Second opinion readings on outside studies: Should we bother?

O. Clark West, MD, is a Professor of Radiology, Department of Radiology, Memorial Hermann Hospital–The University of Texas Medical School at Houston, Houston, TX.

“Dr. West, would you please take a quick look at the images on this DVD on a new trauma patient?” asks the PGY-3 trauma surgery resident. “Could you also give us a final reading on those?”

Over the past 7 years, my thinking on how to manage “outside images” at our trauma center has completely changed. In 2005, when images on transferred patients arrived on film, or increasingly on CD, one of our radiology residents would look at the studies, correlate them with any written materials — which might consist of a handwritten note or, occasionally, a final report — and answer any questions the trauma surgery resident might have. Attending radiologists would occasionally be consulted for problematic cases; the attending trauma surgeon would also usually be involved in this consultation.

Documentation of this process, however, was not uniform. Formal second-opinion reports were infrequently created, usually only incases of a major discrepancy between the outside report and our consultation. Most often, the communication was entirely verbal. Sometimes, the trauma resident would synthesize the essence of the consultation and record his or her opinion in the clinical notes. With few exceptions, the consultation was not documented in the radiology information system.

That system had one overarching advantage, however — minimizing the amount of time and effort radiologists spent consulting on outside images. I believed then, and I do now, that our consultations were useful in identifying the patient’s injuries and reviewing them with the trauma surgery team. However, the behavior of our trauma surgeons and other critical-care colleagues would suggest that the benefit of our verbal consultation was limited. Examinations were frequently repeated so a final written report could be issued by the attending radiologist.

Admittedly, there were widely recognized problems with this informal, curbside, undocumented consultation. Leonard Berlin recommended, “When asked by a physician for a curbstone consultation, radiologists should, if at all possible, convert the oral consultation into a formal written report, even offering to do so as a courtesy and without remuneration.”¹ Rich Duszak outlined some requirements to creating a billable, second-opinion radiology report. These included 1) a written report, 2) medical necessity at a higher standard that justifies a second report, and 3) an appropriate -77 modifier.² Despite these opinions and knowing we were on thin ice with our informal verbal consultation model, many radiologists continued to limp along with it.

Several recent trends, however, have forced me to reconsider our management of outside images. First, computed tomography (CT) scanning has become the primary imaging modality for evaluation of the multiply-injured patient. Increasingly, CT was being repeated after the patient reached our center to obtain “standardized” images and a formal written report in the hospital computer system. While it was only mildly distressing to repeat a few radiographs, repeating full body CT scans was just too much. At a time when Image Gently and ImageWisely were becoming the topic of conversation, we could not justify continuing to needlessly repeat CT scans.

Second, the images received from referring emergency centers were increasingly of higher quality. In particular, CT protocols in use atour center were widely adopted throughout the community, including the use of thin-section multiplanar reformation (MPR) for face, spine,chest, and abdominal imaging. Rationalizing that in-house CTs were of higher quality and contained more information than outside studies,and thus worth repeating, lost its credibility.

Third, an increasingly large fraction of transferred patients arrived with a complete imaging work-up. The practice of obtaining a few radiographs and shipping the patient out was disappearing. A way to use “outside images” more effectively was needed. In 2011, after several years of planning, our practice implemented a web-based mechanism for transferring DICOM images from community emergency centersto our trauma center. With outside images arriving directly to our picture archive and communication system (PACS), how could we continue repeating studies?

Second-opinion report system

In response to these pressures, our practice developed a comprehensive system for issuing second-opinion reports on images received when patients are transferred to our center.

The key tenets of our program are:

• A physician requesting a second-opinion report must provide relevant patient history, similar to ordering a new examination.
• Medical necessity must be documented.

Valid reasons for soliciting a second opinion include:

• A questionable finding from the initial interpretation where another physician’s expertise is needed.
• A change in diagnosis resulting from a second interpretation of the results of the procedures.
• The second-opinion report represents a higher level of care (radiologist expertise). The higher level of care justification is most appropriate for multitrauma patients transferred for expert care, pediatric patients transferred for surgical care, and stroke patients transferred for aggressive stroke intervention.
• An acutely ill or injured patient arriving at our hospital with images but without complete radiology reports. (Re)-interpretation of the images is necessary for immediate care.
• Upon review of outside images and reports, the admitting physician at our center has questions requiring expert radiologist consultation.

The simultaneous review of all original reports from the community hospital is necessary for good medical practice. Radiology clerks assist us in finding original reports when they exist. Second-opinion consultations should occur physician-to-physician to optimize information exchange. When outside images do not meet our strict criteria for second-opinion reporting, the images are uploaded to the PACS and held for reference.

Start-up issues

Implementing this program required a great deal of cooperation. Radiology clerks were trained to upload CDs and DVDs and to manage studies received electronically. Radiology department team leaders were taught how to create orders for each outside examination and to link these images with the appropriate order in PACS. This is a time-consuming process requiring technologists who are not equally versed in all imaging modalities to perform a series of repetitive but complex operations. Our colleagues in the emergency center and throughout the hospital assisted us by appropriately requesting second-opinion reports and in supplying original reports when available.

Start-up required several months, including a painful period of several weeks when every incoming study received a second-opinion report. The situation improved with adoption of the rules outlined above. Our trauma surgeons, emergency physicians, and colleagues throughout the hospital seem satisfied with the improved availability of second-opinion reports. An initial analysis of reimbursement reveals that second-opinion readings are compensated almost at the same rate as primary in-house examinations. In practice, we issue second-opinion reports most frequently on CT and magnetic resonance imaging (MRI). Most radiography examinations are uploaded for reference only. When a study must be repeated in our hospital, we usually use the preceding outside study for reference and do not issue a second-opinion report.

While we are pleased with our initial experience with electronic image transfer and second-opinion reporting, we are working toward improvements. For example, our current electronic image transfer system lacks the means to capture and transmit original reports from the sending hospital. Currently, we rely on paper reports sent by facsimile, which are then scanned into the PACS. Within a year, we should have a standardized method of capturing reports as images.

In addition, the process of creating orders and linking them with images is labor-intensive and needs to be automated. Developing an intelligent interface between our electronic image transfer system and PACS is needed. Finally, a feedback mechanism is needed so the sending radiologist can see the content of our second-opinion reports. This could stimulate the sharing of information and improve the quality of care at both centers.

You may now be wondering why we would undertake this much extra work. Very likely you wish to avoid adopting a similar process in your hospital.

I would submit that we have a duty to our patients to make maximum use of the imaging evaluation that they received before being transferred to a tertiary care facility. We can no longer justify repeating imaging studies for physician convenience and ease of billing. In CT, the radiation and cost of a repeated scans are too high. For all imaging examinations, the technical fees for repeating examinations are high. The relatively modest professional charges for providing duplicate second-opinion readings are far less than the cost of repeating entire examinations, particularly in CT and MRI.

As more hospitals adopt second-opinion reading programs, their value can be more rigorously assessed to determine if the time and expense is justified. Such evidence is important in convincing third-party payers that a streamlined method for reimbursing second-opinion reporting is warranted.

In the meantime, you may wish to start planning for electronic image transfer and second-opinion reading in your practice. It is a trend that is unlikely to go away.

References

1. Berlin, L. Malpractice issues in radiology: Curbstone consultations. AJR. Am J Roentgenol. 2002;178:1353-1359.
2. Duszak, R. Another unpaid second opinion. JACR J Am Coll Radiol. 2005;2:793-794.