Diabetic foot infections contribute to 75,000 hospitalizations in the US each year. There is a lifetime incidence of 15-25% of foot infections in diabetics. The investigators hypothesize that ferumoxytol (FDA approved for use in CKD patients, unlike gadolinium) will improve the accuracy of MR imaging in the diagnosis of osteomyelitis by allowing for the specific imaging of macrophages that have taken up the iron-based compound. In this study the investigators will image 12 patients with suspected diabetic osteomyelitis using ferumoxytol-contrasted MRI, and evaluate the feasibility of this approach.

Prior to enrollment, patients will get baseline labs to ensure they are candidates for the study. These will include a pregnancy test (if not previously acquired within 1 week of enrollment in study), basic metabolic panel (if not already obtained within 48 hours of enrollment), a complete blood count (if not already obtained within 48 hours of enrollment), and a hemoglobin A1C (If not obtained within 4 weeks prior to enrollment).

MRI will be performed, and standard precautionary measures for MRI imaging will be followed.

Patients with a GFR below 40 or AKI (based on clinical suspicion in accordance with standard UNM protocols for contraindications to gadolinium): Will obtain non-contrasted baseline MRI. Within 72 hours of baseline MRI, patients will obtain ferumoxytol contrasted MRI (6 patients).

Patients with a GFR above 40 and no AKI: Will get an MRI with and without gadolinium contrast on day 0 and ferumoxytol contrasted MRI within 72 hours (6 patients).

T1 with fat sat three planes, T2* and T2 with fat sat three planes will be obtained on a UNM MRI machine. FLASH sequences will also be obtained.

24 hours after administration of ferumoxytol,a ferumoxytol-contrasted MRI will be performed with the same acquisition parameters.

Ferumoxytol will be administered at a dose of 7 mg/kg up to 510 mg. The drug will be given as an infusion over at least 15 minutes.

MRI images will be read and interpreted by at least one musculoskeletal radiologist as diagnostic for osteomyelitis, diagnostic for not being osteomyelitis, or indeterminate based on marrow enhancement. The images and their interpretations will be downloaded to PACS, where researchers involved in the project will have access to the images. Signal intensity will be recorded in areas of interest as determined by signal abnormalities on the MRI sequences. Background noise will be measured in the area next to the area of interest. A signal to noise ratio (SNR) will be calculated by dividing signal intensity by the background noise standard deviation. The relative enhancement (RE) will be calculated by the equation: (SIpost - SIpre)/SIpre x 100, where SIpost is the signal intensity 24 hours after administration of contrast, and SIpre is signal intensity prior to administration of contrast.

Patients will be monitored in the hospital for at least 24 hours after administration of ferumoxytol, obtaining vital signs every 4-8 hours.

The patient's CBC (complete blood count) and basic metabolic panel will be obtained on days 0 (the day of ferumoxytol administration), and 1 (the day of ferumoxytol-contrasted MRI acquisition), monitoring specifically white blood count and creatinine.

Patients will be offered a follow-up appointment at the CTSC as indicated above. Follow up imaging will be obtained during this visit as is clinically indicated. As described above, at this time Dr. Unruh, Langsjoen and/or Dr. Neuwelt will decide if patient definitely had osteomyelitis (which requires a tissue diagnosis), may have had osteomyelitis-won't be used for accuracy calculations, or did not have osteomyelitis (based on tissue diagnosis or clinical progression of the wound). The results of this decision will be used as a reference standard. Bone biopsy and/or amputation will only be obtained as clinically indicated by the primary medical team. Per standard protocols, it is likely that bone biopsy and/or amputations will be obtained during the patient's inpatient hospital stay.

At this follow up visit at the CTSC, patient's charts will be reviewed to determine duration of antibiotics administered and whether patient had bone biopsy and/or amputation (and results).

Patients with a medical indication for bone biopsy will have this procedure within 72 hours of ferumoxytol administration if this is medically indicated and feasible in order to optimize scientific yield of the study.

Patients undergoing bone biopsy or amputation will have their tissues examined by pathology faculty. Pathologists will analyze the specimen for evidence of osteomyelitis as well as ferumoxytol content within the analyzed tissue (specifically noting presence within bone or marrow), as determined by an anti-dextran monoclonal antibody (specific for the dextran coating on ferumoxytol). Prussian blue stains will be obtained to measure iron content, and specific stains for macrophages will also be performed. Tissues will be analyzed under the direction of Dr. Bocklage at the UNM human tissue repository.

We also plan on performing quantitative studies to measure iron content and dextran content within the tissues.

Ferumoxytol will be given to patients as an off-label drug as a contrast agent for MRI. Ferumoxytol is FDA-approved for iron replacement in CKD patients.

Inclusion criteria:

• patients in which a diagnosis of pedal osteomyelitis is being entertained,
• 18 years or older,
• a medical indication for an MRI and have diabetes.

Exclusion criteria:

• patients that are septic and clinically unstable as determined by the primary medical teams.
• Patients with a contra-indication to MRI (metal in body, etc) or hemochromatosis will be excluded.
• Pregnant patients will be excluded.
• Patients unable to understand the consent form and prisoners will be excluded.
• Female patients with hct >48% and males with hct>52% will be excluded.