Keywords
osteoid osteoma - microwave ablation - air insulation
Introduction
Osteoid osteomas are benign bone tumors that involve the long bones and spine most
commonly. The most common presentation in these cases is painful scoliosis (concave
on the side of the lesion) with or without soft tissue swelling. Pathologically, the
tumor contains three concentric regions: innermost nidus (neoplastic meshwork of osteoblasts,
dilated vessels, and osteoid), middle fibrovascular rim, and outermost reactive sclerosis.
The actual cause of pain is the release of prostaglandins by the nidus. The diagnosis
of osteoid osteoma is based on imaging. On computed tomography (CT), the nidus is
seen as a central hypodensity with or without central calcification and surrounding
sclerosis. On magnetic resonance imaging (MRI), the nidus shows low signal intensity
on T1-weighted images and variable signal intensity on T2-weighted images with enhancement
of the nidus on postcontrast images. Surgery and percutaneous ablation are considered
as treatment options for these lesions, with ablation proving to have a lower hospitalization
period and fewer complications as compared to surgery.
We report the case of a young female with an osteoid osteoma in the dorsal spine,
which was managed by microwave ablation (MWA) (with air as the insulating agents for
preventing thermal injury to the spinal cord) as surgical option was refused by the
patient. The patient was symptom-free after the procedure without any neurological
deficit.
Technique
A 25-year-old female presented to the Department of Interventional Radiology, with
painful scoliosis for 1 year. The visual analog pain score (VAS) of the patient was
10. CT scan showed a well-defined lesion (8 mm in size) with a central nidus and surrounding
sclerosis in the right lamina of the T11 vertebra, located 5 mm away from the spinal
cord ([Fig. 1]). MRI showed a central low signal-intensity nidus on T1-weighted images with mild
edema in the surrounding right lamina. Postcontrast images showed enhancement of the
nidus. A diagnosis of spinal osteoid osteoma was made. The patient was categorized
as class I according to the American Society of Anaesthesiologists' classification.
Her clinical examination, airway, spinal anatomy, and preoperative laboratory investigations
were within normal limits. She was planned for CT-guided MWA of the lesion under general
anesthesia.
Fig. 1 Preprocedural axial computed tomography section showing the nidus with surrounding
sclerosis in right lamina of T11 vertebral body—white arrow.
The lesion was localized under CT guidance and after the induction of general anesthesia,
a 11-gauge bone access cortical drill (Arrow On Control, Teleflex Inc., Morrisville,
North Carolina, United States) was used to traverse the bone cortex. Subsequently,
the coaxial advancement of a 13-gauge coring needle was performed into the nidus using
intermittent imaging guidance. Then, the coring needle was removed and a 16-gauge,
liquid-cooled microwave antenna with a 10cm shaft and 3.5mm exposed tip (ECO Medical
Saberwave Microwave tumor ablation system, Jiangsu, China) was placed in the lesion.
At the same level, a spinal needle (22 gauge) was inserted into the epidural-space
under CT guidance ([Fig. 2]), using an interlaminar approach (the spinal needle was placed through the interlaminar
space and ligamentum flavum into the epidural space on the same side of the lesion).
Subsequently, 10cc of air was infused in aliquots. A repeat CT scan was performed
to assess the completeness of the air-cuff around the spinal cord. An air-cuff of
5 mm thickness outlining the epidural space was seen with an increased distance between
the spinal cord and the lesion ([Fig. 3]). MWA was performed at 40 watts for 60 seconds (single cycle) with an ablation zone
of 18 mm (along the length of the antenna) x 10mm (width of the ablation zone).
Fig. 2 Intraprocedural axial computed tomography section showing the microwave antenna (#)
placed in the lesion along with the spinal needle (*) for injection of air.
The procedure was uneventful. Postprocedure CT showed the ablation tract as a linear
hypodensity in the location of the lesion with an intact air-cuff around it ([Fig. 4]). There was complete resolution of pain on postprocedure day 1(VAS 0/10), without
any sensory or motor deficit. The patient was followed up by a monthly clinical examination.
No recurrence of pain was reported by the patient.
Fig. 3 Intraprocedural axial computed tomography section showing air-cuff (white arrow)
around the spinal cord.
Fig. 4 Postprocedural axial computed tomography section showing the tract (white arrow)
of the microwave antenna used for ablation with air-cuff (*) around the cord.
Discussion
Thermal ablation is now an established first-line treatment modality for osteoid osteoma.
It includes a variety of modalities like radio-frequency ablation (RFA), MWA, and
laser ablation. Due to its good efficacy and safety profile, MWA is gradually being
established as a treatment option for osteoid osteoma. In MWA, percutaneous microwave
antennae are placed under CT guidance, which deliver high-energy alternating electromagnetic
waves leading to frictional heat generation and peak temperatures up to 150 °C. This
leads to coagulative necrosis and protein denaturation. The target temperature is
reached rapidly in MWA as compared to RFA, thus leading to a shorter ablation time.
In the pediatric population, a shorter ablation time is extremely important due to
concerns regarding cumulative anesthetic use.
An important consideration of any form of thermal ablation for spinal lesions is the
injury to the spinal cord and surrounding nerve roots. Many studies have tried to
analyze the major factors responsible for heat transmission to the spinal canal during
thermal ablation. These factors include distance of the neural elements to the heat
source, a possible insulating effect of the intervening cortical and medullary bone
and the length of the antenna/probe used. According to Dupuy et al, both the cortical
and cancellous bone have a thermal protective effect for the spinal cord.[1] They also postulated a protective effect of the cerebrospinal fluid (CSF) pulsations
and epidural blood flow. However, Nour et al conducted an experiment using RFA on
pigs, and found serious neurologic injuries despite an intact intervening cortex.[2] As the studies on the protection of the neural elements by natural barriers like
bone and CSF have conflicting results, additional neural-protection techniques must
be used during thermal ablation of spinal lesions, for an additional margin of safety.
These neural-protection techniques can be either active or passive. Passive techniques
include using temperature monitoring tools like thermocouples, intraoperative monitoring
of nerves by evoked potential electrodes or by electromyography. Active protection
techniques include insulation by air and hydrodissection. Apart from gas insulation
and hydrodissection, the availability of the rest of the devices is limited.
Both hydrodissection and gas dissection increase the distance between the probe and
the structure to be protected and hence help in producing a safe ablation zone. Whenever
hydrodissection is used with RFA, 5% dextrose in water is preferred over normal saline
as the latter has a high electrical conductivity.[3] However, for MWA, any fluid can be used. As far as gas dissection is concerned,
it can also help in the protection of the neural elements, due to its insulating effect
and increase in distance between the probe and the spinal cord. The use of air as
well as carbon dioxide insulation[3]
[4]
[5] has been successfully demonstrated, with RFA. However, the literature on the use
of air/carbon dioxide insulation with MWA for spinal osteoid osteoma is extremely
scarce.
This report suffers from various limitations. Carbon dioxide is considered a superior
protective agent than air for thermal ablation, as it is a better insulator than air
and is less soluble than air (reducing the chances of air embolism).[4]
[5] However, specialized prefilled syringes are required for using carbon dioxide, which
are not available at our institution. Nevertheless, there have been no reports of
air embolism with such minute quantities of air injected epidurally under imaging
guidance. Although MWA shows consistent surface temperatures with less variation,
in comparison to RFA, it is preferred to use a thermocouple in the epidural space
for temperature monitoring. However, it was not used in this study due to its nonavailability
at our institution.
Thermal protective techniques like epidural injection of air may provide adequate
insulation and neuroprotection during thermal ablation of spinal osteoid osteoma.
However, further studies with larger sample sizes and longer follow-up periods are
required to confirm the findings.
