| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Brigham and Women's Hospital, Harvard Medical School,
75 Francis St., Boston, MA 02115.
2 Department of Radiology, University of Massachusetts Medical Center,
Worcester, MA.
3 Department of Orthopedic Surgery, Brigham and Women's Hospital, Harvard
Medical School, Boston, MA.
4 Department of Radiology, St. Joseph's Hospital and Medical Center, Phoenix,
AZ.
Received April 30, 2006;
accepted after revision February 5, 2007.
This publication was made possible by grant 41RR019703 from the National
Institutes of Health (NIH). Its contents are solely the responsibility of the
authors and do not necessarily reflect the views of the NIH.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twenty-seven biopsy-proven metastatic lesions of soft tissue (n = 17) and bone (n = 10) in 22 patients (15 men, seven women; age range, 24–85 years) were managed with MRI-guided percutaneous cryotherapy. The mean lesion diameter was 5.2 cm. Each lesion was adjacent to or encasing one or more critical structures, including bowel, bladder, and major blood vessels. A 0.5-T open interventional MRI system was used for cryoprobe placement and ice-ball monitoring. Complications were assessed for all treatments. CT or MRI was used to determine local control of 21 tumors. Pain palliation was assessed clinically in 19 cases. The mean follow-up period was 19.5 weeks.
RESULTS. Twenty-two (81%) of 27 tumors were managed without injury to adjacent critical structures. Two patients had transient lower extremity numbness, and two had both urinary retention and transient lower extremity paresthesia. One patient had chronic serous vaginal discharge, and one sustained a femoral neck fracture at the ablation site 6 weeks after treatment. Thirteen (62%) of the 21 tumors for which follow-up information was available either remained the same size as before treatment or regressed. Eight tumors progressed (mean local progression-free interval, 5.6 months; range, 3–18 months). Pain was palliated in 17 of 19 patients; six of the 17 experienced complete relief, and 11 had partial relief.
CONCLUSION. MRI-guided percutaneous cryotherapy for metastatic lesions of soft tissue and bone adjacent to critical structures is safe and can provide local tumor control and pain relief in most patients.
Keywords: interventional radiology • MRI • musculoskeletal system • oncology • percutaneous ablation
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MRI is temperature sensitive and can be used to monitor thermal ablation [9–12]. MRI has been used successfully to monitor cryotherapy for tumors of the liver [9] and kidney [10]. MRI also can be used to visualize radiofrequency ablation changes as an area of decreased signal intensity surrounded by a rim of hyperintensity on T2-weighted and STIR images [11]. In radiofrequency ablation, however, unlike in cryotherapy, MRI can be used for monitoring only when the radiofrequency generator is inactivated or when the radiofrequency energy is filtered. MR images are distorted when the generator is activated [12]. Furthermore, the corresponding margin of necrosis falls somewhere within the rim of hyperintensity and is not visualized as a sharply demarcated edge [11].
During cryotherapy, frozen tissue, referred to as an ice ball, is readily visible as a signal void on MR images obtained with routine pulse sequences. Cryotherapy has been used successfully for decades in the surgical management of malignant lesions of the liver, prostate, and elsewhere [13]. Cryotherapy also has been shown effective in the management of both benign and malignant bone tumors [13–15]. Cryotherapy is curative in the management of benign aggressive bone lesions, and it is effective for local control of metastatic disease [14]. Cryotherapy can be performed percutaneously under imaging guidance [9, 10]. Sonography can be used to guide cryoprobe placement and show the effects of freezing. However, because of shadowing behind the proximal edge of the ice ball, sonography does not depict the circumference of the ice ball in a single view. With MRI, cryotherapy can be monitored in such a way that the entire circumference of the ice ball can be visualized [9, 10].
MRI-guided percutaneous cryotherapy combines the capability of ablation of tumors with the low morbidity of percutaneous approaches and the advantages of MRI in depicting the volume of ablation during the procedure. In this investigation, we sought to determine the safety and feasibility of percutaneous MRI-guided cryotherapy for refractory and painful metastatic lesions of soft tissue and bone adjacent to critical structures.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although we also perform CT-guided radiofrequency ablation of metastatic lesions of bone and soft tissue, we specifically selected for MRI-guided cryotherapy patients who had tumors adjacent to critically important structures. Additional criteria were that the patient be finished with radiation therapy or chemotherapy, be in hemodynamically stable condition, have normal or correctable hemostatic measurements, have no contraindications to MRI, and have no history of active ischemic heart disease, which was defined as recent myocardial infarction, recent symptoms of angina, or ischemic changes on ECG. The last criterion was necessary because ST and T-wave segments cannot be monitored during MRI, and silent ischemia can go undetected [16, 17]. All patients were seen by medical and surgical oncologists, and the decision to treat the patient with MRI-guided percutaneous cryotherapy was made by consensus.
Evaluation before the procedure included an interventional radiology consultative visit and an MRI examination with a 1.5-T system (Signa, GE Healthcare). The MRI protocol was as follows: T2-weighted fast spin-echo images (TR/TE, 3,000–6,850/77–107; echo train length, 8–16; slice thickness, 3–6 mm; interslice gap, 1 mm; field of view, 14–38 cm), breath-hold fast-recovery fast spin-echo images (2,000–5,117/91–114; echo train length, 12–22; slice thickness, 3–5 mm; interslice gap, 0.5–1 mm; field of view, 18–40 cm), or single shot fast spin-echo images (1,234–6,502/32–367; slice thickness, 4–6 mm; gap, 1 mm; field of view, 24–40 cm); T1-weighted spin-echo images (150–800/4.2–14; slice thickness, 3–6 mm; interslice gap, 0.5–1 mm; field of view, 18–40 cm), fast spin-echo images (500–667/10–14; echo train length, 0–4; slice thickness 3–4 mm; interslice gap, 4–4.5 mm; field of view, 20–34 cm), or spoiled gradient-recalled acquisition in the steady state (SPGR) images (285–450/2.1 and 4.7; dual echo; flip angle, 90°; slice thickness, 5 mm; interslice gap, 1 mm; field of view, 38–40 cm); and T1-weighted dynamic imaging before and after IV administration of 20 mL of gadopentetate dimeglumine 469.01 mg/mL (Magnevist, Berlex Laboratories) with SPGR images (210–500/4.2; flip angle, 75°; slice thickness, 5–6 mm; interslice gap, 0–1 mm; field of view, 20–44 cm), fast acquisition with multiphase enhanced SPGR images (5.1–7.5/1.4–2.1; flip angle, 10°; effective slice thickness, 2.5–3.0 mm; interslice gap, 0 mm; field of view, 38–40 cm), or spin-echo images (500–800/9–16; slice thickness, 3.0–7.0 mm; interslice gap, 0.5–2.0 mm; field of view, 18–40 cm).
MRI with an open-configuration 0.5-T system (Signa SP, GE Healthcare) [18–20] was performed before the procedure to plan patient positioning, to select a surface coil and imaging sequences, and to plan the interventional approach. A preprocedural evaluation by an anesthesiologist also was performed. Preprocedural laboratory tests included prothrombin time, partial thromboplastin time, hematocrit, WBC count, platelet count, and serum creatinine and serum myoglobin concentrations.
Twenty-seven metastatic lesions of soft tissue and bone (mean diameter, 5.2 cm; range, 3.0–10.0 cm) were managed in 22 patients (15 men, seven women; age range, 24–85 years; mean age, 56 years) in 24 treatment sessions. One patient had three tumors, and three patients had two tumors each. The other 18 patients had one tumor each. In four patients, two tumors were managed during the same session. The third tumor in the patient with three tumors was managed in a separate session. One patient was treated for the same tumor after it recurred. All patients were treated for local control of tumors refractory to other treatments. In addition to local tumor control, 24 tumors in 19 patients were managed for palliation of pain. Previous treatments included surgical resection in two patients, surgical resection and chemotherapy in one patient, surgical resection and radiation therapy in one patient, chemotherapy and radiation therapy in three patients, and surgical resection, chemotherapy, and radiation therapy in 15 patients.
Seventeen of the 27 tumors were located in soft tissue, either adjacent to bone or involving muscle and soft tissue not arising from an organ. Ten of the 27 tumors were located in bone (four in the appendicular skeleton and three in the axial skeleton). Four tumors were located predominantly in soft tissue, but the tumor exhibited invasion into adjacent bone. Ten of 27 tumors were located in the presacral soft tissue, three in the rectus abdominus muscle, three in vertebrae (T9, L4, and L5), two in the small-bowel mesentery, two in iliac bone, two in the proximal part of the humerus, two in the proximal part of the femur, and one each in the external obturator muscle, the subcutaneous tissues of the back, and the sacrum. All patients with presacral or sacral tumors had undergone abdominoperineal resection.
The primary malignant tumors were rectal carcinoma (n = 10), colon carcinoma (n = 4), lung carcinoma (n = 3), thoracic mesothelioma (n = 3), ocular melanoma (n = 2), renal cell carcinoma (n = 1), testicular teratoma (n = 1), and pelvic leiomyosarcoma (n = 1). Two tumors were adenocarcinomas of unknown origin. All tumors were adjacent to one or more critical structures, including bowel (n = 14), peripheral nerve (n = 13), bladder (n = 8), major blood vessel (n = 6), ureter (n = 5), tendon (n = 5), spinal nerve (n = 4), skin (n = 4), spinal cord (n =1), urethra (n = 1), prostate (n = 1), vaginal cuff (n = 1), and spermatic cord (n = 1). Five patients had symptoms caused by the tumor abutting the adjacent structures, resulting in L5 sensory deficit (n = 1), difficulty urinating (n = 1), pain radiating into the lower extremity (n = 1), and lower extremity motor weakness (n =1). One patient had hematuria. In 10 patients, tumors completely encased the sacral plexus (n = 9), ureters (n = 2), S1 nerve (n = 1), and the periprostatic neurovascular bundle (n = 1), causing urinary incontinence (n = 1), S1 sensory defect (n = 1), erectile dysfunction (n = 1), and hydronephrosis (n = 2). During the treatments, MRI monitoring was used to avoid injury to adjacent structures and not structures encased by the tumor. It was accepted that encased structures would be ablated along with the tumor to achieve a reasonable degree of local tumor control and pain palliation.
|
|
Under MRI guidance, MRI-compatible biopsy needles (18–22 gauge) (E-Z-EM) were placed into the tumor. MRI was used to confirm the position of the needle. All 27 tumors were diagnosed with percutaneous biopsy, either with MRI-guided biopsy immediately before cryoablation (n = 18) or with CT-guided biopsy as a separate procedure (n =9). One to five cryoprobes (mean number, 1.6) were placed in each tumor, depending on its size. During procedures on eight tumors, 15 cryoprobes were placed with a coaxial technique. In these tumors, MRI-compatible 3- and 4-mm-diameter bone biopsy needles (MRI Devices) were needed to breach cortical bone. After the biopsy specimen was obtained, cryoprobes were inserted in a coaxial configuration through the outer cannula of the bone biopsy needles (Figs. 1A and 1B). For all other tumors, a trocar technique was used to insert cryoprobes alongside the guiding biopsy needle.
Two 15-minute freezes separated by a 10-minute thaw were used at each cryoprobe position in all but three tumors. In one patient, two tumors involving different extremities were managed in the same session. One freeze-and-thaw cycle was used at each cryoprobe position to minimize overall anesthesia time. Because of proximity of the ice ball to the skin in another patient, only one freeze-and-thaw cycle was used on a tumor. The goal of each procedure was to ablate as much tumor as possible while avoiding damage to critical structures. To meet this goal for two tumors, the cryoprobes were repositioned into areas of tumor not encompassed by the initial ice ball. Similarly, during procedures on six tumors, cryoprobes were withdrawn 1.5–4 cm to new positions to produce overlapping ice balls and increase tumor coverage. Probe temperatures reached an average of –130°C during treatment.
MR images were acquired in at least two planes to monitor freezing and thus control the extent of ablation to both achieve tumor coverage and avoid critical structures in the field. The MRI protocol included repeated (every 1–3 minutes) acquisitions as follows: T1-weighted fast spin-echo images (500–800/22–26; echo train length, 8, slice thickness, 7–8 mm; field of view, 18–28 cm), T2-weighted fast spin-echo images (4,000–6,000/92–110; echo train length, 8–16; slice thickness, 5–10 mm; field of view, 16–30 cm), or fast multiplanar gradient-recalled images (51/9.8–9.9; flip angle, 60°; slice thickness, 8–10 mm; field of view, 30–32 cm).
Supplemental Prophylactic Measures
A warming catheter (Endo-Care) was inserted into the urethra before
ablation of a metastatic lesion of rectal cancer that involved the prostate
and was adjacent to the urethra. Warm water at 38°C was circulated to
prevent urethral injury (Figs.
2A and
2B). In five patients with
tumor adjacent to one or both ureters, ureteric stents were placed
cystoscopically before treatment and removed 2–4 weeks after treatment.
In one patient with a tumor adjacent to the prostate, a transurethral bladder
catheter routinely placed for less than 1 day in all procedures was kept in
place for 1 week to prevent prostatic edema from causing bladder outlet
obstruction. One week after cryoablation, one patient who underwent treatment
for masses in both the femur and humerus underwent surgical placement of
intramedullary rods to stabilize the bone in case of pathologic fracture.
During all procedures, warm saline soaks were applied to the cryoprobe skin
entry site to prevent frost along the cryoprobes from freezing the skin.
During procedures on four subcutaneous tumors, this measure was taken
specifically to prevent the nearby ice ball from injuring the skin.
|
|
Postprocedural Care and Clinical, Imaging, and Laboratory Evaluation
At the end of each treatment session, axial T2-weighted fast spin-echo
images (4,000–6,000/92–110; echo train length, 8–16; slice
thickness, 5–10 mm; field of view, 16–30 cm) of the treated region
were obtained. Patients were observed for 1–2 hours in the recovery room
and then were admitted to the hospital for a minimum of 24 hours. Serum
hematocrit, WBC count, prothrombin time, partial thromboplastin time, platelet
count, and creatinine and myoglobin concentrations were measured 6, 12, and 24
hours and 1 week after the procedure. The first five patients treated
underwent routine prophylaxis of myoglobinemia-induced renal insufficiency.
The prophylaxis consisted of mannitol (25 g IV at the time of the procedure)
followed by IV hydration and alkalinization of urine (normal saline solution
with 150 mEq/L sodium bicarbonate at 150 mL/h for 24 hours)
[21]. The other patients were
treated with this regimen only if serum myoglobin concentration increased to
more than 1,000 ng/mL (normal value, 0–100 ng/mL) during the recovery
period.
As part of the assessment of complications and local tumor control, 1.5-T MRI was performed 24–48 hours after each procedure with the pulse sequences and parameters used in preprocedural MRI. A musculoskeletal faculty radiologist with 10 years of experience interpreting MR images, who did not perform the treatments, measured the tumor on the preprocedural MR images. This radiologist qualitatively assessed the percentage of tumor cryonecrosis on 24- to 48-hour postprocedural MR images using the following categories: less than 25%, 26–50%, 51–75%, and 76–100%. Tumor cryonecrosis was defined as previously enhancing tumor that did not become enhanced after treatment.
For 21 of the 27 treated tumors, local tumor control was assessed on MR images or CT scans obtained a mean of 5 months (range, 1.5–18 months) after treatment. Follow-up images were not available for six tumors in five patients. Two of these patients were lost to follow-up after the clinic visit 1 week after the procedure; two patients had immediate rapid clinical progression of systemic malignant disease; and one patient underwent surgical resection of the tumor 6 weeks after cryoablation. Imaging follow-up data were collected until local progression of the treated tumor, systemic progression of disease necessitating chemotherapy, or death occurred. Imaging follow-up results were categorized as regression (tumor smaller than original size), stability, or local progression (tumor larger than original size). Clinical outcome, including death and systemic progression of disease necessitating chemotherapy or other treatment, was tabulated by chart review (mean follow-up period, 19.5 weeks; range, 1–80 weeks).
For the 24 tumors in 19 patients treated for pain, baseline and follow-up pain ratings, pain medication doses, and effect of pain on daily activities were recorded when available. Pain was considered successfully palliated when there was a reduction in the pain rating or a decrease in the dose of pain medication during the follow-up period. The levels of pretreatment pain referable to the 24 tumors were mild for four, moderate for 11, and severe for nine tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
In the cases of five (19%) of the 27 tumors, cryoablation of spinal and paraspinal tumors in two patients caused transient lower extremity tingling, numbness, and weakness that resolved in 1 week. In one patient, numbness around the knee persisted. Two other patients, who underwent cryoablation of presacral tumors, experienced gluteal, perineal, scrotal, and thigh numbness that resolved after a period of months. Both patients also had urinary retention and needed to perform self-catheterization. Despite follow-up imaging evidence of regression of a large presacral metastatic lesion of colon cancer abutting the vaginal cuff, the fifth patient had chronic serous vaginal discharge approximately 2 weeks after cryoablation. Two years later, the patient had extensive cellulitis of the buttocks and an abscess at the treatment site that necessitated percutaneous drainage. Percutaneous biopsy of the treated site showed no evidence of tumor.
Femoral neck fracture necessitated surgical resection and placement of a prosthesis in one patient 6 weeks after ablation of a metastatic lesion of renal cell carcinoma. The surgical specimen revealed no tumor. Another patient experienced transient, mild hypotension immediately after cryoablation of mesenteric and subcutaneous metastatic lesions of ocular melanoma. Myocardial infarction was ruled out, and CT of the abdomen and pelvis revealed no evidence of hemorrhage.
Local Tumor Control and Clinical Outcomes
Twenty-four (89%) of 27 treated tumors exhibited 76% or greater tumor
necrosis. The other three tumors had 51–75% tumor necrosis. Cryoablation
was not repeated, however, because of the presence of residual tumor involving
the epidural space in one patient, the bladder wall in another, and the L5
nerve in the third patient. Four (19%) of 21 tumors for which clinical
follow-up (mean, 9 months; range, 4–18 months) findings were available
regressed (Figs. 5A,
5B, and
5C). One of these tumors
necessitated two treatment sessions because recurrence was detected after the
initial session. In another case, a patient who had locally metastatic rectal
cancer invading the prostate was treated for urinary retention that improved
for 4 months after treatment.
|
|
|
Nine (43%) of the 21 tumors were stable at follow-up evaluation (mean, 2.6 months; range, 1.5–7 months). Four patients with six of these tumors had systemic progression of malignant disease and needed chemotherapy (mean time to systemic progression, 5 months; range 2–11 months). Systemic progression led to the deaths of three of these patients (mean time to death, 4 months). Eight (38%) of the 21 tumors progressed locally (mean local progression-free interval, 5.6 months; range, 3–18 months). Seven of the patients were treated with chemotherapy for systemic progression of disease (mean time to systemic progression, 3.9 months; range, 3–7 months). In one patient, the tumor did not progress until 18 months after treatment and was surgically resected. During laparotomy, malignancy was found elsewhere in the abdomen, and chemotherapy was instituted.
Pain Control
Palliation was achieved in 17 (89%) of 19 patients (24 tumors) treated for
pain. Complete pain relief was achieved by six patients (mean follow-up
period, 8.9 weeks; range, 1–44 weeks). Three of these patients had
severe, two moderate, and one mild pain before treatment. Ten patients
experienced a reduction in pain. Five of these patients needed less medication
(mean follow-up period, 15.5 weeks; range, 1–80 weeks). One patient had
an initial decrease in pain but later needed an epidural pump. Two patients
had no pain reduction.
Laboratory Results
In the first 24 hours after all procedures, there was an immediate but
variable increase in myoglobin concentration (maximum observed concentration,
2,688 ng/mL; mean, 693 ng/mL; range, 84–2,688 ng/mL), but the level
returned to normal in 1 week. Serum myoglobin concentration increased to more
than 1,000 ng/mL (normal value, 0–100 ng/mL) after five procedures and
prompted treatment with urine alkalinization, mannitol, and hydration as
described earlier. Serum creatinine values remained normal in all patients.
WBC count increased to slightly greater than normal (normal range,
4–10.8 x 109/L) after four procedures and returned to
baseline within 1 week after the procedure. All patients had normal
prothrombin and partial thromboplastin times before cryoablation, and the
values did not change after the procedure. Platelet count had a trend toward a
decrease in the first 24 hours after the procedure in nine patients but was
less than 150 x 109/L in only two patients. In all patients,
platelet count normalized within 1 week of cryoablation. Mild decreases in
hematocrit occurred in the first 24 hours after the procedure in all patients
but were unaccompanied by signs or symptoms of hemorrhage.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Approximately one half of patients with metastatic disease have poorly controlled pain [23, 24]. These patients often have exhausted conventional therapies, including radiation therapy, chemotherapy, and surgery, and rely on the use of oral and IV medication for pain management. These treatments often are limited. Patients either respond poorly or have serious side effects. Improved means of reducing pain would have considerable influence on the quality of life of these patients. For patients with symptomatic lesions refractory to conventional therapies, percutaneous ablation is a direct response to patients and referring physicians who have turned to new techniques of tumor control. The situation is analogous to the use of other therapies in the control of incurable cancer.
We were successful in the use of cryotherapy to palliate pain in 17 of 19 patients. This success rate is consistent with the results of radiofrequency ablation [6, 7] and supports the notion that percutaneous tumor ablation can be used for pain relief in patients with unresectable tumors. Furthermore, our main objective was to show the safety of ablating tumors adjacent to critical structures, and all but five tumors were managed without injuring these structures. In the management of two presacral tumors, the sacral plexus was encased by tumor, and postprocedural sensory deficits probably were related to ablation of the sacral plexus branches. Similar side effects in two other patients were transient and resolved in months.
Fracture of the femoral neck 6 weeks after cryoablation of painful metastatic renal cell carcinoma might have been prevented by prophylactic surgical placement of an intramedullary rod or by reinforcement of the treated area with cement injection akin to vertebroplasty or acetabuloplasty. The patient who sustained this fracture was the first of seven in this series to undergo cryoablation of metastatic lesions of bone. Prophylactic rod placement was undertaken after two subsequent cryoablation procedures. Because ablation can further weaken bone structure, prophylactic intramedullary rod placement should be considered, particularly in the management of tumors involving weight-bearing bones.
In all cases, MRI was used successfully to monitor and control ice-ball
deposition so that the margins of the ice ball were kept separate from the
critical structure at risk. There were no injuries to normal structures
outside the treatment area delineated by the MRI-depicted ice ball.
Furthermore, because ice-ball margins represent a nonlethal temperature (
0°C), ice balls can abut these structures and not harm them.
We used MRI for all aspects of cryotherapy: planning, targeting, monitoring, controlling the treatment, and assessing treatment response. The interventional MRI system helped with targeting through clear delineation of the tumor and surrounding anatomic structures, real-time imaging, and multiplanar and interactive capabilities. The MRI system was used during treatment to monitor coverage of the tumor by the ice ball and whether adjacent normal structures were being affected. Conventional 0.5-T MRI sequences were well-suited for depicting the ice ball as a well-marginated signal void. This feature allowed us to control the cryoablation zone by repositioning the cryoprobes or stopping the freeze in one or more cryoprobes before the ice balls extended into adjacent critical structures. Thus we were able to strike a balance between ablating a maximum volume of tumor and avoiding injury to adjacent structures. Eradicating a tumor requires extension of the ablation zone beyond the margins of the tumor into surrounding tissues. Because this maneuver might have significantly increased the risk of injury to adjacent critical structures in our patients, the goal of treatment was tumor control rather than eradication of tumor.
Although MRI was used to avoid freezing adjacent critical structures at risk, we used additional safety measures to further reduce the risk of injury. Ureteral stents, Foley catheters, and a urethral warming catheter provided additional protection during this initial trial. Our experience with monitoring made us confident that ice balls can be controlled well enough to avoid most structures. However, because nondilated ureters can be difficult to visualize on 0.5-T MRI, ureteral stents may still be warranted. Warming of the skin also is needed to manage tumors involving subcutaneous tissues.
Complications due to sonography- and CT-guided percutaneous tumor ablation are uncommon. In an initial trial [6] of radiofrequency ablation of bone metastasis, there were no complications, but patients with tumors abutting critical structures may have been deliberately excluded from the trial. Complications are rare after CT-guided radiofrequency ablation of osteoid osteoma [25]. These tumors, however, are much smaller than the lesions managed in our series and therefore require a much smaller treatment volume. In addition, osteoid osteoma is rarely situated adjacent to structures other than bone and soft tissue. Complications of sonography- and CT-guided radiofrequency ablation of liver tumors are uncommon, but major ones, including death, have been reported as a result of perforation of adjacent colon, stomach, and small bowel [26]. In the past these complications may have been caused at least in part by the inability to monitor the margins of the ablation effect with sonography and CT.
Not surprisingly, some proponents [27] of radiofrequency ablation have recommended not managing lesions with CT- or sonography-guided radiofrequency ablation when the lesion abuts critical structures. Choi et al. [28] reported effective and safe radiofrequency ablation of hepatic tumors abutting the gastrointestinal tract. In addition, critical structures can be displaced during ablation by percutaneous instillation of water in radiofrequency ablation [29] and saline solution in cryotherapy [10]. Critical structures encased by tumor rather than adjacent to it are difficult to avoid. Encased critical structures either may not be visible or may have to be sacrificed for management of the entire tumor or for palliation of pain. Transient nerve deficits and urinary retention in two of our patients were attributable to effects of cryoablation on encased sacral plexus nerves.
The use of MRI to target, monitor, and control ablation of metastatic lesions of bone and soft tissue is supported by the advantages of MRI in depicting tumors of the musculoskeletal system. MRI is ideal for imaging bone and soft-tissue tumors and has been used to guide biopsy of musculoskeletal lesions [30]. An MRI-compatible coaxial drill system and other MRI-compatible devices with needlelike probes for gaining access to subcortical bone are means for biopsy and management of bone tumors [31].
We believe that MRI-guided cryotherapy is advantageous in relation to CT-guided cryotherapy. Although CT attenuation of frozen tissue is lower than that of surrounding unfrozen soft tissue, the difference in attenuation is small compared with the difference in signal intensity of frozen and unfrozen tissue [32]. MRI, unlike CT, can be used to monitor cryoablation in close to real time in multiple planes, does not involve the use of ionizing radiation, and distinctly depicts both ice ball and tumor. Tumors typically have increased signal intensity on T2-weighted images, whereas ice balls cause a signal void. On CT, both tumor and ice ball are hypodense and often cannot be differentiated.
Our study was limited because it was a feasibility study with a relatively small number of patients. We used a retrospective study design and therefore did not gather detailed information regarding treatment response. In three cases, assessment of pain palliation was complicated by the presence of pain in other regions of the body. We were unable to assess local tumor control in patients with disease that progressed systemically and was managed with chemotherapy. Had we continued to follow our patients after other therapies were instituted, we would not have known whether subsequent tumor control was due to our treatment or to other treatments.
In summary, MRI-guided percutaneous cryotherapy for metastatic lesions of soft tissue and bone is safe and feasible in anatomic locations adjacent to critical structures. Pain palliation and tumor control can be achieved safely with the use of MRI to monitor treatment during the procedure. Although CT- and sonography-guided radiofrequency ablation can be used to manage many lesions, percutaneous MRI-guided cryotherapy should be considered in the management of tumors adjacent to critical structures.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
