Interobserver Variability of Colposcopic Interpretations and Consistency with Final Histologic Results

Objective. To evaluate the interobserver reliability of colposcopy and the prediction of final histologic results. Materials and Methods. Visibility of the squamocolumnar junction, presence and grading of atypical transformation zone (ATZ), and presence and grading of cervical intraepithelial neoplasia (CIN) were assessed by nine expert colposcopists in 100 cervigrams. Pairwise κ, group κ, and class-specific κ values were computed. Frequency and degree of interobserver disagreement on CIN and ATZ grading and correlation with histologic results were studied. Results. The median pairwise κ and the group κ values were excellent for ATZ detection, acceptable for the squamocolumnar junction, and poor for the other variables; the presence of ATZ had a sensitivity for CIN of 90.2% and a specificity of 48.6%; the colposcopic impression of high-grade CIN had a sensitivity of 54.4% and a specificity of 88%. Conclusions. Some colposcopic predictions are reproducible between observers, predictive of the histologic results, and can be integrated in the algorithm for the management of patients with abnormal Pap smear results.

[1]  D. Santini,et al.  Independent determinants of inaccuracy of colposcopically directed punch biopsy of the cervix. , 2003, Gynecologic oncology.

[2]  Yvonne C Collins,et al.  Strength of correlations between colposcopic impression and biopsy histology. , 2003, Gynecologic oncology.

[3]  T. Wright,et al.  Assessment of the Cervix After Acetic Acid Wash: Inter‐Rater Agreement Using Photographs , 2002, Obstetrics and gynecology.

[4]  P. Elson,et al.  The colposcopic impression. Is it influenced by the colposcopist's knowledge of the findings on the referral Papanicolaou smear? , 2001, The Journal of reproductive medicine.

[5]  M. Schiffman,et al.  Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic estimates from the ASCUS-LSIL Triage Study. , 2001, JAMA.

[6]  M. Sideri,et al.  Operator variability in disease detection and grading by colposcopy in patients with mild dysplastic smears , 1995, Cancer.

[7]  P. Kenemans,et al.  Observer agreement on interpreting colposcopic images of CIN. , 1995, Gynecologic oncology.

[8]  P. Cristoforoni,et al.  Computerized Colposcopy: Results of a Pilot Study and Analysis of Its Clinical Relevance , 1995, Obstetrics and gynecology.

[9]  S. Ciatto,et al.  Evaluation of the Sensitivity of Cervicography in a Consecutive Colposcopic Series , 1992, Tumori.

[10]  I Russell,et al.  Statistics--with confidence? , 1991, The British journal of general practice : the journal of the Royal College of General Practitioners.

[11]  G. Wilbanks,et al.  An international terminology of colposcopy: report of the Nomenclature Committee of the International Federation of Cervical Pathology and Colposcopy. , 1991, Obstetrics and gynecology.

[12]  J. Paavonen,et al.  Observer Variability in the Scoring of Colpophotographs , 1990, Obstetrics and gynecology.

[13]  J. R. Landis,et al.  The measurement of observer agreement for categorical data. , 1977, Biometrics.

[14]  Jacob Cohen A Coefficient of Agreement for Nominal Scales , 1960 .

[15]  M. Toma A quality-control program for colposcopic practice , 2005 .

[16]  J. Andrade,et al.  Colposcopic scoring system for biopsy decisions in different patient groups. , 2000, European journal of gynaecological oncology.

[17]  Z. Rokyta Diagnostic reliability of prebioptic methods in the prediction of a histological basis of cervical lesions and its correlation with accuracy of colposcopically directed biopsy in patients with cervical neoplasia. , 2000, European journal of gynaecological oncology.

[18]  H. D. de Vet,et al.  Interobserver variation in histopathological grading of cervical dysplasia. , 1990, Journal of clinical epidemiology.