Word and deed Pap test malpractice case turns on lack of followup
Kenneth M. Gatter, MD, JD
This article explores some of the legal issues that arose for
one pathologist who was caught between a Pap smear interpretation
of low-grade intraepithelial lesion, or LSIL, and high-grade intraepithelial
lesion, or HSIL. She interpreted the Pap smear as LSIL and added
a note recommending colposcopy followup.
This article will examine the approach that one court of appeals took when faced
with this situation. It will also show the importance of expert
witness testimony and the importance of your clinical colleagues.
Two points to keep in mind. First, the standard of care regarding the followup
for LSIL and HSIL has changed since this case took place, and it
is likely to change again. Second, the case teaches principles that
apply to the practice of pathology generally.
First, let us review the differences in clinical management between a Pap test
interpretation of LSIL and HSIL, because, as we will see in the
legal case, clinical followup can be critical.
The recent publication of the large ASCUS-LSIL Triage Study, or ALTS trial, concluded that
LSIL cytology is best managed by colposcopy.1 However, the ALTS
group reached this conclusion only because it could not identify
another “useful triage strategy.”
In 2002 the American Society for Colposcopy and Cervical Pathology, or ASCCP,
published its 2001 consensus guidelines for cervical cytological
abnormalities and noted that previous recommendations for women
with LSIL included repeat cytology.2 The reasoning for
followup with repeat cytology for LSIL is that most women with LSIL
either have no lesion or have CIN I and only a small risk of progression
to invasive cancer. (Followup cytology of LSIL will detect the small
percentage of cases that progress.) Noting the high rate of loss
to followup for repeat cytologies, the 2001 ASCCP consensus guidelines
recommend colposcopy for all women with LSIL, except for postmenopausal
women and adolescent patients, both of whom may receive followup
by repeat cytology. The ASCCP guideline for women with an HSIL interpretation is colposcopy.
In summary, the standard of care for obstetrician-gynecologists at least before
2002 seems to include repeat cytology within a reasonable time,
though many obstetrician-gynecologists might prefer colposcopy.
Let us return to the legal case of the pathologist caught between the diagnostic
conundrum of LSIL and HSIL sometime in the middle 1990s. The names
have been changed. Nothing herein is intended as legal advice or
CAP endorsement of a particular view.
Ms. Kelsy, age 32, went to Dr. Oliver Bass (OB) for her gynecological
examination and routine Pap smear.
The OB sent the Pap smear to Cyto-lab. Dr. Pete, a cytopathologist, interpreted
the Pap as low-grade dysplasia (LSIL) and in a comment recommended
that the clinician perform colposcopy to “rule out a possible
high grade lesion.”
During this time, the director of Cyto-lab was Dr. Somly, who did not see the Pap smear.
Twelve months later Ms. Kelsy went to see Dr. OB for a followup Pap, which was
interpreted as HSIL. The subsequent biopsy showed invasive squamous
cell carcinoma. Ms. Kelsy’s cervical cancer was stage IIB.
She died 20 months later. Dr. OB’s followup instructions to
Ms. Kelsy after the first Pap smear are not clear.
Ms. Kelsy’s estate filed a wrongful death action asserting negligence against
Dr. OB, Cyto-lab, Dr. Pete (the pathologist), Mr. Smith (the cytotechnologist
who screened the first Pap), and Dr. Somly (the laboratory director).
The plaintiff’s theory was that the defendants’ negligence
decreased Ms. Kelsy’s chance of survival, and an expert witness
for the plaintiff testified to the decreased survival from a one-year
delay in diagnosis. More specifically, the complaint against Dr.
Pete, the pathologist and employee agent of Cyto-lab, and against
Mr. Smith, the cytotechnologist and employee of Cyto-lab, alleged
that their failure to detect early signs of malignancy on Ms. Kelsy’s
Pap smear was negligent. The complaint against Dr. OB alleged that
failure to perform colposcopy and cervical biopsies fell below the
standard of care and constituted negligence.
During discovery, the plaintiff asked the court to dismiss the allegations against
the cytotechnologist, Dr. Somly (the director), and Cyto-lab. The
case against the pathologist and the obstetrician-gynecologist went
to the jury, and the jury returned a verdict in favor of the two
doctors. However, the plaintiff appealed, claiming two errors—one
involving the case against the obstetrician-gynecologist and one
involving the case against the pathologist.
Interpretation and recommendation
The issue on appeal involving the pathologist was whether the trial
court made a mistake in requiring the jury to find that the pathologist
was negligent not only in her interpretation but also in the recommendation
for additional colposcopic evaluation. If the mistake was bad enough—if
it “substantially prejudiced the plaintiff”—then
the case would go back for a new trial and the patient/ plaintiff
would get another chance to convince the jury of the pathologist’s negligence.
The jury had, in effect, said the note saved the pathologist. The plaintiff objected
and wanted the jury to consider only the interpretation and nothing
in the note when determining negligence. In support of this approach,
the plaintiff argued that the evidence supported the obstetrician-
gynecologist having performed a biopsy on Ms. Kelsy had the Pap
smear been interpreted as HSIL. In other words, the issue on appeal
was to what extent the determination of negligence can take into
consideration a recommendation mentioned in the note.
How the court answered this question was critical because it was likely to determine
whether the jury held the pathologist liable for medical malpractice.
A jury considering the diagnosis and the recommendation together
is significantly more likely to conclude that the pathologist was not negligent.
How did the court go about answering this question? It stated that this kind
of jury interrogatory, or direct question to the jury, must address
determinative issues and be based on presented evidence. In this
case, the defendant pathologist’s expert testified that Dr.
Pete’s interpretation and recommendation met the standard
of care, even though the expert said she thought there were a few
high-grade cells among the “numerous” low-grade cells.
The expert testified that the interpretation was a matter of degree.
Whereas some pathologists might diagnose high grade, others would
diagnose low grade. She emphasized that what was important was that
Dr. Pete recommended that the clinician follow up with colposcopy.
Dr. Pete’s expert explicitly said that in her opinion Dr.
Pete’s “interpretation and recommendation” ...
“met the standard of care.” (Emphasis added.)
This sounds like a reasonable and accurate description of how a pathologist
diagnoses many Pap smears and some surgical specimens. Often the
pathologic changes are in a continuum and at the border between
two categories. Dr. Pete dealt with this uncertainty by trying to
straddle two categories: diagnosing one and advising followup to
rule out the other. The notion that inherent vagaries exist in a
diagnosis is, however, a difficult concept for many people, including
some clinicians, to grasp. The court was no exception, and it concluded
that the plaintiff’s version of the jury instructions, and
not the defendant’s, was the proper one.
The appellate court pointed to specific language in the defendant’s expert
testimony. The expert had testified that she identified high-grade
cells on the Pap smear during her review. Nevertheless, in response
to a question from the defense attorney about whether Dr. Pete’s
“interpretation and recommendation” met the standard
of care, the expert clearly stated that it did. The expert went
on about how some people might report it as high grade and others
as low grade. She then said “the important thing” was
that Dr. Pete gave a recommendation “in addition to giving
her diagnosis.” The court’s opinion highlighted this
portion of the expert’s testimony and concluded that the diagnosis
and the recommendation were two different things. The court emphasized
that Dr. OB would have proceeded to colposcopy had the diagnosis
been high-grade dysplasia, and the court noted there was evidence
that the low-grade diagnosis was below the standard of care.
How the issue was handled on remand is not part of the public record, but the
court’s determination means that the defense must focus on
whether the diagnosis of low grade was within the standard of care
without the benefit of the note recommendation.
Thoughts for the cytopathologist
The appellate decision starkly illustrates the vagaries of the practice
of pathology. Dr. Pete did not make a glaring mistake and the jury
did not declare her negligent, but the appellate court decision
probably led to a negotiated outcome. The story shows the importance,
and the problematic character, of knowing the clinical implications
of your diagnosis, knowing your clinicians, and communicating effectively.
Presumably, had Dr. OB gone ahead and performed colposcopy, as the
note recommended, he would have done a biopsy. It is likely that
the biopsy would have shown either high-grade dysplasia or invasive
carcinoma and there would not have been a lawsuit. The problem for
the pathologist is that it is often difficult to know the clinical
implications in certain cases because the clinical implication depends
on the particular physician and the overall circumstances.
One way to ensure a better understanding of the clinical implications is to talk to
the clinician about a specific case and ask. For example, Dr. Pete
might have talked to Dr. OB, mentioning that although she diagnosed
LSIL, she was more concerned that there might be HSIL than she usually
was when she made a diagnosis of LSIL. This is, however, an impractical
solution in most cases because such time-consuming telephone conversations
are impractical in busy pathology and gynecology practices.
Another approach is to incorporate a calculation of risk and known clinical implications
of the diagnosis. If Dr. Pete was on the cusp between HSIL and LSIL,
a diagnosis of HSIL would more likely have resulted in a subsequent
biopsy. The biopsy result was likely to be either LSIL or HSIL.
On the other hand, Dr. Pete diagnoses LSIL, and the biopsy result,
if there is one, is either LSIL or HSIL. In either case, there is
a risk of a discrepancy, but in one the clinical followup is more
predictable. Following this calculation, one could argue that when
the cytopathologist is on the cusp between HSIL and LSIL, the preferred
choice is HSIL. The downside to this approach is that it may lead
to repeated biopsies and perhaps cone biopsies to search for the
high-grade lesion. This is expensive and subjects the patient to
added risk. A pathologist using this approach may lose credibility
with clinical colleagues.
A third approach maintains that the diagnosis in this setting is best ASC-US, cannot
rule out high-grade dysplasia (ASC-H).3 In support of
this diagnosis is that this is a good description of what Dr. Pete
encountered. She saw squamous cells and wasn’t sure what they
represented, but she was worried they might be high-grade dysplasia.
Also in support is that clinicians will be more comfortable if no
high-grade lesion is identified. An argument in opposition is that
people historically associate ASC-US with cells that don’t
quite meet the criteria for low grade, and if these cells are at
least low-grade dysplasia, then the cytologist should at least make this diagnosis.
Alternatively, the pathologist could diagnose “LSIL, cannot rule out HSIL,”
though this is not in the current Bethesda System recommendations.
This may have been what Dr. Pete really wanted to say. Placing the
“cannot rule out HSIL” within the interpretive line
would promote effective communication since clinicians would be
less likely to skip over information in a note, as is possible for
busy gynecology practices receiving large numbers of Pap test reports.
There is a risk, however, that pathologists would overuse this diagnosis,
further frustrating clinicians.
Perhaps HPV testing offers some relief. A diagnosis of “LSIL, cannot rule
out HSIL” could be followed by HPV DNA testing, which would
allow the clinician to incorporate the Pap interpretation, the HPV
risk profile, and the clinical setting. Moreover, HPV testing would
provide additional evidence that might help show the pathologist
met the standard of care. The recent ALTS trials, which suggest
that cytology in conjunction with HPV testing is at least as sensitive
as immediate colposcopy, may support this approach. On the other
hand, if recommendations support additional HPV testing only after
colposcopy, the same problem Dr. Pete faced arises. Of course, clinicians
and investigators will have to work out the details.4,5
Finally, some people may advocate including the recommendation in the diagnostic
line instead of within the comment. However, I am not familiar with
any cytopathologists using this approach, and I see two problems.
First, the approach does not guarantee that the court will not draw
the same distinction it did in Dr. Pete’s case between the
diagnosis and the note stating HSIL can’t be ruled out and
recommending colposcopy followup. Plaintiff’s attorneys will
concentrate on separating out the “diagnosis” from the
“recommendation,” and the same issue will arise when
evaluating jury interrogatories. Second, clinicians may not respond
well to seeing a recommendation within the diagnostic line, and
the 2001 Bethesda consensus statement advises that suggestions for
followup be “phrased in the form of a suggestion” and
include a qualifying phrase.
Thoughts for the expert witness
The expert was also caught in a conundrum between testifying that
her opinion was that the Pap was HSIL or LSIL. Her problem was that
she said she thought the Pap was HSIL but had to explain that LSIL,
particularly with the note recommending colposcopy, was within the
standard of care. Recall that the jury found for Dr. Pete: Apparently
the expert did her job convincingly. In retrospect, however, the
expert’s finessing of the HSIL/LSIL conundrum opened the window
for the plaintiff and appellate court. Perhaps an expert who testified
that one cluster of cells suspicious for high grade did not meet
the quantitative criteria for HSIL would have been more beneficial
for the defendant pathologist. The Bethesda classification does
not mention quantitative criteria, and this would probably be an
area of conflicting expert testimony.
From my perspective, the problem with an expert testifying that
the Pap was LSIL is that the jury may not have believed it was LSIL.
After all, the plaintiff died a short time later. Had the jury not
believed the expert it is likely that it would have returned with
a verdict against the pathologist. Instead, the expert’s testimony
comes across as honest and reasonable, an accurate description of
how a cytopathologist works. It supports the credibility of the
pathologist and her story.
Another glaring problem is that this case has great potential for the two defendants,
the pathologist and the obstetrician-gynecologist, to point fingers
at one another. Many pathologists reading this will argue that the
obstetrician-gynecologist was clearly negligent, not only in not
following up with colposcopy as a matter of course for LSIL, but
also for not doing colposcopy when a request is made by the pathologist.
On the other hand, clinicians are often frustrated by pathologists’
interpretations because they are not definitive. Recall in this
case that the plaintiff argued, and had evidence, that the obstetrician-gynecologist
would have been more aggressive had the diagnosis been HSIL. The
result could have easily been that the pathologist’s lawyers
attack the obstetrician-gynecologist for negligent followup and
the obstetrician-gynecologist’s lawyers attack the pathologist
for a negligent Pap interpretation. In a scenario such as this,
both physicians would appear negligent.
In summary, the diagnostic conundrum between LSIL and HSIL that
Dr. Pete faced remains problematic. The legal ramifications today
are probably less than they were at the time of this case because
the current standard of care is clearer about proceeding to colposcopy
after a Pap smear interpretation of LSIL. However, as the recent
ALTS trials suggest, the standard of care for followup of LSIL and
HSIL is constantly evolving.
In Dr. Pete’s case, the jury understood that pathology diagnosis is a spectrum
and requires clinical correlation, but the appellate court did not.
(The latest Bethesda System emphasizes that a Pap is only an interpretation
and not a diagnosis, but this is a distinction that will be lost
on many courts.) In some ways, Dr. Pete made the best diagnosis.
She knew she had at least LSIL but was suspicious that she had something
worse and wanted the clinician to investigate. But at the end of
the day, the appellate court saw a plaintiff, Ms. Kelsy, who died
from cervical cancer that should have somehow been diagnosed earlier.
Faced with the diagnostic dilemma between LSIL and HSIL, Dr. Pete might have been
better off diagnosing either “HSIL,” “ASC-US,
cannot rule out HSIL,” or “LSIL, cannot rule out HSIL”
because of the increased likelihood that clinical followup for any
one of these was colposcopy. On the other hand, Dr. Pete’s
diagnosis should not be thought of as negligent. Remember, no court
ever said it was. As seen before, the critical issues remain the
effective communication between pathologists and clinicians and
the pathologist knowing the clinical implications of the diagnosis.
Many pathologists reading this article will be frustrated by the lack of advice about
how to avoid the problems Dr. Pete faced. I know of no guaranteed
way to avoid the problems. I do think this case raises the question,
Why not include “LSIL, cannot rule out HSIL” within
the interpretive line?—a designation not within current Bethesda
recommendations. Dr. Pete’s case does provide an effective
reminder about the importance of malpractice insurance with good
defense attorneys and good expert witnesses.
- ASCUS-LSIL Triage Study (ALTS) Group. A randomized trial on
the management of low-grade squamous intraepithelial lesion cytology
interpretations. Am J Obstet Gynecol. 2003;188(6): 1393–1400.
- Wright TC, Cox JT, Massad LS, et al. 2001 consensus guidelines
for the management of women with cervical cytological abnormalities.
JAMA. 2002; 287: 2120–2129.
- Solomon D, Davey D, Kurman R, et al. The 2001 Bethesda System:
terminology for reporting results of cervical cytology. JAMA.
2002; 287: 2114–2119.
- ASCUS-LSIL Triage Study (ALTS) Group. Results of a randomized
trial on the management of cytology interpretations of atypical
squamous cells of undetermined significance. Am J Obstet Gynecol.
- Monk BJ, Brewster, WR. Does the ALTS trial apply to the community-based
practitioner? (Editorial). Am J Obstet Gynecol. 2003;188(6):1381–1382.
Dr. Gatter is an assistant professor in the Department of Pathology at Oregon
Health Sciences University, Portland, and adjunct professor of law
at Willamette University College of Law, Salem, Ore. He is a member
of the CAPInsurance Committee.
AUTHOR’S ACKNOWLEDGMENT: Thanks to Anne Rader, MD, and William Rodgers, MD,
PhD, two cytopathology boarded colleagues; William Petty, MD, an
obstetrician-gynecologist; and Edward Fody, MD, a pathologist and
chair of the CAP Insurance Committee, for reviewing the article
and providing helpful insights and suggestions.