August 2011
Feature Story

Anne Paxton

A “decision diamond” marks the point in a flowchart where the algorithm branches depending on the answer to a question, and in clinical pathology, that question tends to be cut and dried: Does the patient’s laboratory result fall outside a certain range? If yes, the result could be a critical value, and would be reported as such to the clinician.

But in surgical pathology, the notion of “critical values” is more nebulous, and decision diamonds are not nearly so easy to devise. If the diagnosis is a clinically unsuspected malignancy or rejection of a transplanted kidney, there may not be much doubt as to whether the pathologist should make the call. But what about crescents in a kidney biopsy specimen? Or absence of villi in uterine contents?

When surgical pathologists are asked what they think critical values or “panic values” would include, “people who are surveyed give answers all over the place,” says Raouf E. Nakhleh, MD, professor of pathology at the Mayo Clinic in Jacksonville, Fla., and chair of the CAP Quality Practices Committee. Among responders, the highest agreement he has seen on any one diagnosis was about 50 percent. “There’s been little consensus as to what is truly a critical diagnosis.”

Now, however, through a project of the CAP’s Pathology and Laboratory Quality Center, better guidance for anatomic pathologists is here. A workgroup of the CAP and the Association of Directors of Anatomic and Surgical Pathology (ADASP) has developed a “Consensus Statement on Effective Communication of Urgent Diagnoses and Significant Unexpected Diagnoses in Surgical Pathology and Cytopathology.”

Approved last month, the statement recommends that each institution set up a policy regarding “urgent” diagnoses and “significant unexpected” diagnoses, communicate such diagnoses within a clinically relevant time frame, and document the communication (The Statement in Brief). It will be published in the Archives of Pathology & Laboratory Medicine, tentatively in early 2012.

“The basic thrust of the statement is No.1, based on the Joint Commission, on CAP lab accreditation, and on CLIA regulations, you need to have a policy in place,” says workgroup member Barry R. DeYoung, MD, clinical professor and co-director of anatomic and surgical pathology, University of Iowa Health Care. “No. 2, it needs to be a policy that works for your institutional characteristics and your clinical staff. And No. 3, you need to show you’re actually implementing and acting on the policy you’ve put together.”

Critical values continue to be a front-burner issue in health care. In July, the National Quality Forum, an influential voluntary standard-setting organization, endorsed a new list of 29 “Serious Reportable Events in Healthcare,” events that are considered largely preventable. Included in the care management events are two new items: “Patient death or serious injury resulting from the irretrievable loss of an irreplaceable biological specimen,” and “Patient death or serious injury resulting from failure to follow up or communicate laboratory, pathology, or radiology test results.”

Specifically in the area of critical values in AP, the College has been aware of the need for recommended practices. A 2003 Q-Probes assessing physician satisfaction with AP laboratory services found the lowest satisfaction score was for communication, including timeliness of reporting, communication of relevant information, and notification of significant abnormal results.

“The College chose to create the Pathology and Laboratory Quality Center so we would have evidence-based guidelines that can be written for pathologists, and since there has been some controversy about critical values in AP, this was adopted in 2010 as an area we wanted to address,” says Dr. Nakhleh, who chairs the workgroup.

The use of the term “critical value” or even “critical diagnosis” may be somewhat misleading in surgical pathology, says Jan F. Silverman, MD, chair of the Department of Pathology and Laboratory Medicine, West Penn Allegheny Health System, Pittsburgh, and professor, Drexel University College of Medicine and Temple University College of Medicine.

“It’s not a specimen that goes on a machine, with an instrument determining the value and a technologist making the call,” Dr. Silverman points out. “It’s a pathologist who needs to communicate the result, and the pathologist is not only calling to report the diagnosis, but there may also be discussion with the clinician over the implications of it. For years, pathologists have been making these phone calls, but they were doing it anecdotally. They didn’t really have guidelines to help them.”

Since 1972, the year that the article by George D. Lundberg, MD, “When to panic over an abnormal value,” was published in MLO, “there’s been an expectation that you have to call within an hour. And in AP we know that’s not the case because the specimen has to go through processing, and it may be a day or more later until the pathologist even gets the specimen or has a slide to look at,” Dr. Silverman says.

Nevertheless, regulatory and accrediting agencies have used the term critical diagnoses or critical values, then tried to apply it to AP even though, because of the time frames, it doesn’t work. “So one of the main drivers for this whole project was to address the nomenclature, “ Dr. DeYoung says.

The workgroup developed its guidelines for two categories of cases. “We are talking about two separate cohorts here,” says Dr. De Young. “The urgent ones, which are kind of the AP analogy to critical values, and the significant unexpected ones. The urgent category we all understand and can get our hands around. The significant unexpected ones are really going to be defined locally by the institution, and they’re going to be highly variable from spot to spot.”

In the clinical lab, some conditions—a high potassium, for example—need to be addressed right away or something bad is going to happen to the patient, Dr. Nakhleh notes. “Most of the time that is not the case in surgical pathology. Rarely does a patient die in the hour of making the diagnosis. In AP it doesn’t make sense to require a call in 30 minutes if you’ve already had the specimen for an entire day. That’s why we changed the term to ‘urgent.’ We do still believe there are situations where you have to urgently call, but we wanted it to be separate from critical.”

One of the workgroup members, Dr. Nakhleh says, did a search of legal cases brought against anatomic pathologists in which claims were made or lawsuits filed because of problems with communication. “In every one of those instances, the problem was not that communication didn’t occur quickly; it was that communication did not occur, period. It was never a matter of life and death because there was a one-hour delay in the communication.”

Obviously, such cases are the extreme, but they indicate what can happen, Dr. Nakhleh says. “In some cases, there were several physicians involved and communication did not get through to all of them, the information didn’t get back to the treating physician, sometimes the treating physician didn’t look at the report, and rarely the patient didn’t come back for followup.”

The reason that the CAP Center decided to produce a consensus statement rather than a practice guideline is that strong evidence for how to deal with critical values in AP doesn’t exist, Dr. Nakhleh notes. “If the evidence is there that something should be done, we would call it a recommendation and the document would be a guideline. Here we called it a consensus statement because the evidence was rather poor. We know that if a diagnosis is not communicated, bad things could happen, but this is not going to be tested in a controlled trial. On the other hand, critical values are already embodied in the law; it’s part of CLIA ‘88, part of accreditation by the Joint Commission and through the CAP Laboratory Accreditation Program. So we felt strongly that we had to deal with this issue in a commonsense way for the practicing pathologist.”

Among pathology departments generally, there is a lot of variability in policies on urgent values in AP right now, says Dr. Silverman. His 2006 article, “Critical values in anatomic pathology” (Arch Pathol Lab Med. 130:638–640), reported on a retrospective review of surgical pathology reports, which found that critical values in surgical pathology are “uncommon but not rare, and the common practice was largely to notify clinicians based on common sense and personal experience,” he says.

A 2007 CAP survey, the results of which were published in 2009 (Nakhleh RE, et al. Arch Pathol Lab Med. 133:1375–1378), showed that “about 25 percent of pathology departments had no policy, and 30 percent had only a policy that was very generic with no examples,” Dr. Silverman says.

Transplant rejection is an example of a diagnosis that Dr. Nakhleh considers urgent. “Frequently clinicians ask us to rush a biopsy because they’re worried something might happen to the patient, so if I get a moderate or severe rejection diagnosis, I usually call. Or if I see an infection such as cytomegalovirus that I don’t expect in a biopsy, that’s something else I’d make a call on.”

But people have to use their judgment, Dr. Nakhleh emphasizes. “If I am looking at a stomach biopsy for gastritis and I see a cancer, it’s very important for me to make sure the clinician knows that. But it’s not so important that I call them in the five minutes that I see the case. I just want to make sure that that communication happens and the diagnosis is not overlooked.”

Dr. DeYoung’s most recent urgent diagnosis was a fungal infection in a transplant patient. Iowa does a fair number of transplants, he notes, “but any infectious disease in the context of someone who’s immunocompromised or immunosuppressed falls under the rubric of ‘that is an urgent diagnosis that needs to get communicated in a timely and prompt manner.’” Almost anyone can encounter such patients, he notes. “If you have patients getting treated for rheumatoid disease, they’re immunosuppressed, so they’re at risk of developing an infection.”

Another recent case Dr. DeYoung encountered was bacteria in the heart valves. “The clinician was aware in that particular case, so it did not lead to any significant change in therapy. But in a case where the patient didn’t carry that diagnosis preoperatively, they would immediately initiate high-dose antibiotic therapy.”

A diagnosis that can come up and should be considered “significant and unexpected,” Dr. Silverman says, would be where a hernia is examined and the pathologist finds a malignancy in it. “That would be totally unexpected and also significant, and it would be recommended that some prompt communication of some sort be made to the surgeon or primary care physician, because it’s probably one of the last things they would be thinking about with a hernia.”

One reason the consensus statement emphasizes collaboration when institutions are developing their AP critical values policy is that customary reporting practices vary from hospital to hospital. For example, rejection of transplant is on the ADASP list, but not on West Penn Allegheny’s, Dr. Silverman says, “because our nephrologists in every case will come up and actually look at the slides after the biopsy and discuss the case with our renal pathologist. Therefore, there is no need to make a phone call at our institution; it would be redundant.” This is why anatomic pathologists have to work with the institution’s medical staff, as the consensus statement calls for, he adds.

On the other hand, there would be fairly good consensus that a malignancy causing superior vena cava syndrome, or a neoplasm causing paralysis, would be a critical diagnosis, Dr. Silverman says. “As soon as you recognize those microscopically, you’re going to want to call the physician, because the longer they sit, potentially the greater the damage that can happen to the patient.”

Similarly, if a patient has endocarditis or needs to have a valve removed, “we’ll make the call,” he says. “In some of those situations, clinically they may already be suspected, but we may not know because the information may not be supplied by the surgeon. In a good percentage of cases they may say ‘Thanks, but the patient is already being treated with antibiotics.’” The presence of mesothelial cells in a heart biopsy specimen is another example of a condition meriting a call. “In a situation like that, maybe the biopsy went too deep and you’re sampling the surface of the heart,” he says.

While the CAP Center and the workgroup consider the consensus statement an important step forward, experience and common sense will continue to play a central role in AP critical values policy. One of the workgroup’s tasks was an extensive literature review that graded various articles based on rather stringent criteria, Dr. DeYoung says. “What we came up with was that there is not a lot of good evidence-based data out there. There is anecdote and personal experience, not a lot of good data. Even our data is not good data by strict definition.”

“I don’t think you can entirely get away from experience, nor should you. Surgical pathologists are all trained MD physicians, and if a voice in the back of your head tells you this is an important finding, then you’d better get ahold of the clinician and let someone know.” For that reason, the consensus statement emphasizes the importance of pathologists’ professional judgment. “We can set some parameters and agree on a list, but if there’s something that falls off the list, you have to get the surgical pathologist to exercise their ‘MD’ and not just their slide-reading persona.”

In developing the consensus statement, the workgroup has tried hard to stay sensitive to practicing pathologists’ needs in addition to quality concerns, Dr. Nakhleh says. The group incorporated an active feedback element and, after a draft version was put on the Internet and public comment solicited, “We actually got more than 400 comments sent back to us, and we used that information to modify our recommendations.”

Some pathologists have expressed concern that the policy recommendations might be burdensome. “We understand, and I think pathologists know, that when you put out a report, the clinician actually has the responsibility to read that report. That’s actually written in law. What we are doing with this, by talking about urgent diagnoses or significant unexpected diagnoses, is trying to make sure that report doesn’t get overlooked. We’re going the extra mile to make sure the patient is safe,” Dr. Nakhleh says.

One area that will continue to be difficult is audits or quality monitors to make sure institutions are following the policies they set. “The folks on the committee recognize that,” Dr. De Young says. “A lot depends on your setup and how easy it is to get information out of your LIS and/or your EMR. LISs vary significantly across the board with regard to the ease of getting at information. But we believe you should make some good faith effort to audit to make sure the policy is being followed. Regulators and inspectors are going to be looking to make sure that loop is closed. Because it’s one thing to have a policy; it’s another thing to have an active policy that’s actually implemented and is standard operating procedure.”

Since his institution has a loose written set of guidelines now, Dr. DeYoung says he and his co-director of AP will use the consensus statement to strengthen and flesh out the policy. “No policy is ever going to cover every contingency,” he says, “but we think the recommendations we’ve put down give folks a good starting point for discussions at their institution.”

The consensus statement is an important milestone in helping institutions agree upon guidelines for handling critical AP diagnoses, Dr. Silverman agrees, but the recommendations will continue to be improved. “They have to be tailored to each institution, and I think the terminology and expectations of time frames need to be fine-tuned. So it’s a work in progress—but it’s a work that’s heading in the right direction.”

Related Links

• The statement in brief