Charting the Advance of Cardiovascular Image & Information Management A Conversation with Thomas Kennedy, PhD*

In this interview with Cath Lab Digest, Dr. Kennedy reflects on the evolution of cardiovascular informatics, comments on current information management challenges in the cath lab, and paints a picture of the integrated cardiology department of the future. In 1997, while co-chair of DICOM Cardiac and Vascular Working Group I, you co-edited, along with Dr. Steven Nissen from the Cleveland Clinic, Digital Cardiac Imaging in the 21st Century: A Primer. The Primer, which was a joint effort of the Working Group and the professional societies in cardiovascular medicine, was intended to educate the cardiology community and help make them make more informed decisions about digital cardiac imaging technology. How have the issues changed since that time? Back in 1997, the cardiology community was still working to understand the implications of filmless operation. Questions broke down along two lines: First, there were questions related to the technology and mechanics of converting a cath lab: How do I connect the x-ray acquisition systems? How do I store the data? How fast can I access studies for review? What is DICOM? Are these systems cost-effective? A second set of questions arose, related to the clinical impact of filmless conversion: Do I lose clinical information by giving up film? What are the effects of image compression? Does the image quality presented at digital review stations match the in-lab image quality? These fundamental questions needed answers before cath lab administrators and clinicians would fully embrace digital archive and review systems. Fortunately, most of these questions have been answered through the experiences of early adopters, and through more controlled studies such as the American College of Cardiology (ACC) study on lossy compression. As a result, we have experienced a steady conversion to filmless operation since 1997. Technology has also advanced dramatically since 1997. All vendors have fully embraced the DICOM Standard for describing and communicating with all cardiovascular modalities. Computer, storage, and network technology has advanced to the point that systems based on proprietary technology are no longer required or even viable. Virtually all systems have converged to the same client-server architecture, although we are seeing an increased use of web-based technology for distribution of studies. All cath lab archive and review system vendors are taking advantage of the favorable cost and performance trends across all the core technologies. High-speed, disk-based storage (RAID), for example, has seen dramatic increases in capacity and performance with nearly proportionate decreases in cost. Networks are faster and more scaleable than they were in 1997. Archive technology has advanced: tape technology is faster with orders of magnitude more storage capacity, and optical media such as CD-R has given way to the higher performing DVD-R. New technologies, such as network attached storage (NAS) and storage area networks (SAN), give departments and entire healthcare enterprises significant opportunities for affordable scalability. It is interesting that while technology improvements have certainly advanced filmless conversion, they also have made it difficult, at first glance, to recognize the differences between all of the solutions offered by vendors today. At the clinical level, I think nearly everyone would agree that digital review is equivalent to film-based review, as long as studies are carefully captured and well-designed image processing algorithms are applied at the workstation to preserve in-lab image quality. Also, most vendors have moved away from low-end image acquisition devices or frame grabbers, since these devices typically degrade image quality to unacceptable levels. Archive and review systems have improved access to prior studies for comparisons and physician-to-physician communication is enhanced for consultations. The biggest change since 1997 has been an increased realization that cath studies are one component of the overall cardiovascular patient record. An isolated cath lab network leaves significant gaps in strategies to improve operational efficiency and reduce costs across all cardiology services. In fact, the demand for a comprehensive cardiovascular information solution is driven in part by the realization that an image-only cath lab network may not be cost effective. Business concerns are as important as the clinical requirements in selection of the information system. Today, the questions in the cath lab are these: How do I improve the workflow in my cath lab? How do I integrate my image management system with my hemodynamic system? How do I effectively control my inventory? How do I integrate clinical reporting? How does an information system help me drive quality initiatives and improve outcomes? Are there similar concerns in other cardiovascular modalities or departments? Absolutely. Other modalities and department services such as echocardiography or nuclear cardiology are experiencing the same pressures as the cath lab. Today, the real trend is integration across the enterprise and creating an end-to-end cardiovascular patient record. Demand has gone from an exclusive focus on image management systems to pan-cardiology information systems that manage all data from all patient encounters -in other words, a true Cardiovascular Information System (CVIS). This trend is an outgrowth of the concepts of the Digital Integrated Cardiovascular Record (DICR) developed by the ACC in collaboration with DICOM Working Group I and first introduced in 1999. From 1997 to today, we’ve gone from an emphasis on enabling technologies like storage, computers, networks, image compression, etc., to a focus on healthcare informatics. The Primer, if re-issued today, would have to answer the questions resulting from the vision of the DICR: How should a comprehensive CVIS be specified? How can the key information needed for management of cardiovascular patients be accessed from a single location? How should the data be presented? How can that data be used to modify my care of the patient? And so on. How does the DICOM Structured Reporting (SR) figure into the Digital Integrated Cardiovascular Record concept? The structured reporting effort of DICOM Working Group I is really an effort to provide some supporting information along with image and waveform data sets in the cath lab environment; other working groups are addressing other specializations like echocardiography. DICOM SR, as it applies to the cath lab, will provide a standard method for interchanging things like waveform annotation, measurements, and procedure information. This data obviously helps the physician put the image and waveform data into its proper clinical context, and will enhance exchange of key information, such as measurements, between devices and reporting applications. I don’t think there should be an expectation, however, that DICOM Structured Reporting will provide a framework for a comprehensive cardiovascular reporting. DICOM SR is really not intended to be a fully expressive model that allows physicians to report the wide range of findings that they may encounter. Instead, I would look for vendors to develop detailed reporting applications that can optionally generate a more distilled DICOM SR file for interchange of studies and data. This model is similar to what we see for the capture of ACC-NCDR data. The ACC-NCDR data elements are not intended to be a complete data set for describing cath studies with all their variability. In practice, most of the ACC-NCDR elements are harvested from the more detailed procedure logs and reports. Continuing with standards activity, what is your assessment of the efforts to apply the Integrating the Healthcare Enterprise (IHE) type of framework to cardiology? The IHE framework went a long way towards solving a nagging problem in radiology PACS implementations: The three core information systems impacting radiology are the Hospital Information System (HIS), Radiology Information System (RIS), and the Picture, Archive, and Communication System (PACS). Each system is designed to manage certain components of the enterprise and department workflow and patient information. While these systems are based on two well-defined standards, HL7 (HIS, RIS) and DICOM (PACS), there happened to be considerable gray areas on how communication should occur between these systems to maximize efficiency. As a result, the benefits promised by a filmless radiology department were either never realized, or realized only after considerable additional cost and aggravation. The IHE framework allowed the vendors to resolve the disconnect between systems and return their focus to optimizing their own applications. Some of the existing elements of the IHE framework could be applied to cardiology and the cath lab to improve workflow, but significant definition is required before a framework can be adopted. In general, the cardiology workflow models are more complex than those seen in radiology, and there are a larger number of data and information sources that must be linked. There was an IHE for cardiology planning meeting during the American Heart Association Conference last autumn, and all participants agreed that there could be some value to an expanded framework for cardiology. The most notable proposal from the meeting is that we need a gap analysis between what was provided for radiology and what cardiology needs, so work will begin to define requirements and priorities for this environment. With the recent announcement of American College of Cardiology support for the IHE for Cardiology initiative, I am sure that we will see vendors and clinicians work together to develop a framework that addresses core workflow issues. With chronic staffing shortages, increasing demands for services, and declining reimbursement, cath labs are stressed and will continue to face challenging times in the years ahead. In what ways can image and information management technology help provide relief from some of these pressures? By streamlining processes, the technology will allow cath labs to operate more efficiently and cost-effectively. It comes down to effective information management. By effective information management, I mean much more than just filmless operation, but rather an integrated environment where data in the cath lab is gathered, codified, and analyzed to drive quality, outcomes, and cost improvements. Most departments continue to transcribe things by paper, and information is copied from one piece of paper to the next. Obviously this creates the potential for lost, incomplete, or even incorrect information for a patient. Most departments are not wired so that data moves from one system to the next. For example, if you introduce an inventory management system into a cath lab independent of the hemodynamic monitoring system, you will be able to track device usage, but it is still difficult to correlate outcomes to those same devices. A series of disconnected information systems ultimately provides little value. The ultimate solution for the cath lab is an integrated information environment where data moves seamlessly from one system to the next in order to build the patient record. In the cath lab, a lot of this integration will occur through the hemodynamic monitoring system. In the ideal environment, the image acquisition and image management systems will tie into the hemodynamic system to link patient demographics, images, and measurements. Procedure information, device usage and other critical information captured in the hemodynamic monitoring system will then flow into the clinical reporting application use by the physician. By the time the physician sits down to review and report the study, the report will be almost completed. The physician will document the findings, digitally sign the report and forward it to a list of pre-defined recipients, including the Hospital Information System. Data integrity will be improved, report turn-around times will be decreased, and billing can be completed sooner. Additionally, the data will be available for analysis at a department level and can be used to answer questions like: What is the inventory usage this month? Which procedures are causing us to use more disposables or devices in the lab? Can we correlate negative outcomes against a particular device so steps can be taken to reduce complications or length of stay after the procedure? Integrating the cath lab really represents the opportunity to embark on quality initiatives that will allow departments to change the way they work. Many labs are installing new high-resolution x-ray systems with flat panel digital detectors. The volume of image data generated by these systems can be up to four times greater than conventional x-ray systems. What kind of issues does this raise with respect to archiving and information management? Frankly, it shouldn’t raise any issues as long as the information management system is scaled appropriately. As I mentioned earlier, the system technology has advanced to the point that the increased loads of these data sets can be handled with standard computer and storage technology. The systems must be scaled to meet the requirements posed by the larger data sets. For example, on the archive side, these larger data sets are an argument for an enterprise-level archive solution such as SAN that could either be disk- or tape-based. There are a number of storage options available, and the choice of a particular technology will depend on the patient volume and practice patterns of a particular lab. You do not want to be locked into any particular storage type, given the pace of technology advancements. High-resolution imaging is not a groundswell that is forcing fundamental changes in system architecture or expensive new solutions. Some have tried to make connecting to these new x-ray systems and managing the high-resolution images as a point of competitive differentiation, but this is mostly marketing hyperbole. The more important concern should be image quality. Vendors must have the knowledge and experience to implement the image processing algorithms required on the review station, maximizing the clinical value of the high-resolution data set. What are the key questions that cath lab and cardiology decision makers need to ask when selecting a vendor for a cardiovascular information system? Obviously, it’s a complex decision-making process, but there are fundamental issues that must be addressed. Beyond the details of image acquisition, review, and storage, the success of CVIS in the cath lab and across all cardiovascular services depends on integration. There is integration of the image management system with the hospital information system to manage orders. There’s the integration with the hemodynamic system, so images, perioperative procedure logging, waveforms, devices and drugs, and other information are linked together to build the procedure report for the physician. Finally, there is the consolidation of data across all sources so that it is available for clinical and business analysis. I believe that the decision team should take a very broad view of cardiovascular information systems particularly in light of the American College of Cardiology vision for the Digital Integrated Cardiovascular Record. A cath study is just one event in the lifetime of a cardiovascular patient. You want an information system that permits data to move with the patient across the entire continuum of care. Vendors whose primary focus is on the image management part of the problem cannot support this model of care. The wall between the office-based record and the hospital-based record is an example of how the continuum of care model breaks down. Today, a physician can know quite a bit about a patient from encounters in the physician’s office. But as a patient comes into the cath lab, the information already obtained by the physician is unknown to the cath lab staff. Core data, such as history of present illness, past history, risk factors, and allergies must be gleaned from the patient. Obviously this process is inefficient and subject to error. Instead, visualize an environment where the patient’s record follows the patient from the office or clinic into the cath lab. Key patient information is automatically available at the hemodynamic system. When the physician goes to generate the procedure report, all the information from both the office and the hemodynamic system are built into the report. The physician provides the diagnosis and care plan and signs the report. This report, in turn, is routed back to the office and the Hospital Information System and is available up on the hospital floor. As the patient moves through the system, the data is aggregated and available to the caregivers at any time. In this scenario, the cath lab is one of many information sources in the lifetime of the patient. This is the direction that the more advanced cardiovascular information systems are going today. The questions customers should now pose to vendors are these: How does your solution help me create a comprehensive Digital Integrated Cardiovascular Record? Can you take responsibility for all facets of a pan-cardiology solution: multi-modality image management, inventory management, hemodynamic recording and procedure logging, clinical reporting, and data mining? Does your solution span the continuum of care between the inpatient and outpatient environments? These are the questions that in many ways, demonstrate the evolution of cardiovascular information systems over the past five years.

NULL