A Single Institutional Experience With the Paris System for Reporting Urinary Cytology: Correlation of Cytology and Histology in 194 Cases

A Single Institutional Experience With the Paris System for Reporting Urinary Cytology:... Abstract Objectives The Paris System for Reporting Urinary Cytology (TPS) is designed to standardize the criteria and terminology used in urinary tract cytology reporting. The aim of this study was to evaluate the impact of implementing TPS and to analyze the correlation with follow-up biopsies in order to assess its reproducibility. Methods Urinary tract cytology specimens with follow-up biopsies over a 2-year period were reviewed and reclassified according to TPS criteria. Surgical follow-up diagnoses were correlated with the initial cytology diagnoses and TPS interpretations, and the results were compared. Results Applying TPS in comparison to our previous reporting system resulted in fewer cases in the atypia category (11.8% vs 24.2%) and higher specificity, accuracy, and predictive value. We observed acceptable interobserver agreement in diagnostic categories of this reporting system. Conclusions TPS improves the overall performance of urinary tract cytology by standardizing the criteria and terminology. Urine, Cytology, Atypia, High-grade urothelial carcinoma, Follow-up Urinary tract cytology is currently the most commonly used test for screening and monitoring of urothelial carcinomas. The available information indicates that urinary cytology is an accurate test for detection of high-grade urothelial carcinoma (HGUC), which has the potential for invasion, metastasis, and a more aggressive course. Urinary tract cytology has known limitations in the diagnosis of low-grade urothelial neoplasms (LGUN), but most LGUN form papillary lesions, which are easily identified and biopsied by cystoscopy examination. There has been recent progress in diagnostic and radiological techniques for urothelial lesions, but urinary tract cytology is still a noninvasive, simple, and low-cost test that is highly sensitive for detection of HGUC, especially flat carcinomas that may be cystoscopically occult.1-4 Urine cytology has challenges when dealing with cases that show some degree of cytologic atypia but do not meet the criteria for malignancy. Various conditions such as inflammation, instrumentation, reactive changes due to treatment effects, or urinary tract calculi can result in cytologic atypia and false-positive results.5,6 The reporting rate of “atypia” diagnoses varies significantly between pathologists and cytotechnologists and even between different institutions, which introduces a management challenge for clinicians. Lack of standard diagnostic criteria and widely accepted terminology in urine cytology reporting led to significant variability between reporting systems, and several classification schemes have been proposed to address this issue.6-8 At the 2013 International Congress of Cytology, the Paris System for Reporting Urinary Cytology (TPS) was introduced to standardize urinary tract cytology reporting. The guidelines provide cytomorphologic criteria for categorizing the cytology results. This system emphasizes the main objective of urine cytology, which is identifying HGUC. TPS followed the lead of widely accepted Bethesda Systems for reporting cervical cytology and thyroid cytology and used a similar terminology.9,10 We sought to utilize previously collected urine cytology specimens with histologic follow-up and to reevaluate them based on the cytomorphologic criteria outlined in TPS. Our goal was to access the utility of TPS in comparison to the previously used classification system at our institution, with special emphasis placed on the rate of “atypia” diagnosis. Materials and Methods Consecutive lower urinary tract cytology cases with follow-up histology specimens from January 2011 to December 2012 were reviewed. The specimens included voided urines, instrumented urines, and bladder wash samples. The follow-up histology specimens were obtained within 6 months after the time of urine sample collection. All specimens were Papanicolaou stained, SurePath (Becton Dickinson, Franklin Lakes, NJ) preparations. When the patient had multiple urinary cytologies preceding the biopsy, the urine sample with the most atypical diagnosis was included in the study. Cases with nondiagnostic biopsies, inadequate urine samples, and nonurothelial malignancies were excluded. The patient’s age, gender, original cytology diagnoses, and follow-up surgical pathology results were recorded. Two cytopathology-boarded pathologists with more than 10 years of experience (F.H. and A.S.) independently reviewed all the cases. The reviewing pathologists assigned a diagnosis to each case using strict adherence to TPS criteria Table 1. They were blinded to previous diagnoses, clinical histories, and follow-up surgical biopsies. Cases with discordant diagnoses between the two reviewing pathologists were re-reviewed in a consensus conference to obtain a final diagnosis. Table 1 Diagnostic Categories of the Paris System for Reporting Urinary Cytology Specimens With Cytomorphologic Criteria of Each Category Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  N:C ratio, nuclear to cytoplasmic ratio. View Large Table 1 Diagnostic Categories of the Paris System for Reporting Urinary Cytology Specimens With Cytomorphologic Criteria of Each Category Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  N:C ratio, nuclear to cytoplasmic ratio. View Large TPS describes seven diagnostic categories and defines cytomorphologic criteria of each group. All TPS categories with corresponding criteria are summarized in Table 1.9-11 In our institution, before the application of TPS, a variety of reporting terminologies were used by different pathologists. For the purpose of this study, the original diagnoses were organized in five groups: negative for malignancy, urothelial atypia, low-grade urothelial carcinoma, suspicious for malignancy, and HGUC. The follow-up histopathologic samples included biopsy and resection specimens. The histology diagnoses were made based on the 2004 World Health Organization/ International Society of Urologic Pathologists consensus guidelines. For the purpose of this study, histology results were separated into three main groups: benign (benign, cystitis glandularis, reactive treatment/ radiation changes), positive for HGUCs (urothelial carcinoma in situ, invasive and noninvasive HGUC), and LGUNs (papilloma, papillary urothelial neoplasia of low malignant potential, and low-grade urothelial carcinoma).12 The new cytologic diagnoses based on TPS and the original classifications were then correlated with the follow-up biopsy results as the gold standard. A 2-year follow-up of the cases with diagnosis of “urothelial atypia” based on the original reporting system was conducted by searching our pathology database. Results A total of 194 consecutive urinary tract cytology cases from years 2011 to 2012, all with follow-up histologic specimens, were included in our study. The samples were obtained from 169 male and 25 female patients (M:F ratio of 7 to 1) with a mean age of 69 years (range, 28 to 92 years). All urine samples were processed by cytocentrifuge and subsequently prepared by the SurePath method and Papanicolaou stain. The follow-up histopathology results were recorded, which included 91 benign specimens (46.9%), 52 LGUNs (26.8%), and 51 HGUCs (26.3%). The diagnoses rendered by both reviewers based on TPS criteria were compared, and discordant interpretations were identified in 29 cases (15%). Discrepant diagnoses (benign vs malignant) was made in only one case. The majority of the differences in interpretations did not pose significant discrepancy because there was no disagreement in the benignity or malignancy of the lesion. The results of each diagnostic category using both reporting systems were recorded Figure 1. The summary of cytology results using both reporting systems and histologic correlations are shown in Table 2 and Table 3. Implementing TPS resulted in classification of 17 more cases as negative for HGUC (NHGUC), which all corresponded to benign or LGUN on follow-up histology. No significant change was observed in the number of HGUCs identified by both systems (26 vs 27). Our study showed an overall decrease in the number of cases in the atypia category (24 less cases, 11.8% vs 24.2%). Upon reclassification, 17 (70%) of the original atypical cases were reclassified as NHGUC, of which 12 correlated with benign histology and five were LGUN on follow-up surgical pathology specimens. The reclassification of the atypia group to NHGUC was based on using TPS criteria, especially nuclear size, because the so-called “atypical” cells did not show the required increased N:C ratio of more than 50%. Seven cases (29%) with original diagnosis of urothelial atypia were reclassified as suspicious for HGUC (SHGUC), of which six (85%) correlated with HGUC on the follow-up histology and one (15%) showed LGUN on biopsy. Table 2 Reclassification of Urine Samples Using the Paris System for Reporting Urinary Cytology Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma. View Large Table 2 Reclassification of Urine Samples Using the Paris System for Reporting Urinary Cytology Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma. View Large Table 3 Classification of Urine Samples Using Authors’ Original Reporting Categories Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; NM, negative for malignancy; SM, suspicious for malignancy; UA, urothelial atypia. View Large Table 3 Classification of Urine Samples Using Authors’ Original Reporting Categories Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; NM, negative for malignancy; SM, suspicious for malignancy; UA, urothelial atypia. View Large Figure 1 View largeDownload slide A, The Paris System for Reporting Urinary Cytology classified 131 cases as negative for high-grade urothelial carcinoma (NHGUC), 23 atypical urothelial cells (AUC), 11 suspicious for high-grade urothelial carcinoma (SHGUC), 27 high-grade urothelial carcinoma (HGUC), and two low-grade urothelial neoplasms (LGUN). B, The original diagnoses include 114 negative for malignancy (NM), 47 urothelial atypia (UA), four suspicious for malignancy (SM), 26 HGUC, and three low-grade urothelial carcinoma (LGUC). Figure 1 View largeDownload slide A, The Paris System for Reporting Urinary Cytology classified 131 cases as negative for high-grade urothelial carcinoma (NHGUC), 23 atypical urothelial cells (AUC), 11 suspicious for high-grade urothelial carcinoma (SHGUC), 27 high-grade urothelial carcinoma (HGUC), and two low-grade urothelial neoplasms (LGUN). B, The original diagnoses include 114 negative for malignancy (NM), 47 urothelial atypia (UA), four suspicious for malignancy (SM), 26 HGUC, and three low-grade urothelial carcinoma (LGUC). A 2-year follow-up was available in 39 of 47 patients with original diagnosis of “urothelial atypia.” In one case, carcinoma in situ was identified in a bladder biopsy that was performed 1 year after the initial urine test. This sample was diagnosed as atypical by both reporting systems. Two samples with initial benign biopsies showed LGUN on follow-up tissue specimens. The 2-year follow-up data are summarized in Table 4. Table 4 Reclassification of 47 Cases With Diagnosis of “Urothelial Atypia” by Original Reporting System: Changes in Diagnosis According to TPS Criteria With Follow-up Tissue Sample Diagnosis and 2-Year Follow-up TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NA, not available; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 4 Reclassification of 47 Cases With Diagnosis of “Urothelial Atypia” by Original Reporting System: Changes in Diagnosis According to TPS Criteria With Follow-up Tissue Sample Diagnosis and 2-Year Follow-up TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NA, not available; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma; TPS, the Paris System for Reporting Urinary Cytology. View Large The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy for detecting HGUC were calculated and then compared between the systems Table 5. TPS showed significantly improved specificity (72% vs 83.9%), PPV (48% vs 62.3%), and diagnostic accuracy (74% vs 81.4%) for detecting HGUCs. The data related to each diagnostic category from both systems were analyzed and compared Table 6. Our results showed high PPV for diagnosis of HGUC by both reporting systems (92% and 96.3%). By implementing TPS, the PPV increased from 50% to 72% for the suspicious group and decreased from 23.4% to 17.4% for the “atypia” group. This finding indicates the cases that were originally classified as “atypia” and showed features more concerning for malignancy, were upgraded to suspicious category by TPS. NPV was similarly high for both reporting systems. Overall, TPS showed improved performance and more accurate results. Table 5 Diagnostic Performance of Authors’ Original Reporting System and TPS for Detecting High-Grade Urothelial Carcinoma   Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4    Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4  NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 5 Diagnostic Performance of Authors’ Original Reporting System and TPS for Detecting High-Grade Urothelial Carcinoma   Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4    Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4  NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 6 Diagnostic Performance of Different Categories of Authors’ Original Reporting System and TPS   Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90    Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90  HGUC, high-grade urothelial carcinoma; NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 6 Diagnostic Performance of Different Categories of Authors’ Original Reporting System and TPS   Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90    Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90  HGUC, high-grade urothelial carcinoma; NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Discussion Urine cytology is a noninvasive and inexpensive test that is widely used for detecting new cases of HGUC and surveillance in patients with a history of urothelial carcinoma.2–4 TPS was designed to improve diagnostic accuracy of urine cytology and standardize the reporting terminology. Given that cytology has a high sensitivity for detecting high-grade lesions while a poor sensitivity for diagnosing low-grade neoplasms, the new system focuses on identifying HGUCs. Another purpose of this scheme is to better define cytomorphologic criteria of different categories and further clarify implication of the “atypia” category in order to help clinicians in patient management.9-11 Several researchers have attempted to better clarify atypical features and have proposed diagnostic criteria for defining the morphologic features associated with neoplasia rather than changes due to reactive/repair.5-8 Our study showed that applying TPS guidelines increased the number of NHGUC diagnoses. These findings were expected because according to TPS all the cases with benign/reactive features, as well as all entities with cytologic changes that are not concerning for HGUC, should be classified in the NHGUC category. Also, more cases with subsequent LGUNs were assigned to the NHGUC category. Hassan et al13 described similar findings, with more benign and LGUN cases being classified as NHGUC after implementing TPS. Implementing TPS reduced the rate of “atypia” diagnoses by 51%, similar to findings that have been reported by others.13-16 In our study, TPS reclassified 36.1% (n = 17) of atypia cases as NHGUC, which all correlated with benign (n = 12) and LGUN (n = 5) on the follow-up histology specimens. Although Granados et al17 have reported that applying TPS guidelines resulted in a significant increase in the “atypia” category, the current study and other similar studies had different results.13-16 By applying TPS criteria, 15% (n = 7) of original cases with atypia diagnoses were reclassified as SHGUC, of which 85% correlated with HGUC on the follow-up histology. These results indicate better diagnostic categorization of atypical cells with features warranting an underlying high-grade neoplasia as the SHGUC category. Cowan et al16 reported that by applying TPS criteria, 40% of the cases with indeterminate diagnosis were reclassified to a higher risk category, thus specimen risk stratification was improved. The predictive value for subsequent HGUC on surgical specimens significantly increased for the suspicious category (50% vs 72%) using TPS criteria. However, association of the “atypia” category with subsequent HGUC was decreased (PPV of 17.4% vs 23.4%), because a subset of cases with higher grade of cytologic atypia were upgraded to the SHGUC category. The aforementioned data indicate that TPS criteria placed more patients with biopsy-proven HGUC into a higher risk category of SHGUC. These findings suggest that prior to implementing TPS in our institution, a wide range of “atypical” features, including benign/reactive to somewhat worrisome for an underlying malignancy, were lumped into atypia category. By applying TPS criteria, a significant portion of these cases were assigned to benign or higher-grade categories, which showed excellent correlation with the follow-up histology, therefore indicating better morphologic criteria were defined by TPS. Our data showed that the category of HGUC from both reporting systems has high predictive value for subsequent HGUC on biopsy (92% and 96.3%), as reported previously.13,16 Other studies reported higher rates of HGUC diagnosis by implementing TPS.14,18 The observed difference probably reflects institutional variation between diagnostic criteria for HGUC before introduction of TPS. TPS showed slightly better sensitivity (72.5% and 74.5%) and NPV (87% vs 90%), which is an important finding because classifying more cases as NHGUC by TPS may raise concern about the sensitivity of urinary tract cytology as a screening test. TPS showed significantly improved specificity (72% to 83.9%), PPV (48% to 62.3%), and diagnostic accuracy (74% vs 81.4%) for detecting HGUC; therefore we concluded it improves the diagnostic accuracy and overall performance of urine cytology. We observed acceptable interobserver agreement. Two cytopathologists who reviewed the urine samples independently agreed on 85% of the cases by applying the strict criteria of TPS. In the majority of the cases with different interpretations, the diagnoses did not show a major discrepancy and agreed on the benign or malignant nature of the pathology, which would result in a low impact on clinical management. Long et al19 recently reported high interobserver variability, including major discrepancies with high clinical impact, but our study did not show that. In summary, we have found that TPS criteria decreased the rate of reporting “atypia” and showed a higher specificity, accuracy, and predictive value. TPS is an important step towards standardizing the criteria and terminology for reporting urinary cytology. The results may have been affected by the study’s limitations. Only cases with follow-up surgical pathology were included in the study, excluding the majority of urinary tract cytology because most urine samples are not accompanied by a biopsy. The slide review and classifications by TPS were performed by two experienced cytopathologists, as opposed to the original diagnoses by several pathologists with different levels of experience. Further studies are needed to evaluate the effect of TPS for reporting urinary cytology on a larger scale and to study the clinician’s experience with the new system and its impact on patient management. References 1. Papanicolaou GN. Cytology of the urine sediment in neoplasms of the urinary tract. J Urol . 1947; 57: 375- 379. Google Scholar CrossRef Search ADS PubMed  2. Koss LG, Deitch D, Ramanathan R, et al.   Diagnostic value of cytology of voided urine. Acta Cytol . 1985; 29: 810- 816. Google Scholar PubMed  3. Bastacky S, Ibrahim S, Wilczynski SP, et al.   The accuracy of urinary cytology in daily practice. Cancer . 1999; 87: 118- 128. Google Scholar CrossRef Search ADS PubMed  4. Planz B, Jochims E, Deix T, et al.   The role of urinary cytology for detection of bladder cancer. Eur J Surg Oncol . 2005; 31: 304- 308. Google Scholar CrossRef Search ADS PubMed  5. Brimo F, Vollmer RT, Case B, et al.   Accuracy of urine cytology and the significance of an atypical category. Am J Clin Pathol . 2009; 132: 785- 793. Google Scholar CrossRef Search ADS PubMed  6. Piaton E, Decaussin-Petrucci M, Mege-Lechevallier F, et al.   Diagnostic terminology for urinary cytology reports including the new subcategories “atypical urothelial cells of undetermined significance” (AUC-Us) and “cannot exclude high grade” (AUC-H). Cytopathology . 2014; 25: 27- 38. Google Scholar CrossRef Search ADS PubMed  7. Owens CL, Vandenbussche CJ, Burroughs FH, et al.   A review of reporting systems and terminology for urine cytology. Cancer Cytopathol . 2013; 121: 9- 14. Google Scholar CrossRef Search ADS PubMed  8. Rosenthal DL, Vandenbussche CJ, Burroughs FH, et al.   The Johns Hopkins Hospital template for urologic cytology samples: Part I-creating the template. Cancer Cytopathol . 2013; 121: 15- 20. Google Scholar CrossRef Search ADS PubMed  9. Rosenthal DL, Wojcik E M, Kurtycz DFI. The Paris System for Reporting Urinary Cytology . New York, NY: Springer; 2016. Google Scholar CrossRef Search ADS   10. Barkan GA, Wojcik EM, Nayar R, et al.   The Paris System for Reporting Urinary Cytology: the quest to develop a standardized terminology. Adv Anat Pathol . 2016; 23: 193- 201. Google Scholar CrossRef Search ADS PubMed  11. VandenBussche CJ. A review of the Paris System for Reporting Urinary Cytology. Cytopathology . 2016; 27: 153- 156. Google Scholar CrossRef Search ADS PubMed  12. Epstein JI, Amin MB, Reuter VR, et al.   The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am J Surg Pathol . 1998; 22: 1435- 1448. Google Scholar CrossRef Search ADS PubMed  13. Hassan M, Solanki S, Kassouf W, et al.   Impact of implementing the Paris System for Reporting Urine Cytology in the performance of urine cytology: a correlative study of 124 cases. Am J Clin Pathol . 2016; 146: 384- 390. Google Scholar CrossRef Search ADS PubMed  14. Malviya K, Fernandes G, Naik L, et al.   Utility of the Paris System in reporting urine cytology. Acta Cytol . 2017; 61: 145- 152. Google Scholar CrossRef Search ADS PubMed  15. Miki Y, Neat M, Chandra A. Application of the Paris System to atypical urine cytology samples: correlation with histology and Urovysion® FISH. Cytopathology . 2017; 28: 88- 95. Google Scholar CrossRef Search ADS PubMed  16. Cowan ML, Rosenthal DL, VandenBussche CJ. Improved risk stratification for patients with high-grade urothelial carcinoma following application of the Paris System for Reporting Urinary Cytology. Cancer Cytopathol . 2017; 125: 427- 434. Google Scholar CrossRef Search ADS PubMed  17. Granados R, Duarte JA, Corrales T, et al.   Applying the Paris System for reporting urine cytology increases the rate of atypical urothelial cells in benign cases: a need for patient management recommendations. Acta Cytol . 2017; 61: 71- 76. Google Scholar CrossRef Search ADS PubMed  18. Roy M, Jain D, Kaushal S, et al.   An institutional experience with the Paris System: a paradigm shift from ambiguous terminology to more objective criteria for reporting urine cytology. Cytopathology . 2017; 28: 509- 515. Google Scholar CrossRef Search ADS PubMed  19. Long T, Layfield LJ, Esebua M, et al.   Interobserver reproducibility of the Paris System for Reporting Urinary Cytology. Cytojournal . 2017; 14: 17. Google Scholar CrossRef Search ADS PubMed  © American Society for Clinical Pathology, 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Journal of Clinical Pathology Oxford University Press

A Single Institutional Experience With the Paris System for Reporting Urinary Cytology: Correlation of Cytology and Histology in 194 Cases

Loading next page...
 
/lp/ou_press/a-single-institutional-experience-with-the-paris-system-for-reporting-50zU9o8MGd
Publisher
Oxford University Press
Copyright
© American Society for Clinical Pathology, 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0002-9173
eISSN
1943-7722
D.O.I.
10.1093/ajcp/aqy043
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives The Paris System for Reporting Urinary Cytology (TPS) is designed to standardize the criteria and terminology used in urinary tract cytology reporting. The aim of this study was to evaluate the impact of implementing TPS and to analyze the correlation with follow-up biopsies in order to assess its reproducibility. Methods Urinary tract cytology specimens with follow-up biopsies over a 2-year period were reviewed and reclassified according to TPS criteria. Surgical follow-up diagnoses were correlated with the initial cytology diagnoses and TPS interpretations, and the results were compared. Results Applying TPS in comparison to our previous reporting system resulted in fewer cases in the atypia category (11.8% vs 24.2%) and higher specificity, accuracy, and predictive value. We observed acceptable interobserver agreement in diagnostic categories of this reporting system. Conclusions TPS improves the overall performance of urinary tract cytology by standardizing the criteria and terminology. Urine, Cytology, Atypia, High-grade urothelial carcinoma, Follow-up Urinary tract cytology is currently the most commonly used test for screening and monitoring of urothelial carcinomas. The available information indicates that urinary cytology is an accurate test for detection of high-grade urothelial carcinoma (HGUC), which has the potential for invasion, metastasis, and a more aggressive course. Urinary tract cytology has known limitations in the diagnosis of low-grade urothelial neoplasms (LGUN), but most LGUN form papillary lesions, which are easily identified and biopsied by cystoscopy examination. There has been recent progress in diagnostic and radiological techniques for urothelial lesions, but urinary tract cytology is still a noninvasive, simple, and low-cost test that is highly sensitive for detection of HGUC, especially flat carcinomas that may be cystoscopically occult.1-4 Urine cytology has challenges when dealing with cases that show some degree of cytologic atypia but do not meet the criteria for malignancy. Various conditions such as inflammation, instrumentation, reactive changes due to treatment effects, or urinary tract calculi can result in cytologic atypia and false-positive results.5,6 The reporting rate of “atypia” diagnoses varies significantly between pathologists and cytotechnologists and even between different institutions, which introduces a management challenge for clinicians. Lack of standard diagnostic criteria and widely accepted terminology in urine cytology reporting led to significant variability between reporting systems, and several classification schemes have been proposed to address this issue.6-8 At the 2013 International Congress of Cytology, the Paris System for Reporting Urinary Cytology (TPS) was introduced to standardize urinary tract cytology reporting. The guidelines provide cytomorphologic criteria for categorizing the cytology results. This system emphasizes the main objective of urine cytology, which is identifying HGUC. TPS followed the lead of widely accepted Bethesda Systems for reporting cervical cytology and thyroid cytology and used a similar terminology.9,10 We sought to utilize previously collected urine cytology specimens with histologic follow-up and to reevaluate them based on the cytomorphologic criteria outlined in TPS. Our goal was to access the utility of TPS in comparison to the previously used classification system at our institution, with special emphasis placed on the rate of “atypia” diagnosis. Materials and Methods Consecutive lower urinary tract cytology cases with follow-up histology specimens from January 2011 to December 2012 were reviewed. The specimens included voided urines, instrumented urines, and bladder wash samples. The follow-up histology specimens were obtained within 6 months after the time of urine sample collection. All specimens were Papanicolaou stained, SurePath (Becton Dickinson, Franklin Lakes, NJ) preparations. When the patient had multiple urinary cytologies preceding the biopsy, the urine sample with the most atypical diagnosis was included in the study. Cases with nondiagnostic biopsies, inadequate urine samples, and nonurothelial malignancies were excluded. The patient’s age, gender, original cytology diagnoses, and follow-up surgical pathology results were recorded. Two cytopathology-boarded pathologists with more than 10 years of experience (F.H. and A.S.) independently reviewed all the cases. The reviewing pathologists assigned a diagnosis to each case using strict adherence to TPS criteria Table 1. They were blinded to previous diagnoses, clinical histories, and follow-up surgical biopsies. Cases with discordant diagnoses between the two reviewing pathologists were re-reviewed in a consensus conference to obtain a final diagnosis. Table 1 Diagnostic Categories of the Paris System for Reporting Urinary Cytology Specimens With Cytomorphologic Criteria of Each Category Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  N:C ratio, nuclear to cytoplasmic ratio. View Large Table 1 Diagnostic Categories of the Paris System for Reporting Urinary Cytology Specimens With Cytomorphologic Criteria of Each Category Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  Cytology Diagnosis  Cytomorphologic Features  Nondiagnostic or unsatisfactory  Instrumented specimens with <10 well-preserved, well-visualized urothelial cells per 10 high-power fields; voided urine specimens lacking appropriate benign urothelial cellularity with <30 mL volume, or specimens with features that obscure characterization of urothelial cells (ie, inflammatory cells, RBC, sperm, or bacteria)  Negative for high-grade urothelial carcinoma  Benign/reactive cells (urothelial, squamous, and glandular), changes caused by a benign process (ie, irradiation, posttreatment effect, calculi, and viral cytopathic effects), tissue fragments and clusters without cytomorphologic changes in presence or absence of instrumentation  Atypical urothelial cells  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of more than 0.5, with the presence of either of the following features: nuclear hyperchromasia, coarse clumped chromatin, or irregular nuclear membranes  Suspicious for high-grade urothelial carcinoma  Nonsuperficial and nondegenerated urothelial cells with increased N:C ratio of at least 0.5-0.7 and moderate to severe hyperchromasia, with the presence of either of the following features: irregular clumped chromatin or marked irregular nuclear membranes  Low-grade urothelial neoplasia (LGUN)  Three-dimensional papillary clusters of urothelial cells with fibrovascular cores is a definitive finding; findings suggestive of LGUN: three-dimensional papillary clusters of urothelial cells without fibrovascular cores or increased number of monotonous single cells  High-grade urothelial carcinoma  At least 5-10 abnormal urothelial cells with N:C ratio of equal to or more than 0.7, moderate to severe nuclear hyperchromasia, coarse clumped chromatin, and markedly irregular nuclear membranes  Other malignancies, primary and metastatic  Primary nonurothelial tumors, direct extension to bladder and metastatic tumors  N:C ratio, nuclear to cytoplasmic ratio. View Large TPS describes seven diagnostic categories and defines cytomorphologic criteria of each group. All TPS categories with corresponding criteria are summarized in Table 1.9-11 In our institution, before the application of TPS, a variety of reporting terminologies were used by different pathologists. For the purpose of this study, the original diagnoses were organized in five groups: negative for malignancy, urothelial atypia, low-grade urothelial carcinoma, suspicious for malignancy, and HGUC. The follow-up histopathologic samples included biopsy and resection specimens. The histology diagnoses were made based on the 2004 World Health Organization/ International Society of Urologic Pathologists consensus guidelines. For the purpose of this study, histology results were separated into three main groups: benign (benign, cystitis glandularis, reactive treatment/ radiation changes), positive for HGUCs (urothelial carcinoma in situ, invasive and noninvasive HGUC), and LGUNs (papilloma, papillary urothelial neoplasia of low malignant potential, and low-grade urothelial carcinoma).12 The new cytologic diagnoses based on TPS and the original classifications were then correlated with the follow-up biopsy results as the gold standard. A 2-year follow-up of the cases with diagnosis of “urothelial atypia” based on the original reporting system was conducted by searching our pathology database. Results A total of 194 consecutive urinary tract cytology cases from years 2011 to 2012, all with follow-up histologic specimens, were included in our study. The samples were obtained from 169 male and 25 female patients (M:F ratio of 7 to 1) with a mean age of 69 years (range, 28 to 92 years). All urine samples were processed by cytocentrifuge and subsequently prepared by the SurePath method and Papanicolaou stain. The follow-up histopathology results were recorded, which included 91 benign specimens (46.9%), 52 LGUNs (26.8%), and 51 HGUCs (26.3%). The diagnoses rendered by both reviewers based on TPS criteria were compared, and discordant interpretations were identified in 29 cases (15%). Discrepant diagnoses (benign vs malignant) was made in only one case. The majority of the differences in interpretations did not pose significant discrepancy because there was no disagreement in the benignity or malignancy of the lesion. The results of each diagnostic category using both reporting systems were recorded Figure 1. The summary of cytology results using both reporting systems and histologic correlations are shown in Table 2 and Table 3. Implementing TPS resulted in classification of 17 more cases as negative for HGUC (NHGUC), which all corresponded to benign or LGUN on follow-up histology. No significant change was observed in the number of HGUCs identified by both systems (26 vs 27). Our study showed an overall decrease in the number of cases in the atypia category (24 less cases, 11.8% vs 24.2%). Upon reclassification, 17 (70%) of the original atypical cases were reclassified as NHGUC, of which 12 correlated with benign histology and five were LGUN on follow-up surgical pathology specimens. The reclassification of the atypia group to NHGUC was based on using TPS criteria, especially nuclear size, because the so-called “atypical” cells did not show the required increased N:C ratio of more than 50%. Seven cases (29%) with original diagnosis of urothelial atypia were reclassified as suspicious for HGUC (SHGUC), of which six (85%) correlated with HGUC on the follow-up histology and one (15%) showed LGUN on biopsy. Table 2 Reclassification of Urine Samples Using the Paris System for Reporting Urinary Cytology Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma. View Large Table 2 Reclassification of Urine Samples Using the Paris System for Reporting Urinary Cytology Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  Histology  Cytology, No.  NHGUC  AUC  SHGUC  HGUC  LGUN  Benign (n = 91, 46.9%)  76  14  1  0  0  HGUC (n = 51, 26.3%)  13  4  8  26  0  LGUC (n = 52, 26.8%)  42  5  2  1  2  Total (n = 194)  131  23  11  27  2  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma. View Large Table 3 Classification of Urine Samples Using Authors’ Original Reporting Categories Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; NM, negative for malignancy; SM, suspicious for malignancy; UA, urothelial atypia. View Large Table 3 Classification of Urine Samples Using Authors’ Original Reporting Categories Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  Histology  Cytology, No.  NM  UA  SM  HGUC  LGUC  Benign (n = 91, 46.9%)  64  26  1  0  0  HGUC (n = 51, 26.3%)  14  11  2  24  0  LGUC (n = 52, 26.8%)  36  10  1  2  3  Total (n = 194)  114  47  4  26  3  HGUC, high-grade urothelial carcinoma; LGUC, low-grade urothelial carcinoma; NM, negative for malignancy; SM, suspicious for malignancy; UA, urothelial atypia. View Large Figure 1 View largeDownload slide A, The Paris System for Reporting Urinary Cytology classified 131 cases as negative for high-grade urothelial carcinoma (NHGUC), 23 atypical urothelial cells (AUC), 11 suspicious for high-grade urothelial carcinoma (SHGUC), 27 high-grade urothelial carcinoma (HGUC), and two low-grade urothelial neoplasms (LGUN). B, The original diagnoses include 114 negative for malignancy (NM), 47 urothelial atypia (UA), four suspicious for malignancy (SM), 26 HGUC, and three low-grade urothelial carcinoma (LGUC). Figure 1 View largeDownload slide A, The Paris System for Reporting Urinary Cytology classified 131 cases as negative for high-grade urothelial carcinoma (NHGUC), 23 atypical urothelial cells (AUC), 11 suspicious for high-grade urothelial carcinoma (SHGUC), 27 high-grade urothelial carcinoma (HGUC), and two low-grade urothelial neoplasms (LGUN). B, The original diagnoses include 114 negative for malignancy (NM), 47 urothelial atypia (UA), four suspicious for malignancy (SM), 26 HGUC, and three low-grade urothelial carcinoma (LGUC). A 2-year follow-up was available in 39 of 47 patients with original diagnosis of “urothelial atypia.” In one case, carcinoma in situ was identified in a bladder biopsy that was performed 1 year after the initial urine test. This sample was diagnosed as atypical by both reporting systems. Two samples with initial benign biopsies showed LGUN on follow-up tissue specimens. The 2-year follow-up data are summarized in Table 4. Table 4 Reclassification of 47 Cases With Diagnosis of “Urothelial Atypia” by Original Reporting System: Changes in Diagnosis According to TPS Criteria With Follow-up Tissue Sample Diagnosis and 2-Year Follow-up TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NA, not available; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 4 Reclassification of 47 Cases With Diagnosis of “Urothelial Atypia” by Original Reporting System: Changes in Diagnosis According to TPS Criteria With Follow-up Tissue Sample Diagnosis and 2-Year Follow-up TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  TPS Interpretation, No.  Tissue Diagnosis Within 6 mo, No.  2-Year Follow-up, No.  NHGUC, 17  Benign, 12  Benign, 7  LGUN, 1  NA, 4  AUC, 23  Benign, 14  LGUN, 4  NA, 1  Benign, 10  LGUN, 1  LGUN, 5  HGUC, 1  NA, 2  HGUC, 4  LGUN, 4  SHGUC, 7  LGUN, 1  NA, 1  HGUC, 6  HGUC, 4  AUC, atypical urothelial cells; HGUC, high-grade urothelial carcinoma; LGUN, low-grade urothelial neoplasia; NA, not available; NHGUC, negative for high-grade urothelial carcinoma; SHGUC, suspicious for high-grade urothelial carcinoma; TPS, the Paris System for Reporting Urinary Cytology. View Large The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy for detecting HGUC were calculated and then compared between the systems Table 5. TPS showed significantly improved specificity (72% vs 83.9%), PPV (48% vs 62.3%), and diagnostic accuracy (74% vs 81.4%) for detecting HGUCs. The data related to each diagnostic category from both systems were analyzed and compared Table 6. Our results showed high PPV for diagnosis of HGUC by both reporting systems (92% and 96.3%). By implementing TPS, the PPV increased from 50% to 72% for the suspicious group and decreased from 23.4% to 17.4% for the “atypia” group. This finding indicates the cases that were originally classified as “atypia” and showed features more concerning for malignancy, were upgraded to suspicious category by TPS. NPV was similarly high for both reporting systems. Overall, TPS showed improved performance and more accurate results. Table 5 Diagnostic Performance of Authors’ Original Reporting System and TPS for Detecting High-Grade Urothelial Carcinoma   Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4    Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4  NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 5 Diagnostic Performance of Authors’ Original Reporting System and TPS for Detecting High-Grade Urothelial Carcinoma   Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4    Authors’ Original Reporting System, %  TPS, %  Sensitivity  72.5  74.5  Specificity  72  83.9  PPV  48  62.3  NPV  88  90  Accuracy  74  81.4  NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 6 Diagnostic Performance of Different Categories of Authors’ Original Reporting System and TPS   Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90    Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90  HGUC, high-grade urothelial carcinoma; NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Table 6 Diagnostic Performance of Different Categories of Authors’ Original Reporting System and TPS   Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90    Authors’ Original Reporting System, %  TPS, %  PPV for diagnosis of HGUC   Atypia  23.4  17.4   Suspicious  50  72   HGUC  92  96.3  NPV   Negative  87  90  HGUC, high-grade urothelial carcinoma; NPV, negative predictive value; PPV, positive predictive value; TPS, the Paris System for Reporting Urinary Cytology. View Large Discussion Urine cytology is a noninvasive and inexpensive test that is widely used for detecting new cases of HGUC and surveillance in patients with a history of urothelial carcinoma.2–4 TPS was designed to improve diagnostic accuracy of urine cytology and standardize the reporting terminology. Given that cytology has a high sensitivity for detecting high-grade lesions while a poor sensitivity for diagnosing low-grade neoplasms, the new system focuses on identifying HGUCs. Another purpose of this scheme is to better define cytomorphologic criteria of different categories and further clarify implication of the “atypia” category in order to help clinicians in patient management.9-11 Several researchers have attempted to better clarify atypical features and have proposed diagnostic criteria for defining the morphologic features associated with neoplasia rather than changes due to reactive/repair.5-8 Our study showed that applying TPS guidelines increased the number of NHGUC diagnoses. These findings were expected because according to TPS all the cases with benign/reactive features, as well as all entities with cytologic changes that are not concerning for HGUC, should be classified in the NHGUC category. Also, more cases with subsequent LGUNs were assigned to the NHGUC category. Hassan et al13 described similar findings, with more benign and LGUN cases being classified as NHGUC after implementing TPS. Implementing TPS reduced the rate of “atypia” diagnoses by 51%, similar to findings that have been reported by others.13-16 In our study, TPS reclassified 36.1% (n = 17) of atypia cases as NHGUC, which all correlated with benign (n = 12) and LGUN (n = 5) on the follow-up histology specimens. Although Granados et al17 have reported that applying TPS guidelines resulted in a significant increase in the “atypia” category, the current study and other similar studies had different results.13-16 By applying TPS criteria, 15% (n = 7) of original cases with atypia diagnoses were reclassified as SHGUC, of which 85% correlated with HGUC on the follow-up histology. These results indicate better diagnostic categorization of atypical cells with features warranting an underlying high-grade neoplasia as the SHGUC category. Cowan et al16 reported that by applying TPS criteria, 40% of the cases with indeterminate diagnosis were reclassified to a higher risk category, thus specimen risk stratification was improved. The predictive value for subsequent HGUC on surgical specimens significantly increased for the suspicious category (50% vs 72%) using TPS criteria. However, association of the “atypia” category with subsequent HGUC was decreased (PPV of 17.4% vs 23.4%), because a subset of cases with higher grade of cytologic atypia were upgraded to the SHGUC category. The aforementioned data indicate that TPS criteria placed more patients with biopsy-proven HGUC into a higher risk category of SHGUC. These findings suggest that prior to implementing TPS in our institution, a wide range of “atypical” features, including benign/reactive to somewhat worrisome for an underlying malignancy, were lumped into atypia category. By applying TPS criteria, a significant portion of these cases were assigned to benign or higher-grade categories, which showed excellent correlation with the follow-up histology, therefore indicating better morphologic criteria were defined by TPS. Our data showed that the category of HGUC from both reporting systems has high predictive value for subsequent HGUC on biopsy (92% and 96.3%), as reported previously.13,16 Other studies reported higher rates of HGUC diagnosis by implementing TPS.14,18 The observed difference probably reflects institutional variation between diagnostic criteria for HGUC before introduction of TPS. TPS showed slightly better sensitivity (72.5% and 74.5%) and NPV (87% vs 90%), which is an important finding because classifying more cases as NHGUC by TPS may raise concern about the sensitivity of urinary tract cytology as a screening test. TPS showed significantly improved specificity (72% to 83.9%), PPV (48% to 62.3%), and diagnostic accuracy (74% vs 81.4%) for detecting HGUC; therefore we concluded it improves the diagnostic accuracy and overall performance of urine cytology. We observed acceptable interobserver agreement. Two cytopathologists who reviewed the urine samples independently agreed on 85% of the cases by applying the strict criteria of TPS. In the majority of the cases with different interpretations, the diagnoses did not show a major discrepancy and agreed on the benign or malignant nature of the pathology, which would result in a low impact on clinical management. Long et al19 recently reported high interobserver variability, including major discrepancies with high clinical impact, but our study did not show that. In summary, we have found that TPS criteria decreased the rate of reporting “atypia” and showed a higher specificity, accuracy, and predictive value. TPS is an important step towards standardizing the criteria and terminology for reporting urinary cytology. The results may have been affected by the study’s limitations. Only cases with follow-up surgical pathology were included in the study, excluding the majority of urinary tract cytology because most urine samples are not accompanied by a biopsy. The slide review and classifications by TPS were performed by two experienced cytopathologists, as opposed to the original diagnoses by several pathologists with different levels of experience. Further studies are needed to evaluate the effect of TPS for reporting urinary cytology on a larger scale and to study the clinician’s experience with the new system and its impact on patient management. References 1. Papanicolaou GN. Cytology of the urine sediment in neoplasms of the urinary tract. J Urol . 1947; 57: 375- 379. Google Scholar CrossRef Search ADS PubMed  2. Koss LG, Deitch D, Ramanathan R, et al.   Diagnostic value of cytology of voided urine. Acta Cytol . 1985; 29: 810- 816. Google Scholar PubMed  3. Bastacky S, Ibrahim S, Wilczynski SP, et al.   The accuracy of urinary cytology in daily practice. Cancer . 1999; 87: 118- 128. Google Scholar CrossRef Search ADS PubMed  4. Planz B, Jochims E, Deix T, et al.   The role of urinary cytology for detection of bladder cancer. Eur J Surg Oncol . 2005; 31: 304- 308. Google Scholar CrossRef Search ADS PubMed  5. Brimo F, Vollmer RT, Case B, et al.   Accuracy of urine cytology and the significance of an atypical category. Am J Clin Pathol . 2009; 132: 785- 793. Google Scholar CrossRef Search ADS PubMed  6. Piaton E, Decaussin-Petrucci M, Mege-Lechevallier F, et al.   Diagnostic terminology for urinary cytology reports including the new subcategories “atypical urothelial cells of undetermined significance” (AUC-Us) and “cannot exclude high grade” (AUC-H). Cytopathology . 2014; 25: 27- 38. Google Scholar CrossRef Search ADS PubMed  7. Owens CL, Vandenbussche CJ, Burroughs FH, et al.   A review of reporting systems and terminology for urine cytology. Cancer Cytopathol . 2013; 121: 9- 14. Google Scholar CrossRef Search ADS PubMed  8. Rosenthal DL, Vandenbussche CJ, Burroughs FH, et al.   The Johns Hopkins Hospital template for urologic cytology samples: Part I-creating the template. Cancer Cytopathol . 2013; 121: 15- 20. Google Scholar CrossRef Search ADS PubMed  9. Rosenthal DL, Wojcik E M, Kurtycz DFI. The Paris System for Reporting Urinary Cytology . New York, NY: Springer; 2016. Google Scholar CrossRef Search ADS   10. Barkan GA, Wojcik EM, Nayar R, et al.   The Paris System for Reporting Urinary Cytology: the quest to develop a standardized terminology. Adv Anat Pathol . 2016; 23: 193- 201. Google Scholar CrossRef Search ADS PubMed  11. VandenBussche CJ. A review of the Paris System for Reporting Urinary Cytology. Cytopathology . 2016; 27: 153- 156. Google Scholar CrossRef Search ADS PubMed  12. Epstein JI, Amin MB, Reuter VR, et al.   The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am J Surg Pathol . 1998; 22: 1435- 1448. Google Scholar CrossRef Search ADS PubMed  13. Hassan M, Solanki S, Kassouf W, et al.   Impact of implementing the Paris System for Reporting Urine Cytology in the performance of urine cytology: a correlative study of 124 cases. Am J Clin Pathol . 2016; 146: 384- 390. Google Scholar CrossRef Search ADS PubMed  14. Malviya K, Fernandes G, Naik L, et al.   Utility of the Paris System in reporting urine cytology. Acta Cytol . 2017; 61: 145- 152. Google Scholar CrossRef Search ADS PubMed  15. Miki Y, Neat M, Chandra A. Application of the Paris System to atypical urine cytology samples: correlation with histology and Urovysion® FISH. Cytopathology . 2017; 28: 88- 95. Google Scholar CrossRef Search ADS PubMed  16. Cowan ML, Rosenthal DL, VandenBussche CJ. Improved risk stratification for patients with high-grade urothelial carcinoma following application of the Paris System for Reporting Urinary Cytology. Cancer Cytopathol . 2017; 125: 427- 434. Google Scholar CrossRef Search ADS PubMed  17. Granados R, Duarte JA, Corrales T, et al.   Applying the Paris System for reporting urine cytology increases the rate of atypical urothelial cells in benign cases: a need for patient management recommendations. Acta Cytol . 2017; 61: 71- 76. Google Scholar CrossRef Search ADS PubMed  18. Roy M, Jain D, Kaushal S, et al.   An institutional experience with the Paris System: a paradigm shift from ambiguous terminology to more objective criteria for reporting urine cytology. Cytopathology . 2017; 28: 509- 515. Google Scholar CrossRef Search ADS PubMed  19. Long T, Layfield LJ, Esebua M, et al.   Interobserver reproducibility of the Paris System for Reporting Urinary Cytology. Cytojournal . 2017; 14: 17. Google Scholar CrossRef Search ADS PubMed  © American Society for Clinical Pathology, 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

American Journal of Clinical PathologyOxford University Press

Published: Jun 6, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off