Keywords
bone tumor - osteoblastic tumor - osteoid osteoma - radiofrequency ablation
Introduction
Osteoid osteoma (OO) is a benign bone forming tumor that was first introduced to the
medial literature as a distinct clinical and pathological entity by Jaffe in 1935.[1] He described five cases of benign osteoblastic tumors composed of atypical bone
and osteoid who underwent surgical excision on the belief that they were an “inflammatory
osteomyelitis” but no pus was identified at the time of surgery. OO is now recognized
to be a relatively common benign skeletal tumor. It is the third commonest primary
benign bone tumor accounting for ∼12.9% of all benign bone tumors.[2]
OO can occur in almost any bone, but there is a predilection for the lower limb with
50% of lesions occurring in the femur or tibia.[2] They usually occur in younger patients with approximately half presenting between
10 and 20 years. OO is uncommon in patients younger than 5 years or older than 40
years[3] and therefore may not be considered in the differential diagnosis of a bone lesion
of a patient after middle age. We describe a case of a histologically proven OO in
patient aged 77 years who was treated with computed tomography (CT)-guided percutaneous
radiofrequency ablation (RFA). We believe this is the oldest recorded case of a patient
with OO in the medical literature to date.
Case Report
A 77-year-old man was referred to our institution with a 6-month history of increasing
pain in swelling in the metacarpal region of his right/left hand. There was no preceding
history of trauma. His medical history was unremarkable. Clinical examination demonstrated
soft tissue swelling on the dorsum of the left hand with no overlying erythema or
warmth. Full blood count was within normal limits, and the serum C-reactive protein
and erythrocyte sedimentation rate were mildly elevated.
X-ray of the hand demonstrated cortical thickening and sclerosis of radial border
of the middle finger metacarpal. Within this area, there was a central focal lucent
area ([Fig. 1]).
Fig. 1 Radiograph of the hand of a 77-year-old man shows cortical thickening of the radial
border of the middle finger metacarpal with a central cortical lucency (white arrow).
Magnetic resonance imaging (MRI) demonstrated a thickened cortex with intense intramedullary
and periosteal edemas. A small central cortical lucency was demonstrated measuring
0.5 × 0.5 cm. Given the patient's age, it was felt that this was most likely to be
osteomyelitis with an intracortical abscess ([Fig. 2]). OO was also considered at this time but on balance was felt less likely due to
patient's age. Other differential diagnoses considered included stress fracture, florid
reactive periostitis, and sclerotic metastasis. As the symptoms were not settling
and blood cultures were negative, the patient underwent a CT-guided biopsy of the
lesion to get a definitive diagnosis.
Fig. 2 (A) Coronal short tau inversion recovery (STIR) sequence of the hand shows cortical
thickening of the metacarpal with intramedullary marrow edema. (B) Axial STIR sequence of the hand showing cortical sclerosis with periosteal and intramedullary
edemas. There is a 5-mm cortical lucency at the radial border of the metacarpal.
The CT scan immediately prior to the biopsy demonstrated cortical sclerosis of the
metacarpal and thickening with a well-defined focal central lucency ([Fig. 3]). Given the CT findings, this was felt to be classical of an OO, and therefore,
the decision was taken to undertake a CT-guided biopsy and percutaneous RFA in the
same sitting. The biopsy was performed with a coaxial 15G Bonopty biopsy system (14G
penetration set) (AprioMed, Sweden), and the sample was sent for histology and microbiology.
CT-guided percutaneous RFA was then performed using the Neurotherm NT1100 RF Generator
(Abbott, United States). A 16G 15-cm RFA probe with a 5-mm exposed tip was placed
into the center of the lesion. Ablation was performed by heating the tip of the probe
to 90 degrees for 6 minutes ([Fig. 4]). This created an ablation zone of ∼5 mm which was felt sufficient to cover the
lesion. Given the risk of a skin burn, the skin and subcutaneous tissues were protected
with generous instillation of local anesthetic which increased the distance between
the active part of the electrode and the skin surface. The skin was monitored with
visual inspection and touched throughout the procedure. No periprocedural complications
were observed and the patient was discharged on the same day. Microbiological cultures
were negative. Histology confirmed fragments of a nidus of an OO with woven bone lined
by osteoblasts ([Fig. 5]). The patient has complete resolution of his symptoms at his 3 months follow-up
and he remains symptom free at 6 months.
Fig. 3 Reformatted coronal computed tomography image showing cortical sclerosis and a cortical
lucency typical of an osteoid osteoma.
Fig. 4 Intraprocedural axial computed tomography image with radiofrequency probe in the
lesion.
Fig. 5 Photomicrograph of the tumor (hematoxylin and eosin staining) showing the typical
features of an osteoid osteoma with thin trabeculae of woven bone surrounded by osteoclasts.
Discussion
There are numerous reports in the medical literature of OO in older individuals. A
large series from the Armed Forces Institute of Pathology which included 225 patients
showed the average age of presentation to be 19 years with an age range from 19 months
to 56 years.[4] Furthermore, there are several case series which have included patients who were
older than 60 years.[2]
[4] The oldest recorded patient with an OO in the medical literature was from a series
from the Mayo Clinic who was 72 years old at the time of diagnosis.[5] The differential diagnosis of OO includes Brodie's abscess, sclerosing osteomyelitis,
stress fracture, osteoma, osteosarcoma, and Ewing's sarcoma. Given our patients demographic,
the most likely diagnosis on the initial imaging was between an osteomyelitis with
a Brodie's abscess and an OO.
The nidus of a classical cortical OO and a Brodie's abscess can have very similar
imaging findings and, in many cases, can be indistinguishable from each other. On
conventional nonenhanced MRI, the nidus of a cortical OO has an intermediate signal
on T1-weighted sequences and is hyperintense signal on fluid sensitive sequences.
If the nidus is sufficiently mineralized, it may be low signal on all sequences. There
is also associated periosteal and intramedullary edemas. On MRI, a Brodie's abscess
can demonstrate the penumbra sign which is a thin rim of higher intensity signal at
the periphery of the abscess on T1-weighted images which represents surrounding granulation
tissue around the central abscess.[6] If the abscess is small, the penumbra sign can be difficult to visualize, however.
CT can be used to help differentiate between the two conditions. In a cortical OO,
the nidus of the lesion is usually surrounded by reactive sclerosis. The nidus is
usually rounded and well defined with a smooth periphery measuring less than 1.5 cm.
The nidus may or may not contain various amount of mineralization. The mineralization
can have different appearances and can be amorphous, ring like, or dense[4] and is usually central in location.[7] Conversely, a Brodie's abscess usually has an irregular outer surface. A sequestrum,
which may or may not be present, is usually located in an eccentric location[7] rather than centrally. The abscess cavity can measure anywhere from a few millimeters
to several centimeters. Other CT findings associated with an OO are the presence of
feeding vessels to the nidus[8]
[9]; however, these could be confused with a thin sinus tract or a cloaca.
High-resolution dynamic contrast MRI may be useful in differentiating an OO from a
Brodie's abscess. Liu et al in 2003 showed the nidus demonstrated a peak enhancement
in the arterial phase with early washout[10] in 82% of cases. A more recent study by Pottecher et al[11] demonstrated early arterial enhancement of the nidus in 93% of cases. Interestingly,
they also showed that the nidus in small and flat bones demonstrated faster contrast
uptake than in long bones as medullary and endosteal lesions did when compared with
intracortical and subperiosteal lesions. Conversely, the enhancement pattern in a
Brodie's abscess is different with slower enhancement with no arterial peak.[12]
Bone scintigraphy can be helpful in the diagnosis of OO. Due to the vascular nature
of the nidus, the early phase of the bone scan will show intense tracer uptake. On
delayed phase imaging, a “double density” sign has been described which represents
the intense tracer uptake of the nidus and the less intense tracer uptake of the surrounding
reactive tissues.[13]
[14] Unfortunately, the pattern of tracer uptake in infection can be similar to that
of an OO, and while it is very sensitive, it has low specificity[15] and cannot reliably differentiate between the two disease processes.
A Brodie's abscess and an OO are treated differently. The mainstay of treatment of
a Brodie's abscess is often surgical drainage often in conjunction with antimicrobial
therapy.[16] The treatment of choice for OO is excision or destruction of the nidus. The preferred
method of destroying the nidus nowadays is percutaneous RFA, which is a well-established
minimally invasive technique with high success rate and minimal complications.[17] This treatment has largely replaced open surgical excision or curettage of the nidus.
In our case where the CT findings were classical for an OO, it was felt reasonable
to proceed to percutaneous biopsy and RFA in the same sitting. When the diagnosis
is in doubt, a biopsy is advisable first and following definitive histology, the appropriate
treatment can be commenced. If a cortical abscess is misdiagnosed and an inadvertent
RFA is performed, it should not burn any bridges for definitive treatment further
down the line, although there is a single case report in the medical literature in
which the inadvertent ablation of a Brodie's abscess led to worsening infection and
abscess formation.[18] Nonsteroidal anti-inflammatory can be useful in the control of symptoms of an OO,
but there are issues with the long-term use of such medications which are exacerbated
in the elderly patient, particularly in those with preexisting renal, pulmonary, and
gastrointestinal-related issues. This, therefore, makes their use as a definitive
treatment strategy for OO less favorable than RFA.[19]
RFA in the hand bones can be performed but should be practiced with caution due to
the close proximity to the skin and adjacent nerves and blood vessels. Preprocedural
ultrasound immediately prior to the placement of the penetration needle will identify
any tendons and vessels in the needle path and therefore can be avoided. In addition,
infiltration of subcutaneous local anesthetic can increase the distance between the
OO and the skin surface thereby reducing the risk of a skin burn. Retraction of the
outer cannula, so that it does not contact the exposed tip of the radiofrequency electrode,
is mandatory and should be checked throughout the procedure, and a CT image should
be saved to confirm this maneuver has been performed. As, with any RFA procedure,
the grounding pad should be carefully placed on the skin, ensuring full skin contract
using the largest available grounding pad on a well-vascularized site to reduce the
risk of grounding pad site thermal injury.[20]
Conclusion
In summary, OO is rare benign bone forming tumor much more frequently encountered
in children and young adults, and hence may be overlooked in the elderly patient which
may lead to a delay in the correct diagnosis and the appropriate treatment. It should
be considered in the differential diagnosis in a painful sclerotic cortically based
lesion in the patient of any age and not just in children and young adults. CT-guided
percutaneous RFA is the treatment of choice with a high success rate and fewer complications
than open surgical excision.
