Reduced access to radiologists forces self-reliance for patient care and preoperative surgical planning
Radiological data made available to surgeons have historically been limited to patients' reports and the occasional image. Surgeons' ability to interpret images has consequently been gained by default and/or experience, rather than through structured learning. The result is that many surgeons have only a narrow set of usable image interpretation skills.
Many factors are now driving the need to change this situation. Digital imaging and the deployment of PACS have not only widened medical specialists' access to radiological images but have also given them the capability to manipulate and analyze image data. Digital images are increasingly being used as a basis for surgical planning. The expansion of clinical governance procedures is also focusing attention on image manipulation as a way of achieving high-quality patient management.
Certain situations are now arising in which clinicians need to understand the images they are presented with. The U.K. Department of Health, for example, has introduced an outsourcing program for diagnostic tests conducted in England. This has resulted in an increasing percentage of radiological reporting being performed at locations that are distant from the imaging site and the patient. Surgeons' access to local imaging support has consequently been reduced. This makes it ever more important for onsite surgeons to have a comprehensive knowledge and understanding of what information is available and what tools may be used to optimize image viewing and analysis.
What images and manipulations a surgeon will need, where the images fit into an overall patient information model, and how the surgeon should view, use, and integrate the information are important considerations. The surgical perspective is different from the radiological approach to image interpretation. Surgeons have a different skill set than radiologists and radiographers. They also have a different mind-set and like to have data presented in 3D rather than in 2D, as is conventional in radiology.
An initiative to generate a surgeon-friendly patient information model is already under way. This should involve the integration of radiological images into a complete information package that also includes relevant pathological, visual, and optical images, as well as physiological data, diagrams and drawings, blood pressure, and other monitoring information. The intention is to provide surgeons with more fully integrated diagnostic information for each patient.
Surgeons do not need to reach the same skill level as radiologists, who are trained to provide full diagnostic reports from images. They should, however, learn basic image interpretation skills and be able to correlate imaging features with human anatomy and surgical implications. In an ideal world, all surgeons would receive a full 3D presentation and an indication of surrounding structures. In reality, however, surgeons need to interpret and analyze whatever 2D images are available and use these for surgical planning.
An awareness of the basic principles of different imaging techniques is invaluable. This will help ensure the best surgical planning, procedures, and patient management. For example, surgeons must:
• Appreciate the effect of x-ray radiation absorption; how does it appear on images, and what do the different gray levels in the images indicate?
• Comprehend windowing; what does changing window levels mean, and how does this vary between different types of images?
• Understand the differences between plain-film x-ray, CT, PET, MRI, ultrasound, and nuclear medicine images and grasp the relevance of diagnosing with different modalities and manipulating images to reach a diagnosis; and
• Recognize the benefits of combining techniques, such as PET and CT, and appreciate which patient groups might benefit from this approach.
The level and scope of information that an individual surgeon needs to know depends on his or her clinical specialization. The images and processing tools used during presurgical templating for a hip replacement procedure (see figure), for instance, may be different from those required when planning how to reconstruct a trauma victim's fractured leg. Both are orthopedic procedures, but each has quite different preoperative planning requirements.
Surgeons also need access to intraoperative imaging. Real-time imaging in the surgical theater can be used to guide the resection of tumors that are close to critical structures, for example. It may also be helpful to modify presurgical images according to intraoperative findings.
Interfacing between imaging systems and surgical systems for direct surgical planning is on the rise. This requires that surgeons be provided with manipulable image sets that can be transferred directly into surgical systems. Surgeons will also need to update, change, and reanalyze the imaging data during surgery. They must consequently understand the presented images and know how to manipulate them with the available tools.
So how are surgeons going to gain this all-important understanding of image interpretation and manipulation? Are existing systems available that will allow them to acquire these skills?
Many good textbooks, interactive CDs, and web-based tools that cover radiological image interpretation and the correlation between imaging and human anatomy are already available. Most, however, have been developed by radiologists and imaging specialists with a view to educating the radiological community. Virtually none are presented from a surgeon's perspective, and few fully exploit the benefits and facilities of the electronic age.
An image learning system designed specifically for surgeons is required. This could use the exact same knowledge resources, image libraries, and core knowledge repositories that are used for radiologists' teaching aids. Additional information specific to surgeons and specific branches of surgery could then be added as appropriate. The path through the information, and the method of presentation should suit the mind-set and thought processes of all surgeons, whatever their individual specialization.
An "imaging for surgeons" learning resource should at least:
• cover the basics of available technologies and modalities;
• provide an overview of applications, indicating which modality is optimal for what type of patient;
• present the clinical conditions to consider when viewing images;
• describe the relationship between surgical anatomy and radiological images;
• explain basic image analysis and manipulation in relation to surgical planning and integration with surgical systems; and
• illustrate how information from images relates to different surgical specialties.
Technology is continually advancing, and PACS and digital imaging are becoming ever more commonplace. The need for surgeons to access information and learning resources related to imaging is increasing day by day. Modern-day clinical medicine, however, is carried out in an environment that demands greater effectiveness and efficiencies and allows less time for learning. Providing the teaching that surgeons require in radiological image interpretation and manipulation is undoubtedly going to be a challenge.
MS. BECKMANN is director of Lanmark, a consultancy firm based in Beaconsfield, U.K.