Osteomyelitis Workup

Updated: Mar 01, 2018
  • Author: Stephen Kishner, MD, MHA; Chief Editor: Murali Poduval, MBBS, MS, DNB  more...
  • Print

Laboratory Studies

A complete blood count (CBC) is useful for evaluating leukocytosis and anemia. Leukocytosis is common in acute osteomyelitis before therapy. The leukocyte count rarely exceeds 15,000/µL acutely and is usually normal in chronic osteomyelitis. Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels are usually increased. [19, 2, 20] In metastatic and some metabolic bone disease, alkaline phosphatase, calcium, and phosphate are elevated, but they are within normal limits in osteomyelitis. [12]

Blood cultures are positive in only 50% of cases of osteomyelitis. [5] They should be obtained before or at least 48 hours after antibiotic treatment. Although sinus tract cultures do not predict the presence of gram-negative organisms, they are helpful for confirming S aureus.


Imaging Studies

The American College of Radiology (ACR) has published imaging guidelines for the diagnosis of suspected osteomyelitis, septic arthritis, and soft-tissue infections in cases not involving the spine or the diabetic foot.  [21]  (See Guidelines.)


Conventional radiography is the initial imaging study at presentation of acute osteomyelitis. It is helpful to interpret current and old radiographs together. (See the image below.)

Osteomyelitis, chronic. Image in a 56-year-old man Osteomyelitis, chronic. Image in a 56-year-old man with diabetes shows chronic osteomyelitis of the calcaneum. Note air in the soft tissues.

Radiographic findings include periosteal thickening or elevation, as well as cortical thickening, sclerosis, and irregularity. Other changes include loss of trabecular architecture, osteolysis, and new bone formation. These changes may not be evident until 5-7 days in children and 10-14 days in adults. Plain films show lytic changes after at least 50%-75% of the bone matrix is destroyed. Therefore, negative radiographic studies do not exclude the diagnosis of acute osteomyelitis.

Healing fractures, cancers, and benign tumors may appear similarly on plain film. Subtle changes may indicate contiguous-focus or chronic osteomyelitis. [4, 2, 5, 22]

Computed tomography

Computed tomography (CT) is useful for guiding needle biopsies in closed infections and for preoperative planning to detect osseous abnormalities, foreign bodies, or necrotic bone and soft tissue. It may assist in the assessment of bony integrity, cortical disruption, and soft-tissue involvement. It may also reveal edema. Intraosseous fistula and cortical defects that lead to soft tissue sinus tracts are also demonstrated on CT.

Although CT may play a role in diagnosis of osteomyelitis, the scatter phenomenon may result in significant loss of image resolution when metal is near the area of inflammation. [4, 2, 5]

Magnetic resonance imaging

Magnetic resonance imaging (MRI) is a very useful modality in detecting osteomyelitis and gauging the success of therapy because of high sensitivity and excellent spatial resolution. The extent and location of osteomyelitis is demonstrated along with pathologic changes of bone marrow and soft tissue. [4] (See the image below.)

Osteomyelitis, chronic. T1- and T2-weighted sagitt Osteomyelitis, chronic. T1- and T2-weighted sagittal MRIs show bone marrow edema in L1 and obliteration of the disk space between L1 and L2.

MRI shows a localized marrow abnormality in osteomyelitis. T1-weighted images typically show decreased signal intensity, whereas T2-weighted images produce increased signal intensity. [4] Increased intensity on T2-weighted images may indicate sinus tracts, which extend from marrow and bone to skin through soft tissue. A decreased intensity on T1-weighted images with no change on T2-weighted images may indicate surgical or posttraumatic scarring of bone marrow.


The presence of fluid collection adjacent to the bone without intervening soft tissue usually suggests osteomyelitis. Other findings on ultrasonography include elevation and thickening of the periosteum. Ultrasonography may also be useful in cases with orthopedic hardware or in patients who are unable to undergo MRI. [23, 24, 25, 22]

Nuclear medicine imaging

Three-phase bone scanning is helpful in evaluating acute osteomyelitic and doubtful diskitis. However, the specificity of this procedure is decreased in secondary osteomyelitis. The bone scan may reveal increased metabolic activity in osteomyelitis, but this finding is indistinguishable from posttraumatic injury or following surgery or cancer. [4, 1] (See the image below.)

Osteomyelitis, chronic. Three-phase technetium-99m Osteomyelitis, chronic. Three-phase technetium-99m diphosphonate bone scans (static component) show increased activity in the heel and in the first and second toes and in the fifth tarsometatarsal joint.

One approach makes use of white blood cells (WBCs) labeled with technetium-99m (99mTc) hexamethylpropylene amine oxime (99mTc-HMPAO) or indium-111 (111In) oxime. This method, when used in the combination of 111In-oxime WBC scanning with 99mTc-sulfur colloid bone marrow scanning, is helpful for evaluating infections of hip prostheses. Isotope accumulates in areas of increased blood flow and new bone formation in the 99mTc polyphosphate scan.

When the imaging is present on the labeled WBC scan but not on the 99mTc bone marrow scan, the test is positive for osteomyelitis. A negative test result may indicate an impaired blood supply to the affected area. When red marrow is present (ie, axial skeleton and spine), WBC scanning is less sensitive for imaging. [4, 1, 22]

Gallium citrate attaches to transferrin, which then leaks into inflamed areas from the bloodstream. Increased uptake may occur in infection, cancer, and sterile inflammatory conditions. Performing a 99mTc scan along with the gallium-67 (67Ga) citrate scan may help distinguish bone and soft-tissue inflammation and show bone detail. [4, 1] However, the scan is most useful in cases of spondylodiskitis. [22]

In the assessment of inflammation of spinal lesions, 2-[18F]fluoro-2-deoxy-D-glucose (18F-FDG) positron emission tomography (PET) may provide high-resolution tomographic images and may represent an alternative to 67Ga citrate scanning. [1] 18F-FDG PET scans have high sensitivity and specificity (97.5% and 86.3%, respectively) in detecting musculoskeletal infections. However, the specificity drops in cases of suspected knee prosthesis infections. [26]



Bone biopsy leads to a definitive diagnosis by isolation of pathogens directly from the bone lesion. [5]  Bone biopsy should be performed through uninfected tissue and either before the initiation of antibiotics or more than 48 hours after discontinuance.

Open or percutaneous needle bone biopsy with histopathologic examination and culture is the routine for the diagnosis of osteomyelitis. This procedure may not be necessary if blood cultures are positive with consistent radiologic findings.

When clinical suspicion is high but blood cultures and needle biopsy have yielded negative results, a repeat needle biopsy or an open biopsy should be performed. A bone sample can be collected at the time of debridement for histopathologic diagnosis in patients with compromised vasculature. To obtain accurate cultures, bone biopsy must be performed through uninvolved tissue. Cultures of the sinus tract may be useful if S aureus and Salmonella species are isolated. [27, 28]


Histologic Findings

Acute osteomyelitis presents with acute inflammatory cells, edema, vascular congestion, and small-vessel thrombosis. In early disease, infection extends into the surrounding soft tissue, which compromises the vascular supply to the bone, as well as host response, surgery, and/or antibiotic therapy.

Large areas of dead bone may form if both medullary and periosteal blood supplies are compromised. Necrotic bone shows extensive resorption and inflammatory exudates on bone biopsy and appears whiter than living bone owing to the loss of blood supply. The development of granulation tissue occurs at the surface of dead bone, which is broken down by proteolytic enzymes, including polymorphonuclear leukocytes, macrophages, and osteoclasts. This occurs most rapidly at the junction of living and necrotic bone. A sequestrum is formed when dead cortical bone is gradually detached from living bone.

Chronic osteomyelitis presents with pathologic findings of necrotic bone, formation of new bone, and polymorphonuclear leukocyte exudation, which is joined by large numbers of lymphocytes, histiocytes, and occasional plasma cells.

The formation of new bone occurs over weeks or months as a vascular reaction to the infection. New bone arises from the surviving fragments of periosteum, endosteum, and cortex in the region of infection along the intact periosteal and endosteal surfaces. It may also occur when periosteum forms an involucrum, which is dead bone surrounded by a sheath of living bone. Involucrum may lead to sinus tracts due to perforations that allow pus to enter surrounding soft tissues and ultimately skin surface. A new shaft forms as the density and thickness of involucrum increases.

As a result of inflammatory reaction and atrophy disuse during the active period of osteomyelitis, surviving bone in the area of infection usually becomes osteoporotic. Bone density increases partially from reuse as the infection subsides and extensive transformation of bone may occur to conform to areas of new mechanical stresses. Over time, old living bone and newly formed bone may appear similar and might be indistinguishable, especially in children.



Two classification systems are commonly used for osteomyelitis.

In 1970, Waldvogel et al classified bone infections on the basis of pathogenesis and proposed the original osteomyelitis staging system. This system classifies bone infections as either hematogenous or osteomyelitis secondary to a contiguous focus of infection. Contiguous-focus osteomyelitis is further classified according to the presence or absence of vascular insufficiency. Both hematogenous and contiguous-focus osteomyelitis may then be classified as either acute or chronic. [29]

In 2003, Cierny-Mader et al developed their staging system, which at present is more commonly used. This system considers host immunocompetence in addition to anatomic osseous involvement and histologic features of osteomyelitis. [30, 1]  The first part of the system specifies four stages, as follows:

  • Stage 1 disease involves medullary bone and is usually caused by a single organism
  • Stage 2 disease involves the surfaces of bones and may occur with deep soft-tissue wounds or ulcers
  • Stage 3 disease is an advanced local infection of bone and soft tissue that often results from a polymicrobially infected intramedullary rod or open fracture; stage 3 osteomyelitis often responds well to limited surgical intervention that preserves bony stability
  • Stage 4 osteomyelitis represents extensive disease involving multiple bony and soft tissue layers; stage 4 disease is complex and requires a combination of medical and surgical therapies, and postoperative stabilization may be needed if the infected bone is an essential weightbearing bone

The second part of the Cierny-Mader classification system describes the physiologic status of the host, as follows:

  • Class A hosts have normal physiologic, metabolic, and immune functions
  • Class B hosts are systemically (Bs) or locally (Bl) immunocompromised; individuals with local and systemic immune deficiencies are labeled as ‘‘Bls’’
  • In Class C hosts, treatment poses a greater risk of harm than osteomyelitis itself; the state of the host is the strongest predictor of osteomyelitis treatment failure, and thus the physiologic class of the infected individual is often more important than the anatomic stage [5]

Other classification systems have been proposed for long-bone osteomyelitis. The Gordon classification classifies long-bone osteomyelitis on the basis of osseous defects, using infected tibial nonunions and segmental defects, as follows [31] :

  • Type A includes tibial defects and nonunions without significant segmental loss
  • Type B includes tibial defects greater than 3 cm with an intact fibula
  • Type C includes tibial defects of greater than 3 cm in patients without an intact fibula

The Ger classification is used to address the physiology of the wound in osteomyelitis, which is categorized as follows [32, 33] :

  • Simple sinus
  • Chronic superficial ulcer
  • Multiple sinuses
  • Multiple skin-lined sinuses

Bone infection persists if appropriate wound management is not undertaken. It is important to cover open tibial fractures with soft tissue early in the disease to prevent infection and ulceration.

The Weiland classification categorizes chronic osteomyelitis as a wound with exposed bone, positive bone culture results, and drainage for more than 6 months. [34] This system also considers soft tissue and location of affected bone. It does not recognize chronic infection if wound drainage lasts less than 6 months. Weiland et al specified the following three types:

  • Type I osteomyelitis was defined as open exposed bone without evidence of osseous infection but with evidence of soft-tissue infection
  • Type II osteomyelitis showed circumferential, cortical, and endosteal infection, demonstrated on radiographs as a diffuse inflammatory response, increased bone density, and spindle-shaped sclerotic thickening of the cortex; other radiographic findings included areas of bony resorption and often a sequestrum with a surrounding involucrum
  • Type III osteomyelitis revealed cortical and endosteal infection associated with a segmental bone defect

The Kelly classification considers the following types of osteomyelitis in adults:

  • Hematogenous osteomyelitis
  • Osteomyelitis in a fracture with union
  • Osteomyelitis in a fracture with nonunion
  • Postoperative osteomyelitis without fracture

This system emphasizes the etiology of the infection along with its relation to fracture healing. [35, 32]