Interpretive trends and pitfalls: a look back at ’03
Jonathan H. Hughes, MD, PhD
The CAP Interlaboratory Comparison Programs in Gynecologic and Nongynecologic
Cytopathology each year create and distribute to participants a summary of data
accumulated from the prior year. These reports are a compilation of the substantial
number of interpretations that are rendered by pathologists, cytotechnologists,
and laboratories on the slides circulated to participants. The Year-End Summary
Reports, or YESR, contain a wealth of information about the performance of practitioners
across a wide variety of settings and identify important trends and diagnostic
pitfalls. While the reader is encouraged to review the complete 2003 reports
for their full content, here are highlights that deserve special mention from
each of the programs.
CAP Interlaboratory Comparison Program in Gynecologic Cytopathology,
or Pap program
Results of 2003 demographics study. A study of participating laboratories was made in 2003 through distribution of a demographics questionnaire in the PAPB mailing. Cytology laboratories of all types participate in the Pap program, with the largest proportion being hospital laboratories. The majority of reporting laboratories (73.8 percent) examine less than 10,000 cytopathology cases per year. Of the laboratories that responded to the survey, 7.1 percent did not examine any conventional slides, 61.6 percent did not examine any ThinPrep preparations, and 85.5 percent did not examine SurePath preparations. Of participating laboratories that are using liquid-based gynecologic cytology, just over 50 percent reported that the technique had been used for more than five years, though a significant proportion had implemented liquid-based cytology within the last year.
Trends in sensitivity and specificity. The sensitivity and specificity of the 2003 Pap program participating laboratories for identifying significant abnormal lesions (200 series) were 98.5 percent and 98.5 percent, respectively. These findings are unchanged from the previous three years. Cytotechnologists continue to demonstrate higher sensitivity (98.9 percent versus 97.8 percent, respectively) and higher specificity (98.3 percent versus 97.5 percent, respectively) than pathologists. The sensitivity for validated conventional slides appears to be slightly higher than for validated ThinPrep slides. Although no validated SurePath slides are yet available, the 2003 data suggest that the sensitivities of ThinPrep and SurePath slides are similar. The specificity of validated ThinPrep slides is slightly higher than that of conventional slides. Preliminary results with the nonvalidated SurePath slides show a specificity that is similar to that of ThinPrep slides.
Formal studies performed by the CAP Cytopathology Committee, using data from the Pap program. Formal studies examining three specific issues in gynecologic cytology were completed in 2003. The reproducibility or precision of gynecologic cytologic interpretation, and its association with the underlying diagnosis, was examined in one study. In a second study, the performance of participants on conventional and ThinPrep slides was compared. The third study demonstrated that the detection of adenocarcinoma in situ of the cervix in cervicovaginal specimens is more difficult than that of other high-grade lesions.
The complete 2003 Pap YESR contains a thorough summary of the performance characteristics of all of the interpretive categories used in the Pap program. These results provide a wealth of data for those cytologists who wish to compare their own laboratory performance with that of Pap program participants as a whole, or who wish to identify important diagnostic trends and pitfalls. In addition, the comparative data generated from the PAPC, PAPM, and PAPK modules may help cytologists make decisions about implementing new technologies in their laboratories. Finally, the 2003 Pap YESR continues the trend of previous Pap program reports by providing a rich source of data for publishing practical and relevant studies that advance the field of cytology and improve interpretive accuracy. Be sure to peruse the 2003 Pap YESR when it arrives in your laboratory. It is the most comprehensive and timely summary available for identifying real-life trends and pitfalls in cervicovaginal cytology.
- Renshaw AA, Davey DD, Birdsong GG, et al. Precision in gynecologic cytologic
interpretation: A study from the College of American Pathologists Interlaboratory
Comparison Program in Cervicovaginal Cytology. Arch Pathol Lab Med.
- Renshaw AA, Young NA, Birdsong GG, et al. Comparison of performance of conventional
and ThinPrep gynecologic preparations in the College of American Pathologists
gynecologic cytology program. Arch Pathol Lab Med. 2004;128:17-22.
- Renshaw AA, Mody DR, Lozano RL, et al. Detection of adenocarcinoma in situ
of the cervix in Papanicolaou tests. Arch Pathol Lab Med. 2004;128:153-157.
CAP Interlaboratory Comparison Program in Nongynecologic Cytopathology, or NGC program
David C. Wilbur, MD
In 2003, the NGC program enrolled 1,522 laboratories, 3,673 pathologists, and 2,969 cytotechnologists. Compared with the 2002 calendar year program, this represents a +16 percent, +17 percent, and +17 percent difference, respectively. The data set represent a variety of practice settings, with the largest group representing voluntary nonprofit hospital laboratories (48 percent of those identifying their setting). All nongynecologic specimen volume levels are also represented, with the largest group (43 percent) having between 1,000 and 5,000 nongynecologic specimens annually. Forty-six percent of laboratories reported a volume of 100 to 499 fine needle aspirations per year, with only 10 percent of respondents having an FNA volume greater than 1,000 per year.
The nongynecologic program is structured in a similar fashion to the Pap program. A quarterly challenge of five slides covering a variety of specimen types and anatomic locations is mailed to participants. Interpretations are reported back to the CAP, generally by fax, and there is immediate feedback with the reference interpretations, a summary, and discussion of the findings, and references. By design, the program is intended to be a self-evaluative quality assurance process. Slides are submitted only by members of the CAP Cytopathology Committee and are initially referenced by two experienced cytopathologists from the committee. Slides are evaluated for technical quality and for the representative nature of the material as compared to the submission diagnosis. Slides are not otherwise validated; hence, the program is not intended to be used as a formal proficiency testing function.
Over the course of the several years of data compiled in the program, performance has been shown to be equal or slightly improved from year to year. The 2003 data show little difference. Minor changes in performance rates were noted, but as in prior years, it is difficult to determine if this represents better interpretations or better material to interpret. Therefore, because of the reiterative nature of the annual summary process, for 2003 it was decided that the report would highlight performance of participants in two anatomic sites: salivary gland aspiration cytology and effusion cytology. These areas had been identified in several past surveys to be areas of considerable challenge to participants.
In the area of salivary gland, significant false-positive and false-negative rates for interpretation were identified. The false-positive rates for the benign categories of pleomorphic adenoma and Warthin's tumor were eight percent and eight percent, respectively. Even greater difficulty was shown in the malignant processes of squamous carcinoma (12 percent), adenoid cystic carcinoma (29 percent), acinic cell carcinoma (62 percent), and lymphoma (32 percent). Dr. Jonathan Hughes, a member of the Nongynecologic Working Group, has headed a committee project looking at the details of these cases, and a publication has been submitted detailing and illustrating the differential features noted in many salivary gland aspiration cases from the program examined. Look for that publication in Archives in the near future.
In the effusion category, Ann Moriarty, MD, headed a project to review good
and poorly performing cases. Effusion cases are the most common entity circulating
in the program. Highlights include the finding of differential patterns noted
on morphology in cases of adenocarcinoma showing little versus greater discordance
from the reference interpretations. In addition, features were noted accounting
for apparent interpretational difficulties in cases of small cell carcinoma
and lymphoma. A review with illustration of these findings was published recently
in Archives of Pathology & Laboratory Medicine (Moriarty, et al. 2004;
128: 513- 518). (See "Recommended
In addition to the commentaries in the two specific categories noted above, the 2003 Year End Summary includes the remainder of the data in all other specimen types and body sites for which at least 100 interpretations were rendered by participants. These data provide a wealth of information awaiting the "miner" or interesting observations in differential diagnosis.
In the future, the Nongynecologic Working Group is planning additions to the program that will include Web-based ancillary material for participants to review. Adding such material will make the program more real-life in that ancillary studies routinely ordered or reviewed, or both, in the process of interpretation, such as radiographic images, immunocytochemistry, and flow cytometric evaluation, will be available.
Dr. Hughes, a member of the CAP Cytopathology Committee, is staff pathologist at Laboratory Medicine Consultants, Las Vegas. Dr. Wilbur, chair of the CAP Nongynecologic Working Group, is director of cytology at Massachusetts General Hospital, Boston.
To view the NGC YESR, visit 2003_ngc_final.pdf.
To view the PAP YESR, visit 2003_pap_final.pdf.