Surgeons swap scalpels for image-guided tools

Radiology and surgery have traditionally been quite separate disciplines, one the very essence of hands-off healthcare, the other unashamedly invasive. Advances in image-guided therapy are now blurring that distinction, according to Prof. Pablo Ros, a professor of radiology at the Hospital de Sant Pau, Autonomous University of Barcelona.

Modern surgery is becoming increasingly less invasive, while radiology is moving in the opposite direction, he told delegates at this year's Computer Assisted Radiology and Surgery conference in Barcelona. The surgeon's "eye" is being replaced by multimodality imaging and the operator's "hand" by interventional therapy delivery systems.

"We are going from open surgery to percutaneous techniques to closed surgery. From invasive procedures to minimally invasive techniques to noninvasive surgery," Ros said during a workshop on surgical PACS and the digital operating room.

The initial stages of this convergence are already being seen in practice. Radiologists traditionally worked only with 2D images of the human anatomy. Now 3D reconstructions from routine multislice CT and MRI examinations are commonplace.

Greater use of 3D images in surgical planning should help minimize the necessity for exploration during surgery and reduce operation times, Ros said. He showed examples from neuroimaging studies in which 3D reconstructions have provided a blueprint for neurosurgery.

"This approach is now being used in the abdomen as well," he said. "How many surgeons do exploratory laparotomy today? The laparotomy is done using CT. You can model the organ deformation and displacement that will occur when you open the abdomen, increase precision in tumor localization, and improve target resection."

Imaging-led planning could be especially helpful when surgery is planned in delicate areas, such as the lower pelvis. The 3D views could reveal how best to remove a tumor without damaging pelvic function. Software programs for virtual hepatectomy are also becoming available, allowing surgeons to plan resection around the liver's vascular anatomy.

Molecular imaging technologies are likely to assume a key role in surgical planning and follow-up, Ros said. Modalities such as PET/CT, near-infrared imaging, and optical coherence technology are greatly improving the sensitivity with which pathology can be detected. It may even be possible in the future to detect nanometer-scale lesions.

The integration of imaging modalities to novel therapy delivery systems is another emerging area. The operating room of the future is more likely to contain an MRI system, allowing procedures such as cryoablation, laser therapy, brachytherapy, and high-intensity focused ultrasound (HIFU) to be guided and monitored in real-time. Practitioners will be able to trace the path of needles, watch areas of tissue freeze and thaw, and track the position of radioactive seeds.

The first clinical MR-guided focused ultrasound system, installed in the U.S. at the Brigham and Women's Hospital in Boston, Massachusetts, is being used to "fry" troublesome fibroids, Ros said. He predicted that the next step will be to apply HIFU to the brain. Studies have shown that the ultrasound technique can induce temporary permeability in the blood-brain barrier. This could provide a window of opportunity for therapy to be delivered directly to target sites.

The continued advance of image-guided surgery and therapy delivery will clearly require cooperation between radiologists and surgeons. Different surgical, therapeutic, and imaging devices will have to work together as well. The multidisciplinary OR will be achieved only with the help of IT, Ros said.

"This is obviously extremely expensive, but at the end it is proven to be helpful because we can detect disease earlier, and we can do surgery better, in a shorter time, and with less invasive methods," he said.

The vision of the OR room of the future, where surgeons work in uncluttered environments and images are projected onto the walls, has still to be translated into reality. No such room yet exists, said Heinz Lemke, Ph.D., a research professor of radiology at the University of Southern California and senior scientific advisor on medical informatics for the Innovation Center for Computer Assisted Surgery at the University of Leipzig in Germany. Individual vendors may have proposed solutions that appear to match this ideal, but none offer full in-tegration with competitors' equipment.

"What we really need is an approach that allows you to plug-and-play with different types of devices, and has the infrastructure that allows you to do so," Lemke said. "We need standards for this. The current standards, such as DICOM, HL7, and the various ISO approaches, are not sufficient to cover demands in the OR."

He envisages that in the future, with the trend toward personalized medicine, information generated outside the OR will also need to be integrated into the surgical environment. Quality metrics should be devised so that the appropriateness of individual surgical plans can be checked, he said.

-By Paula Gould