In this setting, the individual pathologist or the pathol-
ogy group in a department may seek an extramural opinion
from an expert consultant. This typically occurs when there
is a difference of interpretation among pathologists in an
institution or the diagnosis is uncertain after internal re-
view. Consultation may also be obtained when the probable
diagnosis is one with which there is little or no experience.
Another category of consultation results from uncertainty
about the diagnosis engendered by a limited or unrepre-
sentative sample, poor histologic preparation, or a patho-
logic change that appears to be on the borderline between
two or more diagnoses. As noted by Leslie et al.,
6
“Second
opinions in anatomic pathology are an integral part of qual-
ity practice … frequent consultation between pathologists
should be fostered in all practice settings and documented
as part of the quality assurance process.”
Several studies have demonstrated the important con-
tribution to patient care of second opinion pathology
consultations, generally in the context of referrals seen at
academic centers. A very encouraging aspect of this practice
is the high degree to which the primary diagnosis has been
confirmed by the consultant. Epstein et al.
7
reported con-
cordant diagnoses (carcinoma vs. not carcinoma) in 98.7%
of 535 prostatic needle biopsies diagnosed as carcinoma.
Nonetheless, the six diagnoses not sustained as carcinoma
were critically important for the 1.3% of patients. A cost
analysis of these results suggested that the saving in medical
expenses for the six patients who did not undergo surgery
substantially exceeded the cost of reviewing all 535 biopsies.
In a subsequent study of 855 core biopsy samples from the
prostate gland seen in consultation, Epstein and colleagues
8
reported a 1.2% rate of unconfirmed carcinomas, a result
that was virtually identical to their 1996 study. Among 844
cases confirmed to be carcinoma, unreported perineural
invasion was detected in 4.3% and unreported periprostatic
invasion was found in 0.5%.
A higher rate of discrepancies was found by Abt et al.,
9
who compared the original- and second-opinion diagnoses
in a broad range of pathology among 777 patients referred
to an academic center. Forty-five diagnostic disagreements
(6%) were regarded as clinically significant, and overall the
level of agreement was 92.1%. Manion et al.
10
reported a
study of 5,629 outside pathology cases examined between
2003 and 2006 as part of the University of Iowa Hospitals
and Clinics policy that requires “… second opinion pathol-
ogy review of pertinent outside material, irrespective of the
nature of the specimen or complexity of the case.” Major
diagnostic disagreements with the potential to change treat-
ment or prognosis were recorded in 132 (2.3%) cases, result-
ing in changes in clinical management in 68 (1.2%). The
most frequent sites of major disagreements were the female
reproductive tract, the gastrointestinal tract, and the skin.
The largest study to date of discordant pathology was re-
ported by Swapp et al.,
11
who reviewed the records of 71,811
cases seen in consultation at the Mayo Clinic between 2005
and 2010. Major disagreements were recorded in 457 (0.6%)
cases. The most frequent sites of discrepant diagnoses were
the gastrointestinal tract (17.5%), lymph nodes (16%), and
bone/soft tissue (10%). Major disagreements were encoun-
tered in 8% of breast cases.
Perkins et al.
12
estimated that diagnoses were inaccurate
in 2% to 4% of breast carcinoma cases, including mistaking
benign for malignant disease or
vice versa
, over- or underdi-
agnoses of invasive carcinoma, or misinterpretation of prog-
nostic markers such as human epidermal growth factor/
neu
receptor (HER2/
neu
). In a study restricted to breast carci-
nomas, Staradub et al.
13
reviewed second-opinion diagnoses
on 346 tumors from 340 patients who had been referred to
the Sage Comprehensive Breast Program at Northwestern
University. Major changes in diagnosis that affected therapy
occurred in 30 (7.8%) cases. Among seven discrepant cases
with an initial diagnosis of ductal carcinoma
in situ
(DCIS),
the second-opinion diagnosis was benign in one and inva-
sive carcinoma in six. Seven other diagnoses were revised
from invasive carcinoma to DCIS. Sixteen changes of mar-
gin status were documented and in three cases revised mar-
gin status coincided with another major change.
Within the United States, several factors have contributed
to the growing number of pathology consultations. Much
of the increase is generated by patients who seek multiple
clinical opinions from different physicians and institutions.
Some patients are primarily concerned with confirmation
of their diagnosis, and one or more consultations may be
obtained directly from pathologists for this reason alone.
The involvement of patients is epitomized by a January 17,
2012,
Wall Street Journal
article titled “What If the Doctor
Is Wrong?,”
14
which recounts the story of a 47-year-old
woman with abdominal tumors. Based on initial tissue
samples, it was thought that she had a rare form of ovarian
carcinoma. When the patient consulted a major cancer cen-
ter, further studies led to a diagnosis of lymphoma.
In addition to consultations initiated by pathologists
seeking opinions from their colleagues, surgeons, medical
oncologists, and other physicians generate some consulta-
tions. The review of “outside” pathology slides should be
mandatory whenever a patient is referred to a physician for
consultation or treatment at an institution other than the
one where the primary diagnosis was rendered,
15
a policy
referred to by one author as “the pathologist’s preventive
medicine.”
16
The office of the physician seeing a patient in
consultation should inform the patient of the necessity of
obtaining pathology material for review in a timely manner
before the office visit. A policy and procedures should be
established for guiding the patient through this process, in-
cluding instructions as to what material is needed and where
it should be sent. The importance of a second review should
be explained, and the patient should be informed that there
will be a charge for this service.
Slides sent for consultation, regardless of the reason,
must be accompanied by documents that confirm the
identity of the specimen with the patient and a copy of the
pathology/cytology report for each specimen represented,
xiv
Introduction
