Reproducibility of the Johns Hopkins Hospital template for urologic cytology samples

Adequacy of voided urine specimens prepared by ThinPrep and evaluated using The Paris System for Reporting Urinary Cytology

INTRODUCTION

The Paris System for Reporting Urinary Cytology (TPS) currently includes a volume recommendation for voided urine specimens of 30 mL based on observations of performance with BD SurePath preparation system. Given that many labs use the Hologic ThinPrep methods for voided urines, an analysis of data acquired with this system is undertaken in this study.

MATERIALS AND METHODS

We identified a total of 744 voided urine specimens. All specimens were processed fresh by ThinPrep at a large academic center that had incorporated a tiered urine cytology reporting template prior to TPS. To determine the optimum binary cutoff thresholds for voided urine volume, the fraction of high-risk cytologic diagnoses-suspicious for high-grade urothelial carcinoma (SHGUC) or worse-was compared below and above various volumes cutpoints.

RESULTS

The cytology diagnosis was inadequate in 1.5%, negative for high-grade urothelial carcinoma in 64%, atypical urothelial cells in 14.2% SHGUC in 6.6%, high-grade urothelial carcinoma (HGUC) in 11.3%, low-grade urothelial neoplasm in 1.9%, and other malignancies in 0.5%. High-risk cytology was diagnosed in 19.1% of specimens ≥25 mL and in 13.5% of specimens <25 mL (P = 0.090). Volume of ≥25 mL was associated with the optimum cutoff for diagnosing SHGUC or HGUC.

CONCLUSIONS

A specimen volume of ≥25 mL is associated with higher rates of diagnosis of HGUC or SHGUC in voided urine specimens processed by ThinPrep. This is similar to the 30 mL cutoff determined for SurePath preparation that was incorporated into TPS and may provide helpful information for ThinPrep labs that are in the process of adopting TPS.

Cytotechnologist Performance for Screening Hürthle Cell Atypia in Indeterminate Thyroid Fine-Needle Aspirates

OBJECTIVES

We have previously shown that specimens diagnosed as containing Hürthle cells have a 12% chance of being malignant if they are classified as atypia of undetermined significance (AUS-HC). The identification of Hürthle cells by cytotechnologists (CTs) during screening can improve cytopathologist efficiency and may prevent diagnostic errors due to the oversights of focal findings. Here, we examine the performance of our institutional CTs when screening for Hürthle cell atypia in thyroid fine-needle aspiration (FNA) specimens.

STUDY DESIGN

Information on 8,814 thyroid cytopathology specimens was retrieved for a 10-year period. Specimens were screened by 1 of 11 CTs. A subsample of cases was categorized either as AUS-HC or suspicious for Hürthle cell neoplasm.

RESULTS

AUS-HC screening diagnoses were more likely to be downgraded to benign but less likely to be upgraded compared to AUS diagnoses with nuclear or microfollicular atypia. AUS-HC represents almost all papillary thyroid carcinoma (PTC) screening diagnoses downgraded to the AUS category, which suggests that even low levels of Hürthle cell atypia can result in PTC being included in the differential diagnosis.

CONCLUSION

Overall, there are few major discrepancies between CT and pathologist diagnoses for specimens containing Hürthle cell atypia.

Subclassifying atypia in urine cytology: what are the helpful features?

INTRODUCTION

The diagnosis "atypical urothelial cells (AUC)" remains an unresolved problem, making many urologists dissatisfied and confused about the management strategy on these cases. To date, a few inspiring attempts were made to subclassify AUC into "atypical urothelial cells of undetermined significance" (AUC-US) and "atypical urothelial cells cannot exclude high grade" (AUC-H). The aim of our study was to investigate the most predictive for high-grade urothelial carcinoma (HGUC) cytomorphologic parameters and whether the proposed classification can be implemented in our institution.

MATERIAL AND METHODS

The electronic medical record system was searched for cytology specimens that were diagnosed as AUC from January 1, 2005 to March 1, 2013 and their relative clinical-pathological follow-up. All specimens were reviewed by an experienced cytopathologist by using 20 published "most predictive" for HGUC criteria.

RESULTS

A total of 162 AUC specimens were reclassified into 3 groups: AUC-H (n = 45), AUC-US (n = 51), and "negative for malignancy" (n = 66). The reclassification of AUC-H and "negative for malignancy" had 79% sensitivity, 77% specificity, 60% positive predictive value, and 89% negative predictive value to histologically proven HGUC diagnosis.

CONCLUSIONS

Our study demonstrated a good correlation between the presence of "HGUC-predictive" cytologic criteria and the final biopsy-proven HGUC in cytologic cases originally diagnosed as "atypical urothelial cells present." We identified 2 of the most predictive for HGUC on follow-up cytomorphologic parameters such as increased nuclear-cytoplasmic ratio >0.7 and coarse chromatin (16 abnormal cells per slide in average). These parameters, along with positive fluorescent in situ hybridization results can help during cytologic evaluation of urine specimens.

Atypical urothelial tissue fragments in noninstrumented voided urine specimens are associated with low but significantly higher rates of urothelial neoplasia than benign-appearing urothelial tissue fragments

BACKGROUND

The interpretation of urothelial tissue fragments (UTF) in voided urine (VU) specimens is controversial. If UTF contain cytomorphologically atypical cells, the diagnosis often becomes more challenging. The authors previously analyzed the outcome of patients with benign-appearing UTF in 274 noninstrumented VU specimens. In this retrospective study, noninstrumented VU specimens containing UTF with atypical cytomorphological features (AUTF) were evaluated and compared with the previous results.

METHODS

The Johns Hopkins Hospital electronic pathology database was searched for VU cases containing UTF over a 5-year period.

RESULTS

A total of 170 noninstrumented VU specimens containing AUTF were identified. Twenty-four specimens had subsequent or coincidental surgical pathology specimens with the following overall rates of neoplasia: high-grade urothelial carcinoma: 8.8% (15 specimens), low-grade urothelial neoplasia: 1.2% (2 specimens), and prostate carcinoma invading the bladder: 0.6% (1 specimen). A total of 49 specimens (28.8%) were diagnosed with urolithiasis on follow-up. Twenty-five cases had follow-up cytology specimens, all of which were negative for malignancy. Of 72 cases without histopathologic, radiologic, or cytopathologic follow-up, 62 (86.1%) had a mean clinical follow-up of 22.5 months and 10 cases did not have clinical follow-up.

CONCLUSIONS

The presence of AUTF in noninstrumented VU is associated with low rates of urothelial neoplasia but a statistically higher risk of urothelial neoplasia than the presence of BUTF (10.0% vs 4.4%; P<.05). In particular, the rate of high-grade urothelial carcinoma is significantly higher in noninstrumented VU specimens containing AUTF than those containing BUTF (8.8% vs 0.7%; P<.0001). In addition, urolithiasis is associated with AUTF in a substantial percentage of noninstrumented VU specimens.

Cytotechnologist performance for screening microfollicular atypia in indeterminate thyroid fine-needle aspirates

INTRODUCTION

We previously identified a high level of accuracy among our cytotechnologists (CTs) for identifying nuclear atypia in thyroid fine-needle aspiration (FNA) specimens. Herewith, we present our CT performance at screening for microfollicular atypia.

METHODS

8,814 thyroid FNA specimens were identified in our archives, all screened by 1 of 11 CTs and signed out by a cytopathologist. A subsample of cases was categorized either as atypia of uncertain significance (AUS) with microfollicular proliferation (AUS-F) or suspicious for a follicular neoplasm (SFN).

RESULTS

The agreement rate was low between CTs and cytopathologists for SFN and AUS-F. Only 55.8% of SFN screening diagnoses were upheld; 27.9% were downgraded to AUS, 10.4% were downgraded to benign, and 5% were upgraded. Of AUS-F screening diagnoses, 35.5% were upheld, 33.7% were downgraded to benign, and 20.2% were upgraded to SFN. Among all cases, two-step discrepancies were uncommon.

CONCLUSION

Most disagreements were one-category discrepancies between AUS-F and SFN. The evaluation of microfollicular atypia is challenging given that certain follicular lesions cannot be definitively diagnosed on cytology, a high level of subjectivity is involved in the interpretation of such lesions, and the presence of nuclear or Hurthle cell atypia may complicate the diagnosis.