In 1996, the European Institute of Oncology, Milan, developed a new method for localizing occult lesions, called radioguided occult lesion localization (ROLL), the idea for which came from experience with radiocolloid injected close to the breast lesion (typically intradermally) to identify the sentinel node. The radioactive particles travel within the lymphatic ducts to be taken up by sentinel lymph nodes, usually in the axilla. It was noticed that some radioactivity always remained at the injection site, and the idea developed that, if the particles were too large to pass into the lymphatic system, all would remain at the injection site. Further, if this material were injected directly into a nonpalpable lesion, the radioactivity would remain there and serve as a beacon during surgical removal.
Studies showed that macroaggregate human albumin colloid, the particles of which have a larger average diameter than the microaggregate used in sentinel node biopsy, were practically immobile. After technetium 99m (99mTc) has been chemically bound to the particles, they could be injected directly into the occult lesion under stereotactic mammographic or ultrasonic control. In the operating room, a gamma ray–detecting probe locates the lesion and has proven invaluable in guiding its complete removal.16,17 After a pilot phase during which the quantity of radioactivity to be injected was optimized, ROLL became the method of choice for locating nonpalpable lesions at the European Institute of Oncology. It is effective, easily reproducible, and has a short learning curve.18
Twenty-four hours before the operation, human albumin macroaggregate (particle size 10-150 nm in diameter; Macrotec, Sorin Biomedica, Saluggia, Italy) labeled with 3.7 MBq of 99mTc in 0.2 mL of saline is injected directly into the lesion under ultrasonic or mammographic control (Fig. 55-1).
Radiotracer injection under stereotactic control (left) and ultrasonic control (right).
For lesions detected ultrasonically, the radiotracer is injected under the guidance of a linear probe attached to a needle biopsy device, which is inserted into the breast manually. The needle tip is positioned at the center of the lesion, as shown by a change of echogenicity at the lesion site. Radiotracer is then injected, followed by an additional minimal quantity of saline to flush the needle and help avoid dispersing the radioactivity. For lesions visible by ultrasound and mammography, injection under ultrasonic control is preferred.
For microcalcifications or other anomalies revealed only mammographically, mammographic equipment attached to a computerized stereotactic system is used to guide injection. Again a small quantity of pure saline (0.2 mL) is injected, and this is followed by a minimum quantity of radiopaque contrast medium via the same needle. The needle is then removed, and a standard orthogonal mammogram is taken a few minutes later to verify exact correspondence between the lesion and injected radiotracer/radiopaque material (Fig. 55-2). If a lesion has been completely removed by preoperative mammotome biopsy, the clip is located by mammography and the radiotracer injected at the clip site under stereotactic guidance.
Injection of radiopaque contrast medium into the cluster of microcalcifications after injection of radiotracer.
Lateral and anterior scintigraphic images are taken after a few minutes and 5 hours later. The lateral image is obtained with the patient prone using a polystyrene block to hold the breast in position and a flexible wire cobalt source to outline the breast contour (Fig. 55-3). The anterior image is obtained with the patient standing, after placing a cobalt 57 point source on the nipple as landmark. The scintigraphic images are assessed for the presence of radioactive contamination. When the hotspot appears as a small, well-delimitated area, the patient is referred for surgery. If the skin is contaminated with radioactivity, it is carefully cleaned and the scan repeated. The current success rate at our institute is 98.8%. Reasons for unsuccessful tracer localization include contamination of the needle track, skin contamination, and intraductal dissemination of the tracer. As with any new technology, the success rate has improved since the initial evaluation published 1998.16
Scintigraphic lateral view after injection of radiotracer.
The hand held gamma ray–detecting probe used for radioguided surgery consists of a metal cylinder approximately 15 to 20 cm long and approximately 1.5 cm in diameter gloved in a transparent sterile sheath. It is connected to a device that transduces the signal into an audio signal and digital display. The intensity and frequency of the audio signal is directly proportional to number of counts detected. Newer probes are wireless (Bluetooth) and may be more convenient to use.
Use of the probe allows the surgeon to immediately locate the skin projection of the radioactive hotspot and hence decide the most appropriate incision (often based on aesthetic considerations), irrespective of the injection site. In our experience, this is a major advantage over determining the location of the lesion mammographically with the needle localization technique. During surgery, the probe is used to guide lesion removal. Once removed, the probe is used to confirm that the radioactivity is completely within the excised specimen and, importantly, that no radioactivity remains in the excision site.
The margins of the removed specimen containing the lesion are oriented, usually on 3 points, with radiopaque clips or surgical thread. If the lesion is only visible mammographically (for example, microcalcifications or post-mammotome clips), the specimen is x-rayed intraoperatively to verify that all has been removed by comparison with the preoperative x-ray. If necessary, the excision is enlarged and more material removed. If the lesion is a small, nonpalpable nodule, intraoperative frozen-section examination can confirm lesion retrieval. If clinically indicated, intraoperative histologic examination may be performed to establish the diagnosis, and in the setting of malignancy, the procedure can be converted to a therapeutic resection.
Contraindications for ROLL
ROLL is not indicated for the localization of extensive microcalcifications. In such cases, traditional needle localization or placement of a cutaneous marker indicating the area including the microcalcifications can be used. ROLL is also contraindicated for multifocal or large, multicentric lesions. Two ROLL procedures should not be performed in the same breast because it is not easy to clearly distinguish the 2 sites of radioactivity using the gamma probe. If more than 1 nonpalpable lesion is present, 1 can be localized by ROLL and additional lesions can be localized by another method, generally needle localization. A retroareolar lesion is also a relative contraindication for ROLL because the radiotracer may be injected intraductally, increasing the risk of tracer spread away from the target site.
The most common complication is movement or dissemination of the injected radioactivity away from point of injection. When this happens, the localization is a failure and another technique, such as needle localization, must be used. The half-life of 99mTc is only approximately 24 hours.