and metaphyseal equivalents are the most commonly involved sites in hematogenous OM, predisposed by the slow-flowing blood in the metaphyseal vessels, leading to bacterial proliferation. Further progression of the disease results in a characteristic pattern of pathological features depending on the patient’s age. There is hyperemia with production and accumulation of pus within the medullary cavity, increasing the intramedullary pressure, which disrupts the intraosseous blood supply. Hypervascular granulation tissue and reactive bone form around the intramedullary pus, forming a well-circumscribed intramedullary abscess known as Brodie’s abscess.[7] These intramedullary abscesses are a characteristic feature of hematogenous OM and are not frequently found in direct OM [Figure 1].[2] In children, the cortical bone is thin and ruptures easily, due to which the intramedullary abscess more frequently spreads into the subperiosteal space. The periosteum is also more loosely attached to the cortex in this age group, leading to the accumulation of pus underneath the periosteum.[8]

- Illustrations showing the pathophysiology of hematogenous spread of osteomyelitis. (a) Bacterial seeding occurs in the highly vascular metaphyses and metaphyseal equivalents. (b) Formation of pus and increased intramedullary pressure which causes impedance to the blood supply, which eventually leads to osteonecrosis. (c) Infection spreads along the vascular channels to the overlying cortex and subperiosteal space. (d) Further increase in intramedullary pressure leads to formation of sinus tracts and cloacae (arrow in d). The necrotic bone fragment (sequestrum - asterisk in d) may get trapped within the infected bone with a surrounding envelope of new bone (involucrum).
Chronic OM
Cases of acute OM who receive inadequate treatment progress to chronic OM. The disruption of intraosseous and periosteal blood supply in the acute phase leads to osteonecrosis, which results in the formation of a necrotic bone fragment called “sequestrum,” which is separated from the living bone by granulation tissue. This devascularized sequestrum acts as a nidus of persistent infection, as it is protected from antibiotics as well as the body’s endogenous immune response. As an attempt to wall off the sequestrum, new bone formation occurs around it, leading to involucrum formation. To normalize the elevated intramedullary pressure, the intramedullary pus eventually decompresses through the bony cortex or involucrum through a tract known as “cloaca,” forming a subperiosteal abscess. Subsequently, continued accumulation of subperiosteal pus leads to rupture of the periosteum and spread of the pus into the surrounding muscles and the overlying skin surface, forming a discharging sinus tract.
Imaging modalities
A plain radiograph should always be obtained first to exclude other possible etiologies. Unless contraindicated, a magnetic resonance imaging (MRI) should then be performed, as it is the best currently available modality for establishing the diagnosis of OM. If MRI is contraindicated, computed tomography (CT) or nuclear medicine studies can be obtained, although these tests are of limited sensitivity and specificity compared to MRI.
The imaging appearances of mimics of OM will be discussed in detail in another section [Tables 1 and 2].
| Modality | Approx. sensitivity | Approx. specificity | Key advantages | Key limitations |
|---|---|---|---|---|
| X-ray (radiograph) | Low early (~0–33% in first 1–2 weeks; upto~90% in late/chronic stage) | Moderate (specific if classic changes present, but many mimics) | Widely available, inexpensive, initial screening tool. Can detect chronic changes (sequestra, sclerosis) and alternative diagnoses. | Insensitivity:often normal in early infection. Lacks marrow/soft-tissue detail. Changes are delayed and nonspecific early (e.g. osteopenia, periostitis). |
| CT | Moderate (~67% sens in some studies) | Low-Moderate (e.g. ~50% spec; better for chronic changes) | Excellent bone detail -best for identifying sequestrum/involucrum. Good for complex anatomy (spine/pelvis) and guiding biopsies. Can detect gas and foreign bodies. Faster than MRI. | Limited soft tissue contrast- misses early marrow edema. Struggles to differentiate active versus inactive infection. Radiation exposure. Metal hardware artifacts can obscure area. |
| MRI | High (~90-100%) | High (~80-100%) | Highest sensitivity for early infection (detects marrow edema within days). Great soft tissue visualization -maps abscesses, sinus tracts. Multiplanar and no ionizing radiation. Particularly useful for spine and foot infections. A normal MRI almost rules out osteomyelitis. | Expensive and may have limited availability. Contraindications/artifacts: can’t use with some implants; metal causes artifacts (though improvinig). Requires patient cooperation (motion sensitive). Possible sedation in kids. Gadolinium contraindicated in certain patients (renal failure) |
| 3-phase Bone scan (Tc-99m) | Very high in general (>90% in adults; can detect osteo~48 h after onset) | Low (poor specificity without additional imaging -uptake in any increased bone turnover) | Highly sensitive- useful for screening and detecting occult or multiple lesions. Whole body imaging in one exam. Early positive before X-ray changes. Widely available. | Lacks specificity- positive in fractures, arthritis, postoperative changes. Low spatial resolution, cannot distinguish bone vs soft tissue infection by itself. Some radiation exposure. |
| Labeled WBC Scan (± marrow scan) | High (indum - WBC~83%; Tc-HMPAO WBC~95%) | High (with marrow scan: Up to~90–95%) | Targets active infection -high specificity especially when combined with marrow imaging. Helpful in equivocal cases (diabetic foot, prosthetic joints) to confirm bone infection. SPECT/CT improves localization. Not affected by bone turnover from other causes as much. | Labor-intensive (blood handling, delayed imaging). Slower results (e.g. 1 day). Moderate radiation. Not as sensitive in low-grade/chronic infections without neutrophils. Poorer images versus MRI/CT. Limited use in children. |
| FDG-PET/CT | High (84–100% in chronic osteo) | High (80–95%) | Whole-body survey in one session - great for multifocal disease. Very effective for chronic osteomyelitis and spinal infections. Rapid results (within 1 h). High NPV - negative scan confidently excludes osteomyelitis. Good anatomical detail with PET/CT fusion | False positives: FDG accumulates in sterile inflammation, healing tissue, and tumors (reduced specificity early post-op). Costly and not universally available. Limited sensitivity in acute infection of small bones (movement can blur PET). Best used after >6–12 months post-surgery in prosthetic cases to avoid confounding uptake. |
CT: Computed tomography, MRI: Magnetic resonance imaging, FDG: Fluorodeoxyglucose, PET: Positron emission tomography, WBC: White blood cells
| Clinical scenario | Preferred imaging modality | Justification |
|---|---|---|
| Initial evaluation of suspected osteomyelitis (any site) | Plain radiograph (X-ray) | First-line imaging to detect bone destruction, fractures, foreign bodies; rules out mimics; inexpensive and quick |
| Early detection and assessment of extent (especially in extremities or spine) | MRI | Highest sensitivity and specificity; detects early marrow edema, abscesses, soft tissue involvement |
| Contraindication to MRI (e.g. pacemaker, severe claustrophobia) | 3-phase bone scan±WBC scan | Bone scan highly sensitive; add WBC scan for specificity in differentiating infection from trauma or degenerative changes |
| Diabetic foot ulcer with concern for osteomyelitis | MRI (or WBC scan if MRI contraindicated) | MRI detects early marrow changes and extent; WBC scan differentiates Charcot vs infection |
| Suspected osteomyelitis with hardware/prosthetic joint infection | Labeled WBC scan+marrow scan OR FDG-PET/CT | MRI is limited by artifact; WBC scan with marrow improves specificity; FDG-PET useful if WBC scan not available |
| Suspected vertebral osteomyelitis/discitis | MRI with contrast | Best for evaluating endplates, disc, paraspinal and epidural abscesses |
| Multifocal or unclear site of infection (e.g. fever or unknown origin, bacteremia) | FDG-PET/CT OR Bone scan | Whole-body imaging; detects multiple sites; FDG-PET more specific; bone scan good screening tool |
| Chronic osteomyelitis with suspected sequestrum or involucrum | CT | Best bone detail; detects cortical destruction, sequestra, cloaca; guides surgical planning |
| Subperiosteal abscess or joint effusion in children | Ultrasound | Identifies collections for drainage; non-invasive and no radiation |
| Monitoring response to therapy or recurrence suspicion | MRI (for local follow-up) OR FDG-PET (for systemic) | MRI monitors abscess resolution; PET detects residual active inflammation/metabolic activity |
CT: Computed tomography, MRI: Magnetic resonance imaging, FDG: Fluorodeoxyglucose, PET: Positron emission tomography, WBC: White blood cells
Imaging findings
The imaging appearance of OM on various modalities depends on the type of spread (indirect or direct) and also on the stage of disease (acute or subacute, or chronic).
Plain radiographs
Plain radiographs are the initial modality of choice in the evaluation of cases of OM; however, they lack sensitivity and specificity in the early stages of OM. Radiographs may be normal in the first 2 weeks of OM. However, they may help in ruling out other differentials of pain, such as fractures or tumors.
The earliest finding on radiographs is soft tissue swelling and effacement of the surrounding fat planes. It has been reported that only when OM compromises 30–50% of the bone mineral content, noticeable changes are seen on radiographs.[9] Bony changes that are seen in the early stages include periosteal reaction in the form of periosteal elevation or thickening, osteopenia, loss of normal appearance of bony trabeculae, or lytic changes. In the subacute stage, intramedullary lytic lesions representing intramedullary abscesses may be found, typically in the metaphyseal regions. The pathognomonic finding in chronic OM is the presence of sequestrum, which is a necrotic bone fragment showing density higher than the adjacent unaffected bone. Surrounding new bone (involucrum) and cloacae are also seen on radiographs of chronic OM.
Radiographs are also useful in follow-up of cases of OM, to rule out disease progression by comparing with the baseline radiographs.[10]
Ultrasound
Ultrasound may play a primary or supplementary role in the imaging evaluation of OM. It is a particularly useful modality in children, when subperiosteal collections may give a clue to the underlying intramedullary infection. Peri-osseous fluid collections or abscesses may also raise a possibility of infection in operated cases with hardware in situ. Ultrasound is also useful in providing guidance for aspiration or biopsy for microbiological analysis.
CT
CT scan has a limited role in the evaluation of OM. The main role of a CT scan is in the demonstration, localization, and sizing of the sequestrum in cases of chronic OM. Other findings in chronic OM include cortical thickening, involucrum, and sinus tracts. Nonspecific findings include effacement of the surrounding fat planes and cortical irregularity. A CT scan may also be utilized for guided aspiration, particularly in cases with deep-seated abscesses. The major drawback of CT scan is its inability to detect bone marrow edema, which makes it a much inferior modality as compared to MRI, especially in cases of acute OM.
MRI
MRI has a high sensitivity of 82–100% and specificity of 75– 99% in the diagnosis of acute OM, and is the investigation of choice, particularly in cases of acute OM.[11] Associated soft-tissue complications, including abscesses, phlegmon, tenosynovitis, or involvement of ligaments, are also well visualized on MRI.
Imaging in all cases of OM is performed in 3 planes, using various sequences. The sequences to be included are T1-weighted (T1W), T2-weighted (T2), and fat-suppressed T2 (T2FS) or short-tau inversion recovery (STIR) sequences.[12]
Diffusion-weighted imaging may prove to be very useful in diagnosing acute OM, particularly in the pediatric age group, as it may help in differentiating cellular marrow from OM. Infected marrow and intramedullary abscesses show significantly more diffusion restriction with lower apparent diffusion coefficient (ADC) values as compared to cellular marrow or reactive marrow changes [Figure 2].[8,13]

- Importance of diffusion-weighted imaging in hematogenous OM. (a) Axial STIR and (b) post-contrast T1FS images show subtle marrow abnormality in tibia with surrounding soft tissue edema. (c) DWI and (d) ADC images at the same level show definite diffusion restriction, confirming the presence of osteomyelitis. The marrow of the adjacent fibula can also be seen in the same section for comparison, showing normal marrow signal on the conventional as well as diffusion-weighted sequences. OM: Osteomyelitis, DWI:Diffusion-weighted imaging, ADC: Apparent diffusion coefficient.
Gadolinium-enhanced contrast MRI (CE-MRI) is not essential for all cases of OM. However, it may serve as a problem-solving tool in certain cases, especially in differentiating phlegmonous changes from abscess formation.[14] Phlegmonous tissue shows diffuse enhancement, whereas abscesses show peripheral contrast enhancement. CE-MRI may also aid in localizing sequestrum, with enhancement seen in the surrounding granulation tissue, and the sequestrum seen as a dark, non-enhancing structure within.[9]
Indirect/hematogenous OM
Acute OM
Marrow changes on MRI can be detected as early as 1–2 days after the onset of infection.[15] Typical marrow changes seen in acute OM are hypointensity on T1W images and hyperintensity on fluid-sensitive sequences [Figure 3]. These findings are, however, nonspecific and may also be seen in other pathologies such as reactive marrow edema or stress reaction. Periosteal reaction may also be identified on MRI, seen as elevation of the hypointense periosteum off the bony cortex [Figure 4]. MRI may, in fact prove to be more sensitive to radiographs in the identification of periosteal reaction in the early stages.

- Acute OM. A 11-year-old child presented with a painful, erythematous and swollen leg. (a) Radiograph of the knee shows peri-articular soft tissue swelling, without obvious any bony abnormality. (b, c) Typical marrow changes seen in acute OM are seen on MRI in the form of hypointensity and “fat globules” on T1-weighted images (green asterisks in b) and hyperintensity on the fluid-sensitive STIR sequence (green asterisks in c). A small subperiosteal collection can also be seen medially causing mild localized periosteal elevation (yellow arrow in c).

- Subperiosteal abscess. (a) Subperiosteal abscess seen superficial to the cortex of the medial femoral condyle (denoted by markers). (b) Illustration showing the formation of a subperiosteal abscess.
Subacute OM
Brodie’s abscess or an intramedullary abscess is a pathognomonic finding of the subacute stage of OM [Figure 5]. It shows a characteristic “penumbra sign” on T1W images. The penumbra sign is caused by central hypointensity of the pus-filled cavity, a thin peripheral hyperintense rim representing the inflammatory granulation tissue, which is further surrounded by a hypointense rim representing reactive sclerosis.

- Subacute OM. (a) Coronal T1W and (b) STIR images of a Brodie’s abscess or intramedullary abscess in a case of subacute OM (white arrows in a and b), showing focal involvement of the adjacent physeal plate and extension of the marrow abnormality into the distal tibial epiphysis. (b) A small subperiosteal collection is also seen medially (yellow arrows). OM: Osteomyelitis, STIR: Short tau inversion recovery
Chronic OM
Chronic OM shows changes secondary to new bone formation in the form of cortical thickening, solid or lamellated type of periosteal reaction, involucrum, and formation of sequestrum [Figure 6]. Sequestrum is a fragment of necrotic bone and acts as a nidus of active infection. It appears hypointense or dark on all sequences. Intervening hyperintense granulation tissue may help in differentiating the sequestrum from the normal marrow on fluid-sensitive fat-suppressed sequences. Involucrum is seen as a thick shell of hypointensity representing new bone surrounding the sequestrum. Cloaca is seen as a cortical defect, draining intramedullary pus into the surrounding soft tissues [Figures 6 and 7].

- Chronic OM. A 22-year-old male presented with a history of discharging wound in the thigh since a year. (a) Radiographs of the femur show new bone formation in the form of cortical thickening, periosteal reaction and involucrum (white arrows). (b) Cloaca is seen as a cortical defect, draining intramedullary pus into the surrounding soft tissues (yellow arrow). (c, d) Sequestrum appears hypointense or dark on all sequences (white arrow in c and d). OM: Osteomyelitis

- Computed tomography scan showing involucrum (arrowheads) and sequestrum with cloaca (*).
Direct OM
The above-described imaging findings of acute, subacute, and chronic stages of OM are typically found in cases of hematogenous OM and are not seen in cases of direct OM. A geographical pattern of intramedullary hypointensity on T1W images appearing darker than the adjacent muscles favors OM [Figure 8].[14] A hazy, reticular pattern of intramedullary T1W hypointensity, irrespective of its appearance on T2-weighted or contrast-enhanced images, indicated reactive marrow changes rather than OM [Figure 9].[14] In direct spread OM secondary to penetrating injury, imaging may also help to identify any retained foreign body, which acts as a nidus for infection [Figure 10].

- Direct contiguous spread OM with overlying skin ulcer in an adult diabetic male. (a) Clinical picture of foot showing the ulcer over medial aspect of forefoot. (b) Frontal and lateral radiographs of foot show a subtle ill-defined lytic area in the medial aspect of the head of 1st metacarpal (white arrow). (c, d) Axial STIR and T1W MR images show confluent marrow abnormality showing a geographic, hypointense pattern on T1W images, adjacent to the skin thickening, suggesting changes of osteomyelitis. OM: Osteomyelitis

- Reactive marrow changes seen adjacent to a non-healing ulcer at the plantar aspect of great toe. (a-c) STIR hyperintensity (yellow asterisk in a) without confluent T1W hypointensity adjacent to the soft tissue abnormality (white arrow in c) suggests reactive marrow edema.

- Direct inoculation. Sagittal short-tau inversion recovery (a) and T1-weighted (b) images show a retained foreign body (asterisks in a and b) in the metatarsal in a patient with a history of thorn prick, which is acting as a nidus for ongoing infection. An intramedullary abscess (white arrows in b) is also seen surrounding the foreign body.
Complications of OM
MRI also helps to detect any associated complications of OM, including tendon or tendon sheath involvement, soft-tissue abscess, joint involvement (septic arthritis), pathological fractures or deformities secondary to osseous destruction, and even squamous cell carcinoma.
Nuclear medicine
Bone scintigraphy has higher sensitivity but low specificity. The adjacent soft-tissue infection is also not well visualized.
Triple-phase bone scan with technetium-99m-labelled methylene diphosphonate (MDP) (Tc99m-MDP) has a high sensitivity for the detection of OM in non-violated bone. However, when the bone has been violated due to prior trauma or surgery, etc., the specificity is lower.[2] Specificity can be improved by combining it with a white blood cell (WBC) scan.[2] In WBC scans, the radionuclides used are Indium-111 or Tc99-HMPAO, whereas bone marrow scans use Tc99-labelled colloid.
Gallium scan with Gallium-67 has a higher specificity than triple-phase bone scan, and a combination of both tests may be useful in evaluating cases of suspected vertebral OM.[2] Gallium scans have a high negative predictive value in the diagnosis of OM.
Fluorodeoxyglucose (FDG) positron emission tomography (PET)-CT has been recognized as an invaluable modality in the assessment of complex OM, especially in cases of equivocal findings on MRI or conventional scintigraphy.[16] It exploits the glucose metabolism of activated inflammatory cells, which is reflected as high FDG uptake for the assessment of active infection.[17] PET scan offers superior resolution as compared to scintigraphy and leukocyte scans, aids in full-body assessment, and is more useful in scenarios dealing with disco-vertebral OM, multifocal disease, and peri-prosthetic joint infections.[18]
Role of imaging in follow-up of OM
The comprehensive evaluation of the cortex, periosteum, marrow, and soft tissue makes MRI the modality of choice for assessing treatment response.[19,20] Follow-up evaluation focuses on determining the adequacy of treatment, identification of recurrent or persistent infections, and further ruling out complications, including abscess formation, sinus tracts, and pathological fractures. The primary clue to successful therapy is the reappearance of normal fatty marrow signal on T1W images and resolution of the fluid-sensitive (STIR/T2 Fat-suppressed) hyperintense signals. Contrast-enhanced MRI further helps in distinguishing abscesses from postoperative collections and granulation tissue, as they typically exhibit rim enhancement. However, marrow changes may persist despite clinical remission, including persistent granulation tissue on imaging, so post-treatment MRI findings should always be interpreted with a thorough clinical and laboratory analysis.
Other than MRI, multiple other radiological modalities play a complementary role in assessing different aspects of follow-up imaging. Radiographs can be used to assess cortical thickening, periosteal reaction, sequestrum, and pathological fractures in post-treatment imaging. It is an inexpensive tool and is widely available. However, it is relatively insensitive to early findings, especially marrow abnormalities.[21]
Ultrasound is an excellent tool to visualise soft-tissue complications such as abscesses, sinus tracts, and subperiosteal collections, and aids in guided aspiration and drainage procedures. Still, the major constraint is suboptimal intraosseous evaluation.[22]
CT scan provides unmatched details about sequestrum, cloacae, and early cortical destruction, even when MRI is suboptimal due to extensive metallic susceptibility.
Nuclear medicine imaging is superlative in providing functional information. Tc99m MDP scintigraphy is sensitive but nonspecific, and uptake may persist despite resolution of infection. Labeled leukocyte scintigraphy (Tc-99m HMPAO or In-111 WBC) is more specific for active infection in postoperative patients; however, it comes at the cost of lower spatial resolution.[16]
Imaging in periprosthetic infections
Radiographs followed by cross-sectional or nuclear medicine imaging are required for imaging suspected infections of operated cases of deep-seated structures such as the spine [Figure 11]. Operated extremities with hardware in situ can be evaluated with radiographs and ultrasound.

- Vertebral implant osteomyelitis. (a) Lateral and AP radiographs show vertebral fixation implant in situ, with subtle sclerosis in D12 and L1 vertebral bodies with irregular end plates (yellow arrow). (b) Sagittal short-tau inversion recovery and (c) T1-weighted images show mild marrow abnormality in the same areas, (d) with enhancement on the post contrast images (yellow arrows in b-d).
Radiographs of infected prosthesis may demonstrate periosteal reaction/periostitis, peri-prosthetic lucency showing widening by >2 mm per year,[23] zones of osteolysis showing blurred margins around prosthetic margins, or increased bone density/sclerosis. Small sequestra may be difficult to visualize on radiographs and may be better demonstrated on a CT scan.
Sometimes, radiographs may appear normal, and abnormalities may be limited to the surrounding soft tissues. In such cases, ultrasound and MRI may prove to be of greater utility. Ultrasound may demonstrate collections or abscesses in the surrounding soft tissues, and may also show hardware loosening [Figure 12]. Ultrasound may also be used to guide aspiration of peri-prosthetic fluid collections.

- Appendicular implant OM. (a) Radiographs show non union of mid shaft femoral fracture with intramedullary fixation nail in situ. (a, b) Loosening of the cancellous fixating screws is appreciated on the radiographs as well as on ultrasound, with associated peri-osseous abscess also seen on the ultrasound image (white arrows in b).
With the advent of metallic artefact reduction sequences, MRI has proven to be one of the most useful modalities in demonstrating marrow abnormalities or soft-tissue collections in cases with peri-prosthetic implants. The presence of sinus tracts or abscesses has a high predictive value for peri-prosthetic infection, whereas the absence of effusion in a prosthetic joint almost rules out infection.[23]
Nuclear medicine studies, including bone scan, WBC scintigraphy, and 18F-FDG-PET, may also be utilized in the evaluation of peri-prosthetic infections. The scintigraphic techniques are not affected by the presence of metallic hardware; however, PET/CT may show susceptibility artefacts.[24] WBC scintigraphy shows a high sensitivity and specificity, and hence is considered to be a gold standard imaging technique in the detection of peri-prosthetic infections.[24] Bone marrow scintigraphy with 99mTc-colloids may be used after WBC scintigraphy in equivocal cases.[24] The role of FDG-PET/CT and PET/MRI in the evaluation of periprosthetic infection is still debatable. It has, however, been observed that a declining trend of FDG uptake correlates with response to ongoing therapy.
Differentiating OM from other causes of peri-prosthetic complications
Precise imaging differentiation between aseptic loosening, periprosthetic joint infection (PJI), and metallosis/aseptic lymphocyte-dominant vasculitis-associated lesion (ALVAL) is critical for surgical decision-making and management. Joint infections usually manifest as abnormal marrow edema, sinus tract infections, and periprosthetic collections. Post-contrast sequences using metal artefact reduction sequences can further help with demonstrating soft-tissue enhancement.[25] Metabolic imaging demonstrates intense peri-prosthetic uptake in infection on FDG-PET and labelled scintigraphy.[26] Aseptic loosening is characterized by well-defined smooth peri-prosthetic lucent areas easily demarcated on radiographs and CT, specifically at the bone-graft interface. Marrow edema and sinus formation are less commonly seen than PJI. ALVAL/Metallosis/Pseudotumor are most often associated with metal-on-metal prostheses, and present with exuberant synovial thickening, solid-cystic periarticular masses on MRI.[27] The assessment of the characteristic imaging patterns is essential for distinguishing infection from aseptic pathologies.
Differential diagnosis of OM
Several conditions can mimic OM both clinically and radiologically, and careful evaluation is essential for accurate diagnosis [Table 3].
| Mimic | Imaging features | Key differentiating points | Imaging clues |
|---|---|---|---|
| Ewing sarcoma | Permeative lytic lesion, layered (onion-skin) periosteal reaction, soft-tissue mass | Affects children/adolescents; systemic symptoms may mimic infection; less likely fever | MRI: Large soft-tissue component, skip lesions; PET/CT: Intense uptake, no abscess |
| Osteoid osteoma | Small, round, lucent nidus with central calcification and surrounding sclerosis | Night pain relieved by NSAIDs; young adults | CT: Clearly shows nidus; MRI: May mimic osteomyelitis with marrow edema, but nidus absent in infection |
| Stress fracture | Linear low T1 signal with surrounding marrow edema; periosteal reaction possible | Pain improves with rest; occurs with repetitive activity | MRI: Fracture line without cortical destruction; CT: Thin fracture line, no sequestrum or abscess |
| Charcot neuroarthropathy | Joint dislocation, subchondral bone fragmentation, periarticular sclerosis | Diabetic with peripheral neuropathy; often painless | MRI: Marrow edema without sinus tract or abscess; WBC scan: negative or diffuse uptake pattern |
| Reactive marrow edema | Mild patchy T2 hyperintensity in marrow; no cortical break or soft tissue changes | Trauma history; resolves with conservative treatment | MRI: Fat saturated images show edema, but no enhancement or abscess; no diffusion restriction |
| Chronic recurrent multifocal osteomyelitis | Multiple metaphyseal bone lesions with sclerosis and mild periostitis | Pediatric patients; systemic autoimmune disorders; sterile cultures | MRI: Multiple sites of bone edema, no abscess; WB-MRI helpful |
CT: Computed tomography, MRI: Magnetic resonance imaging, PET: Positron emission tomography, WBC: White blood cells, NSAIDs: Nonsteroidal anti-inflammatory drugs
Ewing sarcoma often presents in children and adolescents as a permeative lytic lesion with an onion-skin type periosteal reaction and a soft-tissue mass. Systemic symptoms may resemble infection, but fever is less common. MRI typically demonstrates a large soft-tissue component with possible skip lesions, while PET-CT shows intense uptake without abscess formation.
In contrast, osteoid osteoma is seen in young adults as a small round nidus with central calcification and prominent surrounding sclerosis. The hallmark is night pain that is dramatically relieved by nonsteroidal anti-inflammatory drugs. CT best demonstrates the nidus, whereas MRI may show marrow edema mimicking infection, but crucially lacks a nidus in true OM.
A stress fracture can also be confused with OM, appearing as a linear low-signal band on T1 images with surrounding marrow edema and sometimes periosteal reaction. Unlike infection, the pain improves with rest and follows a history of repetitive activity. MRI clearly shows the fracture line without cortical destruction, while CT demonstrates a thin fracture line without sequestrum or abscess.
Charcot arthropathy, most commonly encountered in diabetic patients, manifests with joint dislocation, subchondral bone fragmentation, and periarticular sclerosis. The degree of pain is strikingly disproportionate to the severity of the imaging findings. MRI shows marrow edema without sinus tracts or abscesses, and WBC scans typically show negative or diffuse uptake rather than focal infection.
A healing fracture can be mistaken for infection, as it demonstrates callus formation, periosteal reaction, and cortical thickening. However, there is typically a recent trauma history, improving pain, and no systemic symptoms. Radiographs and CT show smooth callus without aggressive features, and MRI lacks sinus tracts or abscess formation.
Chronic recurrent multifocal OM is an aseptic form of OM primarily affecting the pediatric age group. It presents as multiple metaphyseal bone lesions with sclerosis and mild periostitis. Strikingly, these cases yield negative blood cultures and bone biopsies, and thus obviously do not respond to antibiotics. An autoimmune etiology has been postulated. MRI often reveals multiple sites of bone edema without abscesses, and whole-body MRI can be particularly useful.
CONCLUSION
Early and accurate detection of OM with prompt treatment is essential to prevent complications. Using the optimal imaging modality and protocol is crucial in the imaging of OM. Pre-contrast T1W and STIR images are the most important sequences in the evaluation of direct spread OM, where confluent T1W hypointensity favors OM, whereas reticular or hazy appearance on T1W images favors reactive marrow changes. On the other hand, STIR sequences and diffusion-weighted imaging play an important role in the evaluation of hematogenous OM.
Ethical approval:
Institutional Review Board approval is not required.
Declaration of patient consent:
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Conflicts of interest:
Suvinay Saxena is on the Editorial Board of the Journal.
Use of artificial intelligence (AI)-assisted technology for manuscript preparation:
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
Financial support and sponsorship: Nil.
References
- Imaging of osteomyelitis of the extremities In: Seminars in Roentgenology. Vol 52. Philadelphia, PA: WB Saunders; 2017. p. :49-54.
- [CrossRef] [PubMed] [Google Scholar]
- The imaging of osteomyelitis. Quant Imaging Med Surg. 2016;6:184.
- [CrossRef] [PubMed] [Google Scholar]
- Clinical characteristics of neonatal and infant osteomyelitis and septic arthritis: A multicenter retrospective study. J Pediatr (Rio J). 2024;100:430-7.
- [CrossRef] [PubMed] [Google Scholar]
- Essentials of Skeletal Radiology United States: Lippincott Williams and Wilkins; 1996.
- [Google Scholar]
- Osteomyelitis In: Feigin RD, Cherry JD, Demmler-Harrison GD, Kaplan SL, eds. Textbook of Pediatric Infectious Diseases (6th ed). Philadelphia, PA, USA: Saunders Elsevier; 2009. p. :725-42.
- [CrossRef] [Google Scholar]
- Bone and joint infections in children. Pediatr Clin N Am. 2005;52:779-94.
- [CrossRef] [PubMed] [Google Scholar]
- Bones, joints and soft tissue tumors In: Kumar V, Abbas AK, Fausto N, eds. Robbins and Cotran Pathologic Basis of Disease (8th ed). Philadelphia, PA: Saunders Elsevier; 2010. p. :1221-2. Ch. 26.
- [CrossRef] [Google Scholar]
- The role of diffusion-weighted imaging and dynamic contrast-enhanced magnetic resonance imaging for the diagnosis of diabetic foot osteomyelitis: A preliminary report. Pol J Radiol. 2022;87:274-80.
- [CrossRef] [PubMed] [Google Scholar]
- Musculoskeletal Imaging: The Requisites (3rd ed). Philadelphia, PA: Mosby Elsevier; 2007. p. :545-6.
- [Google Scholar]
- Radiographic imaging in osteomyelitis: the role of plain radiography, computed tomography, ultrasonography, magnetic resonance imaging, and scintigraphy. Semin Plast Surg. 2009;23:80-9.
- [CrossRef] [PubMed] [Google Scholar]
- A case of acute osteomyelitis: An update on diagnosis and treatment. Int J Environ Res Public Health. 2016;13:539.
- [CrossRef] [PubMed] [Google Scholar]
- Diagnosis of osteomyelitis in children: Utility of fat-suppressed contrast-enhanced MRI. AJR Am J Roentgenol. 2009;192:1232-8.
- [CrossRef] [PubMed] [Google Scholar]
- Acute osteomyelitis, septic arthritis and discitis: Differences between neonates and older children. Eur J Radiol. 2006;60:221-32.
- [CrossRef] [PubMed] [Google Scholar]
- T1-weighted MRI characteristics of pedal osteomyelitis. AJR Am J Roentgenol. 2005;185:386-93.
- [CrossRef] [PubMed] [Google Scholar]
- Infection: Musculoskeletal. Pediatr Radiol. 2011;41(Suppl 1):S127-34.
- [CrossRef] [PubMed] [Google Scholar]
- Radionuclide imaging of spinal infections. Eur J Nucl Med Mol Imaging. 2012;39:892-909.
- [CrossRef] [PubMed] [Google Scholar]
- Fluorine-18-FDG PET and technetium-99m antigranulocyte antibody scintigraphy in chronic osteomyelitis. J Nucl Med. 1998;39:2145-52.
- [Google Scholar]
- FDG-PET for osteomyelitis and prosthetic infections. Semin Nucl Med. 2013;43:367-76.
- [CrossRef] [PubMed] [Google Scholar]
- The role of imaging techniques to define a peri-prosthetic hip and knee joint infection: Multidisciplinary consensus statements. J Clin Med. 2020;9:2548.
- [CrossRef] [PubMed] [Google Scholar]
- Imaging of musculoskeletal infections: The radiologist's perspective. Radiology. 2007;244:625-40.
- [Google Scholar]
- Chronic recurrent multifocal osteomyelitis: a great clinical and radiologic mimic in need of recognition by the pathologist. Apmis. 1999;107:369-79.
- [CrossRef] [PubMed] [Google Scholar]
- Imaging orthopedic implant infections. Diagn Interv Imaging. 2012;93:547-57.
- [CrossRef] [PubMed] [Google Scholar]
- The accuracy of diagnostic imaging for the assessment of chronic osteomyelitis: A systematic review and meta-analysis. J Bone Joint Surg Am. 2005;87:2464-71.
- [CrossRef] [PubMed] [Google Scholar]
- Role of nuclear medicine in diagnosis of the infected joint replacement. Radiographics. 2001;21:1229-38.
- [CrossRef] [PubMed] [Google Scholar]
- Nuclear medicine imaging of bone infections. Clin Radiol. 2016;71:632-46.
- [CrossRef] [PubMed] [Google Scholar]
- MRI after arthroplasty: Comparison of MAVRIC and conventional fast spin-echo techniques. AJR Am J Roentgenol. 2011;197:W405-11.
- [CrossRef] [PubMed] [Google Scholar]

