Return to CAP Home
Printable Version

  For AFIP, real-time telepathology
  goes far afield

  cap today

January 2005
Feature Story

The 31st Combat Support Hospital in Baghdad processes all manner of tissue at unpredictable rates with only a single pathologist, far removed from any kind of consultative ability or friendly academic center.

But remote military outposts like this have a real-time lifeline thanks to the dynamic telepathology network constructed over the past few years by the Armed Forces Institute of Pathology at Walter Reed Army Medical Center, Washington, DC.

"They are processing tissue, and on occasion consultation is required," says Keith Kaplan, MD, MAJ, medical director of cytopathology and the Army telepathology program, who championed the system and won funding for it. "The question I asked is, ’Is this as good as looking through the microscope? If this were a biopsy from a relative, would I be comfortable with someone reviewing it by telepathology?’"

The 21-site system, which uses digital cameras linked to robotically controlled microscopes so the distant pathologist can zoom in on or move around a slide as needed, was used 282 times in the first 11 months of 2004, Dr. Kaplan says. The Internet-based system combines quicker turnaround time than commercial overnight couriers and the manipulative capabilities not possible with e-mailed photos, winning over skeptics and inspiring pathologists to find uses for it that even Dr. Kaplan would not have predicted.

"The big thing is, it helps you with difficult diagnoses. You have an expert in the field," says Dave Mendoza, MD, chief of anatomic pathology at Blanchfield Army Community Hospital in Ft. Campbell, Ky., who first used the system in 2003 at the 121st General Hospital in Korea. "I’m still using it over here. It’s great to have an expert— somebody who does nothing but look at skin, or nothing but look at breast all day."

Another user is Barbara Crothers, DO, chief of anatomic pathology at Walter Reed and Dr. Kaplan’s boss, who was stationed in Korea in 2002 and 2003. "When you’re in a remote site, and you’ve got only one other practicing pathologist with you, you just need someone else to bounce the case off of," she says.

One nuisance for those stationed overseas is that courier deliveries take longer— if they can get them at all, Dr. Kaplan says. "For Korea and Germany it’s really been a benefit," he says. "They’ve told me it’s actually decreased hospital stays because they’ve been able to get the turnaround. People upload these cases, and they get a call within a couple of hours."

Or at least the next morning. "I could put a case on as I was leaving at the end of the day, and the next morning have a consultant’s report faxed to me, waiting for me, kind of overnight, while I slept," Dr. Mendoza says. A package from Korea might take two or three days to reach the AFIP, and then "they have to accession it, they have to look at it. The typical turnaround time was two weeks before we’d get a faxed report."

Dr. Crothers says the real-time nature of the system made a huge difference in Korea. "We had a couple of cases whereby if I had had to actually send the cases off to AFIP, it would have taken several weeks—as opposed to a day or two." She recalls the overnight routine Dr. Mendoza described but cites one drawback: "We could load only one slide at a time. They couldn’t call us up and say, ’Could you load up another slide?’"

The system is not without other drawbacks: It’s not suited to multiple-slide cases, those requiring special stains, and certain others, and it’s afflicted with the same occasional technical difficulties as any Internet-based technology.

Still, at this point, only technophobes are avoiding it entirely, says Joseph P. Pulcini, MD, LTC, chief of anatomic pathology at Brooke Army Medical Center at Ft. Sam Houston, Tex., who counts himself among the early skeptics.

"I was very vocally saying this was a waste," Dr. Pulcini says of the system, which, for the first five systems, was $48,000 per site. (The price for additional sites was lower.) He knew it would cost "a couple million dollars to get this up and running," and wondered, How often is this going to make a difference in terms of patient care, to have an answer in a few hours versus just sending a case out and having it in a few days? And how efficient is it when you’re navigating a slide using a mouse and a robot to focus? For the number of times it would make a real difference, he thought, you could put someone on a plane with a first-class ticket to the Mayo Clinic.

"And then I actually started using it," Dr. Pulcini says. "The learning curve is not that steep." It’s mainly older pathologists and others uncomfortable with technology who avoid it. "They’re not part of the video-game generation. ’What is that thing in the corner?’ They don’t want to know." But over time, it has become an accepted part of the anatomic pathology service. "I work in a large bureaucracy, obviously, which is where telepathology is most useful—when you have a big, closed system like the Army. We all work for the same people." Dr. Pulcini now uses it about twice a month.

Eric Berg, MD, COL, chief of the Department of Pathology at Blanchfield, says his department’s use of the system has jumped noticeably, to one or two times per day, since Dr. Mendoza and another younger pathologist came to work at the hospital. "They use it quite a bit," Dr. Berg says. "It’s increased by virtue of the new people."

Dr. Kaplan thinks part of the resistance has come because earlier generations of telepathology, which has been around in one form or another for nearly three decades, did not deliver as promised. Skeptics say: "We’ve tried this, and it didn’t work. We’ve heard about this before, and it doesn’t work. I’m too busy to do what’s functional here. I don’t have time to try what’s new and experimental, somebody’s science project."

"The idea of telepathology has been tried for years," Dr. Mendoza agrees. "Until now, the resolutions on the screen just weren’t high enough. There wasn’t enough throughput on the Internet to get good imagery. There’s still room for improvement, but we’re light years from where we were even five years ago."

Within the Army, usage has at least doubled over a year ago, when fewer hospitals had systems installed and fewer pathologists used it where it was in place, Dr. Kaplan says. "Initially, there was a bit of a lull, a little bit of lag. Then people started using it," he says. "We’re working on a system whereby I can track how many cases are being sent from where to where."

Dr. Kaplan says he’s surprised by the amount of use the system has gotten for specialties like hematopathology, hematology, and cytology. "The lack of oil has not been a major hindrance," he says. "I use it for cytology at 40X or 60X and have yet to defer a case because of the image quality." He predicted a lot of reluctance, but the hematopathologists have turned out to be the biggest proponents. "I thought for sure they’re not going to want to look at [a sample] unless it’s dunked in oil, at 100X, on their own microscope," he says.

Dr. Crothers says she sees it used most often for dermatopathology. "We frequently will get a second or third opinion from someone in AFIP for a case we’re pretty sure about but need a little hand-holding with," she says.

At Blanchfield in Kentucky, Dr. Berg says it’s used most often on breast, prostate, and skin biopsies. "We use it half the time for patient satisfaction as much as our own," he says. "You don’t want to leave them hanging all weekend with a breast biopsy that’s in doubt."

Dr. Pulcini says he’s found it most useful for one-slide cases in unusual fields like oral pathology, but he figures Brooke’s experience, because of its large size, is different from most. "I’m still skeptical about, OK, how often does this make a difference in terms of, do you have to use telepath versus just sending the case out?’" he says, but adds, "We have a lot of subspecialty-trained pathologists here. Our experience here is unusual because we’re more set up to be consultants rather than need consultation. A smaller place more often needs the help."

Remote telepathology is not useful for larger and complicated cases, say Dr. Kaplan and others, though all have slightly different takes on when not to use it.

"The images are 99 percent as good as what you see through your microscope. But sometimes you need that one percent; you want to do your own steps and stains and so forth," Dr. Kaplan says. "I suggest people not use them for complicated pigmented lesions. It also has been an issue for very minute biopsies, prostate needle-core biopsies, or very thin biopsies where you’re looking for very small foci of disease."

Dr. Crothers would send out any multiple-slide breast case, "knowing that our consultant is going to need to do more than review a couple of images," and most cases that would require "further workup through molecular markers, chemical stains, or any other kinds of special stains," she says. "It’s somewhat more limited because you would have to do those special tests first. That would be more time-consuming. In those cases, it’s easier to just package up a case and send it out."

Dr. Berg has never used the system for more than a three-slide case. "It’s pretty much limited to small biopsies, or single slides of interest in larger cases," he says, adding, "It’s not useful for frozen sections—you don’t get an answer quickly enough."

Dr. Pulcini would not recommend remote telepathology for cytology. "There’s a lot of screening," he says. "You have to look for a lot of individual, abnormal cells. There’s a lot of three-dimensionality involved." He shares Dr. Berg’s skepticism about frozen sections but thinks the system can work for that purpose "in a pinch."

"It’s very hard to call AFIP and say, ’I’ve got a frozen, drop what you’re doing.’ The guy 2,000 miles away is not at your beck and call," he says.

Another limitation of the system is one that plagues any Internet-based technology: interrupted connectivity, which can be worse in the Army because of the elaborate system of firewalls set up to meet security requirements.

"There may be hospital firewalls and base firewalls; there may be different levels of security," Dr. Kaplan says. "There are some routers in Europe and Asia that we don’t control. If those go down, that is beyond our control. And if the hospital LAN is down somewhere, it doesn’t work. In Iraq, they didn’t do cases [remotely] for a couple of months because they’ve had, as you can imagine, some issues with power."

Dr. Mendoza remembers that the firewalls in Korea made the system harder to set up there than stateside. "It was a little bit harder tunneling through," he says.

"We have firewalls in place that the average private hospital would never dream of," Dr. Pulcini says. "We have to be careful who gets let in."

Dr. Crothers cites another internal Army problem that surely affects most large bureaucracies. "The information management people are constantly upgrading, downloading, and changing things on the network," she says. "They don’t always check to make sure what kind of impact their changes are going to have on our system. Sometimes our system is not functional, and that will put us up for a day or two."

Dr. Kaplan predicts that dynamic telepathology as it’s now known will be eclipsed in perhaps five years by whole-slide telepathology systems, which enable the pathologist to capture the entire slide at high resolution and display it at multiple resolutions, with the goal of imaging all slides in a multi-slide case. (Related article: "Digital imaging, telepathology: what’s the holdup?" July 2004.)

Virtual slides will combine the ease of e-mailed stills with the manipulative capabilities of the robotic microscopes. "Virtual slides are the Holy Grail of all this," Dr. Kaplan says. "What we have now works well, but you’re still dependent on routers and the Internet being up and the robotic microscopes. For the military, [virtual slides] are going to be an awesome tool."

Dr. Pulcini says the virtual slides will have "fewer moving parts" and "might be more glitch-free" than the robotic microscope, but he sees some of the same problems—as well as a new one. "Bandwidth becomes an issue with a virtual slide. It’s a huge file," he says.

But whether it’s virtual slide or robotic microscope, the "ultimate hurdle," Dr. Pulcini says, "is that it takes longer to look at the specimen." If it’s a large specimen with a lot of subtle features, you’re moving around with a mouse rather than moving a real slide with your fingers.

"The advantage to [either] system is, it’s real time," he adds. "And although this doesn’t usually happen, you can be on the phone with the other pathologist. And you usually get a fax a couple of hours later."

Ed Finkel is a writer in Evanston, Ill. The opinions and assertions contained in this story are not to be construed as official or reflecting the official views of the U.S. Department of the Army or Department of Defense.

The Army telepathology network uses systems manufactured by Trestle Corp., Irvine, Calif.