Outcomes in diabetic foot ulcer patients with isolated T2 marrow
signal abnormality in the underlying bone: should the diagnosis
of Bosteitis^ be changed to Bearly osteomyelitis^?
Received: 31 January 2017 /Revised: 28 March 2017 /Accepted: 18 April 2017 / Published online: 11 May 2017
Objective To evaluate the variability of clinical treatment and
outcomes based on reporting of diabetic foot ulcer MRI findings
of adjacent marrow T2 hyperintensity with normal T1 signal.
Materials and methods A retrospective review was conducted
of 46 MRI examinations evaluating diabetic foot ulcers that
demonstrated normal T1 marrow signal, but T2 marrow
hyperintensity deep to the ulcer. The cohort was divided based
on MRI report impressions into three groups; Bosteitis without
osteomyelitis^ (OW), Bosteitis but cannot exclude early
osteomyelitis^ (OCEO) and Bearly osteomyelitis^ (EO).
Patient demographics (age, gender) and accessory MRI find-
ings of ulcer and sinus tract depth were recorded. Initial clin-
ical assessment and medical treatment (route and duration of
antibiotics), healing versus disease progression and histology
or microbiology results were recorded.
Results The isolated marrow T2 signal hyperintensity was re-
ported as OW in 12 patients, OCEO in 18, and EO in 16. No
statistical difference in clinical assessment was demonstrated
between the OW, OCEO, and EO groups. Pathological condi-
tion was available in 15 patients within 0–7 days (mean
2.4 days) of the MRI examination, with 14 (93%) of these
positive for osteomyelitis by histopathology or positive cul-
tures. Initial diagnosis of or progression to osteomyelitis was
Conclusion Treatment of suspected osteomyelitis is heavily
determined by clinical factors. Patients who initially demon-
strate only T2 marrow signal abnormality under a diabetic
ulcer are eventually diagnosed as osteomyelitis in 61% of
cases and deserve aggressive treatment as early osteomyelitis
when meeting clinical parameters.
Isolated T2 marrow signal
Diabetes-related foot complications account for up to 20% of
all diabetic related North American hospital admissions, with
up to $1.5 billion spent annually in the USA on diabetic foot
ulcer care [1, 2]. Foot ulcers are by far the most common
etiopathogenesis for the development of osteomyelitis in the
diabetic foot . Diabetic foot ulcers pose treatment challenges
to clinicians, as the concomitant neuropathy and vasculopathy
contribute to ulceration, poor soft-tissue healing, and the risk
for progression to osteomyelitis in the underlying bone.
Therefore, early clinical recognition and treatment is needed
to attempt to prevent development of osteomyelitis [4–6].
Magnetic resonance imaging is widely considered the im-
aging modality of choice for investigating possible osteomy-
elitis following plain radiography [4, 6, 7]. MRI is also useful
in establishing the pre-surgical extent of disease when the
diagnosis of osteomyelitis is already clinically established
. MRI is 90% sensitive and 79% specific in pooled analysis
of its efficacy in the diagnosis of diabetic pedal osteomyelitis
. The positive predictive value of MRI is very high when
high T2 and confluent intermediate T1 signal is appreciated in
bone marrow subjacent to a soft-tissue ulcer . The NPV
approaches 100% when the bone marrow signal is normal on
all sequences [10, 11].
The specificity of MRI is decreased when there is increased
T2 signal or marrow edema without associated confluent
* Dennis Duryea
Department of Radiology, H066, Milton S. Hershey Medical Center,
500 University Drive, PO Box 850, Hershey, PA 17033, USA
Skeletal Radiol (2017) 46:1327–1333