REVIEW ARTICLE

The Paris System for Reporting Urinary Cytology:

The Quest to Develop a Standardized Terminology
Güliz A. Barkan, MD, FIAC,* Eva M. Wojcik, MD, MIAC,*
Ritu Nayar, MD, MIAC,w Spasenija Savic-Prince, MD,z
Marcus L. Quek, MD, FACS,y Daniel F.I. Kurtycz, MD,8
and Dorothy L. Rosenthal, MD, FIACz

Abstract: The main purpose of urine cytology is to detect high-

grade urothelial carcinoma. With this principle in mind, The Paris
M ore than 5 decades ago, Dr George Papanicolaou
hypothesized that microscopic evaluation of exfo-
liated cells in the urine was a potentially useful method to
System (TPS) Working Group, composed of cytopathologists, detect urinary tract malignancies. Since then, urinary tract
surgical pathologists, and urologists, has proposed and published a
standardized reporting system that includes specific diagnostic
cytology has been plagued by less than a stellar literature
categories and cytomorphologic criteria for the reliable diagnosis of that showed problems with sensitivity, accuracy, and
high-grade urothelial carcinoma. This paper outlines the essential reproducibility. Particularly troublesome is the low sensi-
elements of TPS and the process that led to the formation and tivity in detecting low-grade noninvasive lesions,1 as well as
rationale of the reporting system. TPS Working Group, organized the lack of standardized diagnostic criteria and wide
at the 2013 International Congress of Cytology, conceived a interobserver variability.
standardized platform on which to base cytologic interpretation of Urine cytology samples comprise a variable, but sig-
urine samples. The widespread dissemination of this approach to nificant percentage of daily nongynecologic case volume in
cytologic examination and reporting of urologic samples and the any cytopathology practice, and continue to be one of the
scheme’s universal acceptance by pathologists and urologists is
critical for its success. For urologists, understanding the diagnostic
more difficult specimens that pathologists encounter.
criteria, their clinical implications, and limitations of TPS is Problems include inadequate cellularity of samples, cellular
essential if they are to utilize urine cytology and noninvasive degeneration before fixation, as well as unrealistic expect-
ancillary tests in a thoughtful and practical manner. This is the first ations for diagnosing low-grade urothelial neoplasms
international/inclusive attempt at standardizing urinary cytology. (LGUN) by cytology. LGUNs are the most prevalent
The success of TPS will depend on the pathology and urology neoplasms that urologists encounter and are for the most
communities working collectively to improve this seminal para- part, readily visualized by cystoscopy. In addition, a
digm shift, and optimize the impact on patient care. standardized/comprehensive reporting system for urinary
Key Words: The Paris System, urine, standardized reporting cytology has been missing that is based on the current
terminology, bladder cancer understanding of the pathogenesis of urothelial carcinoma
(UC), and the clinical significance of various types of uri-
(Adv Anat Pathol 2016;23:193–201) nary tract neoplastic lesions. Over 10 years ago, there was
an attempt to create such reporting guidelines.2 However,
the lack of widespread input of the cytopathology com-
munity most certainly explains why it has never been gen-
erally implemented. In recognition of the need to correct
From the Departments of *Pathology; yUrology, Loyola University this situation, an international panel of cytopathologists
Healthcare Systems, Maywood; wDepartment of Pathology, and an urologist with interest in urinary tract cytology
Northwestern Memorial Hospital, Chicago, IL; zInstitute of convened in Paris in May of 2013 at the 18th International
Pathology, University Hospital Basel, Basel, Switzerland; Congress of Cytology organized by the International
8Department of Pathology and Laboratory Medicine, University of
Wisconsin School of Medicine and Public Health, Wisconsin State Academy of Cytology. The goal was to discuss ways to
Laboratory of Hygiene, Madison, WI; and zDepartment of Path- improve the reporting and performance of urinary cytol-
ology, The Johns Hopkins University, Baltimore, MD. ogy. The value of ancillary tests in the screening and
Because of the nature of this paper an IRB approval is waived. diagnosis of urinary neoplasms was also included for con-
The Paris System for reporting urinary cytology was presented for the
first time at the Annual USCAP meeting in San Diego in March sideration. The original group that met in Paris included
2014. cytopathologists (Drs Dorothy L. Rosenthal, Eva M.
The authors have no funding or conflicts of interest to disclose. Wojcik, Güliz A. Barkan, Lukas Bubendorf, Rana S.
Reprints: Güliz A. Barkan, MD, FIAC, Department of Pathology, Hoda, Ritu Nayar, Stefan E. Pambuccian, Eric Piaton,
Loyola University Healthcare System, 2160 S. First Ave., Building
110, Room 2238, Maywood, IL 60153 (e-mail: gbarkan@ Momin T. Siddiqui, Margareta Strojan-Fležar, and Phil-
lumc.edu). ippe Vielh) and a urologist (Dr Marcus L. Quek).
All figures can be viewed online in color at http://www.anatomic
pathology.com.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
“The Paris System for Reporting Urinary Cytology: The Quest to PATHOGENETIC BASES OF THE PARIS SYSTEM
Develop a Standardized Terminology” is jointly published by Advances FOR REPORTING URINARY CYTOLOGY
in Anatomic Pathology, Acta Cytologica, and The Journal of the
American Society of Cytopathology on behalf of Güliz A. Barkan, Eva
According to current scientific data, UC is divided into
M. Wojcik, Ritu Nayar, Spasenija Savic-Prince, Marcus L. Quek, 2 major groups, low grade and high grade, based on 2
Daniel F.I. Kurtycz, and Dorothy L. Rosenthal. separate pathogenetic pathways and biological behavior.3–5

Adv Anat Pathol  Volume 23, Number 4, July 2016 www.anatomicpathology.com | 193
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Barkan et al Adv Anat Pathol  Volume 23, Number 4, July 2016

Approximately 70% of bladder UCs are nonmuscle inva- criteria are considered, standardizes information between
sive (TA/T1) papillary tumors that are usually morpho- institutions, and provides essential therapeutic and prog-
logically categorized as low-grade urothelial carcinoma nostic details. Litigation experience stresses that medical
(LGUC). They have a good prognosis, but may be asso- malpractice claims can be won or lost based on the quality
ciated with recurrence and “progression” to high-grade and content of the medical record10 and patient manage-
urothelial carcinoma (HGUC) in approximately 10% to ment based on the pathologic/cytologic report.
15% cases. The remaining 30% are muscle-invasive (ZT2) In the United States, widespread implementation of
tumors, which are histologically categorized as high grade Electronic Health Records is central to federal government
and are associated with worse overall survival than LGUC. goals for improving health care quality, safety, and efficiency.
The most common molecular alteration in low-grade non- The need for a common diagnostic terminology is clearly
invasive tumors is an activating mutation of FGFR3 expressed by the National Committee on Vital and Health
(fibroblast growth factor receptor 3). This mutation is Statistics. “If information in multiple locations is to be
associated with overall favorable disease characteristics.6 In searched, shared, and synthesized when needed, we will need
contrast, muscle-invasive tumors show a wide range of y common vocabularies for personal, clinical, and public
genomic alterations, with the most commonly seen deletion health information.”11 The standardization of the pathology
or mutation of p53 occurring in about 70% of those reporting language is a key element to fulfill this mandate.12,13
tumors. There is a significant body of literature that com- The Bethesda System (TBS) for Reporting Cervical
bines gene expression analysis, whole-genome array, Cytology terminology, initiated in 1988,14 led the way for
Comparative Genomic Hybridization analysis, and muta- standardized reporting in cytopathology. The goals of TBS
tional analysis of FGFR3, PIK3CA, KRAS, NRAS, TP53, terminology were to (1) communicate clinically relevant
CDKN2A, and TSC1 with resultant identification of 2 information from the laboratory to the health care pro-
separate neoplastic pathways with 2 intrinsic molecular vider; (2) be uniform and reasonably reproducible across
signatures.4 This genetic evidence has lead to the provoca- different pathologists and laboratories, and with enough
tive question of whether these are 2 separate diseases: one, flexibility to be adopted in a wide variety of laboratory
LGUC, associated with an overall good prognosis, and the settings and geographic locations; and (3) reflect the most
other, HGUC, associated with a mortality rate of approx- current understanding of cervical neoplasia. TBS also
imately 60%. Therefore, the conclusion of the first meeting addressed specimen adequacy, correlated morphology with
of The Paris System (TPS) Working Group was that the biology of disease process, “lumped” biologically equiv-
new reporting system would concentrate primarily on the alent entities, and recognized the reality and poor repro-
detection of HGUC while minimizing the detection of ducibility of “atypia.” TBS has seen successful, realizing
LGUC, as cytology has a high sensitivity of detecting the widespread international implementation leading to the
former with a poor sensitivity for the latter. This new desired standardized terminology, management guide-
paradigm became the guiding principle of The Paris System lines,15–18 and to funding of research.19 It has become a
for Reporting Urinary Cytology. model for subsequent development of standardized cyto-
pathology and histopathology reporting consensus
efforts20,21 in other body sites.
STANDARDIZATION OF THE REPORTING In 2009, Crothers et al22 described major elements of
SYSTEM quality nongynecologic cytology reporting and encouraged
Anatomic pathologists serve as consultants to their the use of standardization. In urinary cytology, despite 2
clinical colleagues and patients, and pathology reports well established genetic pathways for the development of
officially document this communication. To help clinicians bladder cancer, and prognostic implication for LGUC and
choose the optimal management options for the patient, HGUC, the morphologic terminology for urinary cytology
reports must accurately and clearly communicate the remained disparate and complex.
cytopathologic findings and outcome probability. To be adopted and widely accepted by the pathology
Pathologists actively use the terms “suspicious,” community, reporting terminology needs to be based on
“indeterminate” or “atypical,” all too often with resultant evidence and consensus. It should be applicable to differ-
failure to provide a clear diagnostic and therapeutic path ent practice settings; be practical, flexible, and concise;
for clinicians. A survey of pathologists and clinicians, per- avoiding redundancy. With this in mind TPS Working
formed by Redman et al,7 documented the need for a more group convened to form a reporting system that would
standardized terminology for reporting cytopathology allow for evolution/change in our understanding of the
results [thyroid fine-needle aspiration (FNA)] and for the disease processes, would correlate patient management
education of clinicians on that terminology. Although with optimal clinical outcomes, and would be understood
pathologists have paid attention to all elements of the and accepted by the health care team taking care of the
pathology reports (tumor staging summaries, etc.8), they patient.
have not focused on the issue of report comprehension. In a
study looking at surgical pathology reports, surgeons mis-
understood pathologists’ reports 30% of the time.9 One of THE UROLOGIST’S PERSPECTIVE
the issues shared by patients and their advocates on web- Urologists depend on cytology to supplement the
sites dedicated to cancer advocacy is that different pathol- routine radiographic and endoscopic evaluation of the
ogists and/or different institutions use different highly urinary tract to ensure that a potentially life-threatening
technical terms to describe the same entities, predictably urothelial malignancy is reliably detected. Although it may
confusing to both patients and their clinicians. seem contradictory to see a “negative” urine cytology
From a legal perspective, pathologists are advised to report in the face of a well-defined papillary bladder tumor
issue synoptic reports. Such reporting makes the pathology on direct cystoscopic visualization, this simply reflects the
report clinically relevant, assures that important diagnostic fact that the majority of bladder cancers are of low-grade

194 | www.anatomicpathology.com Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Adv Anat Pathol  Volume 23, Number 4, July 2016 Paris Classification for Urinary Cytology

cytomorphology and noninvasive. Most urologists under- interpreting the reports. To improve the clinical utility of
stand the inherent limitations of cytology in diagnosing low urine cytology, it is important for both urologists and
grade and noninvasive lesions due to their cellular cohe- cytopathologists to communicate effectively with each
siveness and lack of nuclear atypia/dysplasia. These tumors other. The clinical history (symptoms, prior treatments)
have a low risk of progression. Alternatively, there is little and cystoscopic findings should be readily available to the
controversy when it comes to the ability of cytology to cytopathologist to optimize the usefulness of the cytology
detect HGUC or carcinoma in situ. These lesions clearly report.
have a potential for recurrence, invasion, metastases, and
morbidity/mortality; therefore, patients with high-grade DIAGNOSTIC CATEGORIES AND MORPHOLOGIC
cytomorphology represent the high-risk population most CRITERIA OF THE PARIS SYSTEM
likely to benefit from surveillance evaluation with non-
A universally accepted and utilized system for
invasive urine cytology.
reporting urinary tract cytopathology does not exist. This
Given the wide differential diagnosis for hematuria
was eloquently demonstrated and documented by Glatz
(both gross and microscopic), the cost-effectiveness of
et al30 via an international telecytology quiz on urinary
voided urine cytology as an initial diagnostic study has been
cytology where the participants failed to agree even on the
questioned.1 Most often, hematuria is not a symptom of
proposed categories. The goal of TPS is not only to define
neoplasia.23 However, in the appropriate clinical setting,
morphologic criteria for the various categories in urinary
urine cytology may play an important adjunctive role,
tract cytopathology, but also to standardize the reporting
because the test is relatively cheap and collection methods
system to be universally acceptable and globally utilized.
are either minimally invasive or noninvasive. The initial
The published diagnostic categories are shown in Table 1,
evaluation for patients at higher risk for bladder cancer
and Figure 1 shows the algorithmic approach to the TPS.
(older age, male, smoking history, occupational exposures)
and those with unexplained irritative urinary symptoms Adequacy
(potentially due to carcinoma in situ) should include urine Unlike surgical pathology, adequacy of the cytopa-
cytology. Several groups also advocate the use of cytology thology specimen is an integral part of the report. For some
in the initial diagnosis and surveillance for HGUC.24–26 specimen types, adequacy has been clearly defined, that is,
This can be performed at the time of cystoscopy during for cervicovaginal cytology,31,32 and FNA specimens of the
which a bladder washing/barbotage may be obtained, thus thyroid33–35; in others, adequacy criteria have been pro-
increasing the cellular yield available for cytologic inter- posed (pancreaticobiliary system cytology,36 endobronchial
pretation. Even for patients who have undergone radical ultrasound guided/endoscopic ultrasound-guided FNAs of
cystectomy with urinary diversion, urine cytology repre- mediastinal and hilar lymph nodes37–39) but are not yet
sents an important means to survey the remnant extra- defined or tested; in most other specimen types there are no
vesical urothelial sites (upper tracts, urethra). well-defined, universally accepted adequacy criteria. Ade-
Although cystoscopy is considered the “gold stand- quacy, in general, ensures that the specimen is representa-
ard” diagnostic technique for detection of bladder cancer, it tive of what is sampled. It is defined according to the type
is by no means perfect. Diagnostic accuracy depends on the of specimen, which may be truly exfoliated specimens
experience of the urologist, the cytopathologist, and the (cerebrospinal fluid, voided urine, serosal cavity fluids), or
clinical suspicion. Knowledge of the results of a urinary forced exfoliative cellular samples, for example, Pap test,
marker has been shown to influence how subtle urothelial bladder washing, or FNA specimens. If the sample contains
abnormalities may be viewed.27 The decision to perform a abnormal cells, no matter how few, the specimen is con-
biopsy of an equivocal lesion is justified if the cytologic sidered “adequate for diagnosis.” Otherwise, the definition
diagnosis is suspicious for high-grade urothelial carcinoma of adequacy is based on the quantification or at least a
(SHGUC) or HGUC. A negative urine cytology coupled semiquantification of the number of cells and/or the volume
with a normal cystoscopy is quite specific and reassuring of voided urine. The adequacy of instrumented urinary
that a potentially lethal high-grade malignancy is most tract specimens was recently addressed by an evidence-
likely absent.28 A diagnosis of a “positive” or “suspicious” based study that prospectively and retrospectively eval-
urine cytology should be thoroughly investigated and fol- uated the cellularity of bladder washing specimens. The
lowed closely, regardless of the cystoscopic findings.29 The results supported the conclusion that 2600 cells or 2 well
conundrum rests with the “atypical” diagnostic category. visualized urothelial cells per high-power field in 10 con-
Some have advocated the use of adjunctive techniques, such secutive high-power fields may serve as an objective meas-
as fluorescence in situ hybridization (FISH) testing, to ure of adequacy in instrumented urine specimens processed
further characterize this cohort and move interpretation using the ThinPrep method.40 Table 2 shows guidelines for
into either a non-neoplastic or neoplastic category. Most
critical is an understanding by the clinician of what the
cytopathologist considers “atypical” and how that relates
to the suspicion for and probability of an underlying TABLE 1. Diagnostic Categories for The Paris System for
malignancy. The smaller the laboratory’s frequency of Reporting Urinary Cytology
“atypical” interpretations, the more meaningful that cat- 1. Nondiagnostic/unsatisfactory
egory is to the clinician. Clearly, there are limitations to 2. Negative for high-grade urothelial carcinoma (NHGUC)
urine cytology. Microscopic morphology is not a perfect 3. Atypical urothelial cells (AUC)
reflection of biological behavior. This may be due to dis- 4. Suspicious for high-grade urothelial carcinoma (SHGUC)
ease-related factors (poor sensitivity for low-grade non- 5. High-grade urothelial carcinoma (HGUC)
6. Low-grade urothelial neoplasm (LGUN)
invasive tumors), the method of sampling (voided vs.
7. Other: primary and secondary malignancies and miscellaneous
instrumented), and the experience of the cytopathologist. lesions
Urologists should understand these limitations when

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. www.anatomicpathology.com | 195
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Barkan et al Adv Anat Pathol  Volume 23, Number 4, July 2016

Similarly, changes associated with urolithiasis, treatment-

related changes, and polyomavirus (BK virus) cytopathic
changes should all be classified as negative for high-grade
urothelial carcinoma (NHGUC).44
Figures 2A and B depict benign urothelial cells clas-
sified under the NHGUC category.

Atypical Urothelial Cells

A major goal of TPS was to clarify the ill-defined
category of “atypia” as much as possible, and minimize the
reporting rate of this category. To date, pathologists have
not agreed upon the general definition of atypia in urinary
tract specimens. Some have defined atypia as “cells that are
reminiscent of, but not diagnostic of, HGUC.” Others
define it as “clusters of urothelial cells, suspicious for
LGUC,” and yet others believe degenerated urothelial cells
should be reported as atypical. As a result, there is a wide
FIGURE 1. Algorithmic approach to diagnosis of urinary cytology interobserver and intraobserver variability, which is the
in The Paris System. HGUC indicates high-grade urothelial car- reason why the rates of atypia vary among institutions from
cinoma; LGUN, low-grade urothelial neoplasms. 1.9% to 23.2%.45,46 In a small survey sent to a voluntary
group of US laboratories, the reported percentages of their
atypia categories range from 0.8% to 22% (mean, 12.9%).
estimating cellularity in instrumented urinary tract speci- A similar survey sent to 20 international groups including
mens. Another study evaluated the volume of voided urine, France, Canada, and Japan showed similar results of atypia
concluding that specimens >30 mL are more likely to be ranging from 1.8% to 23.7% (mean, 13.75%).
cellular/satisfactory.42,43 A review of the literature47–49 and surveys sent out to
Regardless of the specimen type (voided urine or TPS groups responsible for the AUC and SHGUC chapters
instrumented), if the urothelial cells are completely concurred on the 4 cytomormorphologic features in pre-
obscured by lubricant, or inflammatory cells, this represents dicting HGUC: nuclear cytoplasmic ratio, hyperchromasia,
an “unsatisfactory” specimen. Conversely, if there are any irregular nuclear membrane, and coarse chromatin. The
atypical cells regardless of the overall cellularity this rep- criteria for the categories were set using these cytomor-
resents a satisfactory specimen. phologic features (Fig. 1, Table 3).
Therefore, the criteria for diagnosing atypical uro-
thelial cells include 1 major and 1 minor criterion. The
Negative for High-grade Urothelial Carcinoma major or required criterion is the presence of non-
The majority of urinary tract specimens fall in this superficial and nondegenerated urothelial cells with an
category. The most common cellular element is benign increased nuclear cytoplasmic (N/C) ratio (> 0.5). The
superficial urothelial cells, followed by intermediate and minor criteria, of which only 1 is required, include: (1)
basal urothelial cells that are more commonly observed in mild nuclear hyperchromasia, (2) irregular nuclear mem-
instrumented specimens. Superficial squamous cells from branes (chromatinic rim or nuclear contour), and (3)
the female genital tract often out-number urothelial cells. irregular, coarse, clumped chromatin. Figure 3 depicts a
Benign glandular cells (from cystitis glandularis), squamous bladder washing specimen with cytologic atypia, hence
cells originating in squamous metaplasia of urothelium or classified under AUC.
external genital tract skin, and rarely benign seminal vesicle In TBS for Reporting Gynecologic Cytology, the cat-
cells also fall into this category. Groups or fragments of egory “atypical squamous cells” typically raises the possi-
urothelial cells that may be seen in both instrumented and bility of a low-grade intraepithelial lesion and “atypical
noninstrumented urine specimens should be classified as squamous cells, a high-grade lesion cannot be excluded”
negative unless the cytomorphology of the cells forming the typically raises the possibility of a high-grade squamous
group fits the criteria outlined under the atypia category. intraepithelial lesion. In TPS in both equivocal categories,

TABLE 2. Guidelines for Estimating Cellularity of Instrumented Urine Specimens

FN20 Eyepiece ¾10 FN20 Eyepiece ¾10 FN20 Eyepiece ¾10 FN20 Eyepiece ¾10
Objective Objective Objective Objective
Prep No. Fields No. Cells/Field No. Fields No. Cells/Field No. Fields No. Cells/Field No. Fields No. Cells/Field
Diameter Area at FN20, for 2644 Cells at FN20, for 2644 Cells at FN20, for 2644 Cells at FN20, for 2644 Cells
(mm) (mm2) ¾10 Total ¾40 Total ¾10 Total ¾40 Total
13 132.7 42.3 62.5 676 3.9 34.9 75.8 559 4.7
20 314.2 100 26.4 1600 1.7 82.6 32 1322 2
Adapted from Solomon and Nayar.41
Adaptations are themselves works protected by copyright. So in order to publish this adaptation, authorization must be obtained both from the owner of
the copyright in the original work and from the owner of copyright in the translation or adaptation.

196 | www.anatomicpathology.com Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Adv Anat Pathol  Volume 23, Number 4, July 2016 Paris Classification for Urinary Cytology

FIGURE 2. Negative for high-grade urothelial carcinoma (bladder washing, ThinPrep, Papanicolaou stain  600). A, A cluster of benign
intermediate and basal type urothelial cells with scant cytoplasm, and normochromatic nuclei. B, A group of reactive superficial
urothelial cells with open chromatin and prominent chromocenters juxtaposed to benign basal type small umbrella cells with scant
cytoplasm. The nuclear contours of all these benign cells are relatively smooth and regular. Please see this image in color online.

AUC and SHGUC, the atypia refers to the probability of cases diagnosed as SHGUC will reveal a higher rate of
HGUC. Of course, the prediction of HGUC is much lower HGUC compared with that of AUC.
in AUC compared with SHGUC. The major or required criteria are the presence of
Usually, management of an AUC diagnosis will have nonsuperficial and nondegenerated urothelial cells with an
routine follow-up akin to the “negative” category. By increased nuclear cytoplasmic (N/C) ratio (> 0.7) and
minimizing the atypia rate we will help guide our urology severe nuclear hyperchromasia. The minor criteria, of
colleagues toward an appropriate management strategy, which only 1 is required, include: (1) irregular nuclear
and reduce patient anxiety related to an indeterminate membranes (chromatinic rim or nuclear contour), (2) very
diagnosis. According to the open ASC web-based forum on dark, irregular, coarse, clumped chromatin. Figure 4
TPS, 97% of the participants agree that there should be a depicts a urine specimen with significant cytologic atypia
diagnostic category of AUC and similarly, 93% of the in a few cells, hence classified under SHGUC.
participants agree that this category should be kept at the
lowest possible rate to maintain clinical significance. High-grade Urothelial Carcinoma
Although urine cytomorphology reporting has evolved
over time from the days of George Papanicolaou and
Suspicious for High-grade Urothelial Carcinoma Leopold Koss, perhaps the 1 concept that has remained
This category includes cases with severe urothelial unchanged is the cytomorphologic characteristics of
atypia, but falls quantitatively short of a definitive HGUC HGUC. HGUC has been well recognized in urinary tract
diagnosis. However, the atypia present is beyond the atypia cytopathology as having the following features: High N/C
defined in the AUC category. Naturally, the follow-up of ratio, nuclear pleomorphism, nuclear membrane

TABLE 3. Comparison of Morphologic Criteria of Abnormal Cells in The Paris System for Reporting Urinary Cytology
N-C Nuclear Chromatinic Mandatory
Ratio Chromasia Rim/Nuclear Chromatin (Major)
Category (1) (2) Membrane (3) Quality (4) Features Minor Features
AUCw > 0.5 Similar to Fine and even Finely 1 2, 3, 4 (one of the features 2-4 noted with “w” must be a
umbrella Or granular second feature identified in the cells of interest in
cells Uneven shape and Or addition to number 1)
Or thicknessw Coarsely
Dark/very clumpedw
darkw
SHGUC* > 0.7 Very dark Uneven shape and Coarsely 1, 2 3, 4 (at least one of the above must be a third feature
thickness clumped identified)
HGUC* > 0.7 Very dark Uneven shape and Coarsely 1, 2 3, 4 (at least one of the above must be a third feature
thickness clumped identified)
*Only difference is the quantity: SHGUC = very few cells, <5 cells; HGUC > 5 to 10 cells.
wOnly 1 minor feature required.
HGUC indicates high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma.

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. www.anatomicpathology.com | 197
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Barkan et al Adv Anat Pathol  Volume 23, Number 4, July 2016

FIGURE 3. Atypical urothelial cells (bladder washing, ThinPrep, FIGURE 4. Suspicious for high-grade urothelial carcinoma(HGUC)
Papanicolaou stain  600). This urine specimen has very rare cells (bladder washing, ThinPrep, Papanicolaou stain 600). This urine
with slightly higher N:C ratio ( > 0.5), in addition to hyper- specimen contains rare cells with high N:C ratio ( > 0.7), irregular
chromasia. This atypical cluster (arrow) cells are enlarged compared nuclear contours, and coarse chromatin. Compared with the
with the neighboring clusters of benign urothelial cells. Degener- benign urothelial cells, the abnormal urothelial cells are hyper-
ative changes make it difficult to further characterize the chromatin chromatic and are all features of HGUC; however, the paucity of
pattern. However, the cytomorphologic changes are sufficient to abnormal urothelial cells (arrows) precludes a definitive diagnosis
classify this case under “AUC.” Please see this image in color online. of HGUC. On follow-up this patient had an invasive HGUC in the
urinary bladder. Please see this image in color online.
irregularity, and severe hyperchromasia.50,51 In addition,
coarse chromatin patterns are well described and illus- category in order to clarify the conspicuous absence of
trated. Other features, such as nuclear and cytoplasmic HGUC. Figure 6 demonstrates LGUN, where the surgical
pleomorphism, eccentrically located nuclei, dense cyto- follow-up was noninvasive LGUC. In TPS, LGUN also
plasm, presence of mitotic figures, and apoptotic bodies are serves as a placeholder, awaiting further understanding of the
also seen in these cases. Prominent nucleoli, isolated molecular biology of the lesion.
malignant cells with enlarged nuclear size and extensive
necrosis have been described as features of HGUC in urine
cytology specimens, with necrosis increasing the possibility
for invasive disease.52 According to TPS, the necessary Other Malignancies: Primary, Metastatic, and
morphologic features to diagnose HGUC include: a mini- Miscellaneous Lesions
mum of 5 to 10 severely abnormal urothelial cells with an Primary malignancies of the urinary bladder, other than
N/C ratio of Z0.7, with cells showing moderate to severe urothelial origin are rare, and typically represent <5% of
hyperchromasia, coarse chromatin, and markedly irregular bladder tumors. They include squamous cell carcinoma,
nuclear membrane. Figure 5 depicts a classic HGUC.

Low-grade Urothelial Neoplasm

Although the main goal of TPS is to detect a HGUC,
low-grade urothelial lesions cannot be discounted. Previous
studies list a number of morphologic features that enabled
the diagnosis of LGUC, such as minimal nuclear enlarge-
ment, nuclear membrane irregularity, density of cytoplasm,
and elongated nuclei.53–56 However, TPS acknowledges that
in the majority of cases a reliable diagnosis of low-grade
carcinoma cannot be made, even with the morphologic fea-
tures listed above. In a recent study by McCroskey et al,57
most of the features described previously as diagnostic for
LGUC were observed almost equally in cases negative for
LGUC regardless of whether the specimens were from the
upper or lower urinary tract. Presence of fibrovascular cores,
a feature extremely rare in urine specimens, is the only
instance when the diagnosis of low-grade papillary lesion in
instrumented urine can be made with confidence. Fibrovas- FIGURE 5. High-grade urothelial carcinoma (bladder washing,
cular cores can be seen in any low-grade papillary lesion, cytospin, Papanicolaou stain  600). This urine specimen has
numerous cells with high N:C ratio (> 0.7), demonstrating nuclear
including papillomas, papillary urothelial neoplasia of low hyperchromasia, coarse chromatin, and irregular nuclear mem-
malignant potential and LGUC. Therefore, for reporting branes. Most cells have a plasmacytoid appearance with eccen-
purposes, “low-grade urothelial neoplasm (LGUN)” is rec- trically placed nuclei. The background has a significant number
ommended as a diagnostic category. This category is to be of red blood cells, which is commonly seen in cytospin prepa-
used sparingly, and in conjunction with the NHGUC rations. Please see this image in color online.

198 | www.anatomicpathology.com Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Adv Anat Pathol  Volume 23, Number 4, July 2016 Paris Classification for Urinary Cytology

FIGURE 6. Low-grade urothelial neoplasm (LGUN) (renal pelvis FIGURE 8. FISH TEST: (Cytospin, original magnification Papani-
brushing, conventional preparation, Papanicolaou stain  400). This colaou stain,  600). Bladder washing 3 months after intravesical
is a very cellular urine specimen composed of monomorphic cells. BCG-treatment for pT1, high-grade urothelial carcinoma. Atyp-
The most striking morphologic features are the presence of fibro- ical urothelial cells in a background of inflammation support the
vascular cores lined by the monomorphic urothelial cells. This is the diagnosis of AUC. However, morphology cannot distinguish
only sine qua nonmorphologic feature to render a diagnosis of between neoplastic cells and reactive cellular changes. Inset of
LGUN. On follow-up this patient did have a subcentimeter low-grade the same atypical cell group: UroVysion FISH clearly shows a
papillary urothelial carcinoma. Please see this image in color online. positive result with gain of the centromeric signals (blue, red, and
green) and homozygous deletion of 9p21 (no yellow signals).
Encircled is an inflammatory cell that shows a regular disomic
signal pattern and serves as internal negative control (original
adenocarcinoma, and small cell carcinoma. Their cytologic magnification  600). Please see this image in color online.
features are the same as those in other parts of the body.
Secondary malignancies in the bladder occur in <10%
of bladder tumors. Most of these are direct invasion from THE USE OF ANCILLARY DIAGNOSTIC TESTING
prostate, cervix, uterus, or gastrointestinal tract. The most IN URINE CYTOLOGY
common distant metastases are malignant melanoma, car- As mentioned above, the diagnosis of “atypical uro-
cinomas of stomach, breast, kidney, and lung. Figure 7 is thelial cells” is inconclusive for malignancy, and creates a
an example of adenocarcinoma of the prostate involving dilemma for the urologist, especially in patients with neg-
the urinary bladder. ative or equivocal findings on ureterocystoscopy. There
have been many ancillary studies used for urine cytology,
but only a few are currently FDA approved to be used in
the laboratory setting; namely: UroVysion FISH (Abbott
Molecular Inc., Des Plaines, IL), ImmunoCyt (Scimedx,
Denville, NJ), BTA stat (Polymedco, Cortlandt Manor,
NY), and NMP 22 (Allere, Waltham, MA). The FDA
approval for these tests are for voided urine specimens only.
Of these, one of the most commonly used to clarify
inconclusive cytologic findings is the UroVysion FISH test
likely due to its morphologic applicability to the cytopathology
laboratory. This multiprobe FISH test was initially developed
to improve the detection of invasive HGUC in voided urines
and is now FDA approved for initial diagnosis and surveil-
lance of patients with hematuria.58 The reported sensitivity
and specificity of the test for detection of HGUC vary widely
in the literature and have been reported from 8% to 100% and
29% to 100%, respectively.59 This variability in the reported
performance of the test may be due to lack of standardization
of the technical testing procedure and test evaluation. These
vulnerabilities include the definition for UroVysion FISH
positivity, prevalence of disease in the population tested, the
FIGURE 7. Prostatic adenocarcinoma involving urinary bladder specimen type (voided urine vs. instrumented specimens), and
(bladder washing, ThinPrep, Papanicolaou stain  600). This urine the cellularity of the urinary specimen used for FISH testing.
specimen demonstrated mostly clusters of cells with high N:C ratio A cytologic diagnosis of “positive for malignancy” has
and prominent nucleoli. Although high-grade urothelial carcinoma
can show clustering and prominent nucleoli, these features are more
a high specificity and positive predictive value of >90% for
commonly observed in adenocarcinomas, especially of prostatic the diagnosis of HGUC. In this scenario, the ancillary tests
origin. This patient did have a history of prostatic adenocarcinoma, does not add any additional clinical benefit, but only
Gleason score 4 + 4 = 8, and the cell block sections showed PSA- unnecessary cost. The UroVysion FISH test can increase
positive tumor cells. Please see this image in color online. the sensitivity of cytology for the detection of LGUC from

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. www.anatomicpathology.com | 199
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Barkan et al Adv Anat Pathol  Volume 23, Number 4, July 2016

TABLE 4. Relative Risk of the Diagnostic Categories Outlined in the Paris System, Based on Studies to Date
Risk of Malignancy
Category (%) Management
Unsatisfactory/nondiagnostic < 5-10 Repeat cytology, cystoscopy in 3 mo if increased clinical
suspicion
Negative for high-grade urothelial carcinoma 0-10 Clinical follow-up as needed
Atypical urothelial cells 8-35 Clinical follow-up as needed
Potential use of ancillary testing
Suspicious for high-grade urothelial 50-90 More aggressive follow-up, cystoscopy, biopsy
carcinoma
Low-grade urothelial neoplasm B10 Need cystoscopy and biopsy to further evaluate grade and stage
High-grade urothelial carcinoma > 90 More aggressive follow-up, cystoscopy, biopsy, staging
Other malignancy > 90 More aggressive follow-up, cystoscopy, biopsy, staging

25% to 60% to 75%, but usually low-grade neoplasms are 3. Netto GJ. Molecular genetics and genomics progress in
clearly visible by cystoscopy and the FISH result will not urothelial bladder cancer. Semin Diagn Pathol. 2013;30:
impact the clinical management. Conversely, in the setting 313–320.
4. Cheng L, Zhang S, MacLennan GT, et al. Bladder cancer:
of atypia with negative or inconclusive findings on cysto-
translating molecular genetic insights into clinical practice.
scopy, a negative UroVysion FISH test makes it very Hum Pathol. 2011;42:455–481.
unlikely that these abnormal cells derive from a HGUC and 5. Knowles MA. Molecular pathogenesis of bladder cancer. Int J
this additional information will help the urologist in further Clin Oncol. 2008;13:287–297.
management of the patient.60 6. van Rhijn BW. Combining molecular and pathologic data to
In general, the ancillary test might be of potential use prognosticate non-muscle-invasive bladder cancer. Urol Oncol.
for clarifying atypia in urinary cytology (Fig. 8) and may be 2012;30:518–523.
able to assist the urologist in clinical management. How- 7. Redman R, Yoder BJ, Massoll NA. Perceptions of diagnostic
ever, testing must be well standardized, performed in the terminology and cytopathologic reporting of fine-needle
hands of experienced cytomorphologists (if it is a mor- aspiration biopsies of thyroid nodules: a survey of clinicians
and pathologists. Thyroid. 2006;16:1003–1008.
phology-based assay), under consideration of cystoscopy 8. Lankshear S, Srigley J, McGowan T, et al. Standardized
findings, and the patient’s medical history. synoptic cancer pathology reports - so what and who cares? A
population-based satisfaction survey of 970 pathologists,
surgeons, and oncologists. Arch Pathol Lab Med. 2013;137:
CONCLUSIONS 1599–1602.
Important ongoing work by TPS Working Group will 9. Powsner SM, Costa J, Homer RJ. Clinicians are from Mars
provide a standardized platform for reporting cytologic and pathologists are from Venus. Arch Pathol Lab Med.
interpretation of urine samples. The relative risk of the 2000;124:1040–1046.
diagnostic categories outlined in The Paris System, based 10. Troxel DB. The pathology report: reducing malpractice risk.
Available at: http://www.thedoctors.com/KnowledgeCenter/
on studies to date are outlined in this paper (Table 4).
PatientSafety/. Accessed March 10, 2016.
Prospective studies to establish successful prediction of 11. Statistics NCoVaH. Information for Health A Strategy for
HGUC by all categories, and clinical outcomes relative to Building the National Health Information Infrastructure Report
each morphologic category will be essential to the suc- and Recommendations from the National Committee on Vital
cessful acceptance and implementation of TPS. For urolo- and Health Statistics. Washington, DC: US Department of
gists, understanding the diagnostic criteria, their clinical Health and Human Services; 2001.
implications, and appreciating the limitations of TPS is 12. Grapsa D, Ekaterini P. Standardized categorical reporting of
necessary if we are to utilize urine cytology and ancillary cytopathology results: the strengths and weaknesses of a
tests in a thoughtful and practical manner. constantly evolving and expanding system. Diagn Cytopathol.
2013;41:917–921.
13. Ali SZ, Leteurtre E. The official nomenclature and terminol-
ACKNOWLEDGMENTS ogies in diagnostic cytopathology: history, evolution, applic-
The authors would like to thank all who were involved in ability and future. Ann Pathol. 2012;32:e3–e7. 389-393.
creating the Paris System of Reporting Urinary Cytology. 14. The 1988 Bethesda System for reporting cervical/vaginal
This was truly a team effort. cytological diagnoses. National Cancer Institute Workshop.
JAMA. 1989;262:931–934.
15. Davey DD, Austin RM, Birdsong G, et al. ASCCP patient
REFERENCES management guidelines: Pap test specimen adequacy and
1. Blick CG, Nazir SA, Mallett S, et al. Evaluation of diagnostic quality indicators. Am J Clin Pathol. 2002;118:714–718.
strategies for bladder cancer using computed tomography (CT) 16. Wright TC Jr, Cox JT, Massad LS, et al. 2001 consensus
urography, flexible cystoscopy and voided urine cytology: guidelines for the management of women with cervical intra-
results for 778 patients from a hospital haematuria clinic. BJU epithelial neoplasia. Am J Obstet Gynecol. 2003;189:295–304.
Int. 2012;110:84–94. 17. Stoler MH, Castle PE, Solomon D, et al. American Society for
2. Layfield LJ, Elsheikh TM, Fili A, et al. Papanicolaou Society Colposcopy and Cervical Pathology. The expanded use of
of C. Review of the state of the art and recommendations of HPV testing in gynecologic practice per ASCCP-guided
the Papanicolaou Society of Cytopathology for urinary management requires the use of well-validated assays. Am J
cytology procedures and reporting: the Papanicolaou Society Clin Pathol. 2007;127:335–337.
of Cytopathology Practice Guidelines Task Force. Diagn 18. Massad LS, Einstein MH, Huh WK, et al. 2012 updated
Cytopathol. 2004;30:24–30. consensus guidelines for the management of abnormal cervical

200 | www.anatomicpathology.com Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.
Adv Anat Pathol  Volume 23, Number 4, July 2016 Paris Classification for Urinary Cytology

cancer screening tests and cancer precursors. Obstet Gynecol. 39. Cameron SE, Andrade RS, Pambuccian SE. Endobronchial
2013;121:829–846. ultrasound-guided transbronchial needle aspiration cytology: a
19. Schiffman M, Solomon D. Findings to date from the ASCUS- state of the art review. Cytopathology. 2010;21:6–26.
LSIL Triage Study (ALTS). Arch Pathol Lab Med. 40. Prather J, Arville B, Chatt G, et al. Evidence-based adequacy
2003;127:946–949. criteria for urinary bladder barbotage cytology. J Am Soc
20. Crippa S, Mazzucchelli L, Cibas ES, et al. The Bethesda Cytopathol. 2014;4:57–62.
System for reporting thyroid fine-needle aspiration specimens. 41. Solomon D, Nayar R. Bethesda System from Reporting
Am J Clin Pathol. 2010;134:343–344. Author reply 345. Cervical Cytology Chapter 1 Chapter 1, 2nd ed. New York:
21. Darragh TM, Colgan TJ, Cox JT, et al. The Lower Anogenital Springer-Verlag; 2004:1–20.
Squamous Terminology Standardization Project for HPV- 42. VandenBussche CJ, Rosenthal DL, Olson MT. Adequacy in
Associated Lesions: background and consensus recommenda- voided urine cytology specimens: The role of volume and a
tions from the College of American Pathologists and the repeat void upon predictive values for high-grade urothelial
American Society for Colposcopy and Cervical Pathology. carcinoma. Cancer Cytopathol. 2016;124:174–180.
Arch Pathol Lab Med. 2012;136:1266–1297. 43. Barkan GA. Enough is enough: adequacy of voided urine
22. Crothers BA, Tench WD, Schwartz MR, et al. Guidelines for cytology. Cancer Cytopathol. 2016;124:163–166.
the reporting of nongynecologic cytopathology specimens. 44. Wojcik EM. What should not be reported as atypia in urine
Arch Pathol Lab Med. 2009;133:1743–1756. cytology. J Am Soc Cytopathol. 2015;4:30–36.
23. Mazhari R, Kimmel PL. Hematuria: an algorithmic approach 45. Renshaw AA. Subclassifying atypical urinary cytology speci-
to finding the cause. Cleve Clin J Med. 2002;69:870. 872-874, mens. Cancer. 2000;90:222–229.
876 passim. 46. Brimo F, Vollmer RT, Case B, et al. Accuracy of urine
24. Hall MC, Chang SS, Dalbagni G, et al. Guideline for cytology and the significance of an atypical category. Am J
the management of nonmuscle invasive bladder cancer Clin Pathol. 2009;132:785–793.
(stages Ta, T1, and Tis): 2007 update. J Urol. 2007;178: 47. Piaton E, Advenier AS, Benaim G, et al. Atypical urothelial
2314–2330. cells (AUC): A Bethesda-derived wording applicable to urinary
25. Babjuk M, Burger M, Zigeuner R, et al. EAU guidelines on cytopathology. Ann Pathol. 2011;31:11–17.
non-muscle-invasive urothelial carcinoma of the bladder: 48. Rosenthal DL, Vandenbussche CJ, Burroughs FH, et al. The
update 2013. Eur Urol. 2013;64:639–653. Johns Hopkins Hospital template for urologic cytology
26. Clark PE, Agarwal N, Biagioli MC, et al. Bladder cancer. samples: part I-creating the template. Cancer Cytopathol. 2013;
J Natl Compr Canc Netw. 2013;11:446–475. 121:15–20.
27. van der Aa MN, Steyerberg EW, Bangma C, et al. Cystoscopy 49. McCroskey Z, Bahar B, Hu Z, et al. Subclassifying atypia in
revisited as the gold standard for detecting bladder cancer urine cytology: what are the helpful features? J Am Soc
recurrence: diagnostic review bias in the randomized, prospec- Cytopathol. 2014;4:183–189.
tive CEFUB trial. J Urol. 2010;183:76–80. 50. Koss LG, Melamed MR, Koss LG. Koss’ Diagnostic Cytology
28. Khan R, Hussain H, Pambuccian S, et al. What is the negative and its Histopathologic Bases, 5th ed. Philadelphia, PA:
predictive value of urinary tract cytology. J Am Soc Lippincott Williams & Wilkins; 2006.
Cytopathol. 2015;4:S21–S22. 51. Papanicolaou GN, Marshall VF. Urine sediment smears as a
29. Aharony S, Baniel J, Yossepowitch O. Clinically unconfirmed diagnostic procedure in cancers of the urinary tract. Science.
positive urinary cytology: diagnostic implications and onco- 1945;101:519–520.
logical outcomes. BJU Int. 2011;108:E179–E183. 52. Reid MD, Osunkoya AO, Siddiqui MT, et al. Accuracy of
30. Glatz K, Willi N, Glatz D, et al. An international telecytologic grading of urothelial carcinoma on urine cytology: an analysis
quiz on urinary cytology reveals educational deficits and of interobserver and intraobserver agreement. Int J Clin Exp
absence of a commonly used classification system. Am J Clin Pathol. 2012;5:882–891.
Pathol. 2006;126:294–301. 53. Raab SS, Slagel DD, Jensen CS, et al. Low-grade transitional
31. Studeman KD, Ioffe OB, Puszkiewicz J, et al. Effect cell carcinoma of the urinary bladder: application of select
of cellularity on the sensitivity of detecting squamous cytologic criteria to improve diagnostic accuracy [corrected].
lesions in liquid-based cervical cytology. Acta Cytol. 2003;47: Mod Pathol. 1996;9:225–232.
605–610. 54. Raab SS, Lenel JC, Cohen MB. Low grade transitional cell
32. Solomon D, Nayar R. The Bethesda System for Reporting carcinoma of the bladder. Cytologic diagnosis by key features as
Cervical Cytology: Definitions, Criteria, and Explanatory identified by logistic regression analysis. Cancer. 1994;74:1621–1626.
Notes, 2nd ed. New York, NY: Springer; 2004. 55. Xin W, Raab SS, Michael CW. Low-grade urothelial
33. Goellner JR, Gharib H, Grant CS, et al. Fine needle aspiration carcinoma: reappraisal of the cytologic criteria on ThinPrep.
cytology of the thyroid, 1980 to 1986. Acta Cytol. 1987;31: Diagn Cytopathol. 2003;29:125–129.
587–590. 56. Murphy WM, Soloway MS, Jukkola AF, et al. Urinary
34. Grant CS, Hay ID, Gough IR, et al. Long-term follow-up of cytology and bladder cancer. The cellular features of transi-
patients with benign thyroid fine-needle aspiration cytologic tional cell neoplasms. Cancer. 1984;53:1555–1565.
diagnoses. Surgery. 1989;106:980–985; discussion 985-986. 57. McCroskey Z, Pambuccian SE, Kleitherms S, et al. Accuracy
35. Cibas ES, Ali SZ. The Bethesda system for reporting thyroid and interobserver variability of the cytologic diagnosis of low-
cytopathology. Am J Clin Pathol. 2009;132:658–665. grade urothelial carcinoma in instrumented urinary tract
36. Pitman MB, Centeno BA, Ali SZ, et al. Standardized cytology specimens. Am J Clin Pathol. 2015;144:902–908.
terminology and nomenclature for pancreatobiliary cytology: 58. Sokolova IA, Halling KC, Jenkins RB, et al. The development
the Papanicolaou Society of Cytopathology guidelines. Diagn of a multitarget, multicolor fluorescence in situ hybridization
Cytopathol. 2014;42:338–350. assay for the detection of urothelial carcinoma in urine. J Mol
37. Alsharif M, Andrade RS, Groth SS, et al. Endobronchial Diagn. 2000;2:116–123.
ultrasound-guided transbronchial fine-needle aspiration: the 59. Dimashkieh H, Wolff DJ, Smith TM, et al. Evaluation of
University of Minnesota experience, with emphasis on useful- urovysion and cytology for bladder cancer detection: a study of
ness, adequacy assessment, and diagnostic difficulties. Am J 1835 paired urine samples with clinical and histologic
Clin Pathol. 2008;130:434–443. correlation. Cancer Cytopathol. 2013;121:591–597.
38. Nayak A, Sugrue C, Koenig S, et al. Endobronchial ultra- 60. Tapia C, Glatz K, Obermann EC, et al. Evaluation of
sound-guided transbronchial needle aspirate (EBUS-TBNA): a chromosomal aberrations in patients with benign conditions
proposal for on-site adequacy criteria. Diagn Cytopathol. and reactive changes in urinary cytology. Cancer Cytopathol.
2012;40:128–137. 2011;119:404–410.

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved. www.anatomicpathology.com | 201
Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.