From an interview with Dr. James Trippi
Storer, Schmidt and Assoc., P.C
Indianapolis, IN
317-924-5444; sxfm24a@prodigy.com

Each year in the U.S. about 4 million patients present to an emergency room with chest pain for which a myocardial infarction ("heart attack") must be ruled out. For 70% of these cases, the problem is non-cardiac in origin or is insignificant enough that it could be treated on an elective, outpatient basis. The problem is: how to quickly separate the wheat from the chaff, so that resources aren’t spent on avoidable tests and in-hospital observation? Here are some problems faced by anyone trying to streamline the rule-out process:

• Over 50% of "rule-out M.I." cases come into the E.R. after 5:00 P.M., when it is harder to get hold of specialists
• Cardiologists are regularly called in from their homes in the wee hours of the morning and during their off-hours, leading to professional overwork and burn-out
• A significant portion of the time spent in the hospital (by patients with what turns out to be benign processes) is spent awaiting studies to be interpreted.

Cardiologist James Trippi came increasingly to realize the inefficiency of the standard "rule-out M.I." protocol (see Figure below) as he settled into private practice in Indianapolis after completing his residency in 1984. There had to be some way to make the time-motion process of a rule-out M.I. more efficient. The key stumbling block seemed to be the lag in getting echocardiograms read -- especially as it became clearer in the cardiology literature that the sequence of a normal resting echo, followed by a normal stress (dobutamine-induced) echo, sufficed to safely rule out serious ischemic heart disease. Conceptually, it was easy to overcome the stumbling block: get rid of the lag, much of which was caused by delays in cardiologists getting to the hospital at night and on weekends.

In 1993 Dr. Trippi began experimenting with the first version of his solution. This was a specially configured Toshiba C4400 laptop computer, equipped with a modem and special software (Review Software for Windows, TomTec Imaging Systems), that would download echocardiograms from the Methodist Hospital emergency room, intensive care unit, or a patient room via standard phone lines. CinÈ-loop echocardiograms were obtained on line with a digitizing system (TomTec Imaging Systems) integrated into the Hewlett-Packard Sonos 1000 echocardiograph. The four echocardiographic views were acquired 50 ms apart and presented in quad-screen systolic eight-frame cinÈ-loops, at a density of 320 x 240 lines. The diagnostic frames for the cinÈ-loops were determined by the highly experienced acquiring technician. Transmission was performed from the patient’s bedside or from an echocardiographic workstation. Transmission speed was up to 14.4 kbps.

The uncompressed 2-D echocardiograms and Dopplers averaged 0.5 MB in size, and took about 10 minutes apiece to transmit for this store-and-forward application. The laptop computer was given each evening to the on-call cardiologist in the seven-person group. It enabled them to receive and interpret cinÈ-loops of the echocardiograms from their homes. Typical diagnoses included wall motion abnormalities, myocardial contusion, pericardial effusion, and valvular disease. In a series of peer-reviewed abstracts and articles published in 1994 and 1995, Trippi’s group documented that tele-echocardiography studies were essentially equivalent in diagnostic value to standard studies read from videotapes. Agreement rates for serious diseases were consistently above 95%. Besides being extremely reliable, they found that the expedited interpretations greatly compressed the amount of time spent ruling out myocardial infarctions. A study, as well as a comparative cost analysis of standard-vs.-telecardiology management of emergencies, will be published this Spring. While the results of this study are embargoed until publication, Dr. Trippi states that there is a significant dollar savings from this application of telecardiology.

Unfortunately, echocardiography telemedicine is presently not reimbursable by third-party carriers. For reimbursement, interpretations using the telecardiology system are repeated on site the next day. However, Dr. Trippi believes that, by improving patient care, shortening hospital stays, and making on-call duties less onerous, the technology more than pays for itself.

In the meanwhile, work on the equipment continues. Techniques for compressing images using the JPEG format are being developed, enabling 10x reductions in the files being sent and stored. A new Micron laptop computer, with 2 GB of memory for storing images, and 40 MB of RAM for faster processing, will allow images to be transmitted and processed in seconds instead of minutes. Also, the laptop system is being adapted for wireless transmission of images, to enable reception of images on the road.

Dr. Trippi reports that the seven cardiologists using the system have been extremely pleased. Over the past 3 years they have seen nearly 750 cases. Currently they are seeing an average of 0.9 telecardiology cases/day, up to 7 cases in one 24-hour period. Our interview concluded with this comment: "We can give better patient care. It is a new way, and a better way, to do things." [CALLOUT]

Suggested Reading:

Trippi J, Kovacs R, Kopp G, Nelson D. Trans-telephonic interpretation and reporting of echocardiograms using a laptop computer. Abstract, 5th Annual Scientific Sessions, Am Soc of Echocardiography, 1994.

Trippi JA, Kamthorn SL, Kopp G, Nelson D, Kovacs R. Emergency echocardiography telemedicine: an efficient method to provide 24-hour consultative echocardiography. J Am Coll Cardiol (in press)

Trippi JA, Kopp G, Kamthorn SL et al. The feasibility of dobutamine stress echocardiography in the emergency department with telemedicine interpretation. J Am Soc Echocardiogr 9:113-8, 1996