How Is Osteomyelitis Diagnosed?

Osteomyelitis is diagnosed through a combination of clinical evaluation, blood tests, imaging, and often a tissue or bone sample to identify the infecting organism. No single test confirms it on its own. In practice, doctors layer these tools: a physical exam and inflammatory blood markers raise suspicion, imaging (most often MRI) localizes the infection, and cultures or biopsy nail down the specific bug causing the problem. The exact diagnostic path depends heavily on where in the body the infection sits, whether it is acute or chronic, and whether hardware like a joint replacement is involved.

What Doctors Look for First

The clinical picture matters more than people realize. In children arriving at emergency departments with acute osteomyelitis, pain is the most common complaint, reported in about 94% of cases, followed by difficulty using the affected limb and fever. On examination, tenderness over the bone and visible swelling are frequent findings, though fever is present during the actual evaluation only about half the time.

In adults, the presentation is more variable. Vertebral osteomyelitis often shows up as persistent back pain with or without fever. Diabetic foot osteomyelitis can be surprisingly subtle, hiding beneath a foot ulcer with little systemic illness. Chronic osteomyelitis may present with a draining wound that has been there for weeks or months. The clinical picture alone cannot confirm the diagnosis, but it shapes every decision that follows.

Blood Tests That Raise or Lower Suspicion

Two inflammatory markers do most of the heavy lifting in initial lab work: erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). Both measure how much inflammation is happening in the body, and both tend to be elevated in osteomyelitis. In children with acute osteomyelitis, elevated ESR and CRP are the most sensitive lab markers, each flagging the infection in over 90% of confirmed cases.¹ That said, these markers are nonspecific. A broken bone, rheumatoid arthritis, or even a bad cold can push them up, so an elevated ESR or CRP does not prove osteomyelitis exists. It just means something inflammatory is going on.

A meta-analysis looking specifically at diabetic foot osteomyelitis found that ESR, CRP, and procalcitonin (PCT) all had pooled sensitivities and specificities in the low-to-mid 70s percent range. None qualified as outstanding standalone tests. ESR performed best overall, rated “excellent” as a biomarker, while CRP and PCT were rated “acceptable.”² In practice, these numbers mean blood tests are good at raising or lowering your suspicion but not good enough to rule the diagnosis in or out by themselves.

Where these markers get more useful is in combination with each other and with a clinical threshold. Research in diabetic patients found that when ESR is below 30 mm/h, osteomyelitis is unlikely. But when ESR climbs above 60 mm/h and CRP exceeds roughly 8 mg/dL, the probability is high enough that treatment should be strongly considered.³ So it is not about any single number but about patterns: how high the markers are, what the clinical picture looks like, and what the imaging shows.

Plain X-Rays and Why They Often Come First

An ordinary X-ray is usually the first imaging test ordered. It is cheap, fast, and widely available. The problem is that in early osteomyelitis, X-rays look completely normal. Bone changes take time to develop on plain film. Soft-tissue swelling may appear early, but actual bone demineralization typically does not show up for about a week. A periosteal reaction (new bone forming along the surface in response to infection) takes roughly two weeks, and clear-cut bone destruction can take three weeks or more to become visible.⁴

Despite this delay, X-rays still serve a purpose. They can rule out other causes of pain, such as a fracture or a tumor, and they establish a baseline for follow-up imaging. For vertebral osteomyelitis, subtle erosions of the vertebral endplates may lag two to eight weeks behind symptoms.⁵ So when an X-ray comes back “normal” but clinical suspicion remains, more advanced imaging is the next step.

MRI as the Preferred Imaging Tool

MRI is widely considered the gold standard for imaging suspected osteomyelitis. It detects changes in bone marrow within days of infection onset, long before X-rays show anything abnormal. For vertebral osteomyelitis specifically, the reported sensitivity, specificity, and accuracy of MRI hover around 97%, 92%, and 94%.⁵ In the lower limb, one study found MRI’s sensitivity and specificity at about 89% and 88%, with an overall diagnostic performance (measured by area under the curve) of 0.92.⁶

MRI excels because it visualizes the bone marrow directly and shows the extent of infection into surrounding soft tissues, abscesses, and sinus tracts. Adding intravenous gadolinium contrast improves the picture further, especially for identifying abscesses and the spread of infection into the spinal canal or nearby tissues.⁵

But MRI is not perfect. In patients with metal implants, the hardware creates artifacts that distort the image. Specialized sequences (sometimes called metal artifact reduction sequences) can reduce this distortion and improve what doctors can see around spinal hardware.⁷ MRI also cannot be used in people with certain pacemakers, ferromagnetic implants, or severe claustrophobia without sedation. And there are scenarios where an initial MRI simply misses the diagnosis.

When MRI Comes Back Normal but the Infection Is There

This happens more often than you might expect, and it can lead to dangerous delays. In one reported case, a child with confirmed bloodstream infection from bacteria had an initial CT and MRI of the spine that both failed to show vertebral osteomyelitis. A nuclear bone scan ultimately identified the infected vertebra, and a follow-up MRI focused on that spot confirmed the diagnosis.⁸ In another case, an adult with persistent Staphylococcus aureus in the blood and ongoing back pain had an initial whole-spine MRI that was negative. Only a repeat MRI, ordered because the clinical suspicion remained high, revealed infection of the lower spine with an abscess compressing the spinal nerves.⁹

The lesson from these cases is that a single negative MRI does not rule out osteomyelitis when clinical suspicion is strong. Repeat imaging, nuclear medicine scans, or a more focused MRI may be needed.

CT, Ultrasound, and Nuclear Medicine Scans

When MRI is unavailable or contraindicated, several alternatives exist, each with different strengths.

CT scanning is particularly useful in chronic osteomyelitis, where its main job is to locate and size pieces of dead bone (called a sequestrum), identify sinus tracts, and guide needle biopsies of deep abscesses.¹⁰ CT also helps distinguish osteomyelitis from degenerative spine changes when MRI findings are ambiguous: the absence of cortical erosion at the endplates on CT makes vertebral infection less likely.⁵

Ultrasound plays a more limited role but is valuable in children. It can confirm and precisely localize a subperiosteal abscess, which is a collection of pus beneath the outer layer of bone. One pilot study found that ultrasound had about 83% accuracy for detecting osteomyelitis and nearly 96% accuracy for identifying subperiosteal abscesses in children, compared with MRI as the reference.¹¹ When ultrasound confirms an abscess, doctors can plan surgical drainage. When it shows no abscess, surgery may be avoided entirely.¹² However, ultrasound can miss early infection or multifocal disease, so a bone scan or MRI is still needed when the clinical picture does not match.¹³

Nuclear medicine techniques, including three-phase bone scans, white blood cell (leukocyte) scans, and FDG-PET/CT, fill in where other imaging falls short. They are especially useful in complex scenarios like post-traumatic osteomyelitis, where the normal anatomy is already disrupted.¹⁴ For chronic osteomyelitis, FDG-PET/CT has a notable advantage: it maps the metabolically active infected bone, and when surgeons use that map to guide their debridement, recurrence rates drop substantially compared with conventional surgery.¹⁵

Identifying the Bug

Knowing that osteomyelitis is present is only half the battle. Identifying which organism is causing the infection determines which antibiotic will work and how long treatment needs to last. Doctors pursue this through blood cultures, bone biopsies, and increasingly through molecular testing.

Blood cultures are the least invasive option. In children with acute blood-borne osteomyelitis, a systematic review of 17 studies found a pooled positivity rate of about 31%, meaning roughly one in three blood cultures will grow the responsible organism.¹⁶ In adults with vertebral osteomyelitis, the yield is higher: one large case series found that about 55% of blood cultures grew an organism, with Staphylococcus aureus accounting for the majority.¹⁷ Because blood cultures are simple and risk-free, they are drawn in virtually every suspected case before antibiotics are started.

When blood cultures are negative or when doctors need tissue confirmation, a bone biopsy becomes necessary. CT-guided needle biopsies are commonly used. One study found that these biopsies yielded the causative organism in about 43% of confirmed osteomyelitis cases, with the sensitivity rising to roughly 54% when histology (tissue analysis under a microscope) was added to microbiology.¹⁸ That sensitivity might seem low, but biopsy results are more reliable than blood cultures at identifying the actual organism growing in the bone, especially when unusual or antibiotic-resistant bacteria are involved. Factors that increase the chance of a positive biopsy include fever, an elevated white blood cell count, and intravenous drug use.¹⁸

Prior antibiotic use is a notorious spoiler. In one imaging-guided biopsy series, 41% of osteomyelitis patients had already received antibiotics before the biopsy, and only 34% of all osteomyelitis cases ended up with a positive culture.¹⁹ This is why guidelines strongly recommend obtaining cultures before starting antibiotics whenever it is safe to delay treatment briefly.

Molecular Testing When Cultures Fail

Standard cultures do not always grow an organism, particularly after antibiotics or in infections caused by slow-growing or fastidious bacteria. This is where molecular diagnostics come in. Techniques like 16S rDNA PCR and metagenomic next-generation sequencing (mNGS) look for bacterial DNA in the tissue sample rather than waiting for live bacteria to grow in a dish.

A meta-analysis of these molecular approaches in joint infections found that 16S rDNA PCR had a pooled sensitivity and specificity of about 80% and 94%, while mNGS achieved about 89% sensitivity and 93% specificity. Among the molecular methods tested, mNGS had the strongest overall diagnostic performance.²⁰ Current expert recommendations advise considering 16S PCR when cultures stay negative but clinical suspicion remains high.²¹

These molecular tools are becoming more widely available, but they are not without limitations. They cannot tell you whether the organism they detect is alive or dead, they cannot provide antibiotic susceptibility results, and contamination can produce false positives. So they complement rather than replace traditional cultures.

The Probe-to-Bone Test for Diabetic Foot Infections

Diabetic foot osteomyelitis is a special diagnostic challenge because the infection often sits beneath a chronic skin ulcer, and patients may have diminished sensation that masks pain. One of the simplest and most practical bedside tests is the probe-to-bone test: a sterile metal probe is gently passed through the ulcer, and if it touches hard bone at the base, osteomyelitis becomes much more likely.

Research shows this test is highly sensitive (about 87%) and specific (about 91%). Its negative predictive value is especially impressive at 98%, meaning that if the probe does not reach bone, infection of the underlying bone is very unlikely.²² However, the positive predictive value is lower, around 57%, because exposed bone in a deep ulcer does not always mean infection. So a positive probe-to-bone test triggers further workup with imaging and labs rather than confirming the diagnosis outright.

Diagnosing Infection Around Metal Implants

When osteomyelitis develops around a prosthetic joint or fracture fixation hardware, the diagnostic rules change. Bacteria form biofilms on implant surfaces, and these biofilms are notoriously difficult to detect with standard cultures of surrounding tissue. Sonication, a technique that uses ultrasound waves to shake bacteria off the implant surface into fluid that is then cultured, significantly improves detection.

Compared with traditional tissue cultures (about 54% sensitivity), sonication fluid cultures achieve around 75% sensitivity for periprosthetic joint infection while maintaining high specificity.²³ Even better results have been reported when sonication fluid is cultured in blood culture bottles rather than on standard plates, with sensitivity reaching about 97% in one study without any loss in specificity.²⁴ The catch is that sonication requires the implant to be removed, so it applies mainly when revision surgery is already planned.

Imaging around metal also presents challenges. Standard MRI sequences produce heavy artifact that can obscure the bone and soft tissue in the exact area you need to evaluate. Metal artifact reduction sequences help, but image quality near hardware still falls short of what you get in native bone. CT handles cortical bone detail well but also suffers from streak artifacts near metal. For these reasons, nuclear medicine scans like labeled white blood cell scintigraphy or FDG-PET/CT sometimes become the preferred imaging approach when hardware is in place.

Telling Osteomyelitis Apart from Bone Tumors

On imaging, osteomyelitis and bone tumors can look alarmingly similar, and misdiagnosis goes in both directions. Both can cause bone destruction, soft-tissue masses, and periosteal reactions. One distinguishing feature on MRI is the “penumbra sign,” a rim of higher signal surrounding the main area of bone abnormality that suggests an abscess. In one study, this sign had a sensitivity of about 73% for osteomyelitis and a specificity of over 99%, meaning it almost never appeared in tumors.²⁵ When imaging alone cannot settle the question, biopsy with both histopathology and cultures becomes essential to tell infection from malignancy.

Emerging Diagnostic Approaches

Researchers are working on faster, more accessible ways to detect bone and joint infections. One line of development involves point-of-care biosensors that measure leukocyte esterase activity in joint fluid. The concept works similarly to a glucose meter: a disposable test strip gives a quantitative reading in minutes at the bedside, rather than requiring samples to be sent to a lab and waiting hours or days.²⁶ Meanwhile, advances in machine learning are being applied to predict whether blood cultures will grow an organism in patients with acute osteomyelitis, using patterns in routine lab data that humans might not spot.²⁷ These tools are still experimental, but they hint at a future where diagnosis involves less waiting and fewer invasive procedures.

For now, diagnosing osteomyelitis remains a layered process. The speed and accuracy of that process depend on the clinical context: a child with sudden limb pain and a high fever follows a different path from an elderly person with diabetes and a slowly worsening foot ulcer, or a patient with a persistently painful hip replacement. Understanding the available tools and their limitations helps explain why the diagnostic journey sometimes takes days or weeks and why repeat testing is not a sign that something went wrong, but that the team is being thorough.

1
PubMed. Clinical Presentation of Acute Osteomyelitis in the Pediatric Emergency Department
2
PubMed. Update of biomarkers to diagnose diabetic foot osteomyelitis: A meta-analysis and systematic review
3
PubMed Central. What are the Optimal Cutoff Values for ESR and CRP to Diagnose Osteomyelitis in Patients with Diabetes-related Foot Infections?
4
Diagnostic and Interventional Imaging. Pitfalls in osteoarticular imaging: How to distinguish bone infection from tumour?
5
PubMed Central. Diagnosis of vertebral osteomyelitis
6
PubMed Central. Osteomyelitis of the Lower Limb: Diagnostic Accuracy of Dual-Energy CT versus MRI
7
PubMed. MRI of spinal hardware: comparison of conventional T1-weighted sequence with a new metal artifact reduction sequence
8
PubMed Central. Difficulties in diagnosing vertebral osteomyelitis in a child
9
PubMed Central. Initial Negative MRI Spine and Persistent Staphylococcus aureus Bacteraemia Lead to Delayed Diagnosis of Lumbar Spondylodiscitis and Epidural Abscess
10
Indian Journal of Musculoskeletal Radiology. Imaging in osteomyelitis: Unveiling the secrets of bone infections
11
PubMed. Ultrasonography for the Detection of Osteomyelitis in Pediatric Patients With Soft Tissue Infection
12
PubMed. Ultrasound localization of subperiosteal abscesses in children with late-acute osteomyelitis
13
Clinical Radiology. Ultrasound in the management of children with acute haematogenous osteomyelitis
14
PubMed Central. Nuclear medicine imaging of posttraumatic osteomyelitis
15
SpringerLink. Is 18 F-FDG PET-CT-guided geographic debridement superior to conventional debridement in appendicular chronic osteomyelitis?
16
Clinical Infectious Diseases. PIDS/IDSA Guideline on Diagnosis and Management of Acute Hematogenous Osteomyelitis in Pediatrics
17
Open Forum Infectious Diseases. Culture Yield in the Diagnosis of Native Vertebral Osteomyelitis
18
PubMed Central. Diagnostic Performance of CT-guided Bone Biopsies in Patients with Suspected Osteomyelitis
19
AJR Am J Roentgenol. Imaging-guided bone biopsy for osteomyelitis: are there factors associated with positive or negative cultures?
20
Clinical Microbiology and Infection. Diagnostic accuracy of 16S rDNA PCR, multiplex PCR and metagenomic next-generation sequencing in periprosthetic joint infections
21
PubMed Central. Benefits of Polymerase Chain Reaction Combined With Culture for the Diagnosis of Bone and Joint Infections
22
PubMed. Probe-to-bone test for diagnosing diabetic foot osteomyelitis: reliable or relic?
23
PubMed Central. The Function of Sonication in the Diagnosis of Periprosthetic Joint Infection After Total Knee Arthroplasty
24
PubMed Central. Improving the microbiological diagnosis of fracture-related infection and prosthetic joint infection through culturing sonication fluid in Bactec blood culture bottles
25
PubMed. Differential diagnosis between osteomyelitis and bone tumors
26
Grantome. Rapid electrochemical biosensor for point-of-care diagnosis of joint infection
27
PubMed Central. A transformer-based deep learning model for identifying the occurrence of acute hematogenous osteomyelitis and predicting blood culture results